Background: Platinum-based chemotherapy followed by second-line immunotherapy has been traditionally the standard of care for locally advanced or metastatic urothelial cancer (La/mUC). However, checkpoint inhibitors are remarkably reshaping the therapeutic landscape for urothelial cancer. Despite these advances, La/mUC remains a highly lethal disease with poor prognosis and limited therapeutic response. Moreover, discrepancies between clinical trial populations and real-world patients often result in different outcomes and toxicity profiles. In this context, oncology networks have become critical in generating Real-World Data (RWD), which can be translated into Real-World Evidence (RWE) to support clinical decision-making in everyday practice. We retrospectively analysed La/mUC patients treated with second-line pembrolizumab within the Campania Oncological Network (Rete Oncologica Campana, ROC).
Patients and methods: This multicenter retrospective study included adult patients (≥ 18 years) with histologically or cytologically confirmed La/mUC, who had previously received chemotherapy and were treated with Pembrolizumab (200 mg every three weeks) across six ROC centres. Primary endpoints were progression-free survival (PFS) and overall survival (OS). Secondary endpoints included objective response rate (ORR), disease control rate (DCR) and safety.
Results: Between January 2021 and November 2023, 132 patients with La/mUC received at least one Pembrolizumab dose. The median age was 67 years (range 30-88), and most were male (73.5%). The majority had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 (20.5%) or 1 (62.1%). Histologically, 115 patients (87.1%) had pure urothelial carcinoma, while 17 (12.9%) had rare urothelial carcinoma subtypes, including squamous, sarcomatoid/carcinosarcoma, micropapillary or nested variants. At the time of initial diagnosis, 45 (34.1%) patients presented with metastatic disease (stage IV), and 72 (54.5%) had undergone radical cystectomy. Metastatic sites included lymph nodes 98 (74.2%), lung 51 (38.6%), bones 45 (34.1%), and liver 21 (15.9%). Liver metastases were more frequent in pure urothelial carcinoma (17.4% vs. 5.9%), while non-urothelial variants showed a higher incidence of peritoneal carcinomatosis (23.5% vs. 1.7%). After a median follow-up of 20 months (95% CI 12.7-32.6), 108 patients (81.8%) experienced disease progression or death. Median PFS was 3.75 months (95% CI: 3.4 to 4.7), while median OS was 7.3 months (95% CI: 6.05-9.33). The ORR was 13.5% (95% CI: 7.4% - 19.7%), DCR was 33.9% (95% CI: 25.6%-42.0%) (driven by a 20.4% stable disease rate). Among 17 patients with rare subtypes of urothelial carcinoma, the ORR was 23.5% (95% CI: 3.3% to 43.7%). Multivariate analysis revealed that metastatic disease at diagnosis (HR = 2.01, p = 0.02) and liver metastases (HR = 2.11, p = 0.02) were independently associated with increased mortality risk (Table 3B). Lymph node-only metastases had a significantly lower risk of disease progression (HR = 0.46, p = 0.004) and death (HR = 0.52, p = 0.02). (). Univariate analysis indicated that prior cystectomy was associated with improved PFS (HR = 0.67, p = 0.04) and OS (HR = 0.54, p = 0.003), suggesting its role as a favourable prognostic factor in this cohort [1] (Table 2 A and 2B). A total of 114 treatment-related AEs were reported, with 79.8% being grade 1-2 and 20.2% grade 3-4. The most common events were asthenia (20.0%), pruritus (10.0%), and myalgia (7.0%). No treatment-related deaths occurred. Pembrolizumab was discontinued in 7.6% of patients due to AEs. The safety profile was manageable; 7.6% of patients discontinued treatment due to AEs (predominantly drug-induced pneumonitis), and 1.5% required permanent corticosteroid replacement for immune-related adrenal insufficiency.
Conclusions: This real-world analysis supports the clinical applicability of second-line Pembrolizumab in La/mUC, demonstrating efficacy and acceptable safety. Prognostic indicators, including metastatic sites and prior cystectomy, significantly impacted outcomes. These findings highlight the value of oncology networks in generating real-world evidence to refine treatment approaches.
© 2026. The Author(s).