BACKGROUND Clear cell renal cell carcinoma (ccRCC) with pancreatic metastases (PM) is paradoxically associated with prolonged overall survival (OS), but the biological basis for this observation remains unclear. METHODS We analyzed matched primary and metastatic samples from an international consortium of patients with PM ( n = 108) and compared them with a previously characterized ccRCC cohort without PM ( n = 273). RESULTS Primary ccRCC tumors associated with PM were dominated by indolent, angiogenic phenotypes, characterized by low-grade histology and reduced mTORC1 activation (all P < 0.001). Tumors of patients with PM were often PBRM1-deficient (80.4% vs. 54.8%, P < 0.001) and rarely harbored BAP1 loss (3.7% vs. 20.7%, P < 0.001). After metastasis diagnosis, patients with PM had significantly longer median OS compared with those without PM (110 vs. 33 months, HR 0.28 [95% CI, 0.19–0.39], P < 0.001). Survival was further prolonged among patients with PBRM1 loss (143 vs. 64 months, HR 0.41 [95% CI, 0.22–0.81], P = 0.008). Notably, PM lesions were typically low-grade and PBRM1-deficient even when more aggressive and evolved clones were present in primary tumors. Finally, PBRM1 loss was associated with preferential response to angiogenesis inhibitors over immune-oncology therapy, reflected by longer time on treatment (32.1 vs. 9.1 months, HR 0.16 [95% CI, 0.06–0.39], P < 0.001). CONCLUSION These findings illustrate selective tropism of indolent, less-evolved, PBRM1-deficient ccRCC clones for pancreatic dissemination. This biological bias likely underlies therapeutic sensitivity and favorable survival, supporting the consideration of PBRM1 status and metastatic tropism in risk stratification and treatment selection. FUNDING NIH Kidney Cancer SPORE grant (P50CA196516); The Cancer Prevention and Research Institute of Texas (RP220294); Endowment from Jan and Bob Pickens Distinguished Professorship in Medical Science and Brock Fund for Medical Science Chair in Pathology.

Copy number aberrations (CNAs) are gains and losses of large genomic segments present across most cancer types and are a hallmark of cancer genomic alterations. However, the processes underlying CNAs and characteristic patterns of CNAs are poorly understood. Bioinformatic advances have identified underlying single nucleotide variant mutational signatures resulting from distinct mutational processes, yet development of algorithms able to uncover similar signatures for CNAs remains less advanced. Using segmented data files from DNA sequencing, six copy number features are extracted for signature determination: segment size, breakpoints, copy number oscillation, changepoint size, copy number, and breakpoints per chromosome arm, along with ploidy. Mixed model approaches and non-negative matrix factorization are utilized to derive CNA signatures across cancer types. The full methodology was packaged in a publicly available, robust R package, CNSigs. To verify reproducibility, we derived five signatures from two independent breast cancer datasets (total n>3000), demonstrating high accuracy (average cosine similarity = 0.89). Pan-cancer application of CNSigs in TCGA resulted in derivation of 13 pan-cancer signatures which were significantly associated with disease-specific survival. Benchmarking CNSigs to two other CNA signature approaches within TCGA demonstrated non-overlapping signatures and favorable compute speed for CNSigs. We evaluated n=24 pairs of tumor and circulating tumor DNA (ctDNA) that demonstrated that CNSigs are detectable and reproducible via ctDNA, with significant association of CNSig11 with metastatic triple-negative breast cancer progression-free survival specifically for taxane chemotherapy. CNSigs association with immunophenotype was evaluated in low-grade glioma and CNSig3 was found to be highly prognostic yet complementary to immune features. The CNSigs allows researchers to easily analyze their own samples to derive copy number signatures and evaluate clinical associations. We demonstrate its potential application in ctDNA and association with treatment response. The development of this package allows further investigation of underlying processes that may be responsible for CNA fingerprints.

