Preparation method and application of co-assembled nano preparation for targeting glioblastoma

Abstract

The invention belongs to the field of new auxiliary materials and new dosage forms of pharmaceutical preparations, and relates to a preparation method and application of a co-assembled nano preparation for targeting glioblastoma. The preparation method comprises the following steps: firstly, synthesizing a nitroimidazole dimer with tumor hypoxia-oxidation-reduction triple response, and assembling the nitroimidazole dimer with AOH1996 and HRO761 to prepare the nano preparation. The molar ratio of the AOH1996 to the HRO761 to the nitroimidazole dimer is 1 to 2 to (0.5 to 2). The key scaffold protein PCNA for DNA repair is blocked by the PCNA inhibitor AOH1996, and recruitment and anchoring functions of the key scaffold protein PCNA on repair factors are destroyed, so that the glioblastoma is induced to enter a mismatch repair defect state. On the basis, a WRN inhibitor HRO761 is further used for promoting large-scale breakage of chromosomes and finally inducing cell death. The nitroimidazole dimer and the nitroimidazole dimer are jointly assembled into the nano preparation, and accurate drug release is realized in response to a hypoxia state and oxidation and reduction signals overexpressed in a tumor microenvironment, so that the heterogeneity of the tumor microenvironment is overcome.

Description

Technical Field

[0001] The invention belongs to the field of new excipients and new dosage forms of pharmaceutical preparations, and relates to a preparation method and application of a co-assembled nano preparation targeting glioblastoma. Background Art

[0002] Malignant tumors seriously threaten human health. Among them, glioblastoma (GBM) is one of the most invasive and challenging malignant tumors of the central nervous system, with an average survival of only 12.1-14.6 months. At present, the standard clinical treatment of GBM mainly includes surgical resection combined with radiotherapy or chemotherapy. However, due to the diffuse infiltrative growth of GBM, the lack of capsule and clear boundaries, it is difficult to completely remove the lesions by surgery. In addition, excessive doses of radiotherapy and chemotherapy may damage normal nerve tissue, leading to severe neurological dysfunction and significantly reducing the quality of life of patients. More importantly, GBM has a high degree of tumor heterogeneity, and different subtypes of tumor cells have significant differences in sensitivity to radiotherapy and chemotherapy, which can easily lead to the selective proliferation of drug-resistant cells and ultimately promote tumor progression. Therefore, how to improve the therapeutic specificity of GBM and overcome its tumor heterogeneity is an important issue that needs to be urgently addressed in the current field of GBM drug development and treatment.

[0003] Synthetic lethality is an emerging cancer treatment strategy that can precisely target tumor-specific genetic weaknesses, selectively kill cancer cells, and reduce damage to normal tissues. Its mechanism of action is based on a specific gene interaction relationship, that is, when a single genetic defect exists, the cell can still survive, but when both complementary genes are damaged, the cell will irreversibly enter the death pathway. Warren helicase (WRN) is an important synthetic lethal target reported in recent years. Especially in tumor cells with mismatch repair defects, microsatellite instability is often accompanied, leading to large-scale amplification of microsatellite repeat sequences and replication fork stagnation. When WRN helicase is inhibited, microsatellite repeat sequences are cleaved by endogenous nucleases, resulting in large-scale chromosome breakage and ultimately inducing cell death. Therefore, WRN inhibitors are considered to have potential therapeutic value for microsatellite instability GBM.

[0004] HRO761 is a small molecule WRN inhibitor currently in Phase I clinical trials for the treatment of solid tumors with high microsatellite instability or mismatch repair deficiency.

[0005]

[0006] However, GBM is not a typical microsatellite unstable tumor, and most of its cell populations still maintain microsatellite stability, which limits the therapeutic effect of WRN inhibitors in microsatellite stable GBM.

[0007] During the mismatch repair process, Proliferating Cell Nuclear Antigen (PCNA) is a key molecular scaffold responsible for recruiting and anchoring various DNA repair factors. By using PCNA inhibitors, mismatch repair defects can be artificially induced in microsatellite-stable GBM cells, making them sensitive to WRN inhibitors. Therefore, initially inducing artificial mismatch repair defects through PCNA inhibition holds promise for overcoming tumor heterogeneity in GBM, increasing the sensitivity of GBM to WRN inhibitors, and realizing the potential for broad and precise killing of different subtypes of GBM.

[0008] AOH1996 is a small molecule PCNA inhibitor that first achieved selective inhibition of the traditionally considered "undruggable" target PCNA. It is currently undergoing a Phase I clinical trial. AOH1996 is an orally active ligand of the replisome isoform component PCNA, stabilizing the interaction between PCNA and RNA polymerase II, leading to proteasome-dependent degradation of rpb1 and lethal DNA damage. AOH1996 also interferes with the interaction between PCNA and its binding proteins, resulting in DNA replication stress and inducing apoptosis. AOH1996 synergistically acts with DNA-damaging agents to inhibit tumor cells.

[0009]

[0010] In addition, the main difficulty in GBM drug treatment lies in the restriction of the blood-brain barrier, which makes it difficult for most chemotherapy drugs to reach the brain tissue and exert their effects. At the same time, the non-specific distribution of chemotherapy drugs in the brain tissue may cause severe neurotoxicity, leading to irreversible damage. Achieving efficient brain-targeted drug delivery and controlling the precise release of drugs in tumor cells is a key technical bottleneck in the current field of brain tumor treatment. Therefore, constructing an intelligent responsive brain-targeted drug delivery system that can actively recognize and cross the blood-brain barrier and, after entering the brain parenchyma, can precisely release to the tumor site, reducing damage to normal nerve tissue, is of great significance for improving the treatment effect of GBM. Summary of the Invention

[0011] In view of the limitations of the prior art, the present invention innovatively proposes a synthetic lethality drug synergy strategy and constructs a drug delivery system with brain targeting ability for it. The specific implementation means are as follows: Select the WRN inhibitor HRO761 and the PCNA inhibitor AOH1996 as the synthetic lethality drug combination. Induce artificial mismatch repair defects through the PCNA inhibitor to increase the sensitivity to the WRN inhibitor, overcome the heterogeneity of GBM subtypes, and induce DNA damage in GBM cells. Secondly, synthesize a nitroimidazole dimer with triple sensitivity to hypoxia-oxidation-reduction in the tumor microenvironment and co-assemble it with HRO761 and AOH1996 to construct a nanoformulation with multiple responses to the tumor microenvironment to overcome the heterogeneity of the tumor microenvironment and achieve precise response release of the drug combination. Finally, perform brain targeting modification on the co-assembled nanoformulation to enable it to cross the blood-brain barrier and be released at a fixed point in GBM cells in the brain to exert an anti-tumor effect, and remain inert in other tissues, organs, and normal brain tissues. The present invention provides a brand-new treatment idea and technical solution for overcoming the heterogeneity of glioblastoma multiforme.

