Systems and methods for bone marrow extraction and cryopreservation

Abstract

Methods are provided for extracting bone marrow cells from bone obtained from a deceased donor, for preparing bone marrow for cryopreservation, and for obtaining desired cells from cryopreserved bone marrow and fresh bone marrow.

Description

[0001] Priority claim and citation of related applications

[0002] This application claims priority to co-pending U.S. Utility Application No. 16 / 734,713, filed on January 6, 2020, U.S. Provisional Application No. 62 / 834,087, filed on April 15, 2019, and entitled “SYSTEMS AND METHODS FOR COLLECTING AND STORING BONE MARROW FOR CLINICAL USE,” and co-pending U.S. Provisional Application No. 62 / 938,480, filed on November 21, 2019, and entitled “SYSTEMS AND METHODS FOR BONE MARROW EXTRACTION AND CRYOPRESERVATION.” The entire disclosures of all three applications are expressly incorporated herein by reference. Background Art

[0003] Currently, bone marrow for clinical purposes is harvested from HLA-matched siblings or best-matched unrelated donors. Other sources of transplants are now used, including unmatched haploidentical relatives or unrelated donors and umbilical cord blood (CB). When transplanted into patients with certain diseases, hematopoietic stem cells (HSCs) in the donor bone marrow engraft in the patient and reconstitute the immune and hematopoietic systems.

[0004] Bone marrow is also a good source of mesenchymal stromal / stem cells (MSCs), which are self-renewing multipotent progenitor cells with multipotential differentiation potential to differentiate into cell types of mesodermal origin, such as adipocytes, osteocytes, and chondrocytes. In addition, MSCs can migrate to sites of inflammation and exert potent immunosuppressive and anti-inflammatory effects through interactions with lymphocytes associated with the innate and adaptive immune systems.

[0005] Currently, bone marrow is usually collected using a trocar through a hole made in the cortical bone, and then a bone marrow aspiration needle and syringe are used to draw the marrow into the syringe. Multiple syringes are often required to extract all the marrow from the bone. The syringes are then removed from the sterile field and each syringe is connected to a collection bag containing an anticoagulant and the marrow is pushed into the bag. This step is often repeated multiple times in both lumbar bones and can result in contamination of the aspirate.

[0006] Sixty years ago it was recognized that stored whole bone marrow (BM) from deceased donors was also a very viable source of HSCs. Recovery of highly functional BM from deceased organ donors is conceptually similar to the procurement of organs and tissues that has been performed for decades, with more than 30,000 organ transplants and 1 million tissue transplants performed each year in the United States alone. Bone marrow HSCs are more robust than sensitive organs and most tissues because these cells have evolved to exist in the hypoxic environment within the BM niche and are therefore able to withstand prolonged ischemia. HSCs are typically in a quiescent (G0) state and therefore require few metabolic substrates and produce little waste. CD34+ HSCs and progenitor cells within deceased organ donor BM have been found to be highly viable. Published viability values ​​for CD34+ cells isolated from organ donor BM (even with non-optimized and unvalidated recovery and processing procedures) are 95.2%, compared to 93.5% for living donor BM. Deceased organ donors are a rich source of viable BM cells and are statistically indistinguishable from living donors in terms of CD34+ viability and total nucleated cells (TNC). The higher yield of CD34+ HSCs and the larger amount of BM from organ donors allow for the storage of multiple BM units (≥2 units, at approximately 2×10 6 CD34+ cells / kg, based on a 70 kg patient) for transplantation into multiple recipients and to ensure re-transplantation in the event of primary transplant failure.

[0007] However, multiple barriers prevent the mainstream use of cadaveric bone marrow. One important barrier is to find a rationalized process to control the extraction and preservation of deceased donor bone marrow and the cell yield from the bone marrow. Current best practices for recovering BM from cadaveric organ donors involve multiple manual steps and require multiple skilled operators. Typically, vertebral bodies (VBs) are recovered by transplant surgeons and first cleaned in the OR before being shipped to the processing laboratory, where they are cleaned again very carefully to remove all tough connective tissue remnants before further processing steps. Next, the VBs are processed in groups of 3, first manually cutting the bones into cubes and then feeding the cubes into a bone grinding system. The ground bone is then tumbled and rinsed multiple times, and finally the cells are concentrated by centrifugation. Since no more than 3 VBs can be processed at a time, this procedure must be repeated three times for each donor. This fully manual current process typically requires 40 hours of total labor, and nearly 11 hours of processing time, with a typical cost of more than $10,000 per donor.

[0008] Another issue regarding the use of cadaveric bone relates to the cryopreservation, storage, and recovery of bone. In particular, the problem relates to the quality of living cells (e.g., HSC) that can be obtained from donor bone, particularly for bones recovered from geographically dispersed locations and transported long distances to cold storage facilities. Each step of the process of recovering bone from a deceased donor involves ischemia, or lack of oxygen to the bone marrow cells. It is known that variations in cold ischemia and warm ischemia times can affect the quality of HSC and progenitor cells derived from cadaveric bone. Current tissue storage guidelines in the United States allow tissue to be recovered from deceased donors up to 24 hours after cardiac arrest, provided that the body is refrigerated within 12 hours after cardiac arrest. However, body cooling is a variable that has not been systematically studied in terms of its relationship to bone marrow recovery. Methods are needed to determine the tolerance limits of warm and cold ischemia, which, if exceeded, may render the quality and function of the recovered cells unusable for therapeutic purposes. Summary of the invention

[0009] The systems and methods disclosed herein provide a needed supplement to existing bone marrow and stem cell sources. Typically, less than half of patients waiting for allogeneic BM transplants receive a transplant. Living donor BM registries, BM cryopreservation and autologous transplants, and umbilical cord blood storage have provided life-saving solutions for thousands of patients with hematologic diseases; however, these methods are still subject to severe limitations related to supply and logistics, and will benefit from this valuable supplement. In addition, although it rarely occurs, adverse events may occur with living bone marrow donations (i.e., the risk of death associated with bone marrow donation is 1:10,000), and although peripheral blood stem cell donations are currently used more, almost all of these donors will experience bone pain, a quarter of the donors will experience significant headaches, nausea or citrate toxicity, and 1 / 5,000 of the donors will experience spleen rupture or other fatal complications. In addition, the long-term effects of stem cell mobilization agents are unclear. The technical feasibility of cadaveric BM storage and donation has been proven in principle, but these large-scale alternative supplies have been abandoned due to the problems directly solved by the present invention.

[0010] The stored BM disclosed herein provides a ready-made mechanism for matching many patients who cannot find a living donor. By allowing transplantation on demand and shortening the waiting time for these patients from several months to only 1-2 days, it can greatly improve the post-transplant survival rate of many patients with rapid disease progression and poor prognosis. And importantly, this method provides a large amount of BM from each donor, enough to achieve implantation of hematopoietic stem cells and progenitor cells (HSPC) to several patients, and can immediately perform repeated BM transplants when needed.

[0011] The methods and systems disclosed herein are capable of supplying large quantities of on-demand bone marrow for national emergency preparedness. The HHS, BARDA's multi-billion dollar biocontainment program, and the Department of Defense have all amply demonstrated the urgent and unmet need for on-demand bone marrow and stem cell transplantation as a medical countermeasure for nuclear accidents or nuclear attacks. The disclosure also provides bone marrow required for emerging applications such as immune tolerance induction. Protocols for processing BM and actual storage of BM from organ donors for long periods of time are critical to the method. In addition, if BM from these donors is stored, patients receiving deceased donor organ transplants today can benefit from this therapy (when such therapy is available in the future), making this method immediately beneficial to major organ transplant recipients. If successful, other promising methods and treatments being studied (including HLA-mismatched unrelated donor (mMUD) BM transplants) have the potential to greatly increase the value of cadaveric BM acquisition and storage using the proposed method, making large supplies of stored bone marrow immediately available to most recipients who need rapid BM transplants, especially to address severe forms of autoimmune diseases, genetic diseases, multiple sclerosis, and type 1 diabetes.

[0012] In one aspect, a method for obtaining bone marrow cells from a deceased donor bone is provided, comprising the steps of obtaining bone from a deceased donor; cleaning the bone of soft tissue; grinding the bone into bone blocks; filtering and washing the ground bone to produce a liquid composition; centrifuging the filtered and washed liquid composition of the ground bone to concentrate the bone marrow cells; and extracting the bone marrow cells into a sterile container for cryopreservation and subsequent isolation of target cells.

[0013] In yet another aspect, a bone cutting tool is provided for preparing bone for grinding in the above method. The bone cutting tool comprises two handles, a knife element, and a ratchet and pawl mechanism for driving the knife element into the bone, wherein the components are interconnected by an elongated pin passing through an opening in the respective components. The pin is removably retained by at least one removable clasp so that the bone cutting tool can be easily disassembled for cleaning and reassembled after cleaning. The bone cutting tool is made of medical grade stainless steel, wherein the surface is passivated to withstand a sterilization environment.

[0014] In another aspect, a method for recovering cells from the bone marrow of a deceased donor is provided, the method comprising the steps of obtaining bone from a deceased donor; processing the bone to extract bone marrow cells from the bone; obtaining a reduced density Ficoll solution having a density of 1.063-1.052 gm / mL; introducing the reduced density Ficoll solution into a centrifuge tube to form a Ficoll gradient; stratifying the extracted bone marrow cells over the Ficoll gradient in the centrifuge tube; centrifuging the tube containing the Ficoll gradient and bone marrow cells; harvesting buffy coat cells from the centrifuge tube; and washing the harvested cells for subsequent use or processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1A-Figure 1D is a view of a manual bone cutting tool according to one aspect of the present disclosure.

[0016] Figure 2 is an illustration of a filtration system according to one feature of the present disclosure.

[0017] Figure 3 is an illustration of a sterile bag containing bone marrow pellet processed according to the methods of the present disclosure.

[0018] Figure 4 yes Figure 3 View of a sterile bag with a clamp clamped on the bag to separate the fat from the bone marrow pellet.

[0019] Figure 5 The setup used to isolate the bone marrow pellet is shown.

[0020] Figure 6 is a perspective view of a cooling box according to an aspect of the present disclosure.

[0021] Figure 7 is a flowchart of the steps of a method according to the present disclosure

[0022] Fig. 8A and Figure 8B are side and perspective views of an automated bone processing system according to one aspect of the present disclosure.

[0023] Fig. 9A and Fig. 9B is a perspective view of a bone clearing station of the system shown in Figures 78A-8B.

[0024] Fig. 10A and Fig. 10B yes Fig. 8A , Figure 8B Perspective and front views of the bone grinding station of the system shown.

[0025] Fig.11 yes Fig. 8A , Figure 8BA perspective view of a screening station of the system shown.

[0026] Figure 12A-12C is a table showing CD34+ cell viability as a function of warm and cold ischemia time with and without body cooling.

[0027] Figure 13A-13C is a table of CFU-total as a function of warm and cold ischemia time with and without body cooling.

[0028] Figure 14A-Figure 14C is a table of CFU-total as a function of warm and cold ischemia time with and without body cooling.

[0029] Fig.15 is a graph of viability threshold as a function of warm and cold ischemia times. DETAILED DESCRIPTION

[0030] For the purpose of promoting an understanding of the principles of the present disclosure, reference will now be made to the embodiments illustrated in the accompanying drawings and described in the following written specification. It should be understood that it is not intended to limit the scope of the present disclosure thereby. It should also be understood that the present disclosure includes any changes and modifications to the illustrated embodiments, and includes further applications of the principles disclosed herein as would normally occur to a person skilled in the art to which the present disclosure pertains.

[0031] The present disclosure provides a clinically oriented research protocol and system that has been modified to be implemented in an industrial environment within a state-of-the-art clean room. Among other things, one aspect of the disclosed system involves debridement of ready-to-use donor bone, initial fragmentation using custom surgical stainless steel cutters, and grinding of the fragmented bone to a fragment size of approximately 3 mm. These modifications provide a system in which a skilled tissue processing technician can process multiple sets of donor bones within a 6-hour time window to produce meaningful amounts of usable marrow.

[0032] The first step of the process described herein is to evaluate the potential source of deceased donor bone marrow. When processing long bones (e.g., tibia) from donors, it has been found that due to the conversion of red bone marrow into yellow with age, red bone marrow is limited to the ends of long bones and varies greatly between donors. It has also been determined that compared with complete red bone marrow (e.g., bone marrow from vertebral bodies or ilium), mixed yellow-red bone marrow is of poor quality, and mixed yellow-red bone marrow contains fat infiltration, complicating subsequent processing steps. Compared with cells obtained from the ilium of the same donor, the best donor long bones in some clinical experiments only produce 1 / 100BM cells / kg. Therefore, it has been determined that long bone processing is preferably performed only in special cases, such as involving additional valuable "universal" HLA types or bone marrow with HIV-resistant δ32 mutations.

[0033] In contrast, the vertebral bodies and iliac bones represent the largest consistent high-quality red marrow pools. Utilizing these two sources optimizes bone marrow recovery, particularly by implementing the industrialized, scalable GMP processes disclosed herein. Completion of the processes disclosed herein results in a cryopreserved final product configuration storing 60-70 mL volumes at a target of 100-150 million TNC / mL in standard blood bags, similar to product configurations that have been used for cryopreserved BM for autologous transplantation.

[0034] Preparation of donor bone

[0035] For purposes of illustration, it is assumed that the donor bone is a vertebral body. However, it should be understood that the methods described herein can be used for ilium, a combination of vertebral bodies and ilium, or other bones suitable for extracting bone marrow and cells from bone marrow, even donor bones with a lower expected yield.

[0036] It is understood that donor bone can be obtained according to a fixed protocol for clinical retrieval. The surgeon or trained OPO (Organ Procurement Organization) personnel can use an osteotome and mallet to retrieve bone from a consented organ and tissue donor. Unprocessed bone is preferably wrapped with a sponge and towel soaked in saline to ensure moisture retention during cryogenic transportation on wet ice at 0-10°F to the processing facility.

[0037] The process of preparing the donor bone can be performed shortly after the bone is obtained from the deceased donor, or after the donor bone is transported to a processing facility in a cryogenic environment. Because the donor bone may experience prolonged ischemia during recovery and transportation to the processing facility, care must be taken to track the duration and type of ischemia - i.e., warm ischemia and cold ischemia. As described in more detail herein, bones that have been subjected to a predetermined period of warm and / or cold ischemia are suitable for obtaining a meaningful amount of viable bone marrow cells.

[0038] In the first step of processing the donor bone, the bone is cleared in an ISO-5 (class 100) environment (biosafety cabinet) with an ISO-7 (class 10,000) background (clean room), and special care is taken to sterilize the bag containing the donor bone, such as by spraying 70% isopropyl alcohol. In one embodiment, the clearing is performed manually using a scalpel, osteotome, and gouge. When processing vertebrae, a spinal segment comprising multiple vertebral levels will generally be provided. In a typical case, the spinal segment is taken from T8 to L5, ten vertebral bodies. During the initial clearing of the spinal segment, when enough soft tissue has been removed to visualize the pedicles, the pedicles are removed using a tissue processing band saw or a bone saw (e.g., a Stryker System 6 Saw (Stryker, Kalamazoo, MI). Special attention is paid to avoid damaging the cortical bone that will expose the cancellous bone to ensure that the anoxic cancellous bone marrow is protected throughout the clearing process. The anterior elements of the vertebral body are retained, while the pedicles and posterior elements are discarded.

[0039] Using a boning knife or tissue handling band saw, separate the vertebral bodies at the intervertebral disc. Remove the intervertebral disc and soft tissue remaining on each vertebral body with a scalpel, scissors, and / or osteotome, leaving a clean, isolated VB. In the case of donor iliac crests, the soft tissue can be removed with a gouge and scalpel, again with extreme care to ensure that the cortical bone is not disrupted. Observe and record any anatomic pathology or damage to the bone as part of the batch record of the bone marrow ultimately obtained from the bone. Discard damaged bone during the recovery process.

[0040] The VB is placed in a sterile bag and immersed in a 10% bleach solution (producing a 5,000 ppm free chlorine concentration) for a predetermined period of time, typically 10-25 minutes. Bleach has broad-spectrum antimicrobial activity, leaves no toxic residues, is not affected by water hardness and is fast-acting. At the end of this period, the bone is transferred to another sterile bag and immersed in a 3% hydrogen peroxide (H2O2) solution. The bag is sealed and briefly shaken to ensure that the entire surface of the bone is in contact with the solution. Most living cells contain catalase, an enzyme that catalyzes the decomposition of H2O2 into H2O and O2. When the H2O2 solution contacts soft tissue but not bone, this decomposition manifests as foam or foam-like substances. The foam level can be observed as an indication of the amount of soft tissue remaining on the bone. This observation can be performed manually by a manual processor, or in another embodiment, by an automated processor. The automated processor combines a visualization device, such as a camera, and object recognition software that can determine the foam level within the bag. The addition of an inert contrast dye can help a manual or automated processor detect the foam level. If any foam or foamy material was observed, the bone was sent back for further processing to remove all remaining soft tissue from the bone.

[0041] Once all soft tissue of the VB or ilium has been cleaned, the bone is transferred to a new sterile bag. The bag contains 1 L of PLASMA-LYTE TM (Multiple Electrolyte Injection from Baxter Healthcare, Ltd.) or other suitable sterile, pyrogen-free, isotonic solution. Close the bag and shake briefly to ensure that the entire bone is completely infused with PLASMA-LYTE. TM touch.

[0042] Extraction of bone marrow

[0043] FROM BAG AND PLASMA-LYTE TM Remove the bone from the VB and use sterile gauze or sponges to absorb any fluid remaining on the VB. In one method, the VB is cut into smaller pieces, such as 1.5 cm, using a saw and / or anvil shears. 2 In order to simplify the process and improve the safety of the handlers, the following are provided: Figures 1A-1DThe custom bone cutting tool 100 shown is used to cut VB into smaller pieces. The bone cutting tool 100 includes a knife member 102 having a blade 102a configured to penetrate and cut bone. The knife member 102 is pivotally connected to a fixed handle member 104 at a pivot 105. The fixed handle member 104 includes a jaw end 104a, which is juxtaposed with the blade 102a to cut through the bone held in the jaw end. Figure 1B As shown, the fixed handle member includes two plates 104d which are spaced apart to receive the knife member therebetween, such as Figure 1B-1D 4. As best shown in FIG. 4, the blade 102a passes between two plates 104d at the jaw end 104a, which ensures that the blade 102a passes through the bone captured by the jaw end 104a. The jaw end 104a may include two recesses 104c separated by a ridge 104b, which engages the bone and helps to retain the bone at the jaw end as the blade 102a passes through the bone. Alternatively, a single recess may be defined at the jaw end, which is configured to retain the bone. The pivot 105 is in the form of a pin that extends through the two plates 104d and the knife member 102 sandwiched between the plates.

[0044] The bone cutting tool 100 includes a lever handle 107 pivotally mounted to the fixed handle 104 at a pivot 109. The pivot may include a biasing element, such as a torsion spring (not shown), which is configured to bias the lever handle 107 away from the fixed handle 104. The lever handle is thus configured to pivot toward the fixed handle when the user grasps and squeezes the two handles, and then pivot away from the fixed handle when the user releases the grip on the handles. It will be appreciated that the lever handle 107 is formed by two plates 107a, wherein the fixed handle 104 is sandwiched between the two plates 107a at the pivot 109. Like the pivot 105, the pivot 109 is in the form of a pin that extends through the two plates 107a and through the fixed handle 104. The two handles 104, 107 include respective gripping plates 106, 108, which are contoured to be grasped by the palm and fingers of the user. Gripping plates 106, 108 connect the paired plates 104d, 107a, which form two handles. The surface of the gripping plates may include anti-slip features to facilitate gripping of the tool.

[0045] The lever handle 107 includes a pawl 112 pivotally mounted to the lever handle at a pivot 113. Like the other pivots, the pivot 113 is a pin that extends through the pair of plates 107a forming the lever handle 107 and through the end of the pawl 112. The pivot 113 includes a biasing element, such as a torsion spring (not shown), which biases the pawl 112 toward the ratchet member 110 of the knife element 102. The end of the pawl 112 is configured to engage with the teeth 110a on the ratchet member 110 to rotate the ratchet member in a counterclockwise direction, thereby rotating the knife element 102 in a counterclockwise direction, as shown in FIG. Figure 1AAs shown. In particular, when the user squeezes the two handles together, the lever handle 107 moves toward the fixed handle 104, which pushes the pawl 112 upward against the teeth 110a of the ratchet to pivot the ratchet upward and counterclockwise. When the user releases the lever handle, the handle moves away from the fixed handle, causing the pawl 112 to slide down the ratchet in a clockwise direction until it reaches another tooth 112a. Repeated squeezing of the two handles causes the pawl to rotate the ratchet successively. The knife element 102 also includes an integral link 103, which is pivotally connected to a free link 114 at a pivot 116. The free link 114 is pivotally connected to the lever handle 107 at a pivot 115. When the pawl traverses the ratchet component 110, the integral link 103 and the free link 114 keep the knife element 102 in place. The pivot 115 of the free link is a pin, as are the other pivots, and includes a biasing element, such as a torsion spring (not shown), which biases the lever handle 107 away from the fixed handle 104. This allows the user to close and release the handle of the tool to successively advance the pawls 112 along the ratchet member 110, which successively advances the blades 102a into the bone.

[0046] In one feature of the bone cutting tool 100 of the present disclosure, the pivots 103, 109, 113, and 115 are configured to allow for complete disassembly of the tool. Complete disassembly is important for complete cleaning and sterilization of the tool between uses. Therefore, the pivots each include a pin and a clasp structure, wherein the clasp holds the components together on the pin. Thus, as Figure 1D As shown, the pin 121 can extend through the wall of the component, such as the opposite wall 107a of the lever handle 107 and through the connected component, such as the corresponding hole in the knife element 102. The retaining ring 122 can engage with the groove 123 at the opposite end of the pin 121 to hold the components together. Optionally, one end of the pin can have an enlarged head, wherein the retaining ring engages the opposite end of the pin. When it is necessary to clean and disinfect the tool 100, all the retaining rings 122 can be removed, all the pins 121 can be removed, and the connected components can be separated. The knife element 102, the fixed handle 104 and the lever handle 107 are thus separated, so that each surface of the components can be effectively cleaned.

[0047] The elements of the bone cutting tool 100 are formed of medical grade stainless steel. The steel is preferably a hardened steel that can withstand the forces required to cut through bone. During the cleaning process, the tool is steam sterilized, which may be harmful to the steel. Therefore, in one feature of the present disclosure, the surface of the stainless steel elements is passivated to prevent oxidation of the steel elements during the sterilization process.

