ANTI-CD56 ANTIBODY-DRUG CONJUGATES AND THEIR USE IN THERAPY

Abstract

The invention relates to antibody-drug conjugates and to their use in therapy, in particular for treating CD56+ cancers.

Claims

1. An anti-CD56 antibody-drug conjugate (ADC) comprising an anti-CD56 antibody and a drug conjugate, wherein the ADC has one or more effective functions mediated by an Fc portion of the anti-CD56 antibody attenuated, wherein the one or more effective functions mediated by the Fc portion are selected from ADCC (Antibody-Dependent Cell-mediated Cytotoxicity) and CDC (Complement-Dependent Cytotoxicity).

2. The anti-CD56 ADC as claimed in claim 1, comprising a mutation of Fc portion of the anti-CD56 antibody.

3. The anti-CD56 ADC as claimed in claim 1, comprising a mutation of the Fc portion that reduces the effective functions mediated by the Fc portion of the anti-CD56 antibody.

4. The anti-CD56 ADC as claimed in claim 1, comprising a mutation of the Fc portion that aims at deglycosylating the Fc portion of the anti-CD56 antibody.

5. The anti-CD56 ADC as claimed in claim 1, comprising a mutation of the Fc portion of the ant-CD56 antibody that aims at suppressing glycosylation at asparagine 297.

6. An anti-CD56 antibody-drug conjugate (ADC) comprising a Fc portion, wherein said Fc portion does not carry a glycosylation.

7. The anti-CD56 ADC as claimed in claim 1, wherein the amino acid sequence of the anti-CD56 antibody comprises a mutation aimed at deglycosylating the Fc portion of said anti-CD56 antibody.

8. The anti-CD56 ADC as claimed in claim 1, wherein the amino acid sequence of the anti-CD56 antibody comprises a mutation aimed at suppressing glycosylation at asparagine 297.

9. The anti-CD56 antibody as claimed in claim 1, wherein the anti-CD56 antibody does not carry a glycosylation at asparagine 297.

10. The anti-CD56 antibody as claimed in claim 1, wherein the amino acid sequence of the anti-CD56 antibody comprises a substitution of asparagine 297 with alanine.

11. The anti-CD56 ADC as claimed in claim 1, wherein the antibody is an IgG antibody.

12. The anti-CD56 antibody-drug conjugate according to claim 1, having the following formula (I): ##STR00037## in which: A is an anti-CD56 antibody or an antibody fragment; the attachment head is represented by one of the following formulae: ##STR00038## the linker is a cleavable linker represented by the following formula: ##STR00039## the spacer is represented by the following formula: ##STR00040## m is an integer from 1 to 10; n is an integer from 1 to 4.

13. The antibody-drug conjugate as claimed in claim 1, in which the cytotoxic drug is selected from methotrexate, IMIDs, duocarmycin, combretastatin, calicheamicin, monomethyl auristatin E (MMAE), monomethyl auristatin F (MMAF), maytansine, DM1, DM4, SN38, amanitin, pyrrolobenzodiazepine, pyrrolobenzodiazepine dimer, pyrrolopyridodiazepine, pyrrolopyridodiazepine dimer, an inhibitor of histone deacetylase, an inhibitor of tyrosine kinase, and ricin, preferably MMAE.

14. The antibody-drug conjugate as claimed in claim 1, having the following formula: ##STR00041##

15. The antibody-drug conjugate as claimed in claim 1, in which the anti-CD56 antibody is m906.

16. The antibody-drug conjugate as claimed in claim 1, with the following formula (Ia): ##STR00042## or with the following formula (Ia): ##STR00043##

17. A composition comprising one or more antibody-drug conjugate(s) as claimed in claim 1.

18. The composition as claimed in claim 17, further comprising paclitaxel, docetaxel, doxorubicin and/or cyclophosphamide, lenalidomide, dexamethasone, carboplatin, etoposide and/or an antibody used in anti-cancer immunotherapy such as an anti-PD1 or an anti-PD-L1.

19. A method of treating a CD56+ cancer in a subject in need thereof comprising administering to the subject an anti-CD56 antibody drug conjugate (ADC) according to claim 1 or a composition according to claim 17.

20. The method of treatment as claimed in claim 19, wherein the CD56+ cancer is selected from the group consisting of melanoma, blastemal tumors, hemopathies such as acute myeloid leukemias, myelomas, blastic plasmacytoid dendritic cell neoplasms and neuroendocrinal carcinomas.

21. The method of treatment as claimed in claim 19, wherein the CD56+ cancer is a neuroendocrine carcinomas selected from the group consisting ofsmall cell lung carcinoma, neuroblastoma or Merkel cell carcinoma, preferably Merkel cell carcinoma.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0178] FIG. 1 represents the HIC (Hydrophobic Interaction Chromatography) profile of an antibody-drug conjugate composition in accordance with the invention. The figure shows that the composition is enriched in DAR4, with more than 50% DAR4.

[0179] FIG. 2 represents a SEC (Size Exclusion Chromatography) analysis of an antibody-drug conjugate composition in accordance with the invention. The figure shows that the composition is extremely homogeneous, with more than 99% of monomer.

[0180] FIG. 3 represents a native mass spectrometric analysis on a Vion IMS Qtof mass spectrophotometer coupled to an Acquity UPLC H-Class system from Waters (Wilmslow, UK) of an antibody-drug conjugate composition in accordance with the invention. This method can be used to clearly identify each type of DAR. The figure shows that the composition is enriched in DAR4, with close to 75% DAR4.

[0181] FIG. 4 represents the expression of CD56 for different cell lines (4 of MCC and one breast cancer control line) by flow cytometry after incubation with the anti-CD56 antibody labeled with the fluorophore FITC (Fluorescein IsoThioCyanate).

[0182] FIG. 5A, FIG. 5B, FIG. 5C, FIG. 5D, FIG. 5E represent the evaluation of the performances of the ADC MF-m906-MMAE, compared with the m906 controls (non-coupled antibody), MF-TTZ-MMAE (ADC Trastuzumab coupled to MMAE) and MMAE (toxin alone) on the MCC cell lines (FIG. 5A: WaGa, FIG. 5B: PeTa, FIG. 5C: MS-1 and FIG. 5D: MKL-2) and breast cancer cell line (FIG. 5E: SK-BR3). The experiments were carried out independently in duplicate (6 biological replicates/experiment). The results are expressed as the mean (+/SEM) of the percentage obtained.

[0183] FIG. 6A-6C represent the evaluation of the specificity of MF-m906-MMAE in the WaGa lines after knock-down of CD56. The WaGa cells were transduced independently with lentiviral vectors containing two distinct shRNAs inducible by doxycycline (Dox) (A, B and C) and targeting the sequences of CD56. After antibiotic selection (puromycin), the cells were exposed to doxycycline for 7 days before evaluation. (A-B-C): Confirmation of knock-down of CD56 by real-time RT-qPCR (A) (*: p<0.05, the control cells without doxycycline were used as a reference), by Western blot (attenuated mass of CD56=140 KDa) (B). (C) Evaluation of the cytotoxic effect of MF-m906-MMAE on cells expressing CD56 (Dox) or not expressing CD56 (+Dox) by using three different shRNAs (Sh RNA A, Sh RNA B and Sh RNA C). The experiments were carried out independently in duplicate (6 biological replicates/experiment). The results are expressed as the mean (+/SEM) of the percentage obtained.

[0184] FIG. 7 illustrates the evaluation of the performances of the conjugate MF-m906-MMAE and the controls (PBS and MF-TTZ-MMAE, a Trastuzumab ADC coupled to MMAE) in a xenograft model of MCC. FIG. 7 represents the plots for relative tumor volume (mean+/SEM) during the study. The relative tumor volumes (RTV) were calculated using the tumor volume for the start of the study as the reference (Volume at d0=100%). Each point represents the mean of the RTVs for the groups treated with MF-m906-MMAE, MF-TTZ-MMAE and PBS, the gray arrows indicating the time of the injections.

[0185] FIG. 8 illustrates the evaluation of the performances of the conjugate MF-m906-MMAE and the controls (PBS and MF-TTZ-MMAE, a Trastuzumab ADC coupled to MMAE) in a xenograft model of MCC. FIG. 8 represents the weight of the tumors at the end of the study in the experimental group and the control groups (*: p<0.05). The horizontal lines are the means, the quartiles and the limits.

[0186] FIG. 9 represents the evaluation of the performances of the ADC MF-m906-MMAE, compared with the controls m906 (non-coupled antibody), MF-TTZ-MMAE (Trastuzumab ADC coupled to MMAE) and MMAE (toxin alone) on the cell line for small cell lung cancer, H69. The experiments were carried out independently in triplicate (6 biological replicates/experiment). The results are expressed as the mean (+/SEM) of the percentage obtained.

