The Society for Immunotherapy of Cancer (SITC) is pleased to present highlights of the latest advances in cancer immunotherapy emerging from the 2025 AACR Annual Meeting.
2025 Scientific Highlights
Addition of perioperative pembrolizumab to the standard of care significantly improves event free survival in patients with head and neck squamous cell carcinoma
CT001. Neoadjuvant and adjuvant pembrolizumab plus standard of care (SOC) in resectable locally advanced head and neck squamous cell carcinoma (LA HNSCC): Phase 3 KEYNOTE-689 study
Ravindra Uppaluri (Brigham and Women's Hospital and Harvard Medical School/Dana-Farber Cancer Institute, Boston, MA, USA) presented results from the first interim analysis of KENOTE-689, a Phase 3 clinical study comparing the addition of neoadjuvant and adjuvant pembrolizumab (pembro) to the standard of care (SOC; surgery followed by postoperative radiotherapy +/- concurrent chemotherapy) to SOC for patients with resectable locally advanced head and neck squamous cell carcinoma (LA HNSCC). 363 patients received perioperative pembro + SOC and 351 patients received SOC. 89% of patients in the perioperative pembro + SOC arm received surgery, compared to 88% of patients in the SOC arm. At a median follow-up of 38.3 months, addition of perioperative pembro to the SOC significantly improved event free survival. Among patients with PD-L1 CPS >= 10, median EFS was 59.7 months for patients who received pembro + SOC (n=234) compared to 26.9 months for patients who received SOC (n=231; HR 0.66, p=0.0022). Pembro + SOC also improved EFS compared to SOC among patients with PD-L1 CPS >= 1, with median EFS of 59.7 months (n=347) vs. 29.6 months (n=335; HR 0.70, p=0.0014), respectively. In the entire patient population, EFS among patients who were treated with pembro + SOC was 51.8 months, compared to 30.4 months for patients who received SOC (HR 0.73, p=0.0041). Perioperative pembro was associated with EFS benefits in all patient subgroups observed. No patients in the SOC arm achieved a major pathological response (mPR), but in the perioperative pembro + SOC arm, 13.5% of patients with CPS >= 10, 9.8% of patients with CPS >= 1, and 9.3% of all patients achieved an mPR. A significant improvement in overall survival (OS) was not observed: median OS was not reached in the pembro + SOC arm and 61.8 months in the SOC arm (HR 0.72, p=0.02), and additional follow-up is ongoing. No new safety signals were observed in the pembro + SOC arm, and incidences of treatment-related adverse events (TRAEs) and TRAEs of Grade 3 or higher were similar between both arms. 4 deaths from TRAEs occurred in the pembro + SOC arm, compared to 1 death in the SOC arm, and immune-mediated AEs occurred in 43.2% of patients who received pembro + SOC. Although survival data are immature , the significant improvements in EFS major pathological response rates suggest that the addition of perioperative pembrolizumab to surgery and adjuvant (chemo)radiotherapy may become a new standard of care for the treatment of patients with resectable locally advanced head and neck squamous cell carcinoma.
Using circulating tumor DNA to determine adjuvant treatment with PD-1 blockade for mismatch repair deficient solid tumors
CT002. Circulating tumor DNA status to direct adjuvant immunotherapy for mismatch repair deficient tumors
Yelena Yuriy Janjigian (Memorial Sloan Kettering Cancer Center, New York, NY, USA) presented a prospective interventional study examining the efficacy of adjuvant PD-1 blockade in preventing recurrence in patients with resected mismatch repair deficient (MMRd) tumors who are minimal residual disease (MRD) positive (detected by circulating tumor DNA (ctDNA)) after surgery and standard of care adjuvant therapy. This study builds on prior results indicating that MMRd tumors are sensitive to immune checkpoint blockade (ICB) and tests the model of post-operative ctDNA positivity as a predictor of metastatic disease. After receiving resection and standard of care adjuvant therapy, patients with microsatellite instability-high (MSI-H) or MMRd solid tumors were tested for the presence of ctDNA. Of 174 patients representing 16 different tumor types, 22 patients (11%) were ctDNA positive, and 13 of those received adjuvant therapy with pembrolizumab (pembro). 6 ctDNA-positive patients did not receive therapy due to the presence of disease before receiving adjuvant pembro, and 152 patients were ctDNA-negative. At a median follow-up of 32.1 months, the 24-month recurrence-free survival (RFS) rate was 94% among patients who initially tested as ctDNA-negative and 64% for patients who were ctDNA-positive and received pembro. Among the 13 ctDNA-positive patients who received adjuvant pembro, 11 (85%) were ctDNA-negative after 6 months of treatment with pembro, and 100% achieved ctDNA clearance after 9 months. Five of the 13 patients who received pembro (38%) experienced disease recurrence even though they had previously achieved ctDNA clearance, potentially identifying a subgroup of patients who are at high risk for disease recurrence. Survival data are immature, but 24-month overall survival (OS) rates were 98% in patients who were originally ctDNA-negative, 92% in patients who were originally ctDNA-positive and received adjuvant pembro, and 67% in patients who were ctDNA-positive and experienced disease recurrence before receiving adjuvant pembro. Although a subset of high-risk patients will likely need additional interventions to prevent disease recurrence, this study provides the first line of evidence that adjuvant therapy with ICB can clear MRD as indicated by ctDNA clearance. This study also further supports ctDNA status as a useful tool to guide treatment decisions surrounding early-stage resectable MSI-H/MMRd cancers.