[0048] Returning to the process steps, particularly the step of extracting the bone marrow, the mass produced by the bone cutting tool is immediately placed in a sterile jar and immersed in 300-500 mL of grinding media. In one aspect of the present system and method, the grinding media is PLASMA-LYTETM -A as a matrix, containing 10U / mL heparin, 2.5% human serum albumin (HSA) and 3U / mL Reagent (Merck KGAA). Heparin is used as an anticoagulant. HSA provides a protein source to prevent cell adhesion and adsorption to the surface, as well as to scavenge reactive oxygen species. It is worth noting that conventional grinding media use DNase, but for the present disclosure, Reagents to replace DNase TM Reagent (Qiagen Sciences LLC). DNases only work on DNA, but modern pharmaceutical biotechnology processes rely on enzymes that can cleave all forms of DNA and RNA and can reduce the viscosity of the solution in which the cells are suspended. Of note, IMDM (Iscove's Modified Dulbecco's Medium) can be used as an alternative to PLASMA-LYTE TM -A, because IMDM is suitable for high-density cell culture with rapid proliferation and is ideal for supporting T and B lymphocytes. Also note that Reagents (C-Lecta GmbH) are equivalent to the same amount of Reagents. Reserve another jar of 300-500 mL of grinding media for collecting the ground bone fragments, and reserve another portion of approximately 100 mL of grinding media for flushing through the grinder during the grinding process to prevent bone fragments from sticking to the jar surface of the grinding component.

[0049] An electric bone grinder or a dedicated bone grinder, such as that of Biorep Technologies Inc (Miami, FL), can be used in an ISO-5 environment within an ISO-7 clean room. If VB and iliac crests from the same donor are being processed, the bone types remain separate. The bones remain submerged in the grinding media at all times during and after the grinding process. Once all donor bone pieces have been ground, the chamber of the bone grinder is thoroughly flushed with fresh processing media. The bone fragments are discharged from the grinder into a canister containing the grinding media.

[0050] Transfer the contents of the tank to a sterile bag. In the next step, filter the contents of the sterile bag to extract the solid components. In one embodiment, the contents of each bag are passed through a series of stainless steel sieves. In this embodiment, a No. 40 (425 μm) sieve is stacked on top of a No. 80 (177 μm) sieve, which is located above a receiving tray for receiving the contents of the liquid filter. The sterile bag containing the output from the grinder is rotated and then poured evenly onto the sieve stack or filter group. Observe the filtration process to ensure that excessive agglomeration does not occur (this may indicate the presence of soft tissue or other contaminants). The bone fragments retained on the surface of the sieve are evenly distributed on the sieve and rinsed with 250 mL of fresh processing medium. In one embodiment, the processing medium used for rinsing is the above-mentioned grinding medium or PLASMA-LYTE containing 2.5% HSA. TM The sieved bone marrow product (approximately 1000 mL in a well-run process) is transferred to sterile packaging for subsequent processing and analysis. The contents of each bag are visually inspected to confirm that the contents do not include any visible bone fragments or soft tissue.

[0051] In another embodiment, the contents of each bag are passed through a bone marrow filtration unit, such as Figure 2 As shown. In this embodiment, the system 150 includes a support 154 configured to support a sterile collection bag 152 containing bone fragments and media from the above-mentioned grinding operation. The support includes a container hanger 155, which is configured to engage the lid 153 of the sterile bag to hang the container. The bottom of the bag includes a discharge assembly 160, which includes a pre-filter 162 protruding into the body of the collection bag. In a specific embodiment, the pre-filter 162 is an 850μm filter. The filter 162 is connected to an output tube 164, which is connected to the input line 171 of the first series filter 170 through a container clamp 166. In a specific embodiment, the first series filter is a 200μm or 500μm filter. The output line 172 of the first series filter is connected to the input line 176 of the second series filter 175. The second series filter is a 200μm or 500μm filter. For the first pass through the filtration system 150, both series filters are initially 500μm. The grind is then rinsed a second time using two 200 μm filters in series. This double pass filtration produces a cleaner suspension and enhances fat removal from the suspension. The second inline filter 175 has an output line 177 that can be coupled to a sterile bag, such as bag 152 for the second filtration pass. On the second pass through the system, the output line 177 of the second inline filter 175 can be coupled to a container fixture 181 of a transfer packaging container 180. The transfer packaging container can be a 600-2000 mL bag to hold the filtered bone marrow product, which can be about 1000 mL in a well-run process.

[0052] For quality control, a small amount of bone marrow, e.g., 0.3 mL, is extracted from the sterile package 152 at the injection site 157 using a syringe and mixed by inversion before withdrawing the sample. The sample can be tested by a hematology analyzer (e.g., a Sysmex hematology analyzer) to determine the TNC (total nucleated cell) content of the sample as an indicator of the TNC content of the bone marrow that is subsequently processed.

[0053] Fat removal and concentration

[0054] The bone marrow product collected from the filtration step is essentially a fat emulsion. The fat content of the suspension obtained from the sieving method disclosed above is greater than the fat content of the suspension obtained from the double-pass filtration system 150. However, in both cases, it is necessary to remove the fat content from the suspension. The suspension obtained from the filtration step is recovered into a 250 mL bag, which is sealed with a tube welder. Pairs of sterile bags and balancing rods are installed in the centrifuge with the bag opening facing down and kept balanced. A volume compensation plate is used to prevent wrinkling of the bag during centrifugation. In one embodiment, the bag is centrifuged at 500×g for 15 minutes at room temperature to concentrate the cells, preferably to 2-3×10 8 / mL. After centrifugation is complete, each bag is hung individually on a ring stand. The different layers within the bag are visible, with the fat layer clearly shown at the top of the supernatant and the bone marrow pellet at the bottom, as shown in Figure 1. Figure 3 As shown. Attach a new sterile bag to the bag removed from the centrifuge. Place a bag clamp or clip 190 on the bag just below the fat layer, as shown. Figure 4 As shown, the bag is clamped or squeezed closed under the fat layer. The sediment is then drained from the centrifuge bag into a new sterile bag, wherein the bag clamp prevents the fat layer from passing through. The sediment is stirred as it is drained to resuspend all of the sediment. After about half of the sediment has drained into the new bag, the tube is closed with a hemostat or tube sealer. A second centrifuge bag is then attached to the new bag containing the sediment, and the contents of the second centrifuge bag are drained into the new bag.

[0055] The result of this step is that new sterile bags containing bone marrow are centrifuged to remove fat. These bags of defatted bone marrow are then centrifuged at 500 x g for 15 minutes at room temperature, using a volume compensation plate to prevent wrinkling of the bags. Each bag is removed and hung on a ring stand, a waste bag is attached to the bag, and the supernatant is removed to the waste bag using a plasma extractor, such as Figure 5As shown. When the precipitate rises or breaks, clamp the tube with a hemostat. The tube is then sealed and cut to remove the bag containing the precipitate from the waste bag and discard the waste bag. Connect the Luer connector to the bag containing the precipitate. Use a large syringe to merge the precipitate in each bag into the bulk bag. Use flushing medium to rinse the bag containing the precipitate into the bulk bag. Invert the bulk bag several times to ensure that all the precipitate is resuspended. A small amount of processed BM, such as 0.5 mL, can be taken out for quality control testing of density and cell count. The test sample can also be evaluated for human leukocyte antigen, CCR5δ32 mutation, and apolipoprotein (APOE), etc.

[0056] Bone marrow cryopreservation

[0057] Based on ten vertebrae and iliac bones obtained from donors, it is expected that each bone donor can produce three or more bags of marrow through the above process. If three bags of marrow are not obtained at the end of the process for a given donor, the donor may be marked as possibly failing overall quality control. It is expected that there is a predetermined volume of marrow in each bag, for example, 70 mL in a 250 mL bag. The predetermined volume is used to calculate the volume of the freezing medium component required for effectively cryopreserving the bone marrow precipitate. The freezing medium is a solution of a flushing medium and a cryopreservation composition. The cryoprotectant composition can be a permeable medium, such as dimethyl sulfoxide (DMSO); 1,2 propanediol; ethylene glycol; glycerol; formamide; glycol or butane 2,3 diol; and / or a non-permeable medium, such as hydroxyethyl starch (HES), dextran, sucrose, trehalose, lactose, raffinose, ribitol, mannitol or polyvinyl pyrrolidone (PVP). 2.5% HSA also provides cryoprotection, cell surface protein stabilization and reactive oxygen scavenging by osmotic pressure. In a preferred embodiment, the cryopreservation medium is DMSO. The flushing medium can be an electrolyte medium, such as PlasmaLyte, Isolyte, IMDM or other electrolyte solutions suitable for perfusion. The freezing medium can also include a concentration of an oxidase to reduce the oxygen content to below atmospheric pressure, for example, to below 3% of the atmospheric pressure concentration. Adding an oxidase produces a low specific gravity composition that can facilitate cryopreservation.

[0058] The freezing medium is prepared by mixing the cryoprotectant and flushing medium based on the calculated total volume of freezing medium required for the volume of bone marrow collected in the previous step. The bag containing the bone marrow is placed on a rocker for mixing, and the freezing medium is introduced into the bag by syringe. The freezing medium is introduced at a specific rate over a predetermined time. In one embodiment, the freezing medium is added at a rate of 10% of the medium per minute for a period of ten minutes. Once the medium is mixed with the concentrated bone marrow, a test sample is extracted by syringe. The remaining mixture of freezing medium and bone marrow is injected into separate cryopreservation bags in predetermined amounts. In one embodiment, 70 mL of the bone marrow mixture is introduced into each cryopreservation bag and the air is aspirated with a syringe. At the end of this process, 8 mL of sample can be removed for sterility testing. Each cryopreservation bag is sealed to form four compartments, which are then separated for storage in a dark box and then stored in a low-temperature freezer. In another embodiment, the separated compartments are stored in a passive cooling box, such as Figure 6 In the cooling box 200 shown.

[0059] When the cell count and sterility of the test sample from a specific marrow batch are verified, the frozen bone marrow bag can be further cooled for long-term storage. In one embodiment, the bag is cooled at a controlled rate to prevent damage to the marrow and cells. In a specific embodiment, the bag is cooled to a temperature suitable for the bag to be put into liquid nitrogen at a rate of -1°C to -40°C per minute. A suitable temperature range is -40 to -100°C. Once the temperature is reached, the bag is further cooled to a temperature below -130°C at a faster rate for storage. The cryopreservation bag is placed in the corresponding compartments 201-203 of the cooling box 200 and the lap cover 205 is closed above the compartment to provide a sealed environment for the contents of the cryopreservation bag. The cooling box is placed in a low-temperature freezer, so that the cooling box produces a cooling rate of -0.5°C / min to -2°C / min (and typically -1°C / min), wherein the nucleation temperature is higher than -20°C. The freezing process continues at a specified rate until the temperature of the marrow reaches a suitable temperature. The suitable temperature for storing the bags is ≤-80℃ or ≤-150℃.

[0060] In another embodiment, the bag is cooled at static room temperature, as opposed to the controlled rate cryopreservation described above. In the passive cooling method, the cooling box is placed in a -86°C freezer until the bag reaches a stable temperature.

[0061] It is contemplated that cryopreservation can be in a variety of forms. For example, cryopreserved bone marrow can be contained in a bag of 1 mL to 5 mL volume or a vial of 0.1-15 mL volume. In a preferred embodiment, a bag with 70 mL of bone marrow is stored in a cooling box in a cryogenic freezer.

[0062] Cryopreserved bone marrow is frozen and stored for later thawing and extraction of desired cells. Thawed bone marrow can be provided for a wide range of treatments, including treatment of leukemia, brain tumors, breast cancer, Hodgkin's disease, multiple myeloma, neuroblastoma, non-Hodgkin's lymphoma, blood cancer, ovarian cancer, sarcoma, testicular cancer, other solid organ cancers, rheumatoid arthritis, multiple sclerosis, diabetes, cystic fibrosis, Alzheimer's disease, inherited immunodeficiency, metabolic disorders, bone marrow failure syndrome and HIV. Bone marrow can also be used to induce immune tolerance to reduce the potential rejection of implants obtained from organ donors. Bone marrow therapy can also be applied to casualties caused by radiation and certain biological weapons.

[0063] Bone marrow is a well-known mesenchymal stromal / stem cell (MSC) source, which can be harvested from previously frozen stored organs and tissue donor bone marrow using the above method. MSC is a self-renewing multipotent progenitor cell with multidirectional differentiation potential, to differentiate into mesodermal derived cell types, such as adipocytes, osteocytes and chondrocytes. In addition, MSC can migrate to the site of inflammation, and play an effective immunosuppressive and anti-inflammatory effect by the interaction between the lymphocytes associated with the innate immune system and the adaptive immune system. MSC can be used to treat osteogenesis imperfecta, cartilage defects, myocardial infarction, Crohn's disease, multiple sclerosis, autoimmune diseases (such as lupus), cirrhosis, osteoarthritis and rheumatoid arthritis. Matched HSC / MSC units can be used for combined transplantation to treat graft versus host disease (GVHD) and hematopoietic stem cell transplantation support.

[0064] The present method provides a system for extracting and storing bone marrow for future clinical use according to the above process steps, such as Figure 7 This approach could eliminate the failure of current methods to match bone marrow donors to hard-to-match groups, such as certain ethnic minorities. Once the bone marrow is cryopreserved and stored, there is no uncertainty about the source of the bone marrow, no need to wait for future recipients, and the bone marrow is available in large quantities for repeated use.

[0065] Automated system for bone marrow recovery

[0066] The present disclosure contemplates automated processes for recovering bone marrow, and even selecting cells from bone marrow. In one aspect, the automated system 209 includes sequential stations, such as Figure 8A-8B The first station 210 of the automated process cleans the VBs to remove all soft tissue. In contrast to the manual process, which operates on one VB at a time, the automated process is configured to clean the entire set of donor VBs (which may be at least ten vertebral bodies). The VBs are mounted on a rack or tray 212, which is configured to support the set of vertebral bodies from a given donor. Figures 9A-9BAs shown, the tray 212 is placed on a conveyor track 216 of a housing 215, wherein the tray is automatically or manually advanced into the interior of the housing. The housing 215 supports a plurality of hydraulic ejectors 220, which direct high-pressure and high-velocity saline jets onto the VB. In known manual processes, manual hydraulic ejectors operating at lower speeds and pressures direct a stream of detergent onto the VB. In manual processes, a detergent is required to clean the soft tissue off the VB. In contrast, the automatic cleaning station 210 of the present disclosure uses a saline medium, wherein the velocity and pressure of the water jet are sufficient to remove all soft tissue from the VB. The automatic cleaning station of the present disclosure includes ejectors configured to produce a direct stream or narrow "V" shaped water / saline jet of high concentrated impact force at different distances. In order to achieve good coverage of the VB, the device includes a number of closely spaced direct ejectors in different directions relative to the VB, which allows for uniform cleaning independent of the location of the VB in the device. Fig. 9A In the embodiment shown, the hydro jets are arranged in an upper row 220 and a lower row 221. The "V" shaped jets are oriented at different angles to achieve complete coverage of the VB surface. Additionally or alternatively, the hydro jets 220, 221 can be configured to swing over the VB tray to ensure complete coverage.

[0067] A visualization device 225 is disposed at the exit of the removal station 210 and is operable to visualize and interpret the VB exiting the station to determine whether all soft tissue has been removed, such as Fig. 9B As shown. If not, the VB is returned to the housing along track 216 for further hydrojet treatment. It is expected that a controller (not shown) may be provided to control the movement of the tray 212 along track 216 and interpret the signals generated by the visualization device 225. The visualization device may include a camera that obtains an image of the VB and the controller may include imaging software capable of identifying soft tissue in the acquired image. At the end of the hydrojet cleaning process, a dye may be applied to the cleaned VB, wherein the dye is absorbed by the soft tissue but not by the bone. Thus, the dye can provide contrast to facilitate distinguishing any remaining soft tissue from the bone. The visualization device 225 can be configured to move the camera throughout the VB (e.g., by translating along the frame 226 and by translating the frame) to view the VB at all angles.

[0068] Back to Figure 8A-8B, once it is determined that all soft tissue has been cleared from the VB, the cleared VB is then fed by conveyor belt 230 to an automated grinding station 240 to produce appropriately sized chunks for tumbling and eventual cell extraction. The manual "cutting" process described above can be variable, time consuming, and potentially unsafe for the operator. The automated system includes a grinding station that combines the steps of "cutting" the VB (i.e., cutting the VB into small chunks) and grinding the cut VB to reduce the VB to 2-3 mm chunks. The track 216 and tray 212 can be configured to place the cleared VB on the conveyor belt 230, which then automatically transports the VB to an input hopper 242 of the grinding station 240, more details as shown. Figures 10A-10B VB is guided through the initial milling cutter module 244 and then through the funnel 246 to the fine milling cutter module 248, as shown. Fig. 10A As shown. Fig. 10B As shown, the initial milling module 242 includes opposed rotating grinders 245 separated by a predetermined gap (e.g., 5-8 mm gap) so that the incoming VB is ground into coarse size segments. The coarsely ground segments are fed to the fine milling module 248, in which a smaller diameter grinder 249 is provided. The fine grinders 249 are separated by a smaller gap (approximately 2-3 mm) to produce finely ground VB segments. Fig. 10A As shown, hopper 246 delivers the coarsely ground segments to a second grinder 248, and hopper 250 directs the finely ground VB segments to a collection tray 252 supported on a plate 253. During the milling operation, a measured volume of a treatment / resuspension medium containing DNAse can be directed through the upper hopper to the milling tool. The medium can be introduced manually during operation of the milling station 240, or can be automatically implemented by a nozzle incorporated into the hopper 242.

[0069] The finely ground VB segments and process media are collected in the collection tray 252 and the plate 253 can be moved manually or automatically to the screen station 260 ( Figure 8A-8B Once at the screen station 260, the contents of the pan 252 fall into a screen drum unit that includes two 12" diameter filter screens - a top #40 screen 262, followed by a finer #80 screen 264, such as Fig.11As shown. Funnel 266 directs the filtered contents to collection container 268. The grounds retained by the filter are rinsed in screening station 260 with a treatment / resuspending medium that does not include DNAse. As described below, the liquid bone marrow product in collection container 268 can be analyzed to determine the cell content, and then concentrated and packaged into a suitable volume for cryopreservation. Alternatively, some or all of the processed bone marrow can be further processed using an automated cell selection method for specialized cell products, such as CD34+ cells. Since a large number of cells can be recovered from a single organ donor using this method, a single donor can produce a variety of product types. In addition, since the source is primary bone marrow (as opposed to peripheral blood mobilized by G-CSF), the cell product will tolerate cryopreservation treatment.

[0070] In one modification, the output from the grinding station 240 or the screening station 260 can be automatically fed to a collection bag for cryogenic processing. In this modification, the lower funnel 250 can be configured to direct the contents to a fluid line connected to a sterile bag. A peristaltic pump can engage the fluid line to pump the output from the grinding station to the sterile bag. A similar arrangement can be engaged to the funnel 266 of the screening station.

[0071] The contents of collection container 268, which is essentially bone marrow pulp, are transferred manually or automatically to an adjacent drum station 270, which includes a mechanical drum 272 and a large disposable container 274 that can contain the entire contents of ten processed VBs and associated processing / resuspension media. Drum 272 has paddles for agitating the grinding pulp to mechanically release the cells. When the tumbling cycle is complete, the contents of the drum are poured through a screen box into container 274. The contents of container 274 can be further processed or prepared for cryogenic storage.

[0072] Isolation of cells from bone marrow

[0073] In one aspect of the present disclosure, a method is provided for selecting CD34 expressing (CD34+) cells from deceased donor bone marrow using a reduced density Ficoll and an immunomagnetic CD34+ cell separation kit. Surprisingly, it has been found that cell separation using a reduced density Ficoll prior to CD34 selection is beneficial for obtaining high purity and high viability CD45 / CD34+ cells from freshly prepared deceased donor bone marrow. On the other hand, it has been found that a conventional density Ficoll is optimal for selecting CD45 / CD34+ cells from thawed cryopreserved deceased donor bone marrow.

[0074] Vertebral sections obtained from a recently deceased donor are processed as described above. Thus, in one embodiment, the bone is cleaned of all soft tissue and then cut into small pieces that are immediately immersed in 500 mL of grinding media. The grinding media can be PLASMA-LYTE TM A injection pH 7.4, multiple electrolytes, Type 1 injection USP (PLASMA-LYTE TM ), 3U / mL denarase and 10U / mL heparin. The sliced ​​VB was ground using a bone grinder, filtered and washed with a washing medium (e.g., PLASMA-LYTE containing 2.5% HSA). TM ) was washed. The entire cell suspension was centrifuged to concentrate the cells to 2-3×10 8 / mL and extract the cell concentrate. A portion or all of the resulting BM preparation can be used immediately for CD34 selection, while the remainder can be used for cryopreservation. The cryopreservation portion involves adding DMSO and 5% HSA to the BM cells at a final concentration of 10%, and bringing the preparation to -86°C by passive cooling or controlled cooling at a rate of approximately -1°C / min, and then plunging the cryopreserved portion into liquid nitrogen.

[0075] For the selection of CD34+ cells, a newly processed BM preparation is used, or a previously cryopreserved portion is thawed and used. Ficoll-Paque PLUS is added to the BM preparation to separate the desired CD34+ cell components in the bone marrow. It has been found that for selecting cells from cryopreserved bone marrow, the conventional density of Ficoll of 1.077 g / mL produces acceptable results. However, in one aspect of the present disclosure, for cell selection from freshly prepared deceased donor bone marrow, the Ficoll density is lower than the conventional density. In particular, the density is reduced by mixing Ficoll-Paque PLUS (density 1.077 g / mL, GE) with Plasma Lyte-A injection pH 7.4 (Baxter Healthcare 2B2544X) in a specific ratio to obtain an overall density of less than 1.077 g / mL, especially an overall density of 1.063-1.052 g / mL. In a specific embodiment, considering the amount, viability and purity of CD34+ cells, it is found that a density of 1.063 g / mL is optimal for separating CD34+ cells.

[0076] In one embodiment, 5mL 1.063g / mL density Ficoll solution is pipetted in a 15mL centrifuge tube, and the BM solution produced from the VB of the deceased donor is carefully layered on the Ficoll gradient. The tube is centrifuged at room temperature for 30 minutes at 400g without rupture. After centrifugation, the buffy coat cells are carefully collected, and washed cells in a phosphate buffered saline (PBS) (MACS buffer, Miltenyi) containing 0.5% HSA and 2mM ethylenediaminetetraacetic acid (EDTA). In a specific embodiment, the obtained cell pellet is resuspended in 10mL PBS with 400g centrifugation for 5 minutes, and then centrifuged for a second time for 5 minutes with 400g.

[0077] The nucleated cells in the separated buffy coat can be counted using Sysmex XP-300. Cellometer Vision (Nexcellom) or flow cytometer can be used to determine the cell count of purified CD34 cells. 20 microliters of AOPI can be added to 20 microliters of cells, and the total viable cells can be determined after mixing. CD34+ cells can be selected by positive immunoisolation method according to the manufacturer's protocol using CliniMAX system (Miltenyi, Bergisch Gladbach, Germany) or EasySep CD34 kit (Stemcell Technologies, Vancouver, BC, Canada). According to the test under various Ficoll densities, it was surprisingly determined that the lower Ficoll density (i.e., 1.063-1.052gm / mL relative to the conventional 1.077gm / mL density) envisioned in the present disclosure produces more optimized cell recovery. Optimization is based on the purity, viability and yield of the selected CD34 cells. The goal of >90% purity and >90% viable CD34+ cells is preferred. Although lower Ficoll density produces higher purity and fewer dead cells, it was surprisingly found that using lower Ficoll density to prepare buffy coats, a greater portion of CD34+ cells present in whole bone marrow of deceased donors before selection were lost. Therefore, the high viability and purity of CD45 / CD34+ cells obtained under conventional Ficoll density gradients also resulted in a substantial loss in yield (about 60% loss of input CD34+ cells).