[0187] FIG. 10 represents the evaluation of binding of mAb-003 on several CD56+ cell lines and equivalent CRISPR cell lines (CD56). Merkel cell carcinoma cell lines are WaGa, PeTa and MKL-1. Small cell lung cancer cell lines are H69 and H209. Prostate cancer cell line is H660.

[0188] FIG. 11 represents the killing effect of MIO-003 and LM on CD56+ tumor MCC and SCLC cell lines.

[0189] FIG. 12 represents the killing effect of MIO-003 and LM on CD56 tumor cell lines.

[0190] FIG. 13 shows ADCC of MIO-001, MIO-002, mAb-001 and mAb-002 on primary NK Cells after 4 h incubation (A) and tumor cell cytoxicity on WaGa MCC cell line after 4 days of incubation (B).

[0191] FIG. 14 represents the killing effect of the antibodies mAb-001, mAb-002, mAb-003, L and L-002 on primary NK cells after 4 h of incubation.

[0192] FIG. 15 represents the killing effect of the ADCs MIO-001, MIO-002, MIO-003, LM and LM-002 on primary NK cells after 4 h of incubation.

[0193] FIG. 16 represents the killing effect of the antibodies mAb-001, mAb-002, mAb-003, L and L-002 on primary NK cells after 4 days of incubation.

[0194] FIG. 17 represents the killing effect of the ADCs MIO-001, MIO-002, MIO-003, LM and LM-002 on primary NK cells after 4 days of incubation.

[0195] FIG. 18 represents the killing effect of the antibodies mAb-001, mAb-002, mAb-003 and L on neutrophils.

[0196] FIG. 19 represents the killing effect of neutrophils exposed to the ADCs MIO-001, MIO-002, MIO-003 and LM.

[0197] FIG. 20 represents the killing effect of the antibodies (mAb-001, mAb-002, mAb-003, L and L-002) towards monocytes.

[0198] FIG. 21 represents the killing effect of ADCs (MIO-001, MIO-002, MIO-003, LM and LM-002) on monocytes.

[0199] FIG. 22 shows cell cycle analysis of CD56+ and CD56 cells incubated with MIO-003 without co-culture and with co-culture.

[0200] FIG. 23 shows the mean relative tumor volume curves (+/SEM) during experiment. Monitoring of tumor volume progression after IV injection twice a week of PBS, MIO-001, MIO-002, MIO-003 or LM at 5 mg/kg. Relative tumor volumes (RTV) were calculated using start point tumor volume as reference (Volume day 0=100%).

[0201] FIG. 24 shows the mean end tumor weight at day=30 after IV injection twice a week of PBS, MIO-001, MIO-002, MIO-003 or LM at 5 mg/kg.

[0202] FIG. 25 shows the relative mice weight (Mean+/SEM) monitored during 30 days after IV injection twice a week of PBS, MIO-001, MIO-002, MIO-003 or LM at 5 mg/kg.

[0203] FIG. 26 shows the mean relative tumor volume curves (+/SEM) during experiment. Monitoring of tumor volume progression after one single IV injection of PBS or MIO-003 at 10, 30, 50 or 70 mg/kg. Relative tumor volumes (RTV) were calculated using start point tumor volume as reference (Volume day 0=100%).

[0204] FIG. 27 shows the mean relative mice weight curves (+/SEM) during experiment. Monitoring of mice weight progression after one single IV injection of PBS or MIO-003 at 10, 30, 50 or 70 mg/kg. Relative tumor volumes (RTV) were calculated using start point tumor volume as reference (Volume day 0=100%).

[0205] FIG. 28 shows the final mean tumor weight at day=30 for each group after one single IV injection of PBS or MIO-003 at 10, 30, 50 or 70 mg/kg.

EXAMPLES

Example 1: Synthesis of a Cytotoxic Conjugate in Accordance with the Invention

General Reaction Scheme

[0206] ##STR00028##

Detailed Reaction Scheme

Preparation of Benzyl Isonicotinate (2)

[0207] ##STR00029##

[0208] Isonicotinic acid (1) (5.00 g; 40.614 mmol; 1.0 eq) was dissolved in thionyl chloride (15 mL; 206.77 mmol; 5.1 eq) and heated under reflux overnight. After returning to ambient temperature, the excess thionyl chloride was eliminated by evaporation under reduced pressure, then the residue obtained was dissolved in anhydrous dichloromethane (55 mL). Benzyl alcohol was added (4.2 mL; 40.614 mmol; 1.0 eq) and the mixture was stirred under reflux for 10 h. After returning to ambient temperature, the reaction medium was neutralized with a saturated solution of sodium hydrogen carbonate and extracted with dichloromethane (3100 mL). The organic phases were combined, washed with a saturated solution of sodium chloride, dried over magnesium sulfate and concentrated under reduced pressure. The product obtained was purified by flash chromatography (SiO.sub.2, cyclohexane/ethyl acetate 50:50) in order to give (2) (6.97 g; 80%) in the form of a colorless oil.

[0209] .sup.1H NMR (300 MHz, DMSO) 8.80 (dd; J=6.1; 1.6 Hz; 2H.sub.1,5), 7.86 (dd; J=6.1; 1.6 Hz, 2H.sub.2,4), 7.56-7.29 (m; 5H.sub.9-13), 5.39 (s; 2H.sub.7).

[0210] .sup.13C NMR (75 MHz, DMSO) 165.0 (1C.sub.6); 151.3 (2C.sub.1,5); 137.2 (1C.sub.3); 136.1 (1C.sub.8); 129.0 (2C.sub.10,12); 128.8 (1C.sub.11); 128.6 (2C.sub.9,13); 123.0 (2C.sub.2,4); 67.4 (1C.sub.7).

[0211] HRMS (ESI): calculated neutral mass for C.sub.13H.sub.11NO.sub.2 [M]: 213.0790; observed: 213.0796.

Preparation of Benzyl 2,6-bis(hydroxymethyl)isonicotinate (3)

[0212] ##STR00030##

[0213] Benzyl isonicotinate (2) (2.48 g; 11.630 mmol; 1.0 eq) was dissolved in methanol (43 mL), stirred at 50 C. and concentrated sulfuric acid (320 L; 6.016 mmol; 0.52 eq) was added. A solution of ammonium persulfate (26.500 g; 116.000 mmol; 10.0 eq) in water (43 mL) was added in two steps: a first rapid addition of 30 droplets; a white suspension was formed, then rapidly, drop by drop, for 5 min. The reaction ran away up to 75 C., then the yellow solution obtained was stirred at 50 C. for an additional 1 h. After returning to ambient temperature, the methanol was evaporated under reduced pressure. 50 mL of ethyl acetate was added, and the medium was neutralized by addition of a saturated solution of sodium hydrogen carbonate. The aqueous phase was extracted with ethyl acetate (3100 mL) and the combined organic phases were washed with a saturated solution of sodium chloride, dried over magnesium sulfate, then concentrated under reduced pressure. The impure product was purified by flash chromatography (SiO.sub.2, dichloromethane/methanol, 95:5) in order to give (3) (1.56 g; 49%) in the form of a beige solid.

[0214] .sup.1H NMR (300 MHz, DMSO) 7.81 (s; 2H.sub.2,4); 7.55-7.32 (m; 5H.sub.9-13); 5.60 (t; J=5.9 Hz; 2H.sub.15,17); 5.40 (s; 2H.sub.7); 4.59 (d; J=5.9 Hz; 4H.sub.14,16).

[0215] .sup.13C NMR (75 MHz, DMSO) 165.0 (1C6); 162.8 (2C.sub.1,5); 138.0 (1C.sub.3); 135.7 (1C.sub.8); 128.6 (2C.sub.10,12); 128.4 (1C.sub.11); 128.3 (2C.sub.9,13); 117.0 (2C.sub.2,4); 66.9 (1C.sub.7); 63.9 (2C.sub.14,16).

[0216] HRMS (ESI): calculated neutral mass for C.sub.15H.sub.15NO.sub.4 [M]: 273.1001; observed: 273.1001.

Preparation of 2,6-bis(hydroxymethyl)isonicotinic acid (4)

[0217] ##STR00031##

[0218] Benzyl 2,6-bis(hydroxymethyl)isonicotinate (3) (1.33 g; 4.867 mmol; 1.0 eq) was dissolved in methanol (50 mL) and the solution was degassed with argon for 15 min. Palladium on carbon, 10% by weight (133 mg) was added and the reaction medium was stirred at ambient temperature in an atmosphere of hydrogen for 2 h. The reaction medium was filtered over dicalite (rinsed with methanol). The filtrate was concentrated under reduced pressure in order to give (4) (849 mg; 95%) in the form of a beige solid.

