Pioneering success of antibodies targeting immune checkpoints such as PD-1 and CTLA4 has opened novel avenues for cancer immunotherapy. Along with impressive clinical activity, severe immune-related adverse events (irAE) due to the breaking of immune self-tolerance are becoming increasingly evident in antibody-based approaches. As a strategy to better manage severe adverse effects, we set out to discover an antagonist targeting PD-1 signaling pathway with a shorter pharmacokinetic profile. Herein, we describe a peptide antagonist NP-12 that displays equipotent antagonism toward PD-L1 and PD-L2 in rescue of lymphocyte proliferation and effector functions. In preclinical models of melanoma, colon cancer, and kidney cancers, NP-12 showed significant efficacy comparable with commercially available PD-1–targeting antibodies in inhibiting primary tumor growth and metastasis. Interestingly, antitumor activity of NP-12 in a preestablished CT26 model correlated well with pharmacodynamic effects as indicated by intratumoral recruitment of CD4 and CD8 T cells, and a reduction in PD-1+ T cells (both CD4 and CD8) in tumor and blood. In addition, NP-12 also showed additive antitumor activity in preestablished tumor models when combined with tumor vaccination or a chemotherapeutic agent such as cyclophosphamide known to induce “immunologic cell death.” In summary, NP-12 is the first rationally designed peptide therapeutic targeting PD-1 signaling pathways exhibiting immune activation, excellent antitumor activity, and potential for better management of irAEs.

Recent advances in achieving highly durable clinical responses via inhibition of immune checkpoint proteins including CTLA-4 and PD-1 have revolutionized the outlook for cancer therapy (1–5). Immunotherapies including checkpoint inhibitors are being considered as the “fifth pillar” of cancer therapy, as witnessed by the approval of a number antibodies targeting PD-1/PD-L1 immune checkpoint signaling pathway spanning an expanding list of indications (6–9). In contrast to blocking CTLA4, antibodies that target PD-1/PD-L1 signaling pathway have stolen the limelight primarily because of the observed efficacy in multiple clinical indications with toxicities that appear to be less common and less severe (10). However, along with impressive clinical activity (response rate of ∼25% with either anti-CTLA-4 or anti-PD-1 as single agent, but >50% with a combination), severe immune-related adverse events (irAE) due to the breaking of immune self-tolerance are becoming increasingly evident. IrAEs ≥ grade 3 severity occurs in up to 43% of patients with CTLA4 agent and ≤20% with PD-1/PD-L1 agents. The incidence of irAEs with antibodies is dose-dependent, with greater toxicity at higher dose levels (11, 12). Sustained target inhibition as a result of a long half-life (>15–20 days) and approximately 70% target occupancy for months are likely contributing to severe irAEs observed in the clinic with antibodies targeting immune checkpoint proteins (13–15). Hence, we set out to develop novel immune checkpoint blockers with potent antitumor activity but with a shorter pharmacokinetic profile as a strategy to better manage severe adverse effects.

Initial efforts in identifying nonantibody-based approaches for checkpoint antagonism was reported by investigators at Harvard University (Cambridge, MA) by demonstrating the immunomodulating activity in an IFNγ release assay in transgenic mouse T cells that express PD-1 (16). Phage display technique employed by researchers at Zhengzhou, Tsinghua and Zhejiang Sci-Tech Universities yielded hydrolysis-resistant D-peptides as metabolically stable PD-L1 antagonists (17). Although these D-peptides exhibit desirable pharmacokinetic profile, a completely nonnative sequence that is also composed of all D-amino acids, may be highly immunogenic. Potential generation of antidrug antibodies against this D-peptide upon dosing in patients may render this agent ineffective in subsequent administrations. We conceptualized an alternate peptide-based approach based on the native sequence to minimize immunogenicity by identifying and stabilizing critical fragments from the discontinuous epitopes at the interface of PD-1 with PD-L1 interaction. In this article we report detailed in vitro and in vivo pharmacologic characterization of the lead peptide NP-12. To our best knowledge NP-12 is the first rationally designed nonantibody-based pharmacologically active immune checkpoint antagonist targeting PD-1 signaling pathways.

Test compound

NP-12 [Ser-Asn-Thr-Ser-Glu-Ser-Phe-Lys(Ser-Asn-Thr-Ser-Glu-Ser-Phe)-Phe-Arg-Val-Thr-Gln -Leu-Ala-Pro-Lys-Ala-Gln-Ile-Lys-Glu-NH2] was synthesized in-house (Aurigene Discovery Technologies Ltd.), according to the processes described in the U.S. patent 8907053, in particular those described in example 2 (18). A scrambled peptide with identical amino acid composition was synthesized using similar synthetic procedure as that of NP-12.

Mouse splenocyte or human peripheral blood mononuclear cell proliferation assay

For all in vitro assays, 10 mmol/L NP-12 stock was prepared in water and diluted further in the assay media to achieve test concentrations. Mouse splenocytes (10 × 106 cells/mL) or isolated peripheral blood mononuclear cells (PBMC; 10–20 × 106 cells/mL) from human blood were treated with 1 μmol/L of carboxyfluorescein succinimidyl ester (CFSE; eBioscience) in prewarmed 1 × PBS/0.1% BSA solution for 10 minutes at 37°C. Excess CFSE was quenched by adding 5 volumes of ice-cold culture media to the cells and incubated on ice for 5 minutes. CFSE-labeled splenocytes/PBMCs were further given three washes with ice-cold RPMI (Gibco) media containing 10% FBS (HyClone). CFSE-labeled splenocytes/PBMCs (1 × 105 cells/well, in 96-well plate) were added to wells containing recombinant PD-L1 (R&D Systems) or PD-L2 (R&D Systems; both 10 nmol/L) and NP-12 peptide or anti-PD-1 antibody (eBioscience) or isotype controls (eBioscience). Cells were stimulated with host-specific anti-CD3 and anti-CD28 antibodies (eBioscience; 1 μg/mL each) and the cell culture was further incubated for 72 hours at 37°C with 5% CO2. After incubation, cells were collected by centrifuging at 500 × g for 5 minutes at 2–8°C and the cell pellet was washed three times with FACS buffer by centrifuging the plates at 200 × g for 5 minutes at 2–8°C. The cells were resuspended in FACS buffer and transferred into FACS tubes and acquired using BD FACSCalibur (model: E6016) with 488-nm excitation and 521-nm emission filters. Each experimental condition was carried out in triplicates. Percent proliferation and rescue for a given test compound concentration was calculated by normalizing individual percent peptide proliferation values to percent anti-CD3 and anti-CD28 antibodies stimulated proliferation. EC50 values were derived by plotting transformed data into nonlinear fit in sigmoidal dose–response curve using GraphPad Prism 5 software.