[0012] To achieve the above object, the present invention adopts the following technical solutions:

[0013] The present invention provides the nitroimidazole dimer or its pharmaceutically acceptable salt as shown below:

[0014]

[0015] Wherein, X is a tumor microenvironment-sensitive chemical bond.

[0016] Further, in the above structural formula, the tumor microenvironment-sensitive chemical bond is a pH-sensitive bond, a redox environment-sensitive bond, an enzyme-sensitive bond, or a hypoxia-sensitive bond. The pH-sensitive bond includes at least one of a hydrazone bond, an imine bond, an acetal bond, and a β-amino ester bond. The redox environment-sensitive bond includes at least one of a monothioether bond, a disulfide bond, a trisulfide bond, a monoselenide bond, a diselenide bond, and a borate ester bond. The enzyme-sensitive bond includes at least one of a metalloproteinase-sensitive bond, an esterase-sensitive bond, a phosphatase-sensitive bond, a transglutaminase-sensitive bond, and a thioredoxin reductase-sensitive bond. The hypoxia-sensitive bond includes at least one of an azo bond and a nitroaromatic bond.

[0017] Further, X is a redox environment-sensitive bond.

[0018] Taking X as a disulfide bond as an example, the synthesis method of the above nitroimidazole dimer specifically includes the following steps:

[0019] Add dithiodibutyric acid, EDCI, and DMAP into a reaction vessel, dissolve them using an organic solvent, activate under an ice bath, and then add 5-nitro-1H-imidazole-1-ethanol dissolved in the organic solvent. After reacting at room temperature overnight, separate and purify the product to obtain the final product, nitroimidazole dimer.

[0020] The organic solvent described is: DMF.

[0021] The purity of the nitroimidazole dimer prepared in the present invention is above 99%.

[0022] The present invention provides a pharmaceutical composition comprising the nitroimidazole dimer or a pharmaceutically acceptable salt thereof, AOH1996, and HRO761 as described above.

[0023] In the pharmaceutical composition described, the molar ratio of AOH1996 to HRO761 is 5:1 - 1:2, preferably 1:1 - 1:2.

[0024] Furthermore, the molar ratio of AOH1996, HRO761 to the nitroimidazole dimer is 1:2:(0.5 - 2), preferably 1:2:(1 - 1.5).

[0025] The nitroimidazole dimer or a pharmaceutically acceptable salt thereof in the present invention can form a co-assembled nano-formulation (AHN7) of the nitroimidazole dimer, AOH1996, and HRO761 driven by intermolecular non-covalent forces with AOH1996 and HRO761.

[0026] The molar ratio of AOH1996 to HRO761 described is 5:1 - 1:2, preferably 1:1 - 1:2.

[0027] The molar ratio of AOH1996, HRO761 to the nitroimidazole dimer is 1:2:(0.5 - 2), preferably 1:2:(1 - 1.5).

[0028] The co-assembled nano-formulation described, in addition to containing HRO761, AOH1996, the nitroimidazole dimer or a pharmaceutically acceptable salt thereof, further contains a PEG modifier / active targeting modifier.

[0029] The molar ratio of AOH1996 to HRO761 described is 5:1 - 1:2, preferably 1:1 - 1:2.

[0030] The molar ratio of AOH1996, HRO761 to the nitroimidazole dimer is 1:2:(0.5 - 2), preferably 1:2:(1 - 1.5). The mass ratio of the total mass of the nitroimidazole dimer or a pharmaceutically acceptable salt thereof, AOH1996, and HRO761 to the mass of the PEG modifier / active targeting modifier is: 1:(0.1 - 1).

[0031] The PEG modifier is an amphiphilic polymer or targeting group such as DSPE-PEG, TPGS, PEG-PLGA or PEG-PCL, and the active targeting modifier is a brain targeting peptide / antibody conjugate, a ligand conjugate, a cell penetrating peptide conjugate, a receptor targeting conjugate, a sugar residue, a hormone, etc. that can be targeted to a specific tissue, selected from DSPE-PEG-SHp, DSPE-PEG-Angiopep, DSPE-PEG-T7, DSPE-PEG-RVG29, DSPE-PEG-cRGD, DSPE-PEG-Lactoferrin, DSPE-PEG-NGR, DSPE-PEG-TAT, DSPE-PEG-iRGD, DSPE-PEG-Mannose, DSPE-PEG-OTC, DSPE-PEG-GE11, DSPE-PEG-CREKA, DSPE-PEG-TH, DSPE-PEG-R8, DSPE-PEG-APOE, preferably a phospholipid-polyethylene glycol-T7 peptide (DSPE-PEG-T7).

[0032] The present invention also provides a method for preparing the nitroimidazole dimer, AOH1996 and HRO761 co-assembled nanoformulation, comprising the following steps:

[0033] Appropriate amounts of PEG modifier / active targeting modifier, HRO761, AOH1996 and nitroimidazole dimer were dissolved in an organic solvent, stirred and mixed thoroughly, and then the mixed solution was slowly added to the aqueous phase. Driven by intermolecular non-covalent forces, the system spontaneously formed uniformly dispersed nanoparticles. Subsequently, the organic solvent was completely removed by reduced pressure rotary evaporation, and finally a pure PEG-modified / actively targeted nitroimidazole dimer, AOH1996 and HRO761 co-assembled nanoformulation was obtained.

[0034] The organic solvent is an organic solvent miscible with water such as ethanol, acetone, tetrahydrofuran, dimethyl sulfoxide, N,N-dimethylformamide, etc., preferably acetone.

[0035] The volume ratio of the organic solvent to water is (0.1-1):1, preferably 0.2-0.5:1.

[0036] Beneficial effects of the present invention:

[0037] (1) The present invention innovatively proposes a synthetic lethality strategy based on the combined action of a WRN inhibitor and a PCNA inhibitor. By blocking the key scaffold protein PCNA for DNA repair with the PCNA inhibitor AOH1996, its recruitment and anchoring functions for repair factors are disrupted, thereby inducing glioblastoma into a mismatch repair defect state. On this basis, the WRN inhibitor HRO761 is further used to cause large-scale chromosome breaks and ultimately induce cell death. This strategy can not only accurately and thoroughly kill glioblastoma, but also effectively overcome the treatment bottleneck brought about by the high heterogeneity of glioblastoma, providing new ideas for the treatment of glioblastoma.

