LBA417 Background: There is no globally accepted standard of care (SOC) for advanced RCC after IO therapy. VEGFR-TKIs are often leveraged in this setting, but were primarily tested in phase 3 studies prior to PD-(L)1 inhibitors becoming SOC in earlier lines of therapy. The phase 3 LITESPARK-011 study (NCT04586231) investigates bel + lenva vs cabo in pts with advanced RCC progressing after anti–PD-(L)1 therapy in the 1L, 2L or adjuvant setting. Methods: Eligible pts were ≥18 yrs old with advanced clear cell RCC (ccRCC) that progressed on or after 1L or 2L anti–PD-(L)1 therapy or ≤6 mo of last dose of adjuvant anti–PD-(L)1 therapy. Pts were randomized 1:1 to bel 120 mg + lenva 20 mg QD vs cabo 60 mg QD. The dual primary endpoints were progression-free survival (PFS) by blinded independent central review (BICR) per RECIST 1.1 and overall survival (OS). Secondary endpoints included objective response rate (ORR, key) and duration of response (DOR) by BICR per RECIST 1.1, and safety. Results are reported for the first (IA1; data cutoff Jun 26, 2024) and second (IA2; data cutoff Apr 9, 2025) interim analysis. Results: 747 pts were randomized to bel + lenva (n = 371) or cabo (n = 376). Median (range) follow-up was 19.6 mo (9.9–39.8) at IA1, and 29.0 mo (19.3–49.2) at IA2. Bel + lenva showed statistically significant improvement in PFS (IA1, IA2) and ORR (IA1) vs cabo (Table). OS results did not reach statistical significance (Table); additional follow-up for OS is ongoing. Median (range) DOR was 23.0 mo (2.0–44.3+) with bel + lenva vs 12.3 mo (1.8+–35.9+) with cabo at IA2. Grade ≥3 TEAEs occurred in 84.1% of pts with bel + lenva and 82.7% with cabo; TEAEs led to death in 5.4% of pts (2 were treatment-related: 1 each thrombotic microangiopathy and pneumonitis) and 3.2% of pts (1 was treatment-related: hemoptysis), respectively. Conclusions: Bel + lenva demonstrated superior PFS and ORR vs cabo in pts with advanced ccRCC following anti–PD-(L)1 therapy. OS favored bel + lenva but did not reach statistical significance and will be tested further at final analysis. The safety profile of bel + lenva was consistent with the profiles of the individual drugs. LITESPARK-011 is the first phase 3 study of a HIF-2α inhibitor combined with a VEGFR-TKI, and the first phase 3 study in RCC to show improved outcomes vs a contemporary VEGFR-TKI. Clinical trial information: NCT04586231 . IA1 IA2 Bel + Lenva, N = 371 Cabo, N = 376 Bel + Lenva, N = 371 Cabo, N = 376 Median PFS (95% CI), mo 14.6 (11.1–16.6) 10.6 (9.2–11.1) 14.8 (11.2–16.6) 10.7 (9.2–11.1) HR (95% CI) 0.74 (0.61–0.89) 0.70 (0.59–0.84) P value (1-sided) .00095* .00007* Median OS (95% CI), mo NR (26.5–NR) 27.4 (23.6–31.4) 34.9 (27.5–NR) 27.6 (24.0–31.4) HR (95% CI) 0.90 (0.70–1.15) 0.85 (0.68–1.05) P value (1-sided) .19322 .06075 ORR, % (95% CI) 52.6 (47.3–57.7) 39.6 (34.6–44.8) 52.6 (47.3–57.7) 40.2 (35.2–45.3) P value (1-sided) .0002* NA NA, not applicable. *Denotes statistical significance.