Mouse splenocyte or human PBMC IFNγ release assay

Mouse splenocytes (1 × 105 cells/well) or isolated PBMCs from human (1 × 105 cells/mL) were added to wells in 96-well plate containing recombinant mouse or human PD-L1 or PD-L2 (both 10 nmol/L) and NP-12 peptide or anti-PD-1 antibody or isotype controls. Cells were stimulated with host-specific anti-CD3 and anti-CD28 antibodies (1 μg/mL each) and the cell culture was further incubated for 72 hours at 37°C with 5% CO2. After 72 hours of incubation, the cell culture supernatants were collected after brief centrifugation of culture plates (200 × g for 5 minutes at 2–8°C) and processed for IFNγ measurement by ELISA following the manufacturer's protocol (mouse, DY-485 and human, DY-285; R&D Systems). Each experimental condition was carried out in triplicates. Percent IFNγ release was calculated as a ratio of test compound IFNγ concentration (subtracted with that in PD-L background control) to anti-CD3+ anti-CD28+ control (subtracted with that in PD-L background control) multiplied by 100. EC50 values were derived by plotting transformed data in sigmoidal dose–response curve using GraphPad Prism 5 software.

Characterization of immune cell population in splenocyte culture upon rescue from the PD-L1–mediated inhibition

Mouse splenocytes were cultured at 1 × 105 cells/well in 96-well culture plate containing RPMI complete media (RPMI + 10% FBS + 1 mmol/L sodium pyruvate + 10,000 units/mL penicillin and 10,000 μg/mL streptomycin) in the presence or absence of recombinant mouse PD-L1 (10 nmol/L) and NP-12 (100 nmol/L) or anti-PD-1 antibody (100 nmol/L, clone J43). Cells were stimulated with anti-mouse CD3e and anti-mouse CD28 antibodies (1 μg/mL each) and cultured for 72 hours at 37°C with 5% CO2 in incubator. Cells were harvested after 72 hours of culture, washed thrice with ice cold FACS buffer, and stained with fluorescent dye–conjugated antibodies for CD4 and CD8 T cells, B cells, natural killer (NK) cells, and regulatory T cells (Treg; eBioscience) by incubating the samples in dark for 30 minutes on ice. After incubation, cells were washed thrice with FACS buffer and the cell suspension was fixed and permeabilized using fixation/permeabilization buffer (eBioscience) followed by intercellular staining with anti-Ki67 and FoxP3 antibodies (eBioscience). These labeled cells were washed and further suspended in FACS buffer and were analyzed in FACSCalibur acquiring at least 50,000 of total events on live gate in scatter plot. Percentage of cells positive to each marker was gated and analyzed for percent Ki67+ and interpreted accordingly.

Disruption of the interaction of PD-1 with PD-L1

The effect of PD-1–derived peptide (NP-12) and a scrambled control peptide (NP-S1; with similar number and type of amino acids as in NP-12), for their ability to inhibit PD-1 interaction with PD-L1 was evaluated by using sulfo-SBED cross-linking method (19, 20). Briefly, 180 μmol/L of sulfo-SBED (Thermo Fisher Scientific) was incubated for 1 hour in dark with 1 μmol/L of PD-1 (R&D Systems). After incubation, excess sulfo-SBED was removed by centrifuging the mixture in a 3-kDa Microcon centrifugal filter tubes at 18,000 × g for 30 minutes at 25°C. Complex was washed twice with 1× PBS by centrifugation. Concentrated reaction mixture was collected by inverting the same filter tubes into fresh vials and by spinning at 950 × g for 3 minutes. In parallel, 200 nmol/L of PD-L1 (R&D Systems) was incubated with different concentrations of NP-12 (200 nmol/L, 500 nmol/L, and 1 μmol/L) or NP-S1 (200 and 500 nmol/L) in separate tubes for 1 hour at room temperature in dark. After incubation, PD-L1 and NP-12/NP-S1 complex were further incubated with sulfo-SBED–PD-1 complex for 1 hour in dark. The complex was then cross-linked in an UV cross-linker for 5 minutes with 15 watts five tubes at 5-cm distance. The complex was detected by running on 8% SDS-PAGE. The proteins were transferred on to a nitrocellulose membrane and the complexes were detected by probing the membrane with streptavidin–horseradish peroxide (HRP; 1:2,000 dilutions in 5% BSA in 1 × TBST). Membrane was exposed to SuperSignal West Pico chemiluminescent substrate (1:1 ratio). The bound protein complexes were captured on to X-ray film and band intensity was quantified by using Bio-Rad Quantity One software.

Plasma protein binding, plasma and metabolic stability of NP-12

NP-12 was tested for plasma protein binding (test concentration 10 μmol/L), plasma stability (test concentration 10 μmol/L), and metabolic stability (test concentration 1 μmol/L) as per the protocols reported in the literature (21–24).

Pharmacokinetics of NP-12 in Balb/c mice

All animal experimental procedures used in these studies including pharmacokinetic, pharmacodynamic, and efficacy experiments were approved by the Institutional Animal Ethical Committee based on the Committee for the Purpose of Control and Supervision on Experiments on Animals (India) guidelines. NP-12 was administered either intravenously or subcutaneously to the animals at a dose of 3 mg/kg to determine the pharmacokinetic parameters using 5% dextrose water as formulation. After administration, blood samples were collected at regular intervals until 24 hours and centrifuged to obtain the plasma fraction. The plasma samples were processed by SPE method and the eluent were analyzed by LC/MS-MS to determine the plasma concentration of the compound. From intravenous administration, plasma concentration after injection (C0 minutes), the area under the concentration−time curve from time zero to infinity (AUC 0−∞), the mean residence time, volume of distribution (Vdss), and clearance (CL) for each mouse were obtained. The maximum plasma concentration (Cmax), time to reach maximum plasma concentration (Tmax), and AUC 0−∞ were obtained from subcutaneous administration of NP-12. On the basis of the intravenous and subcutaneous parameters, bioavailability of NP-12 was calculated.