[0038] (2) The present invention designed and synthesized a nitroimidazole dimer with triple responses to tumor hypoxia-oxidation-reduction, and co-assembled it with AOH1996 and HRO761 to prepare a nanoformulation. The prepared nanoformulation has a particle size less than 300 nm, with an optimal range of 50 - 200 nm, and a polydispersity index (PDI) below 0.2, showing good uniformity, stability and reproducibility. Moreover, the co-assembled nanoformulation can respond to the overexpressed hypoxia, oxidation and reduction signals in the tumor microenvironment to achieve precise drug release, thereby overcoming the heterogeneity of the tumor microenvironment.

[0039] (3) The present invention uses T7 peptide to modify the surface of the co-assembled nanoformulation, endowing it with the ability to cross the blood-brain barrier. Relying on the highly expressed transferrin receptors on glioblastoma and the blood-brain barrier, through receptor-mediated transcytosis, the drug can efficiently cross the blood-brain barrier and be precisely delivered to the glioblastoma lesion. This strategy can not only maximize the anti-tumor activity, but also reduce the systemic off-target toxicity, providing a safe and efficient precise treatment plan for glioblastoma. Description of the Drawings

[0040] Figure 1 It is the mass spectrometry diagram of the nitroimidazole dimer in Example 1 of the present invention.

[0041] Figure 2 It is of the nitroimidazole dimer in Example 1 of the present invention 1 HNMR spectrum.

[0042] Figure 3 It is the colloidal stability diagram of the co-assembled nanoformulation formulation in Example 3 of the present invention.

[0043] Figure 4 It is the long-term storage stability diagram of the co-assembled nanoformulation formulation in Example 3 of the present invention.

[0044] Figure 5 It is the particle size distribution diagram of AHN7 in Example 3 of the present invention.

[0045] Figure 6Transmission electron micrograph of AHN7 in Example 3 of the present invention.

[0046] Figure 7 In vitro drug release profile of AHN7 in Example 5 of the present invention.

[0047] Figure 8 γH2AX immunofluorescence image in Example 6 of the present invention.

[0048] Figure 9 Quantification graph of γH2AX immunofluorescence in Example 6 of the present invention.

[0049] n.s. No significant difference, P<0.05 is considered to have a significant difference (one-way ANOVA).

[0050] Figure 10 Comet assay images of each formulation in Example 6 of the present invention.

[0051] Figure 11 Quantification graph of tail moment in comet assay of each formulation in Example 6 of the present invention.

[0052] n.s. No significant difference, P<0.05 is considered to have a significant difference (one-way ANOVA).

[0053] Figure 12 Quantification graph of DNA content in the tail of comet assay of each formulation in Example 6 of the present invention. n.s. No significant difference, P<0.05 is considered to have a significant difference (one-way ANOVA).

[0054] Figure 13 In vivo tumor fluorescence signal images of each formulation in Example 7 of the present invention.

[0055] Figure 14 Quantification graph of in vivo tumor fluorescence signal of each formulation in Example 7 of the present invention.

[0056] Figure 15 Ex vivo fluorescence signal images of major tissues and brain of each formulation in Example 7 of the present invention.

[0057] Figure 16 Quantification graph of ex vivo fluorescence signal of major tissues and brain of each formulation in Example 7 of the present invention.

[0058] Figure 17 Ex vivo fluorescence signal images of brain of each formulation in Example 7 of the present invention.

[0059] n.s. No significant difference, P<0.05 is considered to have a significant difference (one-way ANOVA).

[0060] Figure 18Bioluminescence image of temozolomide-resistant glioblastoma in mice in Example 8 of the present invention.

[0061] Figure 19 Quantitative bioluminescence image of temozolomide-resistant glioblastoma in mice in Example 8 of the present invention. n.s. indicates no significant difference, and P < 0.05 is considered to have a significant difference (one-way ANOVA).

[0062] Figure 20 Curve graph of the body weight change of mice in Example 8 of the present invention.

[0063] Figure 21 Survival curve graph of mice in Example 8 of the present invention.

[0064] Figure 22 Graph of the median survival time of mice in Example 8 of the present invention.

[0065] Figure 23 H&E section image of the mouse brain tissue in Example 8 of the present invention. Detailed implementation manners

[0066] The present invention will be further described in detail below in conjunction with the embodiments.

[0067] Example 1: Synthesis of nitroimidazole dimer

[0068] Under nitrogen protection, dithiodibutyric acid (5.0 mmol), EDCI (10.0 mmol), and DMAP (1 mmol) were successively added to a three-necked reaction flask, and dissolved with DMF as a solvent and stirred for 30 min to form a homogeneous system. Subsequently, the reaction solution was placed in an ice-water bath and stirred for activation for 1 hour. After the activation was completed, 5-nitro-1H-imidazole-1-ethanol (5 mmol) pre-dissolved with DMF was added dropwise, and the ice-bath condition was maintained and stirred continuously to avoid excessive local concentration. After the addition of the materials was completed, the ice-bath device was removed and the reaction system was heated to room temperature, and the reaction was carried out overnight by magnetic stirring. After the reaction was completed, ice water was added to quench the reaction, and the precipitate was removed by filtration after vigorous stirring. The filtrate was washed with saturated sodium bicarbonate solution and saturated sodium chloride solution respectively, concentrated to a viscous state by a rotary evaporator, and finally the product was purified with a preparation solution, and the detection wavelength was 277 nm to obtain the final product nitroimidazole dimer.

[0069] The structure of the nitroimidazole dimer in Example 1 was determined by mass spectrometry and nuclear magnetic resonance hydrogen spectroscopy, and the results are as Figure 1 、 Figure 2 shown. The results of spectral analysis are as follows:

[0070] 11H NMR (400 MHz, Chloroform-d) δ 8.02 (d, J = 1.1 Hz, 2H, f), 7.60 (d, J = 1.1 Hz, 2H, g), 4.66 (dd, J = 5.7, 4.4 Hz, 4H, d), 4.45 (dd, J = 5.7, 4.4 Hz, 4H, e), 2.66 (t, J = 7.0 Hz, 4H, a), 2.42 (t, J = 7.3 Hz, 4H, c), 1.97 (p, J = 7.1 Hz, 4H, b). MS (ESI) m / z: [M+H] + Calcd for C 18 H 24 N6O8S2, 517.1131; found, 517.1176 [M+H] + and 539.0998 [M+Na] + .