Clinical efficacy and safety of neoadjuvant durvalumab for mismatch repair deficient solid tumors
CT003. Non operative management of mismatch repair deficient tumors
Andrea Cercek (Memorial Sloan Kettering Cancer Center, New York, NY, USA) presented a study of the efficacy of neoadjuvant immune checkpoint blockade (ICB), specifically anti-PDL1, for early-stage mismatch repair deficient (MMRd) solid tumors. Preliminary results from a prior clinical trial have indicated that neoadjuvant treatment with ICB for locally advanced MMRd rectal cancers results in a 100% complete clinical response (cCR) rate with durable responses, eliminating the need for surgery. Multiple studies have indicated that MMRd tumors have a high response rate to ICB regardless of tumor type, suggesting the promising results in MMRd rectal cancers may be extended in a tumor-agnostic manner. Patients with early-stage locally advanced MMRd solid tumors received neoadjuvant durvalumab for 6 months. Patients who achieved a cCR after neoadjuvant therapy underwent a non-operative follow-up every 4 months. Patients who exhibited residual disease after neoadjuvant durvalumab therapy received standard of care treatment for their tumor type. 103 patients with early-stage MMRd solid tumors eligible for curative intent surgery completed treatment with neoadjuvant durvalumab. Cohort 1 consisted of 49 patients with locally advanced MMRd rectal cancer, and Cohort 2 consisted of 54 patients with a variety of MMRd cancers, including colon, gastric/gastroesophageal junction, genitourinary, and gynecologic cancers. Dostarlimab was well-tolerated, with 35% of patients experiencing no adverse events and 5% of patients experiencing grade 3 or 4 adverse events. In Cohort 1, all 49 patients (100%) achieved a cCR. At a median follow-up of 30.2 months, the 2-year recurrence-free survival (RFS) rate was 96%. In Cohort 2, 35/54 patients (65%) achieved a cCR and 19 exhibited an incomplete clinical response. Urothelial, hepatobiliary pancreatic, colon, and gastric cancers were represented among the patients who achieved a cCR. At a median follow-up of 14.9 months, the 2-year RFS rate was 85%, although few patients in Cohort 2 had reached the two-year post-treatment landmark. Across both cohorts, 81% of patients who completed 6 months of treatment achieved cCR, and 79% avoided surgery. Of the 5 recurrences, most took place in patients’ lymph nodes, and one patient experienced a local regrowth. Three of the 4 patients who were rechallenged with PD-1 blockade for recurrent disease achieved a complete response to therapy, and one patient is currently responding to rechallenge with PD-1 blockade. With regards to early indicators of response to neoadjuvant therapy, patients with cCRs exhibited circulating tumor DNA (ctDNA) and tumor bed biopsy clearance in 1.5 and 1.4 months, respectively. Patients with a cCR achieved ctDNA clearance rapidly and remained negative, while patients who exhibited incomplete responses and recurrences lagged in achieving ctDNA negativity. In addition to supporting the use of ctDNA clearance as an indicator of response and recurrence, these data strongly support the use of ICB for the treatment of early-stage MMRd cancers of all types, dramatically reducing the use of current standard of care treatments such as chemotherapy, radiation, and surgery.
Efficacy and safety of olaparib combined with pembrolizumab for homologous recombination repair-mutated or homologous recombination deficient tumors
CT004. KEYLYNK-007: Tumor agnostic trial of olaparib plus pembrolizumab in homologous recombination repair mutation- and homologous recombination deficiency- positive advanced cancers
Timothy A. Yap (University of Texas MD Anderson Cancer Center, Houston, TX, USA) reported data from KEYLYNK-007, a phase 2 tumor-agnostic trial of the poly ADP ribose polymerase inhibitor (PARPi) olaparib in combination with pembrolizumab (pembro) in homologous recombination repair-mutated (HRRm) or homologous recombination deficient (HRD-positive) cancers. This trial builds upon previous studies that indicate that HRRm and HRD cells rely on PARP enzymes for DNA repair. Increased DNA damage promotes immunogenicity and adaptive upregulation of PD-L1; thus, inhibition of PARP enzymes may have a synergistic effect on immune checkpoint inhibitors in HRRm, including BRCA1/2 mutated (BRCA1/2m) or HRD-positive tumors. 332 patients were enrolled in the study. Cohort 1 consisted of 132 patients with BRCA1/2m tumors, excluding ovarian and breast cancers, Cohort 2 consisted of 104 patients with HRRm tumors without BRCA mutations (non-BRCA), and Cohort 3 consisted of 96 patients with wild type HRR HRD-positive tumors. The overall response rate (ORR) was 27% in Cohort 1 (BRCAm) with 11 complete responses (CRs) and a disease control rate of 64.4%. The median duration of response (DOR) was 19.1 months. In Cohort 2 (non-BRCA HRRm) the ORR was 27% with 2 CRs and a disease control rate of 60.3%. Median DOR was 8.3 months. In Cohort 3 (HRR wild type HRD-positive), ORR was 12.5%, with 5 CRs and a DCR of 62.5%. Median DOR was 11.5 months. The 15-month durations of response for Cohorts 1, 2, and 3 were 53.7%, 47.6%, and 41.7%, respectively. Median progression free survival for Cohorts 1, 2, and 3 were 4.4 months, 3.7 months, and 4.1 months, respectively, and median overall survival was 14.0 months, 10.4 months, and 10.8 months, respectively. Responders included patients with advanced breast, ovarian, pancreatic, urothelial, renal, cervical, esophageal, colorectal, and HRRm prostate cancers. 30% of patients in the study experienced treatment-related adverse events (TRAEs) of grades 3 or higher, and 5% of patients discontinued treatment due to TRAEs. The combination of olaparib plus pembrolizumab produced a manageable safety profile and promising clinical results in HRRm/HRD-positive tumors, especially in patients with BRCA1 or BRCA2 mutations. Furthermore, responses were observed in multiple indications where olaparib and/or pembrolizumab are not approved. More time is needed to fully quantify the durability of these responses, and analyses of predictive biomarkers of response are ongoing.