Syngeneic mouse studies

In all in vivo tumor growth inhibition (TGI) studies, tumor volumes were measured two times weekly using digital calipers and the volume was expressed in mm3 using the formula V = 0.5a × b2, where a and b are the long and short diameters of the tumor, respectively. Body weights and clinical signs were monitored twice a week. NP-12 was dissolved in 5% dextrose water for all the in vivo studies, except for B16F10 mouse melanoma and Renca tumor models where 1 × PBS was used. Fresh formulation was prepared every day. Compound and vehicle controls were dosed subcutaneously once a day at a dosing volume of 10 mL/kg body weight.

For syngeneic efficacy studies in CT26 colon carcinoma model, a suspension of 2 × 106 cells (source: ATCC, cultured under ATCC-suggested conditions) were injected subcutaneously to the 6- to 8-weeks old male Balb/c mice (in-house). Dosing started 5 days after cell implantation when average tumor volumes were around 60 mm3 (n = 10). Dosing continued for 17 days. Anti-PD-1 antibody (Clone J43 from BioXCell) was dosed intraperitoneally at 100 μg/animal once every week. Tumor volumes measured on day 17 were used to calculate percent TGI. Tumor volume of each animal is compared with the average tumor volume of vehicle control group to calculate TGI and error bars. Five animals from each of the treatment groups (satellite arms) were sacrificed after 17 days of dosing and analyzed for CD69 and PD-1 expression on CD4 and CD8 T cells by flow cytometry. Briefly, the samples (blood and tumor) were collected on the day 17 of the study. PBMCs from whole blood were collected by density gradient centrifugation, by over laying samples on histopaque (1083) and centrifuged at 800 × g for 20 minutes. Opaque layer comprising of PBMCs were collected and stained with fluorophore-conjugated antibodies (CD4-APC, CD8-FITC, CD69-PE, and PD-1-PE; eBioscience) and incubated for 20 minutes on ice. The samples were washed and further suspended pellets in FACS buffer and acquired using BD FACSCalibur. Tumor samples collected were subjected to mechanical disruption and infiltrating cells were isolated by density gradient centrifugation as detailed above.

For the experiment with cyclophosphamide, 100 mg/kg of cyclophosphamide was dosed once intraperitoneally on day 5 after cell implantation (first day of dosing for all other groups).

For evaluating inhibition of metastasis in B16F10 mouse melanoma model, 0.1 × 106 B16F10 cells (source; ATCC, cultured under ATCC-suggested culture conditions) were injected intravenously to the tail vain of C57B6 mice (in-house) on day 0 and dosing started on day 1. Animals were dosed every day for a period of 14 days and monitored for clinical signs and body weight loss. At the end of 14 days study period, animals were sacrificed and discrete metastatic foci were counted with naked eye. Percent inhibition of metastasis in each treatment group was calculated as reduction in average metastatic foci in each group compared with that in vehicle group.

For syngeneic study with Renca cells, a suspension of 0.5 × 105 cells (source: ATCC, cultured under per ATCC-suggested conditions) were injected orthotopically to the anterior capsule of right kidney of 6- to 8-weeks-old female Balb/c mice (in-house). Mice were randomized (n = 15 per group) 5 days after the recovery period from the day of surgery. For vaccination, Renca cells (1 × 106) were irradiated at 160 Gy by gamma irradiation (Gamma Irradiation Chamber-5000, Department of Atomic Energy, Board of Radiation and Isotope Technology, Mumbai, Maharashtra, India) and injected subcutaneously on day 4, 7, and 10 post cell implantations. Animals were dosed starting from 5th day of cell implantation and dosing continued for 21 days. Animals were sacrificed on day 21 to record tumor weight (difference in weight between left and right kidneys). Average tumor weight in the treatment arms was compared with that of vehicle control arm to calculate percent TGI.

Statistical analysis

Data analysis was performed using GraphPad Prism software. One-way ANOVA was used to determine the statistical significance. Unless noted otherwise, error bars in figures represent the SE of independent determinations.

Identification of lead candidate

As an alternative to antibody-based approach, we sought to discover and develop peptide-based immune checkpoint antagonists capable of targeting PD-1/PD-L1 signaling pathways. We reasoned that such therapeutic agents would likely allow better management of irAEs due to relatively shorter pharmacokinetic exposure. The rational design strategy was initiated by synthesizing and analyzing strands and loops of PD-1 from the interface of PD-1/PD-L1 interaction for functional antagonism. The interacting strands and loops sequences from the interface were based on the crystal structure reported by Lin and colleagues (25). Initially individual strands and loops were designed, synthesized, and screened in a functional assay at a single concentration of 100 nmol/L for rescue of mouse splenocyte proliferation in the presence of recombinant PD-L1. Among the preliminary hits, NK-14 peptide designed on the basis of sequences from FG loop and G-strand was identified as a starting point. NK-14 rescued PD-L1–mediated inhibition of proliferation of mouse splenocytes to the tune of 52% (Table 1). Addition of D-strand to NK-14 resulted in NK-15, which showed marginal increase in activity. Further addition of sequences from BC loop to NK-15 resulted in NK-56 that increased the rescue to 87%. Because addition of BC loop increased activity significantly, we wanted to test the tandem repeat of BC loop in the NK-56 backbone. But as observed with peptide NK-77, the tandem repeat of BC loops spaced with one lysine resulted in compromised activity. We then decided to incorporate BC loop as a branch and toward achieving this design we first introduced a branching point using lysine as in peptide NP-03 with retention of activity. At the branching point in the N-terminus of linear sequence we introduced BC loop (NP-12), as well as other loops from the PD-1 ectodomain such as CC' loop (NL-17) and FG loop (NM-26). Among the designed peptides, NP-12 was able to fully rescue the inhibition of proliferation mediated by PD-L1 in mouse splenocyte assay. An all D-amino acids version of NP-12 was inactive in the same assay.