860 Background: On treatment biomarkers that connect molecular clearance to clinical outcomes are needed in metastatic urothelial carcinoma (mUC). We tested whether circulating tumor DNA (ctDNA) kinetics, especially ctDNA negativity, track response and survival during enfortumab vedotin plus pembrolizumab (EV+pembro) treatment. Methods: This single-center retrospective study included 25 mUC patients treated with EV+pembro at The Ohio State University Comprehensive Cancer Center between August 2023 and August 2025. A tumor informed assay (Signatera) measured ctDNA longitudinally. Metrics included baseline detectability, early decrease (≤60 days), ctDNA negativity (0 MTM/mL) at any time, and rise after nadir. PFS and OS were measured from treatment start using Kaplan Meier and log rank methods. Hazard ratios were derived from univariate Cox regression analysis. Fisher’s exact test was used to assess associations with the ORR. Results: Median age 69 yrs; 68% male; ECOG 0–1 = 80%. Primary tumor site was bladder (80%) or upper tract (20%). Median follow up was 15.4 months. Baseline ctDNA was available for 18/25 patients and was detectable in all 18 (100%). Among those with baseline ctDNA, an on-treatment decrease occurred in 15/18 (83%) patients, including 13/18 (72%) within ≤60 days. Overall, ctDNA negativity was achieved in 15/25 (60%) patients. The best overall response was 5 CR, 14 PR, 4 SD, and 2 PD, for an ORR of 76%. ctDNA negativity aligned with response: the ORR was 100% (15/15) in patients who achieved ctDNA negativity versus 40% (4/10) in those who did not (p = 0.0012). ctDNA negativity was also associated with significantly improved PFS (HR 0.16, 95% CI 0.05–0.54; p = 0.003) and OS (HR 0.08, 95% CI 0.01–0.68; p = 0.021). Early decrease (≤60 days) showed non-significant trends toward improved PFS (HR 0.88, 95% CI 0.22–3.47; p = 0.86) and OS (HR 0.46, 95% CI 0.05–4.12; p = 0.49). The median PFS for the entire cohort was 9.59 months; median OS was not reached. Among 13 patients with a ctDNA rise after nadir, the median time from nadir to rise was 126 days, and the median time from rise to radiographic progression was 115 days, indicating potential lead time. In exploratory analyses, TP53 mutations (n = 11) trended toward shorter PFS (HR 4.44; p = 0.061) and lower ctDNA negativity (45.5% vs 88.9% without mutation; p = 0.07). TSC1 mutations (n = 3) trended toward shorter PFS (HR 4.51; p = 0.095) but showed no association with ctDNA negativity (p = 1.0). Conclusions: In EV+pembro treated mUC, achieving ctDNA negativity is a powerful on treatment biomarker, coinciding with a near universal radiographic response and substantially longer PFS and OS. An early decrease in ctDNA similarly indicated treatment activity and trended toward improved outcomes. These hypothesis generating findings support the prospective validation of ctDNA guided management strategies.

BACKGROUND AND OBJECTIVE:Lynch syndrome (LS) and germline DNA damage repair (DDR) mutations have been described in urothelial carcinoma (UC). However, the incidence in unselected cohorts and the impact on outcomes remain unclear. DESIGN:We performed a retrospective study to determine the frequency of pathogenic/likely pathogenic (P/LP) germline variants in 78 known cancer predisposition genes among 273 patients with urinary tract cancer (UTC) of the bladder, renal pelvis, ureter, and/or urethra and validated the frequency in an independent cohort of 5972 patients. We identified variants in germline whole-exome sequencing that were overrepresented in UTC. We measured associations with treatment-related outcomes. RESULTS AND LIMITATIONS:In unselected cohorts, 9.3-9.5% of the patients with UTC harbored a variant in a cancer predisposition gene. No clinicodemographic variable predicted the presence of a P/LP variant. LS was found in 0.7-0.8% of the patients who were more likely to have upper tract disease and strong personal and family histories of malignancy. Non-Lynch DDR variants occurred in 6.6-8.0% of the patients, most often in CHEK2, BRCA1, BRCA2, and ATM. Very small cohorts of patients with UC and LS or a DDR variant responded well to immune checkpoint inhibitors (ICIs) or platinum chemotherapy, respectively. Standard variant calling methods may miss large deletions. CONCLUSIONS:The incidence of P/LP germline variants in UTC is clinically meaningful, and more than one-third of the patients are missed with current guidelines. The response of UC with LS to ICI and with DDR to platinum is hypothesis generating and warrants further investigation.

748 Background: Metastatic neuroendocrine bladder cancer is a rare and aggressive malignancy characterized by poor prognosis and limited effective treatment options. Standard chemotherapy has shown limited success, necessitating novel therapies. We hypothesized that adding immunotherapy to chemotherapy would improve overall survival (OS) in these patients. Methods: We conducted a retrospective cohort study of 1,410 patients diagnosed with metastatic neuroendocrine bladder cancer between 2004 and 2020 using the National Cancer Database (NCDB). Treatment groups included no treatment, chemotherapy alone, and combined immunochemotherapy. Survival outcomes were analyzed using Kaplan-Meier curves, log-rank tests, and Cox proportional hazards regression models, adjusting for age, tumor size, Charlson-Deyo comorbidity score, and residential area. Results: Median OS was 1.8 months (95% CI: 1.6-2.0) for patients receiving no treatment (n=460), 9.6 months (95% CI: 8.9-10.1) for chemotherapy alone (n=860), and 11.7 months (95% CI: 9.6-14.8) for combined immunochemotherapy (n=90) (P < 0.001 for immunochemotherapy vs. both other groups). Further analysis revealed that first-line immunotherapy initiated within 30 days of chemotherapy (n=21) significantly improved OS (median 11.6 months; 95% CI: 6.8–16.1; P=0.0374) compared to chemotherapy alone. Notably, immunotherapy initiated after 30 days (n=69, median OS 12.4 months; 95% CI: 9.0–14.8; P=0.0264), whether as first-line or second-line therapy, also showed superior efficacy to chemotherapy alone. These findings suggest that if first-line immunotherapy is not administered, subsequent immunotherapy should still be considered. Multivariable Cox regression analysis showed that immunochemotherapy compared to chemotherapy alone was associated with a lower hazard for death (HR = 0.80; 95% CI: 0.66-0.97; P = 0.0242), while patients receiving no treatment had an increased risk of death (HR = 3.50; 95% CI: 3.00-3.90; P < 0.0001). Additional factors associated with increased mortality included older age, larger tumor size, and higher Charlson-Deyo score. Conclusions: Adding immunotherapy to chemotherapy significantly improved OS in metastatic neuroendocrine bladder cancer. First-line immunochemotherapy showed substantial benefits, but late initiation also improved outcomes. These findings support incorporating immunotherapy into first-line protocols and consider it for subsequent lines. Prospective clinical trials are needed to optimize immunotherapeutic strategies for this aggressive malignancy.

