VARIANTS OF SAC7D AND THEIR USE IN CANCER THERAPY

Abstract

The invention relates to variants of OB-fold proteins, hi particular of the Sac7d family that bind PD-L1 or HSP110 and are able to be used alone or in combination for cancer treatment.

Claims

1. A polypeptide comprising a variant of a member of a Sac7d family binding to human PD-L1 and inhibiting the liaison of PD-L1 with PD1, wherein the variant comprises from 4 to 20 mutated residues in an interface of binding of the member of the Sac7d family to its natural ligand, wherein the mutated residues in the interface of binding of the member of the Sac7d family to its natural ligand are selected from V2, K3, K5, K7, Y8, K9, G10, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, D36, N37, G38, K39, T40, A44, S46, E47, K48, D49, A50 and P51 of Sac7d.

2. The polypeptide of claim 1, wherein the variant comprises the Y8M, V26L, S31L, R42L and A44F mutations, or the Y8I, V26I, S31L, R42M, and A44L mutations, with numbering corresponding to a position in the Sac7d sequence of SEQ ID NO: 1.

3. The polypeptide of claim 1 or 2, wherein the variant comprises the Y8M, W24T, V26L, M29A, S31L, T33R, R42L and A44F mutations, or Y8I, W24R, V26I, M29Y, S31L, T33K, R42M, and A44L mutations with the numbering corresponding to a position in the Sac7d sequence of SEQ ID NO: 1, wherein the variant also binds to mouse PD-L1.

4. The polypeptide of claim 1, wherein the variant further comprises at least one mutation selected from D16E, N37Q and M57L, with the numbering corresponding to a position in the Sac7d sequence of SEQ ID NO: 1.

5. (canceled)

6. The polypeptide of claim 1, which is selected from Sac7d from Sulfolobus acidocaldarius, Sac7e from Sulfolobus acidocaldarius, SSo7d from Sulfolobus solfataricus, Ssh7b from Sulfolobus shibatae, Ssh7a from Sulfolobus shibatae, DBP7 from Sulfolobus tokodaii, Sis7a from Sulfolobus islandicus, Mse7 from Metallosphaera sedula, Mcu7 from Metallosphaera cuprina, Aho7a from Acidianus hospitalis, Aho7b from Acidianus hospitalis, Aho7c from Acidianus hospitalis and Sto7 from Sulfurisphaera tokodaii.

7. The polypeptide of claim 1 comprising a sequence selected from SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 60, SEQ ID NO: 61, SEQ ID NO: 62, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59, SEQ ID NO: 39, SEQ ID NO: 40, SEQ ID NO: 41, SEQ ID NO: 36, SEQ ID NO: 37, SEQ ID NO: 38, and amino acids 1-57 of these sequences.

8. (canceled)

9. (canceled)

10. The polypeptide of claim 1 comprising SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, or SEQ ID NO: 23 or amino acids 2-54 of these sequences.

11. (canceled)

12. The polypeptide of claim 1 comprising a sequence selected from SEQ ID NO: 71, SEQ ID NO: 68, SEQ ID NO: 50, SEQ ID NO: 47, and amino acids 1-54 of these sequences.

13. (canceled)

14. The polypeptide of claim 1, wherein the variant of the member of the Sac7d family binding to human PD-L1 is conjugated to an organic molecule.

15. The polypeptide of claim 1, wherein the variant of the member of the Sac7d family binding to human PD-L1 is conjugated to another polypeptide.

16. The polypeptide of claim 15, wherein the other protein is a variant of the a Sac7d family that binds to HSP110 or to EGFR.

17. The polypeptide of claim 16, wherein the variant of the Sac7d family that binds to HSP110 or to EGFR comprises a polypeptide selected from SEQ ID NO: 83, SEQ ID NO: 80, SEQ ID NO: 75, SEQ ID NO: 72, SEQ ID NO: 73, SEQ ID NO: 74, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28 or SEQ ID NO: 29, and amino acids 1-57 of these sequences.

