All Patients N=38 Complete response mg/kg q2w. 1 mg/kg q2w. 6 mg/kg q3w n=10. All Patients N= mg/kg. 10 mg/kg. q3w n=10.

Transcription

1 58P Phase /, Open-Label, Multiple Ascending Dose Trial of AGEN34, an Anti PD- Monoclonal Antibody, in Advanced Solid Malignancies: Results of Dose Escalation in Advanced Cancer and Expansion Cohorts in Subjects With Relapsed/Refractory Cervical Cancer Charles Drescher, Kathleen N. Moore, Joyce Liu, 3 David M. O Malley, 4 Edward Wenge Wang, 5 Judy Sing-Zan Wang, 6 Vivek Subbiah, 7 Breelyn A. Wilky, 8 Guojun Yuan, 9 Christopher D. Dupont, 9 Min Lim, 9 Ana M. Gonzalez, 9 David Savitsky, 9 Sara Coulter, 9 Olga Shebanova, 9 Ed Dow, 9 Waldo Ortuzar, 9 Igor Proscurshim, 9 Jennifer S. Buell, 9 Robert Benjamin Stein, 9 Hagop Youssoufian 9 Swedish Medical Center, Seattle, WA; Stephenson Cancer Center at the University of Oklahoma Health Sciences Center/Sarah Cannon Research Institute, Oklahoma City, OK; 3 Dana-Farber Cancer Institute, Boston, MA; 4 The Ohio State University College of Medicine, Columbus, OH; 5 City of Hope Comprehensive Cancer Center, Duarte, CA; 6 Florida Cancer Specialists/SCRI, Sarasota, FL; 7 The University of Texas MD Anderson Cancer Center, Houston, TX; 8 Sylvester Comprehensive Cancer Center, Miami, FL; 9 Agenus Inc. or subsidiary thereof (current or former employee), Lexington, MA BACKGROUND Inhibition of the programmed cell death protein- (PD-) pathways by blockade of receptor-ligand interactions has been demonstrated in numerous clinical trials to result in objective clinical response and increased survival in solid tumor indications (Figure ). -4 PD- signaling is mediated by ligands: PD- ligand (PD-L, or CD74) and PD- ligand (PD-L, or CD73). 5 Upon binding to PD-L or PD-L, PD- signaling in T cells can potently attenuate TCR signaling, resulting in diminished cytokine and proliferative responses, reduced T-effector cell cytolytic activity, and impaired central memory T-cell differentiation. 6 Figure. Pathways Affected by PD- 3 Restoration of T-cell activation Overcomes tumor resistance to apoptosis Enhances T-cell mediated killing of tumor cells AGEN34 is a novel, fully human monoclonal immunoglobulin G4 (IgG4) designed to block PD- from interacting with its ligands PD-L and PD-L. AGEN34 blocks interaction with PD-L and PD-L, mediates human T-cell activation in vitro, and activates T cells in vivo in non-human primates. 7 The current study (NCT34699) evaluates AGEN34 in patients with advanced, refractory malignancies, including cervical cancer. Here, we present an interim analysis of the data collected, cleaned, and analyzed as of July 6, 8. OBJECTIVE Phase : To assess the safety and tolerability of different AGEN34 dose levels in patients with metastatic and/or locally advanced solid tumors in order to determine recommended phase dosing Phase : To assess the efficacy and safety of AGEN34 in patients with recurrent, unresectable, or metastatic cervical cancer METHODS This is a phase, open-label, dose-escalation trial of AGEN34 with a phase expansion in patients with select solid tumors currently ongoing. Primary outcome measures were to assess: Phase : Occurrence of dose-limiting toxicities (DLTs) in subjects in dose escalation during the first days of treatment Phase : Best overall response (BOR), according to Response Evaluation Criteria in Solid Tumors, version. (RECIST.), as determined by an independent endpoint review committee (IERC), in patients with advanced cervical cancer Secondary outcome measures were to assess: Safety and tolerability, PK profile, and immunogenicity BOR, duration of response, progression-free survival, and overall survival (with a -year time frame; phase only), per RECIST., as determined by the IERC (phase only) and investigator (both phases). Exploratory outcome measures were to: Evaluate biological responses to AGEN34 in blood/serum Association of PD-L expression with clinical responses (phase ) Phase : Dose Escalation for Recommended Phase Dosing Phase consisted of a standard 3+3 dose-escalation design with the following escalating dose levels and schedules: Part A:, 3, and administered intravenously Part B: 6 and administered intravenously AGEN34 was administered for up to years or until confirmed drug or withdrawal from the trial occurred. Eligible patients included male and female patients (aged 8 years) with a histologically or cytologically confirmed diagnosis of a metastatic or