[0219] .sup.1H NMR (300 MHz, DMSO) 7.78 (s; 2H.sub.2,4); 5.54 (s broad; 2H.sub.9,11); 4.59 (s; 4H.sub.8,10).

[0220] .sup.13C NMR (75 MHz, DMSO) 166.7 (1C.sub.6); 162.5 (2C.sub.1,5); 139.4 (1C.sub.3); 117.3 (2C.sub.2,4); 64.0 (2C.sub.8,10).

[0221] HRMS (ESI): calculated neutral mass for C.sub.8H.sub.9NO.sub.4 [M]: 183.0532; observed: 183.0526.

Preparation of Methyl 6-((2,6-bis(hydroxymethyl)pyridin-4-yl)amidohexanoate (5)

[0222] ##STR00032##

[0223] 2,6-bis(hydroxymethyl)isonicotinic acid (4) (50 mg; 0.273 mmol; 1 eq) was dissolved in anhydrous N,N-dimethylformamide (3.0 mL), the solution was cooled to 0 C., then HATU (156 mg; 0.410 mmol; 1.5 eq) and 2,6-lutidine (147.0 L; 1.260 mmol; 4.7 eq) were added. The activation solution was stirred at 0 C. for 15 min, then methyl 6-aminohexanoate (59 mg; 0.322 mmol; 1.2 eq) was added. The walls of the flask were rinsed with 2 mL of anhydrous N,N-dimethylformamide and the reaction medium was stirred at ambient temperature for 15 h. The reaction mixture was diluted in ethyl acetate, washed three times with a saturated solution of sodium chloride, dried over magnesium sulfate, filtered and concentrated under reduced pressure. The product was purified by flash chromatography (dichloromethane/methanol, 90:10) in order to give (5) (76 mg; 91%) in the form of an off-white solid.

[0224] .sup.1H NMR (300 MHz, DMSO) 8.79 (t; J=5.6 Hz; 1H.sub.7); 7.71 (s; 2H.sub.2,4); 5.50 (t; J=5.8 Hz; 2H.sub.16,18); 4.57 (d; J=5.8 Hz; 4H.sub.15,17); 3.57 (s; 3H.sub.14); 3.25 (m; 2H.sub.8); 2.30 (t; J=7.4 Hz; 2H.sub.12); 1.62-1.45 (m; 4H.sub.9,11); 1.37-1.21 (m; 2H.sub.10).

[0225] .sup.13C NMR (75 MHz, DMSO) 173.3 (1C.sub.13); 165.1 (1C.sub.6); 161.8 (2.sub.1,5); 142.9 (1C.sub.3); 115.8 (2C.sub.2,4); 64.1 (2C.sub.15,17); 51.2 (1C.sub.14); 38.5 under DMSO peak (1C.sub.8); 33.2 (1C.sub.12); 28.6 (1C.sub.9); 25.9 (1C.sub.11); 24.2 (1C.sub.10).

[0226] HRMS (ESI): calculated neutral mass for C.sub.15H.sub.22N.sub.2O.sub.5 [M]: 310.1529; observed: 310.1526.

Preparation of Methyl 6-(2,6-bis(bromomethyl)pyridin-4-yl)amidohexanoate (6)

[0227] ##STR00033##

[0228] Methyl 6-((2,6-bis(hydroxymethyl)pyridin-4-yl)amidohexanoate (5) (55 mg; 0.177 mmol; 1 eq) was taken up into suspension in anhydrous acetonitrile (10.5 mL) then phosphorus tribromide (50 L; 0.532 mmol; 3.0 eq) was added dropwise. The reaction medium was stirred at 45 C. for 2 h. The solution was cooled to 0 C., neutralized with water (10 mL) and extracted with ethyl acetate (315 mL). The combined organic phases were washed with a saturated solution of sodium chloride, dried over magnesium sulfate and concentrated under reduced pressure. The product was purified by flash chromatography (SiO.sub.2, cyclohexane/ethyl acetate 60:40) in order to give (6) (57 mg; 74%) in the form of a white solid.

[0229] .sup.1H NMR (300 MHz; DMSO) 8.83 (t, J=5.6 Hz; 1H.sub.7); 7.84 (s; 2H.sub.2,4); 4.74 (s; 4H.sub.15,16); 3.57 (s, 3H.sub.14); 3.31-3.20 (m; 2H.sub.8); 2.31 (t; J=7.4 Hz; 2H.sub.12); 1.64-1.45 (m; 4H.sub.9,11); 1.39-1.22 (m, 2H.sub.10).

[0230] .sup.13C NMR (75 MHz, DMSO) 173.3 (1C.sub.13); 163.8 (1C.sub.6); 157.5 (2C.sub.1,5); 144.2 (1C.sub.3); 120.8 (2C.sub.2,4); 51.2 (1C.sub.14); 38.9 under DMSO peak (1C.sub.8); 34.1 (2C.sub.15,16); 33.2 (1C.sub.12); 28.5 (1C.sub.9); 25.9 (1C.sub.11); 24.2 (1C.sub.10).

[0231] HRMS (ESI): calculated neutral mass for C.sub.15H.sub.20Br.sub.2N.sub.2O.sub.3 [M]: 433.9841; observed: 433.9832.

Preparation of 6-(2,6-bis(bromomethyl)pyridin-4-yl)amidohexanoic Acid (7)

[0232] ##STR00034##

[0233] Methyl 6-(2,6-bis(bromomethyl)pyridin-4-yl)amidohexanoate (6) (57 mg; 0.131 mmol; 1.0 eq) was dissolved in tetrahydrofuran (4 mL) and a solution of hydrated lithium hydroxide (8 mg; 0.327 mmol; 2.5 eq) in water (4 mL) was added slowly. The reaction medium was stirred at ambient temperature for 8.5 h. The tetrahydrofuran was evaporated off under reduced pressure and the aqueous residue was treated with an aqueous 1N hydrochloric acid solution and extracted with ethyl acetate (310 mL). The combined organic phases were washed with a saturated solution of sodium chloride, dried over magnesium sulfate and concentrated under reduced pressure. The product was purified by flash chromatography (dichloromethane/methanol, 90:10) in order to give (7) (44 mg; 80%) in the form of a white solid.

[0234] .sup.1H NMR (300 MHz; DMSO) 12.00 (s; 1H.sub.14); 8.83 (t; J=5.5 Hz; 1H.sub.7); 7.84 (s; 2H.sub.2,4); 4.74 (s; 4H.sub.15,17); 3.31-3.21 (m; 2H.sub.8); 2.21 (t; J=7.3 Hz; 2H.sub.12); 1.60-1.46 (m; 4H.sub.9,11); 1.39-1.25 (m; 2H.sub.10).

[0235] .sup.13C NMR (75 MHz; DMSO) 174.4 (1C.sub.13); 163.8 (1C.sub.6); 157.5 (2C.sub.1,5); 144.1 (1C.sub.3); 120.8 (2C.sub.2,4); 39.0 under DMSO peak (1C.sub.8); 34.1 (2C.sub.15,16); 33.6 (1C.sub.12); 28.6 (1C.sub.9); 26.0 (1C.sub.11); 24.2 (1C.sub.10).

[0236] HRMS (ESI): m/z calculated for C.sub.14H.sub.19Br.sub.2N.sub.2O.sub.3 [M+H].sup.+: 420.9757; observed: 420.9752.

Preparation of MMAE 6-(2,6-bis(bromomethyl)pyridin-4-yl)amido-N-hexanamide-valine-citrulline-p-aminobenzoyl Carbamate (8) (8)

[0237] ##STR00035##

[0238] In an inert atmosphere, in darkness and under anhydrous conditions, 6-(2,6-bis(bromomethyl)pyridin-4-yl)amidohexanoic acid (7) (13.2 mg; 0.0313 mmol; 2.28 eq) was dissolved in anhydrous acetonitrile (1.2 mL), then N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline (EEDQ) (21.2 mg; 0.0857 mmol; 6.25 eq) was added. The activation medium was stirred under 25 C. for 1 h 20. A solution of the trifluoroacetic acid salt of MMAE valine-citrulline-p-aminobenzoyl carbamate (17.0 mg; 0.0137 mmol; 1.0 eq), dissolved in anhydrous N,N-dimethylformamide (300 L) in the presence of N,N-diisopropylethylamine (9.4 L; 0.0537 mmol; 3.92 eq), was added to the activation medium. The reaction medium obtained was stirred at 25 C. for 1 h. The mixture was diluted 2-fold with N,N-dimethylformamide and purified by semi-preparative high pressure liquid chromatography (t.sub.R=22.1 min; on the Gilson PLC 2050 system [ARMEN V2 (pump) and ECOM TOYDAD600 (UV detector)], UV detection at 254 nm at 25 C.; Waters XBridge C-18 column; 5 m (250 mm19.00 mm); elution carried out with 0.1% of trifluoroacetic acid (by volume) in water (solvent A), and acetonitrile (solvent B); gradient 20% to 100% of B for 32 min, then 100% of B for 6 min at 17.1 mL/min) in order to give (8) (18.2 mg; 87%) in the form of a white solid.