[0078] Therefore, according to one aspect of the present disclosure, for fresh preparation, the optimal density of Ficoll for selecting CD45 / CD34+ cells with purity and viability>90% is less than 1.077, especially 1.063-1.052. This Ficoll density provides a higher yield of CD45 / CD34+ cells with purity and cell viability similar to the traditional Ficoll density method.

[0079] In another aspect of the present disclosure, CD34+ cells can be initially obtained from freshly prepared deceased donor bone marrow using the reduced density Ficoll-Paque described above. The BM can be cryogenically frozen and then the CD34+ cells can be obtained using the traditional density Ficoll-Paque. This method essentially allows for the selective recovery of cells from deceased donor bone marrow - either using the modified Ficoll density before freezing, or using the traditional Ficoll density after freezing and thawing.

[0080] Recovery of MSCs from treated bone marrow

[0081] In another feature of the systems and methods disclosed herein, a method for recovering mesenchymal stem cells (MSCs) from enzymatically digested vertebral body (VB) bone fragments is provided, which are byproducts of the VB grinding and elution steps of the methods described herein. In this method, a mixture of collagenase and neutral protease is used to obtain the highest possible vertebral adherent MSC (vBA-MSC) yield. MSCs can be recovered from cryopreserved VB bone fragments that are later processed according to the present disclosure. In a specific aspect, recombinant Clostridium histolyticum collagenase, consisting of two active isoforms, is used in an effective amount during MSC extraction. In the presence of Mesencult medium, the cell mixture released by digesting the VB bone fragments is cultured on tissue-coated plastic to select proliferative vBA-MSCs. Freshly digested preparations and different passages of VB-MSCs can be characterized by flow cytometry, colony forming unit-fibroblast (CFU-F) potential, population doubling time (PDT), and in vitro tri-lineage (adipogenic, chondrogenic and osteogenic) differentiation.

[0082] Therefore, the present disclosure contemplates the use of a combination of purified collagenase and neutral protease to optimize the method of digesting and recovering MSCs from vertebral fragments. In a specific embodiment, the collagenase is DE collagenase (Vitacyte), which is composed of purified Clostridium histolyticum collagenase and Paneibacillus polymyxa neutral protease. According to one aspect of the present disclosure, the optimal neutral protease concentration and collagenase concentration (C1 and C2 collagenase) and the optimal ratio of solution volume (mL) to bone fragment weight (mg) are determined.

[0083] According to this process, VB bone fragments are placed in a TM, 2.5% human serum albumin and 10% dimethyl sulfoxide (DMSO) and incubated at 4°C for 1 hour. The solution was removed and the bone fragments were cooled to -86°C at a rate of approximately 1°C / min and then plunged into liquid nitrogen. After 24-48 hours in liquid nitrogen, the bone fragments were quickly thawed in a water bath set at 37°C, then washed in saline and digested using the above collagenase / protease solution.

[0084] The optimal volume-to-weight ratio was found to be 5:1 and the optimal incubation time was 2.5 hours. The optimal protease produced a neutral protease activity of 19.6U / mL. On the other hand, it was found that the total viable MSC cell count was generally insensitive to the collagenase concentration. It was also found that regardless of the C1 / C2 ratio, the yields produced by the recombinant collagenase subtypes C1 and C2 were similar to those of the purified collagenase. A more detailed description of the MSC recovery process of the present disclosure can be found in the technical article in Appendix A of the present application, the entire disclosure of which is incorporated herein by reference.

[0085] Predicting cell viability based on ischemia time

[0086] As described above, the ischemic time of the donor bone affects the viability of the cells extracted using the above-mentioned process. According to the present disclosure, total ischemia is defined as the time interval from the time of death (the point at which the donor's arterial system is cross-clamped and circulation stops) to the beginning of cell recovery from the bone. For the purpose of statistical modeling, the total interval can be divided into three consecutive and mutually exclusive time components: (a) warm ischemic time (WIT)-starting from the time of death, until the bone is recovered and packed on ice or the body is placed in a cooler; (b) body cooling time (BCT)-starting from the body being placed in a cooler, until the bone is packed on ice; and (c) cold ischemic time (CIT)-starting from when the bone is packed on ice, until the start of the process for extracting cells (such as HSPC). Therefore, total ischemic time = (WIT) + (BCT) + (CIT). For the case where whole body cooling is not used, BCT is zero and total ischemic time = (WIT) + (CIT).

[0087] In addition to total ischemic time, a variable corresponding to treatment experience can also be incorporated into the viability assay. It is well known that learning curves have a significant impact on the results, so to control for this fact, the variable EXP can be defined as the number of donors treated before the current donor - i.e., for the i-th donor, EXP = i-1. Other variables can include bone type (e.g., vertebral body and iliac bone), donor sex, and donor age.

[0088] In one aspect, the outcome variable is: the survival rate of a specific cell population (e.g., CD34+ cells), the percentage of cells detected after treatment, 5The total number of colony-forming units (CFU) of nucleated cells and the number of colony-forming units (CFU) per 10 5 The number of CFU-granulocyte macrophages (CFU-GM) detected per nucleated cell.

[0089] According to the present disclosure, an ordinary least squares (OLS) beta regression model can be used to predict the outcome variable, where a linear regression model is used for CFU and CFU-GM, and a beta regression model is used for the proportion of live CD34+ cells, or CD34+%, where 0 < (CD34+%) < 1. The beta regression equation for predicting CD34+% is:

[0090] (1) η = ln[pCD34* / (1-pCD34*)]

[0091] =β0+β1(WIT)+β2(BCT)+β3(BCT 2 )+β4(CIT)+β5(CIT 2 )

[0092] Where:

[0093] pCD34* =[1±100(%CD34±)] / 102 , which is the transformation of the target variable

[0094] β0 = constant (intercept)

[0095] β1 = warm ischemia time (WIT) correlation coefficient

[0096] β2 = Body Cooling Time (BCT) Correlation Coefficient

[0097] β3 = Body cooling time squared (BCT 2 ) Correlation coefficient

[0098] β4 = cold ischemia time (CIT) correlation coefficient

[0099] β5 = square of cold ischemia time (CIT 2 ) Correlation coefficient

[0100] The inverse link function is applied to the linear predictor variable η so that the result is the expected value of the outcome variable pCD34*, that is, the percentage of live CD34+ cells expected to be extracted from the donor bone. The inverse link function is:

[0101] (2)

[0102] Or replace η with the above equation (1):

[0103] (3)

[0104]

[0105] In this embodiment, the mathematical model predicts the proportion of viable CD34+ cells that can be extracted from donor bone subjected to specific ischemic conditions. The value of E[pCD34*] is between 0 and 1 because it is the ratio of the number of viable CD34+ cells to the total number of CD34+ cells in the bone sample.

[0106] In one embodiment, the coefficients of the beta regression calculation used to predict CD34+ % have the following values:

[0107] β0=3.5000

[0108] β1=-0.01996

[0109] β2=-0.181

[0110] β3=0.007

[0111] β4=-0.111

[0112] β5=0.002

[0113] Each β coefficient β0, β1, β2, β3, β4, β5 corresponds to the intercept, WIT, BCT, BCT 2 , CIT and CIT 2 , as described above.

[0114] Predictions for total colony forming units (CFU) and CFU-GM numbers can be obtained using the following linear regression model:

[0115] (4)η=β0+β1(WIT)+β2(BCT)+β3(BCT 2 )+β4(CIT)

[0116] A linear regression model for determining the CFU outcome variable may have the following coefficient values:

[0117] β0=756.5084

[0118] β1=-9.10826

[0119] β2=-95.03639

[0120] β3=3.45603

[0121] β4=-4.53349,

[0122] Each β coefficient β0, β1, β2, β3, β4 corresponds to the intercept, WIT, BCT, BCT 2 and CIT, as described above.

[0123] The linear regression model used to determine the CFU-GM outcome variable can have the following form:

[0124] (5)η=β0+β1(WIT)+β2(BCT)+β3(CIT)

[0125] With the following coefficient values:

[0126] β0=104.1805

[0127] β1=-8.11295

[0128] β2=-5.52927

[0129] β3=0.08872.

[0130] The above model is a base model or unadjusted model that only considers ischemia-based variables and does not consider experience, bone type, donor gender, and donor age variables. A fully adjusted model for CD34+% considering all variables may have the following form:

[0131] (6) η=β0+β1(experience)+β2(bone type)+β3(WIT)+β4(BCT)+β5(BCT 2 )+β6(CIT)+β7(CIT 2 )

[0132] With the following corresponding coefficient values:

[0133] %CD34+

[0134]

[0135] The fully adjusted CFU model is as follows:

[0136] (7)η=β0+β1(experience)+β2(equipment x experience)+β3(bone type)+β4(WIT)+β5(BCT)+β6(BCT 2 )+β7(CIT)+β8(CIT 2 )

[0137] CFU

[0138]

[0139] The coefficient β1 attempts to quantify the effect of the number of donors processed (i.e., experience) on cell mass and viability. In the fully tuned CFU model, the coefficient β2 corresponds to the experience of a specific device based on the assumption that devices can have different learning trajectories. One or both of these coefficients can be modified or even eliminated.

[0140] CFU-GM

[0141]

[0142]

[0143] Applying these models to the observed data can be used to determine the effect of ischemic time variables on CD34+%, such as Figure 12A-12C As shown in the table, the effect on total CFU, such as Figure 13A-13C As shown in the table, and the effect on the amount of CFU-GM, such as Figure 14A-Figure 14C The data in these tables can be used to determine whether a particular donor bone can produce enough cells to warrant further processing of the donor bone. In other words, the prediction model can be used to establish ischemic tolerance limits and HSPC quality acceptance criteria. For example, for the CD34+% outcome variable, a predicted value of more than 80% may be required to consider a particular donor bone.

[0144] The above model and Figure 12a- Fig. 14C The examples shown in the table show that acceptable HSPC quality levels can be achieved despite the inevitable extended ischemia times when bone must be procured by geographically dispersed OPOs and transported long distances to processing centers. Even under these conditions, a favorable combination of warm and cold ischemia times can be achieved, resulting in CD34+% viability in the range of 80-90%. The model also shows that freezing the body prior to bone recovery (a practice common in tissue recovery) is less beneficial in the context of bone marrow recovery. For example, when whole body cooling was used, the CD34+ viability averaged 72.75%, while when body cooling was not used, the average was slightly less than 90%. These models suggest that the best practice is to dispense with body cooling and move the recovered bone to a cold ischemia environment as quickly as possible. The model further suggests that limiting the WIT (warm ischemia time) to less than eight (8) hours and limiting the CIT (cold ischemia time) to less than 40 hours will optimize the chances of recovering a meaningful amount of viable cells from donor bone.

[0145] The model disclosed in this paper is based on Fig.15 The graph shown in predicts viability, where a threshold of 80% CD34+ cell viability was determined to be acceptable. As reflected in the graph, the relationship between warm ischemia time and cold ischemia time follows a curve from a point where WIT is 10 hours and CIT is 18 hours to a point where WIT is 1 hour and CIT is 27 hours.

[0146] A more detailed description of the disclosed method for predicting cell viability can be found in Appendix B of the present application, the entire disclosure of which is incorporated herein by reference.

[0147] The characterization description of the present disclosure should be considered as illustrative rather than restrictive. It should be understood that only certain embodiments have been presented, and all changes, modifications, and further applications that fall within the spirit of the present disclosure are desired to be protected.

[0148] Appendix A

[0149] Identification and characterization of a large source of primary human mesenchymal stem cells that are tightly attached to the bone surface of the vertebral body medullary cavity

[0150] Brian H. Johnstone, PhD 1,2 ,Hannah M. Miller,MS 1,2 ,Dongsheng Gu,MD,PhD 1 ,Sreedhar Thirumala,PhD 3 , Madelyn R. Beck, MS 3 ,Michael LaFontaine,PhD 2 Gerald Brandacher, MD 4 ,Erik J. Woods, PhD 1,2,3 *

[0151] 1 Ossium Health, Inc.5754W.74 th St., Indianapolis, IN 46278

[0152] 2 Department of Biomedical Sciences, College of Osteopathic Medicine, Marian University, Indianapolis, IN, USA

[0153] 3 Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN, USA

[0154] 4Department of Plastic and Reconstructive Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, USA

[0155] *Correspondence address: Erik J. Woods, PhD, HCLD (ABB), Ossium Health, Inc., 5754 W.74 th St.,Indianapolis,IN,46278,USA.E-mail: Erik@OssiumHealth.com

[0156] summary

[0157] Therapeutic allogeneic mesenchymal stem / stromal cells (MSC) are currently used in clinical trials to evaluate their effectiveness in treating many different disease indications. The final commercialization and widespread distribution will require further improvements in the production process to produce MSCs economically enough to meet the needs of the plan. The key factor causing the current high cost of goods (COG) of MSC production is the need to create a master cell bank (MCB) from multiple donors, resulting in variability in large-scale production operations. Therefore, the availability of a single donor large reservoir of primary MSCs will greatly benefit the cell therapy market by reducing the costs associated with production.

[0158] We have found an abundant population of cells that possess all the hallmarks of MSCs that are intimately associated with the vertebral body (VB) bone matrix and are only released by proteolytic digestion. Here we demonstrate that these vertebral bone adherent (vBA) cells possess all the MSC properties defined by the International Society of Cell and Gene Therapy (ISCT) (e.g., plastic adherence, surface marker expression, and trilineage differentiation), and have therefore been referred to as vBA-MSCs to distinguish this population from loosely associated MSCs recovered by aspiration or flushing of the bone marrow (BM) compartment.

[0159] Experimental banking and expansion were performed using vBA-MSCs obtained from three deceased donors and demonstrated that an average of 2.9 × 10 MSCs were obtained at one passage from 100 g of digested VB bone fragments. 8 ±1.35×10 8The pool size of vBA-MSCs was 100,000. The cells in each pool showed robust proliferation over a total of 9 passages without a significant decrease in the number of population doublings. The theoretical average total yield with limited expansion over 4 passages would produce 2 trillion (2×10 12 ) cells, equivalent to 10 6 cells / kg for 30,000 doses for an average 70 kg patient. Thus, we have established a sufficient new source of MSCs that will benefit the cell therapy market by overcoming production and regulatory inefficiencies due to variability between donors.

[0160] Keywords

[0161] Human mesenchymal stromal cells / stem cells

[0162] MSC

[0163] vertebral body

[0164] marrow

[0165] Cell therapy

[0166] regenerative medicine

[0167] Mass production

[0168] abbreviation

[0169] introduction

[0170] The strong activity and high expandability of mesenchymal stromal / stem cells (MSCs) have attracted great attention from commercial entities in developing "off-the-shelf" allogeneic MSC therapeutics derived from a limited number of donors. The development of allogeneic "universal donor"-based cell therapies allows controlled production with careful attention to comprehensive assessment of quality (e.g., consistency, potency, purity, and safety) for each production batch at significantly reduced costs compared to the production of individual batches of autologous cells for individual donors (e.g., currently occurring in chimeric antigen receptor (CAR) T cell therapies).

[0171] An inherent challenge in scaling up the production of cell-based therapeutics is the size of the production run. For some indications, effective doses of MSCs are as high as 1 × 10 9 cells / dose, which would require the production of 10 trillion (10×10 12) cells to meet potential demand at an affordable price [1-4]. Even at this production level, the cost of goods (COG) per dose of MSCs could still exceed $100,000, assuming economies of scale [3]. A significant driver of production costs, which scale in proportion to batch size, is the need to replenish the master cell bank (MCB) by isolating MSCs from new donors due to the limited volume of tissue and fluid that can be safely obtained from healthy volunteers and the limited expansion potential of MSCs isolated from each donor [5,6]. MSCs are scarce in all tissues, for example, BM aspirates contain approximately 0.001-0.01% total nucleated cells (TNCs) [7]. Given that BM aspirates from healthy volunteers are limited to 100 ml (50 ml from both iliac crests) for donor safety, the total yield of fresh, unpassaged MSCs is approximately 2×10 4 / donor. Expansion to trillions of cells would require seeding 1×10 7 In the past, 150 donors had a reserve of MSCs, thus limiting cell proliferation to 9 population doublings [8]. This number was in addition to the cells reserved for quality control measurements of the expanded MCB and working cell bank (WCB). Therefore, the number of MSCs available from each donor was more than 3 orders of magnitude lower than the optimal amount for the initial stage of expansion.

[0172] The need to continually replenish the cell bank with fresh cells from new donors introduces inconsistency into the manufacturing process due to the variability observed between MSCs derived from different donors that are otherwise matched in attributes (e.g., age and health status) [6,9,10]. The economic impact of donor-to-donor variability and the resulting impact on production costs is considerable. In a study examining large-scale production of multiple batches of MSCs derived from different donors, a 1.8-fold difference in cumulative population doublings was found between five different BM donors during a 30-day culture period [9]. This resulted in a >13-day difference in process time to produce a batch of 350 million MSCs. In addition to the logistical burden of coordinating batch runs, there is a corresponding increase in media costs, which is also a key cost driver for the production of cell-based therapeutics [1,3,8]. Furthermore, the authors found >18% variability in colony-forming potential and >50% variability in interleukin-6 expression, adding the additional complexity of quality control validation of potency for each batch derived from an individual donor. Similarly, single-center experience with clinical production of 68 batches of MSCs from BM recovered from 59 human volunteer donors observed population doubling times that varied more than 2-fold (46.8–141 h), with an average of 71.7 h, yielding a range of 1.9 × 10 7 to 5.43×10 9 (average 5.46×10 8 ) MSCs in a final batch size

[10] .

[0173] In addition to the direct economic burden imposed by increasing the COG / production run, there is also a regulatory burden due to the associated costs caused by the need to update the cell bank. MCB serves as a reserve of starting cultures for all production runs using cells from a specific donor. Regulatory requirements for quality and safety assessments of MCB are expensive and time-consuming

[11] . Among the three primary parameters (e.g., safety, efficacy, and consistency) required to assess the suitability of produced cell therapy products, efficacy is the most problematic because it involves the characteristics of individual donors and the spectrum of changes that occur with expansion as described above. This is especially true when the MSC group approaches the limit of expansion and enters aging that severely limits its efficacy

[12] . For these reasons, population doubling limits are important factors for regulatory agencies and are generally unresolved factors when submitting to the FDA

[13] .

[0174] Reducing the economic and regulatory burden of producing multiple batches of MCB per year to meet the needs of large-scale production requires the identification of large reservoirs of unmanipulated MSCs. A potential solution may come from the rich tissue containing MSCs that is typically discarded after routine medical procedures or that is available at autopsy. Adipose-derived stem / stromal cells (ASCs) are obtained from selective procedures that typically produce several liters of tissue and have recently been extensively studied; however, they are primarily used for autologous use [14, 15]. Direct isolation from bone containing the medullary cavity obtained by medical procedures or cadavers yields a higher percentage of MSCs than those present in aspirates (approximately 0.04%), most likely reflecting the absence of peripheral blood contamination

[16] . Approximately 5×10 9 Total nucleated cell counts have been obtained from BM of vertebral bodies (VBs) recovered from deceased organ donors, with each VB containing approximately 2 × 10 6 MSCs, or approximately 2 × 10 MSCs per recovered typical spinal 9VB segment. 7 Total MSCs

[17] . In addition, the ilium, sternum, ribs and bones of the long bones are sources of BM from which MSCs can be recovered[18-20]. Thus, the VB cavity of BM from a typical deceased donor alone yields >3 orders of magnitude more MSCs than can be obtained from a healthy human donor.

[0175] In addition to cells obtained by eluting or aspirating BM, another MSC population is tightly associated with the medullary cavity bone structure [21-23]. Bone-adherent MSCs (BA-MSCs), first identified in rodent long bones, have subsequently been isolated from human bone fragments obtained from the condyles and vertebrae of the long bones

[24] . We have discovered another source of MSCs, termed vertebral BA-MSCs (vBA-MSCs), which remain attached to VB bone fragments after extensive washing to remove BM cells and can only be released by digestion with proteases. The frequency and functionality of vBA-MSCs are equivalent to those of eluted VB BM-MSCs. Here, we present these data and establish a new source of MSCs that can be used in a large-scale production process to produce a total of more than 10 cells from individual donors. 15 thus, meeting the most optimistic demand levels for decades and overcoming current barriers to commercial-scale production [2, 8].

[0176] Materials and methods

[0177] Sources of tissues and cells

[0178] After obtaining informed consent for research purposes from surviving family members, after cardiac death of brain-dead organ donors, vertebrae are recovered. Each recovered vertebra is assigned a unique ID. The inclusion criteria for donor selection are brain death, 12 to 55 years of age, non-septicemia, and disease-free and pathogen-free. BM aspirated from living donors of three healthy volunteers is purchased from Lonza (Walkersville, MD). Expansion of living donor MSCs cryopreserved in 2 passages is purchased from Lonza. Related donor characteristics are shown in Table 1.

[0179] Deceased Donor Tissue Procurement and Shipping

[0180] Previously developed clinical retrieval methods [16, 25] were combined with subsequent experience from the ongoing VCA transplant immune tolerance clinical trial at Johns Hopkins University (ClinicalTrials.gov ID: NCT01459107) to form the basis of the procurement and transport protocol. Rationalized organ procurement organization (OPO) retrieval procedures were used in conjunction with dedicated kits and centralized training on retrieval and transport procedures. Retrieved bones were shipped to Ossium Health (Indianapolis, Indiana). Vertebral sections were purchased by six different OPO partners: Gift of Hope (Itasca, Illinois); Donor Alliance (Denver, Colorado); Iowa Donor Network (North Liberty, Iowa); MidAmerica Transplant (St. Louis, Missouri); and Nevada Donor Network (Las Vegas, Nevada). Bone was recovered from consenting organ and tissue donors by OPO personnel using osteotomes and mallets in accordance with IRB-approved protocols. The recovered bone was wrapped in lap sponges and towels and immersed in saline to ensure moisture retention during transport. The wrapped specimens were transported overnight on wet ice to one of two processing facilities.

[0181] Manual removal

[0182] Once received, the soft tissue was manually removed using a scalpel and chisel in a biosafety cabinet. Once visible pedicles were removed using a tissue processing band saw or a Stryker system 6 saw (Stryker, Kalamazoo, MI), leaving only the connected vertebral bodies. The vertebral bodies were separated and the intervertebral discs and soft tissue were removed with a scalpel. Care was taken to ensure that the cortical bone was not damaged to preserve and protect the hypoxic cancellous bone marrow during the entire removal process.

