CXCL9:SPP1 ratio: Macrophage polarization and outcomes with pembrolizumab or enfortumab vedotin plus pembrolizumab in metastatic urothelial cancer.

Abstract

4573 Background: Enfortumab vedotin plus pembrolizumab (EV+P) is standard first-line treatment for metastatic urothelial cancer (mUC), yet responses vary considerably. Identifying resistance mechanisms may guide patient selection and inform novel therapeutic strategies. Single-cell RNA sequencing (RNA-Seq) across tumor types has revealed dichotomous tumor-associated macrophage (TAM) populations: SPP1+ TAMs exhibit tumor-promoting programs while CXCL9+ TAMs express programs associated with anti-tumor immunity. The CXCL9:SPP1 gene expression ratio (CS ratio) from bulk RNA-Seq captures this balance and correlates with outcomes across solid tumors (Bill et al, Science, 2023). We explored the CS ratio's association with outcomes in mUC patients receiving pembrolizumab monotherapy or EV+P. Methods: We retrospectively analyzed mUC patients who underwent molecular profiling at Caris Life Sciences using NGS (592-gene panel, DNA WES, or RNA WTS). CS ratios were calculated and patients stratified into quartiles (Q1–Q4; Q4=highest ratio). PD-L1 expression was assessed by IHC (22c3, positive (+) CPS ≥ 10%) while TMB-High was defined as ≥ 10 Mut/Mb. TME composition was inferred using quanTIseq. Real-world overall survival (rwOS) was derived from insurance claims and calculated from biopsy to last contact or time on treatment (TOT) using Kaplan-Meier analysis while Hazard ratio (HR) was calculated by Cox proportional hazard method. Mann-Whitney U and χ²/Fisher exact tests were applied with adjustment for multiple comparisons. Results: Among 6,708 mUC patients, 1,299 received pembrolizumab and 561 received EV+P. Lower CS ratios (SPP1+ TAM predominance) were associated with shorter time on treatment and rwOS (Table 1). Low CS ratio was also associated with reduced PD-L1 expression (15.5% vs 66.7%, p < 0.001) and TMB-H prevalence (34.6% vs 52.3%, p = 0.006). Low CS ratio tumors demonstrated reduced immune infiltration except for neutrophil enrichment. Conclusions: Lower CS ratios associate with shorter time on treatment and rwOS with both pembrolizumab and EV+P. To our knowledge, this is the first analysis linking macrophage transcriptional states with EV+P resistance. Further validation of the CS ratio in patients treated with EV + P or P is warranted. CS ratio and associated outcomes with pembrolizumab monotherapy and EV+P. Pembrolizumab Monotherapy EV + P CS Ratio TOT (months) OS (months) TOT (months) OS (months) CS-Q1 2.7 (2.1-3.0) 5.7 (4.4-8.2) 7.6 (6.2-8.8) 17.8 (13.8-19.8) CS-Q2 3.1 (2.2-4.1) 7.2 (5.4-8.8) 6.9 (6.0-8.0) 16.4 (13.2-19.4) CS-Q3 4.5 (3.9-5.6) 13.5 (10.7-16.8) 9.2 (8.1-10.8) 17.2 (14.4-20.3) CS-Q4 6.9 (4.9-8.7) 18.7 (13.1-23.7) 10.6 (9.3-12.2) 21.8 (16.5-28.1) CS-Q4 vs. CS-Q1 HR 0.56, 95% CI 0.47-0.67,p < 0.001 HR 0.53, 95% CI 0.44-0.64,p < 0.001 HR 0.69, 95% CI 0.54-0.87, p = 0.00189 HR 0.62, 95% CI 0.46-0.83, p = 0.00148