18. (canceled)

19. A nucleic acid molecule coding for the polypeptide of claim 1.

20. A pharmaceutical composition comprising the polypeptide of claim 1 or a nucleic acid molecule coding for the polypeptide, or the nucleic acid of claim 19, and a pharmaceutically acceptable carrier

21. A method for producing the polypeptide of claim 1 comprising: (a) culturing a cell culture wherein cells have been transformed by a nucleic acid molecule coding for the polypeptide; and and (b) recovering the polypeptide.

22. (canceled)

23. A method for treating cancer comprising administering a polypeptide of claim 1 to a subject in need thereof.

24. The method of claim 23, wherein the peptide is administered in combination with chemotherapy or treatment with CAR-T cells.

25. A method for treating cancer comprising simultaneously, separately, or sequentially administering a composition comprising a polypeptide of claim 1 and a chemotherapy agent or CAR-T cells.

Description

DESCRIPTION OF THE FIGURES

[0256] FIG. 1: alignment of proteins of the Sac7d family

[0257] FIG. 2: binding of B11 to both human (upper curve) and mouse (lower curve) proteins.

[0258] FIG. 3: inhibition of binding of PD1 and PD-L1 using a variant (B11) protein binding to PD-L1.

[0259] FIG. 4: inhibition of binding of PD1 and PD-L1 using a variant (B11) protein binding to PD-L1, alone (triangles) or as a dimer (reverse triangles), Pembrolizumab (squares) and atezolizumab (circles) were used as controls.

[0260] FIG. 5: mean tumor weight (mg) measured in the different experimental groups at the end of study. NF1=anti-HSP110 variant NF2: anti-PD-L1 variant. NF1-NF2: dimer of both variants.

[0261] FIG. 6: Data analysis of chicken CD3 expression in tumor.

[0262] FIG. 7: data analysis of chicken MMD expression in tumor.

[0263] FIG. 8: number of dead and surviving embryos at the end of the study.

[0264] FIG. 9: mean tumor weight (mg) measured in the different experimental groups after 10 days of treatment on the CAM. Mean values±SEM of tumor weight measured for each experimental group.

[0265] FIG. 10: mean tumor weight (mg) measured in the different experimental groups after 10 days of treatment on the CAM. Mean values±SEM of tumor weight measured for each experimental group.

[0266] FIG. 11: mean tumor weight (mg) measured in the different experimental groups at the end of study. Mean values±SEM of tumor weight (mg).

[0267] FIG. 12: Relative Quantity of chicken CD3 Expression (±SEM) in Tumor per Group.

[0268] FIG. 13: data analysis of chicken MMD expression in tumor. Relative Quantity of chicken MMD Expression (±SEM) in Tumor per Group.

[0269] FIG. 14: representation of PDL1 structure, with its two immunoglobulin-like domains (Ig like V type (aa19-aa127)) and (Ig like C2 type (aa133-aa225)), in complex from right to left with PD1 (PDB 4ZQK, A), Atezolizumab (PDB 5XXY, B), Avelumab (PDB 5GRJ, C) and Durvalumab (PDB 5XJ4, D).

EXAMPLES

Example 1. Characterization of a Variant of Sac7d Binding to PD-L1

[0270] Efficacy of the PDL1 neutralizing Nanofitins was evaluated using the PD1/PDL1 blockade bioassay from Promega. The assay consists of two genetically engineered cell lines provided in a thaw and use format: Jurkat T cells expressing human PD-1 and a luciferase reporter driven by an NFAT response element (PD-1 Effector Cells) and CHO-K1 cells expressing human PD-L1 and an engineered cell surface protein designed to activate cognate TCRs in an antigen-independent manner (PD-L1 aAPC/CHO-K1 Cells). When the two cell types are co-cultured, the PD-1/PD-L1 interaction inhibits TCR signaling and NFAT-RE-mediated luminescence. Neutralization of the PD11PDL1 interaction releases the inhibitory signal and results in TCR activation and NFAT-RE-mediated luminescence.