440 Background: The phase 1/2 KEYMAKER-U03 Substudy 03B (NCT04626518) is being conducted to evaluate combination treatments for previously treated advanced ccRCC. We present results for targeted therapy–containing regimens from arm B4 (pembro + belzutifan [HIF-2α inhibitor]), arm B5 (lenvatinib [VEGF-TKI] + belzutifan), and the reference (ref) arm (pembro + lenvatinib). Methods: Adults with histologically confirmed locally advanced/metastatic ccRCC and disease progression on or after PD-(L)1 inhibitor and VEGF-TKI treatment were randomly assigned 1:1 to arms open for enrollment. Arms B4 and B5 had a safety lead-in phase where ~10 patients (pts) were initially enrolled before randomization. Treatment doses were pembro 400 mg IV Q6W + belzutifan 120 mg PO QD (arm B4), lenvatinib 20 mg PO QD + belzutifan 120 mg PO QD (arm B5), or pembro 400 mg IV Q6W + lenvatinib 20 mg PO QD (ref arm). Primary end points were safety and ORR per RECIST v1.1 by blinded independent central review (BICR). Secondary end points included DOR, clinical benefit rate (CBR; CR + PR + SD ≥6 months), and PFS per RECIST v1.1 by BICR, and OS. Efficacy was evaluated in all enrolled (allocated and randomized) pts; safety was evaluated in all pts who received ≥1 dose of treatment. No formal comparisons across arms occurred. Enrollment was planned for 50 pts in each arm, although enrollment would be stopped if the 6-mo PFS rate was ≤40%. Results: Overall, 62 pts were assigned to arm B4, 64 to arm B5, and 73 to the ref arm. Median (range) follow-up was 16.6 mo (6.5-38.7) in arm B4, 17.6 mo (6.5-35.9) in arm B5, and 19.4 mo (6.7-33.2) in the ref arm. Efficacy is reported in the table. Grade 3-5 treatment-related AEs (TRAEs) occurred in 26/62 pts (42%) in arm B4, 38/63 pts (60%) in arm B5, and 36/73 pts (49%) in the ref arm. TRAEs led to death in 2 pts in arm B5 (cerebral hemorrhage and intracranial hemorrhage) and 1 pt in the ref arm (esophageal perforation). Conclusions: Lenvatinib + belzutifan (arm B5) exhibited durable antitumor activity and a safety profile consistent with the individual profiles of the drugs. Results from Substudy-03B support further investigation of lenvatinib + belzutifan combination for pts with advanced RCC, as in LITESPARK-011. Clinical trial information: NCT04626518 . Arm B4Pembro + belzutifann = 62 Arm B5Lenvatinib + belzutifann = 64 Ref armPembro + lenvatinibn = 73 ORR (95% CI), % 19 (10-31) 47 (34-60) 40 (29-52) CR, n (%) 2 (3) 1 (2) 0 (0) PR, n (%) 10 (16) 29 (45) 29 (40) CBR (95% CI), % 32 (21-45) 59 (46-72) 58 (45-69) DOR, median (range), mo Not reached (1.4+-33.0+) 22.1 (1.4+-32.8+) 8.3 (2.6+-25.6+) PFS, median (95% CI), mo 5.4 (2.8-6.9) 12.5 (5.9-26.3) 9.4 (6.9-11.2) 6-mo PFS rate, % 42 63 67 OS, median (95% CI), mo 27.4 (12.6-not reached) 32.3 (22.4-not reached) Not reached (21.8-not reached) 12-mo OS rate, % 68 80 82

