Continuous upstream manufacturing process of an antibody product, apparatus to perform the process and antibody product produced thereby

Abstract

This invention provides an adapted perfusion manufacturing method or continuous perfusion manufacturing method, which includes automatically controlling the perfusion rate based on biomass, ensuring a more efficient method. Therefore, the method according to the invention is more manufacturing-friendly and operation-friendly. Apparatus for performing this method and biological products produced by this method are also provided.

Description

Technical Field

[0001] This invention relates to methods of biotechnology, and in particular to the automation of methods (preferably continuous manufacturing methods) for manufacturing biological products such as antibodies. Prior Technology

[0002] Despite advancements in manufacturing, new biopharmaceuticals (protein-based drugs) require novel and optimized manufacturing methods to avoid negative product quality effects, such as protein aggregation. This impacts upstream manufacturing, downstream manufacturing, storage, and application.

[0003] This new protein-based drug includes antibodies, such as bispecific and / or monoclonal antibodies. Bispecific antibody systems are artificial proteins that can simultaneously bind two different types of antigens. They are known in several structural forms and their applications in cancer immunotherapy and drug delivery have been explored (Fan, Gaowei; Wang, Zujian; Hao, Mingju; Li, Jinming (2015). "Bispecific antibodies and their applications". Journal of Hematology & Oncology. 8: 130).

[0004] Generally speaking, bispecific antibodies can be IgG-like, i.e., full-length bispecific antibodies, or non-IgG-like bispecific antibodies that are non-full-length antibody constructs. Full-length bispecific antibodies typically retain the structure of a traditional monoclonal antibody (mAb) with two Fab arms and an Fc region, the difference being that the two Fab sites bind different antigens. Non-full-length bispecific antibodies lack the entire Fc region. These include chemically linked Fabs, Fab regions alone, and various types of bivalent and trivalent single-chain variable fragments (scFvs). Fusion proteins that mimic the variable domains of two antibodies also exist. Among these newer forms, the BiTE® bispecific T-cell conjugate molecule is the most likely to be further developed (Yang, Fa; Wen, Weihong; Qin, Weijun (2016). "Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies". International Journal of Molecular Sciences. 18 (1): 48).

[0005] Bispecific molecules, such as BiTE®, are recombinant protein constructs prepared from two flexibly linked antibody-derived binding domains. One binding domain of the BiTE® molecule is specific to a selected tumor-associated surface antigen on target cells; the second binding domain is specific to CD3 (a subunit of the T cell receptor complex on T cells). Through its specific design, the BiTE® antibody construct is uniquely suited for transiently binding T cells to target cells while simultaneously and potently activating the inherent cytolytic potential of T cells against target cells. Important further development of the first-generation BiTE® molecules, AMG 103 and AMG 110, which have entered clinical trials (see WO 99 / 54440 and WO 2005 / 040220), provides a bispecific molecule that binds to a context-independent epitope at the N-terminus of the CD3ε chain (WO 2008 / 119567). BiTE® molecules binding to this selected epitope not only exhibit cross-species specificity for the CD3ε chain in humans, tamarisks (Callithrix jacchus), woolly tamarins (Saguinus oedipus), or squirrel monkeys (Saimiri sciureus), but also, due to recognizing this specific epitope (rather than the CD3-binding epitope in previously described bispecific T-cell conjugating molecules), do not nonspecifically activate T cells to the same extent observed with previous generations of T-cell conjugating antibodies. This reduction in T-cell activation is associated with less or reduced T-cell redistribution in patients identified as a risk factor for side effects.

[0006] Currently, antibodies, such as bispecific antibodies, are typically produced using fed-batch culture methods. Fed-batch culture is a well-known biotechnological technique in which one or more nutrients (substrates) are fed into a bioreactor during culture, and one or more products remain in the bioreactor until the end of the run (Tsuneo Yamanè, Shoichi Shimizu: Fed-batch Techniques in Microbial Processes [Feed-batch Techniques in Microbial Processes] (1984) Advances in Biochem Eng. / Biotechnol [Biochemical Engineering / Advances in Biotechnology], 30: 147-194). Therefore, bispecific antibody products accumulate in fed-batch methods, easily leading to product quality loss, such as due to aggregation, trimming, or certain chemical degradation reactions. Moreover, no product is obtained until the end of the run. Furthermore, method-related impurities such as host cell proteins (HCPs) also accumulate in the bioreactor during fed-batch methods. Downstream removal of such impurities is often challenging and requires additional measures and resources to ensure the quality of the final product. The necessary repetitive growth phases, requiring a new cell culture growth stage for each new run, impair the overall productivity of fed-batch production. Furthermore, large bioreactors requiring significant space and energy are needed to achieve sufficient product quantities produced by fed-batch equipment. Therefore, there is a need for improved upstream manufacturing methods, particularly for the production of bispecific antibodies, that increase both product quantity and quality to provide sufficient product quantities at commercial scale with a quality that minimizes product rejection during downstream processing. Given the cost of large-scale cell culture methods and the growing demand for larger quantities and lower costs of bioproducts supplied to patients with significant unmet medical needs, new process methods offering even incrementally improved capabilities in recombinant protein production and recovery are valuable. In this regard, perfusion or continuous perfusion production of bioproducts in bioreactors is a promising strategy that offers gains in method productivity, flexibility, and efficiency. Since continuous methods are more complex to operate than perfusion or fed-batch methods, greater automation is required to achieve robust results. To improve the robustness of higher biomass methods, an automated feeding strategy based on biomass is proposed. Summary of the Invention

[0007] Surprisingly, adaptive perfusion or continuous perfusion manufacturing methods, including biomass-based automated perfusion rate control, can be provided. This biomass-based automated perfusion rate control ensures a more efficient method that is automated and less prone to human error, such as that caused by operators. Therefore, the method according to the invention is more manufacturing-friendly and operationally friendly. In this respect, improved yields of (bispecific) antibody products can be obtained. Even though continuous manufacturing methods for producing proteins such as antibodies are known (e.g., Cattaneo et al., US 2017 / 0204446 A1), such methods are not suitable for the specific needs of bispecific antibodies, which tend to aggregate, trim, and chemically degrade in upstream manufacturing steps, resulting in lower product quantity and quality. Adjusting the reactor biomass (cell concentration) is one of the main levers for achieving these gains. Proof-of-concept experiments showed that high biomass (up to 40%) with a constant feed rate resulted in lower viability (>75%), while high biomass (up to 50%) with a biomass-based feed rate resulted in higher viability (>90%). Furthermore, in the context of this invention, it has been found that achieving and maintaining a high biomass in a bioreactor, i.e., at least 30%, 40%, 50%, 60%, 70%, 75%, 80%, or even 90% of the filled cell volume (PCV, i.e., the percentage of solids in the cell suspension in the bioreactor), results in less lactate production. Lactate is known to be detrimental to biotechnological production methods because cells typically cease to grow in high lactate environments, which means a decrease in biomass and ultimately, yield and productivity.

[0008] Therefore, in one aspect, within the context of this invention, it is contemplated to provide an upstream manufacturing method for producing antibody products by applying automated (rather than manual (i.e., non-automatic)) measurement and regulation of the perfusion rate in a perfusion bioreactor (basic setup shown in Figure 1), the method comprising the following steps: (i) A liquid cell culture medium is provided in the perfusion bioreactor, the liquid cell culture medium comprising at least one mammalian cell culture capable of expressing the antibody product, and wherein the cells have a concentration of at least 1 x 10^5 cells / mL (viable cell density, VCD) when inoculated in the perfusion bioreactor. (ii) A first control loop is provided for measuring and regulating the culture medium level in the bioreactor, the first control loop comprising a level probe for measuring the culture medium level in the bioreactor relative to a set point, a osmotic pump calibrated to measure the permeation rate (volume / time), and a level control device receiving inputs from the level probe and the osmotic pump, which, in response to the inputs from the level probe and the osmotic probe, enables the culture medium pump (feed pump) to correct to the culture medium feed rate of the bioreactor, or wherein the level control device receiving inputs from the level probe and the culture medium pump, which, in response to the inputs from the level probe and the culture medium probe, enables the osmotic pump to correct outflow from the bioreactor; wherein the measurement of the culture medium level in the bioreactor is performed at preset fixed time intervals; (iii) A second control loop is provided for measuring and regulating the biomass in the bioreactor, the second control loop comprising a permittivity probe or Raman probe, preferably a permittivity probe, for measuring the biomass in the bioreactor, and a biomass control device for receiving input from the biomass permittivity probe or Raman probe, which, in response to the input, enables a discharge pump to correct the discharge rate from the bioreactor; wherein the biomass measurement in the bioreactor is performed at preset fixed time intervals; (iv) An integrated first and second control loop is provided by connecting the biomass control device and the level control device to an integrated unit, wherein the integrated unit is capable of performing automatic infusion rate calculation, wherein the infusion rate is a function of the biomass value, preferably based on the following equation: Perfusion rate (mL / min) = a function of biomass value (permeability, PCV, VCD, spectral value) and / or Perfusion rate [mL / min] = Perfusion rate based on permittivity (constant) [cm / pF / d] x Permittivity value [pF / cm] The constant is the permeation rate [1 / d] divided by the permittivity [pF / cm], wherein the permittivity value is 0.5 to 120 pF / cm during the first period (growth phase) in the bioreactor when the biomass increases to approximately the predetermined biomass setpoint, and / or 25 to 100 pF / cm during the second phase (production phase) when the biomass stabilizes after reaching the predetermined biomass setpoint.

[0009] (v) The integrated unit automatically corrects or maintains the perfusion rate, and in response to the biomass measured at preset fixed time intervals, the integrated unit sends signals to the osmotic pump and / or the culture medium pump to increase or decrease the pump rate.

[0010] In this aspect of the invention, it is envisioned that in step (i), the cells are inoculated in the bioreactor at a concentration of at least 7 x 10^5 cells / mL.

[0011] In this aspect of the invention, it is envisioned that in step (iv), the biomass setpoint is equal to at least 30 x 10^6 cells / mL of VCD, preferably 30 x 10^6 cells / mL (if the manufacturing method is a fed-batch based method) and 65 x 10^6 cells / mL (if the manufacturing method is a continuous manufacturing method).

[0012] In this aspect of the invention, it is envisioned that in step (iv), the cell culture is grown for at least 4 days, preferably at least 7 days, and more preferably at least 12 or 14 days.

[0013] In this aspect of the invention, it is envisioned that in step (ii), the preset fixed time interval corresponds to at most 1 minute, preferably 30 seconds, more preferably at most 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds, and preferably 1 second.

[0014] In this aspect of the invention, it is envisioned that in step (iii), the preset fixed time interval corresponds to at most 1 minute, preferably 30 seconds, more preferably at most 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds, and preferably 1 second.

[0015] In this aspect of the invention, it is envisioned that in step (v), the preset fixed time interval corresponds to at most 1 minute, preferably 30 seconds, more preferably at most 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds, and preferably 1 second.

[0016] In this aspect of the invention, the capacitance during the growth stage is envisioned to be 0.70 to 120 pF / cm, more preferably 0.73 to 70.7 pF / cm, and even more preferably 1 to 20 pF / cm or 100 to 117 pF / cm (if the manufacturing method is a continuous manufacturing method).

[0017] In this aspect of the invention, a cell-specific perfusion rate based on capacitance is envisioned to be 0.01 to 0.049 cm / pF / d during the growth phase, preferably 0.015 to 0.04 cm / pF / d, more preferably 0.02 to 0.04 cm / pF / d, and most preferably 0.0266 to 0.04 cm / pF / d. In the case of discontinuous manufacturing, the upper limit can be higher, such as up to 0.2 cm / pF / d, and more preferably up to 0.13 cm / pF / d.

[0018] In this aspect of the invention, it is envisioned that the applied perfusion rate corresponds to 0.01 to 0.1 nL / cell / d during the growth phase, preferably 0.02 to 0.08 nL / cell / d, and more preferably 0.027 to 0.076 nL / cell / d during the growth phase.

[0019] In this aspect of the invention, it is envisioned that the capacitance during the production stage is 55 to 85 pF / cm, more preferably 60 to 75 pF / cm, and even more preferably 62 to 73 pF / cm.

[0020] In this aspect of the invention, it is envisioned that the cell-specific perfusion rate based on capacitance is 0.01 to 0.04 cm / pF / d during the production phase, preferably 0.01 to 0.035 cm / pF / d, and more preferably 0.01 to 0.0266 cm / pF / d.

[0021] In this aspect of the invention, it is envisioned that the applied perfusion rate corresponds to a CSPR of 0.01 to 0.49 nL / cell / d during the production phase, preferably 0.015 to 0.04 nL / cell / d, and particularly preferably 0.023 to 0.035 nL / cell / d.

[0022] In this aspect of the invention, it is envisioned that the production phase requires at least 14 days, preferably at least 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 or 32 days (if the production method is a continuous manufacturing method) and at least 3 days, preferably 4 or 5 days (if the production method is based on a feed-in batching method).

[0023] In this aspect of the invention, the upstream manufacturing method is envisioned as a pouring method (e.g., batch feeding) or a continuous pouring (continuous manufacturing) method.

[0024] In this aspect of the invention, the antibody product is envisioned to be a full-length antibody, such as a monoclonal antibody, such as an IgG antibody, preferably targeting PD-1, or a non-full-length bispecific molecule.

[0025] In this aspect of the invention, it is envisioned that the antibody products are full-length antibodies or molecules based on full-length antibodies or fragments thereof, which are preferably bispecific.

[0026] In this aspect of the invention, it is envisioned that the antibody product is a fusion protein, preferably an anti-PD-1 mAb / IL-21 mutant protein fusion protein.

[0027] In this aspect of the invention, the antibody product is envisioned as a bispecific non-full-length antibody molecule comprising first and second binding domains that bind to target cells and effector cells, respectively.

[0028] In this aspect of the invention, it is envisioned that the bispecific molecule includes a half-life extension portion, which is preferably selected from human serum albumin (HAS), HAS binding domain or Fc-based half-life extension portion derived from IgG antibody, and most preferably scFc half-life extension portion.

[0029] In this aspect of the invention, a bispecific molecule is envisioned as a bispecific T-cell conjugate molecule.

[0030] In this aspect of the invention, it is envisioned that the first binding domain of the bispecific molecule binds to at least one target cell surface antigen selected from the group consisting of: CD19, CD33, EGFRvIII, MSLN, CDH19, FLT3, DLL3, CDH3, EpCAM, CD70, MUC17, CLDN18, BCMA, and PSMA.

[0031] In this aspect of the invention, it is envisioned that the second binding domain of the bispecific antibody product binds to CD3.

[0032] In this aspect of the invention, the first binding structural domain is envisioned to include a VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from the group consisting of: and a VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from the group consisting of: (a) CDR-H1 as depicted in SEQ ID NO: 1, CDR-H2 as depicted in SEQ ID NO: 2, CDR-H3 as depicted in SEQ ID NO: 3, CDR-L1 as depicted in SEQ ID NO: 4, CDR-L2 as depicted in SEQ ID NO: 5, and CDR-L3 as depicted in SEQ ID NO: 6, (b) CDR-H1 as depicted in SEQ ID NO: 29, CDR-H2 as depicted in SEQ ID NO: 30, CDR-H3 as depicted in SEQ ID NO: 31, CDR-L1 as depicted in SEQ ID NO: 34, CDR-L2 as depicted in SEQ ID NO: 35, and CDR-L3 as depicted in SEQ ID NO: 36, (c) CDR-H1 as depicted in SEQ ID NO: 42, CDR-H2 as depicted in SEQ ID NO: 43, CDR-H3 as depicted in SEQ ID NO: 44, CDR-L1 as depicted in SEQ ID NO: 45, CDR-L2 as depicted in SEQ ID NO: 46, and CDR-L3 as depicted in SEQ ID NO: 47, (d) CDR-H1 as depicted in SEQ ID NO: 53, CDR-H2 as depicted in SEQ ID NO: 54, CDR-H3 as depicted in SEQ ID NO: 55, CDR-L1 as depicted in SEQ ID NO: 56, CDR-L2 as depicted in SEQ ID NO: 57, and CDR-L3 as depicted in SEQ ID NO: 58, (e) CDR-H1 as depicted in SEQ ID NO: 65, CDR-H2 as depicted in SEQ ID NO: 66, CDR-H3 as depicted in SEQ ID NO: 67, CDR-L1 as depicted in SEQ ID NO: 68, CDR-L2 as depicted in SEQ ID NO: 69, and CDR-L3 as depicted in SEQ ID NO: 70, (f) CDR-H1 as depicted in SEQ ID NO: 83, CDR-H2 as depicted in SEQ ID NO: 84, CDR-H3 as depicted in SEQ ID NO: 85, CDR-L1 as depicted in SEQ ID NO: 86, CDR-L2 as depicted in SEQ ID NO: 87, and CDR-L3 as depicted in SEQ ID NO: 88, (g) CDR-H1 as depicted in SEQ ID NO: 94, CDR-H2 as depicted in SEQ ID NO: 95, CDR-H3 as depicted in SEQ ID NO: 96, CDR-L1 as depicted in SEQ ID NO: 97, CDR-L2 as depicted in SEQ ID NO: 98, and CDR-L3 as depicted in SEQ ID NO: 99, (h) CDR-H1 as depicted in SEQ ID NO: 105, CDR-H2 as depicted in SEQ ID NO: 106, CDR-H3 as depicted in SEQ ID NO: 107, CDR-L1 as depicted in SEQ ID NO: 109, CDR-L2 as depicted in SEQ ID NO: 110, and CDR-L3 as depicted in SEQ ID NO: 111. (i) CDR-H1 as depicted in SEQ ID NO: 115, CDR-H2 as depicted in SEQ ID NO: 116, CDR-H3 as depicted in SEQ ID NO: 117, CDR-L1 as depicted in SEQ ID NO: 118, CDR-L2 as depicted in SEQ ID NO: 119, and CDR-L3 as depicted in SEQ ID NO: 120, (j) CDR-H1 as depicted in SEQ ID NO: 126, CDR-H2 as depicted in SEQ ID NO: 127, CDR-H3 as depicted in SEQ ID NO: 128, CDR-L1 as depicted in SEQ ID NO: 129, CDR-L2 as depicted in SEQ ID NO: 130, and CDR-L3 as depicted in SEQ ID NO: 131, (k) CDR-H1 as depicted in SEQ ID NO: 137, CDR-H2 as depicted in SEQ ID NO: 138, CDR-H3 as depicted in SEQ ID NO: 139, CDR-L1 as depicted in SEQ ID NO: 140, CDR-L2 as depicted in SEQ ID NO: 141, and CDR-L3 as depicted in SEQ ID NO: 142, (l) CDR-H1 as depicted in SEQ ID NO: 152, CDR-H2 as depicted in SEQ ID NO: 153, CDR-H3 as depicted in SEQ ID NO: 154, CDR-L1 as depicted in SEQ ID NO: 155, CDR-L2 as depicted in SEQ ID NO: 156, and CDR-L3 as depicted in SEQ ID NO: 157, (m) CDR-H1 as depicted in SEQ ID NO: 167, CDR-H2 as depicted in SEQ ID NO: 168, CDR-H3 as depicted in SEQ ID NO: 169, CDR-L1 as depicted in SEQ ID NO: 170, CDR-L2 as depicted in SEQ ID NO: 171, and CDR-L3 as depicted in SEQ ID NO: 172, (n) CDR-H1 as depicted in SEQ ID NO: 203, CDR-H2 as depicted in SEQ ID NO: 204, CDR-H3 as depicted in SEQ ID NO: 205, CDR-L1 as depicted in SEQ ID NO: 206, CDR-L2 as depicted in SEQ ID NO: 207, and CDR-L3 as depicted in SEQ ID NO: 208; (o) CDR-H1 as depicted in SEQ ID NO: 214, CDR-H2 as depicted in SEQ ID NO: 215, CDR-H3 as depicted in SEQ ID NO: 216, CDR-L1 as depicted in SEQ ID NO: 217, CDR-L2 as depicted in SEQ ID NO: 218, and CDR-L3 as depicted in SEQ ID NO: 219; (p) CDR-H1 as depicted in SEQ ID NO: 226, CDR-H2 as depicted in SEQ ID NO: 227, CDR-H3 as depicted in SEQ ID NO: 228, CDR-L1 as depicted in SEQ ID NO: 229, CDR-L2 as depicted in SEQ ID NO: 230, and CDR-L3 as depicted in SEQ ID NO: 231; and (q) CDR-H1 as depicted in SEQ ID NO: 238, CDR-H2 as depicted in SEQ ID NO: 239, CDR-H3 as depicted in SEQ ID NO: 240, CDR-L1 as depicted in SEQ ID NO: 241, CDR-L2 as depicted in SEQ ID NO: 242, and CDR-L3 as depicted in SEQ ID NO: 243.

[0033] In this aspect of the invention, it is envisioned that perfusion culture is continuously operated for at least 7 days, preferably at least 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 or 28 days, and most preferably at least 35 days, by feeding and discharging additional cells from the bioreactor at a defined cell-specific perfusion rate to maintain the biomass set point.

[0034] As a second aspect of the invention, it is envisioned to provide an apparatus for performing a continuous upstream manufacturing method as described in claim 1, the apparatus comprising a perfusion bioreactor, the first control loop, the second control loop, and an integrated unit.

[0035] As a third aspect of the invention, it is envisioned to provide (bispecific) antibody products produced by the upstream manufacturing method as described in claim 1. Simple Explanation of the Diagram

[0036] [ [picture] [1] illustrates one arrangement of an automated continuous manufacturing method according to the present invention, wherein (i.) the permeation pump is controlled by a horizontal controller that receives input from a horizontal probe and a feed pump, or wherein (ii.) the feed pump is controlled by a horizontal controller that receives input from a horizontal probe and a permeation pump. The second case (ii.) is preferred in the embodiments of the present invention.

[0037] [ [picture] [2] shows viability (A), PCV (B), capacitance-specific perfusion rate (CSPR) (C), and biomass-specific perfusion rate (BSPR, (D)) and cell culture values ​​during the production phase of the continuous perfusion method. The control method had an artificial time-based feed rate [volume / day] and a fixed PCV. The test conditions had a manual PCV-based feed [volume / day / %PCV] and a manually increased PCV setpoint. This test condition resulted in good cell growth, i.e., PCV increased to 50%, with high viability (after day 23). Required setpoint: 0.078 L / day from day 23 to day 33.

[0038] [ [picture] [3] shows the metabolite values ​​(lactate (A), osmolarity (B), glucose (C), and ammonia (D)) during the production phase of the continuous perfusion method with a manually controlled PCV-based feed rate [volume / day / PCV]. The control method had a manually controlled time-based feed rate [volume / day] and a fixed PCV. The test conditions had a manually controlled PCV-based feed [volume / day / %PCV] and manually incremented PCV setpoints. The trends in metabolite changes were similar between the test and control conditions.

[0039] [ [picture] [4] shows viability (A), PCV (B), CSPR (C), and BSPR (D) under capacitance-based automated perfusion rate [cm / pF·day] control during the growth phase of the continuous perfusion method. The control method had an artificial time-based feed rate [volume / day] during the growth phase. The test conditions had capacitance-based automated feeding (0.04 cm / pF·day for days 4–10 and 0.03 cm / pF·day for days 11–14). These test conditions resulted in good cell growth.

[0040] [ [picture] [5] shows the metabolite values ​​(lactate (A), osmolarity (B), glucose (C), and ammonia (D)) controlled by an automated perfusion rate [cm / pF·day] based on capacitance during the growth phase of the continuous perfusion method. The control method had an artificial time-based feed rate [volume / day] during the growth phase. The test conditions had an automated feed based on capacitance (0.04 cm / pF·day for days 4–10 and 0.03 cm / pF·day for days 11–14). The trends in metabolite changes were similar between the test and control conditions.

[0041] Figure 6 shows the CSPR (A) and BSPR (B), PCV (C), and permeability (D) values ​​during the production phase of the continuous infusion method. A fixed feed rate [volume / day] was maintained from days 12 to 26. Automatic infusion rate control based on capacitance was implemented from days 27 to 32 at a rate of 0.0277 cm / pF·day. Three PCV setpoints (19%, 23%, and 27%) were tested throughout the experiment from days 12 to 32. PCV was more tightly controlled during the capacitance-based automatic feed from days 27 to 32 compared to the fixed feed rate from days 12 to 26.

[0042] [ [picture] [7] The production stages of the continuous perfusion method are shown in the perfusion product concentration (titer) (A) and cell viability (B), as well as lactate (C) and osmolarity (D). A fixed feed rate [volume / day] was used from day 12 to 26. Automatic perfusion rate control based on capacitance was performed from day 27 to 32 at a rate of 0.0277 cm / pF·day. Three PCV setpoints (19%, 23%, and 27%) were tested throughout the experiment from day 12 to 32. Higher titers were observed at higher PCV setpoints under fixed feed rate and capacitance-based feed conditions. Titer, lactate, and osmolarity were more stable at a fixed BSPR from day 27 to 32. Titer increased over time from day 12 to 26 at a fixed feed rate.

[0043] [ [picture] [8] shows the VCD (A), production rate (B), viability (C), and product concentration (D) (titer) in permeate for the growth and production phases of continuous production of CD70xCD3 bispecific T cell conjugate molecules. Control conditions (triangle symbols) had a fixed feed rate [volume / day]. Test conditions had capacitance-based autoperfusion rates at three levels for both growth and production phases: high rate (cross symbol; 0.065 cm / pF.day for days 0–6 and 0.035 cm / pF.day for days 7–12), medium rate (dash symbol; 0.03, 0.017), and low rate (circle symbol; 0.02, 0.01). Under non-steady-state cell culture conditions, the highest rate resulted in higher yields, titers, and lower viability. The lowest rate resulted in steady-state cell culture operation with stable viability but lower yields and titers. The medium rate group performed similarly to the control group, falling between the high rate and low rate groups.

[0044] [ [picture] [9] shows the productivity (A), product concentration (titer) (B), VCD (C), and viability (D) of the PD1 x IL21 mutant protein antibody construct in the perfusion method. Control and historical conditions had a fixed feed rate [volume / day]. Test conditions had an automated perfusion rate based on capacitance during the growth and production phases (0.12 cm / pF.day for days 3–8 and 0.03 cm / pF.day for days 9–15). Test conditions had higher VCD and viability, and lower and / or similar productivity compared to control and historical conditions.

[0045] [ [picture]

[10] shows the productivity (A), product concentration (titer) (B), VCD (C), and viability (D) of the PD1 mAb infusion method. Control and historical conditions had a fixed feed rate [volume / day]. Test conditions had an automated infusion rate based on capacitance during the growth and production phases (0.12 cm / pF·day for days 3–8 and 0.03 cm / pF·day for days 9–15). The test conditions had higher productivity, lower titer, VCD, and viability compared to historical conditions. Implementation

[0046] This invention provides a perfusion or continuous perfusion method for manufacturing biopharmaceuticals, namely therapeutic proteins, particularly antibodies and bispecific molecules. The invention envisions adapting the upstream process to the specific needs of manufacturing bispecific antibodies. Compared to standard fed-batch manufacturing solutions known in the art, the upstream process not only helps increase productivity but also helps reduce space requirements. More importantly, the continuous manufacturing method of the present invention—preferably a continuous upstream manufacturing method—is particularly suitable for bispecific antibodies and envisions producing higher product quality, i.e., less aggregated bispecific antibodies, with higher monomer content compared to fed-batch manufacturing. In particular, the method according to the invention, in itself, does not require new components or instruments, but rather new combinations of known components and instruments, and applies new, very high-frequency online measurements and program control directly reacting to said measurements, to facilitate automation of a continuous manufacturing method typically within one minute, preferably only one second.

[0047] This invention is based on the precise and timely measurement of biomass during the growth and production stages to control the production process of bioproducts such as antibodies, antibody constructs, or bispecific T-cell conjugate molecules. In the context of this invention, biomass is preferably measured by biocapacitance measurement, but it can also be measured manually offline. However, in the context of this invention, Raman spectroscopy is also conceived as an advantageous method for determining biomass and thus controlling the production process of the desired bioproduct using Raman probes. Cell cultures controlled according to this invention can generally be controlled as a steady state in continuous perfusion or in a non-steady mode in either continuous or non-continuous perfusion methods.

[0048] In the context of this invention, Raman spectroscopy is understood as a spectroscopic technique utilizing Raman scattering or inelastic scattering from a monochromatic laser, and it is used to study the rotational, vibrational, and other modes of a system. The interaction between phonons and the laser results in an energy shift, which provides information about the phonon modes in the system. Typically, a laser beam is used to illuminate the sample. Electromagnetic radiation from the laser's midpoint is collected by a lens and passed through a collimator. Molecules are excited from their ground state to excited states and then relax to vibrationally excited states. This produces Stokes Raman scattering. If the molecules are assumed to be in a vibrational state, it is called anti-Stokes Raman scattering. Molecules require a change in polarizability to exhibit Raman scattering. The intensity of Raman scattering depends on the change in polarizability, while the Raman shift depends on the vibrational energy levels involved. Advanced versions of Raman spectroscopy include stimulated Raman spectroscopy; surface-enhanced Raman spectroscopy; and resonance Raman spectroscopy. Raman spectroscopy results can be correlated with biomass and lead to biomass measurements.

[0049] In the context of this invention, permittivity (usually denoted by the Greek letter ε (epsilon)) is a measure of the polarizability of a dielectric. Materials such as the (external) cell membrane of (living) cells, which have high permittivity, are more polarized in response to an applied electric field than materials with low permittivity, thus storing more energy in the electric field. Therefore, as described herein, permittivity plays an important role in determining capacitance. Typically, the electric displacement field D produced by an applied electric field E is εE. More generally, permittivity is a thermodynamic function of the state. It may depend on the frequency, magnitude, and direction of the applied electric field. The SI unit of permittivity is farads per meter (F / m). The capacitance of a capacitor is based on its design and architecture, meaning it does not change with charging and discharging. The capacitance formula for a parallel-plate capacitor is as follows: Where A represents the area of ​​a plate, d represents the distance between the plates, and ε represents the permittivity of the dielectric between the two plates.

[0050] This method does not require the additional step of converting biocapacitance (pF / cm) to viable cell density (VCD - 1E6 cells / mL) as in prior art methods. Similarly, the constant applied is technically not a cell-specific perfusion rate (CSPR - mL / 1E6 cells·day), but a capacitance-specific perfusion rate (cm / pF·day), which also differs from typical methods found in prior art.

[0051] In the context of this invention, cytokines of 1 are used. The cell line (typically CHO cells in the context of this invention) remains unchanged. As proposed in the prior art, it is possible to correlate biocapacitance with VCD, but this model will also depend on cell diameter, which can fluctuate. Biocapacitance correlates better with biomass as measured by filler cell volume (PCV). Here, PCV measurements are used only to adjust the biocapacitance setpoint for the continuous manufacturing process of the bispecific molecule, not for cytokines.

[0052] In the prior art, Dowd et al. adjusted the values ​​hourly based on two biocapacitance measurements (converted to VCD using a model). The method of the system according to the invention acquires biocapacitance readings more frequently (e.g., per second) and calculates the perfusion (osmosis) rate per second to obtain an optimized biomass-specific perfusion rate (BSPR), which increases productivity, for example, by allowing for higher PCV and increasing product quality (e.g., by avoiding high lactate levels).