[0271] Briefly, 0.5 mL of thaw-and-use PD-L1 aAPC/CHO-K1 cells suspension was transfered to 14.5 mL of cell recovery medium (90% Ham's F12/10% FBS) and mixed well by gently inverting 1-2 times. 100 μL of the cell suspension were dispensed to each of the inner 60 wells of two 96-well, white, flat-bottom assay plates. 100 μL of cell recovery medium were added to each of the outside wells of the assay plates. The assay plates was then covered with a lid, and incubated for 16-20 hours in a 37° C. 5% CO2 incubator. After overnight incubation, the medium was removed by flicking the plate. 40 μL of the test item concentration ranges and 40 μL of the thaw and use PD-1 effector cells suspension, all prepared in assay buffer (99% RPMI1640/1% FBS), were immediately added to the inner wells. Subsequently, 80 μL of assay buffer were added to the outside wells and the plate was covered and incubated for six hours in a 37° C., 5% CO2 incubator. After the incubation period, the plate was allowed to equilibrate to ambient temperature for 5-10 minutes. 80 μL of Bio-Glo™ Reagent were added to the inner 60 wells of the assay plates and the plate was allowed to incubate at ambient temperature for 5-30 minutes. Luminescence was finally measured and data analysis was performed on GraphPad Prism software VB.

[0272] A Sac7d variant (B11, SEQ ID NO. 21) was found to inhibit PD1/PD-L1 binding (FIG. 3). A correlation between affinity and efficacy of neutralization was also observed (not shown). A dimer of B11 was also shown to be as efficient as controls (FIG. 4). B11 is also named NF2 in the examples.

[0273] This variant was also found to interact with both the human and mouse PD-L1 proteins (FIG. 2). This property is of particular interest as such species crossing is not observed often for PD-L1 binders, and is useful for pre-clinical studies. Affinity of B11 to PD-L1 proteins (Surface Plasmon Resonance) is about K.sub.D=1.5×10.sup.−8 M (human) and K.sub.D=3.2×10.sup.−9 M (mouse).

[0274] Maturation of the protein made it possible to determine that the key amino acids that need to be present for obtaining binding to human PD-L1 are 18, 126, L31, M42 and L44. R24, Y29, K33 are further needed for binding to mouse PD-L1.

Example 2. Characterization of a Variant of Sac7d Binding to HSP110

[0275] A Sac7d variant (A-C2, SEQ ID NO: 27) was found to bind to HSP110 and to inhibit

[0276] HSP110-Stat3 binding.

[0277] A-C2 favored expansion of anti-tumoral M1 macrophages. A-C2 is also named NF1 in the examples.

[0278] Maturation of the protein made it possible to determine that the key amino acids that need to be present for obtaining binding to human PD-L1 are W8, R9, W21, D22, W24, Y31, H33, M40 and K44.

Example 3. In Vivo Experiments—Anti-HSP110 Variant Alone or in Combination with Anti-PD-L1 Variant

Material and Methods

[0279] All in vivo experiments were performed using the CAM technology, corresponding to the graft of cancer cells onto the chorioallantoic membrane (CAM) of chick embryos (see Kroiss et al, Oncogene. 2015 May 28;34(22):2846-55; Green et al, PLoS One. 2009 Aug. 21;4(8):e6713).

[0280] MDA-MB-231 human breast carcinoma cell line were grafted on the CAM of 9-days old chick embryos, with treatments between days 10 and 17 (1 to 8 days prior to grafting, followed by analysis for toxicity and tumor growth inhibition.