501 Background: Since its approval, PD-1/PD-L1 IO therapies are used greatly among patients with mRCC. However, outcomes in the post-IO setting are not well characterized. This study aims to characterize the treatment patterns and clinical outcomes of patients with mRCC who have experienced progression following IO therapy. Methods: A retrospective review of electronic health records from a US community oncology network included patients with mRCC who progressed on their initial IO therapy and received subsequent therapies between 01/01/2018 and 01/31/2023. The index regimen was defined as the therapy administered after the initial IO therapy, and the index date marked the initiation of this index regimen. The index line of therapy (LOT) was defined as the treatment setting in which index regimen was used. Descriptive statistics were used to summarize patient characteristics and treatment patterns. Real-world progression free survival (rwPFS) and overall survival (OS) were analyzed by Kaplan–Meier method. Log-rank tests were performed to compare outcomes across various Index LOTs and index regimen classes. Results: The study identified 299 patients with a median age of 68 years (IQR 60-75). Among them, 72.7% were male, 84.7% were white, and 87.4% had an intermediate/poor risk score according to the international metastatic RCC database consortium. The index regimen was given as follows: 1.7% in first line (1L), 67.9% in 2L, and 30.4% in 3L. Tyrosine kinase inhibitor (TKI) monotherapy was the most common index regimen class (62.2%) with cabozantinib being the most used regimen (46.5%). Over a median follow up of 11.3 months, median (95% CI) rwPFS and OS were 6.0 (5.4, 6.8) and 14.3 (12.1, 16.4) months, respectively. No differences were observed in rwPFS and OS across index LOTs (rwPFS p= 0.763; OS p= 0.885) or index regimen classes (rwPFS p= 0.721; OS p= 0.747, Table 1.). Conclusions: Treatment outcomes in the post-IO setting for mRCC patients are limited. TKI monotherapies are the most common treatments in this setting. No difference was found in OS or rwPFS across index LOTs and index regimen classes. Novel treatments are needed to improve clinical outcomes in these setting. Subgroups N Median rwPFS (95% CI) Median OS (95% CI) Overall 299 6.0 (5.4, 6.8) 14.3 (12.1, 16.4) Index Regimen 2L* 203 6.0 (5.0, 7.1) 14.1 (11.5, 16.3) Index Regimen 3L 91 6.0 (5.1, 8.2) 14.5 (11.6, 20.1) Index Regimen Class: IO + IO** 25 3.6 (2.8, 6.7) 19.7 (12.4, 33.3) Index Regimen Class: IO + TKI 52 5.4 (3.6, 8.0) 13.4 (9.6, 23.8) Index Regimen Class: TKI Only 186 6.7 (5.6, 8.5) 14.4 (11.6, 16.7) Index Regimen Class: Other 28 5.4 (2.9, 8.0) 13.9 (6.9, 26.9) *Due to small sample sizes of the index regimen 1L, log-rank test was performed between index regimen 2L and 3L subgroups. **Due to small sample sizes of IO only subgroup, log-rank test was performed among IO + IO, IO + TKI, TKI only, and other subgroups.

552 Background: Substudy 03B of the phase 1/2 KEYMAKER-U03 trial (NCT04626518) was designed to evaluate combination treatments with pembro and investigational agents for previously treated ccRCC. We present results from arms containing only immunotherapy regimens: arm B1 (quavonlimab [qmab; anti–CTLA-4] coformulated with pembro), arm B2 (favezelimab [fave; anti–LAG-3] coformulated with pembro), and arm B3 (pembro + MK-4830 [anti-ILT4]), and the reference (ref) arm (pembro + lenvatinib). Methods: Adults with histologically confirmed ccRCC and disease progression on or after PD-(L)1 inhibitor and VEGF-TKI treatment were enrolled. Patients (pts) in arm B1 could not have received prior anti–CTLA-4 inhibitors. Pts were randomly assigned 1:1 to arms open for enrolment. Treatments were qmab/pembro 25 mg/400 mg IV Q6W in arm B1, fave/pembro 800 mg/200 mg IV Q3W in arm B2, pembro 200 mg IV Q3W + MK-4830 800 mg IV Q3W in arm B3, and pembro 400 mg IV Q6W + lenvatinib 20 mg PO QD in ref arm. Primary end points were safety and confirmed ORR per RECIST v1.1 by blinded independent central review (BICR). Secondary end points included CBR (CR + PR + SD ≥6 months), DOR, and PFS per RECIST v1.1 by BICR, and OS. End points were evaluated in each arm separately; no formal comparisons were made across arms. Efficacy was evaluated in all enrolled pts and safety was evaluated in all pts who received ≥1 dose of treatment. Enrollment was planned for 50 pts in each arm, although enrollment would be stopped if the 6-mo PFS rate was ≤40%. Results: Overall, 20 pts were assigned to arm B1, 20 to arm B2, 24 to arm B3, and 73 to ref arm. Median follow-up was approximately 2 years across arms (B1, 20.6 mo; B2, 23.3 mo; B3, 30.0 mo; ref, 19.4 mo). Efficacy is reported in the table. Grade 3 or 4 treatment-related AEs (TRAEs) were reported in 5/20 pts (25%) in arm B1, 2/19 pts (11%) in arm B2, 2/23 pts (9%) in arm B3, and 36/73 pts (49%) in the ref arm. No grade 5 TRAEs occurred. Conclusions: Preliminary data from qmab/pembro (arm B1) and pembro + lenvatinib (ref arm) showed antitumor activity in pts with ccRCC that progressed on anti–PD-(L)1 and VEGF-TKI therapy. Fave/pembro (arm B2) and pembro + MK-4830 (arm B3) arms did not show clinical activity in this setting. The safety profile of each arm was manageable. Additional arms containing targeted therapies will be reported separately. Clinical trial information: NCT04626518 . Arm B1Qmab/pembron = 20 Arm B2Fave/pembron = 20 Arm B3Pembro + MK-4830n = 24 Ref Pembro + lenvatinibn = 73 ORR (95% CI), % 30 (12-54) 0 (0-17) 0 (0-14) 40 (29-52) CR, n (%) 0 0 0 0 PR, n (%) 6 (30) 0 0 29 (40) CBR (95% CI), % 35 (15-59) 5 (0-25) 0 (0-14) 58 (45-69) DOR, median (range), mo Not reached(5.6+-19.4+) — — 8.3 (2.6+-25.6+) PFS, median (95% CI), mo 5.5(1.5-not reached) 1.6 (1.4-2.8) 1.5 (1.3-1.6) 9.4 (6.9-11.2) 6-mo PFS rate, % 45 20 4 67 OS, median (95% CI), mo Not reached(7.8-not reached) Not reached(9.4-not reached) 11 (7-18) Not reached(21.8-not reached) 12-mo OS rate, % 75 61 44 82