[0071] Example 2: Screening of Synergistic Ratio

[0072] Using the human glioblastoma U87 cell model, the optimal synergistic ratio of AOH1996 and HRO761 was screened, and the cytotoxicity of their combined use was evaluated. First, mixed solutions of AOH1996 and HRO761 with different molar ratios (5:1, 4:1, 3:1, 2:1, 1:1, 1:2, 1:3, 1:4, 1:5) were prepared, and the toxicity changes of each group of drugs against U87 cells were determined by the MTT method. During the experiment, U87 cells were seeded into 96-well plates at a density of 2000 cells per well. After overnight adherent culture, the original medium was discarded, and the medium containing different concentrations of drugs was added, and incubation was continued for 48 hours. Subsequently, 1 mg / mL MTT solution was added and incubated for 3.5 - 5 hours. After removing the supernatant, DMSO was added to dissolve the formazan crystals, and the absorbance was measured at a wavelength of 490 nm to calculate the half-maximal inhibitory concentration (IC 50 ). In addition, the IC 50 of AOH1996 or HRO761 used alone was used as a control, and the combination index (CI 50 ) was calculated. The calculation formula is as follows:

[0073]

[0074] where D A and D B are the concentrations of AOH1996 and HRO761 in the mixed solution, respectively, and IC 50，A and IC 50，B are the IC 50 values of AOH1996 and HRO761, respectively. CI 50 < 1 indicates a synergistic effect, and CI 50= 1 represents an additive effect, CI 50 > 1 indicates an antagonistic effect.

[0075] The IC at each ratio 50 and CI 50 values are shown in Table 1. The results show that when the molar ratio of AOH1996 to HRO761 is in the range of 5:1 to 1:2, both drugs exhibit a synergistic effect, and the CI 50 value of the 1:2 ratio is the lowest, indicating that this formulation can achieve the best synergistic anti-tumor effect. Therefore, AOH1996:HRO761 = 1:2 was determined as the optimal synergistic ratio and used in subsequent experiments.

[0076] Table 1. IC of AOH1996 and HRO761 at different molar ratios 50 and CI 50

[0077]

[0078] Example 3. Construction of Brain-Targeted Co-Assembled Nanopreparations

[0079] According to the synergistic ratio screening in Example 2, the optimal molar ratio of AOH1996 to HRO761 was determined to be 1:2. Based on this formulation, a co-assembled nanopreparation of AOH1996, HRO761, and nitroimidazole dimer was prepared using microfluidic technology. The specific operation is as follows: The accurately weighed AOH1996, HRO761, and nitroimidazole dimer were dissolved in a tetrahydrofuran / acetone mixed solvent at different molar ratios (1:2:0, 1:2:0.5, 1:2:1, 1:2:1.5, 1:2:2, 1:2:3, 1:2:5) together with the brain-targeting modification material DSPE-PEG-T7 (mass ratio 20%, w / w), and an ultrasonic oscillation was performed to form a uniform drug-loaded solution. Subsequently, two-phase laminar flow mixing was achieved through a coaxial microfluidic chip at an accurate flow rate (organic phase 0.5 mL / min, aqueous phase 2 mL / min), and co-assembly was driven by solvent diffusion under intermolecular non-covalent interactions at 30 °C to form a co-assembled nanopreparation. Free drugs were removed by tangential flow filtration, and organic solvents were removed using a rotary evaporator to finally obtain a pure brain-targeted co-assembled nanopreparation.

[0080] Characterization of brain-targeted co-assembled nanoparticles: The prepared co-assembled nanoparticles were diluted 10-fold with ultrapure water, and then the hydrodynamic diameter, polydispersity index (PDI), and Zeta potential were measured using a Malvern particle size analyzer. After the supernatant was filtered through a 0.22-μm microporous membrane, quantitative analysis was performed using a high-performance liquid chromatography system equipped with a C18 reverse-phase chromatographic column (4.6×250 mm, 5 μm). The mobile phase was a methanol-water gradient system, and the detection wavelengths for AOH1996, HRO761, and nitroimidazole dimer were set at 300 nm, 320 nm, and 277 nm, respectively.

[0081] The calculation formulas for encapsulation efficiency (EE%) and drug loading (DL%) are as follows:

[0082] EE% = (drug content in the formulation / total drug amount input) × 100%;

[0083] DL% = theoretical drug loading × encapsulation efficiency × 100%.

[0084] Colloidal stability of brain-targeted co-assembled nanoparticles: The prepared co-assembled nanoparticles were mixed with phosphate buffer solution (PBS, pH 7.4) containing 10% fetal bovine serum (FBS) at a volume ratio of 1:10 and placed in a constant temperature shaker at 37°C (oscillation frequency 200 rpm) to simulate the in vivo microenvironment. Samples were taken at preset time points (0, 2, 4, 8, 12, 24, 48 hours), and the hydrodynamic diameter and PDI were measured using a Malvern particle size analyzer.

[0085] Long-term storage stability of brain-targeted co-assembled nanoparticles: The prepared co-assembled nanoparticles were stored in a closed container in the dark at 4°C in a refrigeration device, and samples were taken on the 1st, 5th, 10th, 20th, 30th, 40th, and 60th days. After the samples were diluted 10-fold with ultrapure water, the hydrodynamic diameter and PDI were measured using a Malvern particle size analyzer.

[0086] Morphological observation of brain-targeted co-assembled nanoparticles: After the prepared co-assembled nanoparticles were diluted 10-fold, 5 μL of the sample was pipetted and dropped onto the surface of a carbon-coated copper grid, and then naturally dried in a clean environment for 12 hours. Subsequently, negative staining was performed using a 0.25% phosphotungstic acid solution. After standing in the dark for 5 minutes, the excess staining solution was removed with filter paper. After the sample was completely dried, the morphological characteristics of the nanoparticles were observed under a transmission electron microscope, and high-resolution images were collected.

[0087] The results of the particle size, PDI, encapsulation efficiency, and drug loading of brain-targeted co-assembled nanoparticles under different molar ratios are shown in Tables 2, 3, and 4.