[0281] In brief, fertilized White Leghorn eggs were incubated at 37.5° C. with 50% relative humidity for 9 days. At that moment (E9), the chorioallantoic membrane (CAM) was dropped down by drilling a small hole through the eggshell into the air sac, and a 1 cm.sup.2 window was cut in the eggshell above the CAM. At least 20 eggs (depending on embryo surviving rate after 9 days of development, there could be more than 20 eggs per group) were used for each group. An inoculum of 1.10.sup.8 MDA-MB-231 cells was added onto the CAM of each egg

[0282] Keytruda® (pembrolizumab) was used as the reference compound. Six experimental groups as described in Table 2.

TABLE-US-00028 TABLE 2 Study groups Group Description Compound Dose (mg/kg) 1 Neg. Ctrl. (Vehicle) PBS — 2 Pos. Ctrl. Keytruda ® 2 3 Exp. Cpd. 1 [1] A-C2 (NF1) 2 4 Exp. Cpd. 1 [2] A-C2 20 5 Exp. Cpd. 2 [1] A-C2-B11 2 6 Exp. Cpd. 2 [2] A-C2-B11 20

[0283] On day E10, tumors began to be detectable. Treatments corresponding to different groups are detailed in Table 3. After each treatment, eggs were individually checked daily for death or visible abnormalities (visual checks).

TABLE-US-00029 TABLE 3 Description of treatment groups injected [C] Final (in 100 [C] Dosing Group Description Treatment μl/egg) (in ovo) Regimen 1 Neg. Ctrl. PBS — — E10, E12, E14, E15, E17 2 Keytruda Keytruda ® 0.12 mg/ 2 mg/kg E10, E12, 100 μl E14, E15, E17 3 NF1 - NF1 0.12 mg/ 2 mg/kg E10, E12,  2 mg/kg 100 μl E14, E15, E17 4 NF1 - NF1  1.2 mg/ 20 mg/kg  E10, E12, 20 mg/kg 100 μl E14, E15, E17 5 NF1-NF2 - NF1-NF2 0.12 mg/ 2 mg/kg E10, E12,  2 mg/kg 100 μl E14, E15, E17 6 NF1-NF2 - NF1-NF2  1.2 mg/ 20 mg/kg  E10, E12, 20 mg/kg 100 μl E14, E15, E17

[0284] Quantitative Evaluation of Tumor Growth

[0285] On day E18, the upper portion of the CAM (with tumor) was removed from all viable embryos with tumors, washed by PBS buffer and then directly transferred in PFA (fixation for 48 hrs.). After that. tumors were carefully cut away from normal CAM tissue and weighed. A one-way ANOVA analysis with post-tests is performed on the data.

[0286] Quantitative Evaluation of immune Cell Infiltration

[0287] On day E18, 6 tumor samples per group were collected to evaluate the infiltration of immune cells. Each tumor sample was removed, washed by PBS buffer and then directly transferred in 4% PFA (fixation for 48 hr). After that, genomic RNA was extracted from fixed tumor (commercial kit) and analyzed by RT-qPCR with specific primers for chicken CD3 and MMD sequences.

[0288] For all points done in qPCR, expression of GAPDH is also analyzed, as reference gene expression, and used to normalize immune biomarker expression between samples. A one-way ANOVA analysis with post-tests is done on the data

[0289] Quantitative Evaluation of Embryonic Toxicity

[0290] Embryonic viability was checked daily. The number of dead embryos was also counted on E18, in combination with the observation of eventual visible gross abnormalities, to evaluate treatment-induced embryo toxicity.

[0291] Significance of Statistical Analysis

[0292] For all analyses, statistical difference between groups are made visible on graphs by the presence of stars with the following meaning:

[0293] No star: No statistical difference (p value>0.05);

[0294] *: 0.05≥p-value>70.01;

[0295] **: 0.01≥p-value>0.001;

[0296] ***: 0.001≥p-value.

Results

[0297] Quantitative Evaluation of Tumor Growth

[0298] FIG. 5 presents the mean tumor weight (mg) measured in the different experimental groups at the end of study.