PD-1 inhibitor, penpulimab, in combination with chemotherapy as first-line therapy for metastatic nasopharyngeal carcinoma
CT011. Penpulimab versus placebo in combination with chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal carcinoma: A global, multicenter, randomized, double-blind, phase 3 trial (AK105-304)
Aditya Shreenivas (City of Hope, Los Angeles, CA, USA) reported results from AK105-304, a phase 3 clinical study studying the addition of penpulimab to chemoradiotherapy for the treatment of recurrent or metastatic nasopharyngeal cancer (NPC). NPC is an aggressive cancer and is endemic in certain areas. Treatment is typically chemotherapy combined with immunotherapy. Prior studies have indicated PD-1 inhibitors combined with chemotherapy is effective as first-line treatment for recurrent or metastatic (R/M) NPC. Penpuimab is an anti-PD-1 monoclonal antibody with a modified Fc subunit, limiting binding to the Fc gamma receptor and increasing stability, potentially improving clinical efficacy and safety, compared to other PD-1 inhibitors. Patients in the penpulimab arm (n=144) received penpulimab in combination with chemotherapy placebo arm for six cycles followed by penpulimab for 24 months or until loss of clinical benefits occurred. Patients in the placebo arm (n=147) received placebo in combination with chemotherapy followed by placebo. 68 patients (46%) in the placebo arm crossed over to the penpulimab arm at the end of treatment. Penpulimab significantly increased progression free survival (PFS) compared to placebo, with median PFS of 9.6 months and 7.0 months, respectively (HR 0.45, p<0.0001), and 24-month PFS rates were 22.8% for the penpulimab arm, compared to 5.5% in the placebo arm. Penpulimab was associated with significant PFS benefits in almost all patient subgroups, except in patients with an ECOG score of 0 and in patients with an Epstein-Barr Virus level less than 500 copies/mL. Overall response rates were 68.1% in the penpulimab group and 63.9% in the placebo group, and responses to penpulimab + chemotherapy were significantly more durable than responses to placebo + chemotherapy (median duration of response 9.8 months vs 5.7 months, respectively, HR 0.40, p<0.001). Penpulimab was associated with a manageable safety profile. Treatment-related adverse events (TRAEs) were similar between the two arms, and as expected, immune-related TRAEs were more common in the penpulimab arm. With its high level of clinical efficacy and its manageable safety profile, penpulimab combined with chemotherapy represents a new effective treatment option for patients with recurrent or metastatic nasopharyngeal carcinoma. On April 23, 2025, penpulimab in combination with chemotherapy was approved by the U.S. Food and Drug Administration (FDA) for the treatment of recurrent or metastatic nasopharyngeal carcinoma.
Immunogenicity of a neoantigen vaccine in immunologically cold tumors
CT012. Personalized neoantigen vaccine with or without pembrolizumab in patients with microsatellite-stable metastatic colorectal cancer
Saurav Daniel Haldar (University of Texas MD Anderson Cancer Center, Houston, TX, USA) reported results of a pilot study of a personalized neoantigen vaccine as monotherapy or in combination with pembrolizumab for microsatellite-stable metastatic colorectal cancer (MSS mCRC). Due to its characterization as an immunologically “cold” tumor type, novel immunotherapy strategies are needed for management and long-term outcomes for MSS mCRC. Recent preclinical and clinical studies suggest personalized peptide vaccines can prime T cells against neoantigens, even in immunologically cold tumors, and neoantigen vaccines may enhance the anti-tumor activity of immune checkpoint inhibitors. This study used a personalized peptide vaccine platform, NeoAg-VAX, which targets up to 10 tumor-derived neoantigens per patient. Specific neoantigens were not isolated to determine which peptide generated the strongest immune response in patients. The TLR7 agonist imiquimod was applied at injection sites as an immune adjuvant. Patients in Cohort A (n=13) received NeoAg-VAX alone, and patients in Cohort B (n=15, including 8 patients from Cohort A retreated) received NeoAg-VAX in combination with pembrolizumab. Patients in both cohorts received up to 15 doses of the vaccine. All patients had been previously treated for CRC. No safety signals were associated with NeoAg-VAX. Vaccination with NeoAg-VAX resulted in a median 5.9 peak fold increase in neoantigen-specific T cells, and 10/11 evaluable patients exhibited a 2-fold or higher increase in neoantigen-specific T cells after vaccination with NeoAg-VAX. Clinical efficacy was limited, with one patient in each cohort achieving a partial response. The disease control rates in Cohorts A and B were 61.5% and 53.3%, respectively. Imaging mass cytometry on pre-treatment metastatic tumors indicated heterogeneity of immune cell composition of metastatic tumors across sites. Neighborhood analyses indicated T cells were closer to tumor cells, and myeloid cells were more distant from tumor cells in lung metastases compared to liver metastases, indicating organ-specific differences in immune cell archetype within the tumor microenvironment may contribute to different responses to immunotherapy. Results from this study indicate that the neoantigen vaccine NeoAg-VAX as monotherapy or in combination with pembrolizumab is safe and immunogenic in a heavily pre-treated population of patients with metastatic MSS CRC. Given the limited clinical efficacy in this setting, studies of NeoAg-VAX are ongoing in patients with minimal residual disease for colorectal cancer and pancreatic ductal adenocarcinoma.
Dose-finding study of SENTI-202: A first-in-class logic gated CAR NK cell therapy for relapsed/refractory hematological malignancies
CT014. First-in-human, multicenter study of SENTI-202, a CD33/FLT3 selective off-the-shelf logic gated CAR NK cell therapy in hematologic malignancies including AML: Clinical data
Stephen A. Strickland (Sarah Cannon Research Institute at TriStar Bone Marrow Transplant, Nashville, TN, USA) reported results of SENTI-202, a CD33/FLT3 logic-gated off-the-shelf CAR NK cell therapy for hematological malignancies, including acute myeloid leukemia (AML). SENTI-202 represents an approach to specifically target AML blasts and leukemia stem cells (LSCs) while sparing hematopoietic stem and progenitor cells (HSPCs). The SENTI-202 gene circuit design causes NK cells to target cells expressing CD33 and/or FLT3, expressed in 95% of patients with AML but inhibits the NK cells from targeting EMCN, which is expressed by non-cancerous HSCs and progenitor cells. Calibrated release of IL-15 enhances SENTI-202 anti-cancer activity and persistence. SENTI-202 has a scalable manufacturing process, and it can be readily available for patients. It is derived from healthy donors, exhibits high potency after thawing, and is intended to be used in an outpatient setting. Patients with relapsed or refractory CD33 and/or FLT3 expressing hematological malignancies participated in the study, and to date, all enrolled patients were diagnosed with AML. Patients received SNTI-202 at two dose levels, Dose Level 1 (1 x 10^9 cells/infusion) or Dose Level 2 (1.5 x 10^9 cells/infusion). Two dosing schedules were used where patients received 3 weekly doses (Schedule I) or 5 semi-weekly doses (Schedule II) of SENTI-202 over a 14-day period. Both dosing schedules included lymphodepletion prior to the first cell infusion. No dose-limiting toxicities were observed among the 3 patients receiving Dose Level 1 on Schedule I, the 3 patients receiving Dose level 1 on Schedule II, nor the 3 patients who received Dose Level 2 on Schedule I. Adverse events (AEs) of grades 3 or 4 were hematologic, unrelated to SENTI-202 and consistent with patients with AML receiving lymphodepletion. SENTI-202-related AEs were low-grade and manageable. Five of 7 patients responded to SENTI-202, representing all three dosing cohorts. 2 of 3 patients who received Dose Level 2 on Schedule 1 achieved a composite complete response (cCR), thus Dose Level 2 on Schedule I (3 weekly doses of 1.5 x 10^9 cells/infusion over 28 days) was determined to be the recommended phase 2 dose (RP2D). Four of 7 patients achieved a cCR, and all 4 complete responders were minimum residual disease negative. All cCRs were ongoing until the data cutoff. All but one patient experienced blast reduction compared to baseline after receiving SENTI-202. All responders exhibited a decrease in LSCs in bone marrow, and LSCs decreased 10-fold or more in patients who achieved a cCR. Patients who achieved a cCR recovered peripheral blood cell counts quickly, and HSPCs were maintained or increased in the bone marrow of patients who achieved cCRs, consistent with the logic-gated design of SENTI-202. SENTI-202 was detected in the peripheral blood of all treated patients, regardless of patient response. These promising early data suggest SENTI-202 may produce durable clinical results while minimizing on-target, off-tumor toxicities, meeting an unmet need for patients with AML. Additional studies of SENTI-202 at the RP2D are ongoing.