locally advanced solid tumor for which no standard therapy is available or standard therapy has failed. Phase : Expansion in With Recurrent, Unresectable, or Metastatic Cervical Cancer The recommended phase dosing of AGEN34, established in phase, is to be administered for a maximum of years or until confirmed drug or withdrawal from the trial occurs. For the phase portion of the study, an Independent Data Management Committee was established to evaluate safety and efficacy, and an IERC was established to adjudicate tumor response. Inclusion and exclusion criteria were the same as phase, except for the following: were female (aged 8 years) and must have a histologically or cytologically confirmed diagnosis of a metastatic or locally advanced, unresectable squamous-cell carcinoma, adenosquamous carcinoma, or adenocarcinoma of the cervix. must have relapsed after a platinum-containing doublet administered for treatment of advanced. must have measurable on imaging based on RECIST.. The null hypothesis to be tested is that the objective response rate (ORR) does not exceed % (H: ORR %), and the alternative hypothesis is that the ORR is greater than % (H: ORR >%). ORR was defined as the proportion of patients with confirmed BOR of partial or complete response. PHASE RESULTS Patient Disposition and Demographics As of July 6, 8, a total of 5 patients had been recruited, enrolled, and treated in the phase dose-escalation portion of the study, with each dosage group comprising patients. Dosages: mg/kg,,,, Demographics and baseline characteristics of the phase cohorts are detailed in Table, with patient disposition detailed in Table. Median (range) time since the most recent recurrence/progression to the administration of the first dose was.7 (.,.) months. 7% of the subjects had a ECOG Performance Status of. Table. Phase Demographics and Baseline Characteristics mg/kg N=5 Median Min, max 3, 77 3, 76 39, 73 4, 7 5, 79 3, 79 Females, n (%) 8 (8%) 8 (8%) 9 (9%) 9 (9%) 8 (8%) 4 (84%) Table. Phase Patient Disposition Discontinued study, n (%) mg/kg N=5 4 (4%) 3 (3%) (%) (%) (%) (%) Death 3 (3%) (%) (%) 6 (%) Withdrawal of consent Continuing in study at time of data extract, n (%) (%) (%) (%) 3 (6%) 6 (6%) 7 (7%) 9 (9%) 9 (9%) 8 (8%) 39 (78%) Median Min, max 49, , , 3 57, 5, 57 45, 464 No DLTs were observed in the phase dose-escalation phase. A summary of the treatment-emergent adverse event (TEAE) profile is provided in Table 3. The most common TEAEs were fatigue (), nausea (), decreased appetite (n=5), diarrhea (n=4), and vomiting (n=3). Table 3. Phase Treatment-Emergent Adverse Events, n (%) mg/kg N=5 TEAEs (%) (%) (%) (%) 9 (9%) 49 (98%) Grade 3 7 (7%) 5 (5%) 6 (6%) 5 (5%) 3 (3%) 6 (5%) Serious 5 (5%) 4 (4%) (%) 5 (5%) (%) 7 (34%) (%) (%) (%) 5 (%) death (%) (%) 3 (6%) TRAEs 8 (8%) 8 (8%) (%) 6 (6%) 4 (4%) 36 (7%) Grade 3 (%) (%) (%) (%) (%) 7 (4%) Serious (%) 4 (4%) (%) 7 (4%) (%) (4%) death (%) (%) TEAE, treatment-emergent adverse event; TRAE, treatment-related adverse event. 36 patients experienced a TEAE that was considered related to treatment (TRAE), with only patients experiencing a TRAE that resulted in treatment ( : hepatitis, n=; pneumonitis, n=). The TRAE of hepatitis was fatal in a 76-year-old Asian female. 7 patients reported a TEAE of grade 3 considered related to treatment: mg/kg : patient (pneumonitis) : patients ( hepatitis, blood alkaline phosphatase increased, blood bilirubin increased, aspartate aminotransferase increased, alanine aminotransferase increased; dyspnea) : patients ( rash; diarrhea) : patient (nausea, vomiting, hyperglycemia) : patient (colitis) 7 patients reported serious TEAEs, with gastrointestinal disorders being the most common system organ class (n=5 patients). The most common preferred term was pneumonitis ( mg/kg, n=;, n=). other serious TEAEs occurred in < patients. 