[0299] Quantitative Evaluation of Immune Cell Infiltration

[0300] Chicken CD3 expression in tumor

[0301] FIG. 6 present data analysis of data analysis of chicken CD3 expression in tumor.

[0302] Chicken MMD Expression in tumor

[0303] FIG. 7 present data analysis of chicken MMD expression in tumor.

[0304] Quantitative Evaluation of Embryonic Toxicity

[0305] FIG. 8 present the number of dead and surviving embryos at the end of the study.

Conclusion

[0306] The aim of this project was to evaluate the toxicity and efficacy of an anti-HSP110 (NF1, A-C2) alone or as a dimer with an anti-PD-L1 (NF2, B11) against tumors initiated from MDA-MB-231 human breast carcinoma cell line in a CAM Model. Keytruda® was used as the Reference compound.

[0307] In terms of tumor growth, when compared to the Negative Control, both NF1 and NF1-NF2 compounds induced a significant tumor growth inhibition at the high dose (20 mg/kg). The effect of NF1-NF2 compound is dose dependent, its performance at 20 mg/kg is evidently better than at 2 mg/kg.

[0308] In terms of immune cells filtration in tumor, an increase tendency of CD3 positive cells infiltration was observed in presence of Keytruda; an increase tendency of MMD positive cells was observed in presence of the tested compounds, with a dose effect for NF1 alone. All these increases are not statistically significant.

[0309] Finally, treatments with both NF1 and NF1-NF2 compounds did not induce any evident embryonic toxicity at all tested doses when compared to Negative Control.

Example 4. Study of Efficacy of NF-B11 Compound on Breast Carcinoma Tumors Initiated from MA-MB-231 Cell Lines

[0310] The CAM model as described in example 3 was used

[0311] Treatments

[0312] At day 10 (E10), tumors began to be detectable. They were then treated during 10 days, every two days (E11, E13, E15, E17) by dropping 100 μl of vehicle (in PBS), Ref Compound (Tecentriq/Atezolizumab), and NF-B11 at two different doses onto the tumor (see Table 4 for concentration).

TABLE-US-00030 TABLE 4 Experiments Group description Molecule Name Concentration Group 1 Negative Ctrl Neg Ctrl PBS (vehicle) Group 2 Positive Ctrl Tecentriq 1.02 μg/mL (Ref. Compound) Group 3 Exp. Group 2 NF-B11 10XIC50 41 μg/mL Group 4 Exp. Group 3 NF-B11 100XIC50 410 μg/mL

Results

[0313] Tumors Growth Table 5 presents the mean values of the tumor weights for the different experimental groups at E18.

TABLE-US-00031 TABLE 5 Mean value, SD, SEM and p-value of tumor weight (mg) for each experimental group Tumor weight % of tumor n (mg) SD SEM regression Neg Ctrl 12 12.19 6.88 1.98 N/A Tecentriq 12 9.92 3.34 0.96 18.62 NF-B11 12 9.25 3.72 1.07 24.10 10XIC50 NF-B11 11 6.45 2.44 0.73 47.06 100XIC50

[0314] FIG. 9 presents the mean tumor weight (mg) measured in the different experimental groups after 10 days of treatment on the CAM.

[0315] Toxicity Table 6 presents the number of dead and surviving embryos after 10 days of treatment in the different experimental groups.

TABLE-US-00032 TABLE 6 Number of dead and surviving embryo for each experimental group. Total Alive Dead % alive % dead Neg Ctrl 16 11 5 69 31 Tecentriq 16 12 4 75 25 NF-B11 16 12 4 75 25 10XIC50 NF-B11 16 12 4 75 25 100XIC50

[0316] In term of results, Tecentriq has no statistical effect compared to negative control while NF-B11 treatment shows a 47% reduction of tumor weight, for the treatment at 50 times the IC50 respectively. For the dose of 10 times the IC50, the tumor reduction is around 24% but is not statistically different compare to negative control.