[0053] The constants in the equations applied to the integrated unit are based on permittivity (not cell density, i.e., CSPR). The units of the constants are cm / pF / day. This is derived by dividing the permeation rate (1 / day) by the permittivity (pF / cm). The constants, or several constants, according to the invention apply to any given molecule. These constants can depend on the cell line, culture medium, and molecular stability.

[0054] Several advantages have been found in the method according to the invention compared to methods described in the prior art. The method is simpler and does not require an additional VCD model. The applied constants can be modified for each molecule / method / cell line / culture medium. However, it is not required to calculate cytokines for each new method. Furthermore, the method according to the invention does not introduce errors in VCD estimation. If cell diameter increases due to some adverse methodological event, the VCD model will become inaccurate. Moreover, the method according to the invention is more sensitive to changes in biomass because the second control loop typically measures once per second. Advantageously, no instability has been observed due to the faster integrated control loop.

[0055] In the context of this invention, it is particularly advantageous to provide equilibrium method parameters particularly suitable for bispecific molecules such as those described herein. Lower CSPR typically results in higher productivity, but the CSPR must not fall below the lower limit described herein because there exists a minimum value beyond which viability will decrease drastically, making the method unusable and subsequently leading to failure. For example, for the CD70 x CD3 bispecific molecule described herein, a preferred CSPR is 0.01 nL / cell / day.

[0056] In the context of this invention, the preferred benefit of CSPR-based feed is that it can increase biomass (and therefore productivity) without significantly impacting viability, as shown in Figure 2. Furthermore, in the context of this invention, CSPR-based feed has a metabolite profile at least similar to the control (manual) one, even though biomass is typically increased, as shown in Figure 3.

[0057] Specific low product concentrations in bioreactors decisively help avoid aggregates, i.e., higher relative and / or absolute monomer concentrations of the product. This is crucial for ensuring product quality and improving the overall economics of the process. The fewer aggregates generated upstream, the less non-quality products must be removed downstream. Product concentrations below 3.5 g / L are associated with a lower likelihood of aggregation. If the highest product concentration is maintained below 1.2 g / L throughout the upstream process, product quality is even better. Even better are product concentrations below 0.5 or even 0.3 g / L. By ensuring a sufficiently high perfusion rate of 1 vvd or preferably at least 2 vvd or higher, a cost-effective production rate with preferably no aggregates can be achieved. This applies to all bispecific antibody products, whether full-length or non-full-length antibodies, such as (single-chain) bispecific molecules.

[0058] In the context of this invention, another surprising aspect is the fact that, for bispecific antibody products, it is preferable to adjust the perfusion rate relative to the viable cell density (VCD) to obtain favorable product quality and quantity. In this regard, the perfusion rate is increased continuously, gradually, or incrementally after inoculation until a preferred set point is reached. Typically, the set point is reached when the biomass set point equals at least 35 × 10^6 cells / mL, preferably at least 65 × 10^6 cells / mL, more preferably at least 71 × 10^6 cells / mL, and most preferably at least 85 × 10^6 cells / mL of average viable cell density (VCD). Generally, the perfusion rate is set to a low value as long as the VCD is low relative to the maximum VCD achieved in the same method. For example, when the VCD is equal to about 0.5 × 10^6 cells, the perfusion rate can be as low as about 0.4 vvd. However, as the VVD increases due to cell growth in the bioreactor, when a biomass setpoint of, for example, 35 x 10^6 cells / mL is reached, the perfusion rate can be increased continuously, gradually, or incrementally from 0.4 VVD to 2 VVD. Preferably, the higher the biomass setpoint, the greater the increase in the perfusion rate. For example, when the biomass setpoint is, for example, at least 65 × 10^6 cells / mL, more preferably at least 71 × 10^6 cells / mL, and most preferably at least 85 × 10^6 cells / mL, the VVD can be set to at least 2, more preferably at least 2.01, 3, 4, 5, 6, or even 6.4.

[0059] Within the context of this invention, it is also envisioned that the perfusion rate be adjusted throughout the continuous manufacturing process, depending on the continuously measured VCD. VCD is understood to be an easily accessible and reliable parameter. The integrated live cell density (IVCD) is understood herein as the area under the VCD curve as a function of time; therefore, for example, it may be preferable to maintain a constant cell-specific perfusion rate (CSPR, nL / cell / day), which in turn contributes to controlled product concentrations in the bioreactor to avoid negative impacts on product quality.

[0060] As a result, it is best to ensure controlled and low product concentrations throughout the continuous upstream manufacturing process – for example, less than 1.2 g / L for full-length bispecific antibodies, less than 0.4 g / L for HLE bispecific molecules, and less than 0.12 g / L for non-HLE bispecific molecules. This results in less product being affected by aggregation, trimming, or other chemical degradation.

[0061] In the context of this invention, CSPR is understood as the ratio of perfusion rate D (bioreactor volume / day) to mean VCD (CV, i.e., mean number of viable cells / mL):

[0062] In the context of this invention, it should also be understood that it is preferable to provide a consistent microenvironment to the cells in the cell culture, regardless of cell density. Therefore, it is preferable to exchange the culture medium at a rate proportional to the cell density. By applying a perfusion rate based on a preferred CSPR, the perfusion rate is analogous to the cell density.

[0063] In the context of this invention, CSPR is preferably applied automatically by a control station with online biomass measurement, such as based on biocapacitance or Raman spectroscopy. This preferably allows for small and / or stable adjustments to D in response to changes in CV. This stable, i.e., continuous response is preferable to gradual, i.e., incremental or discrete changes in D. Minimal CSPR is the rate at which the minimum amount of nutrients required to meet cellular needs and support high productivity is delivered. At high cell densities, such as in high cell density cultures (HCDC), the application of minimal or near-minimum CSPR is of particular practical importance. In the context of this invention, HCDC is, for example, directed to cell cultures with a VCD of at least 65 × 10^6 cells / mL, preferably at least 71 × 10^6 cells / mL or even at least 85 × 10^6 cells / mL. HCDC with a VCD of at least 100 × 10^6 cells / mL is also envisioned.

[0064] In the context of this invention, a typical minimum CSPR is 0.01 nl / cell / day. Within the preferred range common to all bispecific molecules or antibodies contemplated herein, some do indeed have more preferred values ​​for optimal product quality and / or quantity. For example, in the context of this invention, the CSPR of the CD19 x CD3 BiTE® molecule is preferably less than 0.04 nl / cell / day, more preferably equal to or less than 0.028 nl / cell / day. For bispecific molecules containing an I2C domain targeting CD3 (SEQ ID NO 26), such as the CD33 x CD3 BiTE® molecule, the CSPR is preferably equal to or less than 0.028 nl / cell / day or at least 0.051 nl / cell / day, more preferably 0.06 to 0.1 nl / cell / day. For full-length bispecific antibodies such as TNF-α x TL1A bispecific antibodies or PD1 inhibitory mAbs, a CSPR (CD-based CSPR) is preferably equal to or less than 0.028 nl / cell / day, more preferably less than 0.2 nl / cell / day or at least 0.051 nl / cell / day, and more preferably 0.06–0.1 nl / cell / day. For the bispecific molecules disclosed herein, such as CD70xCD3 bispecific T-cell conjugate molecules, lower capacitance-based CSPR values ​​have been found to advantageously lead to higher productivity compared to classic fed-batch production at the cost of lower titration (see, for example, Figure 9). Therefore, the proposed method has proven suitable for automated and automated production processes of biopharmaceuticals, particularly bispecific T-cell conjugate molecules, to ensure resource security while simultaneously improving productivity.

[0065] In the context of this invention, "cell culture" or "culture" refers to the growth and reproduction of cells outside of multicellular organisms or tissues. Suitable culture conditions for mammalian cells are known in the art. See, for example, *Animal cell culture: A Practical Approach*, edited by D. Rickwood, Oxford University Press, New York (1992). Mammalian cells can be cultured in suspension or attached to a solid substrate.

[0066] The term "mammalian cell" refers to any cell derived from or originating from any mammal (e.g., human, hamster, mouse, green monkey, rat, pig, cow, or rabbit). For example, a mammalian cell may be an immortalized cell. In some embodiments, the mammalian cell line is a differentiated cell. In some embodiments, the mammalian cell line is an undifferentiated cell. Non-limiting examples of mammalian cells are described herein. In the context of this invention, the preferred type of mammalian cell line is GS-KO cell. Further examples of mammalian cells are known in the art.

[0067] As used herein, the term "cell culturing medium" (also known as "culture medium," "cell culture media," or "tissue culture medium") refers to any nutrient solution used to grow cells (e.g., animal or mammalian cells) and typically provides at least one or more of the following components: energy (usually in the form of carbohydrates, such as glucose); one or more of all essential amino acids, typically twenty basic amino acids, plus cysteine; vitamins and / or other organic compounds typically required in low concentrations; lipids or free fatty acids; and trace elements, such as inorganic compounds or naturally occurring elements, typically required in very low concentrations (usually in the micromolar range).

[0068] Cell culture media include those that are typically used and / or are known to be used in any cell culture method, such as, but not limited to, batch, extended batch, fed-batch, and / or perfusion or continuous cell culture.

[0069] "Growth" cell culture medium or fed medium refers to a cell culture medium typically used during the exponential growth phase ("growth phase") and which is sufficiently intact to support cell culture during this phase. Growth cell culture media may also contain selectants that confer selective marker resistance or viability to host cell lines. Such selectants include, but are not limited to, genimycin (G4118), neomycin, hygromycin B, puromycin, bleomycin, methionine sulfinimide, methotrexate, glutamate-free cell culture media, glycine-deficient cell culture media, hypoxanthine and thymidine, or thymidine alone.

[0070] "Production" cell culture medium or fed culture medium refers to cell culture typically used during the transition period at the end of exponential growth and during subsequent transition and / or production phases when protein production takes over. Such cell culture media are sufficiently intact to maintain the required cell density, viability, and / or product titers during this phase.

[0071] "Perfusion" cell culture medium, or fed culture medium, is a cell culture medium typically used to maintain cell culture by perfusion or continuous culture methods, and is sufficiently complete to support cell culture during this process. Perfusion cell culture medium formulations can be richer or more concentrated than basal cell culture medium formulations to suit methods for removing used medium. Perfusion cell culture medium can be used during both the growth and production phases.

[0072] The term "0.5 × volume" refers to approximately 50% of the volume. The term "0.6 × volume" refers to approximately 60% of the volume. Similarly, 0.7 ×, 0.8 ×, 0.9 ×, and 1.0 × represent approximately 70%, 80%, 90%, or 100% of the volume, respectively.

[0073] The term "culture" or "cell culture" refers to the maintenance or proliferation of mammalian cells under a set of controlled physical conditions.

[0074] The term "culture of mammalian cells" refers to a liquid culture medium containing multiple mammalian cells that is maintained or proliferated under a set of controlled physical conditions.

[0075] The term "liquid culture medium" refers to a fluid containing sufficient nutrients to allow cells (e.g., mammalian cells) to grow or proliferate in vitro. For example, a liquid culture medium may contain one or more of the following: amino acids (e.g., 20 amino acids), purines (e.g., hypoxanthine), pyrimidines (e.g., thymidine), choline, inositol, thiamine, folic acid, biotin, calcium, nicotinamide, pyridoxine, riboflavin, thymine, cyanocobalamin, pyruvate, lipoic acid, magnesium, glucose, sodium, potassium, iron, copper, zinc, and sodium bicarbonate. In some embodiments, the liquid culture medium may contain serum from mammals. In some embodiments, the liquid culture medium does not contain serum from mammals or another extract (defined liquid culture medium). In some embodiments, the liquid culture medium may contain trace amounts of metals, mammalian growth hormones, and / or mammalian growth factors. Another example of a liquid culture medium is a basic culture medium (e.g., a medium containing only inorganic salts, a carbon source, and water). Non-limiting examples of liquid culture media are described herein. Further examples of liquid culture media are known in the art and are commercially available. Liquid culture media may contain mammalian cells at any density. For example, as used herein, a certain volume of liquid culture medium removed from a bioreactor may be substantially free of mammalian cells.

[0076] In the context of this invention, "bioreactor" refers to a container suitable for perfusion cell culture, wherein at least steps (i) to (iii) of this invention are performed. A bioreactor may be a disposable container, for example, made of plastic material, or a reusable container, for example, made of stainless steel.

[0077] The term "stirring" refers to agitating or otherwise moving a portion of liquid culture medium in a bioreactor. This is done, for example, to increase the concentration of dissolved O2 in the liquid culture medium within the bioreactor. Stirring can be performed using any method known in the prior art, such as instruments or propellers. Exemplary apparatus and methods for stirring a portion of liquid culture medium in a bioreactor are known in the art.

[0078] The term "continuous method" refers to a method in which a fluid is continuously fed through at least a portion of a system. For example, in any of the exemplary continuous biomanufacturing systems described herein, a liquid culture medium containing a recombinant therapeutic protein is continuously fed into the system and a therapeutic protein active pharmaceutical ingredient is discharged from the system during system operation.

[0079] The term "feed-batch bioreactor" is a prior art term and refers to a bioreactor in which multiple cells (e.g., mammalian cells) are contained in a first liquid culture medium, wherein the culture of the cells present in the bioreactor involves periodically or continuously adding a second liquid culture medium to the first liquid culture medium without substantially or significantly removing either the first or second liquid culture medium from the cell culture. The second liquid culture medium may be the same as the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is a concentrated form of the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is added as a dry powder.

[0080] The term "trimming" refers to the partial cleavage of proteins, usually through proteolysis.

[0081] The term "degradation" typically refers to a larger entity, such as a peptide or protein, breaking down into at least two smaller entities, one of which may be significantly larger than the others.

[0082] The term "deacetylation" refers to any chemical reaction in which the acetylation functional group in the side chain of an amino acid (usually aspartic acid or glutamic acid) is removed or converted into another functional group. Typically, aspartic acid is converted into aspartic acid or isoaspartic acid.

[0083] The term "aggregation" generally refers to the direct attraction between molecules, such as through van der Waals forces or chemical bonds. In particular, aggregation is understood as the accumulation and condensation of proteins together. Aggregates can include amorphous aggregates, oligomers, and amyloid fibrils, and are often referred to as high molecular weight (HMW) molecules, i.e., molecules with a higher molecular weight than the pure product molecule, which is a non-aggregating molecule, and are here often referred to as low molecular weight (LMW) molecules or monomers.

[0084] In this paper, acidic species are generally understood to be included among the variants commonly observed when analyzing antibodies using charge-based separation techniques such as isoelectric focusing (IEF) gel electrophoresis, capillary isoelectric focusing (cIEF) gel electrophoresis, cation exchange chromatography (CEX), and anion exchange chromatography (AEX). These variants are referred to as acidic or basic species compared to the dominant species. When analyzing antibodies using IEF-based methods, acidic species are typically variants with lower apparent pI, while basic species are variants with higher apparent pI.

[0085] A preferred permeability probe according to the present invention is the Incyte probe, which applies an alternating electric field to a culture and measures the resulting polarization and depolarization of living cells and microorganisms by means of a permeability reading (permeability / area). This signal can be correlated with the density of living cells because only living cells can be polarized. Dead cells have leaky membranes and cannot be polarized. Therefore, this method is insensitive to dead cells, cell debris, and microcarriers.

[0086] Incyte is a type of permittivity-based sensor that responds only to living cells.

[0087] The term "residence time" generally refers to the time a particular product molecule exists in a bioreactor, that is, the time from its biotechnological production to its separation from the bioreactor chamber.

[0088] "Product quality" is typically assessed by the presence of trimming, degradation, deacetylation, and / or aggregation. For example, a product (molecule) containing less than 40%, preferably less than 35%, or even 30%, 25%, or 20% of HMW species can be considered to have better product quality. Furthermore, better product quality compared to products manufactured by methods different from those of this invention (e.g., fed-batch processing) is associated with the substantial absence of residual host cell proteins (HCPs) and the substantial absence of trimming, degradation, and deacetylation, or with a significant reduction in HCP concentration, trimming, degradation, and / or deacetylation. In the context of this invention, methods known in the art for assessing product quality include cation exchange-high performance chromatography (CEX-HPLC) for charge change analysis, trypsin peptide mapping for chemical modifications, host cell protein (HCP) ELISA, reducing capillary electrophoresis-sodium dodecyl sulfate (RCE-SDS), and size exclusion-high performance liquid chromatography (SE-HPLC).

[0089] The term "antibody product" refers to a "secreted protein" or "secreted recombinant protein," and specifically to a protein (e.g., a recombinant protein) that initially contains at least one secretion signaling sequence during translation in mammalian cells and is at least partially secreted into the extracellular space (e.g., liquid culture medium) by enzymatic cleavage of the secretion signaling sequence in mammalian cells. Those skilled in the art will understand that a "secreted" protein does not need to be completely dissociated from the cell to be considered a secreted protein.

[0090] The term bispecific antibody product encompasses bispecific antibodies, such as full-length IgG-based antibodies and their fragments, which are typically referred to herein as bispecific molecules.

[0091] The term "antibody construct," or alternatively, a bispecific T-cell binding molecule or bispecific molecule, refers to a molecule whose structure and / or function are based on the structure and / or function of an antibody (e.g., a full-length or intact immunoglobulin molecule, typically comprising two untruncerated heavy chains and two light chains) and / or extracted from the variable heavy chain (VH) and / or variable light chain (VL) domains of an antibody or fragment thereof. Thus, an antibody construct is capable of binding to its specific target or antigen. Furthermore, the domain that binds to its binding partner according to the invention is understood herein as the binding domain of an antibody construct according to the invention. Typically, the binding domain according to the invention contains minimum structural requirements for the antibody to allow target binding. This minimum requirement can be defined, for example, by the presence of at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region), preferably all six CDRs. Alternative methods for defining the minimum structural requirements of an antibody include defining the antibody epitope within the specific target structure, the protein domain of the target protein constituting the epitope region (epitope cluster), or by referencing a specific antibody that competes with the defined antibody epitope. The antibodies upon which the constructs of the present invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies.

[0092] The binding domain of the antibody construct or bispecific T-cell binding agent molecule according to the present invention may, for example, include the CDR group mentioned above. Preferably, those CDRs are contained within the framework of the antibody light chain variable region (VL) and the antibody heavy chain variable region (VH); however, they do not necessarily include both. For example, the Fd fragment has two VH regions and typically retains some antigen-binding function of the intact antigen-binding domain. Other examples of antibody fragments, antibody variants, or binding domains include (1) Fab fragments, a monovalent fragment having VL, VH, CL, and CH1 domains; (2) F(ab')2 fragments, a bivalent fragment having two Fab fragments connected in the hinge region by disulfide bridges; (3) Fd fragments having two VH and CH1 domains; (4) Fv fragments having a single arm of antibody with VL and VH domains; (5) dAb fragments having a VH domain (Ward et al., (1989) Nature [Nature] 341: 544-546); (6) separate complementarity-determining regions (CDRs); and (7) single-chain Fv (scFv), the latter being preferred (e.g., derived from scFV libraries). Examples of antibody constructs or bispecific molecules according to the present invention are described, for example, in the following: WO 00 / 006605, WO 2005 / 040220, WO 2008 / 119567, WO 2010 / 037838, WO 2013 / 026837, WO 2013 / 026833, US 2014 / 0308285, US 2014 / 0302037, WO 2014 / 144722, WO 2014 / 151910 and WO 2015 / 048272.

[0093] Additionally, within the definition of "binding domain" or "binding structure domain," fragments of full-length antibodies, such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab')2, or "r IgG" ("half antibody"). Antibody constructs or bispecific T-cell conjugate molecules according to the present invention may also contain modified fragments of the antibody (also called antibody variants), such as scFv, di-scFv or di(s)-scFv, scFv-Fc, scFv-zipper, scFab, Fab2, Fab3, biantibodies, single-chain biantibodies, tandem biantibodies (Tandabs), tandem di-scFv, tandem tri-scFv, "multi-antibodies" (such as tri- or tetra-antibodies), and single-domain antibodies such as nanoantibodies or monovariable domain antibodies containing only one variable domain, which may be VHH, VH, or VL that specifically binds to antigens or epitopes independently of other V regions or domains.

[0094] As used herein, the terms "single-chain Fv," "single-chain antibody," or "scFv" refer to a single polypeptide chain antibody fragment containing variable regions from both the heavy and light chains, but lacking constant regions. Generally, single-chain antibodies further include a polypeptide linker between the VH and VL domains, which allows it to form the desired structure that will allow antigen binding. Single-chain antibodies are discussed in detail in: Pluckthun, *The Pharmacology of Monoclonal Antibodies*, Vol. 113, edited by Rosenburg and Moore, Springer-Verlag, New York, pp. 269–315 (1994). Various methods for generating single-chain antibodies are known, including those described in: U.S. Patent Nos. 4,694,778 and 5,260,203; International Patent Application Publication No. WO 88 / 01649; Bird (1988) Science 242: 423-442; Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85: 5879-5883; Ward et al. (1989) Nature 334: 54454; Skerra et al. (1988) Science 242: 1038-1041. In specific embodiments, single-chain antibodies may also be bispecific, multispecific, human and / or humanized, and / or synthetic.

[0095] Furthermore, the definition of the term "antibody construct" or bispecific T-cell conjugate molecule includes monovalent, divalent, and polyvalent / multivalent constructs, and therefore includes bispecific constructs that specifically bind to only two antigen structures, and multispecific constructs that specifically bind to more than two (e.g., three, four, or more) antigen structures by means of different binding domains. Additionally, the definition of the term "antibody construct" or bispecific T-cell conjugate molecule includes molecules consisting of only one polypeptide chain and molecules consisting of more than one polypeptide chain, which may be identical (homodimer, homotrimer, or homooligomer) or different (heterodimer, heterotrimer, or heterooligomer). Examples of the antibodies and their variants or derivatives identified above are particularly described in the following: Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999); Kontermann and Dübel, Antibody Engineering, Springer, 2nd edition 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

[0096] As used herein, the term "bispecific" refers to an antibody construct that is "at least bispecific," meaning it contains at least a first binding domain and a second binding domain, wherein the first binding domain binds to one antigen or target (e.g., a target cell surface antigen), and the second binding domain binds to another antigen or target (e.g., CD3). Therefore, the antibody construct according to the invention comprises specificity against at least two different antigens or targets. Furthermore, the specificity of the bispecific T-cell conjugate molecule according to the invention is characterized by targeting two different cells simultaneously, namely effector T cells and target cells, and aggregating them together to achieve a therapeutic effect. Therefore, in the context of the invention, the term "bi-" specifically represents dual-cell targeting, i.e., targeting and aggregating two different cells together. For example, the first domain preferably does not bind to the extracellular epitopes of one or more of the species described herein. The term "target cell surface antigen" refers to an antigenic structure expressed by a cell and present on the cell surface such that it is accessible to the antibody construct as described herein. It may be a protein, preferably the extracellular portion of a protein; or a carbohydrate structure, preferably the carbohydrate structure of a protein, such as a glycoprotein. It is preferably a tumor antigen. The term "bispecific antibody construct" in this invention also covers multispecific antibody constructs, such as trispecific antibody constructs, which include three binding domains, or constructs having more than three (e.g., four, five...) specificities.

[0097] Given that the antibody constructs or bispecific molecular systems according to the present invention are (at least) bispecific, they are not naturally occurring and are distinctly different from naturally occurring products. Therefore, the "bispecific" antibody constructs or immunoglobulin systems have at least two artificial hybrid antibodies or immunoglobulins with different binding sides having different specificities. Bispecific antibody constructs or bispecific molecules can be produced by a variety of methods, including hybridoma fusion or Fab' fragment linkage. See, for example, Songsivilai and Lachmann, Clin. Exp. Immunol. [Clinical Experimental Immunology] 79: 315-321 (1990).

[0098] The antibody constructs or bispecific molecules of the present invention may or may not contain peptide linkers (spacer peptides) at least two binding domains and variable domains (VH / VL). According to the present invention, the term "peptide linker" comprises an amino acid sequence by which the amino acid sequences of one (variable and / or binding) domain and another (variable and / or binding) domain of the antibody constructs or bispecific molecules of the present invention are linked to each other. Peptide linkers can also be used to fuse a third domain with other domains of the antibody constructs of the present invention. A key technical feature of such peptide linkers is that they do not contain any polymerization activity. Suitable peptide linkers are those described in U.S. Patents 4,751,180 and 4,935,233 or WO 88 / 09344. Peptide linkers can also be used to link other domains or modules or regions (such as half-life extension domains) to the antibody constructs of the present invention.

[0099] The antibody construct of the present invention is preferably an "in vitro generated antibody construct." This term refers to an antibody construct as defined above, wherein all or part of the variable region (e.g., at least one CDR) is generated in a non-immune cell selection process, such as in vitro phage display, protein wafers, or any other method in which the ability of a candidate sequence to bind to an antigen can be tested. Therefore, this term preferably excludes sequences generated solely by genomic rearrangements in immune cells of an animal. "Recombinant antibody" refers to an antibody produced using recombinant DNA technology or genetic engineering.

[0100] As used herein, the term "monoclonal antibody" (mAb) or monoclonal antibody construction system refers to an antibody obtained from a substantially homogeneous population of antibodies, meaning that, apart from possibly naturally occurring mutations and / or post-translational modifications (e.g., isomerization, amideation) that may be present in small amounts, the individual antibody systems constituting that population are identical. Monoclonal antibodies exhibit high specificity against a single antigenic side or determinant on an antigen, compared to conventional (multiclonal) antibody formulations that typically consist of different antibodies targeting different determinants (or epitopes). In addition to their specificity, monoclonal antibodies also have the advantage of being synthesized from hybridoma cultures and therefore not contaminated by other immunoglobulins. The modifier "monoclonal" indicates the characteristic of antibodies obtained from a substantially homogeneous population of antibodies and should not be construed as requiring the antibody to be produced by any particular method.

[0101] For the preparation of monoclonal antibodies, any technique that provides antibodies produced from continuous cell line cultures can be used. For example, the monoclonal antibody to be used can be prepared by the hybridoma method first described by Koehler et al., Nature, 256: 495 (1975), or by a recombinant DNA method (see, for example, U.S. Patent No. 4,816,567). Examples of other techniques for producing human monoclonal antibodies include the three-source hybridoma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss (1985), 77-96).

[0102] Hybridomas can then be screened using standard methods such as enzyme-linked immunosorbent assay (ELISA) and surface electroplasmic resonance (BIACORE™) analysis to identify one or more hybridomas that produce antibodies that specifically bind to a specified antigen. Any form of the relevant antigen can be used as an immunogen, such as recombinant antigens, naturally occurring forms, any variants or fragments thereof, and their antigenic peptides. Surface electroplasmic resonance, as used in the BIAcore system, can be used to increase the efficiency of phage antibodies binding to epitopes of target cell surface antigens (Schier, Human Antibodies Hybridomas, 7 (1996), 97-105; Malmborg, J. Immunol. Methods, 183 (1995), 7-13).

[0103] Another exemplary method for preparing monoclonal antibodies includes screening protein expression libraries, such as phage display or ribosome display libraries. Phage display is described, for example, in: Ladner et al., U.S. Patent No. 5,223,409; Smith (1985) Science 228: 1315-1317; Clackson et al., Nature 352: 624-628 (1991); and Marks et al., J. Mol. Biol. 222: 581-597 (1991).

[0104] In addition to using display libraries, relevant antigens can be used to immunize non-human animals, such as rodents (e.g., mice, hamsters, rabbits, or rats). In one embodiment, the non-human animal includes at least a portion of the human immunoglobulin gene. For example, it is possible to engineer mouse strains with antibody production defects using large fragments of human Ig (immunoglobulin) gene loci. Using hybridoma technology, antigen-specific monoclonal antibodies derived from genes having the desired specificity can be generated and selected. See, for example, XENOMOUSE™, Green et al. (1994) Nature Genetics 7: 13-21, US 2003-0070185, WO 96 / 34096, and WO 96 / 33735.

[0105] Monoclonal antibodies can also be obtained from non-human animals and then modified using recombinant DNA techniques known in the art, such as humanization, deimmunization, and chimerism. Examples of modified antibody constructs include humanized variants of non-human antibodies, "affinity-matured" antibodies (see, for example, Hawkins et al. J. Mol. Biol. [Journal of Molecular Biology] 254, 889-896 (1992) and Lowman et al., Biochemistry [Biochemistry] 30, 10832-10837 (1991)) and antibody mutants with altered functions of one or more effectors (see, for example, U.S. Patent 5,648,260, Kontermann and Dübel (2010), cited above, and Little (2009), cited above).

[0106] In immunology, affinity maturation is a process by which B cells produce antibodies with increased affinity for antigens during an immune response. Due to repeated exposure to the same antigen, the host produces antibodies with progressively greater affinity. Similar to the natural prototype, in vitro affinity maturation is based on the principles of mutation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random mutations are introduced into the CDR using radiation, chemical mutagens, or error-prone PCR. Furthermore, genetic diversity can be increased through strand shuffling. Two or three rounds of mutation and selection using display methods (such as phage display) typically produce antibody fragments with affinity in the low nanomolar range.

[0107] Preferred types of amino acid substitution changes in antibody constructs involve substituting one or more hypervariable residues of the parent antibody (e.g., humanized or human antibody). Generally, one or more resulting variants selected for further development will have improved biological properties relative to the parent antibody that produced them. A convenient method for generating such substituted variants involves phage-displayed affinity maturation. Briefly, several hypervariable sides (e.g., 6-7 sides) are mutated to produce all possible amino acid substitutions on each side. The resulting antibody variants are displayed monovalently from filamentous phage particles as fusions with the M13 gene III product packaged within each particle. The biological activity (e.g., binding affinity) of the phage-displayed variants is then screened as disclosed herein. To identify candidate hypervariable sides for modification, alanine scanning mutagenesis can be performed to identify hypervariable residues that significantly contribute to antigen binding. Alternatively or additionally, analyzing the crystal structure of the antigen-antibody complex to identify contact points between the binding domain and, for example, human target cell surface antigens may be advantageous. According to the techniques described herein, these contact residues and adjacent residues are candidates for substitution. Once such variants are generated, they are screened as described herein, and antibodies exhibiting superior properties in one or more relevant assays can be selected for further development.

[0108] The monoclonal antibodies and antibody constructs of this invention specifically include "chimeric" antibodies (immunoglobulins) wherein a portion of the heavy chain and / or light chain is identical or homologous to a corresponding sequence in an antibody derived from a specific species or belonging to a specific antibody class or subclass, while the remainder of one or more chains is identical or homologous to a corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass, and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences], 81: 6851-6855 (1984)). Chimeric antibodies of interest herein include "primate-derived" antibodies, which contain a variable domain antigen-binding sequence derived from non-human primates (e.g., Old World monkeys, apes, etc.) and a human constant region sequence. Various methods for preparing chimeric antibodies have been described. See, for example, Morrison et al., Proc. Natl. Acad. Sci USA [Proceedings of the National Academy of Sciences of the United States of America] 81: 6851, 1985; Takeda et al., Nature [Nature] 314: 452, 1985; Cabilly et al., U.S. Patent No. 4,816,567; Boss et al., U.S. Patent No. 4,816,397; Tanaguchi et al., EP 0171496; EP 0173494; and GB 2177096.