Table 1.

Structure–activity relationship studies leading to the discovery of NP-12

Structure–activity relationship studies leading to the discovery of NP-12
Structure–activity relationship studies leading to the discovery of NP-12

NP-12 disrupts the interaction of PD-1 with PD-L1

After observing the functional antagonism of both PD-L1 and PD-L2, we sought to determine whether the functional antagonism is the result of disruption of the interaction of PD-1 with PD-L1. Recombinant human PD-1 was labeled with sulfo-SBED containing an amine-reactive NHS-ester group and UV light activatable aryl azide group and cross-linked with recombinant human PD-L1 in the presence or absence of NP-12. Successful label transfer to the interacting protein (PD-L1) as a result of complex formation was determined by Western blot analysis using streptavidin–HRP. The complex formation of PD-1 with PD-L1 was inhibited to the extent of 65% and 93% at 2.5- and 5-fold molar excess addition of NP-12 with respect to PD-L1 concentration. In contrast, addition of scrambled peptide did not disrupt the complex formation. Schematic representation of cross-linking assay and dose-dependent disruption as determined by Western blot analysis with streptavidin–HRP is represented in Fig. 1A and B. Increase in PD-1–PD-L1 interaction at lower concentration of NP-12 was an outlier and was not reproducible. In a cell-based binding assay, the preferential binding of FITC-labeled NP-12 to PD-L1–overexpressing CHO-K1 cells as compared with the parental cells expressing low levels of PD-L1 indicates that NP-12 binds to PD-L1 (Supplementary Fig. S1A–S1E).

Figure 1.

NP-12 disrupts interaction of PD-1 with PD-L1 in an SBED cross-linking assay. Schematic representation (A) and dose-dependent disruption as determined by Western blot analysis with streptavidin-HRP (B). NP-12 was tested at 200 (1× relative to PD-L1), 500 (2.5× relative to PD-L1), and 1,000 nmol/L (5× relative to PD-L1).

Figure 1.

NP-12 disrupts interaction of PD-1 with PD-L1 in an SBED cross-linking assay. Schematic representation (A) and dose-dependent disruption as determined by Western blot analysis with streptavidin-HRP (B). NP-12 was tested at 200 (1× relative to PD-L1), 500 (2.5× relative to PD-L1), and 1,000 nmol/L (5× relative to PD-L1).

Close modal

NP-12 restores the proliferation and IFNγ rescue from mouse splenocyte and human PBMC inhibited by recombinant PD-L1 or PD-L2 proteins

Functional antagonism of PD-L1 or PD-L2 signaling by NP-12 was evaluated by monitoring the rescue of T-cell activation inhibited in the presence of these ligands in mouse splenocyte or human PBMC cultures. Mouse splenocytes or human PBMCs stimulated with anti-CD3 and anti-CD28 antibodies are known to activate T cells inducing proliferation and cytokine secretion. Also, during their activation, these cells start expressing immune checkpoint receptor PD-1. In this study, recombinant PD-L1 and PD-L2 proteins were used to inhibit T-cell proliferation stimulated in the presence of anti-CD3 and anti-CD28 antibodies (Fig. 2A and D). Human PBMCs isolated from individual donors in which proliferation and IFNγ release were inhibited by PD-L1 and PD-L2 (a minimum inhibition of 50% for proliferation and 70% for IFNγ release) were used in the assays for analyzing the test agents. In these functional assays, in the presence of PD-L1 or PD-L2, anti-PD-1 antibody and the peptide antagonist NP-12 showed a dose-dependent rescue of proliferation or IFNγ secretion. Four independent experiments were performed using mouse splenocytes and human PBMCs from four independent mice/human donors. Average EC50 values for antibody or NP-12 are presented in Supplementary Table S1A and S1B. Representative EC50 graphs plotted from splenocytes from 1 mouse and PBMCs from 1 human donor are presented in Figs. 2B, C, E, and F and3B, C, E, and F).

Figure 2.

In vitro functional activity of NP-12. PD-L1–mediated rescue of proliferation (assay format in A) in mouse splenocyte (B), and human PBMCs (C). PD-L2–mediated rescue of proliferation (assay format in D) in mouse splenocyte (E) and human PBMCs (F).

Figure 2.

In vitro functional activity of NP-12. PD-L1–mediated rescue of proliferation (assay format in A) in mouse splenocyte (B), and human PBMCs (C). PD-L2–mediated rescue of proliferation (assay format in D) in mouse splenocyte (E) and human PBMCs (F).

Close modal
Figure 3.

In vitro functional activity of NP-12. PD-L1–mediated IFNγ release (assay format in A) in mouse splenocyte (B) and human PBMCs (C). PD-L2–mediated IFNγ release (assay format in D) in mouse splenocyte (E) and human PBMCs (F).

Figure 3.

In vitro functional activity of NP-12. PD-L1–mediated IFNγ release (assay format in A) in mouse splenocyte (B) and human PBMCs (C). PD-L2–mediated IFNγ release (assay format in D) in mouse splenocyte (E) and human PBMCs (F).

Close modal

In mouse splenocyte culture, the anti-PD-1 antibody (clone J43) showed a dose-dependent response in rescuing the proliferation (Fig. 2B and E) with average EC50 values of 18.2 ± 6.1 and 18.6 ± 10.2 nmol/L against PD-L1 and PD-L2, respectively (Supplementary Table S1A). Isotype control (Armenian hamster IgG), did not rescue the proliferation indicating the specificity of the rescue to mouse PD-1 antagonism. The peptide antagonist NP-12 rescued the proliferation in the mouse splenocyte assay system (Fig. 2B and E) with average EC50 values of 17 ± 3.6 and 16.6 ± 3.1 nmol/L against rmPD-L1 and rmPD-L2, respectively (Supplementary Table S1A).

