Figure S13 shows H4K20 methylation levels, chromatin binding of BRCA1 and BARD1, and focal accumulation of BRCA1, RPA, and RAD51 in WT and SUV4-20H1/H2 KO DU145 cells treated with etoposide.

Figure S4 shows the lethal synergy profiles of various prostate cancer cell lines for the combination of etoposide with A-196, and of PC3 cells with EZH2 inhibitor.

Figure S5 shows the synergistic cytotoxicity of A-196 and etoposide across various cancer cell types.

Figure S14 shows the absence of toxicity of the treatment A-196 and etoposide in athymic nude male mice.

Figure S11 shows the time-lapse analysis of cell cycle progression in DU145 cells treated with the combination of A-196 and etoposide.

Figure S2 shows Fraction Genome Altered, MSI MANTIS Score, Aneuploidy Score, Buffa Hypoxia Score, and Winter Hypoxia Score in patients stratified by SUV4-20H2 expression levels.

Figure S1 shows SUV4-20H2 expression across Gleason scores, prostate cancer patient categories, TP53 mutation status, ERG structural variants, two neoplasm disease lymph node stages, and with or without SPOP mutations and TMPRSS2 structural variants.

Figure S3 shows cell cycle profile and proliferation of DU145 cells treated with various concentration of the chemical drug A-196.

Figure S8 shows the levels of 53BP1 and γ-H2AX staining in DU145 WT and DU145 SUV4-20H KO cells treated with etoposide.

Figure S9 shows the phosphorylation levels of RPA and DNAPKcs in A-196-treated cells.

Figure S10 shows the molecular and cellular responses of castration-resistant 22RV1 cells to combined treatment with etoposide and A-196.

Reliable prognostic biomarkers are needed to guide treatment decisions for patients with metastatic kidney cancer receiving immune checkpoint inhibitors (ICIs). Radiomics biomarkers leveraging quantitative imaging features from longitudinal CT scans have demonstrated prognostic utility across multiple tumor types, including advanced non-small cell lung cancer (NSCLC). Evaluating such biomarkers in additional tumor types may further enhance personalized treatment approaches. Evaluate the performance of a deep learning biomarker for predicting overall survival (OS) in metastatic kidney cancer patients receiving ICIs. Serial computed tomography response score (Serial CTRS) is a fully automated deep learning radiomics biomarker that predicts OS by analyzing paired baseline and early-treatment thoracic CT scans, typically capturing thoracic and upper abdominal disease. Serial CTRS was previously validated in advanced NSCLC patients receiving programmed death-ligand 1 (PD-L1) ICIs, demonstrating superior OS prediction compared to conventional RECIST and tumor volume metrics in retrospective real-world and clinical trial datasets. This study retrospectively analyzed paired baseline (within 90 days prior to ICI initiation; median: 21 days prior) and follow-up (28–120 days post-ICI initiation; median: 80 days) thoracic CT scans from 117 metastatic kidney cancer patients treated with ICIs within Providence Health System. Of these, 87 patients (median age: 66 years; IQR: 59–72) with available paired scans were included. Predictive performance of Serial CTRS for OS was assessed using Cox proportional hazards models, concordance index (C-index), and area under the receiver operating characteristic curve (ROC-AUC) for OS at 6, 12, and 24 months. Serial CTRS demonstrated significant association with OS, yielding robust risk stratification (C-index: 0.73; 95% CI: 0.65–0.80). ROC-AUC values showed strong predictive accuracy for OS at 6 months (0.87; 95% CI: 0.79–0.94), 12 months (0.78; 95% CI: 0.66–0.90), and 24 months (0.73; 95% CI: 0.60–0.87). Kaplan-Meier analysis using predetermined thresholds from prior NSCLC datasets revealed clear survival stratification among Serial CTRS groups: low- versus high-survival probability (HR=5.01; 95% CI: 2.15–11.68), low- versus intermediate-survival probability (HR=3.25; 95% CI: 1.58–6.69), and intermediate- versus high-survival probability (HR=1.84; 95% CI: 0.82–4.15). Serial CTRS demonstrated robust and significant predictive utility for OS in metastatic kidney cancer patients receiving ICIs, consistent with previous validation in advanced NSCLC. Its fully automated methodology, which requires no manual lesion annotations, may facilitate scalable and objective clinical implementation enhancing prognostication and optimizing treatment stratification in oncology clinical trials and clinical practice. Further prospective validation exploring integration of Serial CTRS into clinical trial designs is warranted. Chiharu Sako, Taly G. Schmidt, Beatriz G. Lourenco, Karishma Sewaramani, Ross McCall, Ryan Beasley, Arpan A. Patel, Dwight H. Owen, Arya Amini, Ronan J. Kelly, Ray D. Page, Jean-Paul Beregi, Stephane Sanchez, Olivier Gevaert, George R. Simon, Ravi B. Parikh, Petr Jordan, Brendan D. Curti. Validation of a Deep Learning Serial Computed Tomography Response Biomarker for Predicting Overall Survival in Metastatic Kidney Cancer Treated with Immune Checkpoint Inhibitors [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr A029.

