TREATMENT OF CANCER WITH A COMBINATION OF AN ANTIBODY THAT BINDS LGR5 AND EGFR AND A TOPOISOMERASE I INHIBITOR

20230084382 · 2023-03-16

    Inventors

    Cpc classification

    International classification

    Abstract

    The invention describes antibodies or functional parts, derivatives and/or analogues thereof that comprise a variable domain that binds an extracellular part of EGFR and a variable domain that binds an extracellular part of LGR5 for use in the treatment of cancer wherein the antibody or functional part, derivative and/or analogue thereof is administered with a topoisomerase I inhibitor.

    Claims

    1-29. (canceled)

    30. A method of treating cancer in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of a topoisomerase I inhibitor and a bispecific antibody or antigen-binding fragments thereof that comprise a variable domain that binds an extracellular part of EGFR and a variable domain that binds an extracellular part of LGR5.

    31. The method of claim 30, wherein the cancer is colorectal, pulmonary, gastrointestinal or ovarian cancer.

    32. The method of claim 31, wherein the cancer is colorectal cancer.

    33. The method of claim 30, wherein the bispecific antibody or antigen-binding fragments thereof and the topoisomerase I inhibitor are administered to the subject concurrently or sequentially.

    34. The method of claim 30, wherein the bispecific antibody or antigen-binding fragment thereof is administered to the subject prior to the topoisomerase I inhibitor.

    35. The method of claim 30, wherein the bispecific antibody or antigen-binding fragments thereof comprise one VH domain that binds EGFR and comprises the amino acid sequence of MF3755 as depicted in FIG. 8 or comprising not more than 5, 4, 3, 2, or 1 conservative amino acid substitutions; and one VH region that binds LGR5 comprises the amino acid sequence of MF5816 as depicted in FIG. 8, or comprising not more than 5, 4, 3, 2, or 1 conservative amino acid substitutions, and wherein the VH region binds antigen in association with a VL region.

    36. The method of claim 30, wherein the bispecific antibody or antigen-binding fragment thereof comprises a common light chain.

    37. The method of claim 36, wherein the common light chain comprises a germline IgVκ1-39 variable region V-segment.

    38. The method of claim 36, wherein the common light chain comprises the kappa light chain V-segment IgVκ1-39*01.

    39. The method of claim 30, wherein the bispecific antibody or antigen-binding fragment thereof comprises a light chain variable region comprising an amino acid sequence that is at least 90%, at least 95%, at least 97%, at least 98%, at least 99% identical or 100% identical to the amino acid sequence set forth in FIG. 9.

    40. The method of claim 30, wherein the topoisomerase I inhibitor is camptothecin or a derivative thereof.

    41. The method of claim 30, wherein the topoisomerase I inhibitor is irinotecan or topotecan.

    42. The method of claim 30, wherein the bispecific antibody or antigen-binding fragments thereof is antibody-dependent cellular cytotoxicity (ADCC) enhanced.

    43. The method of claim 30, wherein the bispecific antibody or antigen-binding fragments thereof is afucosylated.

    44. A method for inhibiting proliferation of a cell that expresses EGFR and LGR5 comprising contacting the cell with a topoisomerase I inhibitor and a bispecific antibody or antigen-binding fragments thereof that comprise a variable domain that binds an extracellular part of EGFR and a variable domain that binds an extracellular part of LGR5.

    45. A pharmaceutical composition comprising a topoisomerase I inhibitor and a bispecific antibody or antigen-binding fragments thereof that comprise a variable domain that binds an extracellular part of EGFR and a variable domain that binds an extracellular part of LGR5.

    46. The pharmaceutical composition according to claim 45, wherein the topoisomerase I inhibitor and the bispecific antibody or antigen-binding fragments thereof are provided in a single formulation.

    47. The pharmaceutical composition according to claim 45, wherein the topoisomerase I inhibitor and the bispecific antibody or antigen-binding fragments thereof are provided in separate formulations.