8 patients reported serious TRAEs: mg/kg : patients ( pneumonitis; diarrhea) : 4 patients ( pneumonitis; adrenal insufficiency; hepatitis; confusional state) : patients ( nausea and vomiting; rash) Of the 38 evaluable patients, 3 partial responses (all confirmed) were noted in patients with cervical, ovarian, and breast cancers in the -mg/kg and 3-mg/kg cohorts (Table 4). Table 4. Phase Summary of Best Overall Response by Treatment, n mg/kg n=8 N=38 Complete response Partial response (5.46) (.38 and 9.97) 3 Stable Overall response determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Numbers in parentheses represent the duration of best overall response for each patient in months. Tumor type where responses were seen: ovarian cancer, cervical cancer, and breast cancer. Pharmacokinetic Profile Serum concentrations of AGEN34 collected and analyzed to date are displayed in Figure. Figure. Pharmacokinetic Profile of AGEN34 Serum AGEN34 Serum AGEN34.. Cohort ( mg/kg ) Cohort 3 ( ) Serum AGEN34 Serum AGEN34 Serum concentrations plotted on a log scale. Arrows indicate estimated treatment day ( or ) Cohort ( ) 5 5 Cohort 4 ( ) Receptor occupancy on circulating effector CD4 + and CD8 + T cells was quantified by flow cytometry (Figure 3). Receptor occupancy values were similar 4 hours post-dosing to those observed weeks later. Furthermore, receptor occupancy levels were consistent over all doses tested and all exposures measured. These results suggest that the maximum possible receptor occupancy was obtained. Figure 3. Receptor Occupancy (RO) on CD4 + and CD8 + T Effector Memory Cells Following Treatment With AGEN CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) mg/kg For each subject in each dose cohort, receptor occupancy is plotted as a function of the concentration of AGEN34 in the serum measured at the same timepoint. Receptor occupancy and AGEN34 concentrations were measured 4 hours after the initial dosing, and weeks later (prior to the administration of the second dose). For each graph, cohorts are indicated by the color and the shape. Concentrations of AGEN34 are indicated on the x-axis, and receptor occupancy is indicated on the y-axis. PHASE RESULTS The recommended phase dosing established in phase was AGEN34. As of July 6, 8, a total of 6 females with cervical cancer had been recruited, enrolled, and treated in the phase portion of the study (Table 5). Table 5. Phase Demographics and Baseline Characteristics Median (min, max) 6.5 (36, 67) ECOG performance status, n (%) (6.7%) 5 (83.3%) Time since progression (months) Median (min, max).69 (.8, 8.) Current cancer status, n (%) Localized (6.7%) Metastatic 5 (83.3%) ECOG, Eastern Cooperative Oncology Group. patient discontinued from the study treatment due to progression (Table 6). Table 6. Phase Patient Disposition Discontinued from study treatment, n (%) (6.7%) Disease progression (6.7%) Continuing on study at time of data extract, n (%) 5 (83.3%) Median (min, max) 38.5 (74, 77) At time of data cut-off, a total of 9 patients (3 from phase and 6 from phase ) with cervical cancer had been treated with AGEN34. The summary of BOR is listed in Table 7. Table 7. Summary of Best Overall Response in Cervical Cancer Phase n= Phase Phase Total Cervical Cancer Population N=9 Complete response Partial response () * Stable (.8) (.6) 3 4 Not evaluable Overall response determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Numbers in parentheses represent the duration of best overall response for each patient in months. Figure 4 depicts the duration of progression-free survival at the time of data cut-off. Posterior data cut of September 8, 8, showed there is a new confirrmed PR out of the 6 patients in the cervical cancer expansion cohort; therefore, there are PRs in the overall cervical cancer group. Figure 4. Progression-Free Survival in Cervical Cancer Received Drug (PR) (Phase ) (Phase ) (Phase ) Partial response starts Ongoing in study Months Progression-free survival determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Best overall response indicated in parentheses. PD, progressive ; PR, partial response; SD, stable. At time of data cut-off, no DLTs were reported for any of the patients, and 5 of the 6 patients reported a TEAE (Table 8). TEAEs occurring in > patient included abdominal pain (n=3) and pyrexia (). 3 patients experienced a TEAE of grade 3, including anemia, international normalized ratio increased, back pain, diarrhea, and abdominal pain. Serious TEAEs occurred in 3 patients, including colitis, abdominal pain, back pain, diarrhea, and pyrexia (n= each). The events of diarrhea and colitis were considered related to treatment. patient had a TRAE of grade 3, which was serious (diarrhea). No TEAEs led to treatment or study withdrawal. Table 8. Phase Summary of Treatment-Emergent Adverse Events, n (%) TEAEs 5 (83.3%) Grade 3 3 (5.%) Serious 3 (5.%) TRAEs 3 (5.