[0109] Antibodies, antibody constructs, antibody fragments, or antibody variants can also be modified by specifically deleting human T-cell epitopes (a process known as "deimmunization") using methods disclosed, for example, in WO 98 / 52976 or WO 00 / 34317. In short, the heavy and light chain variable domains of antibodies can be analyzed against peptides that bind to class II MHC; these peptides represent potential T-cell epitopes (as defined in WO 98 / 52976 and WO 00 / 34317). To detect potential T-cell epitopes, a computer modeling method known as "peptide threading" can be applied, and in addition, databases of human MHC class II binding peptides, as described in WO 98 / 52976 and WO 00 / 34317, can be searched for motifs present in the VH and VL sequences. These motifs bind to any of the 18 major MHC class II DR allotypes and thus constitute potential T-cell epitopes. Detected potential T-cell epitopes can be eliminated by substituting a small number of amino acid residues in the variable domain, or preferably by substituting a single amino acid. Typically, conserved substitutions are performed. Usually, but not exclusively, amino acids common to the position in human immunoglobulin sequences can be used. Human sequences are disclosed, for example, in: Tomlinson et al. (1992) J. MoI. Biol. [Journal of Molecular Biology] 227: 776-798; Cook, GP et al. (1995) Immunol. Today [Contemporary Immunology] Vol. 16 (5): 237-242; and Tomlinson et al. (1995) EMBO J. [Journal of the European Society for Molecular Biology] 14: 14: 4628-4638. The V BASE catalog provides a comprehensive catalog of human immunoglobulin variable region sequences (edited by Tomlinson, LA. et al. MRC Centre for Protein Engineering, Cambridge, UK). These sequences can be used as a source of human sequences, for example, for frame regions and CDRs. Common human frame regions can also be used, such as those described in U.S. Patent No. 6,300,064.

[0110] "Humanized" antibodies, antibody constructs, variants, or fragments thereof (such as Fv, Fab, Fab', F(ab')2, or other antigen-binding sequences of antibodies) are antibodies or immunoglobulins with a major human sequence containing one or more minimal sequences derived from non-human immunoglobulins. For most, humanized antibodies are human immunoglobulins (receptor antibodies) in which residues from the hypervariable region (also known as the CDR) of the receptor are replaced by residues from a hypervariable region of a non-human (e.g., rodent) species (donor antibody) (such as mouse, rat, hamster, or rabbit) with the desired specificity, affinity, and capability. In some cases, Fv framework region (FR) residues of human immunoglobulins are replaced by corresponding non-human residues. Furthermore, as used herein, "humanized antibodies" may also include residues not found in either the receptor antibody or the donor antibody. These modifications are made to further improve and optimize antibody performance. Humanized antibodies may also contain at least a portion of the immunoglobulin constant region (Fc) of human immunoglobulins. For more details, see Jones et al., Nature, 321: 522-525 (1986); Reichmann et al., Nature, 332: 323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2: 593-596 (1992).

[0111] Humanized antibodies or fragments thereof can be generated by replacing sequences of Fv variable domains that do not directly participate in antigen binding with equivalent sequences of human Fv variable domains. Exemplary methods for generating humanized antibodies or fragments thereof are provided by: Morrison (1985) Science 229: 1202-1207; Oi et al. (1986) BioTechniques 4: 214; and US 5,585,089; US 5,693,761; US ​​5,693,762; US 5,859,205; and US 6,407,213. Those methods involve isolating, manipulating, and expressing nucleic acid sequences encoding all or part of the immunoglobulin Fv variable domains from at least one of the heavy or light chains. Such nucleic acids can be obtained from hybridomas that generate antibodies against predetermined targets as described above, as well as other sources. The recombinant DNA encoding humanized antibody molecules can then be colonized into a suitable expression vector.

[0112] Humanized antibodies can also be produced using transgenic animals, such as mice that express human heavy and light chain genes but not endogenous mouse immunoglobulin heavy and light chain genes. Winter describes an exemplary CDR transplantation method (US Patent No. 5,225,539) that can be used to prepare the humanized antibodies described herein. All CDRs of a particular human antibody can be replaced with at least a portion of non-human CDRs, or only some CDRs can be replaced with non-human CDRs. Only the desired number of CDRs needed to bind the humanized antibody to the predetermined antigen needs to be replaced.

[0113] Humanized antibodies can be optimized by introducing conserved substitutions, shared sequence substitutions, germline substitutions, and / or reversion mutations. Such modified immunoglobulin molecules can be prepared using any of several techniques known in the art (e.g., Teng et al., Proc. Natl. Acad. Sci. USA, 80: 7308-7312, 1983; Kozbor et al., Immunology Today, 4: 7279, 1983; Olsson et al., Meth. Enzymol., 92: 3-16, 1982, and EP 239 400).

[0114] The terms "human antibody," "human antibody construct," and "human binding domain" include antibodies, antibody constructs, and binding domains having antibody regions that substantially correspond to variable and constant regions or domains of human immunoglobulin sequences known in the art, including, for example, those described by Kabat et al. (1991) (cited above). The human antibodies, antibody constructs, or binding domains of the present invention may, for example, contain amino acid residues not encoded by human immunoglobulin sequences in, for example, CDR, and particularly CDR3 (e.g., mutations introduced by random or flanking specific mutagenesis in vitro or by somatic mutations in vivo). The human antibodies, antibody constructs, or binding domains may have at least one, two, three, four, five, or more sites where amino acid residues not encoded by human immunoglobulin sequences are substituted. However, the definitions of human antibody, antibody construct, and binding domain used herein also include "fully human antibodies," which contain only human antibody sequences that are not artificially and / or genetically modified, such as those derived using technologies or systems like Xenomouse. Preferably, "fully human antibodies" do not contain amino acid residues that are not encoded by human germline immunoglobulin sequences.

[0115] In some embodiments, the antibody construction system of the present invention is a "separated" or "substantially pure" antibody construct. When used to describe the antibody constructs disclosed herein, "separated" or "substantially pure" means that the antibody construct has been identified, isolated, and / or recovered from the components of its production environment. Preferably, the antibody construct does not associate with, or substantially does not associate with, any other components from its production environment. Contaminating components of its production environment, such as those produced by recombinant transfected cells, are typically substances that interfere with the diagnostic or therapeutic use of the peptide and may include enzymes, hormones, and other proteins or non-protein solutes. The antibody construct may, for example, comprise at least about 5% by weight or at least about 50% by weight of the total protein in a given sample. It should be understood that, depending on the circumstances, the isolated protein may comprise from 5% to 99.9% by weight of the total protein content. By using inducible promoters or high-performance promoters, peptides can be prepared at significantly higher concentrations, allowing for increased concentration levels. This definition includes the production of antibody constructs in a variety of organisms and / or host cells known in the art. In a preferred embodiment, the antibody construct (1) is purified to the extent necessary to obtain at least 15 N-terminal or internal amino acid residues using a spin-cup sequencer, or (2) can be purified to homogenization by SDS-PAGE under non-reducing or reducing conditions using Coomassie blue or, more preferably, silver staining. However, isolated antibody constructs are typically prepared by at least one purification step.

[0116] The term "binding domain" in this invention characterizes a domain that (specifically) binds to / interacts with / recognizes a given target epitope or a given target side terminal on a target molecule (antigen) (e.g., CD33 and CD3, respectively). The structure and function of the first binding domain (recognizing, for example, CD33), and preferably a second binding domain (e.g., recognizing CD3), are based on the structure and / or function of an antibody, such as a full-length or intact immunoglobulin molecule, and / or extracted from the variable heavy chain (VH) and / or variable light chain (VL) domains of the antibody or a fragment thereof. Preferably, the first binding domain is characterized by the presence of three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). The second binding domain preferably also includes the minimum structural requirements of the antibody that allow target binding. More preferably, the second binding domain comprises at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). It is envisioned that the first and / or second binding domains are generated or available through phage display or library screening methods, rather than by transplanting CDR sequences from pre-existing (monoclonal) antibodies into a scaffold.

[0117] According to the present invention, the binding domain is in the form of one or more polypeptides. Such polypeptides may include protein and non-protein portions (e.g., chemical linkers or chemical cross-linking agents, such as glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments, and peptides typically having fewer than 30 amino acids) comprise two or more amino acids coupled to each other via covalent peptide bonds (forming an amino acid chain).

[0118] As used herein, the term "peptide" describes a group of molecules that typically consist of more than 30 amino acids. Peptides can further form polymers, such as dimers, trimers, and higher-order oligomers, i.e., composed of more than one polypeptide molecule. The polypeptide molecules forming such dimers, trimers, etc., can be identical or different. Therefore, the corresponding higher-order structures of such polymers are referred to as homodimers or heterodimers, homotrimers or heterotrimers, etc. An example of a heteropolymer is an antibody molecule, which in its naturally occurring form consists of two identical light polypeptide chains and two identical heavy polypeptide chains. The terms "peptide," "polypeptide," and "protein" also refer to naturally modified peptides / polypeptides / proteins, where the modification is achieved, for example, by post-translational modifications (such as glycosylation, acetylation, phosphorylation, etc.). When mentioned herein, "peptide," "polypeptide," or "protein" can also be chemically modified, such as polyethylene glycol-modified. Such modifications are well known in the art and are described below.

[0119] Preferably, the binding domains for target cell surface antigens and / or the binding domains for CD3ε are human binding domains. Antibodies and antibody constructs containing at least one human binding domain avoid some of the problems associated with antibodies or antibody constructs having non-human (such as rodents, e.g., mice, rats, hamsters, or rabbits) variable and / or constant regions. The presence of such rodent-derived proteins can lead to rapid clearance of the antibody or antibody construct or can induce an immune response against the antibody or antibody construct in the patient. To avoid using rodent-derived antibodies or antibody constructs, human or fully human antibodies / antibody constructs can be generated by introducing human antibody function into rodents to produce fully human antibodies in rodents.

[0120] The ability to select and recombinant megabase-sized human loci in YAC and introduce them into mouse lines provides a powerful method for elucidating the functional components of very large or coarsely located loci and for generating useful human disease models. Furthermore, replacing mouse loci with their human equivalents using this technique can provide unique insights into the performance and regulation of human gene products during development, their communication with other systems, and their involvement in disease induction and progression.

[0121] A key practical application of this strategy is the "humanization" of the mouse humoral immune system. Introducing human immunoglobulin (Ig) loci into mice where endogenous Ig genes are inactivated provides an opportunity to study the fundamental mechanisms of antibody programmed expression and assembly, as well as their role in B cell development. Furthermore, this strategy can provide an ideal source for the production of fully human monoclonal antibodies (mAbs)—a significant milestone contributing to the prospects of antibody therapy in human diseases. Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic responses inherent in mouse or mouse-derived mAbs, thereby increasing the efficacy and safety of administered antibody / antibody constructs. The use of fully human antibodies or antibody constructs is anticipated to offer significant advantages in treating chronic and relapsing human diseases requiring repeated compound administration, such as inflammation, autoimmune diseases, and cancer.

[0122] One approach to achieving this goal is to engineer mouse strains with antibody-deficient production using large fragments of human Ig loci. These mice are expected to generate a large repertoire of human antibodies in the absence of mouse antibodies. Large human Ig fragments will maintain significant variable genetic diversity and appropriate regulation of antibody production and expression. By utilizing mouse mechanisms for antibody diversification and selection, and by exploiting the lack of immune tolerance to human proteins, the regenerated human antibody repertoire in these mouse strains should produce high-affinity antibodies against any antigen of interest, including human antigens. Using hybridoma technology, antigen-specific human mAbs with desired specificity can be readily generated and selected. The generation of the first XenoMouse mouse strain demonstrates this general strategy (see Green et al., Nature Genetics, 7: 13-21 (1994)). The XenoMouse strain was engineered using yeast artificial chromosomes (YACs) containing germline fragments of 245 kb and 190 kb sizes from the human heavy chain locus and the κ light chain locus, respectively, containing core variable and constant region sequences. YACs containing human Ig were demonstrated to be compatible with the mouse system for rearranging and expressing antibodies, and these YACs were able to replace inactivated mouse Ig genes. This was demonstrated by their ability to induce B cell development, produce an adult-like human repertoire of fully human antibodies, and generate antigen-specific human mAbs. These results also showed that introducing human Ig loci containing a greater number of V genes, additional regulatory elements, and a larger portion of the human Ig constant region could substantially reproduce a complete repertoire characteristic of the humoral response to infection and immunity. The work of Green et al. has recently been extended to introducing germline YAC fragments of megabase sizes from the human heavy chain locus and the κ light chain locus, respectively, to introduce a human antibody repertoire greater than approximately 80%. See Mendez et al., Nature Genetics, 15: 146-156 (1997) and U.S. Patent Application No. 08 / 759,620.

[0123] The origin of the XenoMouse mouse is further discussed and described in the following: U.S. Patent Application Serial Nos. 07 / 466,008, 07 / 610,515, 07 / 919,297, 07 / 922,649, 08 / 031,801, 08 / 112,848, 08 / 234,145, 08 / 376,279, 08 / 430,938, 08 / 464,584, and 08 / 464. 582, Serial No. 08 / 463,191, Serial No. 08 / 462,837, Serial No. 08 / 486,853, Serial No. 08 / 486,857, Serial No. 08 / 486,859, Serial No. 08 / 462,513, Serial No. 08 / 724,752 and Serial No. 08 / 759,620; and U.S. Patent Nos. 6,162,963; 6,150,584; 6,114,598; 6,075,181 and 5,939,598; and Japanese Patent Nos. 3,068 180 B2, 3,068 506 B2 and 3,068 507 B2. See also Mendez et al., Nature Genetics 15: 146-156 (1997) and Green and Jakobovits, J. Exp. Med. 188: 483-495 (1998), EP 0 463 151 B1, WO 94 / 02602, WO 96 / 34096, WO 98 / 24893, WO 00 / 76310 and WO 03 / 47336.

[0124] In an alternative approach, other companies, including GenPharm International, Inc., utilize the "microlobe" approach. In the microlobe approach, exogenous Ig loci are mimicked by incorporating fragments (individual genes) from Ig loci. Thus, one or more VH genes, one or more DH genes, one or more JH genes, mu constant regions, and second constant regions (preferably γ constant regions) are formed into constructs for insertion into animals. The method is described in the following: U.S. Patent Nos. 5,545,807 and 5,545,806; 5,625,825; 5,625,126; 5,633,425; 5,661,016; 5,770,429; 5,789,650; 5,814,318; 5,877,397; 5,874,299; and 6,255,458 (respectively Lonberg and Kay), U.S. Patent Nos. 5,591,669 and 6,023,010, and U.S. Patent No. 5,612, by Surani et al. 205;5,721,367; and 5,789,215, and U.S. Patent Nos. 5,643,763 to Choi and Dunn, and U.S. International Patent Application Nos. 07 / 574,748, 07 / 575,962, 07 / 810,279, 07 / 853,408, 07 / 904,068, 07 / 990,860, 08 / 053,131, 08 / 096,762, 08 / 155,301, 08 / 161,739, 08 / 165,699, and 08 / 209,741 to GenPharm. See also EP 0 546 073 B1, WO 92 / 03918, WO 92 / 22645, WO 92 / 22647, WO 92 / 22670, WO 93 / 12227, WO 94 / 00569, WO 94 / 25585, WO 96 / 14436, WO 97 / 13852, and WO 98 / 24884, and U.S. Patent No. 5,981,175. Further see Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994), Tuaillon et al. (1995), and Fishwild et al. (1996).

[0125] Kirin also demonstrated the generation of human antibodies from mice in which large segments or entire chromosomes were introduced via microcell fusion. See European patent applications 773 288 and 843 961. Xenerex Biosciences is developing a technology for the potential generation of human antibodies. In this technology, SCID mice are reconstructed with human lymphocytes (e.g., B and / or T cells). The mice are then immunized with an antigen and an immune response against the antigen is generated. See U.S. Patent Nos. 5,476,996; 5,698,767; and 5,958,765.

[0126] Human anti-mouse antibody (HAMA) responses have led the industry to produce chimeric or other humanized antibodies. However, certain human anti-chimeric antibody (HACA) responses are expected, particularly with long-term or multiple-dose use of antibodies. Therefore, it is desirable to provide antibody constructs containing human binding domains against target cell surface antigens and human binding domains against CD3ε to eliminate the problems and / or effects of HAMA or HACA responses.

[0127] The terms “(specific) binding”, “(specific) recognition”, “(specific) targeting” and “(specific) reaction” refer to, according to the present invention, the binding domain interacts or specifically interacts with a given epitope or a given target side end on a target molecule (antigen) (here: target cell surface antigen and CD3ε, respectively).

[0128] The term "epitope" refers to a side of an antigen on which a binding domain (such as an antibody or immunoglobulin, or a derivative, fragment, or variant of an antibody or immunoglobulin) specifically binds. "Epitope" is antigenic, and therefore the term is sometimes also referred to herein as "antigen structure" or "antigen determinant." Thus, the binding domain is the "antigen interaction side." The binding / interaction is also understood to define "specific recognition."

[0129] An epitope can be formed by consecutive amino acids or by discontinuous amino acids juxtaposed in the ternary folding of a protein. A linear epitope is an epitope in which the primary sequence of amino acids constitutes the identified epitope. A linear epitope typically contains at least 3 or at least 4, and more commonly at least 5 or at least 6 or at least 7, such as about 8 to about 10 amino acids, in a unique sequence.

[0130] In contrast to linear epitopes, "conformal epitopes" are epitopes in which the primary sequence of the amino acids constituting the epitope is not the sole defining component of the recognized epitope (e.g., an epitope in which the primary sequence of the amino acids is not necessarily recognized by the binding domain). Typically, conformational epitopes contain an increased number of amino acids compared to linear epitopes. Regarding the recognition of conformational epitopes, the binding domain recognizes the antigen, preferably the three-dimensional structure of a peptide or protein or fragment thereof (in the context of this invention, an antigen structure with a binding domain is included within an antigen protein on the target cell surface). For example, when a protein molecule folds to form a three-dimensional structure, certain amino acids and / or polypeptide backbones forming the conformational epitope are juxtaposed, enabling the antibody to recognize the epitope. Methods for determining epitope conformation include, but are not limited to, X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, and site-directed spin labeling and electron paramagnetic resonance (EPR) spectroscopy.

[0131] The following describes a method for epitope localization: When a region (continuous amino acid segment) in a human target cell surface antigen protein is exchanged / replaced with a corresponding region of a non-human, non-primate target cell surface antigen (e.g., mouse target cell surface antigen, but others such as chicken, rat, hamster, rabbit, etc. are also possible), a decrease in binding of the binding domain is expected, unless the binding domain is cross-reactive with the non-human, non-primate target cell surface antigen used. The decrease is preferably at least 10%, 20%, 30%, 40%, or 50% compared to binding to the corresponding region in the human target cell surface antigen protein; more preferably at least 60%, 70%, or 80%; and most preferably 90%, 95%, or even 100%, thereby setting the binding to the corresponding region in the human target cell surface antigen protein to 100%. The expression of the above-described human target cell surface antigen / non-human target cell surface antigen chimera in CHO cells is envisioned. It is also envisioned that human target cell surface antigen / non-human target cell surface antigen chimeras fuse with the transmembrane domains and / or cytoplasmic domains of different membrane-bound proteins (such as EpCAM).

[0132] In alternative or additional methods for epitope localization, several truncated forms of extracellular domains of human target cell surface antigens can be generated to identify specific regions recognized by binding domains. In these truncated forms, different extracellular target cell surface antigen domains / subdomains or regions are progressively deleted starting from the N-terminus. It is envisioned that the truncated target cell surface antigen forms can be expressed in CHO cells. It is also envisioned that the truncated target cell surface antigen forms can fuse with transmembrane domains and / or cytoplasmic domains of different membrane-binding proteins (such as EpCAM). It is also envisioned that the truncated target cell surface antigen forms may encompass a message peptide domain at their N-terminus, such as a message peptide derived from a mouse IgG heavy chain message peptide. Further envisioning that the truncated target cell surface antigen forms may encompass a v5 domain at their N-terminus (after the message peptide), which would allow verification of their correct representation on the cell surface. It is anticipated that those truncated target cell surface antigen forms that no longer encompass the target cell surface antigen region recognized by the binding domain will experience reduced or lost binding. The binding reduction is preferably at least 10%, 20%, 30%, 40%, or 50%; more preferably at least 60%, 70%, or 80%, and most preferably 90%, 95%, or even 100%, thereby setting the binding to the entire human target cell surface antigen protein (or its extracellular region or domain) as 100%.

[0133] Another method for determining the contribution of specific residues of target cell surface antigens to the recognition of antibody constructs or binding domains is alanine scanning (see, for example, Morrison KL and Weiss GA. Cur Opin Chem Biol. [New Insights in Chemical Biology] 2001 June; 5 (3): 302-7), in which each residue to be analyzed is replaced by alanine, for example, via site-directed mutagenesis. Alanine is used because it has a non-giant, chemically inert methyl functional group, but still mimics the secondary structure references of many other amino acids. In cases where the size of conserved mutant residues is required, giant amino acids (such as valine or leucine) can sometimes be used. Alanine scanning is a well-established technique that has been used for a long time.

[0134] The interaction between the binding domain and the epitope or epitope-containing region means that the binding domain exhibits considerable affinity for the epitope / epitope-containing region on a specific protein or antigen (here: target cell surface antigen and CD3, respectively), and generally does not show significant reactivity with proteins or antigens other than target cell surface antigen or CD3. "Considerable affinity" includes binding with an affinity of about 10⁻⁶ M (KD) or stronger. Preferably, binding is considered specific when the binding affinity is about 10⁻¹² to 10⁻⁸ M, 10⁻¹² to 10⁻⁹ M, 10⁻¹² to 10⁻¹⁰ M, 10⁻¹¹ to 10⁻⁸ M, and more preferably about 10⁻¹¹ to 10⁻⁹ M. Whether the binding domain specifically reacts with or binds to the target can be readily tested, in particular, by comparing the reaction of the binding domain with the target protein or antigen with the reaction of the binding domain with proteins or antigens other than target cell surface antigen or CD3. Preferably, the binding domains of the present invention substantially or substantially do not bind to proteins or antigens other than target cell surface antigens or CD3 (i.e., the first binding domain preferably does not bind to proteins other than target cell surface antigens, and the second binding domain does not bind to proteins other than CD3). It is envisioned that the antibody constructs according to the present invention are characterized by superior affinity characteristics compared to other HLE forms. Therefore, this superior affinity indicates a prolonged half-life in vivo. The longer half-life of the antibody constructs according to the present invention can reduce the duration and frequency of administration that typically contributes to improving patient compliance. This is particularly important because the antibody constructs of the present invention are particularly beneficial for highly debilitated or even multiple cancer patients.

[0135] The terms "substantially / truly non-binding" or "cannot bind" mean that the binding domain of the present invention does not bind to proteins or antigens other than target cell surface antigens or CD3, that is, it does not show more than 30% reactivity with proteins or antigens other than target cell surface antigens or CD3, preferably no more than 20%, more preferably no more than 10%, and particularly preferably no more than 9%, 8%, 7%, 6%, or 5%, thereby setting the binding to target cell surface antigens or CD3 as 100%, respectively.

[0136] It is believed that specific binding is achieved through a specific motif in the amino acid sequence of the binding domain and the antigen. Therefore, binding is achieved as a result of its primary, secondary, and / or tertiary structures, as well as secondary modifications of said structures. The specific interaction between the antigen-interacting side and its specific antigen can lead to simple binding between said side and the antigen. Furthermore, the specific interaction between the antigen-interacting side and its specific antigen can alternatively or additionally lead to signal initiation, for example, due to inducing changes in antigen conformation, antigen oligomerization, etc.

[0137] The term "variable" refers to the portion of an antibody or immunoglobulin domain that exhibits sequence variability and participates in determining the specificity and binding affinity of a particular antibody (i.e., "one or more variable domains"). The pairing of variable heavy chains (VH) and variable light chains (VL) together forms a single antigen-binding side end.

[0138] Variability is not uniformly distributed throughout the variable domains of an antibody; it is concentrated in subdomains of each of the heavy and light chain variable regions. These subdomains are called "hypervariable regions" or "complementarity-determining regions" (CDRs). The more conserved (i.e., non-hypervariable) portions of the variable domains are called "framework" regions (FRMs or FRs) and provide a scaffold for the six CDRs in three-dimensional space to form an antigen-binding surface. The naturally occurring variable domains of the heavy and light chains each contain four FRMs (FR1, FR2, FR3, and FR4), which are predominantly β-sheet configurations connected by three hypervariable regions that form loops connecting the β-sheet structure and, in some cases, form part of the β-sheet structure. The hypervariable regions in each chain are closely clustered together by the FRMs and, together with hypervariable regions from the other chain, contribute to the formation of antigen-binding side ends (see Kabat et al., cited above).

[0139] The term "CDR" and its plural "CDR" refer to the complementary determinant regions (CDRs) of three binding features constituting the light chain variable regions (CDR-L1, CDR-L2, and CDR-L3) and three binding features constituting the heavy chain variable regions (CDR-H1, CDR-H2, and CDR-H3). CDRs contain most of the residues responsible for the antibody-antigen specific interaction and thus contribute to the functional activity of the antibody molecule: they are the main determinants of antigen specificity.

[0140] Precisely defined CDR boundaries and lengths are subject to different classification and numbering systems. Therefore, CDRs can be referenced by Kabat, Chothia, contact, or any other boundary definition (including the numbering system described herein). Despite the different boundaries, each of these systems has a degree of overlap in terms of constituting the so-called "hypervariate region" within a variable sequence. Therefore, CDR definitions according to these systems can differ in length and boundary regions relative to adjacent frame regions. See, for example, Kabat (methods based on cross-species sequence variability), Chothia (methods based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al., cited above; Chothia et al., J. MoI. Biol [Journal of Molecular Biology], 1987, 196: 901-917; and MacCallum et al., J. MoI. Biol [Journal of Molecular Biology], 1996, 262: 732). Another standard system characterizing the antigen-binding side is the AbM definition used by the AbM antibody modeling software from Oxford Molecular. See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains in: Antibody Engineering Lab Manual (edited by Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). Regarding the two residue identification techniques defining overlapping regions rather than identical regions, they can be combined to define heterozygous CDRs. However, numbering according to the so-called Kabat system is preferred.

[0141] Typically, CDR formation can be classified as canonical ring structures. The term "canonical structure" refers to the main chain conformation adopted by the antigen-binding (CDR) ring. From comparative structural studies, five of the six antigen-binding rings have been found to have only a limited library of usable conformations. Each canonical structure can be characterized by the torsion angle of the polypeptide backbone. Thus, corresponding rings between antibodies can have very similar three-dimensional structures, but most of the rings exhibit high amino acid sequence variability (Chothia and Lesk, J. MoI. Biol. [Journal of Molecular Biology], 1987, 196: 901; Chothia et al., Nature [Nature], 1989, 342: 877; Martin and Thornton, J. MoI. Biol. [Journal of Molecular Biology], 1996, 263: 800). Furthermore, there is a relationship between the ring structure adopted and the surrounding amino acid sequence. The conformation of a particular canonical class is determined by the length of the ring and the amino acid residues located at key positions within the ring and within the conserved framework (i.e., outside the ring). Therefore, a particular canonical class can be assigned based on the presence of these key amino acid residues.

[0142] The term "canonical structure" can also include considerations regarding the linear sequence of the antibody, such as those cataloged by Kabat (Kabat et al., cited above). The Kabat numbering scheme (system) is a widely adopted standard for numbering the amino acid residues of variable domains of antibodies in a consistent manner and is the preferred scheme for the application of this invention, as mentioned elsewhere herein. Other structural considerations can also be used to determine the canonical structure of an antibody. For example, those differences not fully reflected by Kabat numbering can be described by the numbering system of Chothia et al., and / or revealed by other techniques (e.g., crystallography and two- or three-dimensional computational modeling). Thus, a given antibody sequence can be placed in a canonical category, which in particular allows for the identification of appropriate chassis sequences (e.g., based on the expectation of including multiple canonical structures in a library). Kabat numbering of antibody amino acid sequences and structural considerations, as described by Chothia et al., cited above, and their significance for interpreting the canonical aspects of antibody structure are described in the literature. The subunit structures and three-dimensional configurations of different classes of immunoglobulins are well known in the art. For a review of antibody structures, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, edited by Harlow et al., 1988.

[0143] CDR3 of the light chain, and especially of the heavy chain, can constitute the most important determinants of antigen binding within the variable regions of both the light and heavy chains. In some antibody constructs, the heavy chain CDR3 appears to constitute the primary contact region between the antigen and antibody. In vitro selection schemes that modify CDR3 individually can be used to alter antibody binding properties or determine which residues contribute to antigen binding. Therefore, CDR3 is typically the largest source of molecular diversity within the antibody binding side. For example, H3 can be as short as two amino acid residues or more than 26 amino acids.

[0144] In classic full-length antibodies or immunoglobulins, each light (L) chain is linked to the heavy (H) chain by a covalent disulfide bond, and two H chains are linked to each other by one or more disulfide bonds, depending on the H chain isotype. The CH domain closest to the VH is usually named CH1. The constant ("C") domain does not directly participate in antigen binding but exhibits various effector functions, such as antibody-dependent, cell-mediated cytotoxicity, and complement activation. The Fc region of an antibody is included within the heavy chain constant domain and can, for example, interact with Fc receptors located on the cell surface.

[0145] The sequences of antibody genes are highly altered after assembly and somatic mutation, and it is estimated that these altered genes encode 10^10 different antibody molecules (Immunoglobulin Genes, 2nd ed., edited by Jonio et al., Academic Press, San Diego, CA, 1995). Therefore, the immune system provides an immunoglobulin repertoire. The term "repertoire" refers to at least one nucleotide sequence, wholly or partially derived from at least one sequence encoding at least one immunoglobulin. One or more sequences can be generated by in vivo rearrangement of the V, D, and J segments of the heavy chain and the V and J segments of the light chain. Alternatively, one or more sequences can be generated from cells in response to rearrangement, such as in vitro stimulation. Alternatively, some or all of one or more sequences can be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods, see, for example, U.S. Patent 5,565,332. A repertoire may include only one sequence or may include multiple sequences, including sequences from a genetic diversity set.

[0146] The term "Fc moiety" or "Fc monomer" in this invention refers to a polypeptide comprising at least one domain functional with a CH2 domain and at least one domain functional with a CH3 domain of an immunoglobulin molecule. As is clear from the term "Fc monomer," a polypeptide comprising those CH domains is a "polypeptide monomer." An Fc monomer may be a polypeptide comprising at least a fragment of an immunoglobulin constant region excluding the first constant region immunoglobulin domain (CH1) of the heavy chain, but retaining at least a functional portion of a CH2 domain and a functional portion of a CH3 domain, wherein the CH2 domain is at the amino terminus of the CH3 domain. In a preferred aspect of this definition, an Fc monomer may be a polypeptide constant region comprising a portion of an Ig-Fc hinge region, a CH2 region, and a CH3 region, wherein the hinge region is at the amino terminus of the CH2 domain. It is envisioned that the hinge region of the present invention promotes dimerization. For example, but not limited to, such Fc polypeptide molecules can be obtained by digesting the immunoglobulin region with papain (which, of course, produces a dimer of two Fc polypeptides). In another aspect of this definition, an Fc monomer can be a polypeptide region comprising a portion of the CH2 and CH3 regions. For example, but not limited to, such Fc polypeptide molecules can be obtained by digesting immunoglobulin molecules with pepsin. In one embodiment, the polypeptide sequence of the Fc monomer is substantially similar to the following Fc polypeptide sequences: IgG1 Fc region, IgG2 Fc region, IgG3 Fc region, IgG4 Fc region, IgM Fc region, IgA Fc region, IgD Fc region, and IgE Fc region. (See, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993)). Because there are some variations among immunoglobulins, and for clarity only, the Fc monomer system refers to the last two heavy chain constant regions of IgA, IgD, and IgG immunoglobulin domains, and the last three heavy chain constant regions of IgE and IgM immunoglobulin domains. As mentioned above, the Fc monomer may also include a flexible hinge at the N-terminus of these domains. For IgA and IgM, the Fc monomer may include a J chain. For IgG, the Fc moiety includes immunoglobulin domains CH2 and CH3, and a hinge between the first two domains and CH2. Although the boundaries of the Fc moiety can vary, an example of a human IgG heavy chain Fc moiety containing the functional hinge, CH2, and CH3 domains can be defined, for example, as P476 containing residue D231 (a residue of the hinge domain—corresponding to D234 in Table 1 below) to the carboxyl terminus of the CH3 domain, respectively, L476 (for IgG4), where according to Kabat numbering. Two Fc moietyes or Fc monomers fused together via peptide linkers define a third domain of the antibody construct of the present invention, which may also be defined as an scFc domain.