Dual immune checkpoint blockade of TIGIT and PD-L1 in patients with PD-L1 high locally advanced unresectable or metastatic non-small cell lung cancer
CT051. SKYSCRAPER-01: A phase III, randomized trial of tiragolumab (tira) + atezolizumab (atezo) versus placebo (pbo) + atezo in patients (pts) with previously untreated PD-L1-high, locally advanced unresectable/metastatic NSCLC
Solange Peters (Centre Hospitalier Universitaire Vaudois Oncology Department, Lausanne, Switzerland) presented data from SKYSCRAPER-01, a Phase 3 placebo-controlled clinical trial investigating the efficacy and safety of anti-TIGIT antibody tiragolumab (tira) combined with atezolizumab (atezo) for PD-L1-high locally advanced unresectable or metastatic non-small cell lung cancer (NSCLC) in the first-line setting. TIGIT is an immune checkpoint highly expressed on activated T cells, NK cells and regulatory T cells. Previous studies suggest that tumor infiltrating lymphocytes in lung cancer may express high levels of TIGIT and PD-1, suggesting that through blockade of TIGIT, tira may augment anti-tumor responses of other ICIs like atezo. In the Phase 2 CITYSCAPE trial of tira + atezo as first-line treatment for advanced lung cancer, tira + atezo was associated survival benefits in patient subgroup with high PD-L1 expression (TPS >= 50%). In SKYSCAPER-01, patients with previously untreated PD-1-high (TPS >= 50%) locally advanced unresectable or metastatic NSCLC were randomized 1:1 to receive tira + atezo or placebo (pbo) + atezo. In the Primary Analysis Set (PAS) of patients with TPS >= 50% per the 22C3 assay, 262 patients were treated with tira + atezo, and 259 patients were treated with pbo + atezo. Tira + atezo was associated with numeric but not statistically significant survival benefits compared to placebo + atezo. In the PAS, the median investigator-assessed progression free survival (PFS) was 7.0 months for tira + atezo and 5.6 months for pbo + atezo (HR 0.78, p=0.02). At a median follow-up of 17.9 months, median overall survival (OS) was 23.1 months with tira + atezo and 16.9 months with pbo +atezo (HR 0.87, p=0.22). Similar results were observed in the Secondary Analysis Set, consisting of patients with PD-L1 TC >= 50% per the SP263 assay. In the tira + atezo arm (n=211), median PFS was 8.3 months, compared to 5.7 months in the pbo + atezo arm (n=209; HR 0.86), and median OS was 24.5 months and 20.6 months, respectively (HR 0.93). Overall response rate (ORR) in the PAS was 45.8% with tira + atezo and 35.1% in pbo + atezo, and median duration of response in the tira + atezo arm and the pbo + atezo arm were 18.0 months and 14.6 months, respectively. The safety profile of tira + atezo was consistent with prior studies. In the safety evaluable set, adverse event (AE) frequency was greater in the tira + atezo arm (n=267) compared to the pbo + atezo arm (n=263). AEs leading to treatment withdrawal occurred in 16.1% of patients in the tira + atezo arm and 6.5% of patients in the pbo + atezo arm. Although no statistically significant clinical benefits were associated with tira + atezo, the consistent numeric improvements in survival and response suggest combining anti-TIGIT and anti-PD-L1 ICI may have potential anti-tumor activity in certain populations of patients with NSCLC. More data is needed to identify the patient population who may benefit from this combination therapy. Other trials are exploring the combination of tira + atezo in other settings, including unresectable stage III NSCLC (SKYSCRAPER 03) and metastatic hepatocellular carcinoma (SKYSCRAPER-14). Furthermore, Phase 3 trials of other molecules targeting TIGIT are underway.
Stressed DNA-damaged tumor cells can activate T cells independently of dendritic cells
3791. Tumor cells surviving DNA damage acquire antigen presenting functions to stimulate anti-tumor immunity
Tiffany R. Emmons (Massachusetts Institute of Technology, Cambridge, MA, USA) presented a study of the antigen-presenting functions of stressed or dying tumor cells and their role in stimulating anti-tumor immunity. This study builds upon previous findings that live stressed tumor cells are necessary to stimulate immune responses. T cell functional assays were performed in which tumor cells were treated with DNA-damaging drugs, co-cultured with dendritic cells (DCs), and CD8+ T cells were added to co-cultured tumor cells and DCs. T cell activation was then measured by IFN gamma or granzyme B production. T cell functional assays of the ID8-OVA murine ovarian cancer model indicated that sub-optimal levels of T cell activation can occur in the presence of stressed ID8-OVA murine ovarian cancer cells in the absence of DCs. When drug-treated tumor cells were treated with IFN gamma to increase surface expression of MHC1, DC-independent T cell activation in the T cell functional assay was significantly higher. Activated T cells in T cell functional assays proliferate, express activation markers CD44 and CD22, produce Granzyme B, and kill tumor cells. T cell functional assays were performed with B16F10 melanoma cells express TRP1 and transgenic T cells with a high affinity for TRP1, and drug- and IFN gamma-treated P16F10 melanoma cells were able to activate TRP1 CD8 T cells. DNA damaged tumor cells treated with IFN gamma also exhibited increased expression of co-stimulatory molecules CD80 and CD86 compared to tumor cells without DNA damage. Results from this study suggest that live, stressed tumor cells are immunogenic and can activate T cells independently of DCs, and the activated T cells exhibit proliferative and tumor-killing activities. Future directions for this work include combining DNA-damaging chemotherapy and immune checkpoint blockade to enhance therapeutic efficacy and clinical use of live, damaged tumor cells as immunogenic adjuvants.