Analysis of T-cell subset proliferation subsequent to initiation of fianlimab as monotherapy or in combination with cemiplimab in (A) CD4 effector memory T cells and (B) CD8 effector memory T cells. Representative dot plots of CD4 and CD8+ T-cell memory subpopulations expressing Ki67+, LAG3+, HLA-DR+, PD-1+, or PD-L1+. Pharmacodynamic assay data from the 40 mg/kg dose cohort could not be generated due to poor quality of samples. DL, dose level; IV, intravenous; Q3W, every 3 weeks.

Abstract Colorectal cancer is a malignant tumor of the digestive tract; however, surgical resection and chemoradiotherapy have limited effects, and targeted therapy and immunotherapy have become novel options for prolonging the survival of patients with colorectal cancer. STAT3, a transcription factor, regulates the expression of various proteins and participates in multiple biological processes in tumors, making it a potential target for cancer therapy. PD-L1 are essential immune checkpoint molecules that regulate the tumor microenvironment and immune escape of tumor cells. The increased expression of PD-L1 induce the internal compositional changes of tumor cells, and regulate multiple processes of tumor cell progression. Attenuated Salmonella typhimurium can target tumor tissue to deliver plasmids and preferentially colonize hypoxic areas within the tumor, helping mount an anti-tumor immune response. Therefore, we constructed a double-interfering plasmid shSTAT3/shPD-L1 to reduce the expression of STAT3 and PD-L1 simultaneously and used attenuated Salmonella enterica serovar typhimurium carrying the plasmid to improve tumor tissue targeting and anti-tumor immune effects to explore the anti-tumor effects of double-interfering RNA at multiple levels in vivo and in vitro and to demonstrate its potential application in the clinical treatment of colorectal cancer.