Introduction:Targeted therapies and immune checkpoint inhibitors (ICIs) have revolutionized the management of metastatic non-small cell lung cancer (NSCLC) over the past decade. Methods:This single-center observational study was conducted to describe programmed death-ligand 1 (PD-L1) testing, choice of therapy, and outcomes for adult patients with stage IV NSCLC initiating first-line therapy from 2017 through 2020, with follow-up through June 2021. Patient characteristics and study assessments were described according to four histomolecular subtypes, defined by histologic characteristics and availability of standard-of-care therapies for molecular subgroups at the time of study conduct. Results:Of 507 eligible patients with metastatic NSCLC, 85 (17%) had squamous NSCLC; 288 (57%) had nonsquamous NSCLC with no actionable genomic alteration; 44 (9%) had nonsquamous NSCLC with KRAS G12C mutation; and 90 (18%) had nonsquamous NSCLC with ROS1, BRAF V600E, EGFR exon 20 insertion, or RET or NTRK genomic alteration. Most tumors were PD-L1 tested. After excluding 40 patients whose PD-L1 testing status was unknown, all but 55 tumors (12%) were tested for PD-L1 expression, and the percentages tested rose from 86% in 2017 to 100% in 2020. From 27% of nonsquamous NSCLC with no actionable genomic alteration to 46% of KRAS G12C-mutated NSCLC had PD-L1 expression ≥ 50%. Use of chemotherapy decreased and use of ICI-chemotherapy combinations increased from 2017 to 2020. In the squamous NSCLC group, single or combination chemotherapy was administered most commonly (42%), whereas ICI-chemotherapy combinations were the most common first-line regimens in the three nonsquamous NSCLC histomolecular groups. For patients with NSCLC and no actionable genomic alterations, ICI-chemotherapy combinations were the most common regimens in 2018-2020 in all but the PD-L1 ≥ 50% category, for whom ICI monotherapy was most common every year except 2020. Median overall survival was 25.0 months (95% CI, 19.1-28.3) for all patients, and, by histomolecular cohort, 14.3 months for squamous NSCLC, 25.3 months for nonsquamous NSCLC with no actionable genomic alteration, not reached for KRAS G12C-mutated NSCLC, and 27.7 months for nonsquamous NSCLC with other genomic alterations. Conclusion:Study findings highlight the increased use of PD-L1 testing over the years from 2017 to 2020 and recent changes in therapy, with decreased use of chemotherapy and increased use of ICI-chemotherapy combinations during the study in each histomolecular group. Moreover, we observed improvements in survival for patients with metastatic NSCLC relative to historical real-world data.