[0147] In one embodiment of the present invention, it is envisioned that the scFc domain as disclosed herein, and the Fc monomers fused to each other, are only included in the third domain of the antibody construct. Consistent with the present invention, the IgG hinge region can be identified by simulation using the Kabat numbers shown in Table 1. Based on the above, it is contemplated that the hinge domain / region of the present invention comprises amino acid residues corresponding to the IgG1 sequence segments D234 to P243 according to the Kabat numbers. Similarly, it is contemplated that the hinge domain / region of the present invention comprises or is composed of the IgG1 hinge sequence DKTHTCPPCP (SEQ ID NO: 182) (variations of the sequence corresponding to segments D234 to P243 shown in Table 1 below are also contemplated, as long as the hinge region still promotes dimerization). In a preferred embodiment of the present invention, the glycosylation site at Kabat position 314 of the CH2 domain in the third domain of the antibody construct is removed by N314X substitution, wherein X is any amino acid other than Q. The substitution is preferably N314G substitution. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (according to the Kabat position): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321). It is also envisioned that the third structural domain of the antibody construct of the present invention comprises, or consists of, the following in the order of amino to carboxyl groups: DKTHTCPPCP (SEQ ID NO: 182) (i.e., hinge)-CH2-CH3-linker-DKTHTCPPCP (SEQ ID NO: 182) (i.e., hinge)-CH2-CH3. In a preferred embodiment, the peptide linker of the above-described antibody construct is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, wherein x is an integer of 5 or greater (e.g., 5, 6, 7, 8, etc. or greater), preferably 6 ((Gly4Ser)6). The construct may further comprise the above-described substituted N314X, preferably N314G, and / or additional substituted V321C and R309C. In a preferred embodiment of the antibody construct of the present invention as defined above, it is envisioned that the second domain binds to an extracellular surface site of the CD3ε chain in humans and / or macaques. [Table 1]: Kabat numbers of amino acid residues in the hinge region [Hinges] [IMGT] [serial number] [IgG, 1] [Amino acid translation] [Kabat] [serial number] [1] I 226 [2] P 227 [3] K 228 [4] S 232 [5] C 233 [6] D 234 [7] K 235 [8] T 236 [9] H 237

[10] T 238

[11] C 239

[12] P 240

[13] P 241

[14] C 242

[15] P 243 In another embodiment of the invention, the hinge domain / region comprises or consists of the following: the IgG2 subtype hinge sequence ERKCCVECPPCP (SEQ ID NO: 183), the IgG3 subtype hinge sequence ELKTPLDTTHTCPRCP (SEQ ID NO: 184) or ELKTPLGDTTHTCPRCP (SEQ ID NO: 185), and / or the IgG4 subtype hinge sequence ESKYGPPCPSCP (SEQ ID NO: 186). The IgG1 subtype hinge sequence may be one of the following: EPKSCDKTHTCPPCP (as shown in Table 1 and SEQ ID NO: 183). These core hinge regions are therefore also conceived in the context of the invention.

[0148] The locations and sequences of the IgG CH2 and IgG CD3 domains can be simulated using the Kabat numbers shown in Table 2. [Table 2]: Kabat numbers of amino acid residues in the CH2 and CH3 regions of IgG [IgG] [Subtype] [CH2 aa] [translate] [CH2 Kabat] [serial number] [CH3 aa] [translate] [CH3 Kabat] [serial number] [IgG, 1] APE… …KAK 244… …360 GQP... [ P ] [GK] 361… …478 [IgG, 2] APP… …KTK 244… …360 GQP... [ P ] [GK] 361… …478 [IgG, 3] APE… …KTK 244… …360 GQP... [ P ] [GK] 361… …478 [IgG, 4] APE… …KAK 244… …360 GQP... [ L ] [GK] 361… …478

[0149] In one embodiment of the invention, the amino acid residues highlighted in bold are removed from the CH3 domain of the first or two Fc monomers.

[0150] A peptide linker in which peptide monomers of the third domain ("Fc moiety" or "Fc monomer") are fused together preferably contains at least 25 amino acid residues (25, 26, 27, 28, 29, 30, etc.). More preferably, the peptide linker contains at least 30 amino acid residues (30, 31, 32, 33, 34, 35, etc.). Even more preferably, the linker contains up to 40 amino acid residues, more preferably up to 35 amino acid residues, and most preferably exactly 30 amino acid residues. A preferred embodiment of such a peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, where x is an integer of 5 or greater (e.g., 6, 7, or 8). Preferably, the integer is 6 or 7, and more preferably, the integer is 6.

[0151] When using linkers to fuse a first domain to a second domain or to fuse either the first or second domain to a third domain, the linker preferably has a length and sequence sufficient to ensure that each of the first and second domains can independently retain its differential binding specificity. For peptide linkers connecting at least two binding domains (or two variable domains) in an antibody construct of the present invention, those linkers containing only a small number of amino acid residues (e.g., 12 amino acid residues or less) are preferred. Therefore, peptide linker systems with 12, 11, 10, 9, 8, 7, 6, or 5 amino acid residues are preferred. Contemplated peptide linkers with fewer than 5 amino acids contain 4, 3, 2, or 1 amino acid, wherein Gly-rich linkers are preferred. Preferred embodiments for peptide linkers used to fuse the first and second domains are depicted in SEQ ID NO: 1. A preferred embodiment of the linker for using peptide linkers that fuse the second and third structural domains is the (Gly)4-linker, referred to as the G4-linker.

[0152] In the context of one of the aforementioned "peptide linkers," a particularly preferred "single" amino acid group is Gly. Therefore, the peptide linker can consist of a single amino acid, Gly. In a preferred embodiment of the invention, the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or greater (e.g., 2 or 3). Preferred linkers are depicted in SEQ ID Nos: 1 to 12. The peptide linkers are characterized by the absence of secondary structure promotion, being known in the art, and described, for example, by Dall'Acqua et al. (Biochem. [Biochemistry] (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol [Molecular Immunology] (1992) 29, 21-30), and Raag and Whitlow (FASEB [Journal of the Federation for Experimental Biology of the United States of America] (1995) 9 (1), 73-80). Furthermore, peptide linkers that do not promote any secondary structure are preferred. The connection between the domains can be provided, for example, by genetic engineering, as described in the examples. Methods for preparing fused and operably linked bispecific single-stranded constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., WO 99 / 54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

[0153] In a preferred embodiment of the antibody construct of the present invention, the first and second domains are formed in the form of an antibody construct selected from the group consisting of (scFv)2, scFv-single-domain mAb, biantibody, and oligomers of any of these forms.

[0154] According to a particularly preferred embodiment, and as described in the appended examples, the first and second domains of the antibody construct of the present invention are "bispecific single-chain antibody constructs," more preferably bispecific "single-chain Fv" (scFv). Although the two domains VL and VH of the Fv fragment are encoded by independent genes, these two domains can be joined by a synthetic linker using a recombination method, as described above, which allows them to be prepared as a single protein chain, wherein the VL and VH regions pair to form a monovalent molecule; see, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA [Proceedings of the National Academy of Sciences] 85: 5879-5883. These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the function of the fragments is evaluated in the same manner as that of complete or full-length antibodies. Therefore, single-chain variable fragments (scFv) are fusion proteins of the heavy chain (VH) and light chain (VL) variable regions of immunoglobulins, typically linked by short linker peptides of about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linkers are usually enriched with glycine for flexibility and with serine or threonine for solubility, and can link the N-terminus of the VH to the C-terminus of the VL, or vice versa. Despite the removal of constant regions and the introduction of linkers, the protein retains the specificity of the original immunoglobulin.

[0155] Bispecific single-chain antibody constructs are known in the art and described in the following: WO 99 / 54440; Mack, J. Immunol. [Journal of Immunology] (1997), 158, 3965-3970; Mack, PNAS [Proceedings of the National Academy of Sciences of the United States of America], (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy], (1997), 45, 193-197; Löffler, Blood [Blood], (2000), 95, 6, 2098-2103; Brühl, Immunol. [Immunology], (2001), 166, 2420-2426; Kipriyanov, J. Mol. Biol. [Journal of Molecular Biology], (1999), 293, 41-56. The techniques described for generating single-chain antibodies (see in particular U.S. Patent 4,946,778; Kontermann and Dübel (2010), cited above, and Little (2009), cited above) can be applied to generate single-chain antibody constructs that specifically recognize one or more selected targets.

[0156] Bivalent (also known as divalent) or bispecific single-chain variable fragments (linked-scFvs or di-scFvs with the form (scFv)2) can be engineered by linking two scFv molecules (e.g., using linkers as described above). If the two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule is preferably called bivalent (i.e., having two valences for the same target epitope). If the two scFv molecules have different binding specificities, the resulting (scFv)2 molecule is preferably called bispecific. Linking can be performed by generating tandem scFvs by producing a single peptide chain with two VH regions and two VL regions (see, for example, Kufer P. et al., (2004) Trends in Biotechnology 22 (5): 238-244). Another possibility is the production of scFv molecules with linker peptides that are too short for the two variable regions to fold together (e.g., about five amino acids), thus forcing the scFv to dimerize. This type is called a biantibody (see, for example, Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90 (14): 6444-8).

[0157] Consistent with this invention, the first domain, the second domain, or both the first and second domains may comprise a single-domain antibody, a variable domain of a single-domain antibody, or at least a CDR. A single-domain antibody comprises only one (monomer) antibody variable domain capable of selectively binding to a specific antigen independently of other V regions or domains. The first single-domain antibody system is engineered from heavy-chain antibodies found in camels, and these are referred to as VHH fragments. Cartilaginous fish also possess heavy-chain antibodies (IgNARs), from which single-domain antibodies, referred to as VNAR fragments, can be obtained. An alternative approach is to split a dimer variable domain from common immunoglobulins, such as those from humans or rodents, into monomers, thus obtaining VH or VL as a single-domain Ab. Although most research on single-domain antibodies is currently based on heavy-chain variable domains, nanoantibodies derived from light chains have also shown specific binding to target epitopes. Examples of single-domain antibodies are so-called sdAbs, nanoantibodies, or single variable-domain antibodies.

[0158] Therefore, (single-domain mAb)2 is a monoclonal antibody construct consisting of (at least) two single-domain monoclonal antibodies, each selected individually from the group consisting of VH, VL, VHH, and VNAR. The linker is preferably in the form of a peptide linker. Similarly, "scFv-single-domain mAb" is a monoclonal antibody construct consisting of at least one single-domain antibody as described above and one scFv molecule as described above. Again, the linker is preferably in the form of a peptide linker.

[0159] Whether an antibody construct competitively binds to another given antibody construct can be measured in competitive assays such as competitive ELISA or cell-based competitive assays. Avidin-conjugated microparticles (beads) can also be used. Similar to avidin-coated ELISA plates, each of these beads can be used as a substrate on which an assay can be performed when reacted with a biotinylated protein. The antigen is coated onto the beads, and then pre-coated with a first antibody. A second antibody is added, and any additional binding is determined. Possible readout methods include flow cytometry.

[0160] T cells, or T lymphocytes, are a class of lymphocytes (which are themselves a type of leukocyte) that play a central role in cell-mediated immunity. There are several subgroups of T cells, each with a distinct function. T cells can be distinguished from other lymphocytes (such as B cells and NK cells) by the presence of the T cell receptor (TCR) on their cell surface. The TCR is responsible for recognizing antigens that bind to the major histocompatibility complex (MHC) molecule and is composed of two distinct protein chains. In 95% of T cells, the TCR consists of alpha (α) and beta (β) chains. When the TCR binds to the antigenic peptide and the MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by associated enzymes, co-receptors, specialized adaptor molecules, and activated or released transcription factors.

[0161] The CD3 receptor complex is a protein complex composed of four chains. In mammals, the complex contains a CD3γ (gamma) chain, a CD3δ (delta) chain, and two CD3ε (eposeridone) chains. These chains associate with the T cell receptor (TCR) and the so-called zeta (ζ) chain to form the T cell receptor CD3 complex and generate activation signals in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (eposeridone) chains are highly associated cell surface proteins of the immunoglobulin superfamily containing a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif essential for TCR signaling, called the immunoreceptor tyrosine-based activation motif, or ITAM for short. The CD3ε molecule is a polypeptide encoded in humans by the CD3E gene located on chromosome 11. The preferred epitopes of CD3ε are contained within amino acid residues 1-27 of the extracellular domain of the human CD3ε. It is envisioned that the antibody constructs according to the present invention typically and advantageously exhibit less nonspecific T cell activation, which is undesirable in specific immunotherapies. This implies a reduced risk of side effects.

[0162] Recruiting T cells via multispecific or at least bispecific antibody constructs for redirected lysis of target cells involves cytolytic synapse formation and delivery of perforin and granzymes. The conjugated T cells are capable of continuous target cell lysis and are unaffected by immune escape mechanisms that interfere with peptide antigen processing and presentation or selective T cell differentiation; see, for example, WO 2007 / 042261.

[0163] Cytotoxicity mediated by the antibody constructs of this invention can be measured in various ways. Effector cells can be, for example, stimulated enriched (human) CD8-positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMCs). If the target cell line is of macaque origin or expressed or transfected with a macaque target cell surface antigen bound to a first domain, the effector cells should also be of macaque origin, such as a macaque T cell line, for example, 4119LnPx. Target cells should express the target cell surface antigen (at least the extracellular domain), such as human or macaque target cell surface antigen. Target cells can be cell lines stably or transiently transfected with the target cell surface antigen (e.g., human or macaque target cell surface antigen) (e.g., CHO). Alternatively, target cells can be naturally expressed cell lines that are positive for the target cell surface antigen. For target cell lines expressing higher levels of the target cell surface antigen on their cell surface, a lower EC50 value is expected. The effector cell to target cell (E:T) ratio is typically about 10:1, but can vary. The cytotoxic activity of the xCD3 bispecific antibody construct targeting cell surface antigen can be measured in a 51Cr-release assay (approximately 18 hours of incubation) or in a FACS-based cytotoxicity assay (approximately 48 hours of incubation). The incubation time (cytotoxic response) can also be modified. Other methods for measuring cytotoxicity are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays (including bioluminescence assays), sulforhodamine B (SRB) assays, WST assays, selection-based assays, and ECIS techniques.

[0164] Preferably, the cytotoxic activity mediated by the xCD3 bispecific antibody construct of the target cell surface antigen of the present invention is measured in a cell-based cytotoxicity assay. It can also be measured in a 51Cr-release assay. It is expressed as an EC50 value, which corresponds to the half-maximal effective concentration (the concentration of the antibody construct that induces a cytotoxic response midway between baseline and maximum). Preferably, the EC50 value of the target cell surface antigen xCD3 bispecific antibody construct is ≤ 5000 pM or ≤ 4000 pM, more preferably ≤ 3000 pM or ≤ 2000 pM, even more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 400 pM or ≤ 300 pM, even more preferably ≤ 200 pM, even more preferably ≤ 100 pM, even more preferably ≤ 50 pM, even more preferably ≤ 20 pM or ≤ 10 pM, and most preferably ≤ 5 pM.

[0165] The EC50 values ​​given above can be measured in various assays. Those skilled in the art will know that when stimulated / enriched CD8+ T cells are used as effector cells, a lower EC50 value can be expected compared to unstimulated PBMCs. Furthermore, a lower EC50 value can be expected when target cells express a high level of target cell surface antigen compared to rats with low target expression. For example, when stimulated / enriched human CD8+ T cells are used as effector cells (and target cells are cells transfected with target cell surface antigen, such as CHO cells or target cell surface antigen-positive human cell lines), the EC50 value of the target cell surface antigen xCD3 bispecific antibody construct is preferably ≤ 1000 pM, more preferably ≤ 500 pM, even more preferably ≤ 250 pM, even more preferably ≤ 100 pM, even more preferably ≤ 50 pM, even more preferably ≤ 10 pM, and most preferably ≤ 5 pM. When using human PBMCs as effector cells, the EC50 value of the target cell surface antigen xCD3 bispecific antibody construct is preferably ≤ 5000 pM or ≤ 4000 pM (especially when the target cell line is a human cell line positive for target cell surface antigen), more preferably ≤ 2000 pM (especially when the target cell line is a cell line transfected with target cell surface antigen such as CHO cells), more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 200 pM, even more preferably ≤ 150 pM, even more preferably ≤ 100 pM, and most preferably ≤ 50 pM or lower. When using macaque T cell lines such as LnPx4119 as effector cells and macaque target cell line transfected with target cell surface antigens such as CHO cells as target cell lines, the EC50 value of the target cell surface antigen xCD3 bispecific antibody construct is preferably ≤ 2000 pM or ≤ 1500 pM, more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 300 pM or ≤ 250 pM, even more preferably ≤ 100 pM, and most preferably ≤ 50 pM.

[0166] Preferably, the target cell surface antigen xCD3 bispecific antibody construct of the present invention does not induce / mediate lysis or substantially does not induce / mediate the lysis of target cell surface antigen-negative cells such as CHO cells. The terms "does not induce lysis," "substantially does not induce lysis," "does not mediate lysis," or "substantially does not mediate lysis" mean that the antibody construct of the present invention does not induce or mediate more than 30%, preferably no more than 20%, more preferably no more than 10%, and particularly preferably no more than 9%, 8%, 7%, 6%, or 5% of the lysis of target cell surface antigen-negative cells, thereby setting the lysis of target cell surface antigen-positive human cell lines as 100%. This is generally applicable to antibody constructs at concentrations up to 500 nM. Those skilled in the art know how to measure cell lysis effortlessly. Furthermore, this specification teaches specific instructions on how to measure cell lysis.

[0167] The difference in cytotoxic activity between the monomeric and dimeric isoforms of a single target cell surface antigen xCD3 bispecific antibody construct is called the "potency gap". This potency gap can be calculated, for example, as the ratio between the EC50 values ​​of the monomeric and dimeric forms of the molecule. Preferably, the potency gap of the target cell surface antigen xCD3 bispecific antibody construct of the present invention is ≤ 5, more preferably ≤ 4, even more preferably ≤ 3, even more preferably ≤ 2, and most preferably ≤ 1.

[0168] The first and / or second (or any other) binding domains of the antibody construct of the present invention preferably possess cross-species specificity for members of the order primates. Cross-species specific CD3 binding domains are described, for example, in WO 2008 / 119567. According to one embodiment, in addition to binding to human target cell surface antigens and human CD3, the first and / or second binding domains also bind to target cell surface antigens / CD3 of primates, including (but not limited to) New World primates (such as marmosets, woolly tamarins, or squirrel monkeys), Old World primates (such as baboons and rhesus macaques), gibbons, and non-human hominids.

[0169] In one embodiment of the antibody construct of the present invention, the first binding domain binds to a human target cell surface antigen and further binds to a macaque target cell surface antigen (such as the target cell surface antigen of a cynomolgus monkey), and more preferably, binds to the macaque target cell surface antigen expressed on surface macaque cells. The affinity of the first binding domain for the macaque target cell surface antigen is preferably ≤ 15 nM, more preferably ≤ 10 nM, even more preferably ≤ 5 nM, even more preferably ≤ 1 nM, even more preferably ≤ 0.5 nM, even more preferably ≤ 0.1 nM, and most preferably ≤ 0.05 nM or even ≤ 0.01 nM.

[0170] Preferably, the affinity gap (determined, for example, by BiaCore or Scatchard analysis) of the antibody construct according to the invention for binding macaque target cell surface antigen to human target cell surface antigen [ma target cell surface antigen: hu target cell surface antigen] is < 100, more preferably < 20, more preferably < 15, further preferably < 10, even more preferably < 8, more preferably < 6, and most preferably < 2. The preferred range of the affinity gap of the antibody construct according to the invention for binding macaque target cell surface antigen to human target cell surface antigen is between 0.1 and 20, more preferably between 0.2 and 10, even more preferably between 0.3 and 6, even more preferably between 0.5 and 3 or between 0.5 and 2.5, and most preferably between 0.5 and 2 or between 0.6 and 2.

[0171] The second (binding) domain of the antibody construct of this invention binds to human CD3ε and / or macaque CD3ε. In a preferred embodiment, the second domain further binds to the CD3ε of marmosets, woolly-crowned tamarins, or squirrel monkeys. Both marmosets and *Saguinus yophile* are New World primates belonging to the subfamily Callitrichidae, while squirrel monkeys belong to the family Cebidae.

[0172] For the antibody construct of the present invention, the second domain that binds to the extracellular domain of human and / or macaque CD3 includes a VL region containing CDR-L1, CDR-L2 and CDR-L3: (a) CDR-L1 as described in SEQ ID NO: 27 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 28 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 29 of WO 2008 / 119567; (b) CDR-L1 as described in SEQ ID NO: 117 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 118 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 119 of WO 2008 / 119567; and I. CDR-L1 as described in SEQ ID NO: 153 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 154 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 155 of WO 2008 / 119567.

[0173] In another preferred embodiment of the antibody construct of the present invention, the second domain that binds to the extracellular domain of the human and / or macaque CD3ε chain comprises a VH region containing CDR-H1, CDR-H2 and CDR-H3: (a) CDR-H1 as described in SEQ ID NO: 12 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 13 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 14 of WO 2008 / 119567; (b) CDR-H1 as described in SEQ ID NO: 30 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 31 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 32 of WO 2008 / 119567; I. CDR-H1 as described in SEQ ID NO: 48 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 49 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 50 of WO 2008 / 119567; (d) CDR-H1 as described in SEQ ID NO: 66 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 67 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 68 of WO 2008 / 119567; I. CDR-H1 as depicted in SEQ ID NO: 84 of WO 2008 / 119567, CDR-H2 as depicted in SEQ ID NO: 85 of WO 2008 / 119567, and CDR-H3 as depicted in SEQ ID NO: 86 of WO 2008 / 119567; (f) CDR-H1 as described in SEQ ID NO: 102 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 103 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 104 of WO 2008 / 119567; (g) CDR-H1 as described in SEQ ID NO: 120 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 121 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 122 of WO 2008 / 119567; (h) CDR-H1 as described in SEQ ID NO: 138 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 139 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 140 of WO 2008 / 119567; (i) CDR-H1 as described in SEQ ID NO: 156 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 157 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 158 of WO 2008 / 119567; and (j) CDR-H1 as described in SEQ ID NO: 174 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 175 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 176 of WO 2008 / 119567.

[0174] In a preferred embodiment of the antibody construct of the present invention, the above three groups of VL CDRs are combined with the above ten groups of VH CDRs in the second binding domain to form a group (30), each group containing 1-3 CDR-L and 1-3 CDR-H.

[0175] For the antibody construct of the present invention, the second domain binding to CD3 includes a VL region selected from the group consisting of: such as the VL region depicted in SEQ ID NO: 17, 21, 35, 39, 53, 57, 71, 75, 89, 93, 107, 111, 125, 129, 143, 147, 161, 165, 179 or 183 of WO 2008 / 119567 or the VL region depicted in SEQ ID NO: 200.

[0176] Also preferably, the second domain associated with CD3 includes a VH region selected from the group consisting of: such as the VH region depicted in SEQ ID NO: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177 or 181 of WO 2008 / 119567 or the VH region depicted in SEQ ID NO: 201.

[0177] More preferably, the antibody construct of the present invention is characterized in that the second domain binding to CD3 includes a VL region and a VH region selected from the group consisting of the following: (a) The VL region as depicted in SEQ ID NO: 17 or 21 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 15 or 19 of WO 2008 / 119567; (b) The VL region as depicted in SEQ ID NO: 35 or 39 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 33 or 37 of WO 2008 / 119567; I. The VL region as depicted in SEQ ID NO: 53 or 57 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 51 or 55 of WO 2008 / 119567; (d) The VL region as depicted in SEQ ID NO: 71 or 75 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 69 or 73 of WO 2008 / 119567; I. The VL region as depicted in SEQ ID NO: 89 or 93 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 87 or 91 of WO 2008 / 119567; (f) The VL region as depicted in SEQ ID NO: 107 or 111 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 105 or 109 of WO 2008 / 119567; (g) The VL region as depicted in SEQ ID NO: 125 or 129 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 123 or 127 of WO 2008 / 119567; (h) The VL region as depicted in SEQ ID NO: 143 or 147 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 141 or 145 of WO 2008 / 119567; (i) the VL region as depicted in SEQ ID NO: 161 or 165 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 159 or 163 of WO 2008 / 119567; and (j) The VL region as described in SEQ ID NO: 179 or 183 of WO 2008 / 119567 and the VH region as described in SEQ ID NO: 177 or 181 of WO 2008 / 119567.

[0178] A further preferred embodiment of the antibody construct of the present invention is that the second domain binding to CD3 includes the VL region as depicted in SEQ ID NO: 200 and the VH region as depicted in SEQ ID NO: 201.

[0179] According to a preferred embodiment of the antibody construct of the present invention, the first and / or second domains have the following form: the VH and VL regions are in the form of a single-chain antibody (scFv). The VH and VL regions are arranged in a VH-VL or VL-VH order. Preferably, the VH region is located at the N-terminus of the linker sequence, and the VL region is located at the C-terminus of the linker sequence.

[0180] A preferred embodiment of the antibody construct of the present invention is characterized in that the second domain binding to CD3 comprises an amino acid sequence selected from the group described in SEQ ID NO: 202, which consists of the following items: SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 of WO 2008 / 119567.

[0181] Covalent modifications of antibody constructs are also included within the scope of this invention and are typically, but not always, performed post-translational. For example, several types of covalent modifications of antibody constructs are introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizer capable of reacting with selected side chain or N- or C-terminal residues.

[0182] Cysteine ​​residues are most commonly reacted with α-haloacetic esters (and corresponding amines), such as chloroacetic acid or chloroacetamide, to yield carboxymethyl or carboxyacetamide methyl derivatives. Cysteine ​​residues can also be derived by reacting with bromotrifluoroacetone, α-bromo-β-(5-imidazolyl)propionic acid, chloroacetamide phosphate, N-alkylmaleimine, 3-nitro-2-pyridyl disulfide, methyl-2-pyridyl disulfide, p-chloromercuric benzoate, 2-chloromercuric-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0183] Histidine residues are derived by reacting with diethyl pyrocarbonate at pH 5.5–7.0, as this agent is relatively specific to the histidine side chain. Bromobenzylmethyl bromide is also useful; preferably, the reaction is carried out in 0.1 M sodium dimethylarsinate at pH 6.0. Lysidine residues and amino-terminal residues react with succinic anhydride or other carboxylic anhydrides. Derivatization with these agents has the effect of reversing the charge of the lysidine residues. Other suitable reagents for derivatizing α-amino residues include imine esters, such as methyl pyridinium imide; pyridoxal phosphate; pyridoxal; chlorine borohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reactions with glyoxylates.

[0184] Arginine residues can be modified by reacting with one or more conventional reagents, including benzoxaldehyde, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Due to the high pKa of the guanidine functional group, the derivatization of arginine residues requires the reaction to be carried out under basic conditions. Furthermore, these reagents can react with lysine groups and the ε-amino group of arginine.

[0185] Tyramine residues can be specifically modified, with particular interest in introducing spectral labeling into tyramine residues through reaction with aromatic diazo compounds or tetranitromethane. Most commonly, N-acetylimazole and tetranitromethane are used to form O-acetylityramine and 3-nitro derivatives, respectively. The chloramine-T method described above is suitable for preparing labeled proteins for radioimmunoassays using iodination of tyramine residues with 125I or 131I.

[0186] The carboxyl side group (aspartic acidyl or glutamineyl) is selectively modified by reacting with a carbodiimide (R'—N=C=N–R'), where R and R' are different alkyl groups as needed, such as 1-cyclohexyl-3-(2- Lino-4-ethyl)carbodiimide or 1-ethyl-3-(4-aza-4,4-dimethylpentyl)carbodiimide. Furthermore, aspartic acid residues and glutamine residues are converted to aspartic acid residues and glutamine residues by reacting with ammonium ions.

[0187] Bifunctional derivatization can be used to crosslink the antibody constructs of this invention to water-insoluble carrier matrices or surfaces for use in a variety of methods. Commonly used crosslinking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters (e.g., esters with 4-azidosalicylic acid), and iso-bifunctional imide esters, including disuccinimide esters such as 3,3'-dithiobis(succinimide propionate), and bifunctional maleimides such as bis-N-maleimide-1,8-octane. Derivatizing agents such as methyl 3-[(p-azidophenyl)dithio]propionic acid produce photoactivated intermediates capable of forming crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates and reactive substrates, as described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440, are used for protein fixation.

[0188] Glutamic acid residues and aspartic acid residues are typically deamined to yield the corresponding glutamine residues and aspartic acid residues, respectively. Alternatively, these residues may be deamined under weakly acidic conditions. Both forms of these residues are within the scope of this invention.

[0189] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl groups of serine or threonine residues, methylation of the α-amino groups of the side chains of lysine, arginine, and histidine (TE Creighton, Proteins: Structure and Molecular Properties, WH Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of N-terminal amines, and amylation of any C-terminal carboxyl group.

[0190] Another type of covalent modification of antibody constructs included within the scope of this invention involves altering the glycosylation pattern of a protein. As is known in the art, the glycosylation pattern can be dependent on the protein's sequence (e.g., the presence or absence of specific glycosylated amino acid residues discussed below) or the host cell or organism from which the protein is produced. Specific expression systems are discussed below.

[0191] Glycosylation of peptides is typically N-linked or O-linked. N-linking refers to the attachment of the carbohydrate moiety to a side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid other than proline) are recognition sequences that enzymatically link the carbohydrate moiety to the asparagine side chain. Therefore, the presence of either of these tripeptide sequences in a peptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylglucosamine, galactose, or xylose to a hydroxy amino acid, most commonly serine or threonine, but 5-hydroxyproline or 5-hydroxylysine can also be used.

[0192] Adding glycosylation sites (for N-linked glycosylation sites) to an antibody construct can be conveniently accomplished by altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences. Alternatives (for O-linked glycosylation sites) can be made by adding or substituting one or more serine or threonine residues into the starting sequence. For convenience, the amino acid sequence of the antibody construct is preferably altered at the DNA level, particularly by mutating the DNA encoding the polypeptide at a preselected base to generate a codon that will be translated into the desired amino acid.

[0193] Another means of increasing the amount of carbohydrate moiety on antibody constructs is by chemically or enzymatically coupling glycosides to proteins. The advantage of these procedures is that they do not require the production of proteins in host cells with glycosylation capabilities for N- and O-linking glycosylation. Depending on the coupling method used, one or more sugars may be linked to (a) arginine and histidine, (b) a free carboxyl group, (c) a free thiol group, such as those of cysteine, (d) a free hydroxyl group, such as those of serine, threonine, or hydroxyproline, (f) an aromatic residue, such as those of phenylalanine, tyrosine, or tryptophan, or (c) an amide group of glutamic acid. These methods are described in WO 87 / 05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem. [CRC Critical Review of Biochemistry], pp. 259–306.