%) Grade 3 (6.7%) Serious (6.7%) TEAE, treatment-emergent adverse event; TRAE: treatment-related adverse event. DISCUSSION AGEN34 is a pharmacologically active, well-tolerated PD- antagonist antibody, demonstrating early signs of clinical activity in the phase and portions of the study in patients with advanced refractory malignancies including cervical cancer. No DLT were observed in the Phase I portion of the study. PK determinations are under ongoing analysis as more information is gathered. Receptor occupancy results from the current study are comparable to data available for commercially available PD- antagonists. The phase expansion in patients with relapsed/refractory cervical cancer is continuing to recruit (NCT34699). References. Topalian SL, Hodi FS, Brahmer JR, et al. N Engl J Med. ;366(6): Jia M, Feng W, Kang S, et al. J Thorac Dis. 5;7(3): Zhang T, Xie J, Arai S, et al. Oncotarget. 6;7(45): Huang G, Sun X, Liu D, et al. Oncotarget. 7;9(3): Latchman Y, Wood CR, Chernova T, et al. Nat Immunol. ;(3): ie SR, Zhang W, Fuse S, et al. J Immunol. ;86(): Chand D, Savistky D, Gonzalez A, et al. [Abstract]. J Immunother Cancer. 7; 5(Suppl ):P3. Disclosures G Yuan, CD Dupont, M Lim, AM Gonzalez, D Savitsky, S Coulter, O Shebanova, E Dow, W Ortuzar, J Buell, RB Stein, H Youssoufian: Agenus Inc.: current or former employment/ consultancy and stock ownership. Acknowledgments Medical writing and editorial support were provided by The Medicine Group, LLC (New Hope, PA, USA), which was funded by Agenus Inc. (Lexington, MA, USA). The licensed antibody AGEN34 was originally developed under a Collaborative Research and Development Agreement between Ludwig Cancer Research, 4-Antibody AG (now Agenus Switzerland Inc.) and Recepta Biopharma S.A. Recepta Biopharma S.A. has exclusive rights to this antibody in Brazil and five other South American countries. Funding This analysis was funded by Agenus Inc. (Lexington, MA, USA). Correspondence Waldo Ortuzar: waldo.ortuzar@agenusbio.com Presented at the 8 Annual Congress of the European Society for Medical Oncology (ESMO), October 9 3, 8, in Munich, Germany.

2 BACKGROUND Inhibition of the programmed cell death protein- (PD-) pathways by blockade of receptor-ligand interactions has been demonstrated in numerous clinical trials to result in objective clinical response and increased survival in solid tumor indications (Figure ). -4 PD- signaling is mediated by ligands: PD- ligand (PD-L, or CD74) and PD- ligand (PD-L, or CD73). 5 Upon binding to PD-L or PD-L, PD- signaling in T cells can potently attenuate TCR signaling, resulting in diminished cytokine and proliferative responses, reduced T-effector cell cytolytic activity, and impaired central memory T-cell differentiation. 6 Figure. Pathways Affected by PD- Restoration of T-cell activation Overcomes tumor resistance to apoptosis Enhances T-cell mediated killing of tumor cells 3 AGEN34 is a novel, fully human monoclonal immunoglobulin G4 (IgG4) designed to block PD- from interacting with its ligands PD-L and PD-L. AGEN34 blocks interaction with PD-L and PD-L, mediates human T-cell activation in vitro, and activates T cells in vivo in non-human primates. 7 The current study (NCT34699) evaluates AGEN34 in patients with advanced, refractory malignancies, including cervical cancer. Here, we present an interim analysis of the data collected, cleaned, and analyzed as of July 6, 8. OBJECTIVE Phase : To assess the safety and tolerability of different AGEN34 dose levels in patients with metastatic and/or locally advanced solid tumors in order to determine recommended phase dosing Phase : To assess the efficacy and safety of AGEN34 in patients with recurrent, unresectable, or metastatic cervical cancer METHODS This is a phase, open-label, dose-escalation trial of AGEN34 with a phase expansion in patients with select solid tumors currently ongoing. Primary outcome measures were to assess: Phase : Occurrence of dose-limiting toxicities (DLTs) in subjects in dose escalation during the first days of treatment Phase : Best overall response (BOR), according to Response Evaluation Criteria in Solid Tumors, version. (RECIST.), as determined by an independent endpoint review committee (IERC), in patients with advanced cervical cancer Secondary outcome measures were to assess: Safety and tolerability, PK profile, and immunogenicity BOR, duration of response, progression-free survival, and overall survival (with a -year time frame; phase only), per RECIST., as determined by the IERC (phase only) and investigator (both phases).