597 Background: Early detection of recurrence in urothelial carcinoma (UC) and renal cell carcinoma (RCC) may improve patient outcomes. Conventional imaging techniques may not identify early molecular relapses that precede clinical relapse. Circulating tumor DNA (ctDNA) has emerged as a promising biomarker for the early detection of disease relapse. This study aims to evaluate the concordance of molecular relapses by ctDNA and radiographic progression in patients with UC and RCC who are off immunotherapy due to intolerance of side effects or completion of treatment. Methods: We conducted a retrospective study of locally advanced/advanced UC and RCC patients who were treated at The Ohio State University James Cancer Hospital from November 2021 to September 2024. Patients who completed planned immunotherapy followed by imaging and ctDNA surveillance were included. ctDNA levels were measured using the Signatera tissue informed assay at intervals of 4 to 12 weeks. Imaging surveillance schedules were at the treating physician discretion. Patient demographics, clinical characteristics, and ctDNA levels were recorded and analyzed. Results: A total of 36 patients were included: muscle-invasive bladder cancer (MIBC, n = 10), metastatic urothelial carcinoma (mUC, n = 10), locally advanced RCC (n = 4), and metastatic RCC (mRCC, n = 12). The median follow-up was 39.5 weeks (range: 1–138 weeks), with a median age of 67.5 years (range: 41–83 years), and 72.2% were male. At the time of immunotherapy discontinuation, ctDNA was negative in all patients except one mRCC patient (ctDNA 0.95 MTM/mL). During surveillance, 9 patients (25%) - 5 mUC and 4 mRCC - became ctDNA-positive, of which, eight showed radiographic progression, while one mRCC patient (ctDNA 0.15 MTM/mL) did not. None of the 27 patients (75%) that continued to have negative ctDNA had radiographic progression. For the 8 patients with progression, the median time from stopping immunotherapy to ctDNA relapse was 17.5 weeks (range: 11–42 weeks). The median time from ctDNA relapse to imaging-confirmed progression was 22 weeks (range: 13–43 weeks). At progression, ctDNA levels ranged from 0.08 to 136.49 MTM/mL (median: 2.33 MTM/mL). Six of the 20 metastatic patients (30%) relapsed despite negative ctDNA at immunotherapy discontinuation. Conclusions: ctDNA positivity occurs much earlier and is highly concordant with radiographic progression in patients with locally advanced/advanced UC and RCC under post-immunotherapy surveillance. These findings suggest that ctDNA is a promising tool for early disease relapse detection in these patients. Further studies are warranted to validate these results and establish standardized ctDNA surveillance protocols.

Background: Bone-predominant metastatic renal cell carcinoma (mRCC) presents significant clinical challenges due to its associated morbidities and poor prognosis. Optimal first-line treatment remains unclear, largely because these patients are often excluded from clinical trials due to difficulties in measuring bone lesions. Emerging evidence suggests that bone metastases exhibit high angiogenesis gene signatures, potentially predicting favorable responses to tyrosine kinase inhibitors (TKIs). Methods: We conducted a multicenter retrospective analysis of patients with bone-predominant mRCC treated at The Ohio State University Comprehensive Cancer Center and Fred Hutchinson Cancer Center from January 2008 to June 2021. Bone predominance was defined as having a greater number of osseous metastases compared to extra-osseous sites using computed tomography or bone scans. Patients receiving first-line TKIs or immune checkpoint inhibitors (ICIs) were included; those treated with combination TKI-ICI therapies were excluded due to limited numbers. Demographic, clinical, and treatment data were collected. Progression-free survival (PFS) and overall survival (OS) were analyzed using Kaplan-Meier methods and compared using the log-rank test. Univariate Cox regression analysis was conducted to identify factors associated with OS. Results: A total of 69 patients with bone-predominant mRCC were identified, with 40 receiving TKIs and 29 receiving ICIs as first-line therapy. Baseline characteristics were comparable between groups. The median OS was significantly longer for patients treated with TKIs compared to those receiving ICIs (41.3 months vs. 19.3 months; log-rank P = 0.036). A trend toward improved median PFS was observed in the TKI group (7.9 months vs. 4.9 months; P = 0.075). Univariate analysis showed that treatment with ICIs was associated with an increased risk of death compared to TKIs (hazard ratio = 1.96; P = 0.040). Objective response rates were higher in the TKI group (22.9% vs. 12.0%), although this difference was not statistically significant (P = 0.332). Conclusions: In this multicenter real-world analysis, first-line treatment with TKIs was associated with significantly improved OS compared to ICIs in patients with bone-predominant mRCC. These findings suggest that TKI-containing regimens may be the preferred front-line therapy for this patient subgroup. Prospective studies are warranted to validate these results and to optimize treatment strategies for bone-predominant mRCC.