[0183] VB was cut into approximately 5 cm pieces using custom surgical grade stainless steel anvil scissors. 3The pieces were small enough to be crushed with a bone grinder. The pieces were immediately immersed in 500 mL of treatment medium (composed of Plasma-Lyte A pH 7.4 (Baxter Healthcare, Deerfield, IL) containing 2.5% human serum albumin (HAS; Octapharma USA Inc., Hoboken, NJ), 3 U / ml Benzonase (EMD Millipore, Burlington, MA) and 10 U / ml heparin (McKesson, Irving, TX)).

[0184] Grinding and elution

[0185] The bone grinder (Biorep Technologies. Inc, Miami, Florida) was assembled in a biosafety cabinet. A two-liter stainless steel beaker containing approximately 250 mL of fresh treatment medium was placed under the grinding head to catch the bone fragments and the medium flowing through. The stainless steel plunger was used to help push the fragments through the grinder. The grinder was rinsed with treatment medium to prevent the bone fragments from drying and adhering to the chamber. Once all bone fragments were ground, the chamber was thoroughly rinsed with fresh treatment medium. The final volume in the stainless steel beaker was one liter.

[0186] Filtration was performed using a bone marrow collection kit (Fresenius Kabi, Lake Zurich, IL) equipped with a flexible pre-filter and an inline filter. The whole bone grind and media were carefully transferred to the bone marrow collection kit. The grind was gently kneaded to allow optimal cell release from the grind. The media was then filtered using two 500 μm filters and two 200 μm filters. The bone grind was rinsed with flushing media using two 500 mL washes. The flushing media was Plasma-lyte containing 2.5% HSA. The whole bone marrow was then collected in a collection bag where samples were taken for experiments.

[0187] Digestion protocol for MSC isolation

[0188] Bone fragments (1 or 100 g) were transferred to 50 ml conical centrifuge tubes or 250 ml WhirlPak bags. DE10 collagenase solution (2 mg / ml; Vitacyte, Indianapolis, Indiana) was added to the bone fragments at a 5:1 (volume: weight) ratio. The tube and / or bag was transferred to a shaking incubator and incubated at 37°C for 2 h while shaking at 125 rpm. Protease activity was neutralized by adding 2% Stemulate (Cook Regentec, Indianapolis, Indiana), and the suspension was filtered through a 70 μm cap filter into a 50 ml conical screw cap tube. The bone fragments retained by the filter were washed with 25 ml Dulbecco's modified phosphate buffered saline (DPBS) solution (containing heparin (10 U / ml) and Benzonase (100 U / ml)), which was combined with the original filtrate. The tube was centrifuged at 350 x g for 5 minutes, the supernatant was aspirated, and the pellet was resuspended in 10 ml DPBS. The suspension was centrifuged again at 350 x g for 5 minutes, the supernatant was aspirated, and the pellet was resuspended in DPBS for analysis.

[0189] Isolation of MSCs from iliac bone and VB BM

[0190] A 1 ml sample of concentrated eluted BM was taken and pipetted into a 50 ml conical flask along with 49 ml DPBS. The flask was centrifuged at 300 x g for 10 minutes, the supernatant was aspirated, and the pellet was resuspended in 10 ml Rooster-Nourish medium (Rooster Bio, Frederick, Maryland). Cells were counted and cultured as described below.

[0191] Cell counting

[0192] The Cellometer Vision (Nexcellom, Lawrence MA) was used to determine the total viable cell count. 20 μl of ViaStain AOPI reagent (Nexcelom) was added to an Eppendorf tube containing 20 μl of cells. Once mixed, 20 μl of this solution was added to a Cellometer slide and total cells, live cells, and viability were counted.

[0193] Cell culture

[0194] Fresh cells were counted at 25,000 viable cells / cm 2 The cells were plated at a density of 1:1 in Rooster-Nourish medium (RoosterBio, Frederick, MD). The cells were cultured in T-225 flasks. After the first culture medium was changed on day 1, non-adherent cells were removed. The culture medium was then changed every 3-4 days until the colonies were approximately 80-90% confluent. The cells were released using TrypLE (ThermoFisher Scientific, Waltham, MA). The passaged cells were plated at 3,000 cells / cm 2 The cells were plated at a density of 1:1, but otherwise followed the same protocol as for freshly plated cells.

[0195] MCBs from three donors (DD5, DD6, and DD7) were generated in As above, fresh primary digests were initially plated at 25,000 viable cells / cm 2 The cells were plated. The cells were released with TrypLE and expanded once more to form MCBs. The entirety of passage 1 cells was resuspended in cryopreservation medium (CryoStor CS10; BioLife Solutions, Bothell, WA) and stored in the vapor phase of liquid nitrogen.

[0196] The cells were passaged up to ten times in a medium consisting of DMEM (Cat. No. 10567014, ThermoFisher, USA), ascorbic acid (248 μM; Cat. No. A2218, Sigma, USA), recombinant basic fibroblast growth factor (10 ng / ml; Cat. No. 233-GMP-025, R&D Systems, USA) and recombinant epidermal growth factor (10 ng / ml; Cat. No. 236-GMP-200, R&D Systems, USA). Cells were harvested at 70-80% confluence and total cell counts were obtained. A portion of the cells was centrifuged at 3000 cells / cm 2 The cells were replated in triplicate wells of six-well plates with the culture medium changed every 3-4 days.

[0197] Phenotypic analysis of MSCs by flow cytometry

[0198] At passage 2, 3 and 4, 1.8 μl of the following single fluorescent conjugated antibodies or dyes CD3, CD14, CD19, CD31, CD34, CD45, HLA-DR, CD73, CD90, CD105, Stro-1 and 7AAD (Supplementary Table S1) were added to different wells of a 96-well V-bottom plate. 100 μl of MACS (Miltenyi BioTec) buffer and 100 μl of cells (200,000 cells) were added to each well containing the antibody. The plate was incubated at 4°C in the dark for 30 minutes, after which the plate was centrifuged at 300 × g for 5 minutes. The cells were washed and resuspended in 200 μl of MACS buffer. An ACEA Biosciences NovoCyte 2060R flow cytometer was used for data collection and the data were analyzed using NovoExpress software (Acea Biosciences San Diego, CA).

[0199] Trilineage differentiation of MSCs

[0200] MSCs after passage 1 were seeded in wells of a 12-well plate containing 3 ml of Mesencult (Stem Cell Technologies, Vancouver, BC) at 8.0×10 4 4.0×10 4 and 2.0×10 4 for chondrogenic, adipogenic, and osteogenic differentiation. Contains 4.0×10 4The hole of each MSC is also plated as a control.After incubation for 2 hours, the Mesencult in the chondrogenesis hole is replaced with Stem Pro chondrogenesis medium (Thermo Fisher Scientific, Waltham, Massachusetts).After one day, the Mesencult in the fat-forming hole and the bone-forming hole is aspirated and replaced with Stem Pro fat-forming medium and Stem Pro bone-forming medium.Every 3 days, the Mesencult in the corresponding differentiation medium and the control well is supplemented.After 14,12 and 16 days, the hole containing chondrocyte, adipocyte and osteocyte is aspirated to remove the medium, washed twice with DPBS, fixed with 4% formalin for 30 minutes, washed once with DPBS, and stained.The Alcian blue (which dyes chondrocyte proteoglycan into blue) in 0.1NHCl is added to the chondrocyte hole for 30 minutes, the dye is aspirated, the hole is washed three times with 0.1N HCl and neutralized with distilled water, and the chondrocyte is visualized under an inverted optical microscope (Nikon). Oil red O (which stains adipocyte fat globules red) was added to the adipocyte wells for 15 minutes, the stain was aspirated, the wells were washed three times with distilled water, and the adipocytes were visualized under an inverted optical microscope. 2% alizarin red (which stains osteocyte calcium deposits red) was added to the osteocyte wells for 3 minutes, the stain was aspirated, the wells were washed three times with distilled water, and the adipocytes were visualized under an inverted optical microscope. All differentiated cells were qualitatively analyzed by visualization of color and phenotypic spectra.

[0201] Population doubling time

[0202] The population doubling time was determined at each passage using the following formula:

[0203] t*log(2) / log(T1 / T0), where t is the time (hours) between initial plating and cell harvest at 90% confluence, T1 is the cell count at harvest, and T o is the initial count at the time of inoculation.

[0204] CFU-F assay

[0205] For freshly digested cells, add 5 ml Mesencult, 20 μl Amphotericin B, and 100 μl Gentamycin to three wells of a 6-well plate. 5 , 5.0×10 5 and 7.5×10 5Cells were added to the first, second and third wells. The plates were placed in an incubator until the colonies were 90% confluent or up to 12 days. During 14 days, the culture medium was changed every 3-4 days. The plates were washed twice with DPBS, and 2 ml of methanol was added to each dish for 5 minutes to fix the cells. After 5 minutes, the methanol was decanted, the plates were air-dried, and the colonies were stained with 1% crystal violet solution. Colonies containing >50 cells were scored. In addition to the cells being plated at 32 cells / cm 2 , 65 cells / cm 2 and 125 cells / cm 2 The passaged cells were similarly assayed except that they were plated at a density of 100 μl.

[0206] T cell suppression assay

[0207] Inhibition of T cell activity was performed according to a previously published protocol with slight modifications

[26] . Briefly, peripheral blood mononuclear cells were isolated from whole blood (10 ml) by Ficoll (GE, Chicago, IL) separation and resuspended in DPBS. Most cells were labeled with carboxyfluorescein succinimidyl ester (CFSE; Sigma, St. Louis, MO) and frozen until use

[27] . In some cases, MSCs from passage 2 or 3 (pre-stimulated for 18–24 h with 100 ng / ml interferon-γ (IFNγ; RnD Systems, Minneapolis, MN)) were resuspended in RoosterNourish (RoosterBio, Frederick, MD) and incubated at 4 × 10 5 , 1×10 5 , 5×10 4 , 2.5×10 4 , 1.5×10 4 , 5×10 3 10 cells / well were added to a 96-well flat-bottom plate. RoosterNourish was added to each well until the volume was 200 μl / well. The plate was placed in a 37°C incubator at 10% CO2 and 5% humidity for at least two hours to allow MSC attachment. Cryopreserved PBMCs were rapidly thawed and plated at 4×10 6The cells were resuspended in Eagle's Minimum Essential Medium (EMEM; Stem Cell Technologies; supplemented with 10% FBS, 100 μg / ml PenStrep, 2 mM L-glutamine, and 100 μM β-mercaptoethanol) at a concentration of 10 cells / ml. The medium was aspirated from the plate containing MSCs, and 100 μl of PBMCs were added to all wells containing MSCs and wells without MSCs. T cells were stimulated by adding 100 μl of supplemented EMEM (supplemented with 40 μg / ml phytohemagglutinin (PHA; Sigma-Aldrich, St. Louis, MO)) to each well containing MSCs and PBMCs. Control wells containing only labeled and unlabeled PBMCs were also included, half of which were stimulated with PHA and the other half were not stimulated. The plate was returned to the incubator. After 4 days, PBMCs were removed from each well and labeled with 5 μl CD3-PE and 5 μl 7AAD before flow cytometry.

[0208] statistics

[0209] GraphPad Prism version 8 was used for statistical analysis (Student's t test). P values ​​< 0.05 were considered significant.

[0210] result

[0211] After removing the soft tissue, separating the VB and subsequently fragmenting it into approximately 1.5 cm 3 The dimensions of a typical vertebral column (typically T8-L5) before and after are shown in Figure 1. Plastic adherent vBA-MSCs possess a typical spindle-shaped morphology in culture ( Figure 1D ). Cells from the donor were expanded through 4 passages (initial plating was considered passage 0) and assayed by flow cytometry. At passages 1-4, vBA-MSCs had negligible levels of hematopoietic stem and progenitor cell surface markers CD14, CD19, CD34, and CD45 and expressed low to nonexistent amounts of human leukocyte antigen DR (HLA-DR) ( Figure 2 A). Levels of cells expressing PECAM1 (CD31) (typically endothelial cells and monocytes) were also low (<7%) at passage 2 (data not shown). In contrast, passaged vBA-MSCs were uniformly positive for CD73, CD90, and CD105. Thus, vBA-MSCs have a characteristic MSC surface marker profile

[28] . In addition, a variable portion of this population (approximately 20% or less, depending on passage number) also expressed the multipotent MSC surface marker Stro-1 [29-32].

[0212] The chondrogenic potential, adipogenic potential, and osteogenic potential of passage 3 vBA-MSCs were determined for each donor. Each vBA-MSC isolate showed the potential to differentiate into chondrocytes, adipocytes, and osteocytes ( Figure 2 B). A fraction of freshly isolated (i.e., never plated) and passaged vBA-MSCs exhibited a high degree of clonal proliferation as determined by fibroblast colony forming unit (CFU-F) potential. The mean CFU-F frequency in freshly digested VB bone fragments was 0.01 ± 0.004% (mean ± SD), which is similar to the frequency of proliferating MSCs in total BM ( Figure 3 )[7]. Proliferating cells were maintained in cell culture and formed colonies at a frequency of 37±3.4% and 27±1.2% after one and two passages, respectively.

[0213] Inhibition of T cell activity is one of the best-studied therapeutic properties of MSCs, providing a rationale for testing them in clinical trials for inflammatory conditions [33,34]. vBA-MSCs from three different donors dose-dependently inhibited the activation of T cells activated by PHA ( Figure 4 ). The maximum inhibition was 89±7% at a 1:1 ratio of vBA-MSC to peripheral blood mononuclear cells (PBMC). A slight but non-significant increase in inhibition was observed at all ratios by pre-treating vBA-MSC with IFN-γ for 18-24 hours prior to performing the inhibition studies. Treatment with IFN-γ has been shown to stimulate the suppressive function of MSCs, with an enhancing effect on senescent cells

[12] . The lack of an enhanced response to IFN-γ induction suggests that the cultured vBA-MSCs retain full immunomodulatory capacity.

[0214] The immunophenotypic profile, trilineage differentiation capacity, CFU-F potential, and immunomodulatory properties of plastic-adherent vBA-MSCs confirmed the classification of these cells as MSCs according to the guidelines published by the International Society for Cell and Gene Therapy (ISCT)

[28] . To further determine their equivalence to MSCs obtained from BM, a comparison was made between vBA-MSCs and MSCs isolated from central BM ( Figure 5 and Figure 6). Both commercially available previously expanded living donor BM-MSCs obtained by cryopreservation at passage 2 (Ex LD BM-MSCs) and freshly isolated MSCs from aspirated BM of living donors (LD BM-MSCs) were used. In addition, MSCs isolated from deceased donor VB BM (DD vBM-MSCs) were included in the comparison. For each source, MSCs from three donors were expanded to passage 2 and cryopreserved. After subsequent thawing, the cells were passaged once and then analyzed. MSCs from all four sources exhibited a substantially consistent immunophenotypic cell surface marker profile, with very low numbers of cells expressing CD14, CD19, CD34, CD45 and HLA-DR, whereas almost all cells expressed CD73, CD90 and CD105 ( Figure 5 A).

[0215] MSCs from each source grew rapidly in culture through 5 passages (the longest period examined), with no difference in population doubling time (PDT) between passages 4 and 5 ( Figure 5 C and Figure 5 D). Ex LD BM-MSCs obtained by pre-expansion to the second passage did show significantly higher PDT at the third passage compared with the other three MSC populations ( Figure 5 B). The CFU-F potential of Ex LD BM-MSCs at the second passage was also significantly lower than that of other MSC groups ( Figure 5 E). No comparison of CFU-F potential was performed for subsequent passages. Finally, the three-lineage differentiation potential was compared and it was found that each MSC population formed adipocytes, chondrocytes, and osteocytes at qualitatively the same frequency in vitro ( Figure 6 ).

[0216] The potential clinical translational utility of vBA-MSCs was assessed by conducting pilot-scale production runs to examine the feasibility of banking and expanding large numbers of cells from individual donors. VB fragments from three different donors were digested to isolate vBA-MSCs. The amount (100 g) was equivalent to approximately one-third of the total VB bone fragment weight that can be obtained from a typical donor. 8 ±1.35×10 8 vBA-MSCs were prepared from the first passage MCB of each donor and cryopreserved en masse, while the remainder was cultured over multiple passages, tracking the total cell yield at each passage ( Figure 7 A). For MCB, passage 1 was considered optimal, showing essentially the same surface marker profile and CFU-F potential as subsequent passages ( Figure 2 and Figure 3). A single further expansion to the second passage was sufficient to generate a population containing 5.17×10 9 ±4.3×10 9 Based on the observed population doublings, two expansions of the entire WCB would be sufficient to produce more than a trillion cells. PDT remained nearly constant between passages 2 and 9, with no indication of decreased growth rates at higher passage numbers. However, there were differences in PDT between donors ( Figure 7 B) Based on the PDT observed for each donor, starting with a seeding stock of 2 million vBA-MSCs, it would take 23, 36, and 29 days to produce one trillion cells from three different donors. These times were calculated using 2D tissue culture flasks and may be different in bioreactors.

[0217] discuss

[0218] The translational potential of MSCs to treat a variety of medical conditions has been idealized for more than a decade; however, despite many demonstrations of this potential in preclinical and early clinical trials, no MSC-based therapy has yet succeeded in late-stage registrational (generally Phase 3 in the United States) clinical trials, although a few have been approved for limited indications in relatively small jurisdictions. The reasons for the slow progress in the approval and subsequent commercialization of therapeutic MSCs, despite intensive research and development efforts by multiple entities, are certainly multifactorial. With hindsight, it appears that attempts to mass-produce MSCs by adopting methods and procedures from the highly successful biopharmaceutical industry may have been a contributing factor [35, 36]. There are many differences between the manufactured product derived from the cells and the cells themselves. Biopharmaceuticals are produced using immortal cell lines with nearly unlimited expansion capacity, allowing the generation of large MCBs from a single seeding stock. In contrast, the limited availability and expansion potential of MSCs requires the generation of multiple MCBs from different donors each year at disproportionately high production costs and regulatory burdens

[36] .

[0219] Here we provide a feasible solution to alleviate these burdens by identifying and characterizing a large reservoir of MSCs from deceased donor vertebrae. Based on the analysis presented here, vBA-MSCs are phenotypically and functionally equivalent to MSCs obtained from central BM. The cells express typical MSC markers (CD73, CD90, and CD105) and lack expression of hematopoietic stem and progenitor cell markers, and express very low levels of HLA class II proteins ( Figure 2 and Figure 5 A). Similar to BM-MSCs, vBA-MSCs have the potential for clonal expansion and can be induced to undergo trilineage differentiation ( Figure 2 , 3, 5 and 6). The passaged vBA-MSCs were fully adapted to suppress T cell activation, showing no difference in activity from previous stimulation by IFN-γ ( Figure 4 The differences in PDT and CFU-F between BM-MSCs expanded at passage 3 (but not subsequent passages) obtained from commercial sources likely reflect slower recovery from cryopreservation at passage 2 ( Figure 5 B). All MSCs were grown to passage 2 and cryopreserved to try to maintain comparability; however, commercially derived expanded BM-MSCs may be grown in different culture media and frozen in different cryopreservation media. Therefore, the cells experienced a lag in thawing and growth to passage 3, which was not apparent in subsequent passages.

[0220] Average 2.4×10 8 The cell bank size of MSCs can be obtained from 100 g of digested VB bone fragments from each of the three donors ( Figure 7 A). Each pool was expanded over a total of 9 passages without a significant decrease in population doubling time ( Figure 7 B) The theoretical yield of fully expanded cells per donor after 9 passages is 4×10 19 (40 million cubed) cells, for an average 70 kg patient, according to 10 6 / kg, equivalent to more than 500 billion doses. Of course, the actual total cell yield will be lower due to the inefficiencies and testing requirements inherent in large-scale production; however, COGs produced in large quantities from a single donor will likely be much lower than equivalent-scale production from multiple donors. Direct production cost savings can further reduce regulatory burdens by using a single donor source for all production activities. The next step in validating the potential cost savings with vBA-MSCs will be to conduct a scaled-up production run, which is currently underway.

[0221] The question we are currently exploring is why some MSC populations are easily removed or may wander freely in the BM, while other MSC populations remain tightly adherent to the bone / connective tissue matrix and may only be released by enzymatic digestion. Determining differences, if any, is complicated by the relatively low frequency of these cells, making them problematic to characterize using common analytical tools such as flow cytometry without first expansion in culture, which induces changes in phenotype and function [37-45]. A previous report found that freshly isolated enzymatic digests of trabecular bone in the pelvic region contained 15-fold higher CFU-F compared to aspirated BM

[24] ; however, we found no similar differences between freshly isolated vBA-MSCs and BM-MSCs. To better understand the dissimilarity between the populations, if any, we are performing single-cell RNA sequencing (scRNA-Seq) of the vBA-MSC transcriptome [46, 47]. We are also continuing to characterize the therapeutic potential of vBA-MSCs by studying the secretome and extracellular vesicles produced by these cells.

[0222] In summary, based on the data presented here, the fundamental properties of vBA-MSCs do not appear different from aspirated BM-MSCs; therefore, these cells may have the potential to seamlessly replace therapeutic applications at significant savings in production and regulatory costs. In addition, other markets that require large quantities of MSCs may also benefit from an abundant source of primary cells. These include tissue engineering and the production of MSC-derived products such as exosomes, as well as biomedical research applications and emerging applications in cosmeceuticals and bioengineered materials. Each of these markets is expected to grow substantially during the ensuing decades, driving annual demand for MSCs in excess of 10 sextillion (1×10 21 ) combined demand for MSCs [2]. The high demand for MSCs in all of these markets in the future could potentially be fully met by vBA-MSCs obtained from the abundant and stable supply of bone containing the medullary cavity of deceased donors from 10,000 organ donors and another 40,000 tissue donors per year in the United States alone.

[0223] surface

[0224] Table 1. Description of donors used in this study

[0225]

[0226] 1 Abbreviations: DD, deceased donor; LD, living donor; BM, bone marrow

[0227] Supplementary Table S1. Description of antibodies and dyes used

[0228]

[0229] 1 Abbreviations: 7-AAD, 7-aminoactinomycin; PE, phycoerythrin; APC, allophycocyanin; PeCy7, phycoerythrin-cyanin 7; FITC, fluorescein isothiocyanate.

[0230] legend

[0231] Figure 1. Processing of a typical vertebral column for the isolation of vBA-MSCs. Soft tissue was removed from the vertebra (usually T8-L5) (A), followed by isolation of the vertebral body (VB) and removal of the disc and remaining soft tissue (B). The VB was ground to approximately 1.5 cm 3 Debris (C) followed by enzymatic digestion to release adherent cells. Plastic-adherent vBA-MSCs develop a typical spindle shape when cultured (D; passage 2 cells).