3 Exploratory outcome measures were to: Evaluate biological responses to AGEN34 in blood/serum Association of PD-L expression with clinical responses (phase ) Phase : Dose Escalation for Recommended Phase Dosing Phase consisted of a standard 3+3 dose-escalation design with the following escalating dose levels and schedules: Part A:, 3, and administered intravenously Part B: 6 and administered intravenously AGEN34 was administered for up to years or until confirmed drug or withdrawal from the trial occurred. Eligible patients included male and female patients (aged 8 years) with a histologically or cytologically confirmed diagnosis of a metastatic or locally advanced solid tumor for which no standard therapy is available or standard therapy has failed. Phase : Expansion in With Recurrent, Unresectable, or Metastatic Cervical Cancer The recommended phase dosing of AGEN34, established in phase, is to be administered for a maximum of years or until confirmed drug or withdrawal from the trial occurs. For the phase portion of the study, an Independent Data Management Committee was established to evaluate safety and efficacy, and an IERC was established to adjudicate tumor response. Inclusion and exclusion criteria were the same as phase, except for the following: were female (aged 8 years) and must have a histologically or cytologically confirmed diagnosis of a metastatic or locally advanced, unresectable squamous-cell carcinoma, adenosquamous carcinoma, or adenocarcinoma of the cervix. must have relapsed after a platinum-containing doublet administered for treatment of advanced. must have measurable on imaging based on RECIST.. The null hypothesis to be tested is that the objective response rate (ORR) does not exceed % (H: ORR %), and the alternative hypothesis is that the ORR is greater than % (H: ORR >%). ORR was defined as the proportion of patients with confirmed BOR of partial or complete response. PHASE RESULTS Patient Disposition and Demographics As of July 6, 8, a total of 5 patients had been recruited, enrolled, and treated in the phase dose-escalation portion of the study, with each dosage group comprising patients. Dosages: mg/kg,,,, Demographics and baseline characteristics of the phase cohorts are detailed in Table, with patient disposition detailed in Table. Median (range) time since the most recent recurrence/progression to the administration of the first dose was.7 (.,.) months. 7% of the subjects had a ECOG Performance Status of. Table. Phase Demographics and Baseline Characteristics mg/kg N=5 Median Min, max 3, 77 3, 76 39, 73 4, 7 5, 79 3, 79 Females, n (%) 8 (8%) 8 (8%) 9 (9%) 9 (9%) 8 (8%) 4 (84%)

4 Table. Phase Patient Disposition Discontinued study, n (%) mg/kg N=5 4 (4%) 3 (3%) (%) (%) (%) (%) Death 3 (3%) (%) (%) 6 (%) Withdrawal of consent Continuing in study at time of data extract, n (%) (%) (%) (%) 3 (6%) 6 (6%) 7 (7%) 9 (9%) 9 (9%) 8 (8%) 39 (78%) Median Min, max 49, , , 3 57, 5, 57 45, 464 No DLTs were observed in the phase dose-escalation phase. A summary of the treatment-emergent adverse event (TEAE) profile is provided in Table 3. The most common TEAEs were fatigue (), nausea (), decreased appetite (n=5), diarrhea (n=4), and vomiting (n=3). Table 3. Phase Treatment-Emergent Adverse Events, n (%) mg/kg N=5 TEAEs (%) (%) (%) (%) 9 (9%) 49 (98%) Grade 3 7 (7%) 5 (5%) 6 (6%) 5 (5%) 3 (3%) 6 (5%) Serious 5 (5%) 4 (4%) (%) 5 (5%) (%) 7 (34%) (%) (%) (%) 5 (%) death (%) (%) 3 (6%) TRAEs 8 (8%) 8 (8%) (%) 6 (6%) 4 (4%) 36 (7%) Grade 3 (%) (%) (%) (%) (%) 7 (4%) Serious (%) 4 (4%) (%) 7 (4%) (%) (4%) death (%) (%) TEAE, treatment-emergent adverse event; TRAE, treatment-related adverse event. 36 patients experienced a TEAE that was considered related to treatment (TRAE), with only patients experiencing a TRAE that resulted in treatment ( : hepatitis, n=; pneumonitis, n=). The TRAE of hepatitis was fatal in a 76-year-old Asian female. 7 patients reported a TEAE of grade 3 considered related to treatment: mg/kg : patient (pneumonitis) : patients ( hepatitis, blood alkaline phosphatase increased, blood bilirubin increased, aspartate aminotransferase increased, alanine aminotransferase increased; dyspnea) : patients ( rash; diarrhea) : patient (nausea, vomiting, hyperglycemia) : patient (colitis) 7 patients reported serious TEAEs, with gastrointestinal disorders being the most common system organ class (n=5 patients). The most common preferred term was pneumonitis ( mg/kg, n=;, n=). other serious TEAEs occurred in < patients. 