Introduction: The treatment efficacy of prostate cancer (PCa) radiotherapy (RT) can be inadvertently affected by the concurrent usage of non-oncologic medications. Many studies have associated the intake of several non-oncologic drugs with cancer specific outcomes. In this study, we report the impact of daily non-oncologic medications including aspirin, metformin, and statins on time to progression for patients with high-risk PCa. Methods: Patients with high- and very high risk PCa (NCCN definition of Gleason score ≥ 8, prostate-specific antigen (PSA) ≥ 20, or ≥cT3a) who received definitive RT at two institutions were included in this analysis. Progression was defined as either biochemical (PSA > nadir + 2 ng/mL), locoregional (prostate or lymph nodes, biopsy-proven), or development of distant metastases. Progression-free survival (PFS) was defined as the time elapsed from the start of RT to progression or last follow-up. Cox proportional hazards models evaluated the associations between non-oncologic medications and PFS. Results: There were 237 patients eligible for this analysis, of which 47 (19.8%) and 178 (75.1%) had at least clinical T3 disease or at least Gleason 8 disease, respectively. During RT, 82 (34.6%), 88 (37.1%), and 29 (12.2%) patients were taking aspirin, statin, or metformin, respectively. Overall, 54 patients (22.8%) experienced disease progression. Neither aspirin nor statin usage had a significant association with PFS. Patients prescribed metformin displayed worse PFS compared to patients not taking metformin (aHR: 2.46, 95% CI: 1.06–5.72). Conclusions: Aspirin and statin usage was not associated with likelihood of progression in this large cohort of patients with high-/very high risk PCa. Metformin use was associated with poorer PFS, albeit with a small event rate due to fewer patients taking metformin. Further studies are needed to clarify the impact of common non-oncologic medication use on outcomes for patients with high-risk PCa.

TPS899 Background: There is an unmet need to develop neoadjuvant therapies for cisplatin-ineligible patients (pts) with muscle-invasive bladder cancer (MIBC). Neoadjuvant immune checkpoint inhibition (ICI) has shown promising early results in this population; however, > 2/3 pts did not achieve pathological complete response (pCR), a surrogate endpoint for improved survival. Fibroblast growth factor receptor inhibitor (FGFRi) may induce synergy with ICI through modulation of the tumor immune microenvironment. FGFR mRNA overexpression by RNA in situ hybridization (ISH) is an emerging biomarker that may predict responses to an FGFRi ± ICI. Futibatinib (FUTI) is an irreversible pan-FGFRi, and durvalumab (DURV) is an anti-PD-L1 antibody. Methods: This is a phase 2, single arm, multicenter study to evaluate the efficacy and safety of neoadjuvant FUTI + DURV before cystectomy. Pts with stage cT2 to cT4aN0M0, urothelial carcinoma predominant histology, cisplatin-ineligible by Galsky criteria, and FGFR1-3 mRNA expression ≥3+ by RNAscope ISH assay are eligible. Twenty-four pts will be treated with FUTI 20 mg PO daily and DURV 1500 mg IV on day 1 of every 28-day cycle for 3 cycles. The primary outcome is pCR (defined as ypT0N0). We hypothesize that FUTI + DURV will improve the pCR rate from 21% to 41%. This trial has 80% power to detect the alternative hypothesis while maintaining a one-sided α = 0.1. The study will begin with a safety lead-in period, during which 6 pts will be enrolled for dose-limiting toxicity (DLT) evaluations. After confirming safety (≤ 2 DLTs), another 18 pts will be enrolled. The study will be monitored for futility using a Bayesian Optimal Phase II design, including 2 interim analyses (12 and 18 pts). If there are ≥ 3 pCRs among the first 12 pts, we will continue enrollment. If there are ≥ 5 pCRs among the first 18 pts, the study will continue. When 24 evaluable pts are enrolled, we will reject the null hypothesis and conclude that the treatment is acceptable if there are ≥ 8 pCRs. The secondary endpoints include safety, pathologic downstaging rate, progression-free survival, and overall survival. Correlative studies will explore immune and molecular predictors of response and resistance in the tumor, blood, and urine. Clinical trial information: NCT06263153 .