Example 5. Study of efficacy of NE-B11 Compound on Breast Carcinoma Tumors Initiated from MDA-MB-231 Cell Lines

[0317] Treatments

[0318] At day 10 (E10), tumors began to be detectable. They were then treated during 10 days, every two days (E11, E13, E15, E17) by dropping 100 μl of vehicle On PBS), Ref Compound (Tecentriq), NF-B11 or NF-B11-B11 compounds (see Table 7 for concentration).

TABLE-US-00033 TABLE 7 Groups for the study Final Group description Molecule Name Concentration Group 1 Negative Ctrl Neg Ctrl 0 (vehicle) Group 2 Positive Ctrl Tecentriq 200x IC50 (Ref. Compound) Group 3 NF-B11 100xIC50 NF-B11 100xIC50 Group 4 NF-B11-B11 10x IC50 NF-B11-B11 10x 10x IC50 Group S NF-B11-B11 100x IC50 NF-B11-B11 100x 100x IC50

[0319] Tumor Growth Analysis

[0320] At day 18 (E18) the upper portion of the CAM was removed, washed in PBS and then directly transferred in PFA (fixation for 48 h). The tumors were then carefully cut away from normal CAM tissue. Tumors were then weighed. A one-way ANOVA analysis with post-tests has been done on these data.

[0321] Significance on Statistical Analysis

[0322] For tumor weight and metastasis, statistical difference between groups are visible on graphs by presence of stars with the following signification:

[0323] No stars: no statistical different (p value>0.05);

[0324] One star (*): 0.05≥p value>0.01;

[0325] Two stars (**): 0.01≥p value>0.001;

[0326] Three stars (*** ): 0.001≥p value.

[0327] Toxicity

[0328] The number of dead embryo evaluates the acute toxicity after 10 days of the treatment.

Results

[0329] Tumors Growth

[0330] FIG. 10 presents the mean tumor weight (mg) measured in the different experimental groups after 10 days of treatment on the CAM

[0331] Toxicity

[0332] Table 8 presents the number of dead and surviving embryos after 10 days of treatment in the different experimental groups.

TABLE-US-00034 TABLE 8 Number of dead and surviving embryo for each experimental group. Total Alive Dead % alive % dead Neg Ctrl 18 18 0 100 0 Tecentriq 16 13 3 81 19 NF-B11 18 17 1 94 6 NF-B11- 17 13 4 76 24 B11 10x NF-B11- 18 17 1 94 6 B11 100x

[0333] The study tested the efficacy of anti-PD-L1 (NF-B11) compounds on tumor initiated from MDA-MB-231 cells.

[0334] In this study, Tecentriq, and NF-B11-B11 at 10 times IC50 have a significant effect on tumor weight (16-17% reduction).

[0335] Concerning toxicity, the ratio of dead embryo is higher in treated group NF-B11-B11 at 10 time IC50, but is lower in treated group NF-B11-B11 at 100X×IC50 showing that there is no specific toxicity due to treatment (at the doses tested).

Example 6. Evaluation of In Vivo Toxicity and Efficacy of 2 Compounds in Tumors Derived from MDA-MB-231 Human Breast Cell Line in a CAM Model

[0336]

TABLE-US-00035 TABLE 9 Compounds and concentration Dose Group Description Compound (mg/kg) 1 Neg. Ctrl. (Vehicle) PBS — 2 Pos. Ctrl. Keytruda ® 2 3 Exp. Cpd. 1 [1] NF2 2 4 Exp. Cpd. 1 [2] NF2 20 5 Exp. Cpd. 2 [1] NF1-NF3 2 6 Exp. Cpd. 2 [2] NF1-NF3 20

[0337] NF1: anti-HSP110 Sac7d variant (A-C2)

[0338] NF2: anti-PD-L1 sac7d variant (B11)

[0339] NF3: variant of Sac7d binding to EGFR

[0340] Treatments

[0341] On day E10, tumors began to be detectable. Treatments corresponding to different groups are detailed in Table 10. After each treatment, eggs were individually checked daily for death or visible abnormalities (visual checks).