[0232] Figure 2 .Surface antigen phenotype and tri-lineage differentiation of vBA-MSCs. (A) Surface antigen expression analysis of 1, 2, 3, and 4 passage vBA-MSCs from 3 different donors (DD1, DD2, and DD23; Table 1) was performed using fluorescence-conjugated antibodies and flow cytometry. The percentage of cells after each passage culture is shown (all cells are gated using side and forward scatter). (B) After staining for chondrocytes (Alcian Blue), adipocytes (Oil Red O), or osteocytes (Alizarin Red), 3rd passage vBA-MSCs grown in expansion medium (BA) or induced to undergo (BB) chondrogenesis, (BC) adipogenesis, or (BD) osteogenesis were imaged as described in Materials and Methods. Images represent the results of vBA-MSCs derived from 3 different donors. All magnifications are 20X.

[0233] Figure 3 Fibroblast colony forming unit (CFU-F) potential of vBA-MSCs isolated from 3 different donors (DD1, DD2 and DD3; Table 1) and plated immediately after isolation by digestion (fresh) or after 1 or 2 passages (P1 and P2). 5×10 5 624 (fresh) or 624 (passaged) total cells were plated in triplicate in wells of 6-well plates and incubated for 14 days with medium changes every 3-4 days.

[0234] Figure 4 . Inhibition of T cell activation by vBA-MSCs. (A) Inhibition at decreasing ratios of PBMC to vBA-MSCs. PBMCs isolated from blood of a single donor were labeled with carboxyfluorescein diacetate succinimidyl ester (CSFE). After washing and addition of 4×10 5vBA-MSCs were adhered in 96-well plates for 2 hours before adding PBMCs. In some experiments, IFN-γ (100 ng / ml) was added 18-24 hours before adding PBMCs. T cells were stimulated with PHA for 4 days. After being labeled with anti-CD3-PE antibodies, cells were recovered from the plate and analyzed by flow cytometry. The percentage of activated T cells was plotted. (B) Representative flow cytometry of PBMCs alone and PBMCs and MSCs after PHA activation without and with PHA activation is shown. Each data point represents the average of 3 different experiments of 3 different donors (DD1, DD2 and DD3). Error bars represent standard deviations. P>0.05 is used for comparison at all PBMC:vBA-MSC ratios + / -IFN-γ.

[0235] Figure 5 . Comparison of vBA-MSCs isolated from deceased donor vertebral body BM and BM aspirated from the iliac crest of living donors with MSCs. (A) Surface marker expression of 3 passage cells was characterized by flow cytometry. The different sources of MSCs were: deceased donor vBA-MSCs (DD vBA-MSCs); MSCs derived from deceased donor vertebral body bone marrow (DD BM-MSCs); BM MSCs aspirated from living donors (LD BM-MSCs); and 2 passages of living donor aspirated BM MSCs obtained from commercial sources (LD ExBM-MSCs). The percentage of cells in the total population after gating to remove debris is shown. Surface marker expression did not differ between cell types. (BD) The population doubling time (PDT) from 2 to 3 passages (B), 3 to 4 passages (C), and 4 to 5 passages (D) were compared. Between passages 2 and 3, LD Ex BM-MSCs grew significantly (*, P < 0.05) slower than vBA-MSCs and LD BM-MSCs. No difference in PDT was observed in the next two passages. (E) Figure 3 The CFU-F assay was performed on the passaged cells as described in . Compared with the MSCs from the other three sources that were also in the second passage, the formation of CFU-F was significantly slower for the LD ExBM-MSCs in the second passage (*, P<0.05). Each column represents the mean ± sd from 3 donors for each MSC source. The specific donors are: LD BM-MSC (donors LD1, LD2 and LD3); LD Ex BM-MSC (donors LD4, LD5 and LD6); vBM-MSC and vBA-MSC (donors DD1, DD2 and DD3). The donor characteristics are listed in Table 1.

[0236] Figure 6Trilineage differentiation of vBA-MSCs with MSCs isolated from vertebral body BM of deceased donors and BM aspirated from the iliac crest of living donors. Figure 2 As described in, for each cell type, cells are cultured and induced to differentiate. There is no qualitative difference in the adipogenic potential, chondrogenic potential and osteogenic potential of the 3rd passage cell from any source in the four sources. The image represents the experiment of 3 different donors of MSC for each source. Magnification is shown for each image.

[0237] Figure 7 .Cumulative population growth of vBA-MSCs from 3 different donors. vBA-MSCs obtained from digestion of VB fragments from 3 donors (DD5, DD6 and DD7) were isolated and expanded to 1 passage to form a master cell bank. A portion of the 1st passage vBA-MSCs from each donor was expanded to 9 passages. (A) Observed and potential cumulative growth yields for each passage of vBA-MSCs from 3 donors. (B) Cumulative vBA-MSC population doublings for 0-9 passages. Population doublings (PD) are calculated based on the initial cell number of the plate and the number of cells recovered before re-plating the cells after each plate reached 80% confluence, and are used to determine the theoretical total cell yield after each passage. The theoretical total yield at 2-9 passages is obtained by exponentiating the PD calculated for each passage (base 2) and multiplying by the cumulative cell number of each previous passage. For each passage, each donor vBA-MSC was plated in triplicate. The coefficient of variance (CV) between the cell numbers obtained from each well was <15%.

[0238]

[0239]

[0240]

[0241]

[0242]

[0243]

[0244]

[0245] References

[0246] 1.Lipsitz,Y.Y.,et al.,A roadmap for cost-of-goods planning to guideeconomic production of cell therapy products.Cytotherapy,2017.19(12):p.1383-1391.

[0247] 2.Olsen,T.R.,et al.,Peak MSC-Are We There Yet?Front Med(Lausanne),2018.5:p.178.

[0248] 3.Pereira Chilimia,T.D.,F.Moncaugeig,and S.S.Farid,Impact ofallogeneic stem cell manufacturing decisions on cost of godds,processrobustness and reimbursement.Biochemical Engineering Journal,2018.137:p.132-151.

[0249] 4.Simaria,A.S.,et al.,Allogeneic cell therapy bioprocess economicsand optimization:single-use cell expansion technologies.Biotechnol Bioeng,2014.111(1):p.69-83.

[0250] 5.Harrison,R.P.,N.Medcalf,and Q.A.Rafiq,Cell therapy-processingeconomics:small-scale microfactories as a stepping stone toward large-scalemacrofactories.Regen Med,2018.13(2):p.159-173.

[0251] 6.Mizukami,A.,et al.,Technologies for large-scale umbilical cord-derived MSC expansion:Experimental performance and cost of gooodsanalysis.Biochemical Engineering Journal,2018.135:p.36-48.

[0252] 7.Pittenger,M.F.,et al.,Multilineage potential of adult humanmesenchymal stem cells.Science,1999.284(5411):p.143-7.

[0253] 8.Chilima,T.D.P.,T.Bovy,and S.S.Farid,Designing the optimalmanufacturing strategy for an adherent allogeneic cell therapy.BioProcessInternational,2016.14(9):p.24-32.

[0254] 9.Heathman,T.R.,et al.,Characterization of human mesenchymalstemcells from multiple donors and the implications for large scale bioprocessdevelopment.Biochemical Engineering Journal,2016.108:p.14-23.

[0255] 10.Lechanteur,C.,et al.,Large-scale clinical expansion of mesenchymalstem cells in the GMP-compliant,closed automated Quantum(R)cell expansionsystem:Comparison with expansion in traditional T-flasks.Stem Cell Research&Therapy,2014.4(8):p.1-11.

[0256] 11.Wuchter,P.,et al.,Standardization of Good Manufacturing Practice-compliant production of bone marrow-derived human mesenchymal stromal cellsfor immunotherapeutic applications.Cytotherapy,2015.17(2):p.128-39.

[0257] 12.chinnadurai,R.,et al.,Immune dysfunctionality of replicativesenescent mesenchymal stromal cells is corrected by IFNgamma priming.BloodAdv,2017.1(11):p.628-643.

[0258] 13.Mendicino,M.,et al.,MSC-based product characterization forclinical trials:an FDA perspective.Cell Stem Cell,2014.14(2):p.141-5.

[0259] 14.Lockhart,R.A.,J.A.Aronowitz,and S.Dos-Anjos Vilaboa,Use of FreshlyIsolated Human Adipose Stromal Cells for Clinical Applications.Aesthet surgJ,2017.37(suppl_3):p.S4-S8.

[0260] 15.Dykstra,J.A.,et al.,Concise Review:Fat and Furious:Harnessing theFull Potential of Adipose-Derived stromal Vascular Fraction.Stem Cells TranslMed,2017.6(4):p.1096-1108.

[0261] 16.Donnenberg,A.D.,et al.,Clinical implementation of a procedure toprepare bone marrow cells from cadaveric vertebral bodies.Regen Med,2011.6(6):p.701-6.

[0262] 17.Ahrens,N.,et al.,Mesenchymal stem cell content of human vertebralbone marrow.Transplantation,2004.78(6):p.925-9.

[0263] 18.Cox,G.,et al.,High abundanCe of CD271(+)multipotential stromalcells(MSCs)in intramedullary cavities of long bones.Bone,2012.50(2):p.510-7.

[0264] 19.Rybka,W.B.,et al.,Hematopoietic progenitor cell content ofvertebral body marrow used for combined solid organ and bone marrowtransplantation.Transplantation,1995.59(6):p.871-4.

[0265] 20.Soderdahl,G.,et al.,Cadaveric bone marrow and spleen cells fortransplantation.Bone Marrow Transplant,1998.21(1):p.79-84.

[0266] 21.Blashki,D.,et al.,Mesenchymal stem cells from cortica lbonedemonstrate increased clonal incidence,potency,and developmental capacitycompared to their bone marrow-derived counterparts.J Tissue Eng,2016.7:p.2041731416661196.

[0267] 22.Siclari,V.A.,et al.,Mesenchymal progenitors residing dose to thebone surface are functionally distinctfrom those in the central bonemarrow.Bone,2013.53(2):p.575-86.

[0268] 23.Yusop,N.,et al.,Isolation and Characterisation of Mesenchymol StemCells from Rat Bone Marrow and the Endos teal Niche:A Comparative Study.StemCells Int,2018.2018:p.6869128.

[0269] 24.Jones,E.,et al.,Larqe-scale extraction and characterization ofCD271+multipotential stromal cells from trabecular bone in health andosteoarthritis:implications for bone regeneration strategies based onuncultured or minimally cultured multipotential stromal cells.ArthritisRheum,2010.62(7):p.1944-54.

[0270] 25.Gorantla,V.S.,et al.,Development and validation of a procedure toisolate viable bone marrow cells from the vertebrae of cadaveric organ donorsfor composite organ grafting.Cytotherapy,2012.14(1):p.104-13.

[0271] 26.Li,M.,et al.,Therapeutic Delivery Specifications IdentifiedThrough Compartmental Analysis of a Mesenchymal Stromal Cell-ImmuneReaction.Sci Rep,2018.8(1):p.6816.

[0272] 27.Quah,B.J.,H.S.Warren,and C.R.Parish,Monitoring lymphocyteproliferation in vitro and in vivo with the intracellular fluorescent dyecarboxyfluorescein diacetate succinimidyl ester.Nat Protoc,2007.2(9):p.2049-56.

[0273] 28.Dominici,M.,et al.,Minimal criteria for defining multipotentmesenchymal stromal cells.The International Society for Cellular Therapyposition statement.Cytotherapy,2006.8(4):p.315-7.

[0274] 29.Gronthos,S.,et al.,Molecular and cellular characterisation ofhighly purified stromal stem cells derived from human bone marrow.J Cell Sci,2003.116(Pt 9):p.1827-35.

[0275] 30.Simmons,P.J.and B.Torok-Storb,Identification of stromal cellprecursors in human bone marrow by a novel monoclonal antibody,STRO-1.Blood,1991.78(1):p.55-62.

[0276] 31.Dennis,J.E.,et al.,The STRO-1+marrow cell population ismultipotential.Cells Tissues Organs,2002.170(2-3):p.73-82.

[0277] 32.Bensidhoum,M.,et al.,Homing of in vitro expanded Stro-1-or Stro-1+human mesenchymal stem cells into the NOD / SCID mouse and their role insupporting human CD34cell engraftment.Blood,2004.103(9):p.3313-9.

[0278] 33.Galipeau,J.,et al.,International Society for Cellular Therapyperspective on immune functional assays for mesenchymal stromal cells aspotency release criterion for advanced phase clinical trials.Cytotherapy,2016.18(2):p.151-9.

[0279] 34.Squillaro,T.,G.Peluso,and U.Galderisi,clinical Trials WithMesenchymal Stem Cells:An Update.Cell Transplant,2016.25(5):p.829-48.

[0280] 35.Galipeau,J.and L.Sensebe,Mesenchymal Stromal Cells:ClinicalChallenges and Therapeutic Opportunities.Cell Stem Cell,2018.22(6):p.824-833.

[0281] 36.Jossen,V.,et al.,Manufacturing human mesenchymal stem cells atclinical scale:process and regulatory challenges.Appl Microbiol Biotechnol,2018.102(9):p.3981-3994.

[0282] 37.Banfi,A.,et al.,Replicative aging ond gene expression in long-termcultures of human bone marrow stromal cells.Tissue Eng,2002.8(6):p.901-10.

[0283] 38.Baxter,M.A.,et al.,study of telomere length reveals rapid aging ofhuman marrow stromal cells following in vitro expansion.Stem cells,2004.22(5):p.675-82.

[0284] 39.Bork,S.,et al.,DNA methylation pattern changes upon long-termculture and aging of human mesenchymal stromal cells.Aging Cell,2010.9(1):p.54-63.

[0285] 40.Bruder,S.P.,N.Jaiswal,and S.E.Haynesworth,Growth kinetics,self-renewal,and the osteogenic potential o fpurified human mesenchymal stem cellsduring extensive subcultivation and following cryopreservation.J CellBiochem,1997.64(2):p.278-94.

[0286] 41.Digirolamo,C.M.,et al.,Propagation and senescence of human marrowstromal cells in culture:a simple colony-forming assay identifies sampleswith the greatest potential to propagate and differentiate.Br J Haematol,1999.107(2):p.275-81.

[0287] 42.Muraglia,A.,R.Cancedda,and R.Quarto,Clonal mesenchymal progenitorsfrom human bone marrow differentiate in vitro according to a hierarchicalmodel.J Cell sci,2000.113(Pt 7):p.1161-6.

[0288] 43.Redaelli,S.,et al.,From cytogenomic to epigenomic profiles:monitoring the biologic behavior of in vitro cultured human bone marrowmesenchymal stem cells.stem Cell Res Ther,2012.3(6):p.47.

[0289] 44.Moravcikova,E.,et al.,Proteomic Profiling of Native Unpassaged andCulture-Expanded Mesenchymal Stromal Cells(MSC).Cytometry A,2018.93(9):p.894-904.

[0290] 45.Bara,J.J.,et al.,Concise review:Bone marrow-derived mesenchymalstem cells change phenotype following in vitro culture:implications for basicresearch and the clinic.Stem Cells,2014.32(7):p.1713-23.

[0291] 46.Choi,Y.H.and J.K.Kim,Dissecting Cellular Heterogeneity UsingSingle-Cell RNA Sequencing.Mol Cells,2019.42(3):p.189-199.

[0292] 47.Hwang, B., JHLee, and D.Bang, Single-cell RNA sequencing technologies and bioinformatics pipelines. Exp Mol Med, 2018.50(8): p.96.

[0293] Appendix B

[0294] Relationship of ischemic time and whole-body cooling to the quality of hematopoietic stem and progenitor cells retrieved from the bone marrow of deceased organ donors

[0295] Erik J. Woods, PhD 1,2,3 *,Aubrey M.Sherry,MS 1,2 ,John R. Woods, PhD 4 ,James Hardin,PhD 5 ,Michael LaFontaine,PhD 2 Gerald Brandacher, MD 6 ,Brian H. Johnstone, PhD 1,2

[0296] 1 Ossium Health, Inc. Indianapolis, IN, USA

[0297] 2 Department of Biomedical Sciences, College of Osteopathic Medicine, Marian University, Indianapolis, IN, USA

[0298] 3 Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN, USA

[0299] 4Richard M. Fairbanks School of Public Health, Indiana University, Indianapolis, IN, USA

[0300] 5 Arnold School of Public Health, University of South Carolina, Columbia, SC, USA

[0301] 6 Department of Plastic and Reconstructive Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, USA

[0302] *Correspondence address: Erik J. Woods, PhD, HCLD (ABB), Ossium Health, Inc., 5754 W.74 th St., Indianapolis, IN, 46278, USA. Email: Erik@OssiumHealth.com

[0303] Relationship of ischemic time and whole-body cooling to the quality of hematopoietic stem and progenitor cells retrieved from the bone marrow of deceased organ donors

[0304] summary

[0305] Deceased organ donors represent an unutilized source of therapeutic bone marrow (BM) that can be recovered in quantities 3-5 times that obtained from living donors, tested for quality, cryopreserved, and stored indefinitely for future use on demand. However, a challenge to building future BM banking systems to a genetically diverse scale will be managing the prolonged ischemia times that inevitably occur when bone procured from geographically dispersed sites is transported to distant facilities for processing. The objectives of this study were to: (a) quantify the relationship between ischemia time and the quality of hematopoietic stem and progenitor cells (HSPCs) derived from deceased donor BM under realistic and scaled procurement and transport conditions; (b) identify ischemia time cutoffs beyond which HSPC quality is adversely affected, and (c) investigate whole body cooling as a strategy for preserving cell viability and function. Bone from 62 deceased donors was analyzed after exposure to varying periods of warm ischemia time (WIT), cold ischemia time (CIT), and body cooling time (BCT). Regression models were developed to quantify the independent associations of WIT, CIT, and BCT with viability and function of retrieved HSPCs. Results showed that in a "real-world" scenario: (a) combinations of warm and cold ischemia times that favor high-quality HSPC retrieval are readily achievable (e.g., CD34+ viability in the 80%-90% range is commonly observed); (b) cooling the body prior to bone retrieval is detrimental to cell viability (e.g., CD34+ viability <73% compared to >89% without body cooling); and (c) the vertebral body (VB) is a superior source of HSPCs compared to the ilium (IL) (e.g., CD34+ viability >80% when VB is the source compared to <74% when IL is the source). Our quantitative model can be used to develop ischemia time tolerances and HSPC quality acceptance criteria and inform emerging BM storage systems seeking to develop data-driven industry standards.

[0306] Keywords: Deceased donor bone marrow, bone marrow storage, bone marrow ischemic time, hematopoietic stem cell transplantation

[0307] introduction

[0308] Deceased donor bone marrow (BM) represents a large, unutilized source of hematopoietic stem and progenitor cells (HSPCs) that can be cryopreserved and stored for future, on-demand use in bone marrow transplantation (BMT) procedures. The appeal of BM banking is based in part on the knowledge that HSPCs can be readily obtained during periods of surge in demand, such as after a mass casualty event (e.g., a nuclear disaster resulting in widespread bone marrow failure) [1,2]. This benefit has been further reinforced by recent successes in infusing donor BM cells to establish transient mixed chimerism and / or peripheral immune modulation in recipients of solid organ and vascular composite allograft (VCA) grafts [3-5]. BM banks from deceased organ donors have established a repository for future tolerance induction procedures using a delayed protocol that has been successful in nonhuman primates [6,7].

[0309] Cryopreservation and storage of BM from deceased organ donors will require the establishment of BM banks that are conceptually similar to cord blood banks. As with cord blood, it is well established that BM remains biologically functional after cryopreservation and can serve as a genetically diverse source of stem cell transplants on demand [8–11]. Importantly, the national network of organ procurement organizations (OPOs) that has been active in the United States (US) for more than 50 years provides an existing, well-functioning infrastructure for the procurement and transportation of bone tissue recovered from deceased donors. However, organizing an organ donor BM procurement and storage system that leverages the existing OPO infrastructure will require a coordinated effort that includes the retrieval and safe transportation of biological material to dedicated BM cell processing centers at a scale suitable for clinical production.

[0310] In the case of living BM donors, a key issue that is not usually considered important is the ischemic time that is inevitably introduced during the recovery and transportation of bones recovered from deceased donors. Before scaling up clinical production systems, it is necessary to determine how changes in warm ischemic time and cold ischemic time affect the quality of HSPCs derived from bones recovered in geographically dispersed locations and transported over long distances to centralized processing facilities. And it will be necessary to determine the upper tolerance limit for both warm ischemia and cold ischemia, which, if exceeded, may render the quality and function of HSPCs unusable for therapeutic use.

[0311] Additionally, there is a need to better understand the impact of whole-body cooling in the context of deceased donor bone retrieval and transport. Current tissue banking guidelines in the United States allow for tissue retrieval from deceased donors up to 24 hours after cardiac arrest, provided the body is refrigerated within 12 hours of cardiac arrest.

[12] However, body cooling is a variable that has not been systematically studied in relation to BM retrieval and is a variable that requires criteria that differ from those established for tissue retrieval.

[0312] Here, we present our results for the first time, which quantify the association between ischemia time and systemic cooling and the quality of HSPCs recovered from cadaveric vertebrae. Our analysis shows that high-quality functional HSPCs can be obtained from deceased donors even after the recovered bone has been subjected to more than 40 hours of cumulative warm and cold ischemia time, provided that systemic cooling, which has been shown to be detrimental to viability, is avoided. These findings will be used to establish warm and cold ischemia time tolerance limits and HSPC quality acceptance criteria for BM from deceased organ donors.

[0313] method

[0314] Study Design

[0315] This was a pragmatic field observational study designed to model the effects of ischemia and body cooling time on the viability and function of HSPCs retrieved from the BM of deceased organ donors

[13] . The study was designed to generate observational findings that could be generalized and applied in a routine practice setting. The external validity (generalizability) of the study was enhanced by ensuring the participation of multiple OPOs operating under normal field conditions. With the exception of special training related to the details of bone retrieval and transport (see below), usual procurement conditions were in effect. As the OPOs were geographically dispersed, the data collected covered the full spectrum of ischemia times that might be seen in a ‘real world’ procurement and transport scenario.

[0316] Donor tissue procurement and transportation

[0317] Previously developed clinical retrieval methods, combined with subsequent experience from the ongoing VCA transplant immune tolerance clinical trial at Johns Hopkins University (Clinical Trial Government Identifier: NCT01459107), formed the basis for the procurement and transport protocols [4,14-16]. However, these protocols required optimization and validation to ensure that multiple OPOs could reliably operate them in a manner that allowed for consistent yields of functionally viable HSPCs after bone retrieval and transport across a wide geographic region. To this end, a simplified OPO retrieval procedure was employed in combination with a dedicated kit and centralized training on retrieval and transport procedures.

[0318] Recovered bone was transported to one of two processing facilities located in Centennial, CO (Facility A) or Indianapolis, IN (Facility B). Vertebral segments (Facilities A and B) and / or iliac crests (Facility A only) were procured by six OPOs: Gift of Hope (Itasca, IL); Donor Alliance (Denver, CO); Iowa Donor Network (North Liberty, IA); Mid-America Transplantation (St. Louis, MO); and Nevada Donor Network (Las Vegas, NV). OPO personnel recovered bone from research-consented organ and tissue donors using osteotomies and mallets in accordance with IRB-approved protocols. Unprocessed bone was wrapped in surgical drapes and towels soaked in saline and placed in triple-seal bags to ensure moisture retention during transport. The wrapped specimens were transported overnight on wet ice to one of the two processing facilities.