Background The combination of monalizumab (anti-NKG2A/CD94) and durvalumab (anti-programmed death ligand-1) may promote antitumor immunity by targeting innate and adaptive immunity. This phase 1/2 study of monalizumab and durvalumab evaluated safety, antitumor activity, and pharmacodynamics in patients with advanced solid tumors.Main body Immunotherapy-naïve patients aged ≥18 years with advanced disease, Eastern Cooperative Oncology Group performance status of 0–1, and 1–3 prior lines of systemic therapy in the recurrent/metastatic setting were enrolled. In part 1 (dose escalation), patients received durvalumab 1500 mg every 4 weeks (Q4W) with increasing doses of monalizumab Q2W/Q4W (n=15). Dose expansion in part 1 included patients with cervical cancer (n=15; durvalumab 1500 mg Q4W and monalizumab 750 mg Q2W) or metastatic microsatellite stable (MSS)-colorectal cancer (CRC) (n=15; durvalumab 1500 mg Q4W and monalizumab 750 mg Q4W). In part 2 (dose expansion), patients with MSS-CRC (n=40), non-small cell lung cancer (NSCLC; n=20), MSS-endometrial cancer (n=40), or ovarian cancer (n=40) received durvalumab 1500 mg Q4W and monalizumab 750 mg Q2W. The primary endpoint was safety. Secondary endpoints included antitumor activity per Response Evaluation Criteria In Solid Tumors version 1.1 (RECIST v1.1). Exploratory analyses included assessment of T-cell and natural killer (NK) cell activation and proliferation in peripheral blood and the tumor microenvironment (TME). The study enrolled 185 patients (part 1, 45; part 2, 140). No dose-limiting toxicities were observed and the maximum tolerated dose was not reached. In part 2, the most common treatment-related adverse events were fatigue (12.1%), asthenia (9.3%), diarrhea (9.3%), pruritus (7.9%), and pyrexia (7.1%). In the expansion cohorts, response rates were 0% (cervical), 7.7% (MSS-CRC), 10% (NSCLC), 5.4% (ovarian), and 0% (MSS-endometrial). Sustained NK cell activation, CD8+ T-cell proliferation, increased serum levels of CXCL10 (C-X-C motif chemokine ligand 10) and CXCL11, and increased tumor infiltration of CD8+ and granzyme B+ cells were observed.Conclusions Although efficacy was modest, monalizumab plus durvalumab was well tolerated and encouraging immune activation was observed in the peripheral blood and TME.Trial registration number NCT02671435.

A, Clinical activity and (B) changes to target lesion over time in patients treated with fianlimab plus cemiplimab. Figure only includes patients who had both baseline and postbaseline target lesion assessments; not all patients had these assessments–therefore some patients may not be shown in this figure. Seven patients (7/47, 14.9%) were not evaluated in this treatment group. DL, dose level; IV, intravenous; Q3W, every 3 weeks; SOD, sum of diameters.

Summary of TEAE regardless of attribution.

A, Clinical activitya and (B) changes to target lesion over time in patients treated with monotherapy to combination therapy. aBest overall response is calculated on the basis of tumor change from the start of combination treatment. bTumor response for this patient was calculated as the best change in tumor size from the start of combination therapy. Triangles denote the last tumor assessment on fianlimab monotherapy. Figure only includes patients who had both baseline and postbaseline target lesion assessments; not all patients had these assessments—therefore some patients may not be shown in this figure. Three patients (3/16, 18.8%) were not evaluated in this treatment group. DL, dose level; IV, intravenous; Q3W, every 3 weeks; SOD, sum of diameters.

PURPOSE:Preclinical data indicate that fianlimab (antilymphocyte activation gene-3) plus cemiplimab (anti-PD-1) enhances antitumor activity. Here, we report prespecified final analyses of the dose-escalation part of a first-in-human, phase 1 study (NCT03005782) of fianlimab as monotherapy and in combination with cemiplimab in patients with advanced malignancies. PATIENTS AND METHODS:Adult patients received 1 to 40 mg/kg of fianlimab plus 350 mg of cemiplimab every 3 weeks (Q3W) across various dose-escalation schedules. Primary objectives were the rate of dose-limiting toxicities, adverse events (including immune mediated), deaths, laboratory abnormalities, and pharmacokinetics. Secondary outcomes were objective response rate, best overall response, duration of response, and antidrug antibody variables. RESULTS:Seventy-eight patients were enrolled (fianlimab + cemiplimab, n = 47; fianlimab monotherapy, n = 31). One patient treated with 3 mg/kg fianlimab + cemiplimab experienced dose-limiting toxicities, including increased blood creatine phosphokinase and myasthenic syndrome. No maximum tolerated dose was reached. Any-grade treatment-emergent adverse events occurred in 90% of patients with fianlimab monotherapy, in 87% of patients with fianlimab + cemiplimab, and in 87% of patients who transitioned from monotherapy to combination therapy. Fianlimab pharmacokinetics were dose proportional and similar in monotherapy and combination therapy. Across patients who received fianlimab + cemiplimab, five achieved a partial response, three of whom experienced a response after transitioning from monotherapy to combination therapy. Fianlimab 1,600 mg Q3W (20 mg/kg in an 80-kg individual) is the selected dose for phase 2 and phase 3 studies. CONCLUSIONS:Fianlimab as monotherapy and in combination with cemiplimab demonstrated acceptable safety and preliminary antitumor activity, which is generally consistent with previous reports of cemiplimab.