    48. A kit comprising the pharmaceutical composition of claim 45.

    Description

    BRIEF DESCRIPTION OF THE DRAWINGS

    [0184] FIG. 1 Human LGR5 sequence.

    [0185] FIG. 2 Human EGFR sequence.

    [0186] FIG. 3 Effect of treatments on tumor volume in M005 orthotopic PDX model of CRC. (a) Injection frequency, dosing and sites of injections during the treatment period; (b) Fold change in tumor volume over time; and (c) Dot plot showing fold change per mouse at 6 weeks.

    [0187] FIG. 4 (a) Tumor volume before and after treatment release in mouse model M005. Treatment was stopped after 9 weeks and tumor volume in the same mice monitored for a further 3 weeks. Numbers beneath each group indicate reasons why not all mice were included at 12 weeks; (b) Body weight in each group over time (Model M005).

    [0188] FIG. 5 (a) Number of mice that had metastases at sacrifice as detected macroscopically or evaluated by H&E staining; (b) Residual disease at week 12.

    [0189] FIG. 6 (a) Mouse model M001: Injection frequency, dosing and site of injection during treatment period; (b) Change in mean tumor volume over time; (c) Dot plot showing tumor volume per mouse at 6 weeks.

    [0190] FIG. 7 (a) Body weight in each group over time (Model M001), combination treatment was not toxic; (b) Treatment with a bispecific MF5816xMF3755 alone or a bispecific MF5816xMF3755+irinotecan blocked metastasis. Metastases were evaluated in tissues macroscopically and by staining with H&E.

    [0191] FIG. 8 (a) Amino acid sequences of heavy chain variable regions of MF5816xMF3755 that together with a common light chain variable region such as the variable region of the human kappa light chain IgVκ1 39*01/IGJκ1*01 form a variable domain that binds LGR5 or EGFR. The CDR and framework regions are indicated in FIG. 8b. A DNA sequence is indicated in FIG. 8c. Additional heavy chain variable regions binding EGFR and LGR5, which are suitable for the generation of bispecific antibodies in combination with a topoisomerase I inhibitor are further disclosed in this figure.

    [0192] FIG. 9 Amino acid sequence of a) a common light chain amino acid sequence. b) common light chain variable region DNA sequence and translation (IGKV1-39/jk1). c) Common light chain constant region DNA sequence and translation. d) IGKV1-39/jk5 common light chain variable region translation. e) V-region IGKV1-39A; f) CDR1, CDR2 and CDR3 of a common light chain.

    [0193] FIG. 10. IgG heavy chains for the generation of bispecific molecules. a) CH1 region. b) hinge region. c) CH2 region. d) CH3 domain containing variations L351K and T366K (KK). e) CH3 domain containing variations L351D and L368E (DE).

    EXAMPLES

    [0194] As used herein “MFXXXX” wherein X is independently a numeral 0-9, refers to a Fab comprising a variable domain wherein the VH has the amino acid sequence identified by the 4 digits depicted in FIG. 8. Unless otherwise indicated the light chain variable region of the variable domain typically has a sequence of FIG. 9b. The light chain in the examples has a sequence as depicted in FIG. 9a. “MFXXXX VH” refers to the amino acid sequence of the VH identified by the 4 digits. The MF further comprises a constant region of a light chain and a constant region of a heavy chain that normally interacts with a constant region of a light chain. The VH/variable region of the heavy chains differs and typically also the CH3 region, wherein one of the heavy chains has a KK mutation of its CH3 domain and the other has the complementing DE mutation of its CH3 domain (see for reference PCT/NL2013/050294 (published as WO2013/157954) and FIGS. 10d and 10e. Bispecific antibodies in the examples have an Fc tail with a KK/DE CH3 heterodimerization domain, a CH2 domain and a CH1 domain as indicated in FIG. 10, a common light chain as indicated in FIG. 9a and a VH as specified by the MF number. For example a bispecific antibody indicated by MF3755xMF5816 has the above general sequences and a variable domain with a VH with the sequence of MF3755 and a variable domain with a VH with the sequence of MF5816.