[0088] Table 2. Characterization of co-assembled nanoparticles with different molar ratios

[0089]

[0090]

[0091] Table 3. Encapsulation efficiency of drugs in co-assembled nanopharmaceuticals with different molar ratios

[0092]

[0093] Table 4. Drug loading and corrected molar ratio of drugs in co-assembled nanopharmaceuticals with different molar ratios

[0094]

[0095] a) Total drug loading is the sum of the drug loadings of AOH1996 and HRO761. b) Molar ratio of AOH1996 to HRO761 corrected according to the encapsulation efficiency

[0096] Table 2 shows that when the molar ratio of AOH1996:HRO761:nitroimidazole dimer is 1:2:0 - 2, the particle size of the prepared nanopharmaceuticals is less than 200 nm and the PDI < 0.2.

[0097] Table 3 shows that when the molar ratio of AOH1996:HRO761:nitroimidazole dimer is 1:2:0, although the co-assembled nanopharmaceuticals can be successfully prepared, their encapsulation efficiency is only between 65 - 85%, resulting in the actual molar ratio of the two drugs deviating from the ideal value to reach 1:2.55, making it difficult to achieve the best synergistic effect. When the nitroimidazole dimer is introduced into the formulation, the encapsulation efficiency of AOH1996 and HRO761 is significantly improved. At the ratios of 1:2:1 and 1:2:1.5, the encapsulation efficiency of both exceeds 90%, and the actual molar ratio is close to the ideal 1:2. However, when the content of the nitroimidazole dimer is further increased to 1:2:2, the encapsulation efficiency begins to decline, while the formulations of 1:2:3 and 1:2:5 result in drug precipitation and cannot form stable nanopharmaceuticals. Therefore, in the range of 1:2:1 - 1:2:1.5, not only is the encapsulation efficiency of the drug high, but also stable nanopharmaceuticals can be formed.

[0098] Further investigate the colloidal stability of the co-assembled nanopharmaceuticals with the ratios of 1:2:1 and 1:2:1.5, and compare them with the 1:2:0 group without the nitroimidazole dimer. The results are as Figure 3 and Figure 4As shown, 1:2:0 remained stable in the initial stage of incubation, but the particle size increased after 8 hours, and precipitation occurred after 12 hours. In contrast, 1:2:1 was stable within 48 hours with no significant change in particle size, while the particle size of 1:2:1.5 increased after 12 hours. The long-term stability experiment further confirmed that the particle size of 1:2:0 increased after storing for 30 days, while 1:2:1 and 1:2:1.5 remained stable within 60 days, indicating that the addition of nitroimidazole dimer not only improved the encapsulation efficiency but also enhanced the colloidal stability of the preparation. However, when the content of nitroimidazole dimer was too high, it would affect the co-assembly process and reduce the preparation stability.

[0099] Comprehensively analyzed, the co-assembled nano-preparations of 1:2:1 - 1:2:1.5 had an encapsulation efficiency of more than 90% for both AOH1996 and HRO761. The actual molar ratio was highly close to the ideal ratio. At the same time, they had a high drug loading of more than 50% and excellent colloidal stability and long-term stability, which could be used for in vivo administration and application after long-term storage. Therefore, the final optimal formulation ratio was determined to be 1:2:1 - 1:1:1.5.

[0100] A co-assembled nano-preparation (AHN7) was constructed with AOH1996, HRO761, and nitroimidazole dimer at a molar ratio of 1:2:1. As Figure 5 and Figure 6 shown, the particle size of the constructed AHN7 was about 105 nm, with a uniform particle size distribution (PDI < 0.15), and the surface potential was about -30 mV, ensuring the biocompatibility of intravenous injection. In addition, the observation results of transmission electron microscopy showed that AHN7 had a uniform spherical structure and good morphology.

[0101] Meanwhile, a co-assembled nano-preparation (AHN7) was constructed with AOH1996, HRO761, and nitroimidazole dimer at a molar ratio of 1:2:1.5. The particle size of the constructed nano-preparation was about 120 nm, PDI < 0.2, and the surface potential was about -30 mV, also ensuring the biocompatibility of injection.

[0102] The following is a co-assembled nano-preparation (AHN7) constructed with AOH1996, HRO761, and nitroimidazole dimer at a molar ratio of 1:2:1 for subsequent research, and the preparation method is the same as that in Example 3.

[0103] Example 4: In vitro release of brain-targeted co-assembled nano-preparation

[0104] Using PBS containing 20% (v / v) ethanol as the release medium, sodium dithionite (Na2S2O4, final concentrations of 0.1 and 0.5 mg / mL) or hydrogen peroxide (H2O2) and glutathione (GSH, final concentrations of 1 and 10 mM) were added respectively to simulate the highly expressed hypoxic, oxidative and reducing substances in the tumor microenvironment, and the blank medium was used as a control. AHN7 was loaded into a pre-treated dialysis bag, sealed and vertically immersed in a centrifuge tube containing 50 mL of the corresponding medium, and placed in a 37 °C constant temperature shaking incubator for dynamic release. Sampling was carried out at time points of 1, 2, 4, 8, and 12 hours. After each sampling of 500 μL, an equal volume of pre-warmed fresh medium of the same kind was immediately added to maintain the sink conditions. Quantitative analysis was performed using a high-performance liquid chromatography system equipped with a C18 reversed-phase chromatographic column (4.6×250 mm, 5 μm). The mobile phase was a methanol-water gradient system, and the detection wavelengths of AOH1996 and HRO761 were set at 300 nm and 320 nm respectively.

[0105] Traditional co-assembled nanomedicines are difficult to achieve precise tumor microenvironment-responsive drug release due to the lack of chemical structures that can respond to the tumor microenvironment. Based on this challenge, this study innovatively designed and synthesized a nitroimidazole dimer for co-assembling with AOH1996 and HRO761 to form a tumor microenvironment-responsive co-assembled nanomedicine. The nitroimidazole dimer can undergo dynamic structural transformation under different microenvironment conditions, thereby regulating the stability of the nanoparticles and promoting drug release. Specifically, in a hypoxic environment, the hydrophobic nitro group of the nitroimidazole group can be reduced to a hydrophilic amino group, inducing a hydrophobic-hydrophilic conversion, accelerating the disassembly of the nanostructure and promoting drug release; in an oxidative environment, the disulfide bond in the dimer can be oxidized to a sulfone or sulfoxide, increasing hydrophilicity, and also accelerating drug release by disrupting the nanostructure stability; while in a reducing environment, the disulfide bond can undergo a sulfur exchange reaction with endogenous reducing agents (such as GSH), resulting in the cleavage of the disulfide bond, also promoting the disassembly of the nanoparticles and drug release. Although single tumor microenvironment response can trigger drug release appropriately, tumor heterogeneity leads to significant differences in the expression levels of single stimulants in different tumor types or different development stages of the same tumor. Therefore, nanodrugs with multiple response characteristics have more advantages in adapting to tumor heterogeneity.