TABLE-US-00036 TABLE 10 Description of treatment groups. Injected [C] Final (in 100 [C] Dosing Group Description Treatment μl/egg) (in ovo) Regimen 1 Neg. Ctrl. PBS — — E10, E12, E14, E15, E17 2 Keytruda Keytruda ® 0.12 mg/ 2 mg/kg E10, E12, 100 μl E14, E15, E17 3 NF2 - NF2 0.12 mg/ 2 mg/kg E10, E12,  2 mg/kg 100 μl E14, E15. E17 4 NF2 - NF2  1.2 mg/ 20 mg/kg  E10, E12, 20 mg/kg 100 μl E14, E15, E17 5 NF1-NF3 - NF1-NF3 0.12 mg/ 2 mg/kg E10, E12,  2 mg/kg 100 μl E14, E15, E17 6 NF1-NF3- NF1-NF3  1.2 mg/ 20 mg/kg  E10, E12, 20 mg/kg 100 μl E14, E15, E17

Results

[0342] Quantitative Evaluation of Tumor Growth

[0343] FIG. 11 presents the mean tumor weight (mg) measured in the different experiment& groups at the end of study. Mean values±SEM of tumor weight (mg).

[0344] Quantitative Evaluation of Immune Cell Infiltration

[0345] Chicken CD3 Expression in Tumor

[0346] FIG. 12 present data analysis of chicken CD3 expression in tumor.

[0347] Chicken MMD Expression in Tumor

[0348] FIG. 13 present data analysis of data analysis of chicken MMD expression in tumor.

[0349] Quantitative Evaluation of Embryonic Toxicity

[0350] Table 11 present the number of dead and surviving embryos at the end of the study.

TABLE-US-00037 TABLE 11 Number of dead and alive embryos per group Group Description Total Alive Dead % Alive % Dead 1 Neg. Ctrl. 23 17 6 73.91 26.09 2 Keytruda 20 14 6 70.00 30.00 3 NF2 - 2 24 15 9 62.50 37.50 mg/kg 4 NF2 - 20 21 14 7 66.67 33.33 mg/kg 5 NF1-NF3 - 2 24 16 8 66.67 33.33 mg/kg 6 NF1-NF3 - 20 21 14 7 66.67 33.33 mg/kg

[0351] First, in terms of tumor growth, when compared to the Negative Control, both NF2 and NF1-NF3 compounds didn't impact tumor growth at the low dose (2 mg/kg); at the high dose (20 mg/kg), we observed the tumor growth regression tendency for both compounds, but the tumor growth inhibition was significant only for NF1-NF3, Second, in terms of tumor immune cells filtration, we observed an increase tendency of CD3 positive cells infiltration in presence of Keytruda, an increase tendency of MMD positive cells in presence of the Sac7d variants, with a dose effect for NF2 alone. All these increases are not statistically significant.

[0352] Finally, treatments with both NF2 and NF1-NF3 compounds did not induce any evident embryonic toxicity at all tested doses when compared to Negative Control.

Example 7. Further Elements

[0353] Nanofitins were derived by ribosome display against human PDL1 using the 200 aa long recombinant ECD from R&D system (ref 156-B7), that consists of the two immunoglobulin-like domains fused to an IgG1 Fc fragment. A large number of different Nanofitins were isolated with a specific signal in ELISA for human PDL1, which was immobilized by physisorption. It was possible to group these proteins into cluster of sequences based on the homology of their variable domains. Candidates were then screened for their ability at engaging their target in a biolayer interferometry set up (Octet RED96). In this setup, the target was immobilized on protein A biosensors to ensure a native and oriented presentation of the target. 60% of the candidates retained their ability at engaging human PDL1 in this set up, out of which only 30% (16% of the total) were found to cross react with murine PDL1. All the cross reactive Nanofitin were restricted within one cluster, highlighting the differential drugability of human and murine PDL1.