84 Background: Lutetium-177-vipivotide tetraxetan ( 177 Lu) was approved in 2022 for PSMA-positive metastatic castration-resistant prostate cancer (mCRPC) patients (pts) treated with prior androgen receptor signaling inhibition (ARSI) and a taxane. The impact of 177 Lu in a non-trial setting is limited. Methods: We retrospectively reviewed records of mCRPC pts who received ≥1 dose of 177 Lu at The Ohio State University from 3/2022-3/2024. Demographics, tumor histology, metastasis sites (mets), treatment (Tx) history, standardized uptake value max (SUVm), and prostate-specific antigen (PSA), at baseline (BL) and post-Tx were collected. SUVm was defined as the highest SUV from a single lesion. Outcomes were PSA response, PSA50, radiographic progression-free (rPFS) by PCWG3, and overall survival (OS). Descriptive statistics, Mann-Whitney, Chi-square, and Cox regression model were used to assess impact. Results: A total of 152 pts were included with a median follow-up of 9 months (0.1-23 m). The median age was 70 years (46-92 y), with 39% having de novo metastatic disease, and 61% had Gleason grade group ≥4. The common mets were bone (91%), lymph node (68%), and visceral (34% with 17% liver and 8% lung). The majority received prior taxanes and ARSI, and 20% radium-223. The median lines of prior Tx was 5.The median BL PSA was 56.3 ng/mL, and the median BL SUVm was 30.4. Post-Tx, 64% showed a PSA response, and 46% achieved PSA50. Among the pts with available imaging, 17% had partial response, 12% stable disease, and 48% disease progression.Tx was discontinued in 62% for:radiographic progression (66%), PSA-only progression (7%), clinical decline (24%), and toxicity (3%). At the data cutoff, mortality was 50%. Median PFS and OS was 6.7 m, and 12.2 m respectively. BL SUVm was associated with improved PSA50 (p=0.01), rPFS (p<0.01), and OS (p=0.01), and liver mets associated with worse OS (p<0.01). Conclusions: In this retrospective study of 177 Lu, we see similar PSA and PSA50 responses to reported trials. However, PFS was shorter, and mortality was higher, likely related to the use of 177 Lu in heavily treated pts with few remaining options. Optimizing biomarkers to predict Tx benefit and resistance are needed.

870 Background: Studies cite that 10-24% of patients (pts) with urothelial carcinoma (UC) have a pathogenic germline variant (var), but few have looked at association with outcomes. We aimed to identify the frequency of germline pathogenic vars (PV) in UC and hypothesized they would be associated with cancer-related outcomes. Here we present germline results from the largest reported, fully clinically annotated cohort of pts with UC. Methods: We performed a single center retrospective review of all pts with a diagnosis (dx) of bladder, renal pelvis, ureter, or urethra cancer who underwent clinical germline targeted panel genetic testing from 2018 to 2024. We also included all pts with the same dx enrolled on an institutional protocol that performed germline whole exome sequencing (WES) for research. We analyzed the 77 genes from WES most frequently included in the targeted panels. We classified all vars as pathogenic, variant of unknown significance (VUS), or benign. Results: We included 267 pts. Median number of genes tested was 77 (range 1-93) with 83.9% of pts completing a panel of ≥77 genes. Of 267 pts, 48 (18%) had a PV, 138 (52%) had a VUS without a PV, and 81 (30%) had no germline var. The most common PVs were in CHEK2 (n = 5/48 [11%]), ATM (9%), MUTYH (9%), MSH2 (9%), TP53 (7%) and MLH1 (7%). Among the 48 pts with a PV, 12 (25%) had Lynch syndrome and 25 (52%) had a non-Lynch DNA Damage Repair gene var. Of 256 pts with urothelial histology, 47 (18%) had a PV. Of 7 pts with pure squamous histology, 1 (14%) had a PV in ATM , 4 (57%) had a VUS. Three pts with rare histologies of small cell, rhabdomyosarcoma, and adenocarcinoma had a VUS in ATM, RET, or POLD1, respectively. Of 48 pts with a PV, only 12 (25%) ever developed metastatic disease. Comparing pts with PVs (n=48) to not (n=219), we found no significant difference in median age at dx, age at dx ≥ 60, sex, race, ethnicity, smoking history, median number of 1 st degree family members with a dx of cancer, or clinical stage at dx. Pts with a PV had no significant difference in primary tumor location, however pts with Lynch syndrome were more likely to have an upper tract tumor (p = 0.002). Adjusting for clinical stage, there was no significant difference in overall survival between pts with a PV compared to without. Three of 12 pts with Lynch received an immune checkpoint inhibitor (ICI) for non-muscle invasive, muscle invasive, or metastatic disease, respectively, and all had prolonged responses without recurrence at 44, 12, and 70 months of follow-up. Conclusions: Germline PVs were detected in 18% of pts with UC. We confirmed the association of Lynch syndrome with upper tract disease and favorable response to ICI. Germline genetics of rare histologic variants warrants further investigation. Our findings support genetic testing in all pts with UC, including early-stage disease. Further study of ICIs for early-stage UC in Lynch syndrome is needed. Analysis of a larger cohort and response to specific treatments is ongoing.