[0194] The removal of carbohydrate moieties present on the starting antibody construct can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposing the protein to the compound trifluoromethanesulfonic acid, or an equivalent compound. This treatment results in the cleavage of most or all sugars except the linker sugar (N-acetylglucosamine or N-acetylglucosamine), while keeping the polypeptide intact. Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. [Journal of Biochemistry and Biophysics] 259: 52 and Edge et al., 1981, Anal. Biochem. [Analytical Biochemistry] 118: 131. Enzymatic cleavage of carbohydrate moieties on polypeptides can be achieved by using various endoglycosidases and exoglycosidases, as described by Thotakura et al., 1987, Meth. Enzymol. [Enzymatic Methods] 138: 350. Glycosylation at potential glycosylation sites can be prevented by using the compound tunicamycin, as described by Duskin et al., 1982, J. Biol. Chem. [Journal of Biochemistry] 257: 3105. Tanimycin blocks the formation of protein-N-glycosidic bonds.

[0195] This document also considers other modifications to the antibody construct. For example, another type of covalent modification of the antibody construct includes linking the antibody construct to various non-protein polymers, including but not limited to various polyols such as polyethylene glycol, polypropylene glycol, polyoxyethylene, or copolymers of polyethylene glycol and polypropylene glycol, in the manner illustrated in U.S. Patent Nos. 4,640,835, 4,496,689, 4,301,144, 4,670,417, 4,791,192, or 4,179,337. Furthermore, as is known in the art, amino acid substitutions can be made at different positions within the antibody construct, for example, to facilitate the addition of polymers such as PEG.

[0196] In some embodiments, the covalent modification of the antibody construct of the present invention includes the addition of one or more labels. The labeling group can be coupled to the antibody construct via spacer arms of various lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and can be used in carrying out the present invention. The terms "label" or "labeling group" refer to any detectable label. Generally, labels fall into several categories depending on the assay in which they will be detected—examples include, but are not limited to: a) Isotope labeling, which may be a radioactive isotope or a heavy isotope, such as a radioactive isotope or a radionuclide (e.g., 3H, 14C, 15N, 35S, 89Zr, 90Y, 99Tc, 111In, 125I, 131I). b) Magnetic markings (e.g., magnetic particles) c) Redox active components d) Optical dyes (including but not limited to chromophores, phosphors, and fluoresces), such as fluorescein groups (e.g., FITC, rhodamine, lanthanide phosphors), chemiluminescent groups, and fluoresces, which may be "small molecule" fluoresces or protein fluoresces. e) Enzymatic groups (e.g., horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase) f) Biotinylated groups g) A predetermined polypeptide epitope recognized by the second reporter (e.g., leucine zipper pair sequence, binding side of the second antibody, metal-binding domain, epitope tag, etc.).

[0197] "Fluorescent labeling" refers to any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include, but are not limited to, luciferin, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methyl-coumarin, pyrene, malachite green, stilbene, fluorescent yellow, waterfall blue J, Texas red, IAEDANS, EDANS, BODIPY FL, LC red 640, Cy5, Cy5.5, LC red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), waterfall blue, waterfall yellow, and R-phycoerythrin (PE) (Molecular Probes, Eugene, Oregon). OR), FITC, Rhodamine and Texas Red (Pierce, Rockford, IL), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, PA). Suitable optical dyes (including fluorophores) are described in Richard P. Haugland's Molecular Probes Handbook.

[0198] Suitable protein fluorescent labeling includes, but is not limited to, green fluorescent proteins, including GFP of the Renilla, Ptilosarcus, or Aequorea species (Chalfie et al., 1994, Science 263: 802-805), EGFP (Clontech Laboratories, Genbank accession number U55762), and blue fluorescent proteins (BFP, Quantum Biotechnologies, Inc., 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9); Stauber, 1998, Biotechniques 24: 462-471; Heim et al., 1996, Curr. Biol. 6: 178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories), luciferase (Ichiki et al., 1993, J. Immunol. [Journal of Immunology] 150: 5408-5417), β-galactosidase (Nolan et al., 1988, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America] 85: 2603-2607) and Renilla (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, US Patent Nos. 5,292,658; 5,418,155; 5,683,888; 5,741,668; 5,777,079; 5,804,387; 5,874,304; 5,876,995; 5,925,558).

[0199] The antibody constructs of this invention may also include additional domains, such as those that facilitate the separation of molecules or involve adaptive pharmacokinetic profiles of molecules. Domains facilitating the separation of the antibody construct may be selected from peptide motifs or auxiliaryly introduced portions that can be captured in a separation method (e.g., a separation column). Non-limiting embodiments of such additional domains include peptide motifs referred to as Myc-tags, HAT-tags, HA-tags, TAP-tags, GST-tags, chitin-binding domains (CBD-tags), maltose-binding protein (MBP-tags), Flag-tags, Strep-tags and their variants (e.g., StrepII-tags), and His-tags. All antibody constructs disclosed herein, characterized by an identified CDR, may include a His-tag domain, typically referred to as a continuous His residue repeating sequence in the amino acid sequence of the molecule, preferably five, and more preferably six His residues (hexahistidine). The His-tag may be located, for example, at the N or C terminus of the antibody construct, preferably at the C terminus. Preferably, the hexahistine tag (HHHHHH) (SEQ ID NO: 199) is linked to the C-terminus of the antibody construct according to the present invention via a peptide bond. Additionally, the PLGA-PEG-PLGA conjugation system can be combined with the polyhistine tag for sustained-release application and improved pharmacokinetic profiles.

[0200] Amino acid sequence modifications of the antibody constructs described herein have also been considered. For example, it may be necessary to improve the binding affinity and / or other biological properties of the antibody construct. Amino acid sequence variants of the antibody construct can be prepared by introducing appropriate nucleotide changes into the nucleic acid of the antibody construct or by peptide synthesis. All amino acid sequence modifications described below should produce antibody constructs that retain the desired biological activity (binding to target cell surface antigens and CD3) of the unmodified parent molecule.

[0201] The term "amino acid" or "amino acid residue" typically refers to an amino acid having a generally accepted definition in the art, such as an amino acid selected from the group consisting of: alanine (Ala or A); arginine (Arg or R); aspartic acid (Asn or N); aspartic acid (Asp or D); cysteine ​​(Cys or C); glutamic acid (Gin or Q); glutamic acid (Giu or E); glycine (Giy or G); histidine (His or H); isoleucine (He or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (VaI or V), but modified, synthetic, or rare amino acids may be used as needed. Generally, amino acids can be grouped into those with nonpolar side chains (e.g., Ala, Cys, He, Leu, Met, Phe, Pro, VaI); negatively charged side chains (e.g., Asp, Giu); positively charged side chains (e.g., Arg, His, Lys); or without polar side chains (e.g., Asn, Cys, Gin, Giy, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0202] Amino acid modifications include, for example, the deletion and / or insertion and / or substitution of residues within the amino acid sequence of an antibody construct. Combinations of deletions, insertions, and substitutions are made to achieve the final construct, conditional on the final construct having the desired characteristics. Amino acid changes can also alter the post-translational processes of the antibody construct, such as changing the number or location of glycosylation sites.

[0203] For example, 1, 2, 3, 4, 5, or 6 amino acids can be inserted, substituted, or deleted in each CDR (depending on its length, of course), while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids can be inserted, substituted, or deleted in each FR. Preferably, amino acid sequence insertion into the antibody construct includes amino and / or carboxyl-terminal fusions in the range of 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 residues, insertion into a polypeptide containing 100 or more residues, and insertion into sequences of single or multiple amino acid residues. Corresponding modifications can also be made within the third domain of the antibody construct of the present invention. Insertion variants of the antibody construct of the present invention include fusions with the N-terminus or C-terminus of an enzyme antibody construct or fusions with a polypeptide.

[0204] Sites of most interest in substitution-induced mutagenesis include (but are not limited to) the CDRs of the heavy and / or light chains, particularly the hypermutation region, but changes in the FRs of the heavy and / or light chains are also considered. Preferred substitutions are conservative substitutions as described herein. Preferably, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids can be substituted in the CDR, and 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids can be substituted in the frame region (FR), depending on the length of the CDR or FR. For example, if the CDR sequence contains 6 amino acids, it is conceivable that 1, 2, or 3 of those amino acids are substituted. Similarly, if the CDR sequence contains 15 amino acids, it is conceivable that 1, 2, 3, 4, 5, or 6 of those amino acids are substituted.

[0205] A useful method for identifying certain residues or regions in an antibody construct that are preferred mutagenic sites is called "alanine scanning mutagenesis," as described by Cunningham and Wells in Science, 244: 1081-1085 (1989). Here, residues or target residue groups (e.g., charged residues such as arg, asp, his, lys, and glu) in the antibody construct are identified and replaced with neutral or negatively charged amino acids (preferably alanine or polyalanine) to affect the interaction between the amino acid and the epitope.

[0206] Then, amino acid positions exhibiting functional sensitivity to substitution are refined by introducing further or other variants at or to the substitution site. Therefore, while the sites or regions used to introduce amino acid sequence changes are predetermined, the nature of the mutation itself need not be predetermined. For example, to analyze or optimize the performance of mutations at a given site, alanine scanning or random mutagenesis can be performed at the target codon or region, and antibody construct variants exhibiting these mutations can be screened for the optimal combination of desired activities. Techniques for substitution mutations at predetermined sites in DNA with known sequences are well-known, such as M13 primer mutagenesis and PCR mutagenesis. Mutants are screened using assays of antigen-binding activity (such as target cell surface antigens or CD3 binding).

[0207] Generally, if an amino acid is substituted in one or all of the CDRs in the heavy chain and / or light chain, it is preferable that the subsequently obtained "substituted" sequence is at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, and particularly preferably 90% or 95% identical to the "initial" CDR sequence. This means that the substitution depends on the length of the degree of identity between the CDR and the "substituted" sequence. For example, a CDR having 5 amino acids is preferably 80% identical to its substituted sequence in order to substitute at least one amino acid. Therefore, the CDR of an antibody construct can have different degrees of identity with its substituted sequence; for example, CDRL1 can have 80% identity, while CDRL3 can have 90%.

[0208] Preferred substitutions (or replacements) are conserved substitutions. However, any substitution (including non-conserved substitutions or one or more of the "exemplary substitutions" listed in Table 3 below) is contemplated, provided that the antibody construct retains its ability to bind to the target cell surface antigen via the first domain and to CD3, CD3ε, via the second domain, and / or that its CDR is identical to the subsequently substituted sequence (with at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, and particularly preferably 90% or 95% identity with the "original" CDR sequence).

[0209] Conservative substitutions are shown under the heading "Preferred Substitutions" in Table 3. If such substitutions result in changes in biological activity, then further substantial changes, such as those designated as "Exemplary Substitutions" in Table 3 or as described below with further reference to the amino acid categories, may be introduced, and desired characteristics may be selected. [Table 3]: Amino acid substitutions primitive Exemplary replacement Better alternative Ala(A) val, leu, ile val Arg I lys, gln, asn lys Asn(N) gln, his, asp, lys, arg gln Asp(D) glu、asn glu Cys I ser、ala ser Gln(Q) asn, glu asn Glu I asp, gln Asp Gly(G) Ala Ala His(H) asn, gln, lys, arg Arg Ile(I) leu, val, met, ala, phe Leu Leu(L) norleucine, ile, val, met, ala Ile Lys(K) arg, gln, asn Arg Met(M) leu, phe, ile Leu Phe(F) leu, val, ile, ala, tyr Tyr Pro(P) Ala Ala Ser(S) Thr Thr Thr(T) Ser Ser Trp(W) tyr, phe Tyr Tyr(Y) trp, phe, thr, ser Phe Val(V) ile, leu, met, phe, ala Leu

[0210] The substantial modification of the biological properties of the antibody constructs of the present invention is accomplished by selecting substitutions that are significantly different in maintaining the following effects: (a) the structure of the polypeptide backbone in the substitution region, such as sheet or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the majority of the side chain. Naturally occurring residues are grouped based on common side chain characteristics: (1) hydrophobic: leucine, met, ala, val, leu, ile; (2) neutral hydrophilic: cys, ser, thr, asn, gln; (3) acidic: asp, glu; (4) basic: his, lys, arg; (5) residues affecting chain orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

[0211] Non-conservative substitutions would require replacing one member of one of these categories with another. Any cysteine ​​residue that does not participate in maintaining the proper conformation of the antibody construct can generally be substituted with a serine residue to improve the oxidative stability of the molecule and prevent aberrant crosslinking. Conversely, one or more cysteine ​​bonds can be added to the antibody to improve its stability (especially in the case of anti-system antibody fragments such as Fv fragments).

[0212] For amino acid sequences, sequence identity and / or similarity are determined using standard techniques known in the art, including but not limited to: local sequence identity algorithms (Smith and Waterman, 1981, Adv. Appl. Math. 2: 482); sequence identity alignment algorithms (Needleman and Wunsch, 1970, J. Mol. Biol. 48: 443); similarity retrieval methods (Pearson and Lipman, 1988, Proc. Nat. Acad. Sci. USA 85: 2444); computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Suite software, Genetics Computer Group, 575 Science Drive, Madison, Wisconsin); and Devereux et al., 1984. The best-matching sequence program described in Nucl. Acid Res. 12: 387-395 is preferably used with default settings or by inspection. Preferably, the identity percentage is calculated using FastDB based on the following parameters: mismatch penalty of 1; vacancy penalty of 1; vacancy size penalty of 0.33; and connection penalty of 30. (See "Current Methods in Sequence Comparison and Analysis", Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp. 127-149 (1988), Alan R. Liss, Inc.)

[0213] A useful example of an algorithm is PILEUP. PILEUP uses progressive pairwise alignment to create multiple sequence alignments from a set of related sequences. It can also draw a dendrogram showing the clustering relationships used to create the alignments. PILEUP uses a simplified version of the progressive alignment method described by Feng and Doolittle, 1987, J. Mol. Evol. [Journal of Molecular Evolution] 35: 351-360; this method is similar to that described by Higgins and Sharp, 1989, CABIOS 5: 151-153. Useful PILEUP parameters include a preset vacancy weight of 3.00, a preset vacancy length weight of 0.10, and a weighted terminal vacancy.

[0214] Another example of a useful algorithm is the BLAST algorithm, described in the following: Altschul et al., 1990, J. Mol. Biol. [Journal of Molecular Biology] 215: 403-410; Altschul et al., 1997, Nucleic Acids Res. [Nucleic Acids Research] 25: 3389-3402; and Karin et al., 1993, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America] 90: 5873-5787. A particularly useful BLAST program is the WU-BLAST-2 program obtained from Altschul et al., 1996, Methods in Enzymology [Enzymological Methods] 266: 460-480. WU-BLAST-2 uses several search parameters, most of which are set to preset values. The adjustable parameters are set to the following values: overlap interval = 1, overlap score = 0.125, word threshold (T) = II. The HSP S and HSP S2 parameters are dynamic values ​​and are determined by the program itself based on the composition of a specific sequence and a specific database used to search for sequences of interest; however, these values ​​can be adjusted to improve sensitivity.

[0215] Another useful algorithm is the vacancy BLAST reported by Altschul et al., 1993, Nucl. Acids Res. [Nucleic Acid Research] 25: 3389-3402. The vacancy BLAST uses BLOSUM-62 instead of a score; the threshold parameter T is set to 9; a two-step method is used to trigger non-vacancy expansion, charging a cost of 10+k for a vacancy length of k; Xu is set to 16, and Xg is set to 40 (for the database search phase) and 67 (for the algorithm's output phase). Vacancy matching is triggered by a score corresponding to approximately 22 bits.

[0216] Generally, the amino acid homology, similarity, or identity between the various variant CDR or VH / VL sequences is at least 60% with the sequence described herein, and more typically at least 65% or 70%, more preferably at least 75% or 80%, and even more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, and almost 100%, with increased homology or identity being preferred. Similarly, the "percentage (%) nucleic acid sequence identity" relative to the nucleic acid sequence of the binding protein identified herein is defined as the percentage of nucleotide residues in the candidate sequence that are identical to the nucleotide residues in the coding sequence of the antibody construct. Specifically, the method utilizes the BLASTN module of WU-BLAST-2 with preset parameters, where the overlap interval and overlap fraction are set to 1 and 0.125, respectively.

[0217] Generally speaking, the nucleotide sequence encoding the various variant CDR or VH / VL sequence has at least 60% homology, similarity, or identity with the nucleotide sequence described herein, and more typically has at least 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, and preferably almost 100% increased homology or identity. Therefore, the "variant CDR" or "variant VH / VL region" has specified homology, similarity or identity with the parental CDR / VH / VL of the present invention, and shares biological functions, including but not limited to at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the specificity and / or activity of the parental CDR or VH / VL.

[0218] In one embodiment, the percentage of human lineage identity of the antibody construct according to the invention is ≥ 70% or ≥ 75%, more preferably ≥ 80% or ≥ 85%, even more preferably ≥ 90%, and most preferably ≥ 91%, ≥ 92%, ≥ 93%, ≥ 94%, ≥ 95%, or even ≥ 96%. Identity with human antibody lineage gene products is considered an important characteristic for reducing the risk of therapeutic proteins triggering drug-resistant immune responses in patients during treatment. Hwang and Foote ("Immunogenicity of engineered antibodies"; Methods 36 (2005) 3-10) demonstrated that reducing the non-human portion of the drug antibody construct leads to a reduced risk of inducing anti-drug antibodies in patients during treatment. Comparison of numerous clinically evaluated antibody drugs and corresponding immunogenicity data revealed the following trend: humanization of the V region of antibodies resulted in lower protein immunogenicity (average 5.1% of patients) compared to antibodies carrying the unchanged non-human V region (average 23.59% of patients). Therefore, V region-based protein therapeutics in antibody construct form require a high degree of identity with human sequences. For the purpose of determining germline identity, the Vector NTI software can be used to align the V region of VL with the amino acid sequences of human germline V and J regions (http: / / vbase.mrc-cpe.cam.ac.uk / ) and calculate the amino acid sequence (in percentage) by dividing the number of identical amino acid residues by the total number of amino acid residues in VL. The same applies to the VH region (http: / / vbase.mrc-cpe.cam.ac.uk / ), except that VH CDR3 can be excluded due to the high diversity of VH CDR3 and the lack of existing human germline VH CDR3 matching pairs. Recombinant technology can then be used to increase sequence identity with human antibody germline genes.

[0219] In another embodiment, the bispecific antibody construct of the present invention exhibits high monomer yield under standard research-scale conditions, such as in a standard two-step purification process. Preferably, the monomer yield of the antibody construct according to the present invention is ≥ 0.25 mg / L supernatant, more preferably ≥ 0.5 mg / L, even more preferably ≥ 1 mg / L, and most preferably ≥ 3 mg / L supernatant.

[0220] Similarly, the yield of the isotype of the dimeric antibody construct can be determined, and the monomer percentage (i.e., monomer:(monomer + dimer)) can be determined accordingly. The yield of the monomeric and dimeric antibody constructs and the calculated monomer percentage can be obtained, for example, in an SEC purification step of culture supernatant from standardized research-scale production in a rolling flask. In one embodiment, the monomer percentage of the antibody construct is ≥ 80%, more preferably ≥ 85%, even more preferably ≥ 90%, and most preferably ≥ 95%.

[0221] In one embodiment, the preferred plasma stability of the antibody construct (the ratio of EC50 with plasma to EC50 without plasma) is ≤ 5 or ≤ 4, more preferably ≤ 3.5 or ≤ 3, even more preferably ≤ 2.5 or ≤ 2, and most preferably ≤ 1.5 or ≤ 1. The plasma stability of the antibody construct can be tested by incubating the construct in human plasma at 37°C for 24 hours, followed by EC50 determination in a 51Cr release cytotoxicity assay. The effector cells in the cytotoxicity assay can be stimulated and enriched human CD8-positive T cells. The target cells can be, for example, CHO cells transfected with human target cell surface antigens. The effector cell to target cell (E:T) ratio can be selected as 10:1. The human plasma bank used for this purpose is derived from blood collected from healthy donors using an EDTA-coated syringe. Cellular components are removed by centrifugation, and the supernatant plasma phase is collected and subsequently pooled. As a control, the antibody construct is diluted immediately before the cytotoxicity assay in RPMI-1640 medium. Plasma stability was calculated as the ratio of EC50 (after plasma incubation) to EC50 (control).

[0222] Furthermore, it is preferable that the monomer-to-dimer conversion of the antibody construct of the present invention is low. The conversion can be measured under different conditions and analyzed by high-performance size exclusion chromatography. For example, the incubation of the monomeric isoform of the antibody construct can be carried out in an incubator for 7 days at 37°C and concentrations such as 100 μg / ml or 250 μg / ml. Under these conditions, it is preferable that the antibody construct of the present invention exhibits a dimer percentage of ≤ 5%, more preferably ≤ 4%, even more preferably ≤ 3%, even more preferably ≤ 2.5%, even more preferably ≤ 2%, even more preferably ≤ 1.5%, and most preferably ≤ 1% or ≤ 0.5% or even 0%.

[0223] Preferably, the bispecific antibody construct of the present invention exists with very low dimerization after multiple freeze / thaw cycles. For example, the antibody construct monomer is adjusted to a concentration of 250 μg / ml in, for example, a universal preparation buffer, and three freeze / thaw cycles are performed (freezing at -80°C for 30 min, followed by thawing at room temperature for 30 min), followed by high-performance SEC to determine the percentage of the original monomeric antibody construct that has been converted into a dimer antibody construct. Preferably, for example, after three freeze / thaw cycles, the dimerization percentage of the bispecific antibody construct is preferably ≤ 5%, more preferably ≤ 4%, even more preferably ≤ 3%, even more preferably ≤ 2.5%, even more preferably ≤ 2%, even more preferably ≤ 1.5%, and most preferably ≤ 1% or even ≤ 0.5%.

[0224] The bispecific antibody construct of the present invention preferably exhibits favorable thermal stability with an aggregation temperature ≥ 45°C or ≥ 50°C, more preferably ≥ 52°C or ≥ 54°C, even more preferably ≥ 56°C or ≥ 57°C, and most preferably ≥ 58°C or ≥ 59°C. The thermal stability parameter can be determined based on the antibody aggregation temperature as follows: An antibody solution with a concentration of 250 μg / ml is transferred to a disposable cuvette and placed in a dynamic light scattering (DLS) device. The sample is heated from 40°C to 70°C at a heating rate of 0.5°C / min, and the radius is measured continuously. The aggregation temperature of the antibody is calculated using the increase in the radius indicating the melting of the protein and aggregate.

[0225] Alternatively, the temperature melt profile can be determined by differential scanning calorimetry (DSC) to ascertain the intrinsic biophysical protein stability of the antibody construct. These experiments were performed using a VP-DSC system from MicroCal LLC (Northampton, MA, USA). Energy uptake of the sample containing the antibody construct was recorded from 20°C to 90°C compared to a sample containing only the preparation buffer. For example, the antibody construct was adjusted to a final concentration of 250 μg / ml in SEC running buffer. To record the corresponding melt profile, the sample temperature was gradually increased. Energy absorbance of the sample and the preparation buffer reference was recorded at each temperature T. The difference between the energy absorbance Cp (kcal / mol / °C) of the sample and the reference was plotted against the corresponding temperature. The melt temperature was defined as the temperature at which the first maximum energy absorbance occurred.

[0226] It is also envisioned that the target cell surface antigen xCD3 bispecific antibody construct of the present invention has a turbidity of ≤ 0.2, preferably ≤ 0.15, more preferably ≤ 0.12, even more preferably ≤ 0.1, and most preferably ≤ 0.08 (as measured by OD340 after concentrating the purified monomeric antibody construct to 2.5 mg / ml and incubating overnight).

[0227] In another embodiment, the antibody construct according to the invention is stable at physiological or slightly lower pH values, i.e., about pH 7.4 to 6.0. The greater the tolerance of the antibody construct at non-physiological pH values ​​(such as about pH 6.0), the higher the recovery rate of the antibody construct eluted from the ion exchange column relative to the total amount of loaded protein. The recovery rate of the antibody construct from an ion (e.g., cation) exchange column at about pH 6.0 is preferably ≥ 30%, more preferably ≥ 40%, more preferably ≥ 50%, even more preferably ≥ 60%, even more preferably ≥ 70%, even more preferably ≥ 80%, even more preferably ≥ 90%, even more preferably ≥ 95%, and most preferably ≥ 99%.

[0228] Furthermore, it is envisioned that the bispecific antibody construct of this invention exhibits therapeutic efficacy or antitumor activity. This can be evaluated, for example, in studies disclosed in the following examples of advanced human tumor xenograft models:

[0229] Those familiar with this technique know how to modify or adjust certain parameters of the study, such as the number of injected tumor cells, the injection site, the number of transplanted human T cells, the amount of bispecific antibody construct to be administered, and the timeline, while still obtaining meaningful and reproducible results. Preferably, the tumor growth inhibition T / C [%] is ≤ 70 or ≤ 60, more preferably ≤ 50 or ≤ 40, even more preferably ≤ 30 or ≤ 20, and most preferably ≤ 10 or ≤ 5 or even ≤ 2.5.

[0230] In a preferred embodiment of the antibody construct of the present invention, the antibody construct system is a single-chain antibody construct.

[0231] Furthermore, in a preferred embodiment of the antibody construct of the present invention, the third structural domain comprises, in the order from amino to carboxyl groups: Hinge-CH2-CH3-Connector-Hinge-CH2-CH3.

[0232] Furthermore, in one embodiment of the invention, one or more of the third domains, specifically the CH2 domain of each (two) polypeptide monomers, comprises an intradomain cysteine ​​disulfide bridge. As is known in the art, the term "cysteine ​​disulfide bridge" refers to a bridge having a general structure R The functional group of [–S–S–]R. This bond is also called an SS bond or a disulfide bridge, and is derived by coupling two thiol groups of a cysteine ​​residue. For the antibody construct of the present invention, it is particularly preferred that the cysteine ​​that forms the cysteine ​​disulfide bridge in the mature antibody construct be introduced into the amino acid sequence corresponding to the CH2 domains of 309 and 321 (Kabat number).

[0233] In one embodiment of the invention, the glycosylation site at Kabat position 314 of the CH2 domain is removed. Preferably, the glycosylation site is removed by N314X substitution, wherein X is any amino acid other than Q. The substitution is preferably N314G substitution. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (according to the Kabat position): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0234] Preferred features of the antibody construct of the present invention, compared to, for example, bispecific heterologous Fc antibody constructs known in the art, may particularly involve the introduction of the aforementioned modifications into the CH2 domain. Therefore, for the construct of the present invention, it is preferred that the CH2 domain of the third domain of the antibody construct of the present invention contains intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321 and / or the glycosylation site at Kabat position 314 is removed by the aforementioned N314X substitution, preferably by N314G substitution.

[0235] In another preferred embodiment of the present invention, the CH2 domain in the third domain of the antibody construct of the present invention contains intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321, and the glycosylation site at Kabat position 314 is removed by N314G substitution.

[0236] In one embodiment, the present invention provides an antibody construct, wherein: (182) The first domain contains two antibody variable domains, and the second domain contains two antibody variable domains; (ii) The first domain contains an antibody variable domain, and the second domain contains two antibody variable domains; (iii) The first domain contains two antibody variable domains, and the second domain contains one antibody variable domain; or (iv) The first domain contains an antibody variable domain, and the second domain contains an antibody variable domain.

[0237] Therefore, the first and second domains can each be binding domains containing two antibody variable domains (such as VH and VL domains). Examples of such binding domains containing two antibody variable domains have been described above and include, for example, the Fv fragment, scFv fragment, or Fab fragment described above. Alternatively, one or both of these binding domains may contain only a single variable domain. Examples of such single-domain binding domains have been described above and include, for example, nanoantibodies or single-variable-domain antibodies containing only one variable domain, which may be VHH, VH, or VL, binding antigens or epitopes independently of other V regions or domains.

[0238] In a preferred embodiment of the antibody construct of the present invention, the first and second domains are fused to the third domain via a peptide linker. The preferred peptide linker has been described above and is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or greater (e.g., 2 or 3). A particularly preferred linker for the fusion of the first and second domains with the third domain is depicted in SEQ ID No: 1.

[0239] In a preferred embodiment, the antibody construct of the present invention is characterized by comprising, in the order from amino to carboxyl groups: (a) First structural domain; (b) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187-189; I. Second structural domain; (d) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID NO: 187, 188, 189, 195, 196, 197 and 198; I. The first polypeptide monomer of the third structural domain; (f) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID Nos: 191, 192, 193, and 194; and (g) The second polypeptide monomer of the third structural domain.

[0240] In one aspect of the invention, the target cell surface antigen bound by the first domain is a tumor antigen, an antigen specific to an immune disorder, or a viral antigen. As used herein, the term "tumor antigen" can be understood as those antigens presented on tumor cells. These antigens can be presented on the cell surface having an extracellular portion, which is typically combined with transmembrane and cytoplasmic portions of the molecule. These antigens can sometimes be presented only by tumor cells and never by normal cells. Compared to normal cells, tumor antigens may be expressed only on tumor cells or may represent tumor-specific mutations. In this case, they are called tumor-specific antigens. More commonly, antigens are presented by both tumor cells and normal cells, and these are called tumor-associated antigens. Compared to normal cells, these tumor-associated antigens may be overexpressed, or because the structure of tumor tissue is less compact than that of normal tissue, making antibody binding on tumor cells accessible. Non-limiting examples of tumor antigens used herein include CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, CD20, CD70, BCMA, and PSMA.

[0241] In the context of this invention, other target cell surface antigens specific to immune disorders include, for example, TL1A and TNF-α. These targets are preferably addressed by the bispecific antibody construct of this invention, preferably a full-length antibody. In a very preferred embodiment, the anti-system of this invention is a heterologous IgG antibody.

[0242] In a preferred embodiment of the antibody construct of the present invention, the tumor antigen is selected from the group consisting of: CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, CD20, CD70, BCMA and PSMA.

[0243] In one aspect of the invention, the antibody construct comprises, in the order of amino groups to carboxyl groups: (a) A first domain having an amino acid sequence selected from the group consisting of: SEQ ID No: 7, 8, 17, 27, 28, 37, 38, 39, 40, 41, 48, 49, 50, 51, 52, 59, 60, 61, 62, 63, 64, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 89, 90, 91, 92, 93, 100, 101, 102, 103, 104, 113, 114, 121, 122 123, 124, 125, 131, 132, 133, 134, 135, 136, 143, 144, 145, 146, 147, 148, 149, 150, 151, 158, 159, 160, 161, 162, 163, 164, 165, 166, 173, 174, 175, 176, 177, 178, 179, 180, 181 (b) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187-189; I. A second structural domain having an amino acid sequence selected from the group consisting of the following SEQ ID Nos: SEQ ID Nos of WO 2008 / 119567: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 or SEQ ID NO: 202; (d) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187, 188, 189, 195, 196, 197 and 198; I. A first polypeptide monomer with a third structural domain, the first polypeptide monomer having a polypeptide sequence selected from the group consisting of SEQ ID No: 17-24 of WO2017 / 134140; (f) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID Nos: 191, 192, 193, and 194; and (g) A second polypeptide monomer with a third domain having a polypeptide sequence selected from the group consisting of WO2017 / 134140, SEQ ID No: 17-24.