[0106] The experimental results are as Figure 7As shown, AHN7 exhibits good time- and concentration-dependent drug release behavior under different stimulation conditions, and AOH1996 and HRO761 show a synchronous release trend under various microenvironmental stimulations. Under non-stimulated conditions, only about 25% of the two drugs are released within 24 hours; in a hypoxic environment (Na2S2O4 0.1 mg / mL), the release amount increases to 50 - 55% within 24 hours, and when the Na2S2O4 concentration is increased to 0.5 mg / mL, the release amount further increases to about 75%; in an oxidative environment, almost complete release can be achieved in only 8 hours; while in a reducing environment (GSH 1 mM), the drug release amount reaches about 65% within 24 hours, and when the GSH concentration is increased to 10 mM, almost all the drugs are released within 24 hours. From the above results, through the co-assembly of nitroimidazole dimers, AHN7 has the ability of triple-responsive drug release to hypoxia-oxidation-reduction, and can achieve precise drug release in the heterogeneous tumor microenvironment, while remaining inert in normal tissues and brain parenchyma, thereby effectively reducing systemic toxic and side effects.

[0107] Example 5: Evaluation of the cytotoxicity and selectivity of brain-targeted co-assembled nanoformulations

[0108] The MTT method was used to evaluate the cytotoxicity of each formulation, and the specific operation process was as follows: Human glioblastoma U87, U251, T89G, LN229 cells and mouse hippocampal neuron HT22 cells were used as models. During the experiment, each cell was seeded into a 96-well plate at a density of 2000 cells per well. After adherent culture overnight, the original culture medium was discarded, and the culture medium containing different concentrations of temozolomide, AOH1996, HRO761, a mixed solution of AOH1996 and HRO761 (Mixed sol, molar ratio: AOH1996:HRO761 = 1:2, the same as the ratio of the two drugs in AHN7) and AHN7 was added, and incubation was continued for 48 hours. Subsequently, 1 mg / mL MTT solution was added and incubated for 3.5 - 5 hours. After removing the supernatant, DMSO was added to dissolve the formazan crystals, and the absorbance was measured at a wavelength of 490 nm to calculate the IC 50 . According to the obtained IC 50 The selection index (SI) was calculated, and the formula was as follows: Selection index = IC 50 of the formulation in tumor cells / IC 50 of the formulation in HT22 cells.

[0109] Table 5. IC 50 a)

[0110]

[0111] a) IC 50Unit: μM

[0112] The IC of each formulation on glioblastoma cell lines 50 As shown in Table 5, there were significant differences in the sensitivity of different genotype tumor cells to the first-line chemotherapy drug temozolomide. The IC of the single-drug administration groups of temozolomide, AOH1996, and HRO761 50 showed a fluctuation range of 2.3 - 2.5 times, indicating that tumor heterogeneity affected the efficacy of single-drug treatment. It is worth noting that the synthetic lethal drug combinations exhibited more stable anti-tumor activity. The IC of Mixed sol and AHN7 on each cell line 50 value fluctuation range narrowed to 1.6 - 1.7 times, and both had strong cytotoxicity. Among them, the cytotoxicity of AHN7 was weaker than that of Mixed sol, mainly because the drug had to be released from the nanoformulation to exert cytotoxicity, with a certain delayed release effect.

[0113] Table 6. Selectivity index of each formulation on different cells

[0114]

[0115] HT22 hippocampal neuronal cells were used to simulate normal brain tissue for drug off-target effect detection. Among the single-drug groups, AOH1996 showed the strongest neurotoxicity, followed by Mixed sol, while the neurotoxicity of AHN7 was significantly reduced. Further calculation of the selectivity index of each formulation, which is used to measure the relative toxicity of drugs to tumor cells and normal cells. The larger the value, the stronger the toxicity of the drug to tumor cells and the lower the toxicity to normal cells. The results are shown in Table 6. The selectivity index of the AOH1996 or HRO761 group was mostly lower than 5 (only the selectivity index of AOH1996 exceeded 5 on U87 cells), indicating obvious non-specific toxicity. In contrast, based on the tumor microenvironment-responsive drug release characteristics, AHN7 increased the selectivity index value to the range of 20 - 40, indicating that while maintaining anti-glioma activity, it significantly reduced neurotoxicity, which originated from its tumor-specific drug release behavior and demonstrated a unique therapeutic advantage.

[0116] Example 6: DNA damage ability of brain-targeted co-assembled nanoformulations

[0117] γH2AX immunofluorescence detection: Seed U87 cells at 1×10 5Cells were seeded at a density of [number of cells] per well in a 24-well plate and incubated at 37 °C in a 5% CO₂ atmosphere for 12 hours to allow complete attachment. After aspirating the original medium, the experimental group was treated with fresh medium containing temozolomide (dose: 250 μM), AOH1996, HRO761, Mixed sol, and AHN7 (total drug dose: 250 nM), and untreated cells were set as the blank control and cultured for an additional 24 hours. After treatment, the cells were fixed with 4% paraformaldehyde at room temperature for 30 minutes, washed three times with PBS, permeabilized with 0.2% Triton X-100 for 15 minutes, and then blocked with 5% BSA blocking solution at room temperature for 1 hour. Mouse anti-γH2AX primary antibody was added and incubated overnight at 4 °C. After washing with PBS, Cy3-labeled fluorescent secondary antibody was incubated for 1 hour in the dark, and the nuclei were stained with DAPI for 5 minutes before mounting. Images were observed and captured using a confocal microscope, and the fluorescence intensity was analyzed using ImageJ software.

[0118] Comet assay (single cell gel electrophoresis): U87 cells were seeded at a density of 2×10 5 cells per well in a 6-well plate and incubated at 37 °C in a 5% CO₂ atmosphere for 12 hours to allow attachment. After aspirating the original medium, the experimental group was treated with fresh medium containing temozolomide (dose: 250 μM), AOH1996, HRO761, Mixed sol, and AHN7 (total drug dose: 250 nM), and untreated cells were set as the blank control and cultured for an additional 24 hours. After collection, the cells were washed with PBS and mixed with 1.5% low melting point agarose and spread on pre-treated slides. After solidification, the slides were immersed in lysis buffer (containing 1% Triton X-100) and incubated at 4 °C in the dark for 2 hours, then transferred to alkaline lysis buffer (pH > 13) and unwound at room temperature for 30 minutes. Alkaline electrophoresis was performed at 25 V and 300 mA for 20 minutes, followed by washing with neutralization buffer and staining with ethidium bromide for 10 minutes. Images were collected using a fluorescence microscope, and the percentage of DNA content in the comet tail was quantified using ImageJ software combined with the OpenComet plugin.