In human PBMC cultures, the anti-human PD-1 antibody (clone J116) also showed a dose-dependent rescue (Fig. 2C and F) with average EC50 values of 40.9 ± 19.4 and 37.5 ± 18 nmol/L against rhPD-L1 and rhPD-L2, respectively (Supplementary Table S1B). Isotype control (mouse IgG1k), did not rescue this activity indicating the specificity of rescue to human PD-1 antagonism. NP-12 was also able to significantly rescue recombinant human PD-L1- and PD-L2–mediated inhibition of in vitro human PBMC proliferation stimulated by anti-CD3 and anti-CD28 antibodies (Fig. 2C and F), and average EC50 values for NP-12 in CFSE-labeled PBMC proliferation assay experiment were found to be 63.3 ± 49.8 and 44.1 ± 20 nmol/L against PD-L1 and PD-L2, respectively (Supplementary Table S1B).

Rescue of IFNγ secretion when stimulated with anti-CD3 and anti-CD28 antibodies in both mouse splenocytes culture and human PBMCs by the appropriate PD-1 antibodies, but not with the isotype control indicated the usefulness of this system to characterize the functional antagonism of PD-L1/PD-L2 (Fig. 3A and D). Anti-mouse PD-1 antibodies rescued IFNγ secretion (Fig. 3B and E) with average EC50 values of 22.5 ± 5.4 and 20.7 ± 10.2 nmol/L against mouse PD-L1 and PD-L2, respectively (Supplementary Table S1A), whereas human PD-1 antibody rescued IFNγ secretion (Fig. 3C and F) with average EC50 value of 33 ± 8.6 and 33.2 ± 12.2 nmol/L against human PD-L1 and PD-L2, respectively (Supplementary Table S1B). NP-12 showed potency in restoring the IFNγ release inhibited by recombinant mouse PD-L1 and PD-L2 (Fig. 3B and E) with average EC50 values of 49.4 ± 15.4 and 51.0 ± 22.5 nmol/L against mouse PD-L1 and PD-L2, respectively (Supplementary Table S1A) and average EC50 values of 84.3 ± 43.2 and 98.8 ± 42.3 nmol/L against human PD-L1 and PD-L2, respectively (Supplementary Table S1B). The clinical relevance of NP-12 is evident from the enhanced the T-cell activity in terms of higher IFNγ secretion in the antigen recall assays, which measures the ability to stimulate antigen-specific memory T cells in PBMCs (Supplementary Figs. S2–S4).

NP-12 rescues proliferation of both CD4 and CD8 immune cell subsets

Because NP-12 showed potent restoration of proliferation and IFNγ release in functional studies, we sought to characterize specific T-cell subsets impacted by the addition of NP-12. Anti-CD3- and anti-CD28 antibody–stimulated mouse splenocytes had an increased CD4 and CD8 T-cell subsets over the unstimulated condition as expected (Fig. 4). In the presence of recombinant PD-L1 both CD4 and CD8 T-cell population decreased (P < 0.0001 as compared with anti-CD3 and -CD28 stimulation). NP-12 at 100 nmol/L was able to inhibit PD-L1 effect on both CD4 and CD8 T cells, indicating that NP-12 was able to revert PD-L1 effects and these effects were comparable with anti-PD-1 antibody (P < 0.0001 as compared with PD-L1 treatment). Interestingly, CD4 Tregs and proliferating Tregs were reduced in NP-12- and anti-PD-1 antibody–treated groups (P < 0.005) as compared with anti-CD3, CD28 stimulation. A reduction in the number of Tregs and an increase in the CD4+CD8+ cells by affecting their proliferation upon NP-12 treatment could be supportive of the potential for an protective antitumor immunity by blocking the PD-1/PD-L1 pathway. An increase in the percentage of proliferating B cells compared with stimulated control (likely an indirect effect or artifact of the mixed cell culture system) and a further increase with NP-12 addition, and no evidence on change in NK cells phenotype were also observed from these experiments.

Figure 4.

Effect of NP-12 and anti-mouse PD-1 antibody (clone J43) on rescue of proliferation on lymphocyte subsets.

Figure 4.

Effect of NP-12 and anti-mouse PD-1 antibody (clone J43) on rescue of proliferation on lymphocyte subsets.

Close modal

NP-12 exhibits desirable absorption, distribution, metabolism and excretion (ADME) and Drug Metabolism and Pharmacokinetics (DMPK) profile

To determine the dose and route of administration for pharmacologic characterization of NP-12, the peptide was evaluated for its ADME and pharmacokinetic properties. Plasma protein binding of NP-12 was found to be 93.9% in mice plasma with a plasma stability of more than 60% remaining at 6 hours at tested concentration of 10 μmol/L. NP-12 showed a half-life of more than 90 minutes in mouse liver microsomes. The pharmacokinetics of single dose of NP-12 administered by subcutaneous route at a dose of 3 mg/kg or intravenously at a dose of 3 mg/kg was studied in male Balb/c mice (Supplementary Fig. S5A and S5B). NP-12 exhibited medium clearance and low volume of distribution. Peak plasma levels were reached between 0.2 and 0.4 hours post subcutaneous dosing. Absolute bioavailability was found to be 77% in mice.

NP-12 inhibits spontaneous hematogenous spread in the B16F10 melanoma model

Because NP-12 was effective in rescuing T cells from both PD-L1 and PD-L2, we tested the efficacy of NP-12 in B16F10 mouse melanoma spontaneous lung metastasis model. B16F10 cells were injected intravenously on day 0 and dosed with NP-12 starting from day 1. The animals were administered with NP-12 subcutaneously at 3 mg/kg/day dose for a period of 14 days. Treatment with NP-12 resulted in a 66% reduction in metastatic nodules as compared with a vehicle control (Fig. 5A).

Figure 5.

In vivo pharmacologic activity of NP-12 in syngeneic mouse models. Anti-metastatic effect in B16F10 mouse melanoma model (A), TGI (B), and modulation of immune activation markers in blood (C) and tumors (D) in CT26 mouse colon carcinoma model.

Figure 5.

In vivo pharmacologic activity of NP-12 in syngeneic mouse models. Anti-metastatic effect in B16F10 mouse melanoma model (A), TGI (B), and modulation of immune activation markers in blood (C) and tumors (D) in CT26 mouse colon carcinoma model.