Baseline characteristics and prior therapies.

Investigator-assessed tumor response rate by Response Evaluation Criteria in Solid Tumors version 1.1.

BackgroundImmune checkpoint inhibitors have shown promising efficacy in multiple malignancies and, therefore, have been increasingly used over the past decade. Clinical data have suggested anti-cancer efficacy associated with immune-related adverse events that could have added healthcare resource utilization and costs.ObjectiveWe used a nationwide dataset to investigate the association between immune-related adverse events and healthcare resource utilization, charges, and mortality among patients receiving various immune checkpoint inhibitors for indicated cancers.MethodsWe performed a retrospective analysis of the National Inpatient Sample to identify patients hospitalized in the USA for immunotherapy between October 2015 and 2018. Data between patients who developed immune-related adverse events were compared to those who did not. Baseline characteristics, inpatient complications, and associated charges were collected and analyzed between these two groups.ResultsPatients who developed immune-related adverse events in the hospital had high incidences of acute kidney injury, non-septic shock, and pneumonia, and managing these complications significantly contributed to higher healthcare resource utilization. The average charge of admission was highest in patients who developed an infusion reaction, followed by colitis, and adrenal insufficiency. In terms of cancer type, renal cell carcinoma had the highest charges, followed by Merkel cell carcinoma.ConclusionsImmune checkpoint inhibitor-based regimens have shifted the treatment landscape among multiple malignancies and their use continues to expand. However, a significant proportion of patients still develop severe adverse effects leading to increased healthcare costs and impacting patients' quality of life. Closer attention should be given to recognizing and managing immune-related adverse events according to guidelines across healthcare facilities and clinical practice settings.

BackgroundIbrutinib, a first-in-class inhibitor of Bruton's tyrosine kinase, is approved for the treatment of various B-cell malignancies and chronic graft-versus-host disease. Based on encouraging preclinical data, safety and efficacy of ibrutinib combined with companion drugs for advanced renal cell carcinoma (RCC), gastric/gastroesophageal junctional adenocarcinoma (GC), and colorectal adenocarcinoma (CRC) were evaluated.MethodsIbrutinib 560 mg or 840 mg once daily was administered with standard doses of everolimus for RCC, docetaxel for GC, and cetuximab for CRC. Endpoints included determination of the recommended phase 2 dose (RP2D) of ibrutinib in phase 1b and efficacy (overall response rate [ORR] for GC and CRC; progression-free survival [PFS] for CRC) in phase 2.ResultsA total of 39 (RCC), 46 (GC), and 50 (RCC) patients were enrolled and received the RP2D. Safety profiles were consistent with the individual agents used in the study. Confirmed ORRs were 3% (RCC), 21% (GC), and 19% (CRC). Median (90% CI) PFS was 5.6 (3.9-7.5) months in RCC, 4.0 (2.7-4.2) months in GC, and 5.4 (4.1-5.8) months in CRC.ConclusionsClinically meaningful increases in efficacy were not observed compared to historical controls; however, the data may warrant further evaluation of ibrutinib combinations in other solid tumours.Trial registrationClinicalTrials.gov, NCT02599324.