[0239] .sup.1H NMR (300 MHz, DMSO) (ppm) 10.04-9.95 (m; 1H); 8.94-8.79 (m; 1H); 8.20-8.06 (m; 2H); 7.98-7.87 (m; 1H); 7.84 (s; 2H); 7.81 (s; 1H); 7.70-7.61 (m; 1H); 7.58 (d; J=8.2 Hz; 2H); 7.38-7.11 (m; 6H); 6.07-5.92 (m; 1H); 5.47-5.37 (m; 1H); 5.15-4.96 (m; 1H); 4.73 (s; 4H); 4.54-4.29 (m; 2H); 4.32-4.12 (m; 1H); 4.05-3.92 (m; 1H); 3.30-3.08 (m; 9H); 3.06-2.93 (m; 2H); 2.91-2.77 (m; 2H); 2.24-2.05 (m; 2H); 2.21-2.11 (m; 3H); 2.02-1.88 (m; 1H); 1.60-1.44 (m; 5H); 1.36-1.13 (m; 4H); 1.08-0.93 (m; 6H); 0.93-0.67 (m; 28H).

[0240] HRMS (ESI): m/z calculated for C.sub.72H.sub.111 Br.sub.2N.sub.12O.sub.14 [M+H].sup.+: 1525.6704; observed: 1525.6700.

Example 2: Synthesis of an Antibody-drug Conjugate in Accordance with the Invention

[0241] Code for synthesized product: MF-m906-MMAE (corresponding to formula (Ia), also designated as the antibody-drug conjugate in accordance with the invention below or, in general, ADC).

[0242] Antibody used to produce the antibody-drug conjugate in accordance with the invention: m906.

Preparation of Solutions

[0243] Bioconjugation buffer: Saline buffer 1X, for example phosphate, borate, acetate, glycine, tris(hydroxymethyl)aminomethane, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid in a pH range comprised between 6 and 9, with a final concentration of NaCl comprised between 50 and 300 mM and a final concentration of EDTA comprised between 0.1 and 10 mM.

[0244] m906 at a concentration comprised between 1 and 10 mg/mL in the bioconjugation buffer.

[0245] Reducing agent: Solution of a reducing agent selected from dithiothreitol, -mercaptoethanol, tris(2-carboxyethyl)phosphine hydrochloride, tris(hydroxypropryl)phosphine in a concentration comprised between 0.1 and 10 mM in the bioconjugation buffer.

[0246] Linker solution: compound 8 in a concentration comprised between 0.1 and 10 mM in a mixture of organic solvents selected from dimethylsulfoxide, N,N-dimethylformamide, methanol, tetrahydrofuran, acetonitrile, N,N-dimethylacetamide, dioxane.

Bioconjugation Reaction

[0247] In an inert atmosphere, the reducing agent (4 to 100 eq) was added to m906 in the bioconjugation buffer (1 mg; 1 eq), then it was incubated in its entirety between 15 C. and 40 C. for 0.25 to 3 h, then the solution of compound 8 (4 to 100 eq) was added in an inert atmosphere and the reaction medium was stirred between 15 C. and 40 C. for 0.5 to 15 h. This reaction was duplicated, in parallel, as many times as were necessary in order to obtain the desired final quantity of ADC, i.e. 18 times.

Purification of ADC

[0248] The reaction mixture was purified over PD-10 (GE Healthcare) with PBS Gibco buffer, pH 7.4, as many times as were necessary in order to eliminate the residual chemical reagents, i.e. purified 3 times.

[0249] Result

[0250] The steps described above enabled 12.87 mg of MF-m906-MMAE (71%) to be obtained.

Example 3: Analyses of Antibody-drug Conjugate (MF-m906-MMAE)

HIC (Hydrophobic Interaction Chromatography) Analysis

[0251] Method and Apparatus

[0252] The ADC MF-m906-MMAE was diluted to 1 mg/mL with PBS, pH 7.4, before being filtered over 0.22 m. 50 g of the product was injected onto a MAbPac HIC-Butyl column, 5 m, 4.6100 mm (ThermoScientific), connected to a HPLC Waters Alliance system (e2695) equipped with a PDA (e2998) set for detection at 280 nm. The ADC MF-m906-MMAE was eluted at a rate of 1 mL/min by a gradient from 100% of buffer A (1.5 M ammonium sulfate, 50 mM monobasic sodium phosphate, 5% isopropanol (v/v), pH 7.0) to 20% of buffer B (50 mM monobasic sodium phosphate, 20% isopropanol (v/v), pH 7.0) in 2 minutes, then to 85% of buffer B in buffer A in 30 minutes, then this gradient was maintained for 1 min. The temperature was maintained at 25 C. throughout the separation.

[0253] The results obtained for MF-m906-MMAE are presented in FIG. 1, and in Table 1 below.

TABLE-US-00001 TABLE 1 MF-m906-MMAE DAR DAR DAR DAR DAR DAR 0 1 2 3 4 5 Retention time (min) 8.65 10.87 12.39 18.36 21.47 25.9 Area (%) 0.39 0.09 5.45 16.20 59.69 18.17 mean DAR 3.89

SEC (Size Exclusion Chromatography) Analysis

[0254] Method and Apparatus

[0255] The ADC MF-m906-MMAE was diluted to 1 mg/mL with PBS, pH 7.4, before being filtered over 0.22 m. 50 g of the product was injected onto an AdvanceBio SEC 2.7 m column, 7.8300 mm (Agilent Technologies), connected to a HPLC Waters Alliance system (e2695) equipped with a PDA (2998), set for detection at 280 nm. The ADC MF-m906-MMAE was eluted at a rate of 1 mL/min by an isocratic buffer C (1 mM monobasic sodium phosphate, 155 mM sodium chloride, 3 mM dibasic sodium phosphate, 3 mM sodium azide, pH 7.0) in 24 minutes. The temperature was maintained at 25 C. throughout the separation.

[0256] The results are presented in FIG. 2 and in Table 2 below.

TABLE-US-00002 TABLE 2 MF-m906-MMAE Aggregates Monomers Retention time (min) 4.862 8.253 Area (%) 0.64 99.36

Native MS Analysis (Native Mass Spectrometry)

[0257] Method and Apparatus

[0258] The mass spectrometric analysis was carded out on a Vion IMS Qtof mass spectrometer coupled to an Acquity UPLC H-Class system from Waters (Wilmslow, UK). Before the MS analysis, the samples (20 ug) were desalinated on a BEH SEC 2.1150 mm 300 desalination column by an isocratic gradient (50 mM ammonium acetate, pH 6.5) at 40 L/min. A bypass valve was programmed to allow the solvent to enter the spectrometer between 6.5 and 9.5 min only. The MS data were acquired in positive mode with an ESI source over a m/z range of 500 to 8000 at 1 Hz and analyzed using UNIFI 1.9 software and the MaxEnt algorithm for the deconvolution.

[0259] The results are presented in FIG. 3 and in Table 3 below.

TABLE-US-00003 TABLE 3 MF-m906- MMAE DAR 0 DAR 1 DAR 2 DAR 3 DAR 4 DAR 5 MW (Da).sup.1 n.o..sup.2 n.o..sup.2 152572 153945 155315 156686 Area (%) 0 0 0.7 17.0 74.5 7.7 mean DAR 3.89 .sup.1G0F/G0F glycosylation .sup.2n.o. = not observed [00036]

Example 4: In Vitro Evaluation of Antibody-drug Conjugate MF-m906-MMAE

4.1. Recognition of Tumor Cells by m906: Merkel Cell Carcinoma

Immunohistochemistry on Tumors From Patients

[0260] Method and Apparatus

[0261] Immunohistochemical labelling with a commercial anti-CD56 antibody (123C3, Ventana, Prediluted) was carried out on a cohort of 90 MCC tumors included in a micro array tissue using a benchMark XT platform and following the instructions of the supplier. The expression of CD56 was evaluated by a person skilled in the art using the following semi-quantitative score: 0: absence of expression, 1: weak and/or heterogenous positivity; 2: intense and diffuse positivity.

[0262] Results: In general, 66% of the cases were positives (n=59) with an intense and diffuse expression (score 2) detected in 37% of the cases (n=33).

Flow Cytometry and Immunohistochemistry on the MCC Lines

[0263] Method and Apparatus

[0264] In order to evaluate the expression of CD56 by the cell lines, the tumor cells were incubated in the presence of an anti-CD56 antibody coupled to FITC (BioLegend, clone: HCD56) or of the control isotype in accordance with the indications supplied by the manufacturer. The cells were then washed twice in PBS+1% fetal calf serum, then analysed.