Close modal

NP-12 inhibits tumor growth and modulates immune activation markers in circulation and in tumors in CT26 colon carcinoma model

CT26 is an immunogenic colon tumor model that demonstrates PD-L1 expression on tumor cells, as well as tumor-infiltrating immune cells in vivo. This model is reported to be responsive to agents targeting PD-1/PD-L axis (26). Hence, to evaluate the single-agent antitumor efficacy and subsequent pharmacodynamic modulation, NP-12 was tested in this model. NP-12 was tested at 0.1 and 1 mg/kg doses along with a vehicle control and a PD-1 antibody control (J43 Clone, 100 μg/animal/week). When compared with the vehicle-treated group, NP-12–treated groups showed significant TGI in a dose-dependent manner. Group treated with 0.1 mg/kg of NP-12 showed a TGI of 35% and the group treated with 1 mg/kg of NP-12 showed 53% TGI (Fig. 5B). Five animals from each of the same treatment groups were also taken up for monitoring immune activation markers in circulation and tumors. Data represented as percentage change as compared with vehicle control. Analysis of T-cell subsets indicated an increase in CD4 and CD8 T cells, activated CD8 CD4 T cells (CD69+ cells) with both NP-12 and PD-1 antibody treatment in blood, whereas significant percentage increase in activated CD4 T cells (CD69+) was evident only with NP-12 treatment (1 mg/kg; P < 0.005; Fig. 5C). In tumors, there was an increase in CD8 T cells and cells with double positive for CD4 and CD8 markers with both NP-12 and PD-1 antibody, whereas increase in activated CD4 T cells (CD69+) was observed with NP-12 at 0.1 and 1 mg/kg (Fig. 5D). A significant percentage increase in activated CD8 T cells (CD69+) was observed with both NP-12 and PD-1 antibody (P < 0.05). Interestingly, decrease in CD8 T-cells expressing PD-1 was observed with both NP-12 and PD-1 antibody treatment, whereas statistically significant decrease in CD4 T-cells expressing PD-1 was observed only with NP-12 treatment (P < 0.05).

NP-12 shows significantly improved antitumor efficacy when combined with tumor vaccination and cyclophosphamide

To test whether the immunologic cell death caused by irradiated cells imparts any additional antitumor effects, NP-12 was administered to animals vaccinated with gamma-irradiated Renca cells (27). When compared with vehicle control group, TGI of 54% was observed in NP-12–treated group and the group treated with NP-12 in combination with the vaccination showed a TGI of 74% as compared with 37% in vaccination alone group (Fig. 6A).

Figure 6.

Antitumor activity of NP-12 in combination with vaccination in mouse Renca renal cell carcinoma model (A), survival advantage (B), and antitumor activity in combination with cyclophosphamide (C) in CT26 mouse colon carcinoma model.

Figure 6.

Antitumor activity of NP-12 in combination with vaccination in mouse Renca renal cell carcinoma model (A), survival advantage (B), and antitumor activity in combination with cyclophosphamide (C) in CT26 mouse colon carcinoma model.

Close modal

Low dose of cyclophosphamide is known to reduce the Treg population and thereby enhance efficacy of immunomodulatory agents (28). NP-12 at 3 mg/kg resulted in 41% TGI as a single agent. Cyclophosphamide alone at 100 mg/kg resulted in 85% TGI on day 25 (Fig. 6C). A TGI of 93% was observed when 3 mg/kg of NP-12 was combined with cyclophosphamide at 100 mg/kg indicating additive antitumor effect. Analysis of survival curves indicated a median survival for cyclophosphamide alone group as 34 days whereas for the combination of NP-12 and cyclophosphamide as 41 days, further supporting an additive survival benefit (Fig. 6B).

It is important to note that NP-12 was well tolerated in all the in vivo studies as indicated by lack any reduction in the body weights or clinical signs in any of the treatment groups during the study period.

NP-12 is a 29-amino acid branched peptide behaving as a PD-1 decoy generated from the selected portions of the human PD-1 receptor. Structure activity analysis revealed that the peptide designs based on incorporation of specific loops and strands in a noncontiguous manner using an extra lysine as a branching point showed desirable activity. NP-12 with sequences or residues taken from BC loop, D strand, FG loop, and G strand arranged in a noncontiguous manner is the first reported rationally designed peptide. The maximal PD-L1 antagonistic activity observed with peptide NP-12 is likely due to the optimal spatial disposition of the critical residues of PD-1 for interaction with PD-L1 from two BC loops (complementary determining region 1) arranged in a branched fashion along with D strand, FG loop, and G strand.

We employed a functional assay–based screening approach for identifying lead compounds. A functional assay–based screening was favored to potentially identify agents including those which modulate the activity either by allosteric binding or by inducing PD-1/PD-L1 complex destabilization rather than potent disruption. Interestingly, similar strategies have been applied to identify functionally potent and efficacious anti-PD-1 antibodies (29, 30). In the functional assays, NP-12 displayed equipotent antagonism toward PD-L1- and PD-L2–mediated T-cell exhaustion. Equipotent antagonism of both PD-L1 and PD-L2 is analogous to that achieved with an anti-PD-1 antibody as opposed to selective PD-L1 antagonism by anti-PD-L1 antibody.

NP-12 was also tested for its ability to directly modulate cells stimulated with anti-CD3 alone, anti-CD28 alone, or anti-CD3 plus anti-CD28 signaling. NP-12 caused no effect on anti-CD3, anti-CD28, or anti-CD3 plus anti-CD28 antibodies–induced splenocyte proliferation without inhibitory proteins like PD-L1 or PD-L2. These observations confirm that NP-12 did not affect T-cell receptor signaling mediated through anti-CD3/CD28 antibodies stimulation. Disruption of binding by NP-12 to PD-1/PD-L1 complex as revealed in the cross-linking study further supports that that the T-cell activation is mediated through the antagonism of PD-1 signaling.

On the basis of the comparable potency in restoring proliferation and IFNγ release from PBMCs/splenocytes from human and mouse and desirable pharmacokinetic profile, NP-12 was evaluated for antitumor activity in syngeneic mouse models of cancer such as colon carcinoma (CT26), renal cell carcinoma (Renca), and melanoma (B16F10). These tumor types were selected on the basis of the published information on the expression of the PD-L1 as an immune escape mechanism (31). Differences in the degree of response in the models are likely due to differences in PD-L1 expression, presence or absence of other immune checkpoint or immune stimulatory signaling, and angiogenesis trafficking immune cells in to tumors. Among the models evaluated, B16F10 melanoma is reported to be poorly immunogenic and generally has not responded well for anti-PD-1 or anti-PD-L1 agents (32).