[0244] In one aspect, the bispecific antibody construct of the present invention is characterized by having an amino acid sequence selected from the group consisting of the following groups and targeting the corresponding target cell surface antigens: (a) SEQ ID No: 27, 28, 37 to 41; CD33 (b) Each of SEQ ID Nos. 48 to 52; EGFRvIII (c) Each of SEQ ID No: 59 to 64; MSLN (d) Each of the CDH19 in SEQ ID No: 71 to 82 (e) Each of DLL3 in SEQ ID No: 100 to 104 (f) SEQ ID No: 7, 8, 17, 113 and 114 CD19 (g) Each of the FLT3 in SEQ ID No: 89 to 93 (h) Each of the CDH3 in SEQ ID No: 121 to 125 (i) Each of the BCMAs in SEQ ID No: 132 to 136 and (j) Each of the PSMAs in SEQ ID No: 143 to 151, 158 to 166 and 173 to 181 (k) Each of the MUC17 in SEQ ID No: 212 and 213 (l) Each of CLDN18 in SEQ ID No: 223, 224, 225, 236 and 237 (m) Each of CD70 in SEQ ID No: 247 and 248

[0245] The present invention further provides a polynucleotide / nucleic acid molecule encoding the antibody construct of the present invention. A polynucleotide is a biopolymer composed of 13 or more nucleotide monomers covalently bonded in a chain. DNA (such as cDNA) and RNA (such as mRNA) are examples of polynucleotides with different biological functions. Nucleotides are organic molecules that act as monomers or subunits of nucleic acid molecules such as DNA or RNA. Nucleic acid molecules or polynucleotides can be double-stranded or single-stranded, linear or circular. It is preferably contained in a vector, which is preferably contained in a host cell. The host cell, for example, can express the antibody construct after transformation or transfection with the vector or polynucleotide of the present invention. For this purpose, the polynucleotide or nucleic acid molecule is operatively linked to a control sequence.

[0246] The genetic code is a set of rules that translates the information encoded within genetic material (nucleic acids) into proteins. Biological decoding in living cells is accomplished by ribosomes linking amino acids in a sequence specified by mRNA, using tRNA molecules to carry the amino acids, and reading out three nucleotides of mRNA at a time. This code defines how the sequence of these nucleotide triplets (called codons) specifies which amino acid will be added next during protein synthesis. With some exceptions, trinucleotide codons in nucleic acid sequences specify a single amino acid. Because the vast majority of genes are encoded using the exact same codon, this particular codon is often called the canonical or standard genetic code. While the genetic code determines the protein sequence of a given coding region, other genomic regions can influence when and where those proteins are produced.

[0247] Furthermore, this invention provides a vector containing the polynucleotide / nucleic acid molecule of this invention. The vector system is a nucleic acid molecule used as a medium for transferring (foreign) genetic material into cells. The term "vector" encompasses, but is not limited to, plasmids, viruses, granules, and artificial chromosomes. Generally, engineered vectors contain origins of replication, multi-strain sites, and selectivity markers. The vector itself is typically a nucleotide sequence, which is usually a DNA sequence containing an insert (transgenic gene) and a larger sequence that acts as the "backbone" of the vector. In addition to the transgenic insert and backbone, modern vectors may encompass other features: promoters, genetic markers, antibiotic resistance, reporter genes, target sequences, and protein purification tags. Vectors called expression vectors (expression constructs) are particularly used to express transgenes in target cells and typically have control sequences.

[0248] The term "control sequence" refers to the DNA sequence necessary for the expression of an operable linking coding sequence in a particular host organism. For example, control sequences applicable to prokaryotes include promoters, optional operon sequences, and ribosome-binding sides. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

[0249] When a nucleic acid is positioned in a functional relationship with another nucleic acid sequence, the nucleic acid is "operably linked." For example, if the DNA of a pre-sequence or secretory leader sequence is to act as a pre-protein involved in polypeptide secretion, then the DNA of the pre-sequence or secretory leader sequence is operably linked to the DNA of the polypeptide; if a promoter or enhancer affects the transcription of a coding sequence, then the promoter or enhancer is operably linked to that sequence; or if the ribosome-binding side is positioned to facilitate translation, then the ribosome-binding side is operably linked to the coding sequence. Generally, "operably linked" means that the linked DNA sequences are contiguous, and in the case of a secretory leader sequence, they are contiguous and in the reading phase. However, enhancers do not need to be contiguous. Ligation is accomplished by joining at a convenient restriction site. If such a site is not available, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

[0250] "Transfection" is the intentional introduction of nucleic acid molecules or polynucleotides (including vectors) into target cells. This term is primarily used for non-viral methods in eukaryotic cells. Transduction is generally used to describe virus-mediated transfer of nucleic acid molecules or polynucleotides. Transfection of animal cells typically involves opening transient pores or "holes" in the cell membrane to allow for the uptake of substances. Transfection can be performed using calcium phosphate, by electroporation, by cell extrusion, or by mixing cationic lipids with substances to create liposomes (which fuse with the cell membrane and store their cargo inside).

[0251] The term "transformation" is used to describe the transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and into non-animal eukaryotic cells (including non-viral ones in plant cells). Therefore, transformation is a genetic alteration in bacteria or non-animal eukaryotic cells resulting from the direct uptake of exogenous genetic material (nucleic acid molecules) from their surroundings via one or more cell membranes and subsequent incorporation. Transformation can be achieved artificially. For transformation to occur, the cell or bacteria must be in a competent state, which may occur as a time-limited response to environmental conditions such as starvation and cell density.

[0252] Furthermore, this invention provides a host cell that is transformed or transfected with the polynucleotide / nucleic acid molecule or vector of this invention. As used herein, the terms "host cell" or "recipient cell" are intended to include any single cell or cell culture that may be or has been a receptor for a vector, exogenous nucleic acid molecule, or polynucleotide encoding an antibody construct of this invention, and / or a receptor for the antibody construct itself. The corresponding substance is introduced into the cell by means of transformation, transfection, etc. The term "host cell" is also intended to include the progeny or potential progeny of a single cell. Because certain changes may occur in subsequent generations due to natural, accidental, or intentional mutations or due to environmental influences, such progeny may not actually be completely identical to the parent cell (in morphology, genome, or all DNA complement), but are still included within the scope of the terminology used herein. Suitable host cells include prokaryotic or eukaryotic cells, and also include, but are not limited to, bacterial, yeast, fungal, plant, and animal cells, such as insect cells and mammalian cells, such as mice, rats, macaques, or humans.

[0253] The antibody construct of this invention can be produced in bacteria. After expression, the antibody construct of this invention is separated from the *E. coli* cell paste by soluble fractionation and can be purified by, for example, affinity chromatography and / or size exclusion. Final purification can be performed similarly to methods used for purifying antibodies, for example, expressed in CHO cells.

[0254] Besides prokaryotes, eukaryotic microorganisms (such as filamentous fungi or yeasts) are suitable colonization or expression hosts for the antibody constructs of this invention. Saccharomyces cerevisiae or common baker's yeast are among the most commonly used lower eukaryotic host microorganisms. However, many other genera, species, and strains are generally available and can be used in this paper, such as *Schizosaccharomyces pombe*, hosts of *Kluyveromyce* such as *Kluyveromyce* lactis, *Kluyveromyce* fragilis (ATCC 12424), *Kluyveromyce* bulgaricus (ATCC 16045), *Kluyveromyce* wickeramii (ATCC 24178), *Kluyveromyce* waltii (ATCC 56500), *Kluyveromyce* drosophilarum (ATCC 36906), *Kluyveromyce* thermotolerans, and *Kluyveromyce* marxianus; *Yersinia* (EP 402 226); *Pichia pastoris* (EP 183 070); *Candida*; *Trichoderma reesei* (EP 402 226); *Trichoderma* var. *reesei* (EP 402 226); *Pichia pastoris* (EP 183 070); *Candida* var. *reesei*; *Trichoderma reesei* var. *reesei* (EP 402 226); *Trichoderma* var. *reesei* (EP 402 226); *Pichia pastoris* (EP 402 226); *Candida* var. *reesei*; *Trichoderma reesei* var. *reesei* (EP 402 226); *Trichoderma ... 244 234); Neurospora crassa; Schwanniomyces, such as Schwanniomyces occidentalis; and filamentous fungi, such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts, such as Aspergillus nidulans and Aspergillus niger.

[0255] Suitable host cells for expressing the glycosylated antibody constructs of this invention are derived from multicellular organisms. Examples of invertebrate cells include plant cells and insect cells. Numerous baculovirus strains and variants from hosts such as the fall armyworm (Spodoptera frugiperda) (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melanogaster (fruit fly), and Bombyx mori mori (silkworm), as well as corresponding permitted insect host cells, have been identified. Many publicly available viral strains for transfection, such as the L-1 variant of the alfalfa silver-striped armyworm (Autographa californica) NPV and the Bm-5 strain of the silkworm NPV, can be used as the viruses described herein, particularly for transfecting fall armyworm cells.

[0256] Plant cell cultures of cotton, maize, potato, soybean, petunia, tomato, Arabidopsis, and tobacco can also be used as hosts. Selective and expression vector systems for producing proteins in plant cell cultures are known to those skilled in the art. See, for example, Hiatt et al., Nature (1989) 342: 76-78; Owen et al. (1992) Bio / Technology 10: 790-794; Artsaenko et al. (1995) The Plant Journal 8: 745-750; and Fecker et al. (1996) Plant Mol Biol 32: 979-986.

[0257] However, the greatest interest is in vertebrate cells, and the propagation of vertebrate cells in cultures (tissue cultures) has become a routine procedure. Examples of useful mammalian host cell lines include: monkey kidney CV1 line transformed from SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293 cells or subselected 293 cells for growth in suspension culture, Graham et al., J. Gen Virol. [Journal of Genetic Virology] 36: 59 (1977)); juvenile hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences] 77: 4216 (1980)); mouse saturated hamster cells (TM4, Mather, Biol. Reprod. [Reproductive Biology] 23: 243-251 (1980)); monkey kidney cells (CVI ATCC CCL 70); and African green monkey kidney cells (VERO-76, ATCC). CRL1587); human cervical cancer cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); Buffalo rat hepatocytes (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human hepatocytes (Hep G2, 1413 8065); mouse mammary tumor (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals N. Y Acad. Sci. [Annals of the New York Academy of Sciences] (1982) 383: 44-68); MRC 5 cells; FS4 cells; and human hepatocellular carcinoma cell line (Hep G2).

[0258] In another embodiment, the present invention provides a method for producing an antibody construct of the present invention, the method comprising culturing a host cell of the present invention under conditions that allow the antibody construct of the present invention to express the antibody construct of the present invention and recovering the resulting antibody construct from the culture.

[0259] As used herein, the term "culture" means the in vitro maintenance, differentiation, growth, proliferation, and / or reproduction of cells in a culture medium under suitable conditions. The term "performance" includes any step involved in generating the antibody constructs of the present invention, including but not limited to transcription, post-transcriptional modification, translation, post-translational modification, and secretion.

[0260] When using recombinant technology, antibody constructs can be generated intracellularly in the periplasmic space or secreted directly into the culture medium. If the antibody construct is generated intracellularly, particulate debris from the host cell or lysed fragments is removed as a first step, for example, by centrifugation or ultrafiltration. Carter et al., Bio / Technology 10: 163-167 (1992) describe a procedure for isolating antibodies secreted into the periplasmic space of *E. coli*. Briefly, the cell paste is thawed in the presence of sodium acetate (pH 3.5), EDTA, and benzyl sulfonate (PMSF) over approximately 30 min. Cell debris can be removed by centrifugation. In cases where antibodies are secreted into the culture medium, the supernatant from such performance systems is typically concentrated first using a commercially available protein concentrator, such as an Amicon or Millipore Pellicon ultrafiltration unit. Any of the foregoing steps may include protease inhibitors (such as PMSF) to inhibit proteolysis, and may include antibiotics to prevent the growth of foreign contaminants.

[0261] The antibody constructs of the present invention prepared from host cells can be recovered or purified using, for example, hydroxyapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography. Depending on the antibody to be recovered, other techniques for protein purification may also be used, such as fractionation on ion-exchange columns, ethanol precipitation, reversed-phase HPLC, chromatography on silica, chromatography on heparin, SEPHAROSE™ chromatography on anion or cation exchange resins (such as polyaspartic acid columns), chromatographic focusing, SDS-PAGE, and ammonium sulfate precipitation. In cases where the antibody constructs of the present invention contain a CH3 domain, Bakerbond ABX resin (JT Baker, Phillipsburg, NJ) can be used for purification.

[0262] Affinity chromatography is a superior purification technique. The most common matrix to which the affinity ligand is attached is agarose, but other matrices are also available. Mechanically stable matrices such as controllably porous glass or poly(divinyl styrene)benzene allow for faster flow rates and shorter processing times than achievable with agarose.

[0263] Furthermore, the present invention provides a pharmaceutical composition comprising the antibody construct of the present invention or an antibody construct generated by the method of the present invention. Preferably, the homogeneity of the antibody construct is ≥ 80%, more preferably ≥ 81%, ≥ 82%, ≥ 83%, ≥ 84% or ≥ 85%, further preferably ≥ 86%, ≥ 87%, ≥ 88%, ≥ 89% or ≥ 90%, even more preferably ≥ 91%, ≥ 92%, ≥ 93%, ≥ 94% or ≥ 95%, and most preferably ≥ 96%, ≥ 97%, ≥ 98% or ≥ 99%.

[0264] As used herein, the term "pharmaceutical composition" refers to a composition suitable for administration to a patient, preferably a human patient. Particularly preferred pharmaceutical compositions of the present invention comprise, preferably, a therapeutically effective amount of one or more antibody constructs of the present invention. Preferably, the pharmaceutical composition further comprises a suitable formulation of one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives, and / or adjuvants. The acceptable components of the composition are preferably non-toxic to the recipient at the doses and concentrations used. Pharmaceutical compositions of the present invention include, but are not limited to, liquid, frozen, and lyophilized compositions.

[0265] The compositions of this invention may contain pharmaceutically acceptable carriers. Generally, as used herein, "pharmaceutically acceptable carrier" means any and all aqueous and non-aqueous solutions, sterile solutions, solvents, buffers (e.g., phosphate-buffered saline (PBS) solutions), water, suspensions, emulsions (such as oil / water emulsions), various types of wetting agents, liposomes, dispersion media, and coatings that are compatible with drug administration, particularly parenteral administration. The use of such media and agents in pharmaceutical compositions is well known in the art, and compositions containing such carriers can be formulated by well-known conventional methods.

[0266] Some embodiments provide pharmaceutical compositions comprising the antibody constructs of the present invention and one or more additional excipients, such as those illustratively described in this section and elsewhere herein. In this regard, excipients can be used in the present invention for a variety of purposes, such as adjusting the physical, chemical, or biological properties of formulations, such as adjusting viscosity and / or the methods of the present invention to improve efficacy and / or stabilize such formulations and methods to prevent degradation and spoilage due to stresses occurring, for example, during manufacturing, transport, storage, pre-use preparation, dosing, and post-use processes.

[0267] In some embodiments, the pharmaceutical composition may contain formulations intended to alter, maintain, or preserve the following aspects of the composition: for example, pH, osmotic pressure, viscosity, transparency, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or permeability (see REMINGTON'S PHARMACEUTICAL SCIENCES, 18th edition, edited by AR Genrmo, 1990, Mack Publishing Company). In such embodiments, suitable formulations may include, but are not limited to: • Amino acids, such as glycine, alanine, glutamine, aspartic acid, threonine, proline, 2-phenylalanine, including charged amino acids, preferably lysine, lysine acetate, arginine, glutamine salts and / or histidine. • Antimicrobial agents, such as antibacterial and antifungal agents • Antioxidants, such as ascorbic acid, methionine, sodium sulfite, or sodium bisulfite; • Buffers, buffer systems, and buffers used to maintain components at physiological pH or slightly lower; examples of buffers include borates, bicarbonates, Tris-HCl, citrates, phosphates or other organic acids, succinates, phosphates, and histidines; for example, Tris buffer with a pH of approximately 7.0–8.5; • Non-aqueous solvents, such as propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate; • Aqueous carriers include water, alcohol / aqueous solutions, emulsions, or suspensions, including brine and buffered media; • Biodegradable polymers, such as polyesters; • Expanding agents, such as mannitol or glycine; • Chelating agents, such as ethylenediaminetetraacetic acid (EDTA); • Isotonic agents and absorption delay agents; • Mixing agents (such as caffeine, polyvinylpyrrolidone, β-cyclodextrin, or hydroxypropyl-β-cyclodextrin); • Filler; • Monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose, or dextrin); the carbohydrates can be non-reducing sugars, preferably trehalose, sucrose, octasulfate, sorbitol, or xylitol; • (Low molecular weight) proteins, peptides, or protein carriers, such as human or bovine serum albumin, gelatin, or immunoglobulins, preferably of human origin; • Coloring agents and flavoring agents; • Sulfur-containing reducing agents, such as glutathione, lipoic acid, sodium thioglycolate, thioglycerol, [α]-monothioglycerol, and sodium thiosulfate. • Diluent; • Emulsifiers; • Hydrophilic polymers, such as polyvinylpyrrolidone; • Salt formation against counterions, such as sodium; • Preservatives, such as antimicrobial agents, antioxidants, chelating agents, inert gases, etc.; examples include: benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide; • Metal complexes, such as Zn-protein complexes; • Solvents and co-solvents (such as glycerol, propylene glycol, or polyethylene glycol); Sugars and sugar alcohols, such as trehalose, sucrose, octasulfate, mannitol, sorbitol or xylitol, stachyose, mannose, sorbitol, xylose, ribose, myoinisitose, galactose, lactitol, ribitol, myoinisitol, galactitol, glycerol, cyclic polyols (e.g., inositol), polyethylene glycol; and polyols; • Suspension agent; • Surfactants or wetting agents (such as pluronics, PEG, dehydrated sorbitan, polysorbate esters (such as polysorbate 20, polysorbate ester), tritium nuclei, glycerol, lecithin, cholesterol, tyloxacin); the surfactant may be a detergent, preferably with a molecular weight > 1.2 KD, and / or a polyether, preferably with a molecular weight > 3 KD; preferred non-limiting examples of detergents are Tween 20, Tween 40, Tween 60, Tween 80 and Tween 85; preferred non-limiting examples of polyethers are PEG 3000, PEG 3350, PEG 4000 and PEG 5000; • Stabilizers, such as sucrose or sorbitol; • Tensile enhancers (such as alkali metal halides, preferably sodium chloride or potassium chloride, mannitol or sorbitol); • Parenteral delivery media, including sodium chloride solution, Ringer's dextran, dextran and sodium chloride, lactated Ringer's solution, or non-volatile oils; • Intravenous delivery mediators, including fluid and nutritional supplements, and electrolyte supplements (such as those based on Ringer's dextran).

[0268] It will be apparent to those skilled in the art that, for example, different components of a pharmaceutical composition (e.g., those listed above) can have different effects, and amino acids can act as buffers, stabilizers, and / or antioxidants; mannitol can act as a build-up agent and / or a tension enhancer; sodium chloride can act as a delivery medium and / or a tension enhancer; and so on.

[0269] In addition to the polypeptides of the present invention as defined herein, it is envisioned that the compositions of the present invention may contain other bioactive agents, depending on the intended use of the compositions. Such agents may be drugs acting on the gastrointestinal system, drugs acting as cell inhibitors, drugs preventing hyperuricemia, drugs suppressing immune responses (e.g., corticosteroids), drugs modulating inflammatory responses, drugs acting on the circulatory system, and / or agents known in the art such as cytokines. It is also envisioned that the antibody constructs of the present invention be used in co-therapies, i.e., in combination with another anticancer drug.

[0270] In some embodiments, the optimal pharmaceutical composition will be determined by those skilled in the art based on, for example, the intended route of administration, delivery method, and desired dosage. See, for example, Remington's Pharmaceutical Sciences, ibid. In some embodiments, such compositions can affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the antibody construct of the present invention. In some embodiments, the primary medium or carrier in the pharmaceutical composition may be aqueous or non-aqueous in nature. For example, suitable mediums or carriers may be water for injection, saline solution, or artificial cerebrospinal fluid, possibly supplemented with other substances common in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are other exemplary mediums. In some embodiments, the antibody constructs of the present invention can be prepared for storage by mixing selected components having the desired purity with optional formulations (REMINGTON'S PHARMACEUTICAL SCIENCES, ibid.) in the form of lyophilized cakes or aqueous solutions. Furthermore, in some embodiments, the antibody constructs of the present invention can be formulated into lyophilized products using suitable excipients (such as sucrose).

[0271] When considering parenteral administration, the therapeutic components of the present invention can be provided as a pyrogen-free, parenterally acceptable aqueous solution containing the desired antibody construct of the present invention in a pharmaceutically acceptable medium. A particularly suitable medium for parenteral injection is sterile distilled water, in which the antibody construct of the present invention is formulated into a suitably preserved sterile isotonic solution. In some embodiments, the preparation may involve formulating the desired molecule with an agent that can provide controlled or sustained release of a product (such as injectable microspheres, bio-erosive particles, polymeric compounds such as polylactic acid or polyglycolic acid), beads, or liposomes. In some embodiments, hyaluronic acid, which has the effect of promoting circulation duration, may also be used. In some embodiments, an implantable drug delivery device may be used to introduce the desired antibody construct.

[0272] Other pharmaceutical compositions will be apparent to those skilled in the art, including formulations involving the formulation of the antibody constructs of the present invention into sustained or controlled delivery / release formulations. Techniques for formulating various other sustained or controlled delivery methods (such as liposome carriers, bio-erectible microparticles or porous beads and reservoir injections) are also known to those skilled in the art. See, for example, International Patent Application No. PCT / US93 / 00829, which describes the controlled release of porous polymer microparticles for delivering pharmaceutical compositions. Sustained-release formulations may comprise a semi-permeable polymer matrix in the form of a molded article (e.g., a membrane or microcapsule). Sustained-release matrices may include polyesters, hydrogels, polylactide (as disclosed in U.S. Patent No. 3,773,919 and European Patent Application Publication No. EP 058481), copolymers of L-glutamic acid and γ-L-glutamic acid ethyl ester (Sidman et al., 1983, Biopolymers 2: 547-556), poly(2-hydroxyethyl-methacrylate) (Langer et al., 1981, J. Biomed. Mater. Res. 15: 167-277 and Langer, 1982, Chem. Tech. 12: 98-105), ethylene vinyl acetate (Langer et al., 1981, ibid.), or poly-D(-)-3-hydroxybutyric acid (European Patent Application Publication No. EP 133,988). Sustained-release components may also include liposomes that can be prepared by any of several methods known in the art. See, for example, Eppstein et al., 1985, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America] 82: 3688-3692; European Patent Application Publication Nos. EP 036,676; EP 088,046 and EP 143,949.

[0273] Antibody constructs can also be embedded in microcapsules (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly(methyl methacrylate) microcapsules, respectively) prepared by coarse-grained drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules) or in coarse-drop emulsions. Such techniques are disclosed in Remington's Pharmaceutical Sciences, 16th edition, edited by Oslo, A. (1980).

[0274] Pharmaceutical compositions intended for internal administration are typically provided as sterile preparations. Sterilization can be achieved through filtration using a sterile membrane filter. When the composition is lyophilized, sterilization can be performed using this method before or after lyophilization and reconstitution. Compositions intended for parenteral administration can be stored in lyophilized form or in solution. Parenteral compositions are typically placed in containers with sterile infusion ports (e.g., intravenous solution bags or vials with stoppers that can be punctured by a hypodermic needle).

[0275] Another aspect of the invention includes self-buffered antibody construct formulations of the invention, which can be used as pharmaceutical components, as described in International Patent Application WO 06138181A2 (PCT / US2006 / 022599). Useful protein stabilizing and formulation materials and methods in this regard are available in a wide variety of descriptions, such as Arakawa et al., "Solvent interactions in pharmaceutical formulations," Pharm Res. 8(3): 285-91 (1991); Kendrick et al., "Physical stabilization of proteins in aqueous solution," in RATIONAL DESIGN OF STABLE PROTEIN FORMULATIONS: THEORY AND PRACTICE, edited by Carpenter and Manning in Pharmaceutical Biotechnology 13: 61-84 (2002); and Randolph et al., "Surfactant-protein interactions," Pharm Biotechnol. 13: 159-75. (2002), with particular reference to the relevant sections concerning excipients and methods for self-buffered protein formulations according to the present invention, especially concerning protein pharmaceutical products and methods for veterinary and / or human medical use.

[0276] According to certain embodiments of the invention, salts can be used to, for example, adjust the ionic strength and / or isotonicity of the formulation and / or improve the solubility and / or physical stability of proteins or other components of the composition according to the invention. It is well known that ions can stabilize the native state of proteins by binding to charged residues on the surface of proteins and by masking charged and polar groups in the protein, thereby reducing the strength of their electrostatic interactions, attractive and repulsive interactions. Ions can also stabilize the denatured state of proteins by specifically binding to denatured peptide bonds (--CONHs). Furthermore, ionic interactions with charged and polar groups in proteins can reduce intermolecular electrostatic interactions, thereby preventing or reducing protein aggregation and insolubility.

[0277] Ionic species have significantly different effects on proteins. Various classification and rating systems have been developed for ions that can be used to formulate pharmaceutical compositions according to the present invention and their effects on proteins. One example is the Hofmeister series, which rates ions and polar nonionic solutes based on their effect on the conformational stability of proteins in solution. Stable solutes are referred to as "lyophilic." Unstable solutes are referred to as "agnostic." High concentrations of lyophilic agents (e.g., >1 molar ammonium sulfate) are typically used to precipitate proteins from solution ("salting out"). Agnostic agents are typically used to denature and / or dissolve proteins ("salting in"). The relative effectiveness of ions in "salting in" and "salting out" defines their position in the Hofmeister series.

[0278] According to various embodiments of the present invention, free amino acids can be used in the antibody construct formulations of the present invention as build-up agents, stabilizers, antioxidants, and for other standard uses. Lysine, proline, serine, and alanine can be used to stabilize proteins in the formulations. Glycine can be used for lyophilization to ensure correct cake structure and properties. Arginine can be used to inhibit protein aggregation in both liquid and lyophilized formulations. Methionine can be used as an antioxidant.

[0279] Polyols include sugars such as mannitol, sucrose, and sorbitol, as well as polyols such as glycerol and propylene glycol, and for the purposes of this discussion, polyethylene glycol (PEG) and related substances. Polyols are lyophilic. They are useful stabilizers in both liquid formulations and lyophilized formulations to protect proteins from physical and chemical degradation processes. Polyols can also be used to adjust the stress of formulations. Mannitol is a useful polyol in preferred embodiments of the invention, and it is commonly used to ensure the structural stability of the cake in lyophilized formulations. It ensures the structural stability of the cake. It is often used in conjunction with lyophilization protectants (e.g., sucrose). Sorbitol and sucrose are preferred agents for adjusting stress and acting as stabilizers to prevent freeze-thaw stress during transport or to prevent the formation of clumps during manufacturing. Reducing sugars (containing free aldehyde or ketone groups), such as glucose and lactose, can yophilize lysine and arginine residues on the surface. Therefore, they are generally not preferred polyols for use according to the invention. Furthermore, sugars that form such reactive substances are not preferred polyols in this invention, such as sucrose, which hydrolyzes under acidic conditions to fructose and glucose, thus resulting in glycosylation. PEG can be used to stabilize proteins and as a cryoprotectant, and can be used in this invention.

[0280] Embodiments of the antibody construct formulations of the present invention further include surfactants. Protein molecules can readily adsorb onto surfaces and denature, subsequently accumulating at air-liquid, solid-liquid, and liquid-liquid interfaces. These effects are generally inversely proportional to protein concentration. These detrimental interactions are generally inversely proportional to protein concentration and are typically exacerbated by physical oscillations, such as those generated during product transport and handling. Surfactants are conventionally used to prevent, minimize, or reduce surface adsorption. Surfactants useful in this regard include polysorbate 20, polysorbate 80, other fatty acid esters of dehydrated sorbitol polyethoxylates, and poloxamer 188. Surfactants are also commonly used to control protein conformational stability. The surfactants used in this regard are protein-specific because any given surfactant typically stabilizes some proteins and destabilizes others.

[0281] Polysorbates are readily oxidatively degraded and typically contain sufficient amounts of peroxides to cause oxidation of protein residue side chains, particularly methionine. Therefore, polysorbates should be used with caution and at their lowest effective concentration. In this respect, polysorbates exemplify the general rule that excipients should be used at their lowest effective concentration.

[0282] Embodiments of the antibody construct formulations of the present invention further include one or more antioxidants. Harmful oxidation of proteins in pharmaceutical formulations can be prevented to some extent by maintaining appropriate levels of ambient oxygen and temperature and avoiding exposure to light. Antioxidant excipients can also be used to prevent oxidative degradation of proteins. In this regard, useful antioxidants are reducing agents, oxygen / free radical scavengers, and chelating agents. The antioxidants used in the therapeutic protein formulations according to the present invention are preferably water-soluble and maintain their activity throughout the shelf life of the product. In this regard, EDTA is a preferred antioxidant according to the present invention. Antioxidants can damage proteins. For example, reducing agents, such as glutathione in particular, can break intramolecular disulfide bonds. Therefore, the antioxidants used in the present invention are particularly selected to eliminate or sufficiently reduce the possibility of themselves damaging proteins in the formulation.

[0283] The formulations according to the present invention may contain metal ions, which are protein cofactors and are essential for the formation of protein coordination complexes, such as zinc, which is essential for the formation of certain insulin suspensions. Metal ions can also inhibit some protein degradation processes. However, metal ions also catalyze physical and chemical processes that degrade proteins. Magnesium ions (10-120 mM) can be used to inhibit the isomerization of aspartic acid to isofaradino. Ca²⁺ ions (up to 100 mM) can increase the stability of human deoxyribonuclease. However, Mg²⁺, Mn²⁺, and Zn²⁺ can destabilize recombinant human deoxyribonuclease (rhDNase). Similarly, Ca²⁺ and Sr²⁺ can stabilize factor VIII, which can be destabilized by Mg²⁺, Mn²⁺, Zn²⁺, Cu²⁺, and Fe²⁺, and its aggregation can be increased by Al³⁺ ions.

[0284] Embodiments of the antibody construct formulations of this invention further include one or more preservatives. Preservatives are essential when the development involves extracting multiple doses of parenteral formulations from the same container more than once. Their primary function is to inhibit microbial growth and ensure the sterility of the product throughout its shelf life or use period. Commonly used preservatives include benzyl alcohol, phenol, and m-cresol. Although preservatives have a long history of use in the parenteral administration of small molecules, the development of protein formulations containing preservatives can be challenging. Preservatives almost always have an unstable effect on proteins (aggregation), and this has become a major factor limiting their use in multi-dose protein formulations. To date, most protein drugs have been formulated for single-use only. However, when multi-dose formulations are possible, they offer the added advantages of patient convenience and increased marketability. A good example is human growth hormone (hGH), where the development of preservative formulations has led to the commercialization of more convenient, reusable injection pens. At least four such pen devices containing hGH preservative formulations are currently commercially available. Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech), and Genotropin (lyophilized, two-chamber cartridge, Pharmacia & Upjohn) contain phenol, while Somatrope (Eli Lilly) is formulated with m-cresol. Several aspects need to be considered during the formulation and development of preservative dosage forms. The effective preservative concentration in the drug product must be optimized. This requires testing the given preservative in the dosage form at concentration ranges that impart antimicrobial efficacy without compromising protein stability.