[0354] The ability of the Nanofitins from this cluster at neutralizing the interaction between PDL1 and PD1 was evaluated in a competition ELISA, and compared to candidates from other clusters. Competition ELISA was performed according to the following procedure. Human PD1 was immobilized on NUNC maxisorp plate (100 μL, 1 μg/mL, 1 hr). The well were blocked using TBS-BSA (20 mM Tris, 150 mM NaCl, 0.5% BSA, pH 7.4) for 1 hr (300 μL/well). Between each of the following steps, each well was washed 3 time using 300 μL of TBS-T (20 mM Tris, 150 mM NaCl, 0.1% Tween 20, pH 7.4). The Nanofitins were mixed with biotinylated human PDL1 (100 nM) in TBS-T (100 μL final) and applied to the wells. Revelation was then carried over by the addition of 100 μl of streptavidin HRP conjugate diluted 1:10000 in TBS-T for 1 h, followed by the addition of 100 μl of o-Phenylenediamine dihydrochloride substrate (Sigma-Aldrich) solution at 1 mg/ml in revelation buffer (0.05 M citric acid, 0.05% hydrogen peroxide), Absorbance at 450 nm was measured using a Varioskan ELISA plate reader (Thermo Scientific).

[0355] Two Nanofitin candidates from the cluster were capable of neutralizing the interaction between human PDL1 and human PD1 with a neutralization potential appearing to be well correlated with their affinity, with the lowest affinity providing the highest neutralization efficiency. It was noted that similar affinity does not directly translate with similar neutralization potential between the different clusters.

[0356] It was shown that the two selected nanofitins (SEQ ID NO: 38 and SEQ ID NO: 59) are targeting the Ig like V type domain of PDL1 ECD on an epitope that is in close proximity or overlapping with the interaction area between human PDL1 and human PD1, and that is shared on murine PDL1. These variants show the highest affinity on human PDL1 and SEQ ID NO: 38/SEQ ID NO: 41 showed the highest level of cross reactivity on human and mouse PDL1, whereas SEQ ID NO: 68/SEQ ID NO: 71 had the best efficacy.

[0357] Modification of D16, N36 and M56 (replacement by E, Q and L respectively) didn't change the affinity or neutralization potential.

[0358] Modifications of the mutated amino acids showed that I8, 126, L31, M42, L44 (and R24, Y29 and K33 for murine binding) are important for maintaining binding and neutralization for one Nanofitin, and that M8, L26, L31, L42 and F44 (and T24, A29 and R33 for murine binding) are important for maintaining binding and neutralization for the other

Example 8. Anti-HSP110 Binders

[0359] Nanofitines were also generated as Hsp110 binders. In brief, screening was performed with Sac7d variants. and the identified and sequenced proteins were grouped among four groups according to their sequence and signature. Among these groups, 20 different Nanofitin binding site compositions could be identified. These 20 Nanofitin candidates were compared for their binding level in ELISA at a fixed concentration (1 μM).

[0360] The variant corresponding to SEQ ID NO: 80 (or SEQ ID NO: 27, A-C2) was determined to present the best affinity and neutralization potential. Substitutions of amino acids, as described above, showed that W8, 022, W24 (which is conserved with regards to the wild-type protein), H33 and K44 are important for maintaining binding and affinity.

[0361] Affinity for this protein was below 1 nM, and it was shown that it was unable to penetrate into eukaryotic cells.

[0362] A combinational immunotherapeutic effect of A-C2 and an anti PD-L1 Nanofitin was shown in vivo in the chicken chorioallantoic membrane (CAM) mod& (see also example 3 above).

[0363] Altogether, these results demonstrate the therapeutic interest of A-C2 in cancer by itself, despite a lack on internalization, through an effect believed to involve macrophages, as well as its complementarity with a protein inhibiting binding of PD-L1 and PD1 (in particular a Nanofitin anti-PD-L1).