Combining a TLR agonist with atezolizumab and radiotherapy to remodel the tumor immune microenvironment of metastatic virus-associated cancers
CT129. Immunity induction with atezolizumab, EIK1001, and radiotherapy in virus-associated tumors: Results of the AGADIR Trial
Antoine Italiano (Institute Bergonié, Bordeaux, France) presented efficacy and safety data from a single-arm Phase 2 trial of combination therapy with atezolizumab, EIK1001, a toll-like receptor 7/8 (TLR 7/8) agonist, and radiotherapy for metastatic virus-associated tumors. The therapeutic approach to this study was to remodel the tumor microenvironment (TME) of virus- associated cancers to make them more immunologically “hot,” enhancing anti-tumor activity of immune checkpoint inhibitors like atezolizumab. Toll-like receptor 7/8 (TLR 7/8) agonist EIK1001 has been shown in previous studies to enhance innate immunity by activating dendritic cells (DCs) and boosting cytokine productions. Stereotactic radiation therapy at a metastatic site may also contribute to stimulating anti-tumor immunity by inducing immunogenic cell death, promoting the release of tumor-associated antigens, and upregulating PD-L1 expression, creating a pro-inflammatory environment favorable for innate and adaptive immune responses. 47 patients with metastatic virus-associated tumors participated in the study, and 87% of patients had tumors related to human papilloma virus. Of the 45 evaluable patients, 33.3% experienced grade 3 treatment-related adverse events (TRAEs), and 2.2% of patients (1 patient) discontinued treatment due to TRAEs. The objective response rate for the study was 19.6%, and the median duration of response was 11.5 months. Median progression free survival was 2.6 months, and median overall survival was 10.4 months. Proteomic profiling of plasma samples indicated strong induction of T cell markers, suggesting induction and systemic mobilization of anti-cancer immunity. Immune profiling of pre- and post-treatment tumor samples exhibited increased levels of intratumoral immune cells in post-treatment samples compared to baseline samples. These results indicate that TLR7/8 agonist EIK1001 in combination with atezolizumab and stereotaxic radiation therapy exhibits a manageable safety profile and induces significant changes in the tumor immune microenvironment of metastatic virus-associated cancers. The clinical activity observed in this study is encouraging, and translational analyses will identify subsets of patients most likely to benefit from treatment are ongoing.
Neoadjuvant therapy with next-generation immune checkpoint inhibitors botensiliamab plus balstilimab in MMR proficient and deficient tumors
CT130. Neoadjuvant botensilimab plus balstilimab in MMR proficient and deficient early stage cancers: First results of the pan-cancer NEOASIS study
Myriam Chalabi (The Netherlands Cancer Institute, Amsterdam, Netherlands) reported safety and efficacy data from NEOASIS, a Phase 2 adaptive pan-cancer study investigating neoadjuvant dual immune checkpoint inhibition with next generation botensilimab (bot; anti-CTLA-4) and balstilimab (bal; anti-PD1) in both mismatch repair proficient (pMMR) and mismatch repair deficient (dMMR) tumors. Prior studies have indicated that neoadjuvant immunotherapy is associated with increased survival in some tumor types, including dMMR colorectal cancers (CRC). Additionally, the next generation immune checkpoint inhibitors bot and bal have shown promising efficacy against non-immunogenic tumors, including pMMR colorectal cancers and sarcomas. One aim of this study was to translate the efficacy seen in neoadjuvant treatment of dMMR CRC to dMMR tumors of other origins. Another aim was to determine whether neoadjuvant treatment with bot and bal induce responses in immunologically “cold” tumors. Two cohorts—pMMR and dMMR—of ten patients each participated in the study. All patients had non-metastatic resectable tumors and had not received prior chemotherapy or immunotherapy. For safety reasons, the first 5 patients in each cohort receive 25 mg bot. Due to the limited number of dose-limiting toxicities, all remaining patients in the cohorts received 50 mg bot. The pMMR cohort consisted of patients with triple negative breast cancer (TNBC; n=6), estrogen-receptor positive (ER+) breast cancer (n=2), Merkel cell carcinoma (n=1), and sarcoma (n=1). 9 patients in the dMMR cohort had CRC, and 1 patient had duodenal cancer. All 10 patients who received 20 mg bot received both cycles of therapy. No immune-related adverse events of grade 3 or higher or DLTs were observed, and no delays in surgery occurred. Among the 10 patients who received 50 mg bot, 1 patient experienced a grade 3 irAE that was fully resolved, and 2 patients did not receive the second cycle of therapy due to liver toxicity. No DLTs or delays of surgery occurred. In the pMMR cohort, the major pathological response (MPR) rate was 70%, which included all 5 patients who received 25 mg bot and 2 of the 5 patients who received 50 mg bot. 5 of 8 patients with breast cancer achieved an MPR. In the dMMR cohort, the MPR rate was 80%, with 70% of patients achieving a pathological complete response (pCR). 3 of the 5 patients who received 25 mg bot achieved an MPR, with 2 of them being pCR. Among patients who received 50 mg bot, all 5 achieved a pCR. The single nonresponder was the patient with duodenal cancer, and surgery revealed lesions of peritoneum, diaphragm and liver that showed pCR upon biopsy. Results from this study indicate that neoadjuvant combination treatment with botensilimab and balstilimab is safe and does not delay surgery. Neoadjuvant treatment generated remarkable responses in different types of pMMR and dMMR tumors, suggesting the potential for neoadjuvant botensilimab and balstilimab for all tumor types. Translational studies of prognostic and predictive biomarkers are ongoing, and recruitment for pMMR and dMMR tumors is ongoing.