Early response assessment and identification of patients (pts) with NSCLC likely to derive long-term benefit from immune checkpoint inhibitors (ICIs) are crucial for treatment planning and drug development. We developed a deep learning pipeline using a large multi-institute real-world dataset to generate a serial CT response score (SerialCTRS) from paired pre-treatment and 12wk scans, estimating probability of 1y OS. External validation applied SerialCTRS to an expansion cohort of a single-arm phase 1 clinical trial of dostarlimab as ≥2L therapy for advanced NSCLC (GARNET [cohort E], NCT02715284). OS hazard ratios (HRs) of SerialCTRS were compared to tumor volume change derived from manual segmentations and 12 wk RECIST 1.1 overall response using the same input scans. Of 67 clinical trial pts, 49 were evaluable for all 3 models predicting OS. Matching group sizes to RECIST overall response, 13 pts (26%, the number with RECIST PD) had low probability of 1y OS (SerialCTRS 0.12-0.64), 21 pts (43%, number with SD) had medium probability (0.64-0.82), and 15 pts (31%, analogous to CR/PR) had high probability of 1y OS (0.83-0.90). HRs of adjacent categories predicting OS showed superior performance of SerialCTRS, especially for SD/medium vs CR/PR/high probability (HR 2.68 [95% CI 1.09-6.59]; Table). Continuous SerialCTRS predicted OS with a concordance index of 0.717 (0.733 in subset with RECIST 12 wk SD) and remained a significant predictor of OS in bivariable survival models controlling for known predictors of OS such as age, ECOG, stage at diagnosis (all n=54), and best overall response from previous therapy (n=49). Deep learning-based SerialCTRS, without manual annotation, improved prediction of OS vs RECIST and tumor volume change in the NSCLC cohort of the GARNET trial. Brenda F. Kurland, Chiharu Sako, Jing He, Marius de Groot, Taly G. Schmidt, Arpan A. Patel, Dwight H. Owen, Arya Amini, Brendan D. Curti, Ronan J. Kelly, Ray D. Page, Aurelie Swalduz, Jean-Paul Beregi, Jan Chrusciel, Stephane Sanchez, Richard Rosenberg, Jakob Weiss, An Liu, Olivier Gevaert, George R. Simon, Ravi B. Parikh, Alma Hart, Petr Jordan, Jasper van der Aart. Deep learning response score using baseline and 12-week RECIST chest CTs enhances overall survival (OS) prediction in advanced NSCLC: external validation in a trial with dostarlimab [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 736.

1559 Background: Identifying advanced non–small cell lung cancer (aNSCLC) patients who derive long-term benefit from immune checkpoint inhibitors (ICIs) remains a significant challenge. Radiomic analyses, particularly leveraging deep learning, hold promise for improving prognostic accuracy beyond tumor size metrics. We developed serialCTRS, a novel biomarker using deep learning to quantify thoracic CT changes from baseline to 3 months post-treatment, predicting overall survival (OS) in patients receiving PD-(L)1 inhibitors. Methods: SerialCTRS was previously trained and validated on a multi-institutional Real-World Dataset (RWD) (training: 1,171 aNSCLC patients, 14,424 CT scans; validation: 612 patients; Sako et al. SITC, 2024). For this study, we retrospectively validated serialCTRS in two distinct cohorts of aNSCLC patients: (1) a clinical trial (N = 52) treated with the PD-1 inhibitor sasanlimab in the second or later line and (2) a fully blinded RWD from Baylor Scott & White Health system (N = 147), an institution not used for training. The pipeline—spanning image quality control, preprocessing, feature extraction, and survival modeling—operated without manual annotations. To enhance interpretability, we developed 3D submodels for prognostic signals related to (i) tumor burden, (ii) body composition, and (iii) lung vasculature. Predictive performance was compared to RECIST 1.1 using concordance index (c-index) and ROC-AUC for 24-month OS (OS24 AUC). Results: SerialCTRS outperformed RECIST in OS prediction and remained a significant predictor after multivariate adjustments with other known predictors including age, sex, PD-L1 TPS, and NLR across both validation cohorts. In the sasanlimab cohort, serialCTRS achieved a c-index of 0.77, surpassing RECIST (0.72), with an OS24 AUC of 0.86 (95% CI: 0.74–0.98). In the Baylor cohort, serialCTRS demonstrated a c-index of 0.68 vs. RECIST (0.62) and an OS24 AUC of 0.76 (0.67–0.86). Submodels targeting individual components achieved c-indices of 0.65 (tumor burden), 0.61 (body composition), and 0.61 (vasculature) in the sasanlimab cohort, and 0.63, 0.61, and 0.59, respectively, in the Baylor cohort. Combining the submodels improved c-indices to 0.69 (sasanlimab) and 0.66 (Baylor), demonstrating complementary signal among radiographic features. Conclusions: SerialCTRS outperformed RECIST 1.1 in predicting OS in independent clinical trial and RWD datasets. Interpretable submodels highlighted the prognostic value of tumor burden, body composition, and vasculature changes. SerialCTRS offers a promising tool for personalizing therapy and accelerating drug development in aNSCLC, with a fully automated pipeline for robust and scalable clinical use. Future work will focus on larger, more diverse cohorts to validate utility in guiding precision oncology.