    Example 1

    Cell Lines

    [0195] Freestyle 293F cells (cat. nr. p/n51-0029) were obtained from Invitrogen and routinely maintained in 293 FreeStyle medium. HEK293T (ATCC-CRL-11268) and CHO-K1 (DSMZ ACC110) cell lines were purchased from ATCC and routinely maintained in DMEM/F12 (Gibco) supplemented with L-Glutamine (Gibco) and FBS (Lonza).

    [0196] The amino acid and nucleic acid sequences of the various heavy chain variable region (VH) are indicated in FIG. 8. Bispecific antibodies EGFR/LGR5, MF3755xMF5814; comprising heavy chain variable regions MF3755 and MF5816 and a common light chain and including modifications for enhanced ADCC from afucocylation, among other LGR5 and EGFR combinations as depicted in FIG. 9a have been shown to be effective in WO2017/069628 (page 138).

    Generation of Bispecific Antibodies

    [0197] Bispecific antibodies were generated by transient co-transfection of two plasmids encoding IgG with different VH domains, using a proprietary CH3 engineering technology to ensure efficient heterodimerisation and formation of bispecific antibodies. The common light chain is also co-transfected in the same cell, either on the same plasmid or on another plasmid. In our applications (e.g. WO2013/157954 and WO2013/157953; incorporated herein by reference) we have disclosed methods and means for producing bispecific antibodies from a single cell, whereby means are provided that favor the formation of bispecific antibodies over the formation of monospecific antibodies. These methods can also be favorably employed in the present invention. Specifically, preferred mutations to produce essentially only bispecific full length IgG molecules are amino acid substitutions at positions 351 and 366, e.g. L351K and T366K (numbering according to EU numbering) in the first CH3 domain (the ‘KK-variant’ heavy chain) and amino acid substitutions at positions 351 and 368, e.g. L351D and L368E in the second CH3 domain (the ‘DE-variant’ heavy chain), or vice versa (see FIGS. 10d and 10e). It was previously demonstrated in the mentioned applications that the negatively charged DE-variant heavy chain and positively charged KK- variant heavy chain preferentially pair to form heterodimers (so-called ‘DEKK’ bispecific molecules). Homodimerization of DE-variant heavy chains (DE-DE homodimers) or KK-variant heavy chains (KK-KK homodimers) hardly occurs due to strong repulsion between the charged residues in the CH3-CH3 interface between identical heavy chains.

    [0198] VH genes of variable domain that bind LGR5 described above were cloned into the vector encoding the positively charged CH3 domain. The VH genes of variable domain that bind EGFR such as those disclosed in WO 2015/130172 (incorporated herein by reference) were cloned into vector encoding the negatively charged CH3 domain. Suspension growth-adapted 293F Freestyle cells were cultivated in T125 flasks on a shaker plateau until a density of 3.0×10e6 cells/ml. Cells were seeded at a density of 0.3-0.5×10e6 viable cells/ml in each well of a 24-deep well plate. The cells were transiently transfected with a mix of two plasmids encoding different antibodies, cloned into the proprietary vector system. Seven days after transfection, the cellular supernatant was harvested and filtered through a 0.22 μM filter (Sartorius). The sterile supernatant was stored at 4° C. until purification of the antibodies.

    IgG Purification

    [0199] Purifications were performed under sterile conditions in filter plates using filtration. First, the pH of the medium was adjusted to pH 8.0 and subsequently, IgG-containing supernatants were incubated with protein A Sepharose CL-4B beads (50% v/v) (Pierce) for 2 hrs at 25° C. on a shaking platform at 600 rpm. Next, the beads were harvested by filtration. Beads were washed twice with PBS pH 7.4. Bound IgG was then eluted at pH 3.0 with 0.1 M citrate buffer and the eluate was immediately neutralized using Tris pH 8.0. Buffer exchange was performed by centrifugation using multiscreen Ultracel 10 multiplates (Millipore). The samples were finally harvested in PBS pH7.4. The IgG concentration was measured using Octet. Protein samples were stored at 4° C.