[0319] Manual Clearance

[0320] Upon receipt, soft tissue was manually removed using a scalpel and chisel in an ISO 5 cleanroom (Facility A) or biosafety cabinet (Facility B). Once visible, the pedicles were removed using a tissue processing band saw or a Stryker System 6 saw (Stryker, Kalamazoo, MI), leaving only the connected vertebral body. The vertebral body was separated at the intervertebral disc using a boning knife (Facility B) or a tissue processing band saw (Facility A). The remaining intervertebral disc and soft tissue were removed using a scalpel, leaving a clean, isolated VB. The iliac soft tissue was removed using a chisel and scalpel. Care was taken to ensure that the cortical bone was not disrupted throughout the removal process to preserve and protect the hypoxic cancellous BM.

[0321] Using a saw and / or incus shears, cut the VB and ilium into 5 cm 3 Fragments, small enough to be broken with a bone grinder. The fragments are immediately immersed in 500mL of processing medium (Iscove's modified Dulbecco's medium, containing 100U / mL DNase, 10U / mL heparin and 2.5% human serum albumin). IMDM is suitable for rapid proliferation of high-density cell cultures and is ideal for supporting T-lymphocytes and B-lymphocytes. DNA enzymes are essential for relieving cell clumping due to the stress associated with autopsy of dead cells releasing DNA and deceased donor-derived BM. Heparin is used as an anticoagulant. HSA provides a source of protein to prevent cell adhesion and adsorption to the surface.

[0322] Grinding and elution

[0323] Electric bone grinders are assembled in ISO-5 clean rooms (facility A), and special bone grinders (Biorep Technologies Inc., Miami, Florida) are assembled in biosafety cabinets (facility B). In either facility, a 2L stainless steel beaker containing 100mL of fresh treatment medium is placed under the grinding head to catch bone fragments and the medium flowing through. If VB and IL from the same donor are processed, the bone types remain separate. The treatment medium is used to rinse the grinder throughout the process to prevent the bone from drying and sticking to the chamber. Once all bones are ground, the chamber is thoroughly rinsed with fresh treatment medium. The final volume in the stainless steel beaker is usually about 750mL.

[0324] Stainless steel sieves were stacked with No. 40 (425 μm) on top of No. 80 (177 μm) and placed on a round receiving tray (WS Tyler, St Catherines, ON). The stainless steel beaker was rotated and poured onto the sieves. The bone fragments were evenly distributed on the top of the sieves and rinsed with 250 mL of fresh treatment medium. Approximately 1000 mL of the sieved BM product was transferred to sterile packaging for final analysis.

[0325] Nucleated cell count

[0326] Aliquots of BM extracts were subjected to red blood cell lysis with ammonium chloride RBC lysis buffer. In a 15 mL conical tube, 4 mL of 9% ammonium chloride was added to 1 mL of BM cell suspension and incubated at room temperature for 5 minutes. After incubation, the lysed samples were filled to the top of the tube with IMDM containing 100 U / mL DNase, 10 U / mL heparin, and 2.5% HSA treatment medium. The lysed samples were centrifuged at 300 × g for 5 minutes and decanted. The samples were then washed with 15 mL of treatment medium, centrifuged at 300 × g for 5 minutes, and decanted. Finally, the lysed cells were resuspended with 1 mL of the same treatment medium. Live nucleated cell counts were obtained using trypan blue and a hemocytometer.

[0327] Flow cytometry

[0328] Flow cytometry was performed using an ACEA Biosciences NovoCyte 2060R equipped with 488 nm and 640 nm lasers. The ISHAGE method was used to count CD45+ and CD34+ cells

[16] . 500 μm of lysed bone marrow extracts were stained with 2 μm each of CD45-FITC, CD34-APC, 7-AAD, and Annexin-PE for 15 minutes. All conjugated antibodies were purchased from BD Biosciences and 7-AAD was purchased from Tonbo Biosciences. Cells were also stained with individual conjugated antibodies for control and compensation. After incubation for 15 minutes, cells were washed with Dulbecco's phosphate-buffered saline, centrifuged, and resuspended in 500 μm PBS. These samples were run directly on the flow cytometer and analyzed using the ISHAGE gating protocol

[16] . For each sample, a total of 100,000 events gated on the singlet gate were collected.

[0329] Colony forming unit (CFU) assay

[0330] First, adjust the concentration of the RBC lysed cell suspension to 10 5 The cells were then vortexed vigorously to achieve adequate mixing. A 3cc syringe was used to remove at least 2.2 mL of Methocult containing cells. 1.1 mL was distributed to two 35 mm non-tissue culture treated culture dishes. The culture dishes were covered and tilted to ensure that the entire plate surface was coated with Methocult. Two culture dishes were placed in a larger 100 mm petri dish with a third uncovered 35 mm culture dish containing sterile DI water to wet the plate. The plate was incubated at 37 ° C, 5% CO2 for 14 days, and then the colonies were scored.

[0331] Number of donors and bone marrow samples used for statistical modeling

[0332] 75 bones from 62 donors were initially received at one of two BM processing facilities. The number of samples with complete data records varied by outcome modeled. Table 1 provides a breakdown of the number received and the number with complete data available for statistical modeling by outcome.

[0333] Table 1. Number of donors and bones available for analysis by outcome

[0334] Totals for complete data used for analysis:

[0335]

[0336] Definition of ischemic time

[0337] Total ischemia is defined as the interval from the time of death (when the donor's arterial system is cross-clamped and circulation stops) to the start of BM recovery in the processing facility. For the purpose of statistical modeling, this total interval is divided into three continuous and mutually exclusive time components: (a) warm ischemia time (WIT): starts from the time of death and ends when the bone is recovered and stacked on ice or the body is placed in a cooler. (b) body cooling time (BCT): starts when the body is placed in a cooler and ends when the recovered bone is stacked on ice. (c) cold ischemia time (CIT): starts when the recovered bone is stacked on ice and ends when processing for HSPC extraction begins. According to these definitions, total ischemia time = (WIT) + (BCT) + (CIT). When body cooling is not used, BCT is coded as zero and total ischemia time = (WIT) + (CIT). Ischemia time is considered the main variable of interest in the predictive outcome model.

[0338] Definition of Experience

[0339] Because this was our first series of BM processed from cadaveric bone, we hypothesized that HSPC quality might improve with learning as we gained more experience with processing. This hypothesis was based on long-established research demonstrating that learning curves have a significant impact on outcomes and costs in both industrial manufacturing

[17] and medical practice settings [18–20]. To control for learning, we created the variable experience (EXPERIENCE), defined as the number of donors processed before the current donor. For the i-th donor, experience was coded as i-1 to indicate that experience was always one less than the sequence number of the current case being processed. Because facility A began processing BM five months earlier than facility B and because facility B had the advantage of participating in and learning from cases processed by facility A, we hypothesized that the two facilities would have different learning trajectories. To account for this possible difference, experience was coded separately for each facility. To identify the facility in the model, we coded facility A = 1 and facility B = 0. The effect of experience was initially estimated in a separate regression model and then incorporated as a covariate in the final adjusted model to control for the effect of learning on outcomes.

[0340] Other covariates

[0341] Other variables tested in the statistical model were: (1) BONE TYPE, vertebral body (VB) and ilium (IL), (representing two sources of BM cells, coded VB=1; IL=0); donor sex (DONOR SEX) (percentage of males); and donor age (DONOR AGE) (years). These additional covariates were considered exogenous factors and were included in the final model only if they were statistically significant or improved the performance of the model.

[0342] Outcome Variable

[0343] Results were defined as markers of potential in vivo efficacy based on three qualitative measures: (a) the proportion of viable recovered CD34+ cells as determined by 7-AAD (%CD34+), (b) the total number of colony forming units (CFU) / 10 5 The total number of nucleated cells (TNC) (CFU-total) plated per cell, and (c) the number of CFU granulocyte-macrophage cells detected / 10 5 Nucleated cells (CFU-GM).

[0344] Summary Statistics

[0345] Donor and treatment facility characteristics, ischemic time, and outcome measures were summarized as means or percentages as appropriate. Crude (unadjusted) comparisons were made between FACILITY (A vs. B), bone type (VB vs. IL), and BODY COOLING (yes or no) using independent groups t-tests or z-tests for proportions.

[0346] Statistical Modeling

[0347] The association of ischemic time with outcome was initially investigated using only ischemic time as a predictor in an unadjusted regression model. Additional models were then estimated to determine the individual association of experience with outcome. Finally, the impact of ischemia was evaluated in a multivariate model controlled for the potential effects of facility, experience, bone type, donor sex, and donor age. Separate models were estimated for each of the three outcomes of interest (%CD34+, CFU-total, and CFU-GM).

[0348] Ordinary least squares (OLS) linear regression was used to test a range of candidate models, including models containing two-way interactions as well as logit and second-order polynomial terms. From these candidates, the best reduced model was selected based on the following criteria: (a) having the greatest explanatory power (highest R 2 (b) The parsimonious model that explains the greatest percentage of variance with the fewest predictors is favored. The adjusted R2 of

[21] is used to prevent over-specification by penalizing models that include more predictors. 2It is used as a comparative indicator of explanatory power when selecting the most parsimonious model. 2 value while maintaining or increasing the adjusted R 2 (c) Models with greater precision (as indicated by relatively smaller standard errors associated with both the model and the model coefficients) are favored. (d) Models with the best fit as judged by evaluation of residual plots are favored. Plot the residuals and visually inspect for discernible patterns and confirm quantitatively by regressing the residuals onto the observed values ​​to reveal possible interactions or potential curvilinear relationships.

[0349] Because %CD34+ is a proportion restricted to a closed unit interval, [0 < (%CD34+) < 1], we found that traditional OLS linear regression produced unrealistic fitted values ​​outside the boundaries of these intervals. To correct for this problem, we replaced linear regression with beta regression in the model of %CD34+

[22] . Beta regression is useful when the response variable is a ratio or proportion measured on a continuous scale and subject to minimum and maximum constraints. We modeled the transformed variable, pCD34* = [100 × (%CD34+) + 1] / 102, which satisfies the distributional assumption of beta regression, namely that the outcome variable must be restricted to an open interval, [0 < (%CD34+) < 1]. In order to be able to report the predicted values ​​in their original percentage units, the beta regression results were back-transformed to obtain:

[0350] Pred(%CD34+)=[102×(Pred(pCD34*))-1 / 100]

[0351] (A technical description of the beta regression model can be found in the online Supplementary Technical Appendix A)

[0352] Model Validation All models were validated using leave-one-out bootstrap cross-validation

[23] , which is done by randomly omitting one observation from the dataset with replacement and reestimating the model on the remaining observations. The resulting model is then used to predict the omitted observation. This process is repeated 200 times, producing 200 models with predictions, model coefficients, standard errors, and 95% confidence intervals. Model parameters are summarized as the mean of the 200 bootstrapped models. Since the bootstrapped model is naive to the omitted observations, this form of validation serves as an estimate of the predictive accuracy that might be seen when the original model is used to predict new observations

[21] . Model coefficients are reported for the original model and compared to the mean coefficients from the 200 cross-validated models ± 95% confidence intervals.

[0353] result

[0354] Table 2 provides sample characteristics, and Figure 1 shows the distribution of total ischemic time for each of the 62 donors as well as the individual ischemic time components WIT, CIT, and body cooling (BCT). Most donors (77.2%) were male. The mean donor age was 41.2 years. The mean ischemic time in hours (± standard error) was 3.6+0.4 for WIT, 7.9+0.9 for BCT, and 19.6+1.2 for CIT. The mean total ischemic time was 31.0+1.2 hours. An average of 2.43+0.64% CD45dim CD34+ HSPCs were recovered from BM samples, of which an average of 79.3+3.0% were viable cells. The BM contained an average of 250.3±49.48 CFU-total / 10 5 Total nucleated cells (TNC) and 38.2+7.78 CFU-GM / 10 5 A TNC for laying planks.

[0355] Table 2. Sample characteristic values ​​are values ​​related to the %CD34+ model

[0356]

[0357] * Average number of cases processed before the current case

[0358]

[0359] Figure 1. Cumulative ischemic time for all donors used in this study for whom complete data were available. Donors were ranked from shortest to longest total ischemic time, which is a composite of WIT, CIT, and BCT.

[0360] Unadjusted comparison

[0361] Comparisons of facilities, bone types, and body cooling are shown in Table 3. The distribution of donor age and donor sex did not differ significantly by facility, bone type, or whether body cooling was used.

[0362] There was a significant difference in the distribution of bone types between the facilities (VBs accounted for 27% of the bones processed at Facility A, compared to 100% at Facility B) because Facility B was structured to receive only VBs. Facility A also had more experience (Facility A = 53 bones processed, compared to Facility B = 24 bones processed; p<0.00001), a significantly longer WIT (Facility A = 3.55 hours, compared to Facility B = 2.13 hours; p = 0.003), and a significantly shorter CIT (Facility A = 19.55 hours, compared to Facility B = 28.38 hours; p = 0.004). The two facilities did not differ in BCT or total ischemia time. The results differed only in CFU-GM counts, which were significantly lower at Facility A than at Facility B (28.38 and 64.31, respectively; p = 0.04). There were no significant differences between the two facilities in the percentage of viable CD34+ or CFU-total. Facility differences were controlled for in the final regression model.

[0363] The differences in BCT according to bone type (Table 3 middle) approached significance (p = 0.09), with treatment with VB associated with a shorter BCT (6.32 hours) compared to IL (8.51 hours). This is because Facility B, which only treated VB, had a shorter BCT than Facility A. The results also differed by bone type. VB produced a higher number of CFU-total compared to IL (341.29 vs. 97.44 / 10 5 cells; p = 0.02) and CFU-GM (50.46 and 18.03 / 10 5 cells; p = 0.04). Bone type was controlled in the final regression model.

[0364] In cases where the body was refrigerated prior to bone retrieval (rightmost portion of Table 3), the mean WIT tended to be significantly shorter (2.65 hours compared to 3.98 hours without body cooling, p=0.04). The same was true for CIT (19.51 hours compared to 28.83 hours without body cooling, p=0.009). Notably, all outcomes were worse when body cooling was employed. The mean %CD34+ viability was 72.75% compared to 89.86% without body cooling (p=0.0001). Similarly, the mean CFU-total counts were 100.16 and 659.00 / 10 with and without body cooling, respectively. 5 The mean CFU-GM counts were 18.52 and 94.85 / 10 TNCs (p=<0.00001). 5 Body cooling was considered in both the initial and final ischemic time regression models.

[0365]

[0366] Ischemia time regression model

[0367] Unadjusted (base) regression models used only WIT, BCT, and CIT as predictors (no adjustment for other covariates). These models are summarized in the Online Supplement, Technical Appendix C, Tables S3-S5. As bone type, facility, and experience were found to be significant variables associated with the outcomes (Table 3), adjusted models were developed to statistically control for the effects of these covariates.

[0368] The beta regression model predicting %CD34+ viability is shown in Table 4. (Details of the beta regression are provided in the Online Supplement, Technical Appendix A). The percentage of viable CD34+ cells recovered decreased significantly with increasing BCT, with the decrease occurring at a decreasing rate as %CD34+ approached zero (linear effect, p=0.002; second-order polynomial effect, p=0.03). A similar curvilinear decrease in %CD34+ occurred in association with increasing CIT (linear effect, p=0.003; second-order polynomial effect, p=0.005). Neither bone type nor WIT was significant. Experience (p=0.09) and facility × experience interaction (p=0.07) approached statistical significance. The odds ratio measures the change in %CD34+ associated with a one-unit change in the associated predictor variable. For example, the odds ratio associated with a one-hour increase in WIT is 0.9663, indicating that each one-hour increase in WIT reduces %CD34+ to 96.63% of its previous value. The predictive validity of the model is demonstrated by the similarity of the estimated parameters of the original model (left panel of Table 4) and the estimated parameters of the bootstrapped model (right panel). The model is statistically significant (p=0.001).

[0369] The linear regression results of CFU-total are shown in Table 5. This reveals the importance of bone type as the source of BM cells. When BM cells were recovered from VB instead of IL, CFU-total increased by 207 / 10 5 The effect of BCT on CFU-total was negative. As BCT increased, CFU-total recovery decreased, with the decrease occurring at a decreasing rate (linear effect, p=0.00005; second-order polynomial effect, p=0.002). The effects of WIT and CIT were not statistically significant. Experience was also not significant, however, experience was retained as model performance improved when experience was statistically controlled. When both bone type and experience were statistically controlled, the explanatory power of the model increased from R 2 =35% to 47%. The adjusted R 2The model also increased from 35% to 40%, indicating that the increase was not the result of over-specification of the model. The model accuracy also improved, as shown by the smaller standard error. The similarity of the estimated parameters of the original model (left sub-figure of Table 5) and the average results of the bootstrap model (right sub-figure) demonstrated the predictive validity of the model. The model was significant (p = 0.000005) and explained 47.3% of the variance in CFU-total.

[0370] The linear regression results for CFU-GM are shown in Table 6. The best CFU-GM model included bone type but did not include experience or facility as control variables. Although bone type was not statistically significant, it was retained in the model because its inclusion shifted the explanatory power from R 2 =32% to 34%, while the adjusted R 2 The association between WIT and BCT and CFU-GM remained unchanged (29%), indicating that the model was not over-specified. When statistically controlled for bone type, WIT and BCT continued to demonstrate a statistically significant association with CFU-GM. WIT reduced CFU-GM by -7.19 / 10 per hour 5 TNC per hour (p = 0.03), while BCT reduced CFU-GM by -5.24 / 10 5 The TNC of each plate (p = 0.00003). CIT had no effect (p = 0.86). The predictive validity of the model is demonstrated by the similarity of the parameters of the original model (left panel of Table 6) and the average results of the bootstrap model (right panel). The model is significant (p < 0.00001) and explains only 34% of the variance in CFU-GM.

[0371]

[0372]

[0373] Predictions from the %CD34+ model

[0374] The series of forecasts generated from the adjusted beta regression model in Table 4 are as follows Figure 2 The predicted patterns illustrate how various combinations of WIT and CIT alter the viability of recovered CD34+ cells. Figure 2The predictions in are what would be expected if body cooling was not employed. Inspection of the range of WIT and CIT values ​​indicates that WIT is more detrimental to cell viability than CIT. When WIT was maintained at 3 h or less, viability of CD34+ cells remained at or above 80% (green area) for up to 24 h of CIT. However, the amount of CIT that could be tolerated progressively shortened as WIT was extended beyond 3 h. We did not test the effect of cryopreservation, so these predictions do not account for potential loss of viability due to subsequent freezing and thawing of retrieved cells. Similar predictions can be made for CFU-total and CFU-GM using the coefficients provided in Tables 5 and 6

[0375]

[0376] discuss

[0377] This pragmatic, field observational study is the largest to date in terms of the total number of donors and the amount of bone obtained and is the first to quantify the effects of ischemia and body cooling time on the quality of HSPCs retrieved from deceased donor bone. The study was designed to generate externally validated data that can be generalized and applied in a routine practice setting. The study covered the full continuum of ischemia time that is likely to occur under normal OPO operating conditions and, unlike previous studies conducted at a single institution, donor bone was retrieved immediately after cardiac arrest (i.e., without body cooling) where bone was retrieved rapidly (i.e., short WIT) and did not require prolonged periods of transport (i.e., reduced CIT) [4,14,15].

[0378] This study had three main objectives: (a) to quantify the statistical relationship between ischemia time and the quality of HSPCs from deceased donors, (b) to identify the cut-off conditions beyond which longer ischemia times adversely affect the quality of HSPCs, and (c) to investigate whole-body cooling as a strategy to preserve cell function and viability.

[0379] The findings convey four main messages.

[0380] First, acceptable levels of HSPC quality can be achieved despite the inevitable prolonged ischemia times when bone must be procured by geographically dispersed OPOs and shipped across borders to distant processing centers. Our analysis showed that under these conditions, a favorable combination of warm and cold ischemia times was easily achieved, resulting in CD34+ cell viability in the 80%-90% range. Overall, the unadjusted mean percentage of viable CD34+ cells recovered was slightly less than 80% (79.3%, Table 2).

[0381] The second message is that freezing the body before bone recovery, a common practice in tissue recovery, is detrimental to the viability and function of HSPCs recovered from cadaveric BM. When whole-body cooling was used, CD34+ cell viability averaged 72.75%; when no body cooling was used, the average viability reached nearly 90% (89.96%, Table 3), suggesting that best practice is to eliminate body cooling and transfer the recovered bone to a cold ischemic environment as quickly as possible.

[0382] Third, the source of BM (bone type) is important. Our analysis shows that VB is a superior source of viable HSPCs compared to IL. In unadjusted comparisons, CD34+ cell viability exceeded 80% when VB was the source, but fell below 74% when IL was the source (Table 3). The reasons for this difference are unclear and may be multifactorial. Variations in the isolation procedures used in the two bone types may be a more important factor than physiological differences. Given that this is the first study comparing BM from deceased donors VB and IL, no other directly comparable data exist. The closest approximation is a comparison of the viability of CD34+ cells recovered from deceased donor VB and BM aspirated from the iliac crest of living donors, which showed no difference [24-26].

[0383] The fourth message conveyed by our analysis is that experience is important and can vary significantly across processing centers. Like most technical activities, processing BM cells from cadaveric bone follows a learning curve. It is well known that industrially manufactured products improve with learning, a phenomenon first documented more than 80 years ago

[27] and subsequently incorporated into standard textbooks of operations management [17,28]. More recently, it has been shown that the learning curve phenomenon extends to the outcomes and costs of medical procedures [18–20]. We observed different learning trajectories among the BM processing centers studied [results are provided in the Online Supplement, Technical Appendix B] and, although we analyzed only two centers, our results suggest that the rate of learning and the shape of the learning curve may vary significantly across centers, which is a factor that should be considered when designing future training programs, BM processing protocols, and accreditation practices. Our analysis suggests that there is a volume-outcome relationship for BM processing and that centers in high-volume regions that have accumulated more processing experience are likely to produce higher-quality BM products compared with low-volume centers.

[0384] Although the present study was not designed to optimize yield or viability, some comparisons can be made with data from previous reports of deceased human donor BM recovery where optimization was the goal. Three previous studies compared BM from a total of 99 deceased donors with BM from a total of 58 living donors [24-26]. In these reports, the percentage of CD34+ cells from deceased donor BM (2.1+0.6%; mean±SD) was not statistically different from BM aspirated from living donors (1.56+0.92%) (p=0.32). This contrasts well with our findings, in which the mean percentage of CD34+ cells recovered was 2.43+0.64% (mean±SD). We did observe greater variability in CD34+ percentages, which may reflect the extreme range of ischemic times and, therefore, the quality of the donor tissue when it arrived for our study.