[0119] When DNA double-strand breaks occur, ATM / ATR kinases are activated and phosphorylate the H2AX histone to form γH2AX foci to recruit DNA repair complexes. The density of these foci is positively correlated with the number of double-strand breaks, so γH2AX is widely used as a quantitative marker for DNA damage. As Figure 8 and Figure 9 shown, Mixed sol and AHN7 showed significantly stronger γH2AX fluorescence intensity than other groups, and there was no significant difference between the two; followed by AOH1996, while temozolomide at a 1000-fold concentration still failed to cause strong DNA damage, indicating that the synthetic lethal drug combination has stronger DNA toxicity than commercially available temozolomide.

[0120] The comet assay (single cell gel electrophoresis) is a highly sensitive method for detecting DNA damage. Through alkaline electrophoresis, damaged DNA fragments migrate to form a comet tail, and the degree of single cell DNA damage is quantitatively evaluated based on parameters such as tail moment and tail DNA content. An increase in the parameter value directly reflects the exacerbation of genomic integrity damage. The comet assay images are as shown in Figure 10 , and the quantitative results are as shown in Figure 11 and Figure 12 . The trend is consistent with the trend of γH2AX fluorescence intensity. Mixed sol and AHN7 have the strongest DNA damage effect, producing the longest comet tails and the highest proportion of tail DNA, and there is no significant difference between the two, which further verifies the ability of the synthetic lethal drug combination to exacerbate DNA damage in tumor cells. In addition, the above results also show that the hypoxia-oxidation-reduction multiple tumor response drug release ability of AHN7 enables it to rapidly release active drugs in tumor cells, and its ability to cause DNA damage is not inferior to that of free drugs.

[0121] Example 8: Biodistribution of Brain-Targeted Co-Assembled Nanopreparations

[0122] Preparation of DiR-labeled brain-targeted co-assembled nanopreparations: Referring to the process of Example 3, except that DiR was added as a tracer probe at a mass ratio of 10% (w / w) during the preparation process, and the other processes were exactly the same, and it was named AHN7-DiR for subsequent fluorescence imaging. Preparation of non-targeted control co-assembled nanopreparations (AHN-DiR): Referring to the process of Example 3, except that DiR was added as a tracer probe at a mass ratio of 10% (w / w) during the nano co-assembly process, and DSPE-PEG-T7 was replaced with an equal mass of non-targeted modification material DSPE-PEG, and the other processes were exactly the same, and it was named AHN-DiR for subsequent fluorescence imaging.

[0123] Construction of glioblastoma in situ model: Immunodeficient Balb / c nude mice were selected and fixed on a brain stereotaxic apparatus under isoflurane gas anesthesia. Through a cranial drill hole, the right caudate nucleus area was accurately located, and 5×10 5 trypsin-digested U87-luc cells (PBS-diluted suspension) were slowly injected with a microsyringe. After the operation, the wound was closed with a medical tissue adhesive. Ten days after the operation, D-luciferin potassium (150 mg / kg) was injected intraperitoneally, and the tumor growth was dynamically monitored with a live imaging system. After verifying the success of the modeling through the bioluminescence signal intensity, the animals were randomly grouped.

[0124] Biodistribution study: DiR free dye, AHN7-DiR and AHN-DiR (equivalent DiR dose of 2 mg / kg) were administered to model mice via the tail vein respectively. At 4, 8, 12 and 24 hours after administration, the animals were anesthetized at regular intervals, and the fluorescence signals in the tumor region were collected using a live imaging system. In addition, the mice were sacrificed at 12 hours, and the heart, liver, spleen, lungs, kidneys and brain tissues were isolated, and the accumulation intensity of DiR in each tissue was quantitatively evaluated by ex vivo fluorescence imaging.

[0125] The experimental results are as Figure 13 and Figure 14 shown. Only extremely low fluorescence signals were detected in the brain tissue for the DiR solution and non-brain-targeted modified AHN-DiR, while the AHN7-DiR modified with the T7 peptide brain targeting showed an obvious time-dependent accumulation of fluorescence signals in the brain. As time went by, the fluorescence signal of AHN7-DiR in the brain reached the peak at 12 hours, indicating that the modification with the T7 peptide enabled AHN7-DiR to successfully cross the blood-brain barrier and accumulate at the glioma target site.

[0126] When the maximum accumulation was reached at 12 hours, the fluorescence signal distributions in the main organs and brain tissue are as Figure 15 and Figure 16 shown. The maximum accumulations of the DiR solution and AHN-DiR occurred in the lungs and liver, and there was no obvious accumulation in the brain. In contrast, AHN7-DiR accumulated significantly in the brain tumor region. The quantitative results of the fluorescence signals in the brain are as Figure 17 shown. The signal intensity of AHN7-DiR was 6.84 times that of the DiR solution and 6 times that of AHN-DiR. The above results indicate that the brain-targeted modified AHN7-DiR constructed in this study can successfully cross the blood-brain barrier and accumulate in the glioma lesion area, laying an important foundation for its anti-tumor effect.

[0127] Example 9: In vivo anti-tumor investigation of the brain-targeted prodrug nanoplatform

[0128] Construction of the orthotopic glioblastoma model: Immunodeficient Balb / c nude mice were selected and fixed on a stereotaxic apparatus under isoflurane gas anesthesia. Through a cranial drill hole, it was accurately positioned to the right caudate nucleus region, and 5×10 5 trypsin-digested U87-luc cells (PBS-diluted suspension) were slowly injected with a microsyringe. After the operation, the wound was closed with a medical tissue adhesive. Ten days after the operation, potassium D-luciferin (150 mg / kg) was injected intraperitoneally, and the tumor growth was dynamically monitored using a live imaging system. After verifying the success of the modeling through the bioluminescence signal intensity, the animals were randomly grouped.