[0265] For the evaluation of the fixation of m906 and trastuzumab (anti-HER2 antibody control) on the tumor cells, immunocytochemical labeling was carried out with the m906 antibody used at 650 ng/mL, or the trastuzumab antibody (Herceptin 150 mg Roche) at 40 ng/mL, then revealed with the aid of a secondary antibody coupled to peroxidase.

Tested Cell Lines

[0266] WaGa (RRID:CVCL_E998). [0267] MS-1 (RRID:CVCL_E995). [0268] PeTa (RRID:CVCL_LC73). [0269] MKL-2 (RRID:CVCL_D027).

[0270] Results:

[0271] WaGa, MS-1, PeTa and MKL-2 cells are Merkel cell carcinoma cell lines (hereinafter MCC). All of the above lines are CD56-positive (FIG. 4).

[0272] SK-BR-3 (RRID:CVCL_0033): HER2-positive breast cancer line, selected as a control because it does not express CD56.

[0273] The fixations of m906 and of trastuzumab (TTZ), included as a control, were tested on the lines by immunohistochemical studies. This analysis demonstrated a fixation of m906 on all of the MCC lines, while no fixation was observed with TTZ (Table 4: confirmation of fixation of m906 on the MCC lines).

TABLE-US-00004 TABLE 4 Cell line WaGa MS-1 PeTa MKL-2 SKBR3 Binding of m906 + + + + Binding of + Trastuzumab + presence of a fixation of the antibody to the target cell revealed by peroxidase, absence of fixation

Imaging Flow Cytometrym906 and Coloration of Intracellular Compartment and Image Acquisition

[0274] Method and Apparatus

[0275] In order to confirm its internalisation into the tumor cells, the antibody m906 was conjugated with Alexa Fluor 750 using the SAIVI Rapid Antibody Labeling Kit (Thermoscientific) following the instructions provided by the manufacturer. The WaGa cells (500 000 cells) were incubated for 30 min at ambient temperature with the conjugate m906-Alexa Fluor 750 in a buffer solution (PBS 1X, 2% fetal calf serum, 0.1% sodium azide). The cells were then fixed with the aid of BD Cytofix/Cytoperm (BD Biosciences) and permeabilized with Permwash (BD Biosciences, diluted to 1/10e) in accordance with the instructions from the manufacturer. The nuclear and lysosomal compartments were labeled with Hoechst 33342 (BD Pharmigen, 1/10000) and an anti-LAMP1 antibody coupled to phycoerythrin-Cyanine 5 (BD Pharmigen, H4A3), respectively. The analysis of the labeled cells was carried out with the aid of an Amnis ImageStreamX Mark II flow cytometry imager (Amnis Corp., part of EMD Millipore, Seattle, WA) equipped with 4 lasers (375 nm, 488 nm, 642 nm and 785 nm (SSC)). The images for the WaGa cells were captured with Inspire imaging flow cytometry software at a magnification of 60X and with an extended depth of field (EDF).

[0276] Compensations were made before each analysis. The cells of interest were identified using the Gradient RMS tool of the Bright Field image (BF: white light). Debris and cellular doublets were excluded from the analysis as a function of the aspect ratio with respect to the zone of the BF image. The surface area and the mean intensity of the intracellular fluorescence (IMF) of the m906 conjugated with Alexa Fluor 750 (channel 12) was evaluated for 2 different incubation times (5 and 30 min) with the aid of the surface mask and cytoplasm mask tools, using the IDEAS v6.2 software. The internalisation score for the m906 conjugated with Alexa Fluor 750 (channel 12) was determined using the internalisation function, allowing the ratio of the intensity of intracellular fluorescence to the intensity of the whole cells to be defined. Cells with internalized antibodies have positive scores. The co-localization assistant using the Bright Detail Similarity (BDS) function was employed in order to quantify the level of co-localization between the anti-LAMP1 antibody conjugated with phycoerythrin-Cyanine 5 (channel 5) and the m906 conjugated with Alexa Fluor 750 (channel 12). The positive score for BDS (n. 1) indicates lysosomal localization for the m906.

[0277] Results

[0278] The m906 anti-CD56 antibody is internalized in the lysosome in WaGa cells expressing CD56 (Table 5). This is a crucial step for the release of drugs by an ADC, which makes m906 a good candidate antibody for the development of an ADC.

TABLE-US-00005 TABLE 5 m906 Intracellular Internal- BDS*.sup.2 Incubation Number fluorescence fluorescence isation score period of cells surface of m906 score for (LAMP1/ (minutes) in focus (MFI*.sup.1) (MFI*.sup.1) m906 m906) 5 8617 930 50046 7.08 1.4 30 5223 580 50464 7.96 1.4 *.sup.1Mean Fluorescence Intensity *.sup.2Bright Detail Similarity

4.2. Cytotoxicity of MF-m906-MMAE on Lines In Vitro

Evaluation of Viability (Proliferation Test)

[0279] Method and Apparatus

[0280] In order to evaluate cell viability, a cytotoxicity test using XTT was carried out. The cells were deposited in a 96-well plate (50 000 cells/well) in 7 replicates. The ADCs MF-m906-MMAE and MF-TTZ-MMAE (ADC control) were added in incremental concentrations. The culture medium acted as a negative control. After 4 days of exposure to the drug, 25 L of reagent XTT was added per well and the absorption was measured at 450 nm after 4 hours of incubation at 37 C. The absorption at 620 nm was used as a reference.

[0281] Results

[0282] The evaluation of the cytotoxicity on cell lines has shown that the ADC MF-m906-MMAE is as cytotoxic as free MMAE (IC.sub.50 between 1-10 nM) for all of the MCC lines. Furthermore, neither the antibody m906 (i.e. non coupled to MMAE), nor the ADC MF-TTZ-MMAE (ADC control) has a cytotoxic effect on these same lines at the lowest effective concentrations tested, demonstrated the absence of intrinsic toxicity of the construct (FIG. 5).

[0283] The evaluation of cytotoxicity on the H69 line (RRID:CVCL_1579), a small cell lung carcinoma cell line, has shown that the ADC MF-m906-MMAE is as cytotoxic as free MMAE (IC.sub.50 between 1-2 nM). Furthermore, neither the antibody m906 (i.e. non coupled to MMAE), nor the ADC MF-TTZ-MMAE (ADC control) has a cytotoxic effect on this line at the lowest effective concentrations tested, demonstrating the absence of intrinsic toxicity of the construct (FIG. 9)

4.3. Confirmation of the Specificity of m906: the Cytotoxic Effect Observed is Dependent on CD56 (FIG. 6)

Plasmids and Lentiviral Transduction

[0284] Method and Apparatus

[0285] Three shRNAs targeting the sequence for CD56 were generated (sequences obtained from Consortium RNAi (A: TRCN0000373085 (SEQ ID NO 9-10)/B: TRCN0000373034 (SEQ ID NO11-12)/C: TRCN0000073460 (SEQ ID NO13-14)) and cloned in a FH1tUTG lentiviral vector, as described above [5]. Note in this construct that the activity of the promoter controlling the transcription of sequences of shRNA can be induced by doxycycline. The lentiviral supernatants were produced in the HEK293T (RRID:CVCL_0063) cells as described above [6-7]. The harvested supernatant was sterilized by filtration (0.45 m) and polybrene was added (1 g/mL) before infection. After 14-20 h of incubation with the supernatants containing the lentiviruses, the target cells were washed then underwent antibiotic selection (puromycin). For invalidation of the expression of CD56 in the tumor lines (knock-down), the cells were exposed to doxycycline for 7 days before analysis.

[0286] Results

[0287] The cytotoxicity induced by MF-m906-MMAE was substantially reduced during CD56 knock-down and this was the case for the three shRNAs (FIG. 6), confirming that the recognition of CD56 by m906 is essential in order to induce a cytotoxicity for MF-m906-MMAE.

Example 5: In Vivo Evaluation of Antibody-drug Conjugate (MF-m906-MMAE)

Therapeutic Performance of m906 in a Xenograft Model of MCC Cell Lines on NOD/SCID Mouse

[0288] Method and Apparatus

Mouse Model

[0289] Twenty 7-week-old female NOD/SCID mice (Janvier Labs) were kept under aseptic conditions. All of the procedures relating to the animals were approved by the local ethics committee (Apafis-10076-2017053015488124 v4). The CD56-positive MCC cell line WaGa [8] was used for tumor induction. The mice, anesthetized with isoflurane, received a subcutaneous injection of 10.sup.7 cells in Matrigel (injection site: back). The tumor size, determined by measuring the width, the length and the height with a calliper and the general condition of the animals were monitored every 2 days throughout the procedure. The tumor volume was determined using the following formula: widthlengthheight/6. When the volume of the tumor reached 50 mm.sup.3, the mice were included in the study and randomly assigned to the experimental or control groups.