Figure S1, Figure S2, Figure S3, Figure S4, Table S1, Table S2, Table S3

The purpose of this study was to assess the effect of pevonedistat, a neural precursor cell expressed, developmentally down-regulated protein 8 (NEDD8)-activating enzyme inhibitor, on the heart rate-corrected QT (QTc) interval in cancer patients. Patients were randomized 1:1 to receive pevonedistat 25 or 50 mg/m 2 on day 1 and the alternate dose on day 8. Triplicate electrocardiograms were collected at intervals over 0–11 hours and at 24 hours via Holter recorders on days −1 (baseline), 1, and 8. Changes from time-matched baseline values were calculated for QTc by Fridericia (QTcF), PR, and QRS intervals. Serial time-matched blood samples for analysis of pevonedistat plasma pharmacokinetics were collected and a concentration–QTc analysis conducted. Safety was assessed by monitoring vital signs, physical examinations, and clinical laboratory tests. Forty-four patients were included in the QTc analysis. Maximum least square (LS) mean increase from time-matched baseline in QTcF was 3.2 milliseconds at 1 hour postdose for pevonedistat at 25 mg/m 2 , while the LSs mean change from baseline in QTcF was −1.7 milliseconds 1 hour postdose at 50 mg/m 2 . The maximum 2-sided 90% upper confidence bound was 6.7 and 2.9 milliseconds for pevonedistat at 25 and 50 mg/m 2 , respectively. Pevonedistat did not result in clinically relevant effects on heart rate, nor on PR or QRS intervals. Results from pevonedistat concentration–QTc analysis were consistent with these findings. Administration of pevonedistat to cancer patients at a dose of up to 50 mg/m 2 showed no evidence of QT prolongation, indicative of the lack of clinically meaningful effects on cardiac repolarization. ClinicalTrials.gov identifier: NCT03330106 (first registered on November 6, 2017).

4553 Background: The oral HIF-2α inhibitor belzutifan demonstrated antitumor activity with manageable safety as monotherapy in pts with heavily pretreated ccRCC and in combination with a VEGF-TKI in pts who previously received a PD-1/L1 inhibitor. KEYMAKER-U03B (NCT04626518) is phase 1/2, multicenter, multi-arm, open-label, adaptive umbrella study. We present preliminary results from arm B5 for belzutifan + lenvatinib. Methods: In all arms, adults with histologically confirmed ccRCC, KPS ≥70%, and disease progression on or after PD-1/L1 inhibitor and VEGF-TKI treatment (in sequence or in combination) were enrolled. In arm B5, pts received belzutifan 120 mg PO QD + lenvatinib 20 mg PO QD. The study comprised a safety lead-in phase to establish the recommended phase 2 dose (RP2D), followed by an efficacy phase. Primary end point for the safety lead-in phase was safety and tolerability to establish RP2D; co-primary end points of the efficacy phase were safety and ORR per RECIST v1.1 by blinded independent central review (BICR). Secondary end points include clinical benefit rate (CBR; CR + PR + SD ≥6 mo), DOR and PFS per RECIST v1.1 by BICR. Response will be presented in pts who had an opportunity to receive ≥2 postbaseline scans due to ongoing enrollment. PFS was evaluated in all enrolled pts and safety in all treated pts. End points will be evaluated in each arm separately; no comparisons will be made across arms. Results: As of September 29, 2022, 32 pts were enrolled and 30 received treatment. Median age was 60.5 y, 78% were male and 78% had intermediate/poor IMDC risk. 23 pts were on treatment at data cutoff date. Median follow up was 6.9 mo (range: 0.1-18.2). Safety of belzutifan 120 mg + lenvatinib 20 mg was manageable. Of 10 evaluable pts in the safety lead-in phase, only 1 experienced a dose-limiting toxicity (grade 1 dyspnea). Among pts who had opportunity for ≥2 postbaseline scans (n = 24), ORR was 50% (95% CI, 29-71; all PRs); CBR was 54% (95% CI, 33-74). Median DOR was not reached (NR; range: 1.4+ to 14.0+ mo); 74% of responders remained in response for ≥12 mo by KM estimation. For all enrolled pts, median PFS was 11.2 mo (95% CI, 4-NR); 6-mo rate was 55%. In the safety analysis, 28 pts (93%) experienced a treatment-related AE (TRAE), most commonly (≥40%) anemia (43%), fatigue (43%), and hypertension (43%). Grade 3-4 TRAEs occurred in 15 pts (50%); most commonly hypertension (27%) and anemia (17%). 1 pt experienced grade 2 hypoxia. No pt died due to a TRAE. Conclusions: Preliminary data from the belzutifan + lenvatinib combination exhibited promising antitumor activity in pts with ccRCC that progressed on PD-1/L1 inhibitors and VEGF-TKIs. Safety findings were consistent with individual profiles of each agent. The combination is being further evaluated in the phase 3 LITESPARK-011 study. Clinical trial information: NCT04626518 .