[0385] The quality of deceased donor CD34+ HSPCs has been compared to that of living donor HSPCs by measuring CD34+ cell viability and CFU potential [24-26]. The mean viability of CD34+ cells recovered from deceased donors was 95.2 ± 3.6%, compared to 93.5 ± 0.35% in the case of living donors. The functional equivalence of deceased and living donor BM HSPCs was determined by comparing the frequency of CFU-GM, which was 105 ± 65 / 10 in deceased donor BM. 5 In contrast, in living donor BM, it was 81.4 ± 17 / 10 5 By comparison, our overall means for both of these quality metrics from deceased donor BM (Tables 2 and 3) were lower, presumably again due to the extreme range of ischemia times and the inclusion of body cooling in our study which negatively impacted mean cell viability. We subsequently used the results reported here to establish limits of 8 h WIT and 30 h CIT and eliminated the practice of body cooling. Following this protocol change, vertebrae from 50 donors meeting these criteria have now been recovered and processed (manuscript in preparation). Mean CD34+ HSPC viability was 90.5 ± 1.9% and mean CFU-GM was 158.3 ± 13.5 / 10 5 The number of plated TNCs was comparable to previous studies [24–26].

[0386] Overall, our study demonstrates the feasibility of recovering large quantities of BM from deceased donors for storage. Establishing a BM bank with adequate HLA diversity requires an adequate source and a stable supply of deceased donor BM bone. Fortunately, the Uniform Anatomical Gift Act of 1968 established a consortium of 58 geographically distributed OPOs. Approximately 10,000 deceased individuals donate their organs and an additional 40,000 donate tissues each year, resulting in the recovery of approximately 30,000 organs and more than 1 million tissues per year (unos.org / data / transplant-trends / accessed 29 November 2019). The large volume of bone recoverable through this network could provide the necessary inventory to justify the establishment of an integrated system for bone procurement, recovery, and transportation associated with BM processing and storage centers. This type of integrated system would require the cooperation and coordination of multiple OPOs, all following agreed-upon operating protocols. Our study demonstrates the feasibility of building such a system using existing OPO infrastructure. In particular, we have demonstrated that protocols can be developed and implemented to maintain a favorable ischemic environment from the point of bone procurement and recovery through cross-country transportation to the BM processing center. Because our data are not limited by variables such as body cooling and ischemic time, they are likely to have a high degree of external validity (generalizability), and the results of our predictive model ( Figure 2 ) can be used to establish realistic ischemia time tolerance limits and HSPC quality acceptance criteria.

[0387] The creation of a BM banking system would offer several distinct advantages over current living donor registries. First, the personal risk to living donors would be eliminated, rather than being ameliorated solely by the current practice of primarily mobilized peripheral blood collection. Second, greater numbers of HSPCs could be recovered from deceased donors compared with living donors, allowing for multiple infusions from the same donor in the event of graft failure. Furthermore, the recovered cells could be packaged in known quantities, tested for quality, and cryopreserved for later use as needed. Because these units are cryopreserved, they can be stored indefinitely

[29] , thereby avoiding the attrition issues that occur with living donor registries. Finally, BM banks could serve as a readily available resource during surges in demand following a mass casualty event, such as a nuclear disaster that results in widespread bone marrow failure [1,2].

[0388] To fully realize these advantages, many logistical and systemic issues will have to be addressed. Chief among these is the recognition that the extended ischemia times introduced during bone retrieval, transport, and processing need to be effectively managed to ensure the quality and yield of the BM product. BM banks are now established and are beginning to show great potential. From the perspective of nascent BM processing facilities, the results of our statistical model can be used to establish quantitative ischemia tolerances and quality acceptance criteria to protect the viability and function of HSPCs derived from cadaveric bone. From a broader policy perspective, our model can also provide a basis for the development of data-driven industry standards for emerging BM storage systems.

[0389] References

[0390] 1.Knebel, AR, et al., Allocation of scarce resources after a nucleardetonation: setting the context. Disaster Med Public Health Prep, 2011.5 Suppl1: p.S20-31.

[0391] 2. Weinstock, DM, et al., Radiologic and nuclear events: contingency planning for hematologists / oncologists. Blood, 2008.111(12): p.5440-5.

[0392] 3.Kawai, T., et al., Long-term results in recipients of combined HLA-mismatched kidney and bone marrow transplantation without maintenance immunosuppression. Am J Transplant, 2014.14(7): p.1599-611.

[0393] 4.Schneeberger, S., et al., Upper-extremitytransp / lntation using a cell-based protocol to minimize immunosuppression. Ann Surg, 2013.257(2): p.345-51.

[0394] 5.Spitzer,T.R.,et al.,Twenty Year Follow Up of HistocompatibilityLeukocyte Antigen-Matched Kidney and Bone Marrow Co-Transplantation forMultiple Myeloma with End Stage Renal Disease:LessonsLeorned.Transplantation,2019.

[0395] 6.Hotta,K.,et al.,Long-term Nonhuman Primate Renal Allograft SurvivalWithout Ongoing Immunosuppression in Recipients of Delayed Donor Bone MarrowTransplantation.Transplantation,2018.102(4):p.e128-e136.

[0396] 7.Yamada,Y.,et al.,Overcoming memory T-cell responses for inductionofdelayed tolerance in nonhuman primates.Am J Transplant,2012.12(2):p.330-40.

[0397] 8.Eckardt,J.R.,et al.,Comparison of engraftrnent and acute GVHD inpatients undergoing cryopreserved or fresh allogeneic BMT.Bone MarrowTransplant,1993.11(2):p.125-31.

[0398] 9.Lioznov,M.,et al.,Transportation and cryopreservation may impairhaematopoietic stem cell function and engraftment of allogeneic PBsCs,but notBM.Bone Marrow Transplant,2008.42(2):p.121-8.

[0399] 10.Stockschlader,M.,et al.,Use of cryopreserved bone morrow inallogeneic bone marrow transplantation.Bone Marrow Transplant,1995.15(4):p.569-72.

[0400] 11.Stockschlader,M.,et al.,Use of cryopreserved bone marrow inunrelated allogeneic transplantation.Bone Marrow Transplant,1996.17(2):p.197-9.

[0401] 12.AATB,Guidance Document,in Evaluation of Body Cooling at StandardD5.400.2013,American Association of Tissue Banks:McLean,VA.p.13.

[0402] 13.Schwartz,D.and J.Lellouch,Explanatory and pragmatic attitudes intherapeutical trials.J Chronic Dis,1967.20(8):p.637-48.

[0403] 14.Donnenberg,A.D.,et al.,clinical implementation of a procedure toprepare bone marrow cells from cadaveric vertebral bodies.Regen Med,2011.6(6):p.701-6.

[0404] 15.Gorantla,V.S.,et al.,Development and validation of a procedure toisolate viable bone marrow cells from the vertebrae of cadaveric organ donorsfor composite organ grafting.Cytotherapy,2012.14(1):p.104-13.

[0405] 16.Sutherland,D.R.,et al.,The ISHAGE guidelines for CD34+celldetermination by flow cytometry.International Society of Hematotherapy andGraft Engineering.J Hematother,1996.5(3):p.213-26.

[0406] 17.Riahi-Belkaoui,A.,The learning curve:a mannagement accountingtool.1986,Westport,Conn.:Quorum Books.xiii,245pages.

[0407] 18.Flood,A.B.,W.R.Scott,and W.Ewy,Does practice make perfect? Part1:The relations betweeen hospital volume and outcomes for selected diagnosticcategories.Medical Care,1984.22(2):p.98-114.

[0408] 19.Flood,A.B.,W.R.Scott,and W.Ewy,Does practice make perfect?Part II:The relation between volumes and other hospital characteristics.MedicaI Care,1984.22(2):p.115-125.

[0409] 20.Woods,J.R.,et al.,The learning curve and the cost of hearttransplantation.Health Serv Res,1992.27(2):p.219-38.

[0410] 21.Harrel Jr,F.E.,Regression modeling strategies with applications tolinear models,logistic regression,and survival analysis.2nd ed.SpringerSeries in Statistics.2001,New York:Springer.582.

[0411] 22.Ferrari,S.L.P.and F.Cribari-Neto,Beta regression for modelingrates and proportions.J.Applied Statistics,2004.31(7):p.799-815.

[0412] 23.Picard,R.and D.Cook,Cross-validation of regressionmodels.J.Am.Stat.Assoc,1984.79(428):p.1303-1313.

[0413] 24.Ahrens,N.,et al.,Mesenchymal stem cell content of human vertebralbone marrow.Transplantation,2004.78(6):p.925-9.

[0414] 25. Rybka, W.B., et al., Hematopoietic progenitor cell content pf vertebral body marrow used for combined solid organ and bone marrow transplantation. Transplantation, 1995. 59(6): p. 871-4.

[0415] 26. Soderdahl, G., et al., Cadaveric bone marrow and spleen cells for transplantation. Bone Marrow Transplant, 1998. 21(1): p. 79-84.

[0416] 27. Wright, T., Factors affecting the cost of airplanes. J Aeronautical Sciences, 1936. 3(2): p. 122-128.

[0417] 28. Green, J.H., Operations Management: Productivity and Profit. 1984: Reston Pub Co. 723.

[0418] 29. Woods, E.J., et al., Off the shelf cellular therapeutics: Factors to consider during cryopreservation and storage of human cells for clinical use. Cytotherapy, 2016. 18(6): p. 697-711.

[0419] Online supplement

[0420] Technical appendix A

[0421] Empirical model

[0422] To account for learning, we created a variable experience, defined as the number of donors processed before the current donor. Donors are numbered consecutively from i=1...n in the order in which they were processed, and experience is coded as i-1 to indicate that experience is always 1 less than the current donor being processed. Facility A began processing bone marrow 5 months before facility B, and because facility B had the advantage of participating in (and learning from) cases processed at facility A, the two facilities had different learning trajectories. To account for this difference, experience is coded separately for each facility. To identify the two facilities in the model, we code facility A=1 and facility B=0.

[0423] Regression Model :

[0424] Results (%CD34+, CFU-total / 10 5 and GM-Total / 10 5 ) was modeled as a linear combination of facility (where the treatment occurred), experience (number of cases treated at the facility before the current case), and the facility x experience interaction. %CD34+ was modeled using beta regression (see Technical Appendix B). The other two outcomes (CFU-total and CFU-GM) were modeled using traditional OLS linear regression. The model has the following linear form:

[0425] [A.1]Y = β0 + β1 (facilities) + β2 (experience) + β3 (facilities × experience)

[0426] in :Y=results(%CD34+、CFU / 10 5 or GM / 10 5 )

[0427] β0=intercept (constant term)

[0428] β1 = coefficient related to facilities

[0429] β2 = coefficient related to experience

[0430] β3 = coefficient related to the facility × experience interaction

[0431] The interaction term β3 accounts for the possibility that facility A may have a different linear relationship with experience (different learning trajectory) compared to facility B.

[0432] Algebra for deriving testable effects

[0433] The model for facility A is:

[0434] [A.2]Y = β0 + β1 (facility A) + β2 (experience) + β3 (facility A × experience)

[0435] =β0+β1(1)+β2(experience)+β3(1×experience)

[0436] =(β0+β1)+(β2+β3)×(experience)

[0437] The model for facility B is simplified because facility B = 0 and the terms associated with β1 and β3 are omitted from the model. Therefore, the model for facility B is:

[0438] [A.3]Y = β0 + β1 (facility B) + β2 (experience) + β3 (facility B × experience)

[0439] =β0+β1(0)+β2(experience)+β3(0×experience)

[0440] =β0+β1(0)+β2(experience)+β3(0)

[0441] =(β0)+(β2)×(experience)

[0442] The difference between equations [A.2] and [A.3] provides insight into the effects that can be tested in the model:

[0443] According to [A.4], the following effects are testable as null hypotheses:

[0444] H01: β0 = 0 tests the significance of the intercept for Facility B relative to the zero intercept. (Results for Facility B when Experience = 0).

[0445] H02: β1 = 0 Tests the significance of the intercept for Facility A relative to the intercept for Facility B. (Results for Facility A relative to Facility B when Experience = 0).

[0446] H03: β2 = 0 tests the significance of the learning slope of facility B (the change in outcomes of facility B associated with each additional learning experience).

[0447] H04: β3=0 Test the significance of the learning slope of facility A relative to the slope of facility B.

[0448] Example Calculations

[0449] The following is a method for predicting CFU-GM / 10 given the regression coefficients and observed data in Table S1. 5 Working examples of the effects of experience:

[0450] Table S1 Regression coefficients and ischemic time values

[0451]

[0452] Assuming both Facility A and Facility B have processed 20 previous donors, what will their respective results be for the 21st donor?

[0453] [A.5] Expected results for Facility B :

[0454] CFU-GM / 10 5 =(β0)+(β2)×(experience)

[0455] =111.91+(-3.57)×(20)

[0456] =40.51

[0457] Explanation for Facility B :

[0458] Starting with no experience (experience = 0), the initial CFU-GM yield (intercept term) is expected to be β0 = 111.91 CFU-GM / 10 5 Each additional donor processed is then expected to convert β2 = -3.57 CFU-GM / 10 5 Subtract from the starting amount for Facility B. For the 21st donor processed, the expected yield for Facility B would be (β0) + (β2 x 20) = 111.91 + (-3.57 x 20) = 40.51 CFU-GM / 10 5 .

[0459] [A.6] Expected results for Facility A :

[0460] CFU-GM / 10 5 =(β0+β1)×(facilities)+(β2+β3)×(facilities×experience)

[0461] =(111.91-99.34)×(1)+(-3.57+4.17)x(1×20)

[0462] =(12.57)+(0.60×20)

[0463] =(12.57)+(12.00)

[0464] =24.62

[0465] Explanation of Facility A :

[0466] Without experience, the yield of facility A was estimated to be β0+β1=[111.91+(-99.34)]=12.57 CFU-GM / 10 5units higher than Facility B. Each additional case processed by Facility A increases the initial yield of Facility A by β2+β3=(-3.57+4.17)=0.60 CFU-GM / 10 5 For the 21st case processed, the yield of Facility A would be (β0+β1)+(β2+β3)×(20)=12.57+(0.60×20)=24.57 CFU-GM / 10 5 Note that the learning slope for facility A is positive, increasing by β3 = 4.17 CFU-GM / 10 with each additional case processed. 5 , while the learning slope of Facility B is negative, with each case minus β2 = -3.57 CFU-GM / 10 5 This illustrates a classical interaction. Expressed in relative terms, each additional learning experience for facility A relative to facility B is associated with β2+β3 = (-3.57+4.17) = 0.60 CFU-GM / 10 5 The net increase is related to .

[0467] These examples illustrate the pattern that emerges for all three results. Facility A, which began processing BM cells before Facility B, starts at a relatively low performance level and improves monotonically with each additional case processed. In contrast, Facility B, which has participated in (and learned from) Facility A's initial work, starts at a higher performance level but does not change significantly or declines slightly with experience.

[0468] Technical Appendix B

[0469] Beta Regression

[0470] In the text, we use CD34+ to indicate the recovered CD34+ cell count and we use %CD34+ to indicate the percentage of viable total CD34+ cells. That is:

[0471] %CD34+ = (live CD34+) / [(live CD34+)+(non-viable CD34+)]

[0472] Because it is a ratio, %CD34+ is restricted to the unit closed interval (0 ≤ %CD34+ ≤ 1), meaning it can take on values of 0% or 100% or any value in between, but it cannot be less than 0% or greater than 100%. Given this restriction, ordinary least squares (OLS) linear regression produces unrealistic fitted values outside the bounds of this interval - some predicted values are less than 0% and some exceed 100%. To correct for this, we considered using beta regression [1] in place of OLS linear regression for the model of %CD34+. Maximum likelihood beta regression is used to model random variables with a beta-distribution, making it particularly useful in situations such as ours, where the dependent variable is a ratio or proportion measured on a continuous scale and bounded by a minimum and maximum value. We continued to use OLS linear regression to model the other two outcome variables, CFU-total and CFU-GM.

[0473] Here we use pCD34 = %CD34+ to denote the percentage of viable CD34+ cells recovered. To ensure that the results are evaluable as variables with a beta-distribution, we transformed pCD34 as follows:

[0474] pCD34* = [1 + 100(pCD34)] / 102

[0475] This transformation restricts pCD34* to the open interval (0 < pCD34* < 1), thus satisfying the distributional assumption that the outcome variable can approach but not equal 0% or 100%. The restricted proportion pCD34* was then modeled using beta regression. For ease of interpretation, the predicted values from the beta regression were back-transformed to obtain:

[0476] Pred(pCD34) = [(102(Pred(pCD34*)) - 1) / 100] = Pred(pCD34) = Pred(%CD34+).

[0477] Beta regression equation :

[0478] The beta regression equation uses the logit link function of the outcome for the η linear predictor. The basic beta regression equation we used to predict pCD34* is:

[0479] [B.1] η = ln[pCD34* / (l - pCD34*)] = β0 + β1(WIT) + β2(BCT) + β3(BCT 2 )β4(CIT) + β5(CIT 2 )

[0480] in :

[0481] β0 = constant (intercept)

[0482] β1 = coefficient related to warm ischemia time (WIT)

[0483] β2 = coefficient related to body cooling time (BCT)

[0484] β3 = coefficient related to the square of body cooling time (BCT 2 )

[0485] β4 = coefficient of correlation with cold ischemia time (CIT)

[0486] β5 = coefficient of correlation with the square of cold ischemia time (CIT 2 )

[0487] Example Calculations :

[0488] To illustrate the calculations, we used the coefficients and ischemia times shown in Table S2.

[0489] Table S2 Regression coefficients and ischemic time values

[0490]

[0491] Using equation [B.1], we solve for η as shown in equation [B.2] below:

[0492] [B.2]η=ln[pCD34* / (1-pCD34*)]=β0+β1(WIT)+β2(BCT)+β3(BCT 2 )+β4(CIT)+β5(CIT 2 )

[0493] =3.500+(-0.01996)(1.92)+(-0.181)(0)+(0.007)(0)+(-0.111)(14.92)+(0.002)(222.606)

[0494] =2.2507688

[0495] Because they are related to the outcome variable through a nonlinear function, the coefficients of the linear predictor η lack a simple intuitive meaning. However, by applying the anti-link function to η, we obtain results that are easier to interpret.

[0496] Anti-join function :

[0497] The inverse link function exp(η) / [1+exp(η)] converts the linear predictor η into the expected value of the outcome variable pCD34*:

[0498]

[0499] By applying the inverse link function to the predicted value, and calculating η = 2.2507688 from equation [B.2], we obtain the expected value E[pCD34*]:

[0500] [B.3]E[pCD34*]=exp(2.2507688) / [1+exp(2.2507688]=0.905≈90.5%

[0501] This result can be interpreted as the expected value of pCD34* for the specific values ​​of the predictors given in Table S2.

[0502] To interpret the results of [B.3] in terms of pCD34* (percentage of viable CD34+ cells), we used the inverse transformation:

[0503]

[0504] Equation [B.4] states that for the values ​​specified in Table S.1, the expected percentage of viable CD34+ cells is pCD34 = %CD34+ = 91.31%.

[0505] Determine the expected effect on pCD34* of a one-unit change in a given predictor variable :

[0506] For any given predictor, the beta regression coefficient can be used to estimate the effect of a one unit change in that predictor on pCD34* while controlling for all other predictors in the equation. This is done by exponentiating the specific regression coefficient under consideration. For example, to calculate the effect of a one hour increment in warm ischemia time (WIT) on the ratio of the percentage of viable CD34+ cells to the percentage of non-viable CD34+ cells, the ratio under consideration is:

[0507]

[0508] According to Table S2, the regression coefficient associated with WIT is β1 = -0.01996. Therefore, the effect of a one-hour increment of WIT on the ratio of the percentage of viable CD34+ cells to the percentage of non-viable CD34+ cells is:

[0509]

[0510] Equation [B.5] states that if WIT increases by one hour, while the other variables in the equation (BCT and CIT) remain constant, the ratio of the percentage of live CD34+ cells to the percentage of non-viable CD34+ cells will decrease by 2%, to 98% of its previous value. In equation [B.3], we previously calculated pCD34*=0.905. Therefore, a one-hour increment in WIT will reduce pCD34* to 0.98×0.903=0.885, a decrease of 2%. Equivalently, this will reduce the predicted pCD34 from 0.913 to 0.893, a decrease of approximately 2%. The multiplication factor of 0.98 is a constant that can be applied to any unit change along the continuous range of warm ischemia time. Factors for other predictors can be obtained in the same manner to estimate the effect of a unit change in BCT and CIT on pCD34*.

[0511] Technical Appendix C

[0512] Unadjusted (baseline) ischemia time regression model

[0513] In the initial (basic) regression model, we used only WIT, BCT and CIT as predictors (no adjustment for other covariates). The results of these models for %CD34+, CFU-total and CFU-GM are shown in Tables S3-S5. The model results are shown in the left panel of the table; the average results of 200 cross-validation models (each model was estimated with one observation omitted from the full data set) are shown in the right panel.

[0514] The beta regression model for %CD34+ is shown in Table S3. The relationship between WIT and %CD34+ was not statistically significant, however, both BCT (linear component, p=0.001 and second-order polynomial component, p=0.01) and CIT (linear component, p=0.001 and second-order polynomial component, p=0.004) were curvilinearly associated with %CD34+. In both cases, CD34+ yield decreased with increasing BCT and CIT, but then increased slightly at the upper limit of BCT and CIT. The odds ratios in Table S3 are continuous variable ratios of live CD34+ cells to total CD34+ cells and describe the effect of a one-unit change in a given predictor. These quantities are obtained by exponentiating the regression coefficients associated with the specific predictor under consideration. For example, the coefficient associated with WIT is β1=-0.01996. (Table S3) Exponentiation produces the following results:

[0515] e β =e -001996 = 0.9802 or 98.02%,

[0516] This is the value of the continuous odds ratio for warm ischemia shown in Table S3. This result shows that, with BCT and CIT remaining constant, the ratio of viable to non-viable CD34+ decreases to 98.02% of its previous value for each unit increase in WIT (one hour). The multiplicative factor of 0.9802 is a constant that applies to a one-unit change anywhere in the continuous range of WIT. Other predictors are provided in Table S3 and can be used to estimate the effect of a one-unit change in BCT or CIT, where the other variables are held constant in the equation. The beta regression prediction equation is statistically significant (p=0.0009).

[0517] The right half of Table S3 shows the average results of the bootstrapped cross-validation, which provides an estimate of the effectiveness of the original model in predicting future observations [2]. If the original model is misspecified, the parameters of the re-estimated bootstrapped model will differ from those of the original model. However, as Table S3 reveals, the model parameters (regression coefficients, standard errors, and 95% confidence intervals) associated with the original model (left panel) are almost identical to the corresponding parameters generated by bootstrapped resampling (right panel), providing evidence of the predictive effectiveness of the original model when applied to future data drawn from the same population. [Technical details on beta regression and example calculations are provided in Technical Appendix B above].