[0285] As can be expected, developing liquid formulations containing preservatives is more challenging than developing lyophilized formulations. Lyophilized products can be freeze-dried without preservatives and reconstituted with a preservative-containing diluent upon use. This shortens the time the preservative is in contact with the protein, thus significantly minimizing the associated stability risks. In the case of liquid formulations, the effectiveness and stability of the preservative should be maintained throughout the product's shelf life (approximately 18 to 24 months). It is important to note that the effectiveness of the preservative should be demonstrated in the final formulation containing the active pharmaceutical ingredient and all excipient components.

[0286] The antibody constructs disclosed herein can also be formulated as immunoliposomes. A "liposome" is a small vesicle composed of various types of lipids, phospholipids, and / or surfactants, which can be used to deliver drugs to mammals. The components of a liposome are typically arranged in a bilayer, similar to the lipid arrangement of a biological membrane. Lipid systems containing antibody constructs are prepared by methods known in the art, such as Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688 (1985); Hwang et al., Proc. Natl. Acad. Sci. USA, 77: 4030 (1980); U.S. Patent Nos. 4,485,045 and 4,544,545; and WO 97 / 38731. Liposomes with extended circulation times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by reverse-phase evaporation using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter with defined pore sizes to produce liposomes of the desired diameter. The Fab' fragment of the antibody construct of the present invention can be conjugated to liposomes via a disulfide exchange reaction, as described in Martin et al. J. Biol. Chem. 257: 286-288 (1982). Chemotherapy agents may be included in the liposomes as needed. See Gabizon et al. J. National Cancer Institute 81 (19) 1484 (1989).

[0287] Once a drug formulation is prepared, it can be stored in sterile vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations can be stored in ready-to-use form or in a form reconstituted before dosing (e.g., lyophilized form).

[0288] The bioactivity of the pharmaceutical composition as defined herein can be determined, for example, by cytotoxicity assays, as described in the following examples, WO 99 / 54440, or by Schlereth et al. (Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy] 20 (2005), 1-12). As used herein, “efficacy” or “in vivo efficacy” refers to the response to treatment with the pharmaceutical composition of the present invention using, for example, standardized NCI response criteria. The success or in vivo efficacy of a therapy using the pharmaceutical composition of the present invention refers to the effectiveness of the composition for its intended use, i.e., the composition’s ability to produce its desired effect, i.e., the depletion of pathological cells (e.g., tumor cells). In vivo efficacy can be monitored by established standard methods for corresponding disease entities, including but not limited to white blood cell counts, differential cytometry, fluorescence-activated cell sorting, and bone marrow aspiration. In addition, various disease-specific clinical chemistry parameters and other established standard methods can be used. In addition, computer-assisted computed tomography, X-ray, and magnetic resonance imaging (MRI) can be used (e.g., for response assessment based on National Cancer Institute standards [Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM, Lister TA, Vose J, Grillo-Lopez A, Hagenbeek A, Cabanillas F, Klippensten D, Hiddemann W, Castellino R, Harris NL, Armitage JO, Carter W, Hoppe R, Canellos GP. Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol. April 1999; 17 (4):

[1244] ), positron emission tomography (PET), white blood cell count, differential, fluorescence-activated cell sorting, bone marrow aspiration, lymph node biopsy / histology, and various lymphoma-specific clinical chemistry parameters (e.g., lactate dehydrogenase) and other established standard methods.

[0289] Another major challenge in developing drugs (such as the drug compositions of this invention) is the predictable modulation of pharmacokinetic properties. To this end, pharmacokinetic profiles of candidate drugs can be established, i.e., profiles of pharmacokinetic parameters that affect the ability of a particular drug to treat a given condition. Pharmacokinetic parameters affecting the ability of a drug to treat a disease entity include, but are not limited to: half-life, distribution capacity, hepatic first-pass metabolism, and serum binding. The efficacy of a given drug agent can be affected by each of the parameters mentioned above. The contemplated features of the antibody constructs of this invention provide for specific FC patterns they contain, such as differences in pharmacokinetic behavior. The extended half-life targeted antibody constructs of this invention preferably exhibit a surprisingly increased in vivo retention time compared to the "canonical" non-HLE form of said antibody constructs.

[0290] "Half-life" refers to the time it takes for 50% of an administered drug to be eliminated through biological processes (such as metabolism and excretion). "First-pass metabolism" refers to the tendency of a drug to be metabolized during its first contact with the liver, i.e., its first passage through the liver. "Volume of distribution" refers to the degree to which a drug remains in various compartments of the body (such as intracellular and extracellular spaces, tissues, and organs) and its distribution within these compartments. "Serium binding" refers to the tendency of a drug to interact with and bind to serum proteins (such as albumin), resulting in a decrease or loss of its biological activity.

[0291] Pharmacokinetic parameters also include bioavailability, lag time (T lag), Tmax, absorption rate, onset time, and / or Cmax for a given amount of drug administered. "Bioavailability" refers to the amount of drug in the blood compartment. "Lack time" refers to the time delay between drug administration and its detection and measurability in blood or plasma. "Tmax" is the time after which the drug reaches its maximum blood concentration, and "Cmax" is the maximum blood concentration achievable with a given drug. The time required to reach the blood or tissue concentration of the drug required to achieve its biological effect is influenced by all parameters. Pharmacokinetic parameters for bispecific antibody constructs exhibiting cross-species specificity (which can be determined in preclinical animal testing in non-chimpanzee primates as outlined above) are also shown, for example, by Schlereth et al. (Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy] 20 (2005), 1-12).

[0292] In a preferred aspect of the invention, the pharmaceutical composition is stable at about -20°C for at least four weeks. As will be apparent from the additional embodiments, the quality of the antibody constructs of the present invention can be tested using different systems relative to the quality of corresponding prior art antibody constructs. These tests are understood to conform to "ICH Harmonised Tripartite Guideline: Stability Testing of Biotechnological / Biological Products Q5C and Specifications: Test procedures and Acceptance Criteria for Biotech Biotechnological / Biological Products Q6B" and are therefore selected to provide stability indicator curves to identify changes in the properties, purity, and potency of the product. The term purity is generally accepted as relative. Due to the effects of glycosylation, deacetylation, or other heterogeneity, the absolute purity of biotechnological / biological products should typically be assessed by more than one method, and the resulting purity value depends on the method. For the purposes of stability testing, purity testing should focus on the method for identifying degradation products.

[0293] To assess the quality of pharmaceutical compositions comprising the antibody constructs of the present invention, analysis can be performed, for example, by analyzing the content of soluble aggregates in solution (HMWS excluded based on size). Preferably, stability at about -20°C for at least four weeks is characterized by a content of less than 1.5% HMWS, more preferably less than 1% HMWS.

[0294] The preferred method for product quality analysis described in this paper is size exclusion-high performance liquid chromatography (SE-HPLC). SE-HPLC is typically performed using a size exclusion column and a UHPLC system, such as a Waters BEH200 size exclusion column (4.6 x 150 mm, 1.7 µm) and a Waters UHPLC system. Pure protein samples are injected with phosphate buffer, for example, containing NaCl (mobile phase: 100 mM sodium phosphate, 250 mM NaCl, pH 6.8), at a flow rate of, for example, 0.4 mL / min for isocratic separation, and the eluent is monitored by UV absorbance at 280 nm. Typically, approximately 6 µg of sample is loaded.

[0295] Before initiating the CM method, vials containing CHO cells expressing bispecific antibody constructs are typically thawed. During scale-up, cells are resuspended in fresh selective growth medium at target viable cell density (VCD). Culture volumes are continuously expanded in shake flasks or bioreactors to produce sufficient cell mass for final inoculation into perfusion production bioreactors (e.g., 10 L or 50 L scale or larger).

[0296] Once the cells are seeded into the production bioreactor at the concentration ranges described herein, there is an initial cell growth phase lasting several days, typically approximately 7 to 28 days, to increase the cell density and biomass to the optimal setpoints described herein and measured by a permittivity probe (Hamilton Bonaduz AG, Switzerland). The production bioreactor is maintained at an optimal pH, typically approximately 6 to 7.4, e.g., pH 6.85, dissolved oxygen, e.g., 64 mm Hg, and approximately 36°C. Several days after the cell growth phase, typically on days 2, 3, 4, 5, 6, 7, 8, 9, or 10, preferably day 4, perfusion culture is initiated using an alternating tangential flow (ATF) filtration system (e.g., Refine Technologies, Hanover, NJ) (with filters (e.g., polyether ether 0.2-μm filters (e.g., GE Healthcare, Pittsburgh, PA)) and a suitable chemically defined perfusion medium at the VVD perfusion rate described herein (e.g., 0.4 VVD perfusion rate for a bioreactor). The perfusion rate is typically increased gradually, e.g., from 0.4 VVD on day 4 to 2 VVD on day 12. VVD. Once the biomass set point is reached on the last day of the gradually increasing VVD, the cell culture temperature is typically lowered, for example, to 33.5°C, and HCCF (i.e., cell-free permeate containing bispecific antibody constructs) collection begins. Cell culture is continued for a period as described herein (e.g., at least 7, 14, 28, or 40 days, preferably at least 28 days) by feeding at a set perfusion rate (typically the highest VVD perfusion rate resulting from the gradual increase) and by allowing additional cell discharge to maintain a better biomass set point. Cell density (via CDV (e.g., Nova Biotech)) is then determined. HCCF (measured by Biomedical, Waltham, MIT) and metabolites (e.g., by NovaFlex, Waltham, MIT) and permeate titers (measured by HPLC analysis) are typically measured throughout the culture period. Preferably, HCCF is collected continuously at room temperature or in increments of, for example, 6, 12, 24, 48, 72, 96, 120, or 144 hours, and proceeded to protein-L capture chromatography. Product quality attributes and method-related impurities from the protein-L eluent (e.g., on days 26, 27, 34, and 40) are analyzed using analytical cation exchange chromatography (CEX-HPLC), peptide mapping, and / or HCP ELISA.

[0297] Trypsin peptide maps for chemical modification Using a device such as the Millipore Microcon 30K, a filter-based method was employed to digest the protein samples of the bispecific antibody constructs. The protein samples were added to a filter, centrifuged to remove the sample matrix, and then denatured in a buffer containing, for example, 6M guanidine hydrochloride (GuHCl) (e.g., Thermo Fisher Scientific, Rockford, Illinois). Reduction was then performed for 30 min at, for example, 37°C with, for example, 500 mM dithiothreitol (DTT) (e.g., Sigma-Aldrich, St. Louis, Michigan), followed by alkylation by, for example, incubation with, for example, 500 mM iodoacetic acid (IAA) (e.g., Sigma-Aldrich, St. Louis, Michigan) in the dark at room temperature for, for example, 20 min. Unreacted IAA was quenched by the addition of DTT. All of the above steps were performed on the filter. The sample was then centrifuged and buffer-exchanged to a digestion buffer (e.g., 50 mM Tris, pH 7.8, containing methionine) to remove any residual DTT and IAA. Trypsin digestion was performed on a filter, e.g., at 37°C using a 1:20 (w / w) enzyme-to-protein ratio for 1 hour. The digestion mixture was collected by centrifugation and then quenched, e.g., by adding 8 M GuHCl to acetate buffer at pH 4.7.

[0298] Liquid chromatography-mass spectrometry (LC-MS) analysis was performed using a high-performance liquid chromatography (UPLC) system (e.g., Thermo U-3000) directly coupled to a mass spectrometer (e.g., Thermo Scientific Q-Exactive). Protein digests were separated by reverse-phase separation using an Agilent Zorbax C18 RR HD column (2.1 × 150 mm, 1.8 µm) (column temperature maintained at 50°C). Mobile phase A was 0.020% (v / v) formic acid (FA) (in water), and mobile phase B was 0.018% (v / v) FA (in acetonitrile (I)). Approximately 5 µg of the digested bispecific antibody construct was injected into the column. The peptide was separated using a gradient (e.g., 0.5% to 36% B over 145 min) at a flow rate (e.g., 0.2 mL / min). Eluted peptides were monitored by MS.

[0299] For peptide identification and modification analysis, data-dependent tandem MS (MS / MS) experiments are typically employed. A full scan is typically acquired, for example, from 200 to 2000 m / z in positive ion mode, followed by, for example, six data-dependent MS / MS scans to identify the peptide sequence. Quantification is based on the mass spectrometry data from the selected ion monitoring, using the following equation: Wherein, % represents the level of the modified peptide, A represents the ion chromatography area of ​​the modified peptide, and A represents the ion chromatography area of ​​the unmodified peptide.

[0300] Host cell protein (HCP) ELISA Rabbit anti-HCP immunoglobulin G (IgG) (Amgen, internal antibody) was coated onto a microtiter plate. After washing and blocking the plate, test samples, controls, and HCP calibration standards were added and incubated. Unbound proteins were washed off the plate, and pooled rabbit anti-HCP IgG-Biotin (Amgen, internal antibody) was added and incubated. After another wash, streptavidin™ horseradish peroxidase conjugate (HRP-conjugate) (e.g., Amersham-GE, Buckinghamshire, UK) was added and incubated. The plate was washed one last time, and the chromogenic substrate tetramethylbenzidine (TMB) (e.g., Kirkegaard and Perry Laboratories, Gaithersburg, MD) was added. Color development was inhibited with 1M phosphate, and the optical density was measured using a spectrophotometer.

[0301] Preferred formulations of antibody constructs that are drug components may, for example, contain components of the formulation described below: • Preparation: Potassium phosphate, L-arginine salt, trehalose, and polysorbate 80, at pH 6.0

[0302] Generally speaking, antibody constructs according to the present invention, which provide a specific FC mode, are typically more stable under a wide range of stress conditions (such as temperature and light stress) compared to antibody constructs with different HLE forms and those without any HLE forms (i.e., "canonical" antibody constructs). This temperature stability can involve both decreasing temperatures (below room temperature, including freezing) and increasing temperatures (above room temperature, including temperatures up to or above body temperature). As those skilled in the art will recognize, this improved stability with respect to stress, which is unavoidable in clinical practice, makes the antibody construct safer because fewer degradation products occur in clinical practice. Consequently, the increased stability implies increased safety.

[0303] One embodiment provides an antibody construct of the present invention or an antibody construct produced by the method of the present invention for the prevention, treatment or relief of proliferative diseases, neoplastic diseases, viral diseases or immune disorders.

[0304] The formulations described herein may be used as pharmaceutical components to treat, alleviate, and / or prevent the pathological medical conditions described herein in patients in need. The term "treatment" refers to both therapeutic treatment and prophylactic or preventative measures. Treatment includes administering or introducing the formulation into the body, isolated tissues, or cells of a patient who has a disease / symptom, symptoms of a disease / symptom, or a predisposition to a disease / symptom, with the aim of curing, restoring, alleviating, altering, remedying, alleviating, improving, or influencing the disease, its symptoms, or the predisposition to a disease.

[0305] As used herein, the term "remission" means any improvement in the disease state of a patient suffering from a tumor or cancer or metastatic cancer as described below, achieved by administering an antibody construct according to the invention to a subject in need. Such improvement may also be considered as slowing or halting the progression of the patient's tumor or cancer or metastatic cancer. As used herein, the term "prevention" means preventing the occurrence or recurrence of a patient suffering from a tumor or cancer or metastatic cancer as specified below, achieved by administering an antibody construct according to the invention to a subject in need.

[0306] The term "disease" refers to any condition that would benefit from treatment with the antibody constructs or pharmaceutical compositions described herein. This includes both chronic and acute symptoms or diseases, including pathological conditions that make mammals susceptible to the diseases under consideration.

[0307] "Variations" are abnormal growths of tissue that usually, but not always, form a mass. When they do form a mass, they are usually called "tumors." Vesicles or tumors can be benign, potentially malignant (precancerous), or malignant. Malignant vegetations are usually called cancer. They often invade and destroy surrounding tissues and can metastasize, meaning they spread to other parts of the body, tissues, or organs. Therefore, the term "metastatic cancer" covers metastases to tissues or organs other than the original tumor. Lymphomas and leukemias are lymphatic vegetations. For the purposes of this invention, they are also covered by the terms "tumor" or "cancer."

[0308] The term "viral disease" describes a disease that results from a viral infection in a subject.

[0309] As used in this article, the term "immune disorder" describes immune disorders that meet the common definition of this term, such as autoimmune diseases, hypersensitivity reactions, and immunodeficiency.

[0310] In one embodiment, the present invention provides a method for treating or alleviating proliferative diseases, neoplastic diseases, viral diseases, or immune disorders, the method comprising the step of administering an antibody construct of the present invention or an antibody construct generated according to the method of the present invention to a subject in need.

[0311] The terms "subjects in need" or "those in need of treatment" include those who already have the disorder as well as those who need to prevent the disorder. Subjects in need or "patients" include people and other mammalian subjects who receive preventative or therapeutic treatment.

[0312] The antibody constructs of this invention are typically designed for specific administration routes and methods, specific dosages and frequencies, specific treatments for specific diseases, bioavailability, and durability. The constituent substances are preferably formulated at concentrations acceptable for the administration sites.

[0313] Therefore, formulations and compositions can be designed according to the present invention for delivery via any suitable administration route. In the context of the present invention, administration routes include, but are not limited to, those specified below. • Local routes (e.g., epidermal, inhalation, nose, eye, ear (auricular / aural), vagina, mucous membrane); • Enteric routes (e.g., oral, gastrointestinal, sublingual, sublipal, buccal, rectal); and • Extra-gastrointestinal routes (such as intravenous, intra-arterial, intra-bone, intramuscular, intracerebral, intravenous, epidural, intrathecal, subcutaneous, intraperitoneal, extra-amniotic, intra-articular, intracardiac, intradermal, intralesional, intrauterine, intrabladder, intravitreal, percutaneous, intranasal, transmucosal, intrasynovial, and intraluminal).

[0314] The pharmaceutical compositions and antibody constructs of the present invention are particularly suitable for parenteral administration, such as subcutaneous or intravenous delivery, for example by injection such as rapid concentration, or by infusion such as continuous infusion. The pharmaceutical compositions can be administered using medical devices. Examples of medical devices for administering pharmaceutical compositions are described in U.S. Patent Nos. 4,475,196; 4,439,196; 4,447,224; 4,447,233; 4,486,194; 4,487,603; 4,596,556; 4,790,824; 4,941,880; 5,064,413; 5,312,335; 5,312,335; 5,383,851; and 5,399,163.

[0315] In particular, the present invention provides uninterrupted dosing of suitable components. As a non-limiting example, uninterrupted or substantially uninterrupted (i.e., continuous) dosing can be achieved by means of a small pump system worn by the patient for metering the inflow of a therapeutic agent into the patient's body. Pharmaceutical components comprising antibody constructs of the present invention can be dosed using said pump system. Such pump systems are generally known in the art and typically rely on the periodic replacement of cartridges containing the therapeutic agent to be infused. When the cartridge in such a pump system is replaced, a temporary interruption may occur in the previously uninterrupted inflow of the therapeutic agent into the patient's body. In this case, the dosing phase before and after cartridge replacement will still be considered within the meaning of pharmaceutical means, and the method of the present invention together constitutes one instance of "uninterrupted dosing" of such a therapeutic agent.

[0316] The continuous or uninterrupted administration of the antibody construct of the present invention can be performed intravenously or subcutaneously using a fluid delivery device or miniature pump system, which includes a fluid drive mechanism for discharging fluid from a reservoir and an actuation mechanism for actuating the drive mechanism. A pump system for subcutaneous administration may include a needle or cannula for penetrating the patient's skin and delivering a suitable composition into the patient's body. The pump system may be directly attached to or connected to the patient's skin independently of a vein, artery, or blood vessel, thereby allowing direct contact between the pump system and the patient's skin. The pump system may be attached to the patient's skin for 24 hours to several days. The pump system may be small in size, with a small-volume reservoir. As a non-limiting example, the reservoir volume of the suitable drug composition to be administered may be between 0.1 and 50 ml.

[0317] Continuous delivery can also be achieved transdermally via patches worn on the skin, which are replaced at regular intervals. Those skilled in the art will recognize patch systems suitable for drug delivery for this purpose. Notably, transdermal delivery is particularly well-suited for uninterrupted delivery because the replacement of the first exhausted patch can be advantageously accomplished simultaneously with placing a new second patch, for example, adjacent to the skin surface of the first exhausted patch, and just before the first exhausted patch is removed. There are no issues of flow interruption or battery failure.

[0318] If the drug composition has been lyophilized, the lyophilized material should be reconstituted in a suitable liquid before administration. This reconstitution can be achieved in, for example, sterile water for injection (BWFI), physiological saline, phosphate-buffered saline (PBS), or the same formulation in which the protein was previously in the lyophilized form.

[0319] The components of this invention can be administered to subjects at appropriate doses, which can be determined, for example, by dose-escalation studies of the antibody constructs of this invention exhibiting the interspecies specificity described herein, administered in increasing doses to non-chimpanzee primates (e.g., rhesus monkeys). As described above, the antibody constructs of this invention exhibiting the interspecies specificity described herein can advantageously be used in the same form for preclinical testing in non-chimpanzee primates and for use as a drug in humans. Dosing regimens will be determined by the attending physician and clinical factors. As is well known in the medical field, the dose for any patient depends on many factors, including patient size, body surface area, age, the specific compound to be administered, sex, time and route of administration, general health condition, and other concurrently administered medications.

[0320] The term "effective dose" or "effective dosage" is defined as an amount sufficient to achieve or at least partially achieve the desired effect. The term "therapeutic effective dose" is defined as an amount sufficient to cure or at least partially stop the disease and its complications in a patient with an existing disease. The effective amount or dosage for this purpose will depend on the condition to be treated (indication), the antibody construct delivered, the treatment context and goals, the severity of the disease, prior therapy, the patient's clinical history and response to the therapeutic agent, the route of administration, the patient's body size (weight, body surface or organ size) and / or condition (age and general health status), and the general state of the patient's autoimmune system. Appropriate dosages may be adjusted according to the attending physician's judgment so that it can be administered to the patient in a single dose or in a series of doses to achieve the best therapeutic effect.

[0321] Depending on the factors described above, typical dosage ranges can be from about 0.1 µg / kg to up to about 30 mg / kg or higher. In specific embodiments, dosage ranges can be from about 1.0 µg / kg to about 20 mg / kg, from 10 µg / kg to up to about 10 mg / kg as needed, or from 100 µg / kg to up to about 5 mg / kg.

[0322] The therapeutically effective amount of the antibody construct of the present invention preferably results in a reduction in the severity of disease symptoms, an increase in the frequency or duration of symptom-free periods, or prevention of damage or disability caused by disease suffering. For treating tumors expressing target cell antigens, the therapeutically effective amount of the antibody construct of the present invention, such as an anti-target cell antigen / anti-CD3 antibody construct, preferably inhibits cell or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% relative to untreated patients. The ability of the compound to inhibit tumor growth can be evaluated in animal models where efficacy is predicted.

[0323] The pharmaceutical composition can be administered as a standalone therapeutic agent or in combination with additional therapies, such as as-needed anticancer therapies, for example, other protein and non-protein drugs. Such pharmaceutical compositions can be administered simultaneously with compositions comprising antibody constructs of the present invention as defined herein, or administered separately at time-defined intervals and doses before or after administration of said antibody construct.

[0324] As used herein, the term "effective and non-toxic dose" refers to a tolerable dose of the antibody construct of the present invention that is high enough to induce pathological cell depletion, tumor elimination, tumor shrinkage, or disease stabilization without or substantially without major toxic effects. Such an effective and non-toxic dose can be determined, for example, by dose escalation studies described in the art, and should be below doses that induce serious adverse side effects (dose-limiting toxicity, DLT).

[0325] As used herein, the term "toxicity" refers to the toxic effects of a drug as manifested in adverse events or serious adverse events. Such adverse events may refer to a lack of systemic drug tolerance and / or a lack of local tolerance after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

[0326] As used herein, the terms "safety," "in vivo safety," or "tolerability" define the administration of a drug without immediately inducing serious adverse events (local tolerance) or without inducing serious adverse events over a longer period of drug administration. For example, "safety," "in vivo safety," or "tolerability" can be evaluated at regular intervals during treatment and follow-up periods. Measurements include clinical evaluations, such as organ manifestations, and screening for laboratory abnormalities. Clinical evaluations can be performed, and deviations from normal findings can be recorded / coded according to NCI-CTC and / or MedDRA standards. Organ manifestations can include criteria such as allergy / immunology, blood / bone marrow, arrhythmia, coagulation, etc., as described, for example, in the Common Terminology Standard for Adverse Events v3.0 (CTCAE). Laboratory parameters that can be tested include, for example, hematology, clinical chemistry, coagulation profiles, and urinalysis, as well as examinations of other body fluids (such as serum, plasma, lymph or cerebrospinal fluid, fluids, etc.). Therefore, safety can be assessed by means of, for example, physical examination, imaging techniques (i.e., ultrasound, X-ray, CT scan, magnetic resonance imaging (MRI), other measures with technical devices (i.e., electrocardiography)), vital signs, by measuring laboratory parameters, and recording adverse events. For example, adverse events in non-chimpanzee primates according to the uses and methods of the present invention can be examined by histopathological and / or histochemical methods.

[0327] The above terms are also mentioned in the following: for example, Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6; ICH Harmonised Tripartite Guideline; ICH Steering Committee meeting on July 16, 1997.

[0328] Finally, the present invention provides a kit comprising the antibody construct of the present invention or an antibody construct produced by the method of the present invention, the pharmaceutical composition of the present invention, the polynucleotide of the present invention, the vector of the present invention, and / or the host cell of the present invention.

[0329] In the context of this invention, the term "kit" refers to two or more components (one of which corresponds to the antibody construct, pharmaceutical composition, carrier, or host cell of this invention) packaged together in a container, receptor, or other container. Therefore, a kit can be described as a group of products and / or devices sufficient to achieve a particular objective, which can be sold as a single unit.

[0330] The kit may include one or more receivers (such as vials, ampoules, containers, syringes, pouches) having any suitable shape, size, and material (preferably waterproof, such as plastic or glass), the one or more receivers containing a dose of the antibody construct or pharmaceutical composition of the present invention suitable for administration (see above). The kit may additionally include instructions for use (e.g., in the form of a booklet or instruction manual), means for administering the antibody construct of the present invention (such as syringes, pumps, infusion sets, etc.), means for reconstructing the antibody construct of the present invention, and / or means for diluting the antibody construct of the present invention.

[0331] The present invention also provides kits for single-dose dosing units. The kits of the present invention may further include a first acceptor comprising a dried / lyophilized antibody construct and a second acceptor comprising an aqueous formulation. In some embodiments of the present invention, kits comprising single-compartment and multi-compartment pre-filled syringes (e.g., liquid syringes and lyophilized syringes) are provided. ▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪

[0332] It should be noted that, unless the context clearly indicates otherwise, the singular forms "a," "an," and "the" include a plural of indicators. Thus, for example, a reference to "a reagent" includes one or more of such different reagents, and a reference to "the method" includes reference to equivalent steps and methods known to those skilled in the art that can modify or replace the method described herein.

[0333] Unless otherwise indicated, the term "at least" preceding a series of elements should be understood to refer to each element in the series. Those skilled in the art will recognize or be able to determine many equivalents of the specific embodiments of the invention described herein using only conventional experimentation. Such equivalents are intended to be covered by the invention.

[0334] The term "and / or" as used herein includes the meaning of "and", "or" and "all or any other combination of the elements connected by the term".

[0335] As used herein, the term "about" or "approximately" means within 20% of a given value or range, preferably within 10%, and even more preferably within 5%. However, it also includes specific numbers, such as about 20 including 20.

[0336] The terms "less than" or "greater than" include specific numbers. For example, "less than 20" means less than or equal to. Similarly, "more than" or "greater than" means more than or equal to, or greater than or equal to, respectively.

[0337] Throughout this specification and the following claims, unless the context otherwise requires, the word "comprise" and its variations such as "comprises" or "comprising" shall be understood to implicitly include the stated whole or step or group of whole or steps, but not exclude any other whole or step or group of whole or steps. When used herein, the term "comprising" may be replaced by the terms "containing" or "including," or sometimes by the term "having."

[0338] When used herein, "consisting of" excludes any element, step, or component not specified in the elements of the claim. When used herein, "consisting substantially of" does not exclude materials or steps that do not substantially affect the essential and novel features of the claim.

[0339] In each example in this article, any one of the terms "comprising / including", "substantially consisting of", and "consisting of" can be replaced by any of the other two terms.

[0340] It should be understood that the present invention is not limited to the specific methods, schemes, materials, reagents, and substances described herein, and therefore can be varied. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the invention as defined solely by the claims.

[0341] All publications and patents (including all patents, patent applications, scientific publications, manufacturers' specifications, instructions, etc.) referenced throughout this specification are incorporated herein by reference in their entirety, whether above or below. Nothing herein shall be construed as an admission that the invention is not entitled to any prior art as a result of such disclosures. Where material incorporated by reference conflicts or is inconsistent with this specification to some extent, this specification shall supersede any such material.

[0342] A better understanding of the invention and its advantages will be gained from the following examples, which are for illustrative purposes only. These examples are not intended to limit the scope of the invention in any way.

[0343] [Example] [1]: Continuous manufacturing: Biomass-based automated feeding to improve productivity and robustness For this method, a 50 L commercial-scale single-use bioreactor was used. The total duration was 40 days of continuous bioreactor operation and 28 days of continuous harvesting. Automated biomass control according to the present invention was applied. The antibody construction system CD19 x CD3 bispecific T-cell conjugate molecule (SEQ ID NO: 17) was tested. The cell line CHO cells was used for expression. Three experiments were conducted at laboratory scale (1.5 L working volume) to investigate biomass-based feeding. In the first experiment, manual PCV-based feeding [volume / day / %PCV] was tested during the production phase of the continuous perfusion method. The PCV setpoint was manually increased after day 23 with a corresponding increase in feed rate (based on BSPR). The BSPR setpoint for days 23-33 was 0.078 L / day. The control method featured manual, time-based feed rate [volume / day] and a fixed PCV. As a result, 50% PCV was obtained, exhibiting high viability (Figures 2-3) and metabolite profiles similar to the control. In the second experiment, an automated feed based on permittivity was tested during the growth phase of the continuous perfusion method. The test conditions had the following BSPR: 0.04 cm / pF·day for days 4–10 and 0.03 cm / pF·day for days 11–14. The control method had an artificial, time-based feed rate [volume / day] during the growth phase. The test conditions produced cell growth and metabolite curves similar to the control (Figures 4 and 5). In the third experiment, three levels of biomass were tested: 19%, 23%, and 27% filled cell volume (PCV). A 14-day controlled process (constant feed rate) and a 6-day investigational biomass-based automated feed were conducted outside of the continuous harvest period. No new hardware was required, but two control loops were integrated for level control and biomass control. Biomass-specific perfusion rate (BSPR) was calculated based on the control setting. As a result, better biomass control was observed under automated biomass-based feeding compared to the control feed (see Figure 6). Similar osmolarity and lactate values ​​were observed across biomass levels (see Figure 7). Titration was comparable across different control strategies, with higher titers at higher biomass levels (see Figure 7). Even with biomass and feed rate controlled within certain limits, small increases in BSPR increased potency. Advantageously, high viability (>89%) was maintained across conditions in the bioreactor.