Antitumor efficacy of NP-12 correlates with intratumoral recruitment of CD4 and CD8 T cells, and a reduction in PD-1 expressing T cells (both CD4 and CD8). Activation of T-cell subsets along with reduction in PD-1+ cells indicates immune modulation upon treatment with NP-12. Increase in activated CD4 T cells (CD69+) only with NP-12 but not with anti-PD-1 antibody (J43) could be due to different kinetics of activation of CD4 population between two treatments.

It is well documented that immunotherapeutic approaches including antagonists of immune checkpoints such as PD-1 and CTLA4 work best in combination with tumor vaccines or therapeutic agents known to induce immunologic cell death. In agreement with these reports, NP-12 also showed greater inhibition of primary tumor growth when combined with a vaccination approach in Renca orthotropic model and in combination with cyclophosphamide in the CT26 subcutaneous tumor model. Analysis of CT26 tumor model survival curves shows a significant survival advantage in combination with cyclophosphamide.

As expected for a peptide agent, NP-12 exhibited a relatively shorter pharmacokinetic exposure with a t1/2 of 0.22 hours. However, even with the shorter pharmacokinetic exposure, once a day dosing resulted in efficacy comparable with that observed with anti-PD-1 antibody indicating that a sustained pharmacokinetic exposure is not needed to achieve efficacy. Efficacy without the continuous drug exposure could be an advantage in managing immune-related toxicities. Antibodies targeting PD-1 and PD-L1 show immune-related toxicities albeit at lower frequency compared with anti-CTLA4 antibodies. Sustained target inhibition as a result of a long half-life (>15–20 days) and approximately 70% target occupancy for months are likely contributing to severe irAEs observed in the clinic with antibodies targeting immune checkpoint proteins. The current management of irAEs with anti-PD-1/PD-L1 includes treatment cessation along with administration of steroids and or anti-TNF agents.

In summary, we have developed first rationally designed peptide with equipotent antagonism against PD-L1 and PD-L2, significant antitumor activity in vivo either as a single agent or with significant additive effect in combination with agents reported to cause immunologic cell death along with desirable pharmacodynamic marker modulations. These results demonstrate the significant advantages of NP-12 in inhibiting PD-1 signaling pathway without continuous drug exposure as in the current clinically approved agents.

No potential conflicts of interest were disclosed.

Conception and design: P.G. Sasikumar, K. Subbarao, R. Shrimali, N. Gowda, M. Ramachandra

Development of methodology: P.G. Sasikumar, R.K. Ramachandra, A.A. Dhudashia, S. Vadlamani, K. Vemula, S. Vunnum, L.K. Satyam, K. Subbarao, R. Nair, R. Shrimali, N. Gowda, M. Ramachandra

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R.K. Ramachandra, S. Adurthi, A.A. Dhudashia, S. Vadlamani, K. Vemula, S. Vunnum, L.K. Satyam, K. Subbarao, R. Nair, R. Shrimali, N. Gowda

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.G. Sasikumar, R.K. Ramachandra, S. Adurthi, A.A. Dhudashia, S. Vadlamani, K. Vemula, S. Vunnum, L.K. Satyam, D.S. Samiulla, K. Subbarao, R. Nair, N. Gowda, M. Ramachandra

Writing, review, and/or revision of the manuscript: P.G. Sasikumar, R.K. Ramachandra, L.K. Satyam, K. Subbarao, R. Nair, N. Gowda, M. Ramachandra

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P.G. Sasikumar, K. Subbarao, N. Gowda, M. Ramachandra

Study supervision: P.G. Sasikumar, L.K. Satyam, D.S. Samiulla, K. Subbarao, N. Gowda, M. Ramachandra