A dual inhibitor of diacyl glycerol kinases enhances T cell activation and exhibits anti-tumor activity in preclinical studies
3789. INCB177054: A novel, potent, orally bioavailable DGK alpha/zeta dual inhibitor enhances T-cell function and demonstrates potent antitumor activity
Xiaodi Ren (Incyte Research Institute, Wilmington, DE, USA) reported a preclinical study characterizing the anti-tumor activity of INCB177054, an inhibitor of diacyl glycerol kinase alpha and zeta (DGK alpha/zeta). DGK alpha/zeta modulate diacyl glycerol levels in T cells, serving as intracellular checkpoints of T cell receptor (TCR) signaling and CD8 T cell proliferation. It was hypothesized that inhibition of DGK alpha/zeta would restore T cell function and enhance anti-tumor immunity. Treatment of human primary T cells with INCB177054 enhanced T cell proliferation and activation, as measured by Cd69 expression and phosphorylation of ERK and NFKB in a dose-dependent manner. INCB177054 enhanced T cell-mediated killing of breast cancer spheroids in vitro and synergized with the anti-PD-1 antibody retifanlimab to enhance interferon gamma production in the MLR assay. To determine whether INCB177054 could enhance T cell activation to low-affinity peptides, spherocytes from OT1 mice incubated with peptide variants representing a wide range of TCR affinities. Addition of INCB177054 enhanced T cell activation even in the presence of low-affinity peptides. To characterize INCB177054-mediated T cell activation in vivo, INCB177054 was administered to OT-1 mice. INCB177054 quickly induced T cell activation, measured by CD69 upregulation. INCB177054 exhibited modest anti-tumor activity in mouse syngeneic models of colorectal cancer (MC38 and CT26), but when administered in combination with anti-PD-1 antibodies, tumor growth was significantly inhibited. This study provides strong evidence that the DGK alpha/zeta dual inhibitor INCB177054enhances T cell activation in vitro and in vivo, and INCB177054 exhibits synergistic anti-cancer activity in combination with anti-PD-1 inhibitors in syngeneic mouse models. A Phase 1/2 study of INCB177054 in patients with advanced or metastatic solid tumors has been registered and will likely begin later this year.
Safety, clinical efficacy, and mechanistic insights of an Fc-engineered CD40 agonist for BCG-unresponsive non-muscle invasive bladder cancer
CT203. Fc-optimized anti-CD40 agonist antibody 2141-V11 for BCG-unresponsive non-muscle invasive bladder cancer: Updates on phase 1 study clinical outcomes and biological correlatives
Juan C. Osorio (Memorial Sloan Kettering Cancer Center, New York, NY, USA) presented results from an investigator-initiated Phase 1 open-label dose-escalation study of intravesical 2141-V11 in patients with Bacillus Calmette-Guerin (BCG)-unresponsive non-muscle invasive bladder cancer (NMIBC) who are ineligible or decline radical cystectomy. 2141-V11 is a novel anti-CD40 antibody agonist engineered to enhance binding to Fc gamma RIIB that results in effective dendritic cell (DC) maturation/activation and tumor-specific T cell responses. Prior studies of preclinical models of bladder cancer indicate that intravesical administration of 2141-V11 is safe and leads to durable anti-tumor immunity. 25 patients participated in the study. Patients received 3 weekly doses of 2141-V11 every three months for a six-month period. 5 doses levels were used, ranging from 0.7 mg to 70.0 mg. 2141-V11 was well tolerated. 56% of patients experienced a treatment-related adverse event (TRAE), and only one grade 3 TRAE were observed; no dose-limiting toxicities were observed. 10 patients (40%) achieved a complete response (CR), which is higher than CR rates from studies of other therapies for NMIBC. 78% of responses lasted 6 months or longer, 65% of responses lasted 12 months or longer, and the median duration of response not reached. No responders received cystectomy. At baseline, no major differences in key effector immune cells were observed in responders compared to non-responders. Patients who achieved a complete response exhibited higher baseline levels of both triads consisting of DCs, CXCL13+ CD8+ T cells, and CXCL13+ CD4+ T cells and tetrads consisting of DCs, CXCL13+ CD8+ T cells, CXCL13+ CD4+ T cells, and B cells. Tertiary lymphoid structures (TLSs) were present in both responders and non-responders at baseline, but post-treatment, responders retained or sustained TLSs while TLS number decreased in non-responders. Although follow-up studies are needed to confirm the clinical benefits observed in this study, 2141-V11 is safe and leads to durable CRs in a subset of patients with BCG-unresponsive non-muscle invasive bladder cancer. Planned experiments include mechanistic studies to determine whether immune triads/tetrads serve as a basis for TLS formation and clinical studies testing rational combinations of 2141-V11 with other therapies.
Safety and efficacy of runimotamab, a bispecific antibody as monotherapy and in combination with trastuzumab for HER2-positive breast cancer
CT204. Phase 1a/b study of runimotamab, a HER2 x CD3 T cell-engaging bispecific antibody, administered as a single agent and in combination with trastuzumab in patients with HER2-expressing breast cancer (BC)
Shanu Modi (Memorial Sloan Kettering Cancer Center, New York, NY, USA) presented results from a Phase 1a/b first-in-humans study of runimotamab, a humanized HER2 x CD3 bispecific T cell engager, as monotherapy and in combination with anti-HER2 monoclonal antibody trastuzumab for patients with HER2-expressing breast cancers (BC). The Phase 1a study, a dose escalation study of runimotamab monotherapy for locally advanced or metastatic relapsed/refractory BC included 20 patients; 60% had confirmed HER2-positive disease, and the median of 7 prior lines of therapy. All patients had received prior chemotherapy. Due to the high levels of on-target/off-tumor treatment related adverse events (TRAEs) associated early doses of runimotamab, including cytokine release syndrome (CRS) in 78% of patients with grade 2 severity, and no observed anti-tumor activity, further dose escalation in Phase 1a was halted. The Phase 1b study investigated runimotamab in combination with trastuzumab. It was hypothesized that co-treatment with trastuzumab and runimotamab causes competition with HER2 binding between . Cancer cells, with their higher expression of HER2, would offer more binding sites for runimotamab than non-cancerous HER2-expressing cells. As a result, this reduces on-target/off-tumor toxicities and increases the therapeutic index of runimotamab. 53 patients with HER2-positive metastatic breast cancer who had previously received treatment with trastuzumab participated in the Phase 1b portion of the study. The population was heavily pretreated, with a median of 8 prior lines of therapy. Although co-administration of runimotamab and trastuzumab decreased IL-6 secretion and T cell activation compared to runimotamab monotherapy, co-administration did reduce runimotamab clearance and allowed for a 60-fold increase in runimotamab dosing. Among patients who receive pharmacologically active doses of runimotamab, 78% of patients in Phase 1a experienced CRS or infusion-related reactions, compared to 31% of patients in Phase 1b. Most dose-limiting toxicities (DLTs) or DLT-equivalent adverse events occurred at runimotamab doses greater than 30/90 mg, so 20/60 mg was chosen are the recommended dose for expansion. The overall response rate in Phase 1b was 19% across all doses and 30% at the 20/60 mg level. Most responses lasted 8 months or more, and 2 patients remain on runimotamab therapy. Tumor biopsies from 24 patients collected 21 to 35 days after treatment initiation indicated no major loss of HER2 expression, including in responders. Co-administration of the HER2/CD3 bispecific antibody runimotamab with trastuzumab represents a novel approach to treating HER2-positive breast cancers and has produced encouraging clinical results in a heavily pre-treated patient population. Currently, physicians and scientists are developing combination strategies to improve the anti-tumor activity of runimotamab combined with trastuzumab.