[0518] Table S4 shows the linear regression results for CFU-total. The coefficients in the linear regression are direct estimates of the effect of a unit change in the associated predictor. BCT was the only statistically significant predictor in Table S4. WIT and CIT were not significant. The relationship between BCT and CFU-total was curvilinear, indicating that each hour increase in BCT decreased CFU-total by -95.03639 / 10 5 cells (p<0.0001) and CFU-total 2 Increase 3.45603 / 10 5 (p = 0.0008). In summary, the combination of linear and second-order polynomial components produces a downward trend in CFU-total attenuated at a decelerating rate. The model parameters (left sub-figure of Table S4) are similar to the average parameters of the bootstrap model (right sub-figure), which provides evidence of the predictive validity of the original model. The model is statistically significant (p = 0.00002) and explains 35% of the variance in CFU-total.

[0519] Table S5 shows the linear regression results of CFU-GM. In this model, the effects of WIT and BCT were statistically significant, while the effect of CIT was not significant. When CIT and BCT remained constant, WIT per hour reduced CFU-GM by -8.11295 / 10 5(p=0.01). When WIT and CIT remained unchanged, BCT per hour reduced CFU-GM by -5.52927 / 10 5 (p<0.000009). The right side of Table S5 shows that the estimated parameters of the bootstrap model are similar to those of the original model, which again provides evidence of the predictive validity of the original model when applied to future data. The model is statistically significant (p=0.00002) and explains 32% of the total variance in CFU-GM counts.

[0520]

[0521]

[0522] References

[0523] 1.Ferrari SLP, Cribari-Neto F.Beta regression for modeling rates andproportions.Journal of Applied Statistics.2004, 31(7):799-815

[0524] 2. Harrel Jr, FE, Regression modeling strategies with applications tolinear models, logistic regression, and survival analysis. 2nd ed. SpringerSeries in Statistics. 2001, New York: Springer.582

Claims

1. A method for obtaining stem cells from human cadaver bones, the method comprising the following steps: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaver bone to extract stem cells; and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT does not exceed 32 hours; and wherein the stem cells comprise CD34+ cells, wherein at least 70% of the recovered CD34+ cells are viable, Wherein when the sum of WIT and CIT is greater than 40 hours and / or WIT is greater than 8 hours and / or CIT is greater than 39.5 hours, fewer recovered CD34+ cells will be viable compared to when the sum of WIT and CIT is 40 hours or less and / or WIT is not less than half an hour to not more than 8 hours and / or CIT is not less than 7 hours to not more than 39.5 hours.

2. The method of claim 1, wherein the method provides at least 10 to 350 CFU-total, wherein CFU-total is detected for every 10 5 The total number of colony-forming units (CFU) of nucleated cells.

3. The method according to claim 1, wherein the method provides at least 10 to 100 CFU-GM, wherein CFU-GM is per 10 5 The number of CFU granulocytes and macrophages detected per nucleated cell.

4. The method of claim 1, wherein the cadaveric bone is derived from a vertebral body, ileum, or a combination thereof.

5. The method of claim 4, wherein the cadaveric bone is derived from multiple cadaveric bones from the same donor.

6. The method of claim 1, wherein the recovered stem cells comprise hematopoietic stem cells (HSCs).

7. The method of claim 1, wherein the recovered stem cells comprise mesenchymal stem cells (MSCs).

8. The method of claim 1, wherein at least 85% of the recovered CD34+ cells are viable.

9. The method of claim 1, wherein the body cooling time (BCT) is less than four hours, wherein the BCT begins when the cadaver is placed in the cooling environment and ends when the cadaver bone is placed in the cooling environment or condition.

10. The method of claim 9, wherein the BCT is less than 1 hour. The method of claim 10 , wherein the BCT is zero hours.

12. The method of claim 9, wherein at least 85% of the recovered CD34+ cells are viable.

13. The method of claim 11, wherein at least 90% of the recovered CD34+ cells are viable.

14. The method of claim 1, wherein the total ischemic time including the sum of WIT, CIT, and body cooling time (BCT) is less than forty hours, wherein the BCT begins when the cadaver is placed in a cooling environment and ends when the cadaver bone is placed in a cooling environment or condition.

15. The method of claim 14, wherein the BCT is less than one hour.

16. The method of claim 15, wherein the BCT is zero hours.

17. The method of claim 14, wherein at least 85% of the recovered CD34+ cells are viable.

18. The method of claim 16, wherein at least 90% of the recovered CD34+ cells are viable.

19. A method for recovering transplantable stem cells from cadaveric bone or cadaveric bone fragments, the method comprising: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaver bone to extract stem cells; and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT shall not exceed 32 hours; The extraction of stem cells includes: (a) obtaining the cadaveric bone and processing the cadaveric bone into cadaveric bone fragments; (b) mixing the cadaveric bone fragments with a grinding medium comprising a nuclease that cleaves both DNA and RNA and one or more of human serum albumin (HSA), heparin, an electrolyte medium, and a growth medium to obtain grinding medium-treated cadaveric bone fragments; (c) processing the grinding media treated cadaveric bone fragments to extract bone marrow cells; and (d) collecting the extracted bone marrow cells to recover the transplantable stem cells.

20. The method of claim 19, wherein the nuclease is present in the grinding medium at a concentration of 3 U / mL.

21. The method of claim 19, wherein the electrolyte medium comprises a sterile, pyrogen-free, isotonic solution.

22. The method of claim 19, wherein the electrolyte medium comprises a pH of 7.

4.

23. The method of claim 19, wherein the grinding medium comprises two or more of HSA, heparin, the electrolyte medium, and the growth medium.

24. The method of claim 23, wherein the grinding medium comprises three or more of HSA, heparin, the electrolyte medium, and the growth medium.

25. The method of claim 19, wherein the growth medium is Iscove's modified Dulbecco's medium (IMDM).

26. The method of claim 19, wherein the processing in step (c) comprises grinding the cadaveric bone fragments to obtain ground cadaveric bone.

27. The method of claim 26, wherein the processing in step (c) further comprises a filtering step, thereby producing a filtered product comprising the extracted bone marrow cells.

28. The method according to claim 27, wherein the filtering is performed using a 425 μm sieve or a 177 μm sieve.

29. The method of claim 28, further comprising combining the filtered product with additional grinding media having the components of step (b) to obtain a diluted bone marrow cell product.

30. The method of claim 29, further comprising the step of further filtering the diluted bone marrow cell product with one or more filters comprising a pore size ranging from 200 μm to 825 μm and / or selected from filters having a pore size of 200 μm, 500 μm or 825 μm.

31. The method of claim 30, further comprising the step of removing fat from the intermediate product.

32. The method of claim 19, wherein in step (a) the cadaveric bone fragments are present in or are divided into 1.5 cm 2 of fragments.

33. The method of claim 19, wherein the cadaveric bone or the cadaveric bone fragment is derived from a vertebral body, ileum, or a combination thereof.

34. The method of claim 33, wherein the cadaveric bone or the cadaveric bone fragment is derived from a plurality of cadaveric bones from the same donor.

35. The method of claim 19, wherein the extracted stem cells comprise hematopoietic stem cells.

36. The method of claim 19, wherein the extracted stem cells comprise mesenchymal stem cells (MSCs).

37. The method of claim 24, wherein the grinding medium comprises the electrolyte medium, 10 U / mL heparin, 2.5% HSA, and 3 U / mL nuclease.

38. The method of claim 19, wherein the transplantable stem cells are suitable for transplantation into a human subject.

39. A method for recovering vertebral bone marrow mesenchymal stromal / stem cells (vBM-MSC) and recovering vertebral adherent mesenchymal stromal / stem cells (vBA-MSC) from cadaveric bone or cadaveric bone fragments, the method comprising the steps of: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaveric bone to extract vertebral bone marrow mesenchymal stromal / stem cells (vBM-MSC); and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT shall not exceed 32 hours; The extraction of vertebral bone marrow mesenchymal stromal / stem cells (vBM-MSC) includes: (a) obtaining cadaveric bone, cadaveric bone fragments, ground cadaveric bone, or preparing ground cadaveric bone from cadaveric bone or cadaveric bone fragments; (b) contacting the ground cadaveric bone with a grinding medium under conditions sufficient to separate the bone marrow comprising vBM-MSCs from the ground cadaveric bone; (c) capturing byproducts of the ground cadaveric bone on a filter or sieve and collecting the filtrate that passes through the filter or sieve, thereby recovering the bone marrow containing the vBM-MSCs; (d) incubating the captured ground cadaveric bone byproducts in a digestion solution comprising collagenase and neutral protease, thereby releasing vBA-MSCs from the captured ground cadaveric bone; and (e) collecting the released vBA-MSCs, thereby recovering the vBA-MSCs; wherein steps (b) and (c) precede step (d); At least 2x10 8 vBA-MSCs were obtained from 100 g of captured ground cadaveric bone byproduct.

40. The method of claim 39, wherein the incubation is up to 2.5 hours.

41. The method of claim 39, wherein the amount of neutral protease is 20 U / ml.

42. The method of claim 39, wherein the collagenase is DE collagenase, the collagenase comprises recombinant collagenase isoform C1, and / or the collagenase comprises recombinant collagenase isoform C2.

43. The method of claim 39, wherein the collagenase is recombinant Clostridium histolyticum collagenase.

44. The method of claim 39, wherein the neutral protease is Paenibacillus polymyxa neutral protease.

45. The method of claim 39, wherein the volume to weight ratio of the digestion solution to the weight of the ground cadaveric bone is 5 to 1.

46. ​​The method of claim 39, wherein the ground cadaveric bone is prepared from the cadaveric bone or the cadaveric bone fragment by grinding the cadaveric bone or the cadaveric bone fragment with a bone grinder in the presence of a grinding media.

47. The method of claim 46, wherein the grinding media comprises two or more of a nuclease that cleaves both DNA and RNA, human serum albumin, heparin, and a sterile, pyrogen-free, isotonic solution.

48. The method of claim 39, wherein the digestion solution is neutralized by providing human platelet lysate.

49. The method of claim 39, further comprising the steps of filtering the ground bone after incubation, washing the ground bone after incubation, and / or collecting a filtrate after incubation, wherein the filtrate comprises vBA-MSCs.

50. The method of claim 48, wherein the filtrate comprising vBA-MSCs is combined with a filtrate comprising vBM-MSCs.

51. The method of claim 39, wherein the vBA-MSCs are phenotypically and functionally equivalent to vBM-MSCs.

52. The method of claim 39, wherein the cadaveric bone or the cadaveric bone fragment is derived from multiple cadaveric bones from the same donor.

53. The method of claim 39, further comprising the step of expanding the vBA-MSCs in the culture for 1 passage, 2 passages, 3 passages, or 4 passages, with the initial plating being considered as passage 0.

54. The method of claim 53, wherein the vBA-MSCs express CD73, CD90 and / or CD105.

55. The method of claim 54, wherein in the passaged cell population, the vBA-MSC has a positivity rate of less than 20% for hematopoietic stem and progenitor cell surface markers CD14, CD19, CD34, CD45 and / or HLA class II proteins, including human leukocyte antigen DR (HLA-DR).

56. The method of claim 39, wherein the captured ground cadaveric bone is rinsed with the grinding medium prior to the incubation, wherein the grinding medium comprises two or more of a nuclease that cleaves both DNA and RNA, human serum albumin, heparin, and a sterile, pyrogen-free, isotonic solution.

57. The method of claim 54, wherein the vBA-MSCs are cultured for at least four passages to produce more than 2 trillion vBA-MSC cells per 100 g of captured ground cadaveric bone byproduct.

58. The method of claim 48, wherein the filtrate comprising vBA-MSCs is not combined with the filtrate comprising vBM-MSCs.

59. A method for optimizing the recovery of stem cells from cadaveric bone or cadaveric bone fragments, the method comprising: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaver bone to extract stem cells; and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT shall not exceed 32 hours; The extraction of stem cells includes: (a) obtaining the cadaveric bone or the cadaveric bone fragment, wherein the cadaveric bone fragment is obtained by processing the cadaveric bone; (b) immersing the cadaveric bone or the cadaveric bone fragment in a solution comprising a bleaching agent, thereby obtaining a bleached bone product; and then (c) immersing the bleached bone product in a solution containing hydrogen peroxide to obtain a treated bone product; (d) processing the treated bone product to extract bone marrow cells; and (e) collecting the extracted bone marrow cells to recover the stem cells contained therein.

60. The method of claim 59, wherein the bleach is a 10% bleach solution.

61. The method of claim 60, wherein the bleach solution produces 5,000 ppm of free chlorine.

62. The method of claim 61, wherein the cadaveric bone or the cadaveric bone fragment is immersed in the bleach for 10 to 25 minutes.

63. The method of claim 59, wherein the hydrogen peroxide is a 3% hydrogen peroxide solution.

64. The method of claim 59, further comprising the step of transferring the bleached bone product from the container containing the bleaching solution to the container containing the hydrogen peroxide solution.

65. The method of claim 59, further comprising the step of agitating the bleached bone product in the hydrogen peroxide solution.

66. The method of claim 59, wherein step (c) of immersing the bleached bone product in a solution comprising hydrogen peroxide comprises: (i) immersing the bleached bone product in a container containing a hydrogen peroxide solution; (ii) detecting foam or foam-like matter associated with the bleached bone product; and (iii) Repeating step (b) and / or step (c) until no foam or foam-like substance is detected.

67. The method of claim 66, further comprising the step of manually removing soft tissue from the bleached bone product associated with the foam or foam-like substance in step (ii).

68. The method of claim 66, wherein an inert contrast dye is added to the solution comprising hydrogen peroxide to enhance the visibility of any foam or foam-like material associated with the bleached bone product.

69. The method of claim 59, further comprising the step of transferring the processed bone product to a container containing an electrolyte solution.

70. The method of claim 59, wherein the processing in step (d) comprises grinding the processed bone product to obtain ground cadaveric bone, wherein the grinding is performed using a bone grinder in the presence of a grinding medium.

71. The method of claim 70, wherein the grinding media comprises two or more of a nuclease that cleaves both DNA and RNA, human serum albumin (HSA), heparin, and an electrolyte mediator.

72. The method of claim 70, wherein the grinding medium facilitates extraction of the bone marrow cells from the ground cadaveric bone.

73. The method of claim 72, wherein the ground cadaveric bone is captured on a filter or sieve and the filtrate passing through the filter or sieve comprises bone marrow cells comprising stem cells.

74. The method of claim 73, further comprising incubating the captured ground cadaveric bone in a digestion solution comprising collagenase and / or neutral protease to extract vertebral bone adherent mesenchymal stromal / stem cells (vBA-MSCs) from the captured ground cadaveric bone.

75. The method of claim 59, wherein the cadaveric bone or the cadaveric bone fragment is derived from a vertebral body, an ilium, or a combination thereof.

76. The method of claim 59, wherein the cadaveric bone or the cadaveric bone fragment is derived from multiple cadaveric bones from the same donor.

77. The method of claim 59, wherein the recovered stem cells comprise hematopoietic stem cells.

78. The method of claim 59, wherein the recovered stem cells comprise mesenchymal stem cells (MSCs).

79. A method for recovering stem cells from cadaveric bone or cadaveric bone fragments, the method comprising: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaver bone to extract stem cells; and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT shall not exceed 32 hours; The extraction of stem cells includes: (a) obtaining the said cadaver bones or said cadaver bone fragments; (b) processing the bone or bone fragment to extract bone marrow cells; (c) obtaining a density-reduced solution mixture of a hydrophilic copolymer of sucrose and epichlorohydrin having a density of 1.077 g / mL and an electrolyte medium, wherein the density-reduced solution mixture provides a density gradient of 1.063-1.052 g / mL; (d) mixing the extracted bone marrow cells with the density-reduced solution mixture; (e) centrifuging the extracted bone marrow cells to separate buffy coat cells containing stem cells; and (f) harvesting the buffy coat cells to thereby recover stem cells including hematopoietic stem cells and / or mesenchymal stem cells.

80. The method of claim 79, wherein after centrifugation in step (e), the buffy coat cells are located in the density gradient at a density of 1.063 g / mL.

81. The method of claim 79, wherein the centrifugation is performed at 400 g for 30 minutes.

82. The method of claim 79, wherein the centrifugation is performed at 400 g for 5 minutes.

83. The method of claim 79, wherein the cadaveric bone or the cadaveric bone fragment has not been cryopreserved prior to processing in step (b).

84. The method of claim 79, wherein the recovered stem cells are 90% or more pure.

85. The method of claim 79, wherein the viability of the recovered stem cells is 90% or greater viability.

86. The method of claim 79, wherein the processing in step (b) comprises dividing the cadaveric bone into cadaveric bone fragments and grinding the cadaveric bone fragments to obtain ground cadaveric bone, wherein the grinding uses a bone grinder in the presence of grinding media.

87. The method of claim 86, wherein the grinding media comprises two or more of an enzyme that cleaves DNA and RNA, human serum albumin (HSA) heparin, and an electrolyte medium.

88. The method of claim 86, wherein the grinding medium facilitates extraction of the bone marrow cells from the ground cadaveric bone.

89. The method of claim 88, wherein the ground cadaveric bone is captured on a filter or sieve and the filtrate passing through the filter or sieve comprises the extracted bone marrow cells.

90. The method of claim 79, further comprising the step of washing the harvested buffy coat cells for subsequent use or further processing after step (f).

91. The method of claim 90, wherein the washing comprises contacting the cells with phosphate buffered saline (PBS) containing human serum albumin (HSA) and ethylenediaminetetraacetic acid (EDTA) or with PBS alone.

92. The method of claim 79, further comprising the step of performing a separation method to separate CD34+ cells from the recovered stem cells.

93. The method of claim 79, wherein the cadaveric bone or the cadaveric bone fragment is derived from a vertebral body, ileum, or a combination thereof.

94. The method of claim 79, wherein the cadaveric bone or the cadaveric bone fragments are derived from multiple cadaveric bones from the same donor.

95. The method of claim 79, wherein the hematopoietic stem cells are CD34+ cells.

96. The method of claim 79, wherein the density gradient provides a recovered stem cell population having a higher purity of CD34+ cells and a higher viability compared to a recovered stem cell population obtained from a conventional hydrophilic copolymer of sucrose and epichlorohydrin gradient.

97. The method of claim 79, wherein the cadaveric bone or the cadaveric bone fragment has been exposed to cryogenic temperatures between 0-10°F.

98. A method for recovering transplantable stem cells, the method comprising: (1) Record the time of a person’s death, at which point the person becomes a corpse; (2) placing the corpse or bones removed from the corpse in a cooling environment; (3) Record the time when step (2) occurs; wherein the time period from step (1) to step (2) is referred to as warm ischemia time (WIT) and step (2) is performed not less than half an hour and not more than 8 hours after step (1); (4) processing the cadaver bone to extract stem cells; and (5) Record the time when step (4) starts; wherein the time period from step (2) to step (4) is referred to as cold ischemia time (CIT) and step (4) is performed not less than 7 hours and not more than 39.5 hours after step (2); Among them, WIT and CIT are different time periods; Where the sum of WIT and CIT is forty hours or less, such that: When the WIT is greater than half an hour, the CIT does not exceed 39.5 hours, and When the WIT is 8 hours, the CIT shall not exceed 32 hours; The extraction of stem cells includes: (a) obtaining surface-sterilized cadaveric bones or cadaveric bone fragments; (b) preparing ground cadaveric bone from the surface-sterilized cadaveric bone or cadaveric bone fragments and treating the ground cadaveric bone in the presence of a grinding medium to obtain a fat emulsion containing extracted bone marrow cells; (c) removing fat from the fat emulsion containing extracted bone marrow cells; and (d) collecting the bone marrow cells extracted after step (c) to recover transplantable stem cells.

99. The method of claim 98, wherein the step of removing the fat from the fat emulsion comprises: placing the fat emulsion containing the extracted bone marrow cells in a centrifugeable container; centrifuging the container under conditions that concentrate or pellet the bone marrow cells and form a fat layer toward the top of the supernatant; and The fat layer is removed and / or the concentrated or precipitated bone marrow cells are removed.

100. The method of claim 99, wherein the cadaveric bone or the cadaveric bone fragment is surface sterilized ex vivo.

101. The method of claim 100, wherein the cadaveric bone is cut ex vivo into the cadaveric bone fragments.

102. The method of claim 99 further comprising the step of cleaning the cadaveric bone to remove soft tissue prior to step (b).

103. The method of claim 102, wherein when the cadaveric bone is a vertebral body, the vertebral body retains its anterior element during cleaning so that its cancellous bone marrow remains hypoxic during cleaning.

104. The method of claim 103, wherein the cadaveric bone or the cadaveric bone fragment is surface sterilized ex vivo.

105. The method of claim 104, wherein the cadaveric bone is cut ex vivo into the cadaveric bone fragments.

106. The method of claim 98, wherein the grinding media comprises two or more of a nuclease that cleaves both DNA and RNA, human serum albumin (HSA), heparin, an electrolyte medium, and a growth medium.

107. The method of claim 98, wherein processing in step (b) comprises grinding the cadaveric bone fragments in a bone grinder to obtain ground cadaveric bone.

108. The method of claim 107, wherein the treating in step (b) comprises tumbling and / or agitating the ground cadaveric bone in the presence of a grinding media to mechanically release bone marrow cells from the ground cadaveric bone.

109. The method of claim 108, further comprising one or two filtering steps to capture the ground cadaveric bone to produce a filtered fat emulsion containing extracted bone marrow cells.

110. The method of claim 109, wherein the one or both filtering steps comprise flushing the captured ground cadaveric bone with a grinding media and combining the flushing stream with a filtered fat emulsion containing extracted bone marrow cells.

111. The method of claim 110, wherein the one or both filtering steps comprise 200 μm and / or 500 μm filters.

112. The method of claim 110, wherein the step of removing the fat from the fat emulsion comprises: placing the filtered fat emulsion containing the extracted bone marrow cells in a centrifugeable container; centrifuging the container under conditions that concentrate or pellet the bone marrow cells and form a fat layer toward the top of the supernatant; and The fat layer is removed and / or the concentrated or precipitated bone marrow cells are removed.

113. The method of claim 110, wherein the cadaveric bone or the cadaveric bone fragments are surface sterilized ex vivo and / or wherein the cadaveric bone is cut into the cadaveric bone fragments ex vivo.

114. The method of claim 110, further comprising the step of cleaning the cadaveric bone to remove soft tissue prior to step (b).

115. The method of claim 114, wherein when the cadaveric bone is a vertebral body, the vertebral body retains its anterior element during cleaning so that its cancellous marrow remains hypoxic during cleaning.

116. The method of claim 110, wherein the grinding media comprises two or more of a nuclease that cleaves both DNA and RNA, human serum albumin (HSA), heparin, an electrolyte medium, and a growth medium.

117. The method of claim 110, further comprising the step of cryopreserving the bone marrow cells in a cryopreservation solution comprising one or more of dimethyl sulfoxide (DMSO), 1,2-propylene glycol, ethylene glycol, formamide, ethylene glycol or butane 2,3-diol, hydroxyethyl starch (HES), dextran, sucrose, trehalose, lactose, raffinose, ribitol, mannitol, and polyvinylpyrrolidone (PVP).

118. The method of claim 117, wherein the cryopreservation solution further comprises an oxidase.

Images