AbstractGT103 is a first-in-class, fully human, IgG3 monoclonal antibody targeting complement factor H that kills tumor cells and promotes anti-cancer immunity in preclinical models. We conducted a first-in-human phase 1b study dose escalation trial of GT103 in refractory non-small cell lung cancer to assess the safety of GT103 (NCT04314089). Dose escalation was performed using a “3 + 3” schema with primary objectives of determining safety, tolerability, PK profile and maximum tolerated dose (MTD) of GT103. Secondary objectives included describing objective response rate, progression-free survival and overall survival. Dose escalation cohorts included GT103 given intravenously at 0.3, 1, 3, 10, and 15 mg/kg every 3 weeks, and 10 mg/kg every 2 weeks. Thirty one patients were enrolled across 3 institutions. Two dose-limiting adverse events were reported: grade 3 acute kidney injury (0.3 mg/kg) and grade 2 colitis (1 mg/kg). No dose-limiting toxicities were noted at the highest dose levels and the MTD was not reached. No objective responses were seen. Stable disease occurred in 9 patients (29%) and the median overall survival was 25.7 weeks (95% confidence interval [CI], 19.1–30.6). Pharmacokinetic analysis confirmed an estimated half life of 6.5 days. The recommended phase 2 dose of GT103 was 10 mg/kg every 3 weeks, however further dose optimization is needed given the absence of an MTD. The study achieved its primary objective of demonstrating safety and tolerability of GT103 in refractory NSCLC.

Background In non-small cell lung cancer, social determinants of health (SDOH) influence treatment, but SDOH with geographic precision are infrequently used in real-world research because of privacy considerations. This research aims to characterize the influence of census tract-level SDOH on treatment for stage I and IIa non-small cell lung cancer.Methods Patients diagnosed between January 1, 2017, and September 30, 2022, with stage I or IIa non-small cell lung cancer in the Syapse Learning Health Network had their addresses geocoded and linked to 6 census tract-level indicators of SDOH (the Centers for Disease Control and Prevention and Agency for Toxic Substances and Disease Registry Social Vulnerability Index, percentage housing burden, percentage broadband internet access, primary care shortage area, and rurality). Clinical and demographic characteristics were ascertained from medical records. Nested multinomial logistic regression models estimated associations between SDOH and initial treatment using 2-sided Wald tests. The collective statistical significance of SDOH was assessed using a likelihood ratio test comparing nested models. Descriptive statistics described time to treatment initiation.Results Among 3595 patients, 58% were initially treated with surgery, 29% with radiation, and 12% with "other." Two SDOH variables were associated with increased relative risk for radiation therapy compared with surgery: living in primary care shortage areas (relative risk = 1.61, 95% CI = 1.23 to 2.10) and living in nonmetropolitan areas (relative risk = 1.45, 95% CI = 1.02 to 2.07). The likelihood ratio test suggested that the 5 SDOH variables collectively improved the treatment model. Further, patients in areas with high Social Vulnerability Index, low internet access, and high housing burden initiated treatment later.Conclusion When using precise estimates of geospatial SDOH, these measures were associated with treatment and should be considered in analyses of cancer outcomes.