Experimental Procedure

[0290] After inclusion, the animals received an intravenous injection of ADC (either MF-m906-MMAE or MF-TTZ-MMAE, 10 mg/kg) or the injection of an equivalent volume of PBS for the control mice. A new injection was made in the event of doubling of the volume of the tumor in the experimental group. The mice were sacrificed 30 days after inclusion or if a critical point was reached (tumor ulceration, loss of 20% of weight or prostration). An autopsy was carried out on the animals and all of the organs were examined macroscopically by a pathologist. The weight and volume of the tumors were evaluated after dissection. Microscopic examination of the tumors, from the lungs and the liver, was carried out in order to detect any potential metastatic progression.

Statistical Analysis

[0291] Continuous data were described using means and limits. Categorical data were described by numbers and percentages of interpretable cases. The combinations were evaluated using Fisher's exact test for the categorical data or by Mann-Whitney or Kruskall Wallis tests for the continuous data. The statistical analyses were carried out using XLStat software (Addinsoft, Paris, France). p<0.05 was considered to be statistically significant.

[0292] Results

[0293] MF-m906-MMAE reduced tumor growth in a murine xenograft model of MCC cell lines.

[0294] The results of the administration of a triple dose of 10 mg/kg of MF-m906-MMAE, MF-TTZ-MMAE or PBS are presented in FIG. 7 (relative tumor volumes) and FIG. 8 (tumor mass at end of study). In the last two groups, used as controls, a similar tumor growth was observed with a mean coefficient for the slope of the growth curve of 105% of the initial tumor volume per day (limits 35-166) and 108% of the initial tumor volume per day (limits 33-318) for the PBS and MF-TTZ-MMAE groups respectively). For the experimental group treated with MF-m906-MMAE, tumor growth was significantly retarded (mean coefficient for the slope of the growth curve of 18% of the initial tumor volume per day (limits 1-53); Kruskal Wallis test: p=0.01). As a consequence, the final median weight of the tumors was reduced in the group treated with MF-m906-MMAE compared with the control animals (FIG. 8) (mean weight 0.7 g (limits: (0.4-1.3) as opposed to 2.03 g (limits: 1.3-3.7) and 1.78 g (limits: 1-2.7), Kruskal Wallis test: p-0.02). After sacrifice, no metastasis was observed either in the experimental group or in the control groups. No signs of MF-m906-MMAE toxicity were observed on the non-tumoral tissues removed during autopsy.

Example 6: Evaluation of the Impact of Glycosylation on Fc Part

1. Products Tested

[0295]

TABLE-US-00006 TABLE 6 Type Reference Description of the product Antibody mAb-001 m906 Antibody mAb-002 Deglycosylated m906: m906 was treated with PNGase F to remove the glycosylation at Asn297 Antibody mAb-003 Deglycosylated m906: Asn297 of the amino acid sequence of m906 was mutated aiming to suppress the glycosylation at Asn297 Antibody L Lorvotuzumab Antibody L-002 Deglycosylated Lorvotuzumab: Lorvotuzumab was treated with PNGase F to remove the glycosylation at Asn297 ADC MIO-001 m906 linked to linker (Pyridine-caproic- (corresponding to ValCitPABcMMAE), i.e. corresponding to the above MF- Formula la m906-MMAE) ADC MIO-002 Deglycosylated MIO-001: MIO-001 was treated with PNGase F to remove the glycosylation at Asn297 ADC MIO-003 Deglycosylated MIO-001: Asn297 of the amino acid sequence of m906 was mutated aiming to suppress the glycosylation at Asn297. ADC LM Lorvotuzumab Mertansine ADC LM-002 Deglycosylated Lorvotuzumab Mertansine: Lorvotuzumab Mertansine was treated with PNGase F to remove the glycosylation at Asp297

2. Binding Specificity to CD56

[0296] 2.1. Material and Methods

[0297] Binding levels of L and mAb-003 have been evaluated on several tumor cell lines: MCC (WaGa, PeTa, MKL-1), SCLC (H69, H209) and prostate cancers (H660) as well as WaGa, PeTa and MKL-1 CD56-knocked out using CRISPR technology.

[0298] The L and mAb-003 were labelled with Alexa Fluor 750 (Alexa Fluor 750 Protein Labelling Kit, Invitrogen) to allow their detection by flow cytometry. The labelled antibodies were incubated with 1.10.sup.5 cells at 4 C. for 30 minutes, then cells were washed twice with PBS supplemented with 1% FCS, and the binding was analysed by Flow cytometry on Cytoflex.

[0299] 2.2. Results and Conclusion

[0300] The results are presented in FIG. 10.

[0301] The mAb-003 presented a strong binding to CD56+ cells without any binding to the CD56 cells (CRISPR cells). L exerted at the same time a lower binding level to CD56+ cells and binding to CD56 cells. These two results could underline a potential side effect that could occur with the unspecific binding of LM to healthy tissues/cells.

[0302] As a conclusion, mAb-003 was a better antibody compared to L to develop a targeted therapy due to a better specificity to tumor cells and no binding to healthy cells.

3. ADC-mediated Cytotoxicity

[0303] 3.1. A. Material and Methods

[0304] To evaluate cell viability and cellular metabolic activity, XTT assays were performed according to standard protocols. WaGa, PeTa, H69, H209, H660 cell lines were plated in 3 replicates in a 96-well plate with 5.10.sup.4 cells/well. LM, MIO-003 and MMAE alone were added in incremental concentrations. Untreated cells were used as viability reference. Cells treated with 1% Triton X100 were used as cell death positive control. After 4 days, 25 L XTT reagent at 1 mg/mL (Alfa Aesar, thermo Fisher) with N-methyl dibenzopyrazine methyl sulfate (PMS) activator (25 M) was added per well and absorbance was measured at 450 nm after 4 h incubation. Absorbance at 620 nm was used as a reference. Three independent experiments have been performed.

[0305] 3.2. Results

[0306] 3.2.1. On CD56+ cells: the results are presented in FIG. 11 and Tables 7 and 8.

TABLE-US-00007 TABLE 7 Merkel Cell WaGa n = 3 PeTa n = 3 Carcinoma cell lines IC50 (pM) IC50 (pM) MIO-003 10280 14921 LM 75610 26942 MMAE alone 3069 1110

TABLE-US-00008 TABLE 8 Small Cell Lung H69 n = 3 H209 n = 3 Cancer cell lines IC50 (pM) IC50 (pM) MIO-003 1949 22144 LM 62625 99985 MMAE alone 887.9 2244

[0307] Tables 7 and 8 show comparative IC50 of MIO-003, LM and MMAE alone on MCC (WaGa and PeTa) and SCLC (H69 and H209) cell lines. IC50 values for MIO-003 ranged from 2 to 22 nM, which is 2 to 32 times more potent than LM. MIO-003 killing effect on CD56+ cells is more important than LM.

[0308] 3.2.2. On CD56 cells (i.e. MCC CRISPR-cell lines, see above): the results are presented in FIG. 12 and Table 9.

TABLE-US-00009 TABLE 9 Merkel Cell WaGa CRISPR MKL-1 CRISPR CD56 Carcinoma CD56 n = 3 n = 3 cell lines IC50 (pM) IC50 (pM) MIO-003 637183 LM 75610 74063 MMAE alone 941.6 855.9

[0309] Table 9 shows comparative IC50 of MIO-003, LM and MMAE alone on CRISPR-MCC cell lines (WaGa and MKL-1).

[0310] MIO-003 is not cytotoxic on CD56 cells even at high doses. In comparison, LM exerts a nonspecific cytotoxicity on CD56 cells at high doses. The killing effect of MIO-003 is driven by target recognition; MIO-003 is more specific towards CD56 than LM.

4. 4. In Vitro Cytotoxicity: Primary NK Cells, Monocytes and Neutrophiles

[0311] 4.1. Material and Methods

[0312] Primary natural killer cells (NK) and monocytes were obtained from blood of healthy adult volunteers at the Etablissement Franais du Sang, according to institutional research protection guidelines agreement N CA-REC-2019-188, Centre Val de Loire, France. Ficoll density centrifugation step (Eurobio) was performed to isolate peripheral blood mononuclear cells (PBMC).

[0313] NK cells were isolated from PBMC by negative selection (NK cell isolation kit human, Miltenyi Biotec) with purity higher than 95%. NK cells were cultured for up to 36 h in RPMI 1640 supplemented with 10% FCS, 1% penicillin and streptomycin, 1% L-glutamine and 100 UI/mL IL-2 at 37 C., 5% CO.sub.2 at 1.10.sup.6 cells/mL.