8 patients reported serious TRAEs: mg/kg : patients ( pneumonitis; diarrhea) : 4 patients ( pneumonitis; adrenal insufficiency; hepatitis; confusional state) : patients ( nausea and vomiting; rash) Of the 38 evaluable patients, 3 partial responses (all confirmed) were noted in patients with cervical, ovarian, and breast cancers in the -mg/kg and 3-mg/kg cohorts (Table 4). Serum concentrations plotted on a log scale. Arrows indicate estimated treatment day Receptor occupancy on circulating effect quantified by flow cytometry (Figure 3). R similar 4 hours post-dosing to those obse receptor occupancy levels were consisten exposures measured. These results sugge receptor occupancy was obtained. Figure 3. Receptor Occupancy (RO) on Memory Cells Following Treatment Wi CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) For each subject in each dose cohort, receptor occupancy is plotted as a function of th same timepoint. Receptor occupancy and AGEN34 concentrations were measured 4 administration of the second dose). For each graph, cohorts are indicated by the color x-axis, and receptor occupancy is indicated on the y-axis.

5 Table 4. Phase Summary of Best Overall Response by Treatment, n mg/kg n=8 N=38 Complete response Partial response (5.46) (.38 and 9.97) 3 Stable Overall response determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Numbers in parentheses represent the duration of best overall response for each patient in months. Tumor type where responses were seen: ovarian cancer, cervical cancer, and breast cancer. Pharmacokinetic Profile Serum concentrations of AGEN34 collected and analyzed to date are displayed in Figure. Figure. Pharmacokinetic Profile of AGEN34 Serum AGEN34 Serum AGEN34.. Cohort ( mg/kg ) Cohort 3 ( ) Serum AGEN34 Serum AGEN34 Serum concentrations plotted on a log scale. Arrows indicate estimated treatment day ( or ) Cohort ( ) 5 5 Cohort 4 ( ) Receptor occupancy on circulating effector CD4 + and CD8 + T cells was quantified by flow cytometry (Figure 3). Receptor occupancy values were similar 4 hours post-dosing to those observed weeks later. Furthermore, receptor occupancy levels were consistent over all doses tested and all exposures measured. These results suggest that the maximum possible receptor occupancy was obtained. Figure 3. Receptor Occupancy (RO) on CD4 + and CD8 + T Effector Memory Cells Following Treatment With AGEN CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) mg/kg For each subject in each dose cohort, receptor occupancy is plotted as a function of the concentration of AGEN34 in the serum measured at the same timepoint. Receptor occupancy and AGEN34 concentrations were measured 4 hours after the initial dosing, and weeks later (prior to the administration of the second dose). For each graph, cohorts are indicated by the color and the shape. Concentrations of AGEN34 are indicated on the x-axis, and receptor occupancy is indicated on the y-axis. PHASE RESULTS The recommended phase dosing establ AGEN34. As of July 6, 8, a total of 6 females wi recruited, enrolled, and treated in the phas Table 5. Phase Demographics and B Median (min, max) ECOG performance status, n (%) Time since progression (months) Median (min, max) Current cancer status, n (%) Localized Metastatic ECOG, Eastern Cooperative Oncology Group. patient discontinued from the study trea progression (Table 6). Table 6. Phase Patient Disposition Discontinued from study treatment, n (%) Disease progression Continuing on study at time of data extract, n (%) Median (min, max) At time of data cut-off, a total of 9 patient phase ) with cervical cancer had been tre The summary of BOR is listed in Table 7. Table 7. Summary of Best Overall Resp Phase n= Phase Complete response Partial response () Stable (.8) Not evaluable Overall response determined by Response Evaluation Criteria in Solid Tumors (RECIST) overall response for each patient in months. Figure 4 depicts the duration of progressi data cut-off. Posterior data cut of September 8, confirrmed PR out of the 6 patients in t cohort; therefore, there are PRs in the Figure 4. Progression-Free Survival in Received Drug Mont Progression-free survival determined by Response Evaluation Criteria in Solid Tumors (R PD, progressive ; PR, partial response; SD, stable.