We hypothesize that a significant number of patients do not respond to PD-1/L1 blockade because there are no pre-existing tumor antigen-specific T-cells, and this can be addressed by combination therapy with an oncolytic virus such as T-VEC. S1607 is a single arm Phase 2 study of T-VEC plus pembro in patients with advanced melanoma after PD-1/L1 inhibitor progression. The primary endpoint is ORR by modified RECIST (progression at the first follow-up disease assessment had to be confirmed). Secondary endpoints include durable response rate (response ≥ 6 months), ORR in injected, non-visceral non-injected, and visceral lesions, PFS, OS and toxicity. In Cohort A patients must have at least one measurable visceral lesion; in Cohort B patients must not have any visceral lesions. Each cohort had an independent accrual goal with a 2-stage design. All received intratumoral T-VEC and pembro 200mg IV every 21 days. Tumor biopsy and research blood are taken at baseline and on Day 28 (both injected and non-injected lesions). Tumor assessments are performed every 12 weeks for up to 2 years. 38 evaluable patients were enrolled. As of July 26, 2022, the median follow up was 28 months. Treatment was well tolerated, with 5/38 (13%) grade 3 AE (no grade 4/5) including injection site reactions, lymphocyte count decrease, and hypoxia. Cohort A was closed after stage I (n=11) with no confirmed responses. In Cohort B (n=27), there were 7 confirmed responses (26%; 2 CR, 5 PR; this rejected H0: ORR = 10%, p=0.01). Clinical outcomes are summarized in Table 1. Baseline tumor mutational burden from 17 patients in Cohort B were not different between responder vs non-responders (p=0.96). Translational study is ongoing for pharmacodynamic confirmation. T-VEC plus pembro in melanoma patients who have progressed on prior anti-PD1/L1 therapy has efficacy in the subset of melanoma patients who have non-visceral metastases. Table 1 Cohort A (Visceral) Cohort B(Non-Visceral) N (%; 95% CI) 11 27 Confirmed PR + CR 0 (0%; 0%-28%) 7 (26%; 11%-46%) Confirmed + Unconfirmed 1 (9%; 0%-41%) 9 (33%; 17%-54%) Durable response 0 (0%; 0%-28%) 4 (15%; 4%-34%) Median PFS in months 2.1 (0.7-5.5) 2.3 (1.9-6.2) INJECTED LESIONS 11 27 Confirmed PR + CR 0 (0%; 0%-28%) 6 (22%; 9%-42%) Confirmed + Unconfirmed, PR + CR 1 (9%; 0%-41%) 8 (30%; 14%-50%) NON-INJECTED, NON-VISCERAL LESIONS 8 19 Confirmed PR + CR 0 (0%; 0%-37%) 3 (16%; 3%-40%) Confirmed + Unconfirmed, PR + CR 0 (0%; 0%-37%) 5 (26%; 9%-51%) VISCERAL LESIONS 11 Confirmed PR + CR 0 (0%; 0%-28%) Confirmed + Unconfirmed, PR + CR 1 (9%; 0%-41%) ACQUIRED RESISTANCE 3 2 Confirmed PR + CR 0 (0%; 0%-71%) 2 (100%; 16%-100%) Confirmed + Unconfirmed, PR + CR 0 (0%; 0%-71%) 2 (100%; 16%-100%) Median PFS in months 2.1 (2.0-4.1) NR (8.0-∞) PRIMARY RESISTANCE 8 25 Confirmed PR + CR 0 (0%; 0%-37%) 5 (20%; 7%-41%) Confirmed + Unconfirmed, PR + CR 1 (13%; 0%-53%) 7 (28%; 12%-49%) Median PFS in months 1.8 (0.3-6.2) 2.1 (1.9-3.3) Citation Format: Siwen Hu-Lieskovan, James Moon, John Hyngstrom, Katie M. Campbell, Gino K. In, Theodore F. Logan, Kari L. Kendra, Ding M. Wang, Douglas B. Johnson, Gary C. Doolittle, Alan Tan, Ann W. Silk, Kenneth F. Grossmann, Christopher W. Ryan, Sapna P. Patel, Shay Bellasea, Michael C. Wu, John M. Kirkwood, Helen X. Chen, Antoni Ribas. Combination of talimogene laherparepvec (T-VEC) with pembrolizumab (pembro) in advanced melanoma patients following progression on a prior PD-1 inhibitor: SWOG S1607 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3275.