PURPOSEThis study developed and validated a novel deep learning radiomic biomarker to estimate response to immune checkpoint inhibitor (ICI) therapy in advanced non-small cell lung cancer (NSCLC) using real-world data (RWD) and clinical trial data.MATERIALS AND METHODSRetrospective RWD of 1,829 patients with advanced NSCLC treated with PD-(L)1 ICIs were collected from 10 academic and community institutions in the United States and Europe. The RWD included data sets for discovery (Data Set A-Discovery, n = 1,173) and independent test (Data Set B, n = 458). A radiomic pipeline, containing a deep learning feature extractor and a survival model, generated the computed tomography (CT) response score (CTRS) applied to the pretreatment routine CT/positron emission tomography (PET)-CT scan. An enhanced CTRS (eCTRS) also incorporated age, sex, treatment line, and lesion annotations. Performance was evaluated against progression-free survival (PFS) and overall survival (OS). Biomarker generalizability was further evaluated using a secondary analysis of a prospective clinical trial (ClinicalTrials.gov identifier: NCT02573259) evaluating the PD-1 inhibitor sasanlimab in second or later line of treatment (Data Set C, n = 54).RESULTSIn RWD Test Data Set B, the CTRS identified patients with a high probability of response to ICI with a PFS hazard ratio (HR) of 0.46 (95% CI, 0.26 to 0.82) and an OS HR of 0.50 (95% CI, 0.28 to 0.92) in the first-line ICI monotherapy cohort, after adjustment for baseline covariates including the PD-L1 tumor proportion score. In Clinical Trial Data Set C, the CTRS demonstrated an adjusted PFS HR of 1.03 (95% CI, 0.43 to 2.47) and an OS HR of 0.33 (95% CI, 0.14 to 0.91). The CTRS and eCTRS outperformed traditional imaging biomarkers of lesion size in PFS and OS for RWD Test Data Set B and in OS for the Clinical Trial Data Set.CONCLUSIONThe study developed and validated a deep learning radiomic biomarker using pretreatment routine CT/PET-CT scans to identify ICI benefit in advanced NSCLC.

e14588 Background: CFH is a complement regulatory protein that protects cells from alternative complement pathway activation. Neutralizing antibodies against CFH on tumor cells increase deposition of C3b, facilitate antibody-dependent-cellular phagocytosis and complement dependent cytotoxicity, inhibit tumor growth, and modulate anti-tumor immunity. GT103 is a first-in class, fully human-derived IgG3 anti-CFH monoclonal antibody that recognizes a tumor specific epitope on CFH and has shown anti-tumor activity in preclinical models. Methods: We conducted a phase Ib, first-in-human, dose escalation trial evaluating safety and preliminary efficacy of GT103 in patients (pts) with advanced NSCLC refractory to standard therapies. Pts received GT103 intravenously across six dose levels/schedules following 3+3 design. Primary objectives were to determine maximum tolerated dose (MTD), recommended phase II dose (RP2D), and pharmacokinetic profile of GT103. Secondary objectives were to assess objective response rate, progression free survival (PFS), and overall survival (OS). Here we present updated safety and efficacy results of the study. Results: 31 pts were enrolled from 6/2020 to 9/2023 (median age 63yrs (range, 23-79), 61% had adenocarcinoma, 32% had previously treated brain metastasis). MTD was not reached. Treatment related AEs (TRAEs) observed in ≥10% pts included fatigue (19%), anemia (16%), diarrhea (16%), and nausea (13%). 3 pts developed ≥ grade 3 TRAE, including acute kidney injury, decreased lymphocyte count, and anemia. The best treatment response was stable disease in 9 (29%) pts. Median(m) PFS and mOS were 6 weeks (95% CI 6.0-6.1) and 25.7 weeks (95% CI 19.1-30.6), respectively. 24-week PFS and OS were 9.7% (95% CI 2.5-22.9%) and 50.6% (95% CI 31.9-66.6%), respectively. PD-L1 expression was available in 25 pts. No statistically significant difference (p=0.33) was noted in mPFS between PD-L1 positive (TPS > 1%) vs negative (TPS ≤1%) pts. KRAS and EGFR mutation status was available for 22pts. mPFS and mOS were numerically longer in pts with KRAS positive vs negative disease. While there was no difference in mPFS, 24-weeks PFS was 0 vs 16.7% (95% CI 4.1-36.5%) EGFR MUT vs wild type disease. Biomarker analysis of complement regulatory proteins, FcGR expression and polymorphisms, and changes in sC5b-9 is underway. Conclusions: GT103 was well tolerated across all dose levels. The RP2D of 10mg/kg IV every 3 weeks was selected, although PK modeling will be performed to optimize the dosing schedule. Encouraging anti-tumor activity was seen in heavily pretreated pts with advanced NSCLC with a subset of patients experiencing stable disease lasting for longer than 6 months. Additional biomarker analysis is ongoing. A phase II trial combining GT103 with pembrolizumab in pts with advanced NSCLC is currently enrolling. Clinical trial information: NCT04314089 .