    IgG Quantification using Octet

    [0200] To determine the amount of IgG purified, the concentration of antibody was determined by means of Octet analysis using protein-A biosensors (Forte-Bio, according to the supplier's recommendations) using total human IgG (Sigma Aldrich, cat. nr. 14506) as standard.

    Mice and Preparation of Cells for Engraftment

    [0201] Tumoroids were grown for seven days before disaggregating into a single cells suspension for injection. For all mouse studies female NOD.CB17/AlhnRj-Prkdcscid/Rj mice (Janvier Labs) aged between 6-8 weeks were used.

    Culture conditions and the Method of Creating Single Cells

    [0202] Organoids derived from a colorectal cancer sample were cultured in 100% Basement Membrane Extracts (BME, Amsbio), at 37° C. and 5% CO2, with media composed of Advanced DMEM/F12 (Invitrogen) supplemented with: 2 mM GlutaMax (Invitrogen), 10 mM HEPES (Invitrogen), 1×B27 retinoic acid free (Invitrogen), 50 ng/mL EGF (Peprotech), 0.1 μg/mL Noggin (Peprotech), Rock-inhibiter Y-27632 (Sigma-Aldrich), 10 nM PGE2 (Sigma-Aldrich), 3 μm SB202190 (Sigma-Aldrich), 10 nM Gastrin (Tocris), 1 μg/ml R-SPO1 (home-made), 10 mM Nicotinomide (Sigma-Aldrich), 1.25 mM N-Acetyl-cysteine (Sigma-Aldrich), 0.5 μM A83-01 (Tocris). The day prior to analysis, the organoids were disaggregated into single cells. To this aim, the organoids were first liberated from the BME by removing the culturing media, and re-suspending the BME in cell recovery solution (BD Biosciences), and incubating for 1 hour on ice. Subsequently, the organoids were centrifuged (all centrifuge steps were for 5 minutes, 200 g at 4° C.). The pellet was re-suspended in 1 mL of 50% Trypsin/EDTA Solution (TE); 50% PBS, and pipetted up and down, and regularly visually assessed until a single cell suspension was achieved. The TE was diluted in 10 mL of PBS and centrifuged. The cells were washed twice in 10 mL of PBS before re-suspending in BME and aliquoting into 50 μL drops on to pre-warmed plates (37° C). The BME drops were left to set for 15 minutes before 500 μL of media were added per drop. After 12 hours the cells were isolated from the BME using cell recovery solution. After 1 hour on ice, the cells were centrifuged, and washed once in 10 mL of PBS containing 0.5% BSA and 0.5 mM EDTA (staining buffer). The pellet was then re-suspended in staining buffer and counted.

    [0203] The Stem Cell and Cancer Group at VHIO has developed a collection of CRC PDX models derived from surgically resected primary tumors (colon and rectum) and liver metastases. PDX models are clinically and molecularly annotated and faithfully represent the clinical epidemiology of mCRC. These models can be injected subcutaneously or orthotopically in the cecum wall of immunodeficient mice. Orthotopic models generate local and distant metastases in lymph nodes, liver, lungs and carcinomatoses, reproducing the advance disease in CRC patients. A set of PDX models with key molecular traits was selected to evaluate the efficacy of anti-LGR5/EGFR bi-specific antibodies of the invention (See Table 1). Several mutant and wild type models were selected in the initial PDX set. Other determinants such as the relative expression of EGFR or LGR5 that could also determine response to the EGFR/LGR5 antibodies developed have been measured in these PDX models (Table 1).

    [0204] PDX models derived from liver metastases of three advanced CRC patients (Table 1) were selected. Two models are KRAS mutant (G13D and G12D for M005 and M001, respectively), of which M005 is also an APC and a PIK3CA 112_112del mutant.