Updated results of the completed dose escalation study of the T cell-targeting bispecific antibody invikafusp alfa for antigen-rich solid tumors
CT205. Updated clinical results, recommended phase 2 dose (RP2D) determination and translational study results for START-001: A phase 1/2 trial of invikafusp alfa, a first-in-class TCR beta chain-targeted bispecific antibody in patients with anti-PD(L)1-resistant, antigen-rich solid tumors
Ryan J. Sullivan (Massachusetts General Hospital Cancer Center, Boston, MA, USA) reported updated results from START-001, phase 1/2 monotherapy trial of invikafusp alfa, a selective dual T cell agonist targeting V beta 6/V beta 10 T cells, for patients with anti-PD(L)-1 resistant, antigen rich (tumor mutation burden-high (TMB-H), microsatellite instability high (MSI-H), mismatch repair deficient (dMMR), or virally associated) solid tumors. Invikafusp alfa is a bispecific antibody, which consists of a V beta T cell receptor (TCR) agonist. The agonist targets V beta 6/V beta 10 T cells, the most prevalent tumor-infiltrating lymphocytes to stimulate T cell proliferation and memory, and wild type IL-2 in cis to IL-2 receptors, which stimulate T cell proliferation and effector function. Preclinical data indicate invikafusp alfa expands populations of CD8+ V beta 6/V beta 10 T cells, especially central memory and effector memory T cells in a variety of tumors. The study consisted of 52 patients with 38 receiving the optimal biologic doses (OBD). 80.8% of patients in the study and 76.3% of patients in the OBD population had received prior anti-PD(L)-1 therapy. Invikafusp alfa stimulated expansion of expansion of V beta 6/V beta 10 T cells in a dose proportional manner, with optimal levels occurring at doses of 0.08 mg/kg and 0.12 mg/kg, the OBD, and 0.08 mg/kg was chosen as the recommended phase 2 dose (RP2D). Invikafusp alfa was also associated with a significant increase in effector memory T cells and cytotoxic T cells. The disease control rate (DCR) was 61%, and among the 42 patients who had received prior anti PD(L)-1 treatment, the DCR was 45%. 52% of patients treated with OBD reported any tumor regression. More anti-tumor activity observed in patients with TMB-H tumors compared to patients with virally associated tumors. The objective response rate in patients with TMB-H tumors was 20% and 30% in patients with TMB-H metastatic colorectal cancer (mCRC). Responders exhibited expansion of T cells in peripheral blood and in tumors. The safety profile of invikafusp alfa was consistent with other T cell activators and manageable, with 71.1 % of patients experiencing grade 1 or 2 cytokine release syndrome. Invikafusp alfa monotherapy exhibits clinical meaningful anti-tumor activity in antigen-rich, especially TMB-H, tumors and in tumors resistant to anti-PD(L)-1 treatment. The U.S. Food and Drug Association (FDA) has granted fast-track designation for invikafusp alfa in TMB-H mCR. A Phase 2 study of invikafusp alfa for TMB-H, dMMR, and MSI-H tumors is ongoing.
Safety and efficacy of AIC100, a third-generation CAR T cell targeting ICAM-1, for advanced thyroid cancers
CT206. ICAM-1 directed chimeric antigen receptor (CAR) T cells (AIC100) in patients with advanced thyroid cancers: Clinical and translational data from the phase 1 dose escalation study
Samer A. Srour (The University of Texas MD Anderson Cancer Center, Houston, TX, USA) reported updated results from a Phase 1 dose escalation trial of AIC100, a third-generation CAR T cellular therapy targeting ICAM1, for advanced relapsed/refractory thyroid cancer. ICAM1/CD54, a member of the immunoglobulin superfamily, is a cell surface glycoprotein and adhesion receptor overexpressed in several cancers, including thyroid cancers, and overexpression of ICAM-1 is associated with poor prognosis. The antigen-targeting portion of AIC100 has been modified to obtain optimal binding affinity for ICAM-1, maximizing clinical efficacy and reducing toxicities. AIC100 also co-expresses somatostatin receptor 2 (SSTR2) as a reporter marker, allowing for non-invasive real-time CAR T cell monitoring with DOTATATE PET scan. 15 patients with ICAM-1 expressing tumors were infused with AIC100 after lymphodepletion. 8 patients had anaplastic thyroid cancer (ATC), 7 patients had poorly differentiated thyroid cancer, and most patients had metastatic disease. Patients received one of three dose levels (DL) of AIC100: 1x 107 (DL1), 1 x 108 (DL2), and 5 x 108 cells/kg (DL3). The safety profile of AIC100 was manageable with 67% of patients experiencing grade 1 or 2 CRS. One heavily pre-treated patient developed therapy-related myeloid neoplasm, but it was unclear if it was caused by ICAM-1. At a median follow-up of 13.6 months, the objective response rate among the 9 patients who received DL2 or DL3 was 22%, with one partial response and one complete response. Both responders had ATC. The disease control rate was 56%, the median progression free survival was 2.9 months, and median overall survival was 3.6 months. Tumor shrinkage or stable disease occurred in 6 of 9 (67%) patients. CAR T cell expansion was observed in all testable patients. Most patients had increased levels of cytokines, including IL-6 and IFN gamma, 4 to 7 days after infusion, and DOTATE PET scans of responders indicated CAR T cell infiltration and expansion at the sites of lesions. With its manageable safety profile and encouraging anti-tumor activity, AIC100 holds potential as a new treatment option for patients with aggressive thyroid cancers and for patients with other ICAM-1 expression solid tumors.