[0314] Monocytes were isolated from PBMC by positive selection using anti-CD14 MicroBeads (Miltenyi Biotec) according to manufacturer recommendations. Monocytes were cultured in serum-free X-VIVO-15 medium (Lonza) at 1.10.sup.6 cells/mL.

[0315] The neutrophils were isolated directly from whole blood of healthy adult volunteers at the Etablissement Franais du Sang, according to institutional research protection guidelines agreement N CA-REC-2019-188, Centre Val de Loire, France, by negative selection, using the kit (MACSxpress Whole Blood Neutrophil Isolation Kit, MiltenyBiotech). The neutrophils were used directly following their isolation and then resuspended in HBSS buffer (1X) (Gibco 15266355), at 1106 cells/mL.

[0316] NK cells, monocytes or neutrophils were incubated in the presence of MIO-001, MIO-002, MIO-003, LM, mAb-001, mAb-002 or mAb-003 at 37 C. 5% of CO2. Cell mortality induced by the different products was revealed by APC-Annexin V and 7-AAD staining (APC Annexin V Apoptosis Detection Kit with 7-AAD, Biolegend) and was measured by flow cytometry. All products were tested after 4 days treatment on NK cells and 4 h on monocytes and neutrophils.

[0317] 4.2. Results and Conclusion

[0318] 4.2.1 On primary NK cells (comparison MIO-001, MIO-002, mAb-001, mAb-002): the results are presented in FIG. 13.

[0319] MIO-001 and mAb-001 exerted a killing effect on primary NK cells after 4 h. However, deglycosylation (i.e. mAb-002 and MIO-002) prevented from ADCC killing effect and unexpected killing of NK cells, as illustrated in FIG. 13. Interestingly, even if NK cells have CD56 proteins at their surface, the deglycosylated ADC (i.e. MIO-002) had no killing effect on NK cells (FIG. 13A) but maintained its ADC-mediated cytotoxicity on tumor cell lines (FIG. 13B).

[0320] The deglycosylation prevented from unexpected killing effect on healthy cells while maintaining anti-tumor activity on tumor cell line.

[0321] 4.2.2. On primary NK cells (comparison of all products of Table 6), after 4 h and 4 days of incubation: the results are presented in FIGS. 14-17.

[0322] All experiments have been conducted with n3 donors. All the mAbs (L, L-002, mAb-001, mAb-002 and mAb-003) and all the ADCs (LM, LM-002, MIO-001, MIO-002 and MIO-003) were compared after 4 hours or 4 days of incubation for their killing effect on primary NK cells.

[0323] After 4 h of incubation, ADCC killing effect was observed either with L (antibody) and LM (ADC). In comparison, the effect of the wild type version of mAb-001 and MIO-001 presented a lower killing effect towards NK cells. Deglycosylation using PNGase F or via mutation allowed to reduce drastically the killing effect of the antibody and the ADC (FIGS. 14 and 15).

[0324] After 4 days of incubation, ADCC killing effect was still observed either with L (40%) and L-002 (20%) (antibodies) (FIG. 16). Killing effect was also observed using mAb-001 and was even greater at high concentration (20 to 60% of killing effect). Deglycosylation of the mAb using PNGase F or via mutation allowed reducing drastically the ADCC killing effect (mAb-002 and mAb-003) on primary NK cells.

[0325] Regarding the ADC killing effect (FIG. 17), we observed first that MIO ADC (even wild type MIO-001) exerted a lower killing effect than LM. All deglycosylated versions of the ADC presented a lower killing effect than the glycosylated ones; LM-002, MIO-002 and MIO-003 exerted a killing effect below 20%.

[0326] Deglycosylation using PNGase F or via mutation allowed to reduce drastically the killing effect of all antibodies and ADCs.

[0327] 4.2.3. On neutrophils, after 4 h of incubation: the results are presented in FIGS. 18-19.

[0328] The killing effect towards neutrophils, which are CD56 negative cells, was about 15% for L and below 10% for mAb-001 at high concentration (FIG. 18). Deglycosylation using PNGase F or via mutation allowed to reduce the toxicity of the antibodies towards neutrophils.

[0329] Same results were observed with the ADCs (FIG. 19). LM presented a killing effect around 15% towards neutrophils. MIO-001, MIO-002 and MIO-003 exerted similar killing effect around 5%, thus reducing the risk of neutropenia related to non-specific binding.

[0330] 4.2.4. On monocytes, after 4 h of incubation: the results are presented in FIGS. 20-21.

[0331] All the antibodies presented negligible effect on monocytes, bellow 5% (FIG. 20).

[0332] LM exerted a killing effect on monocytes (FIG. 21) from 5 to 10% at highest concentration. MIO-001, MIO-002, MIO-003 and LM-002 exerted negligible killing effect on monocytes.

[0333] The deglycosylation using PNGase F or via mutation prevented from killing effect on monocytes.

5. Bystander Effect

[0334] Cleavable linker involving MMAE are particularly interesting in the context of solid tumor due to the capability to exert bystander effect; after first internalisation and apoptosis phenomena in the targeted tumor cell, free active MMAE is able to reach other tumor cells in the microenvironment and potentialize the global activity even in case of heterogeneous tumors.

[0335] Bystander effect of MIO-003 has been experimentally demonstrated by cell cycle analysis after coculture of CD56+ and CD56 cells.

[0336] 5.1. Material and Methods

[0337] A co-culture of WaGa CD56-positive, i.e. CD56 wild-type (WT) cells, and WaGa CD56-negative i.e. WaGa CD56 knockout (KO) cells, separated by an insert with 0.4 m pore-size (Corning). In 6-well cell culture plates, 1.10.sup.6 CD56-negative cells per well were seed in the bottom chamber and the same number of CD56-positive cells were seed in the upper chamber.

[0338] Cells were incubated with MIO-003 at 5 or 50 nM for 6 days, and untreated cells were used as control. Cell cycle analysis was performed on cells in the upper and in the bottom inserts separately. Cells were fixed using ice-cold ethanol (90%) for at least 1 h followed by treatment with a propidium iodide solution (PBS supplemented with 1% FCS, 0.1 mg/mL propidium iodide, and 0.1 mg/mL RNAse A) for 15 minutes. Cell cycle was then analyzed by flow cytometry on Cytoflex.

[0339] 5.2. Results and Conclusion

[0340] G2/M arrest and a significant increase of the cells at apoptotic sub-G1 phase induced and MIO-003 (FIG. 22) was shown in a concentration and time dependent manner in MCC cell lines using cell viability assays and flow cytometry assays for cell cycle analysis, illustrating the fact that MIO-003 exerted a bystander effect.

6. Dose Ranging Efficacy In Vivo

[0341] 6.1. Material and Methods

[0342] Xenograft mice model: eighteen 7-week-old females NOD/SCID (Janvier Labs) mice were maintained under aseptic conditions. All animal procedures were approved by local ethics committee (Apafis #26772-2020072715262737 v2). WaGa cell line was used for tumor induction, as previously described in [9]. Mice received one subcutaneous injection of 1.10.sup.7 WaGa cells with 10% matrigel on the back. General state and weight were monitored twice a week during the procedure. Tumor volume was measured with a caliper and tumor volume was calculated according to the formula: /6widthlengthheight. When tumor volume reached 150 mm.sup.3, mice (n=6/group) randomly received intravenous injections into the tail vein of an ADC (MIO-001, MIO-002, MIO-003 or LM) at 5 mg/kg or a volume-equivalent injection of PBS twice a week. Mice were sacrificed 30 days after inclusion or when reaching one predefined endpoint (tumor ulceration, weight loss >20%, or prostration). Tumors, heart, lungs, spleen, and liver were removed for macroscopic examination. After dissection, tumor weight and volume were assessed. To detect metastatic spread and potential toxicity of ADCs, whole tumors and organs (heart, lungs, spleen, and liver), were formalin-fixed and paraffin embedded for microscopic evaluation to evaluate architecture changes, inflammatory infiltrate, necrosis, vascular changes, fibrosis and liver steatosis.

[0343] 6.2. Results and Conclusion

[0344] Stagnation of the tumor was observed after treatment with MIO-001 (FIG. 23). All other groups receiving ADCs exerted frank tumor regression. The final weight of residual tumors is given in Table 10.

TABLE-US-00010 TABLE 10 MIO- MF-L- MIO- mAb-003 MIO- Groups PBS 002 LM MMAE 003 mertansine 001 Mean 3.05 0.00 0.03 0.08 0.03 0.00 0.23 weight (g)
Table 10: mean end tumor weight (g) for each group after administration of products at 5 mg/kg twice a week.
MF-L-MMAE corresponds to L conjugated to MMAE with 6-(2,6-bis(bromomethyl)pyridin-4-yl)amido-N-hexanamide-valine-citrulline-p-aminobenzoyl carbamate; mAb-003 mertansine corresponds to mAb-003 conjugated to mertansine via lysine conjugation.