    [0205] Model M005: 120 NOD-SCID mice were given orthotopic cecum wall injections of 1×10.sup.6 tumor cells derived from the M005 PDX model, where the model was generated essentially as described in Puig et al., A Personalized Preclinical Model to Evaluate the Metastatic Potential of Patient-Derived Colon Cancer Initiating Cells, Clin Cancer Res; 19 (24), 6787-6801 (2013), which is incorporated in its entirety into this application. These human tumor cells were derived from a CRC liver metastasis and contain mutations in the KRAS gene (KRAS G13) and in the PIK3CA gene (PIK3CA 112_112del). See UK Sanger Institute, featured 18 different tissue types carrying mutation PIK3C C420R (cancer.sanger.ac.uk/cosmic, mutation ID COSM757). From 15 days post-injection, weekly CT imaging was used to monitor mice and detect primary tumors in the cecum. Treatments were initiated after at least 80% of animals had a primary tumor growing in the cecum. The following 18 mice were excluded: which died after surgery (#5), with no primary tumor (#7), too small or too large tumors (#2 and #1 respectively), low body weight (#2), general signs of illness (#1).

    [0206] Remaining 102 mice were treated according to FIG. 3a and imaged weekly with microCT. The frequency and size of metastatic lesions was also determined by histological evaluation of liver and lungs (Hematoxylin and eosin stain (H&E) staining) Peritoneal carcinomatoses were detected macroscopically upon necropsy and later confirmed by histology.

    [0207] Model M001: M001 PDX model was generated essentially as described in Puig et al., A Personalized Preclinical Model to Evaluate the Metastatic Potential of Patient-Derived Colon Cancer Initiating Cells, Clin Cancer Res; 19 (24), 6787-6801 (2013), which is incorporated in its entirety into this application; See UK Sanger Institute, featured 18 different tissue types carrying mutation PIK3C C420R (cancer.sanger.ac.uk/cosmic, mutation ID COSM757). In the second orthotopic model, injected human tumor cells were originally derived from a CRC liver metastasis with mutations: KRAS G12D and PIK3CA-C420R. Injections of tumor cells were similarly done as above. The following 18 mice were excluded: which died after surgery (#11), with no primary tumor (#2), mice with too large tumors (#2), low body weight (#1), and general signs of illness (#2). Dosing and treatment regime were according to FIG. 6a.

    [0208] At week 6, all mice treated with vehicle or the bispecific EGFR/LGR5, comprising MF3755 and MF5816 only were sacrificed; roughly half of the mice treated with irinotecan or the bispecific EGFR/LGR5, comprising MF3755 and MF5816+irinotecan were also sacrificed.

    Results

    Analysis Model M005

    [0209] The mean tumor volume in mice treated with the bispecific EGFR/LGR5, comprising MF3755 and MF5816 alone was lower than mice given vehicle, but not as low as that of mice treated with irinotecan alone. Surprisingly, mice receiving the bispecific EGFR/LGR5, comprising MF3755 and MF5816+irinotecan combination treatment had a lower tumor volume compared to all other groups of mice (FIG. 3b, 3c). Interestingly, after treatment release, the bispecific EGFR/LGR5, comprising MF3755 and MF5816 prolonged the tumor growth-blocking effect of irinotecan as seen in the fold change in tumor volume (FIG. 4a).

    [0210] The primary tumors from all mice were harvested at sacrifice and analyzed for frequency and size of metastatic lesions. FIG. 5a shows the numbers of mice found to have metastatic lesions at sacrifice, demonstrating that mice treated with the bispecific EGFR/LGR5, comprising MF3755 and MF5816 or irinotecan, either alone or in combination, had fewer metastases than untreated mice.

    [0211] Tissues were also analyzed in mice which were subjected to treatment release (9 weeks) and sacrifice after a 3-week treatment-free period. Smaller tumors were found to contain necrotic cells or only a small number of tumor cells, whereas most of the larger tumors had abundant tumor cells (FIG. 5b). This analysis showed that tumor volume and cecum weight were positively correlated in treated mice, 3 weeks after treatment release (P<0.0001 for Pearson correlation coefficient).