Safety and efficacy of a mutant KRAS peptide vaccine in combination with checkpoint blockade for colorectal cancer
CT264. Mutant KRAS peptide vaccine combined with ipilimumab/nivolumab in advanced mismatch repair proficient/microsatellite stable (MMRp/MSS) colorectal cancer: Preliminary analysis from a phase I study
Hejia Henry Wang (Johns Hopkins University, Baltimore, MD, USA) presented a first in-human Phase 1 trial of a mKRAS-VAX, a neoantigen vaccine targeting mutant KRAS (mKRAS), in combination with ipilimumab and nivolumab (ipi/nivo) in patients with heavily pretreated mismatch repair proficient/microsatellite stable (MMRp/MSS) colorectal cancer (CRC). Immune checkpoint inhibitors have exhibited minimal clinical activity in the 95% of CRC cases that are MMRp/MSS, partly due to the lower expression of tumor neoantigens. mKRAS is an oncogenic driver of approximately 40% of all cases of CRC. Prior studies indicate that mKRAS is immunogenic: an mKRAS G12D G12R peptide vaccine was associated with an mKRAS specific T cell response and recurrent free survival in the adjuvant setting for CRC and pancreatic cancer. mKRAS-VAX is a pooled synthetic long peptide vaccine targeting the six most common KRAS mutations (G12D, G13D, G12V, G12C, G12A, G12R), and as an off-the-shelf shared antigen vaccine, treatment with mKRAS-VAX can be initiated rapidly for the metastatic setting. 12 patients with chemorefractory metastatic MMRp/MSS colorectal cancer and no prior immunotherapy exposure were treated with mKRAS-VAX in combination with ipi/nivo. 7 of the 12 patients had active liver metastases. All adverse events (AEs) associated with mKRAS-VAX were low-grade. No new safety signals were detected with ipi/nivo, and 3 patients did discontinue ipi/nivo due to AEs related to checkpoint blockade. Patient T cell responses in peripheral blood were quantified ex vivo with IFN gamma-ELISpot assays. 8 of 12 patients were considered immune responders with a significant increase in tumor-specific mKRAS T cell response. 2 of 12 patients responded to therapy, and one of the responders had liver metastases. Of the 5 patients who experienced disease control, 3 had liver metastases. There was no correlation between the magnitude of peripheral KRAS T cell response and clinical response. To more closely examine the T cell receptor (TCR) trafficking of responders and non-responders, baseline and on-treatment peripheral blood samples and tumor biopsies underwent TCR beta sequencing. Samples from one non-responder who showed a significant increase in T cell responses in the ex vivo ELISpot assay, indicated low levels of KRAS-specific and treatment-enriched TCRs at baseline, with a modest increase after vaccination. Samples from a responder who exhibited low levels of T cell responses in the ELISpot assay also exhibited low levels of KRAS-specific and treatment-enriched TCRs at baseline, but these levels dramatically increased after vaccination. There was a significant post-vaccine decrease in Shannon diversity in the responder compared to the nonresponder, indicating greater expansion of T cell clones is associated with response to mKRAS-VAX combined with ipi/nivo. This study represents the first time an mKRAS vaccine combined with ipi/nivo demonstrated clinical efficacy in chemorefractory MMRp MSS CRC. These results also suggest monitoring of mKRAS-specific and treatment-enriched TCRs may be better predictors of response than ex vivo peripheral blood IFN gamma-ELISpot. A Phase 1b trial of mKRAS-VAX combination with next-generation immune checkpoint inhibitors balstilimab and botensilimab in patients with metastatic MMRp/MSS CRC or pancreatic adenocarcinoma in the first-line maintenance setting is currently enrolling patients.
Therapeutic efficacy and safety of engineered TIL therapy in which the novel intracellular checkpoint CISH is disrupted for metastatic gastrointestinal cancers
CT269. First-in-human trial in patients with metastatic colorectal cancer using CRISPR-engineered tumor infiltrating lymphocytes in which the intracellular immune checkpoint CISH is inhibited
Emil Lou (University of Minnesota Medical School, Minneapolis, MN, USA) reported results from a first-in-human trial of the safety and anti-tumor activity of a genetically edited tumor infiltrating lymphocyte (TIL) therapy for metastatic gastrointestinal cancers. TILs were genetically modified by CRISPR-Cas9 to knock out Cytokine inducible SH2 containing protein (CISH), a novel PD-L1-independent intracellular checkpoint target. CISH is associated with non-functional TILs, and previous studies indicate that knockout (KO) of CISH induces TIL neoantigen reactivity, activation of TILs, and synergistic interaction with PD-1 blockade in vivo. 12 patients with metastatic treatment-refractory gastrointestinal cancer participated in the study. Tumors were resected for TIL harvest, and TILs underwent an expansion protocol and CRISPR-Cas9-mediated KO of CISH. KO of CISH occurred with high efficiency (over 90%) and no off-target editing was detected. CISH KO, neoantigen reactive TILs were expanded and administered to patients after non-myeloablative lymphodepletion. High-dose IL-2 was administered after infusion stimulate TIL persistence. Median progression free survival was 57 days, and median overall survival was 129 days. One young adult with drug-refractory metastatic CRC achieved a durable complete response of over 24 months. Adverse events were related to lymphodepleting chemotherapy and high-dose IL-2. Persistence and expansion of unique TCR clonotypes in neoantigen-responsive TILs were identified, and these patterns were consistent with spikes in CISH-edited alleles, as detected by next-generation sequencing. In the patient with the complete response, clonotypes exhibiting persistence of one year or more exhibited reduced or undetectable expression of CISH, suggesting TIL persistence and TCR repertoire were associated with CISH KO. Results from this study support the potential safety and efficacy of ablation of the CISH immune checkpoint. Current studies include the inhibition of CISH through gene or cellular therapies as well as the development of small molecules of CISH inhibitors.