e17013 Background: Bleomycin induced pulmonary toxicity (BIP) occurs in 4-10% of treated patients. Preclinical reports suggest G-CSF use may increase BIP risk, due to neutrophil migration to lung tissues. Limited real-world clinical data is available and G-CSF use may reduce risk of treatment related toxicity with curative bleomycin combination treatment. Methods: We conducted a retrospective study of patients with GCT (Germ cell tumor) and HL (Hodgkin’s lymphoma) receiving bleomycin containing chemotherapy and at-least on dose of G-CSF at The James Cancer Center from 2010-2022 using the institutional database. BIP was diagnosed mainly using clinical symptoms complemented by radiological evidence when possible. We reviewed 361 patients, 145 in GCT group and 216 in HL group and clinically relevant information was recorded.Descriptive statistics for the study variables were presented using frequencies and percentages overall. The bivariate associations between G-CSF use and clinical outcomes were assessed using chi-square tests or fisher’s exact tests for the overall sample, and by chemotherapy type. We used SPSS 26.0 for all the statistical analyses, and an alpha level of 0.05 (two-sided) was used for all the tests. Results: Data on 361 patients, including 145 GCT and 216 HL patients treated with at least one dose of bleomycin was available. The median age was 36.3 years in the GCT group and 44.2 years in the HL group.144 patients in the GCT and 50 patients in the HL group received at least 1 dose of G-CSF. BIP occurred in 4.8% in the GCT and 7.4% of patients in the HL group. Only 2 out of 16 patients developing BIP in the HL group received G-CSF. All patients with BIP developed symptomatic disease. On bivariate analysis, no association was noted between BIP and G-CSF in either group (p-value =1.00 in GCT and p-value = 0.32 in HL). Febrile neutropenia (FN) was reported in 4 patients in the GCT group and 29 in the HL group. In the HL subgroup, 28 (28/29) patients developing FN did not receive G-CSF (p-value=0.0008). All patients who developed FN required hospital admission. Conclusions: Use of G-CSF in patients receiving bleomycin was not associated with an increased incidence of BIP. Incidence of FN was higher in patients not receiving G-CSF leading to hospital admissions. [Table: see text]

Preclinical and clinical data suggest that androgen receptor signaling strongly contributes to bladder cancer development. The roles of the androgen receptor in bladder carcinogenesis have obvious implications for understanding the strong male sex bias in this disease and for potential therapeutic strategies as well. In this review, we summarize what is known about androgen receptor signaling in urothelial carcinoma as well as in tumor-infiltrating immune cells, reviewing preclinical and clinical data. We also highlight clinical trial efforts in this area.

Blood-based liquid biopsy is increasingly used in clinical care of patients with cancer, and fraction of tumor-derived DNA in circulation (tumor fraction; TFx) has demonstrated clinical validity across multiple cancer types. To determine TFx, shallow whole-genome sequencing of cell-free DNA (cfDNA) can be performed from a single blood sample, using an established computational pipeline (ichorCNA), without prior knowledge of tumor mutations, in a highly cost-effective manner. We describe assay validation of this approach to facilitate broad clinical application, including evaluation of assay sensitivity, precision, repeatability, reproducibility, pre-analytic factors, and DNA quality/quantity. Sensitivity to detect TFx of 3% (lower limit of detection) was 97.2% to 100% at 1× and 0.1× mean sequencing depth, respectively. Precision was demonstrated on distinct sequencing instruments (HiSeqX and NovaSeq) with no observable differences. The assay achieved prespecified 95% agreement of TFx across replicates of the same specimen (repeatability) and duplicate samples in different batches (reproducibility). Comparison of samples collected in EDTA and Streck tubes from single venipuncture in 23 patients demonstrated that EDTA or Streck tubes were comparable if processed within 8 hours. On the basis of a range of DNA inputs (1 to 50 ng), 20 ng cfDNA is the preferred input, with 5 ng minimum acceptable. Overall, this shallow whole-genome sequencing of cfDNA and ichorCNA approach offers sensitive, precise, and reproducible quantitation of TFx, facilitating assay application in clinical cancer care.