[0344] [Example] [2] Production of CD70xCD3 bispecific molecules (SEQ ID NO: 248) under controlled biomass-based methods. This method utilizes an extended perfusion approach (continuous manufacturing) using ATF technology for 15–28 days. Cell culture is conducted in a non-steady-state mode. Autofeeding based on capacitance was tested at three CSPR levels during the growth phase (0.02, 0.03, and 0.065 cm / pF / day) and the production phase (0.01, 0.017, and 0.035 cm / pF / day) and compared with a continuous production control. Autofeeding based on biomass (circles) with similar CSPR performed similarly to the control (triangles). At lower CSPR (stripes), there was higher productivity and lower viability during the production phase. Conversely, at higher CSPR (crosses), productivity was lower but viability was higher (see Figure 8). [ [surface] [4] [:] [CD70xCD3] [Scope of Bispecific Molecules (Automated Feeding Based on Capacitance in Unsteady Continuous Manufacturing)] [ ] [parameter] [unit] [Minimum] [maximum] Duration of growth phase Days 0 12 VCD during the growth stage 1E6 cells / mL 4.5 116.2 Permeability during the growth stage pF / cm 6.5 114 growth stage CSPR based on permittivity cm / pF / day 0.02 0.065 Production stage CSPR based on permittivity cm / pF / day 0.01 0.035 growth stage CSPR based on VCD pL / cell / day twenty three 85 Production stage CSPR based on VCD pL / cell / day 15 44

[0345] [Example] [3] A bispecific PD1 IL21 mutant protein molecule was generated. This method employed a 15-day perfusion approach using ATF technology. A CSPR value was tested in duplicate during both the growth and production phases. A higher CSPR (0.12) was tested on days 3–8, and a lower CSPR (0.03) was tested on days 9–15 (see Figure 9). Compared to the control, the titer was lower in the automated feeder, but the productivity was similar. This may be due to lower cell growth under automated feeder conditions (see Figure 9). The CSPR rate and time were optimized to increase VCD and titer to 0.03–0.05 cm / pF / day during the growth phase and 0.01–0.025 cm / pF / day during the production phase. [ [surface] [5] [:] [PD1 IL21] [Scope of Mutant Protein Bispecific Molecules (Automated Feeding Based on Capacitance in Perfusion Methods)] [ ] [parameter] [unit] [Minimum] [maximum] Duration Days 15 15 VCD 1E6 cells / mL 1.1 38.6 permittivity pF / cm 3.6 71 CSPR based on permittivity cm / pF / day 0.03 0.12 CSPR based on VCD nL / cell / day 0.05 0.23 [ ]

[0346] [Example] [4] PD1 mAb was generated. This method employed a 15-day perfusion approach using ATF technology. A CSPR value was tested in duplicate during the growth and production phases. A higher CSPR (0.08) was detected on days 3–8, and a lower CSPR (0.015) was detected on days 9–15. The titer was slightly lower in the autofeeding system compared to the control, but the productivity was higher (see Figure 10). This is likely due to the lower VCD under autofeeding conditions. Lower viability was expected alongside the higher productivity. [ [surface] [6] [:] [PD1 mAb] [Scope (Automatic feeding based on permittivity in infusion methods)] [ ] [parameter] [unit] [Minimum] [maximum] Duration Days 15 15 VCD 1E6 cells / mL 0.8 56.4 permittivity pF / cm 5.7 96 CSPR based on permittivity cm / pF / day 0.015 0.08 CSPR based on VCD nL / cell / day 0.018 0.17 [ ]

[0347] none

[0348] none <![CDATA[ <110> Amgen Inc. (USA) <![CDATA[ <120> Automated perfusion control based on biomass in biopharmaceutical manufacturing <![CDATA[ <130> 32243 / 54411 / PC]]> <![CDATA[ <150> US 62 / 861,297 <![CDATA[ <151> 2019-06-13 <![CDATA[ <160> 248 ]]> <![CDATA[ <170> PatentIn Version 3.5 <![CDATA[ <210> 1]]> <![CDATA[ <211> 15]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VL CDR1]]> <![CDATA[ <400> 1]]> Lys Ala Ser Gln Ser Val Asp Tyr Asp Gly Asp Ser Tyr Leu Asn 1 5 10 15 <![CDATA[ <210> 2]]> <![CDATA[ <211> 7]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VL CDR2]]> <![CDATA[ <400> 2]]> Asp Ala Ser Asn Leu Val Ser 1 5 <![CDATA[ <210> 3]]> <![CDATA[ <211> 9]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VL CDR3]]> <![CDATA[ <400> 3]]> Gln Gln Ser Thr Glu Asp Pro Trp Thr 1 5 <![CDATA[ <210> 4]]> <![CDATA[ <211> 5]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VH CDR1]]> <![CDATA[ <400> 4]]> Ser Tyr Trp Met Asn 1 5 <![CDATA[ <210> 5]]> <![CDATA[ <211> 17]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VH CDR2]]> <![CDATA[ <400> 5]]> Gln Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe Lys 1 5 10 15 Gly <![CDATA[ <210> 6]]> <![CDATA[ <211> 15]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[<223]]> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VH CDR3]]> <![CDATA[ <400> 6]]> <![CDATA[Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp Tyr 1 5 10 15 <![CDATA[ <210> 7]]> <![CDATA[ <211> 111]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19 VL]]> <![CDATA[ <400> 7]]> Asp Ile Gln Leu Thr Gln Ser Pro Ala Ser Leu Ala Val Ser Leu Gly 1 5 10 15 Gln Arg Ala Thr Ile Ser Cys Lys Ala Ser Gln Ser Val Asp Tyr Asp 20 25 30 Gly Asp Ser Tyr Leu Asn Trp Tyr Gln Gln Ile Pro Gly Gln Pro Pro 35 40 45 Lys Leu Leu Ile Tyr Asp Ala Ser Asn Leu Val Ser Gly Ile Pro Pro 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Asn Ile His 65 70 75 80 Pro Val Glu Lys Val Asp Ala Ala Thr Tyr His Cys Gln Gln Ser Thr 85 90 95 Glu Asp Pro Trp Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys 100 105 110 <![CDATA[ <210> 8]]> <![CDATA[ <211> 124]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[<221> Features not yet classified]]> <![CDATA[<223> CD19 VH]]> <![CDATA[<400> 8]]> Gln Val Gln Leu Gln Gln Ser Gly Ala Glu Leu Val Arg Pro Gly Ser 1 5 10 15 Ser Val Lys Ile Ser Cys Lys Ala Ser Gly Tyr Ala Phe Ser Ser Tyr 20 25 30 Trp Met Asn Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Gln Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe 50 55 60 Lys Gly Lys Ala Thr Leu Thr Ala Asp Glu Ser Ser Ser Thr Ala Tyr 65 70 75 80 Met Gln Leu Ser Ser Leu Ala Ser Glu Asp Ser Ala Val Tyr Phe Cys 85 90 95 Ala Arg Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp 100 105 110 Tyr Trp Gly Gln Gly Thr Thr Val Thr Val Ser Ser 115 120 <![CDATA[<210> 9]]> <![CDATA[<211> 5]]> <![CDATA[<212> PRT]]> <![CDATA[<213> Artificial Sequence]]> <![CDATA[<220>]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VH CDR1]]> <![CDATA[ <400> 9]]> Arg Tyr Thr Met His 1 5 <![CDATA[ <210> 10]]> <![CDATA[ <211> 17]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VH CDR2]]> <![CDATA[ <400> 10]]> Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe Lys 1 5 10 15 Asp <![CDATA[ <210> 11]]> <![CDATA[ <211> 10]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VH CDR3]]> <![CDATA[ <400> 11]]> Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr 1 5 10 <![CDATA[ <210> 12]]> <![CDATA[ <211> 10]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VL CDR1]]> <![CDATA[ <400> 12]]> Arg Ala Ser Ser Ser Val Ser Tyr Met Asn 1 5 10 <![CDATA[ <210> 13]]> <![CDATA[ <211> 7]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VL CDR2]]> <![CDATA[ <400> 13]]> Asp Thr Ser Lys Val Ala Ser 1 5 <![CDATA[ <210> 14]]> <![CDATA[ <211> 9]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VL CD]]>R3 <![CDATA[ <400> 14]]> Gln Gln Trp Ser Ser Asn Pro Leu Thr 1 5 <![CDATA[ <210> 15]]> <![CDATA[ <211> 119]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[<220>]]> <![CDATA[<221> Features yet to be classified]]> <![CDATA[<223> CD3 VH]]> <![CDATA[<400> 15]]> Asp Ile Lys Leu Gln Gln Ser Gly Ala Glu Leu Ala Arg Pro Gly Ala 1 5 10 15 Ser Val Lys Met Ser Cys Lys Thr Ser Gly Tyr Thr Phe Thr Arg Tyr 20 25 30 Thr Met His Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe 50 55 60 Lys Asp Lys Ala Thr Leu Thr Thr Asp Lys Ser Ser Ser Thr Ala Tyr 65 70 75 80 Met Gln Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr Trp Gly Gln Gly 100 105 110 Thr Thr Leu Thr Val Ser Ser 115 <![CDATA[ <210> 16]]> <![CDATA[ <211> 108]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD3 VL]]> <![CDATA[ <400> 16]]> Val Asp Asp Ile Gln Leu Thr Gln Ser Pro Ala Ile Met Ser Ala Ser 1 5 10 15 Pro Gly Glu Lys Val Thr Met Thr Cys Arg Ala Ser Ser Ser Val Ser 20 25 30 Tyr Met Asn Trp Tyr Gln Gln Lys Ser Gly Thr Ser Pro Lys Arg Trp 35 40 45 Ile Tyr Asp Thr Ser Lys Val Ala Ser Gly Val Pro Tyr Arg Phe Ser 50 55 60 Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu Thr Ile Ser Ser Met Glu 65 70 75 80 Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gln Gln Trp Ser Ser Asn Pro 85 90 95 Leu Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lys 100 105 <![CDATA[ <210> 17]]> <![CDATA[ <211> 504]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> CD19xCD3 scFv BLINCYTO incl connector and his label]]> <![CDATA[ <400> 17]]> Asp Ile Gln Leu Thr Gln Ser Pro Ala Ser Leu Ala Val Ser Leu Gly 1 5 10 15 Gln Arg Ala Thr Ile Ser Cys Lys Ala Ser Gln Ser Val Asp Tyr Asp 20 25 30 Gly Asp Ser Tyr Leu Asn Trp Tyr Gln Gln Ile Pro Gly Gln Pro Pro 35 40 45 Lys Leu Leu Ile Tyr Asp Ala Ser Asn Leu Val Ser Gly Ile Pro Pro 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Asn Ile His 65 70 75 80 Pro Val Glu Lys Val Asp Ala Ala Thr Tyr His Cys Gln Gln Ser Thr 85 90 95 Glu Asp Pro Trp Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys Gly 100 105 110 Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Val 115 120 125 Gln Leu Gln Gln Ser Gly Ala Glu Leu Val Arg Pro Gly Ser Ser Val 130 135 140 Lys Ile Ser Cys Lys Ala Ser Gly Tyr Ala Phe Ser Ser Tyr Trp Met 145 150 155 160 Asn Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile Gly Gln 165 170 175 Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe Lys Gly 180 185 190 Lys Ala Thr Leu Thr Ala Asp Glu Ser Ser Ser Thr Ala Tyr Met Gln 195 200 205 Leu Ser Ser Leu Ala Ser Glu Asp Ser Ala Val Tyr Phe Cys Ala Arg 210 215 220 Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp Tyr Trp 225 230 235 240 Gly Gln Gly Thr Thr Val Thr Val Ser Ser Gly Gly Gly Gly Ser Asp 245 250 255 Ile Lys Leu Gln Gln Ser Gly Ala Glu Leu Ala Arg Pro Gly Ala Ser 260 265 270 Val Lys Met Ser Cys Lys Thr Ser Gly Tyr Thr Phe Thr Arg Tyr Thr 275 280 285 Met His Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile Gly 290 295 300 Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe Lys 305 310 315 320 Asp Lys Ala Thr Leu Thr Thr Asp Lys Ser Ser Ser Thr Ala Tyr Met 325 330 335 Gln Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys Ala 340 345 350 Arg Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr Trp Gly Gln Gly Thr 355 360 365 Thr Leu Thr Val Ser Ser Val Glu Gly Gly Ser Gly Gly Ser Gly Gly 370 375 380 Ser Gly Gly Ser Gly Gly Val Asp Asp Ile Gln Leu Thr Gln Ser Pro 385 390 395 400 Ala Ile Met Ser Ala Ser Pro Gly Glu Lys Val Thr Met Thr Cys Arg 405 410 415 Ala Ser Ser Ser Val Ser Tyr Met Asn Trp Tyr Gln Gln Lys Ser Gly 420 425 430 Thr Ser Pro Lys Arg Trp Ile Tyr Asp Thr Ser Lys Val Ala Ser Gly 435 440 445 Val Pro Tyr Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu 450 455 460 Thr Ile Ser Ser Met Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gln 465 470 475 480 Gln Trp Ser Ser Asn Pro Leu Thr Phe Gly Ala Gly Thr Lys Leu Glu 485 490 495 Leu Lys His His His His His 500 <![CDATA[ <210> 18]]> <![CDATA[ <211> 14]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's CDR-L1 <![CDATA[ <400> 18]]> Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 1 5 10 <![CDATA[ <210> 19]]> <![CDATA[ <211> 7]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's CDR-L2 <![CDATA[ <400> 19]]> Gly Thr Lys Phe Leu Ala Pro 1 5 <![CDATA[ <210> 20]]> <![CDATA[ <211> 9]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C CDR-L3]]> <![CDATA[ <400> 20]]> Val Leu Trp Tyr Ser Asn Arg Trp Val 1 5 <![CDATA[ <210> 21]]> <![CDATA[ <211> 5]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's CDR-H1]]> <![CDATA[ <400> 21]]> Lys Tyr Ala Met Asn 1 5 <![CDATA[ <210> 22]]> <![CDATA[ <211> 19]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's CDR-H2]]> <![CDATA[ <400> 22]]> Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp Ser 1 5 10 15 Val Lys Asp <![CDATA[ <210> 23]]> <![CDATA[ <211> 14]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[<2]]> 23> Synthetic polypeptides]]> <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's CDR-H3]]> <![CDATA[ <400> 23]]> <![CDATA[His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp Ala Tyr 1 5 10 <![CDATA[ <210> 24]]> <![CDATA[ <211> 125]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C VH]]> <![CDATA[ <400> 24]]> Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 1 5 10 15 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 20 25 30 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 35 40 45 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 50 55 60 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 65 70 75 80 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 85 90 95 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 100 105 110 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser 115 120 125 <![CDATA[ <210> 25]]> <![CDATA[ <211> 109]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's VL]]> <![CDATA[ <400> 25]]> Gln Thr Val Val Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly 1 5 10 15 Thr Val Thr Leu Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly 20 25 30 Asn Tyr Pro Asn Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly 35 40 45 Leu Ile Gly Gly Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe 50 55 60 Ser Gly Ser Leu Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val 65 70 75 80 Gln Pro Glu Asp Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn 85 90 95 Arg Trp Val Phe Gly Gly Gly Thr Lys Leu Thr Val Leu 100 105 <![CDATA[ <210> 26]]> <![CDATA[ <211> 249]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> I2C's VH-VL]]> <![CDATA[ <400> 26]]> Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 1 5 10 15 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 20 25 30 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 35 40 45 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 50 55 60 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 65 70 75 80 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 85 90 95 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 100 105 110 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser Gly Gly Gly 115 120 125 Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Thr Val Val 130 135 140 Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly Thr Val Thr Leu 145 150 155 160 Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 165 170 175 Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly Leu Ile Gly Gly 180 185 190 Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe Ser Gly Ser Leu 195 200 205 Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val Gln Pro Glu Asp 210 215 220 Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn Arg Trp Val Phe 225 230 235 240 Gly Gly Gly Thr Lys Leu Thr Val Leu 245 <![CDATA[<210> 27]]> <![CDATA[<211> 122]]> <![CDATA[ <212> PRT]]> <![CDATA[ <213> Artificial sequence <![CDATA[ <220> ]]> <![CDATA[ <223> Synthetic polypeptides <![CDATA[ <220> ]]> <![CDATA[ <221> Unclassified features <![CDATA[ <223> E11's CD33 ccVH]]> <![CDATA[ <400> 27]]> Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Cys Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 ...

Claims

1. An upstream manufacturing method for producing an antibody product by applying automated measurement and regulation of the perfusion rate in a perfusion bioreactor, the method comprising the steps of: (i) providing a liquid cell culture medium in the perfusion bioreactor, the liquid cell culture medium comprising at least one mammalian cell culture capable of expressing the antibody product, and wherein the cells have a concentration of at least 1 x 10^5 cells / mL (viable cell density, VCD) when inoculated in the perfusion bioreactor; (ii) A first control loop is provided for measuring and regulating the culture medium level in the bioreactor, the first control loop comprising a level probe for measuring the culture medium level in the bioreactor relative to a set point, a osmotic pump calibrated to measure the permeation rate (volume / time), and a level control device receiving input from the level probe and the osmotic pump, which, in response to the input from the level probe and the osmotic probe, enables the culture medium pump (feed pump) to correct to the culture medium feed rate of the bioreactor, or wherein the level control device receiving input from the level probe and the culture medium pump, in response to the input from the level probe and the culture medium probe, enables the osmotic pump to correct the outflow from the bioreactor; wherein the measurement of the culture medium level in the bioreactor is performed at preset fixed time intervals; (iii) A second control loop is provided for measuring and regulating biomass in the bioreactor, the second control loop comprising a permittivity probe or Raman probe for measuring biomass in the bioreactor, and a biomass control device for receiving input from the biomass permittivity probe or Raman probe, which, in response to the input, enables an exhaust pump to correct the discharge rate from the bioreactor; wherein the biomass measurement in the bioreactor is performed at preset fixed time intervals; (iv) integrated first and second control loops are provided by connecting the biomass control device and the level control device to an integrated unit, wherein the integrated unit is capable of performing automatic perfusion rate calculation, wherein the perfusion rate is a function of the biomass value based on the following equations: perfusion rate (mL / min) = function of biomass value (permissivity, PCV, VCD, spectral value) and / or perfusion rate [mL / min] = perfusion rate (constant) based on permittivity [cm / pF / d] x permittivity value [pF / cm] The constant is the permeation rate [1 / d] divided by the permittivity [pF / cm], and the permittivity value is 0 during the first period (growth phase) when the biomass in the bioreactor increases to approximately the predetermined biomass setpoint.The perfusion rate is 5 to 120 pF / cm, and / or 25 to 100 pF / cm in the second stage (production stage) of biomass stabilization after reaching the predetermined biomass setpoint, and (v) the perfusion rate is automatically corrected or maintained by the integrated unit, which, in response to biomass measured at preset fixed time intervals, sends signals to the osmotic pump and / or the culture medium pump to increase or decrease the pump rate.

2. The method as described in claim 1, wherein the upstream manufacturing method is a discontinuous manufacturing method, which is a pouring method and / or a feed-in batching method; or a continuous manufacturing method, which is a continuous pouring method.

3. The method as claimed in claim 1, wherein in step (i), the cells have a concentration of at least 7 x 10^5 cells / mL when inoculated in the bioreactor.

4. The method as described in claim 1, wherein in step (iv), the biomass setpoint is equal to at least 30 x 10^6 cells / mL of VCD.

5. The method as claimed in claim 1, wherein in step (iv), the biomass setpoint is equal to at least 30 x 10^6 cells / mL of VCD when the manufacturing method is a non-continuous manufacturing method, and 65 x 10^6 cells / mL when the manufacturing method is a continuous manufacturing method.

6. The method as described in claim 1, wherein in step (iv), the cell culture is grown for at least 4 days, or at least 7 days, or at least 12 days or 14 days.

7. The method as described in claim 1, wherein in step (ii), the preset fixed time interval corresponds to at most 1 minute or 30, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds.

8. The method as described in claim 1, wherein in step (iii), the preset fixed time interval corresponds to at most 1 minute or 30, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds.

9. The method as described in claim 1, wherein in step (v), the preset fixed time interval corresponds to at most 1 minute or 30, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 or 0.5 seconds.

10. The method as described in claim 1, wherein the capacitance during the growth phase is 0.70 to 120 pF / cm or 0.73 to 70.7 pF / cm.

11. The method as claimed in claim 1, wherein if the manufacturing method is a continuous manufacturing method, the capacitance during the growth stage is 1 to 20 pF / cm or 100 to 117 pF / cm.

12. The method as claimed in claim 1, wherein the cell-specific perfusion rate based on capacitance during continuous manufacturing is 0.01 to 0.049 cm / pF / d, or 0.015 to 0.04 cm / pF / d, or 0.02 to 0.04 cm / pF / d, or 0.0266 to 0.04 cm / pF / d, or wherein, In discontinuous manufacturing scenarios, the cell-specific perfusion rate based on capacitance is up to 0.2 cm / pF / d, or up to 0.13 cm / pF / d.

13. The method as claimed in claim 1, wherein the applied perfusion rate corresponds to CSPR of 0.01 to 0.1 nL / cell / d, or 0.02 to 0.08 nL / cell / d, or 0.027 to 0.076 nL / cell / d during the growth phase.

14. The method as described in claim 1, wherein the capacitance during the production stage is 55 to 85 pF / cm, or 60 to 75 pF / cm, or 62 to 73 pF / cm.

15. The method as claimed in claim 1, wherein the cell-specific perfusion rate based on capacitance is 0.01 to 0.04 cm / pF / d, or 0.01 to 0.035 cm / pF / d, or 0.01 to 0.0266 cm / pF / d during the production phase.

16. The method as claimed in claim 1, wherein the applied perfusion rate corresponds to a CSPR of 0.01 to 0.49 nL / cell / d, or 0.015 to 0.04 nL / cell / d, or 0.023 to 0.035 nL / cell / d during the production phase.

17. The method as described in claim 1, wherein the production stage requires at least 14, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 or 32 days, wherein the method is a continuous manufacturing method, or at least 3, 4 or 5 days, wherein the production method is a non-continuous manufacturing method.

18. The method as described in claim 1, wherein the antibody product is a full-length antibody or a non-full-length molecule.

19. The method as described in claim 1, wherein the antibody product is a full-length antibody against PD-1.

20. The method of claim 18, wherein the antibody product is a full-length antibody or a molecule based on a full-length antibody or a fragment thereof, the full-length antibody or a molecule based on a full-length antibody or a fragment thereof being bispecific, i.e., the full-length antibody or a molecule based on a full-length antibody or a fragment thereof binds to target cells and / or effector cells, respectively.

21. The method as described in claim 20, wherein the bispecific antibody product is an anti-PD-1 mAb / IL-21 mutant protein fusion protein.

22. The method of claim 20, wherein the antibody product is a bispecific non-full-length molecule comprising first and second binding domains that bind to target cells and effector cells, respectively.

23. The method as described in claim 20, wherein the bispecific molecule is a bispecific T-cell binding agent molecule.

24. The method of claim 23, wherein the bispecific molecule comprises a half-life extension portion selected from human serum albumin (HAS), an HAS-binding domain, a heterologous Fc domain, or an Fc-based half-life extension portion derived from an IgG antibody, or an scFc half-life extension portion.

25. The method as described in claim 22, wherein the first binding domain of the bispecific antibody construct binds to at least one target cell surface antigen selected from the group consisting of: CD19, CD33, EGFRvIII, MSLN, CDH19, FLT3, DLL3, CDH3, EpCAM, CD70, MUC17, CLDN18, BCMA, and PSMA.

26. The method as described in claim 22, wherein the second binding domain of the bispecific antibody product binds to CD3.

27. The method as described in claim 22, wherein the first binding domain comprises a VH region containing CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of: (a) CDR-H1 as depicted in SEQ ID NO: 1, CDR-H2 as depicted in SEQ ID NO: 2, CDR-H3 as depicted in SEQ ID NO: 3, CDR-L1 as depicted in SEQ ID NO: 4, CDR-L2 as depicted in SEQ ID NO: 5, and CDR-L3 as depicted in SEQ ID NO: 6, (b) CDR-H1 as depicted in SEQ ID NO: 29, CDR-H2 as depicted in SEQ ID NO: 30, CDR-H3 as depicted in SEQ ID NO: 31, and CDR-L3 as depicted in SEQ ID NO:

6. (c) CDR-L1 as depicted in SEQ ID NO: 34, CDR-L2 as depicted in SEQ ID NO: 35, and CDR-L3 as depicted in SEQ ID NO: 36; (d) CDR-H1 as depicted in SEQ ID NO: 42, CDR-H2 as depicted in SEQ ID NO: 43, CDR-H3 as depicted in SEQ ID NO: 44, CDR-L1 as depicted in SEQ ID NO: 45, CDR-L2 as depicted in SEQ ID NO: 46, and CDR-L3 as depicted in SEQ ID NO: 47; (e) CDR-H1 as depicted in SEQ ID NO: 53, CDR-H2 as depicted in SEQ ID NO: 54, CDR-H3 as depicted in SEQ ID NO: 55, CDR-L1 as depicted in SEQ ID NO: 56, CDR-L2 as depicted in SEQ ID NO: 57, and CDR-L3 as depicted in SEQ ID NO:

58. The CDR-L3 depicted in SEQ ID NO: 58, (e) the CDR-H1 depicted in SEQ ID NO: 65, the CDR-H2 depicted in SEQ ID NO: 66, the CDR-H3 depicted in SEQ ID NO: 67, the CDR-L1 depicted in SEQ ID NO: 68, the CDR-L2 depicted in SEQ ID NO: 69, and the CDR-L3 depicted in SEQ ID NO: 70,(f) CDR-H1 as depicted in SEQ ID NO: 83, CDR-H2 as depicted in SEQ ID NO: 84, CDR-H3 as depicted in SEQ ID NO: 85, CDR-L1 as depicted in SEQ ID NO: 86, CDR-L2 as depicted in SEQ ID NO: 87, and CDR-L3 as depicted in SEQ ID NO: 88; (g) CDR-H1 as depicted in SEQ ID NO: 94, CDR-H2 as depicted in SEQ ID NO: 95, CDR-H3 as depicted in SEQ ID NO: 96, CDR-L1 as depicted in SEQ ID NO: 97, CDR-L2 as depicted in SEQ ID NO: 98, and CDR-L3 as depicted in SEQ ID NO: 99; (h) CDR-H1 as depicted in SEQ ID NO: 105, CDR-L2 ... The CDR-H2 depicted in SEQ ID NO: 106, the CDR-H3 depicted in SEQ ID NO: 107, the CDR-L1 depicted in SEQ ID NO: 109, the CDR-L2 depicted in SEQ ID NO: 110, and the CDR-L3 depicted in SEQ ID NO: 111, (i) the CDR-H1 depicted in SEQ ID NO: 115, the CDR-H2 depicted in SEQ ID NO: 116, the CDR-H3 depicted in SEQ ID NO: 117, the CDR-L1 depicted in SEQ ID NO: 118, the CDR-L2 depicted in SEQ ID NO: 119, and the CDR-L3 depicted in SEQ ID NO: 120, (j) the CDR-H1 depicted in SEQ ID NO: 126, the CDR-H2 depicted in SEQ ID NO: 127, and the CDR-L3 depicted in SEQ ID NO: 120 ... The CDR-H3 depicted in SEQ ID NO: 128, the CDR-L1 depicted in SEQ ID NO: 129, the CDR-L2 depicted in SEQ ID NO: 130, and the CDR-L3 depicted in SEQ ID NO: 131,(k) CDR-H1 as depicted in SEQ ID NO: 137, CDR-H2 as depicted in SEQ ID NO: 138, CDR-H3 as depicted in SEQ ID NO: 139, CDR-L1 as depicted in SEQ ID NO: 140, CDR-L2 as depicted in SEQ ID NO: 141, and CDR-L3 as depicted in SEQ ID NO: 142, (l) CDR-H1 as depicted in SEQ ID NO: 152, CDR-H2 as depicted in SEQ ID NO: 153, CDR-H3 as depicted in SEQ ID NO: 154, CDR-L1 as depicted in SEQ ID NO: 155, CDR-L2 as depicted in SEQ ID NO: 156, and CDR-L3 as depicted in SEQ ID NO: 157, (m) CDR-H1 as depicted in SEQ ID NO: 167, and CDR-L3 as depicted in SEQ ID NO: 137, SEQ ID NO: 138, SEQ ID NO: 139, SEQ ID NO: 139, SEQ ID NO: 140, SEQ ID NO: 140, SEQ ID NO: 141, and SEQ ID NO: 142, (l) CDR-H1 as depicted in SEQ ID NO: 152, SEQ ID NO: 153, SEQ ID NO: 154, SEQ ID NO: 155, SEQ ID NO: 155, SEQ ID NO: 156, and SEQ ID NO: 157, (m) CDR-H1 as depicted in SEQ ID NO: 167, SEQ ID NO: 140, SEQ ID NO: 140, SEQ ID NO: 141, and SEQ ID NO: 142, SEQ ID NO: 143, SEQ ID NO: The CDR-H2 depicted in SEQ ID NO: 168, the CDR-H3 depicted in SEQ ID NO: 169, the CDR-L1 depicted in SEQ ID NO: 170, the CDR-L2 depicted in SEQ ID NO: 171, and the CDR-L3 depicted in SEQ ID NO: 172,(n) CDR-H1 as depicted in SEQ ID NO: 203, CDR-H2 as depicted in SEQ ID NO: 204, CDR-H3 as depicted in SEQ ID NO: 205, CDR-L1 as depicted in SEQ ID NO: 206, CDR-L2 as depicted in SEQ ID NO: 207, and CDR-L3 as depicted in SEQ ID NO: 208; (o) CDR-H1 as depicted in SEQ ID NO: 214, CDR-H2 as depicted in SEQ ID NO: 215, CDR-H3 as depicted in SEQ ID NO: 216, CDR-L1 as depicted in SEQ ID NO: 217, CDR-L2 as depicted in SEQ ID NO: 218, and CDR-L3 as depicted in SEQ ID NO: 219; (p) CDR-H1 as depicted in SEQ ID NO: 226, CDR-H2 as depicted in SEQ ID NO: 203, CDR-H2 as depicted in SEQ ID NO: 204, CDR-H3 as depicted in SEQ ID NO: 205, CDR-L1 as depicted in SEQ ID NO: 206, CDR-L2 as depicted in SEQ ID NO: 207, and CDR-L3 as depicted in SEQ ID NO:

208. CDR-H2 as depicted in SEQ ID NO: 227, CDR-H3 as depicted in SEQ ID NO: 228, CDR-L1 as depicted in SEQ ID NO: 229, CDR-L2 as depicted in SEQ ID NO: 230, and CDR-L3 as depicted in SEQ ID NO: 231; and (q) CDR-H1 as depicted in SEQ ID NO: 238, CDR-H2 as depicted in SEQ ID NO: 239, CDR-H3 as depicted in SEQ ID NO: 240, CDR-L1 as depicted in SEQ ID NO: 241, CDR-L2 as depicted in SEQ ID NO: 242, and CDR-L3 as depicted in SEQ ID NO:

243.

28. The method as claimed in claim 1, wherein the biomass set point is maintained by feeding and discharging additional cells from the bioreactor at a defined cell-specific perfusion rate, and the perfusion culture is continuously operated for at least 7, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 or 28 days, or at least 35 days.

29. An apparatus for performing a continuous upstream manufacturing method as described in claim 1, the apparatus comprising a perfusion bioreactor, the first control loop, the second control loop, and an integrated unit.

30. A bispecific antibody product produced by the upstream manufacturing method as described in claim 1.

Images