[0129] In vivo anti-tumor pharmacodynamic evaluation: Based on the orthotopic model of nude mice with glioblastoma, the anti-tumor effects of temozolomide, Mixed sol and AHN7 were systematically evaluated. Temozolomide (60 mg / kg) was administered by intraperitoneal injection, and Mixed sol and AHN7 (where AOH1996 was 5 mg / kg and HRO761 was 16.63 mg / kg) were injected via the tail vein. Administration started on day 0, and the drug was administered every 3 days for a total of 5 dosing cycles. During the dosing cycle (days 0, 4, 8, 12, 16), the tumor bioluminescence intensity was dynamically monitored by intraperitoneal injection of potassium D-luciferin (150 mg / kg) combined with a live imaging system for quantitative analysis. When the mouse body weight decreased to 80% of the initial body weight baseline, the experimental endpoint criteria were implemented and the mice were euthanized. On day 16, some animals were sacrificed to obtain brain tissues for H&E pathological analysis, and the survival rate observation experiment was continuously carried out on the remaining individuals. The body weight changes and survival status were recorded every 48 hours until the end of the study.

[0130] The tumor growth conditions are shown in Figure 18 and Figure 19 as follows. The mouse survival conditions are shown in Figure 20 , Figure 21 and Figure 22 as follows. In the PBS control group, the tumors in mice grew exponentially, accompanied by a rapid decrease in body weight, and the median survival time was only 18 days. The treatment group with the clinical first-line drug temozolomide showed limited anti-tumor effects, only extending the median survival time to 22 days, and all mice died within 24 days after drug administration, suggesting that it is still difficult to break through the high heterogeneity of glioblastoma. It is worth noting that although Mixed sol showed potent cytotoxicity in in vitro cell experiments, as a free drug, it could not cross the blood-brain barrier and reach the glioma target site, and at the same time might cause systemic toxicity. Eventually, there was no significant difference in the median survival time between the treatment group and the control group, and the survival curve decayed earlier, emphasizing the key role of drug delivery strategies in glioma treatment.

[0131] In contrast, AHN7 demonstrated significant therapeutic advantages. The tumor growth rate was lower than that of the control group, and the trend of body weight loss in mice was slower. The median survival time was extended to 38 days. As shown in Figure 23 , H&E section analysis further verified that AHN7 significantly inhibited the growth of glioblastoma. Temozolomide also had a certain anti-glioma effect, while the tumor size in the Mixed sol treatment group was close to that of the PBS control group, and almost no tumor proliferation was inhibited, which was consistent with the pharmacodynamic results.

[0132] In summary, the brain-targeted co-assembled nanomedicine AHN7 constructed in the present invention has a powerful anti-glioblastoma ability, and its main advantages can be attributed to the following three points: (1) inducing artificial mismatch repair defects in tumor cells by the PCNA inhibitor AOH1996, thereby enhancing their sensitivity to the MRN complex inhibitor HRO761, and the two form a synthetic lethal drug combination that can induce strong DNA damage; (2) utilizing the hypoxia-oxidation-redox triple-responsive property of the nitroimidazole dimer to achieve spatio-temporal controlled release of the drug in the tumor microenvironment, overcoming the problem of insufficient drug release caused by microenvironmental heterogeneity, and reducing systemic toxicity; (3) enabling the nanomedicine to cross the blood-brain barrier and reach the glioblastoma lesion through T7 peptide-mediated transcytosis, overcoming the limitation that free drugs cannot cross the blood-brain barrier. The AHN7 proposed in the present invention provides an innovative treatment strategy for glioblastoma and a choice of convertible formulation products.

Claims

1. Nitroimidazole dimer or a pharmaceutically acceptable salt thereof: Among them, X is a redox environment-sensitive bond, preferably a disulfide bond.

2. A pharmaceutical composition comprising the nitroimidazole dimer or a pharmaceutically acceptable salt thereof as claimed in claim 1, AOH1996 and HRO761.

3. The pharmaceutical composition according to claim 2, characterized in that, The molar ratio of AOH1996 to HRO761 is 5:1 - 1:2, preferably 1:1 - 1:

2.

4. The pharmaceutical composition according to claim 2 or 3, characterized in that, The molar ratio of AOH1996, HRO761 to the nitroimidazole dimer is 1:2:(0.5 - 2), preferably 1:2:(1 - 1.5).

5. The co-assembled nanopreparation of the pharmaceutical composition according to any one of claims 2-4, characterized in that, The co-assembled nanoformulation comprises a nitroimidazole dimer or a pharmaceutically acceptable salt thereof, AOH1996, HRO761 and a PEG modifier / active targeting modifier, and the molar ratio of AOH1996, HRO761 to the nitroimidazole dimer is 1:2:(0.5 - 2), preferably 1:2:(1 - 1.5).

6. The co-assembled nanoformulation according to claim 5, characterized in that, The PEG modifier is an amphiphilic polymer or targeting group of DSPE-PEG, TPGS, PEG-PLGA or PEG-PCL, and the active targeting modifier is a brain targeting peptide / antibody conjugate, ligand conjugate, cell-penetrating peptide conjugate, receptor targeting conjugate, sugar residue, hormone capable of targeting specific tissues, selected from DSPE-PEG-SHp, DSPE-PEG-Angiopep, DSPE-PEG-T7, DSPE-PEG-RVG29, DSPE-PEG-cRGD, DSPE-PEG-Lactoferrin, DSPE-PEG-NGR, DSPE-PEG-TAT, DSPE-PEG-iRGD, DSPE-PEG-Mannose, DSPE-PEG-OTC, DSPE-PEG-GE11, DSPE-PEG-CREKA, DSPE-PEG-TH, DSPE-PEG-R8, DSPE-PEG-APOE, preferably phospholipid-polyethylene glycol-T7 peptide.

7. The co-assembled nanopreparation according to claim 5 or 6, characterized in that, The mass ratio of the total mass of the nitroimidazole dimer or a pharmaceutically acceptable salt thereof, AOH1996 and HRO761 to the mass of the PEG modifier / active targeting modifier is: 1:(0.1 - 1).

8. Use of the nitroimidazole dimer or a pharmaceutically acceptable salt thereof as claimed in claim 1, or the pharmaceutical composition as claimed in any one of claims 2 - 4, or the co-assembled nanoformulation as claimed in any one of claims 5 - 7 in the preparation of an anti-tumor drug, and the anti-tumor drug is preferably a drug targeting glioblastoma.

9. Use of the nitroimidazole dimer or a pharmaceutically acceptable salt thereof as claimed in claim 1, or the pharmaceutical composition as claimed in any one of claims 2 - 4, or the co-assembled nanoformulation as claimed in any one of claims 5 - 7 in the preparation of a drug for treating glioblastoma heterogeneity.

10. Use of the nitroimidazole dimer or a pharmaceutically acceptable salt thereof according to claim 1 in the preparation of a drug for regulating the release of AOH1996 and HRO761 in a co-assembled nanoformulation and regulating the stability of the co-assembled nanoformulation.

Images