[0345] During the experiment, the mice weight was monitored; a mild loss of weight was observed in the group receiving MIO-002 but no weight loss in the other groups.

[0346] The best efficacy results were observed in groups treated with MIO-002, MIO-003 and LM, with same mean end point tumor weight of 101.03 g.

7. Maximal Tolerated Dose Evaluation on Tumor Bearing Mice

[0347] 7.1. Material and Methods

[0348] Xenograft mice model: eighteen 7-week-old females NOD/SCID (Janvier Labs) mice were maintained under aseptic conditions. All animal procedures were approved by local ethics committee (Apafis #26772-2020072715262737 v2). WaGa cell line was used for tumor induction, as previously described in [9]. Mice received one subcutaneous injection of 1.10.sup.7 WaGa cells with 10% matrigel on the back. General state and weight were monitored twice a week during the procedure. Tumor volume was measured with a caliper and tumor volume was calculated according to the formula: /6widthlengthheight. When tumor volume reached 150 mm.sup.3, mice were randomly assigned to the different groups. After inclusion, animals (n=4/group) received single intravenous injections into the tail vein of MIO-003 at several doses 10, 30, 50 and 70 mg/kg or a volume-equivalent injection of PBS twice a week. Mice were sacrificed 30 days after inclusion or when reaching one predefined endpoint (tumor ulceration, weight loss >20%, or prostration). Tumors, heart, lungs, spleen, and liver were removed for macroscopic examination. After dissection, tumor weight and volume were assessed. To detect metastatic spread and potential toxicity of ADCs, whole tumors and organs (heart, lungs, spleen, and liver), were formalin-fixed and paraffin embedded for microscopic evaluation to evaluate architecture changes, inflammatory infiltrate, necrosis, vascular changes, fibrosis and liver steatosis.

[0349] 7.2. Results and Conclusion

[0350] After a single IV injection, using MIO-003, we observed tumor growth stagnation for the dose of 10 mg/kg and frank regression of the tumor for all the other doses (i.e. from 30 mg/kg to 70 mg/kg), without re-growth of the tumors (FIGS. 26 and 28). At 70 mg/kg, mice weight variation exceeded 20% which was our ethical end point (FIG. 27).

Sequence Listing

[0351]

TABLE-US-00011 TABLE11 Sequencenumber Sequencetype Aminoacidsequence SEQIDNO:1 CDR1ofthelight QSLLHSNGYN chainofm906 SEQIDNO:2 CDR2ofthelight YLG chainofm906 SEQIDNO:3 CDR3ofthelight CMQSLQTPWT chainofm906 SEQIDNO:4 CDR1oftheheavy GGTFTGYYMHW chainofm906 SEQIDNO:5 CDR2oftheheavy NSGGTNYAQ chainofm906 SEQIDNO:6 CDR3oftheheavy LSSGYSGYFDYWGQG chainofm906 SEQIDNO:7 Lightchainofm906 DVVMTQSPLSLPVTPGEPASIS CRSSQSLLHSNGYNFLDWYLQ KPGQSPQLLIYLGSNRASGVP DRFSGSGSGTDFTLKISRVEA DDVGVYYCMQSLQTPWTFGH GTKVEIKRTVAAPSVFIFPPSD EQLKSGTASVVCLLNNFYPRE AKVQWKVDNALQSGNSQESV TEQDSKDSTYSLSSTLTLSKAD YEKHKVYACEVTHQGLSSPVT KSFNRGEC SEQIDNO:8 Heavychainof EVQLVQSGAEVKKPGSSVKVS m906 CKASGGTFTGYYMHWVRQAP GQGLEWMGWINPNSGGTNYA QKFQGRVTMTRDTSISTAYME LSRLRSDDTAVYYCARDLSSG YSGYFDYWGQGTLVTVSSAST KGPSVFPLAPSSKSTSGGTAA LGCLVKDYFPEPVTVSWNSGA LTSGVHTFPAVLQSSGLYSLSS VVTVPSSSLGTQTYICNVNHKP SNTKVDKKVEPKSCDKTHTCP PCPAPELLGGPSVFLFPPKPK DTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKT KPREEQYNSTYRVVSVLTVLH QDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLP PSRDELTKNQVSLTCLVKGFY PSDIAVEWESNGQPENNYKTT PPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNH YTQKSLSLSPGK SEQIDNO:9 RNAi TCCCAGCGTTGGAGAGTCCA TRCN0000373085 AATTCTCGAGAATTTGGACTC Forward TCCAACGCTTTTTTCGGG SEQIDNO:10 RNAi AAAAAAGCGTTGGAGAGTCC TRCN0000373085 AAATTCTCGAGAATTTGGACT Reverse CTCCAACGCT SEQIDNO:11 RNAi TCCCCGTTCCCTGAAACCGT TRCN0000373034 TAAACTCGAGTTTAACGGTTT Forward CAGGGAACGTTTTT SEQIDNO:12 RNAi CGGGAAAAACGTTCCCTGAA TRCN0000373034 ACCGTTAAACTCGAGTTTAAC Reverse GGTTTCAGGGAACG SEQIDNO:13 RNAi TCCCCATGTACCTTGAAGTG TRCN0000073460 CAATCTCGAGATTGCACTTCA Forward AGGTACATGTTTTT SEQIDNO:14 RNAi CGGGAAAAACATGTACCTTG TRCN0000073460 AAGTGCAATCTCGAGATTGC Reverse ACTTCAAGGTACATG SEQIDNO:15 Variabledomainof DVVMTQSPLSLPVTPGEPASIS thelightchainof CRSSQSLLHSNGYNFLDWYLQ m906 KPGQSPQLLIYLGSNRASGVP DRFSGSGSGTDFTLKISRVEA DDVGVYYCMQSLQTPWTFGH GTKVEIKR SEQIDNO:16 Variabledomainof EVQLVQSGAEVKKPGSSVKVS theheavychainof CKASGGTFTGYYMHWVRQAP m906 GQGLEWMGWINPNSGGTNYA QKFQGRVTMTRDTSISTAYME LSRLRSDDTAVYYCARDLSSG YSGYFDYWGQGTLVTVS SEQIDNO:17 Heavychainof EVQLVQSGAEVKKPGSSVKVSCK m906(N297A) ASGGTFTGYYMHWVRQAPGQGL EWMGWINPNSGGTNYAQKFQGR VTMTRDTSISTAYMELSRLRSDDT AVYYCARDLSSGYSGYFDYWGQG TLVTVSSASTKGPSVFPLAPSSKS TSGGTAALGCLVKDYFPEPVTVS WNSGALTSGVHTFPAVLQSSGLY SLSSVVTVPSSSLGTQTYICNVNH KPSNTKVDKKVEPKSCDKTHTCPP CPAPELLGGPSVFLFPPKPKDYLNI SRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYAST YRVVSVLTVLHQDWLNGKEYKCK VSNKALPAPIEKTISKAKGQPREPQ VYTLPPSRDELTKNQVSLTCLVKG FYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQ QGNVFSCSVMHEALHNHYTQKSLSL SPGK

References Cited in the Format [Reference Number]

[0352] 1. Feng, Y. et al., MAbs, May-Jun; 8(4): 799-810, 2016 [0353] 2. Goyon, A et al., J. Chromatogr. B, 1065-1066, 35-43, 2017 [0354] 3. Fekete, S. et al., J. Pharm. Biomed. Anal., 130, 3-18, 2016 [0355] 4. Barran, P. et al., EuPA Open Proteomics, 11, 23-27, 2016 [0356] 5. Houben, R. et al., International journal of Cancer, 130, 847-856, 2012 [0357] 6. Angermeyer S. and al., J. Invest. Dermatol., 133(8):2059-2064.2013 [0358] 7. Aragaki M. et al., Biochem. Biophys. Res. Commun., 368(4):923-929, 2008 [0359] 8. Houben R, et al., J. Virol.; 84(14):7064-7072, 2010 [0360] 9. Esnault et al., British Journal of Dermatology, 2021, DOI 10.1111/bjd.20770 [0361] 10. M H Tao, S L Morrison, J Immunol 143, 2595-2601 (1989) [0362] 11. M R Walker, J Lund, K M Thompson, R Jefferis, Biochem J 259, 347-353 (1989) [0363] 12. L C Simmons, et al., J Immunol Methods 263, 133-147 (2002) [0364] 13. Katsuyoshi Masuda, Yoshiki Yamaguchi, Noriko Takahashi, Royston Jefferis, Koichi Kato, FEBS Letters, Volume 584, Issue 15, 2010, Pages 3474-3479, ISSN 0014-5793, https://doi.org/10.1016/j.febslet.2010.07.004