This work was supported by Aurigene Discovery Technologies Limited, Bangalore, India.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Sharpe
AH
. 
Introduction to checkpoint inhibitors and cancer immunotherapy
.
Immunol Rev
2017
;
276
:
5
8
.
2.
Farkona
S
,
Diamandis
EP
,
Blasutig
IM
. 
Cancer immunotherapy: the beginning of the end of cancer?
BMC Med
2016
;
14
:
73
.
3.
Couzin-Frankel
J
. 
Cancer immunotherapy
.
Science
2013
;
342
:
1432
3
.
4.
Topalian
SL
,
Drake
CG
,
Pardoll
DM
. 
Immune checkpoint blockade: a common denominator approach to cancer therapy
.
Cancer Cell
2015
;
27
:
450
61
.
5.
Shekarian
T
,
Valsesia-Wittmann
S
,
Caux
C
,
Marabelle
A
. 
Paradigm shift in oncology: targeting the immune system rather than cancer cells
.
Mutagenesis
2015
;
30
:
205
11
.
6.
Greil
R
,
Hutterer
E
,
Hartmann
TN
,
Pleyer
L
. 
Reactivation of dormant anti-tumor immunity – a clinical perspective of therapeutic immune checkpoint modulation
.
Cell Commun Signal
2017
;
15
:
5
.
7.
Vanpouille-Box
C
,
Lhuillier
C
,
Bezu
L
,
Aranda
F
,
Yamazaki
T
,
Kepp
O
, et al
Trial watch: immune checkpoint blockers for cancer therapy
.
Oncoimmunology
2017
;
6
:
e1373237
.
8.
Oiseth
SJ
,
Aziz
MS
. 
Cancer immunotherapy: a brief review of the history, possibilities, and challenges ahead
.
J Cancer Metastasis Treat
2017
;
3
:
250
61
.
9.
Jardim
DL
,
de Melo Gagliato
D
,
Giles
FJ
,
Kurzrock
R
. 
Analysis of drug development paradigms for immune checkpoint inhibitors
.
Clin Cancer Res
2018
;
24
:
1785
94
.
10.
Naidoo
J
,
Page
DB
,
Li
BT
,
Connell
LC
,
Schindler
K
,
Lacouture
ME
, et al
Toxicities of the anti-PD-1 and anti-PD-L1 immune checkpoint antibodies
.
Ann Oncol
2015
;
26
:
2375
91
.
11.
Kourie
HR
,
Klastersky
J
. 
Immune checkpoint inhibitors side effects and management
.
Immunotherapy
2016
;
8
:
799
807
.
12.
Day
D
,
Hansen
AR
. 
Immune-related adverse events associated with immune checkpoint inhibitors
.
BioDrugs
2016
;
30
:
571
84
.
13.
Brahmer
JR
,
Tykodi
SS
,
Chow
LQM
,
Hwu
WJ
,
Topalian
SL
,
Hwu
P
, et al
Safety and activity of anti–PD-L1 antibody in patients with advanced cancer
.
N Engl J Med
2012
;
366
:
2455
65
.
14.
Topalian
SL
,
Hodi
FS
,
Brahmer
JR
,
Gettinger
SN
,
Smith
DC
,
McDermott
DF
, et al
Safety, activity, and immune correlates of anti–PD-1 antibody in cancer
.
N Engl J Med
2012
;
366
:
2443
54
.
15.
Brahmer
JR
,
Drake
CG
,
Wollner
I
,
Powderly
JD
,
Picus
J
,
Sharfman
WH
, et al
Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates
.
J Clin Oncol
2010
;
28
:
3167
75
.
16.
Sharpe
AH
,
Butte
MJ
,
Oyama
S
inventors
;
President and Fellows of Harvard College, assignee
. 
Modulators of immunoinhibitory receptor PD-1, and methods of use thereof
.
International patent number WO/2011/082400A2
. 
2011 Jul 7
.
17.
Chang
HN
,
Liu
BY
,
Qi
YK
,
Zhou
Y
,
Chen
YP
,
Pan
KM
, et al
Blocking of the PD-‐1/PD‐-L1 interaction by a D-peptide antagonist for cancer immunotherapy
.
Angew Chem Int Ed
2015
;
54
:
11760
4
.
18.
Sasikumar
PGN
,
Ramachandra
M
inventors
;
Aurigene Discovery Technologies Limited
,
assignee
. 
Immunosuppression modulating compounds
.
United States patent US8907053B2
. 
2014
Dec 9
.
19.
Hermanson
G
.
Bioconjugate techniques
. 2nd ed.
London
:
Academic Press
; 
2008
.
20.
Horney
MJ
,
Evangelista
CA
,
Rosenzweig
SA
. 
Synthesis and characterization of insulin-like growth factor (IGF)-1 photoprobes selective for the IGF-binding proteins (IGFBPs) photoaffinity labeling of the igf-binding domain on igfbp-2
.
J Biol Chem
2001
;
276
:
2880
9
.
21.
Crespi
CL
,
Stresser
DM
. 
Fluorometric screening for metabolism-based drug–drug interactions
.
J Pharmacol Toxicol Methods
2000
;
44
:
325
31
.
22.
Obach
RS
. 
Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: an examination of in vitro half-life approach and nonspecific binding to microsomes
.
Drug Metab Dispos
1999
;
27
:
1350
9
.
23.
Di
L
,
Kerns
EH
,
Hong
Y
,
Chen
H
. 
Development and application of high throughput plasma stability assay for drug discovery
.
Int J Pharm
2005
;
297
:
110
9
.
24.
Kariv
I
,
Cao
H
,
Oldenburg
KR
. 
Development of a high throughput equilibrium dialysis method
.
J Pharm Sci
2001
;
90
:
580
7
.
25.
Lin
DY
,
Tanaka
Y
,
Iwasaki
M
,
Gittis
AG
,
Su
HP
,
Mikami
B
, et al
The PD-1/PD-L1 complex resembles the antigen-binding Fv domains of antibodies and T cell receptors
.
Proc Natl Acad Sci U S A
2008
;
105
:
3011
6
.
26.
Lau
J
,
Cheung
J
,
Navarro
A
,
Lianoglou
S
,
Haley
B
,
Totpal
K
, et al
Tumour and host cell PD-L1 is required to mediate suppression of anti-tumour immunity in mice
.
Nat Commun
2017
;
8
:
14572
.
27.
Hillman
GG
,
Reich
LA
,
Rothstein
SE
,
Abernathy
LM
,
Fountain
MD
,
Hankerd
K
, et al
Radiotherapy and MVA-MUC1-IL-2 vaccine act synergistically for inducing specific immunity to MUC-1 tumor antigen
.
J Immunother Cancer
2017
;
5
:
4
.
28.
Heylmann
D
,
Bauer
M
,
Becker
H
,
van Gool
S
,
Bacher
N
,
Steinbrink
K
, et al
Human CD4+CD25+ regulatory T cells are sensitive to low dose cyclophosphamide: implications for the immune response
.
PLoS One
2013
;
8
:
e83384
.
29.
Fenwick
C
,
Pellaton
C
,
Farina
A
,
Radja
N
,
Pantaleo
G
. 
Identification of novel antagonistic anti-PD-1 antibodies that are non-blocking of the PD-1/PD-L1 interaction
.
J Clin Oncol
2016
;
34
:
3072
.
30.
Scheuplein
F
,
Ranganath
S
,
McQuade
T
,
Wang
L
,
Spaulding
V
,
Vadde
S
, et al
Abstract B30: discovery and functional characterization of novel anti-PD-1 antibodies using ex vivo cell-based assays, single-cell immunoprofiling, and in vivo studies in humanized mice
.
Cancer Res
2016
;
76
:
B30
.
31.
Lechner
MG
,
Karimi
SS
,
Barry-Holson
K
,
Angell
TE
,
Murphy
KA
,
Church
CH
, et al
Immunogenicity of murine solid tumor models as a defining feature of in vivo behavior and response to immunotherapy
.
J Immunother
2013
;
36
:
477
89
.
32.
Chen
S
,
Lee
LF
,
Fisher
TS
,
Jessen
B
,
Elliott
M
,
Evering
W
, et al
Combination of 4-1BB agonist and PD-1 antagonist promotes antitumor effector/memory CD8 T cells in a poorly immunogenic tumor model
.
Cancer Immunol Res
2015
;
3
:
149
60
.