5057 Background: Lutetium-177-PSMA vipivotide tetraxetan (Lu177) is approved for men with PSMA-positive metastatic castrate-resistant prostate cancer (mCRPC) previously treated with androgen receptor signaling inhibitors and taxane chemotherapy. The tumor genomic alterations conferring sensitivity or resistance to Lu177 are unknown. We investigated the association of genomic variants with clinical outcomes in men with mCRPC treated with Lu177. Methods: Somatic & germline DNA sequencing data were examined from men treated with ≥1 cycle Lu177 at The James Cancer Hospital of The Ohio State University from 3/2022 – 10/2023. Pathogenic variants from tissue or blood testing were identified. Sets of genes with similar biological function were annotated, including homologous recombination repair (HRR) genes ( BRCA1, BRCA2, ATM, CDK12, CHEK1, CHEK2, PALB2, RAD51B, RAD51C, RAD51D, RAD54L), established tumor suppressor genes ( TP53, RB1, PTEN) and cell cycle genes ( CCND1, CCND3, CDK4, CDK6, CDK12, CDKN1B, CDKN2B) associated with prostate cancer. The impact of genomic variants on PSA50 response (Chi-Square & Fisher’s exact test), median overall survival (OS) and median progression-free survival (PFS) (Kaplan-Meier method) and OS or PFS (Cox proportional hazards model) was assessed. Results: Of 153 patients treated with Lu177, 120 underwent genomic sequencing: 109 had somatic testing (76 tissue-based, 33 blood-based), 50 had germline and 39 had both somatic & germline testing. Patients with amplification of cell cycle genes (n=8) had poorer OS of 5.1 months (mos.) vs 13.4 mos. for those without amplification (p=0.001, HR 4.26, 95% CI 1.79-10.11), and shorter PFS of 4.3 vs 7.4 mos. (p=0.027, HR 2.40, 95% CI 1.10-5.22). Patients with loss or mutation of HRR genes (n=32) had OS of 13.7 vs 12.2 mos. for those with no HRR alterations, but this was not significant (p=0.101, HR 0.57, 95% CI 0.29-1.12). No other genomic alterations were significantly associated with OS or PFS and none were associated with PSA50 response. The table lists the most frequently altered genes in PSA50 responders vs non-responders (differences not significant by Fisher’s exact test). Conclusions: Genomic alterations warrant investigation in larger cohorts as prognostic and predictive biomarkers for Lu177 therapeutic response. Our data suggest that amplification of genes impacting cell cycle regulatory pathways may be of particular interest. Our findings support continued efforts to determine the impact of cyclin dependent kinase (CDK) 4/6 inhibitors for high-risk and metastatic prostate cancer, either alone or in combination with agents such as Lu177.[Table: see text]

Abstract Background: Categorizing breast cancer HER2/ERBB2 expression as “positive” or “negative” is no longer sufficient, with evidence that treatment response and outcomes are associated with HER2 “low” status and HER2 intratumoral heterogeneity. We hypothesized that interrogating HER2 heterogeneity (HER2het) across multiple spatial resolutions would more accurately capture HER2 diversity and be associated with clinical outcomes. Methods: We interrogated tumor cell and microenvironmental features by profiling 1,113,204 single cells in tissue sections from 171 HER2+/HER2low cancers via custom 25-marker high dimensional multiplexed immunofluorescence (HDmIF) using NeoGenomics MultiOmyx, with adjacent section HER2 immunohistochemistry (IHC). We developed novel metrics to concurrently: 1) interrogate HER2 heterogeneity at four spatial resolutions; 2) use machine learning to translate HER2 IF to IHC, termed ‘HAIQu’ (HER2 Automated Immunofluorescence Quantification), scoring HER2 IF expression according to ASCO/CAP guidelines; 3) delineate HER2 signaling phenotypes at the tumor cell level based on six HER2-related proteins; 4) evaluate immunophenotype of 23 immune cell types. We evaluated the association of these novel HER2het metrics with patient clinicopathologic features, recurrence-free survival (RFS), overall survival (OS), and diverse antibody markers representing tumor cell intrinsic processes and tumor-immune microenvironment (TME). Results: 1166 regions of interest were analyzed from 208 unique tumors profiled. Median follow-up from diagnosis was 143 months and 98.9% (n=183/185) received HER2-directed therapy in the (neo)adjuvant or metastatic setting. Our HAIQu scoring system effectively translated adjacent section HER2 IF to IHC with 97.9% concordance between HAIQu and clinical IHC scoring. Single-cell phenotypic analyses of 392,984 HER2+/PanCK+ tumor cells’ concurrent expression of six HER2-positive breast cancer related proteins (HER2, HER3, EGFR, pAKT, ER, KI67) using an unsupervised neural network-based self-organizing map approach resulted in 7 HER2 signaling cell phenotypes. Most patient samples are dominated by a single cell cluster but, intriguingly, Cluster 1 cells (EGFR-low) predominate in tumors with high HER2 cell membrane heterogeneity (ANOVA p=0.003). Evaluation of immunophenotype in hormone receptor-negative, HER2+ tumors, demonstrated significant association with immune cluster and recurrence-free survival (RFS; log-rank p=0.024) with zero RFS events among immune-high versus median survival of only 53.5mo among immune low-PDL1 low tumors. A multivariable Cox proportional hazards model including single cell HER2-heterogeneity (only significant metric on univariate), receptor subtype, and immunophenotype cluster demonstrated significant association with RFS (overall model log-rank p=0.005) and each significantly contributed to the model (all p< 0.05). Conclusions: We present novel metrics of HER2 heterogeneity via HDmIF, which offer detailed characterization of the diversity of HER2 expression in a large, clinically-annotated cohort with long-term follow-up. Identification of a strong association between immunophenotype and RFS supports further investigation of the highly immune activated subsets of ER-/HER2+ breast cancer. Strong correspondence of HER2 IF and IHC and our HAIQu methodology offers a pathway to translation of HER2het metrics to clinical practice. Table 1. Association of HER2 Heterogeneity Metrics with Recurrence-Free Survival Citation Format: David Tallman, Anna Juncker-Jensen, Harry Nunns, Kevin Gallagher, Heather LeFebvre, Karen Yamamoto, Katharine Collier, Mark Vater, Ava Strahan, Mathew Cherian, Ashley Pariser, Preeti Sudheendra, Bhuvaneswari Ramaswamy, Margaret Gatti-Mays, Ainura Kyshtoobayeva, Zaibo Li, Daniel Stover, Kai Johnson. Novel Metrics of HER2 Heterogeneity in HER2-Positive and HER2-Low Breast Cancer via High Dimensional Multiplexed Immunofluorescence Spatial Profiling [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-14-03.