    Analysis Model M001

    [0212] The mean tumor volume in mice treated with the bispecific EGFR/LGR5, comprising MF3755 and MF5816 alone was very similar to that of mice treated with irinotecan alone. However, mice receiving the bispecific EGFR/LGR5, comprising MF3755 and MF5816+irinotecan combination treatment had a lower tumor volume than any of the other group of mice (FIG. 6b,c). No toxicity was observed in mice receiving the combination of the bispecific EGFR/LGR5, comprising MF3755 and MF5816+irinotecan (FIG. 7a).

    [0213] Histological analysis to determine metastatic lesions at sacrifice demonstrated that mice treated with bispecific EGFR/LGR5, comprising MF3755 and MF5816 or irinotecan, either alone or in combination, had fewer metastases than untreated mice (FIG. 7b).

    [0214] Two orthotopic models M005 and M001 were tested with the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 and chemotherapy drug irinotecan, alone and in combination for their potential to inhibit tumor growth and metastatic potential. In M005, the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 and irinotecan alone were able to delay primary tumor growth, but combination of the two demonstrated the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 and irinotecan promote a superior response. After treatment release, combined treatment completely eliminated primary tumors in five out of five surviving mice. For irinotecan monotherapy this was the case in just 1 out of 14 mice. Again, this indicates that the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 potentiates complete tumor regression induced by chemotherapy. In terms of metastatic potential, the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 blocked the formation of distant metastases, as did irinotecan. No metastases were seen in mice treated with combined irinotecan+the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816.

    [0215] The results from model M005 were confirmed in model M001 model. The bispecific antibody EGFR/LGR5, comprising gMF3755 and MF5816 and irinotecan alone were equally effective at delaying primary tumor growth in M001, however, when administered together, the combined treatment appeared to be more effective than either agent given alone.

    [0216] Statistical analysis (ANCOVA) on tumor volumes at week 6 of data shown in FIG. 6c showed that treatment significantly reduced tumor volume between all of the groups except between irinotecan and the bispecific antibody comprising MF3755 and MF5816. (Vehicle vs. MF3755 and MF5816, p<0.0001; vehicle vs. irinotecan, p<0.0001; vehicle vs. irinotecan+MF3755 and MF5816, p<0.0001; MF3755 and MF5816 vs. irinotecan, p<0.6429; MF3755 and MF5816 vs irinotecan+MF3755 and MF5816, p<0.0001; irinotecan+MF3755 and MF5816 vs. irinotecan, p<0.0001.)

    [0217] In M001 combined treatment was not more toxic than irinotecan alone. In terms of metastatic potential, EGFR/LGR5, MF3755 and MF5816 blocked the formation of distant metastases, as did irinotecan. No metastases were seen in mice treated with combined irinotecan+the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816.

    [0218] In conclusion, using two orthotopic CRC tumor models, it has been found that a combination treatment of the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 and irinotecan results in tumor regression to a greater degree compared to when these drugs are administered alone. In addition, metastasis was found to be blocked in the bispecific antibody EGFR/LGR5, comprising MF3755 and MF5816 and irinotecan alone or combined treatments.

    TABLE-US-00001 TABLE 1 Characteristics of PDX models originating from liver metastases of CRC patients. LGR5, EGFR and nuclear β-catenin were determined by immunofluorescence quantification. Mutation status of Wnt signaling (APC, RSPO, RNF43, ZNRF3) and oncogenic (KRAS, PIK3CA, TP53) proteins were determined by genomic analysis. Sensitivity of the PDX models (grown subcutaneously) to WNT inhibitors is indicated in the dark cells. The PDX model T108 was not used in the further experiments. PDX Nuclear RSPO ID LGR5 EGFR β-cat APC FUSIONS RNF43 ZNRF3 KRAS PIK3CA TP53 MSI M005 1398.06 348.29 10868 MUT WT WT WT MUT G13D MUT WT MSS T108 4331.36 NA NA WT MUT WT WT WT WT MUT NA M001 5757.52 483.81  2501 WT WT WT WT MUT G12D MUT MUT NA