6 timated treatment day ( or ). ating effector CD4 + and CD8 + T cells was igure 3). Receptor occupancy values were those observed weeks later. Furthermore, e consistent over all doses tested and all sults suggest that the maximum possible ned. cy (RO) on CD4 + and CD8 + T Effector tment With AGEN CD, 4 hrs post-dose AGEN34 (µg/ml) CD, Pre-dose. AGEN34 (µg/ml) mg/kg ted as a function of the concentration of AGEN34 in the serum measured at the ions were measured 4 hours after the initial dosing, and weeks later (prior to the ndicated by the color and the shape. Concentrations of AGEN34 are indicated on the PHASE RESULTS The recommended phase dosing established in phase was AGEN34. As of July 6, 8, a total of 6 females with cervical cancer had been recruited, enrolled, and treated in the phase portion of the study (Table 5). Table 5. Phase Demographics and Baseline Characteristics Median (min, max) 6.5 (36, 67) ECOG performance status, n (%) (6.7%) 5 (83.3%) Time since progression (months) Median (min, max).69 (.8, 8.) Current cancer status, n (%) Localized (6.7%) Metastatic 5 (83.3%) ECOG, Eastern Cooperative Oncology Group. patient discontinued from the study treatment due to progression (Table 6). Table 6. Phase Patient Disposition Discontinued from study treatment, n (%) (6.7%) Disease progression (6.7%) Continuing on study at time of data extract, n (%) 5 (83.3%) Median (min, max) 38.5 (74, 77) At time of data cut-off, a total of 9 patients (3 from phase and 6 from phase ) with cervical cancer had been treated with AGEN34. The summary of BOR is listed in Table 7. Table 7. Summary of Best Overall Response in Cervical Cancer Phase n= Phase Phase Total Cervical Cancer Population N=9 Complete response Partial response () * Stable (.8) (.6) 3 4 Not evaluable Overall response determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Numbers in parentheses represent the duration of best overall response for each patient in months. Figure 4 depicts the duration of progression-free survival at the time of data cut-off. Posterior data cut of September 8, 8, showed there is a new confirrmed PR out of the 6 patients in the cervical cancer expansion cohort; therefore, there are PRs in the overall cervical cancer group. Figure 4. Progression-Free Survival in Cervical Cancer Received Drug (PR) (Phase ) (Phase ) (Phase ) Partial response starts Ongoing in study Months Progression-free survival determined by Response Evaluation Criteria in Solid Tumors (RECIST) v.. Best overall response indicated in parentheses. PD, progressive ; PR, partial response; SD, stable. At time of data cut-off, no DLTs were repo 5 of the 6 patients reported a TEAE (Tabl TEAEs occurring in > patient included a (). 3 patients experienced a TEAE of grade international normalized ratio increased, abdominal pain. Serious TEAEs occurred in 3 patients, inc back pain, diarrhea, and pyrexia (n= eac The events of diarrhea and colitis were patient had a TRAE of grade 3, which No TEAEs led to treatment discontinuatio Table 8. Phase Summary of Treatmen, n (%) TEAEs Grade 3 Serious TRAEs Grade 3 Serious TEAE, treatment-emergent adverse event; TRAE: treatment-related adverse event. DISCUSSION AGEN34 is a pharmacologically act antagonist antibody, demonstrating ea in the phase and portions of the st refractory malignancies including cerv No DLT were observed in the Phase I PK determinations are under ongoing is gathered. Receptor occupancy results from the to data available for commercially ava The phase expansion in patients wit cancer is continuing to recruit (NCT3 References. Topalian SL, Hodi FS, Brahmer JR, et al. N Engl J. Jia M, Feng W, Kang S, et al. J Thorac Dis. 5; 3. Zhang T, Xie J, Arai S, et al. Oncotarget. 6;7(4 4. Huang G, Sun X, Liu D, et al. Oncotarget. 7;9( 5. Latchman Y, Wood CR, Chernova T, et al. Nat Imm 6. ie SR, Zhang W, Fuse S, et al. J Immunol. 7. Chand D, Savistky D, Gonzalez A, et al. [Abstract 5(Suppl ):P3. Disclosures G Yuan, CD Dupont, M Lim, AM Gonzalez, D Savitsk W Ortuzar, J Buell, RB Stein, H Youssoufian: Agenus consultancy and stock ownership. Acknowledgments Medical writing and editorial support were provided Hope, PA, USA), which was funded by Agenus Inc. ( antibody AGEN34 was originally developed under Development Agreement between Ludwig Cancer R Switzerland Inc.) and Recepta Biopharma S.A. Rece rights to this antibody in Brazil and five other South Funding This analysis was funded by Agenus Inc. (Lexington Correspondence Waldo Ortuzar: waldo.ortuzar@agenusbio.com Presented at the 8 Annual Congress of the Europ (ESMO), October 9 3, 8, in Munich, Germany.