Genitourinary (GU) malignancies are common and associated with significant morbidity and mortality. In patients with localized GU cancers, surgical resection or definitive radiation remain the mainstays of treatment. Despite definitive treatment, many patients with high-risk localized disease experience recurrence. There is growing interest in using neoadjuvant immunotherapy to improve outcomes. This narrative review summarizes the current evidence for neoadjuvant immunotherapy in patients with localized high-risk GU cancers including renal cell carcinoma, urothelial carcinoma, prostate cancer, penile squamous cell carcinoma, and testicular germ cell tumors. We also discuss ongoing clinical trials and candidate biomarkers to optimize patient selection and improve treatment outcomes.

e16626 Background: Studies cite that 10-24% of patients (pts) with urothelial carcinoma (UC) have a pathogenic germline mutation (mut), but few have looked at their association with clinical outcomes. We aimed to identify the frequency of germline pathogenic variants and variants of unknown significance (VUS) in all stages of UC, and hypothesized that they would be associated with cancer-related outcomes. Methods: We performed a retrospective review at our institution of all pts with a diagnosis of bladder, renal pelvis, ureter, or urethral cancer who underwent germline genetic testing from 2018 to 2024. Results: Sixty-four pts were included. The median number of genes tested was 77 (range 5-91) with 87.5% completing a panel of ≥67 genes. Of the 64 pts, 7 (11%) had a pathogenic germline mut, 18 (28%) had a VUS without a pathogenic mut, and 39 (61%) had no germline mut. Pathogenic muts were found in EPCAM, MAX, MSH2, MSH6, MUTYH, PMS2, and POT1. Among the 7 pts with a pathogenic mut, 5 (71%) had Lynch syndrome and 1 (14%) had a non-Lynch DNA Damage Repair gene mut in POT1. Sixty-two pts had urothelial histology, 1 pt had with rhabdomyosarcoma with a VUS in RET, and 1 pt had adenocarcinoma with a VUS in POLD1. Comparing pts with pathogenic muts (n=7) to those with no mut (n=39), we found no significant difference in median age at diagnosis (62 [range 46-76] vs 57 [range 20-86]), age at diagnosis ≥ 60 (4 [57%] vs 17 [44%]), sex (4 [57%] vs 13 [33%] female), race, ethnicity, smoking history (3 [43%] vs 15 [38%] never smoked), median number of 1st degree family members with a diagnosis of cancer (2 [range 1-6] vs 2 [0-4]), or clinical stage at diagnosis. However, there were significantly more pts with upper tract tumors in the group with a pathogenic mut compared to the group with no mut (6 [86%] vs 11 [28%], p=0.0072). Comparing the group with a VUS (n=18) to the group with no mut (n=39), there were no significant differences in demographic variables. However, variants among Blacks or African Americans are more likely to be classified as VUS (3/4 [75%] vs 14/59 [24%], p=0.02013). VUSes were also significantly overrepresented in pts with metastatic disease (9 [50%] vs 7 [18%], p=0.0242). Adjusting for clinical stage, pts with muscle-invasive UC and a pathogenic variant had significantly shorter overall survival (OS) compared to pts with no mut (log rank p= 0.034, median OS 43.2 months vs not reached). There was no significant difference in OS for non-muscle invasive or metastatic disease. Conclusions: Pathogenic germline variants were detected in 11% of pts with UC, were more common in upper tract disease, and were associated with shorter OS for muscle-invasive UC. VUS were found in rare histologic variants, which warrants further investigation, and were more common in metastatic disease and non-White pts, underscoring the need for studies in populations of diverse race and ethnicity. Our findings support genetic testing in all pts with UC. Validation in a larger cohort is ongoing.