7 sing established in phase was females with cervical cancer had been in the phase portion of the study (Table 5). ics and Baseline Characteristics 6.5 (36, 67) (6.7%) 5 (83.3%).69 (.8, 8.) (6.7%) 5 (83.3%) e study treatment due to osition n (%) (6.7%) (6.7%) extract, n (%) 5 (83.3%) 38.5 (74, 77) of 9 patients (3 from phase and 6 from ad been treated with AGEN34. in Table 7. erall Response in Cervical Cancer Phase mg/kg Phase Total Cervical Cancer Population N=9 * (.8) (.6) 3 4 Solid Tumors (RECIST) v.. Numbers in parentheses represent the duration of best of progression-free survival at the time of ber 8, 8, showed there is a new atients in the cervical cancer expansion PRs in the overall cervical cancer group. urvival in Cervical Cancer SD) (PR) (Phase ) (Phase ) (Phase ) Partial response starts Ongoing in study Months iteria in Solid Tumors (RECIST) v.. Best overall response indicated in parentheses. e. At time of data cut-off, no DLTs were reported for any of the patients, and 5 of the 6 patients reported a TEAE (Table 8). TEAEs occurring in > patient included abdominal pain (n=3) and pyrexia (). 3 patients experienced a TEAE of grade 3, including anemia, international normalized ratio increased, back pain, diarrhea, and abdominal pain. Serious TEAEs occurred in 3 patients, including colitis, abdominal pain, back pain, diarrhea, and pyrexia (n= each). The events of diarrhea and colitis were considered related to treatment. patient had a TRAE of grade 3, which was serious (diarrhea). No TEAEs led to treatment or study withdrawal. Table 8. Phase Summary of Treatment-Emergent Adverse Events, n (%) TEAEs 5 (83.3%) Grade 3 3 (5.%) Serious 3 (5.%) TRAEs 3 (5.%) Grade 3 (6.7%) Serious (6.7%) TEAE, treatment-emergent adverse event; TRAE: treatment-related adverse event. DISCUSSION AGEN34 is a pharmacologically active, well-tolerated PD- antagonist antibody, demonstrating early signs of clinical activity in the phase and portions of the study in patients with advanced refractory malignancies including cervical cancer. No DLT were observed in the Phase I portion of the study. PK determinations are under ongoing analysis as more information is gathered. Receptor occupancy results from the current study are comparable to data available for commercially available PD- antagonists. The phase expansion in patients with relapsed/refractory cervical cancer is continuing to recruit (NCT34699). References. Topalian SL, Hodi FS, Brahmer JR, et al. N Engl J Med. ;366(6): Jia M, Feng W, Kang S, et al. J Thorac Dis. 5;7(3): Zhang T, Xie J, Arai S, et al. Oncotarget. 6;7(45): Huang G, Sun X, Liu D, et al. Oncotarget. 7;9(3): Latchman Y, Wood CR, Chernova T, et al. Nat Immunol. ;(3): ie SR, Zhang W, Fuse S, et al. J Immunol. ;86(): Chand D, Savistky D, Gonzalez A, et al. [Abstract]. J Immunother Cancer. 7; 5(Suppl ):P3. Disclosures G Yuan, CD Dupont, M Lim, AM Gonzalez, D Savitsky, S Coulter, O Shebanova, E Dow, W Ortuzar, J Buell, RB Stein, H Youssoufian: Agenus Inc.: current or former employment/ consultancy and stock ownership. Acknowledgments Medical writing and editorial support were provided by The Medicine Group, LLC (New Hope, PA, USA), which was funded by Agenus Inc. (Lexington, MA, USA). The licensed antibody AGEN34 was originally developed under a Collaborative Research and Development Agreement between Ludwig Cancer Research, 4-Antibody AG (now Agenus Switzerland Inc.) and Recepta Biopharma S.A. Recepta Biopharma S.A. has exclusive rights to this antibody in Brazil and five other South American countries. Funding This analysis was funded by Agenus Inc. (Lexington, MA, USA). Correspondence Waldo Ortuzar: waldo.ortuzar@agenusbio.com Presented at the 8 Annual Congress of the European Society for Medical Oncology (ESMO), October 9 3, 8, in Munich, Germany.