Teacher-student collaborated multiple instance learning for pan-cancer PDL1 expression prediction from histopathology slides.

Programmed cell death ligand 1 (PDL1), as an important biomarker, is quantified by immunohistochemistry (IHC) with few established histopathological patterns. Deep learning aids in histopathological assessment, yet heterogeneity and lacking spatially resolved annotations challenge precise analysis. Here, we present a weakly supervised learning approach using bulk RNA sequencing for PDL1 expression prediction from hematoxylin and eosin (H&E) slides. Our method extends the multiple instance learning paradigm with the teacher-student framework, which assigns dynamic pseudo-labels for intra-slide heterogeneity and retrieves unlabeled instances using temporal ensemble model distillation. The approach, evaluated on 12,299 slides across 20 solid tumor types, achieves a weighted average area under the curve of 0.83 on fresh-frozen and 0.74 on formalin-fixed specimens for 9 tumors with PDL1 as an established biomarker. Our method predicts PDL1 expression patterns, validated by IHC on 20 slides, offering insights into histologies relevant to PDL1. This demonstrates the potential of deep learning in identifying diverse histological patterns for molecular changes from H&E images.

Real-world outcomes in locally advanced or metastatic urothelial carcinoma following platinum and PD-1/L1 inhibitor therapy.

Aim: To investigate real-world overall survival (rwOS) and real-world progression-free survival (rwPFS) in locally advanced/metastatic urothelial carcinoma postplatinum and postprogrammed death receptor-1/death ligand 1 inhibitors. Patients & methods: Adult patients diagnosed with locally advanced/metastatic urothelial carcinoma from 1 January 2011 to 31 December 2018 and treated with taxane monotherapy or any therapy postplatinum and post-PD-1/L1 inhibitors were included from a nationwide electronic health record-derived oncology database. Results: Median rwOS among 72 patients treated with taxane monotherapy was 7.6 months (95% CI: 5.2-14.4) and rwPFS was 2.9 months (95% CI: 2.4-4.0). Among 208 patients treated with any therapy, median rwOS was 8.9 months (95% CI: 7.3-10.6) and rwPFS was 3.6 months (95% CI: 2.7-4.7). Conclusion: Short duration of rwOS and rwPFS were observed, highlighting the need for effective and safe treatments in this patient population.

Lay abstract Few studies have evaluated survival outcomes in patients with advanced urothelial cancer who have disease relapse after chemotherapy and PD-1/L1 inhibitor therapy in clinical practice. In this study, we used electronic health records from a nationwide cancer database to assess survival in adult patients who received further treatment in this setting from 2011 to 2018. Among 72 patients who were treated with taxane monotherapy after chemotherapy and a PD-1/L1 inhibitor, average overall survival was 7.6 months and progression-free survival was 2.9 months. Among 208 patients who were treated with any therapy, average overall survival was 8.9 months and progression-free survival was 3.6 months. These results highlight the need for safer and more effective therapies in patients with advanced urothelial cancer who have disease relapse after chemotherapy and PD-1/L1 inhibitor therapy.

Enfortumab vedotin after PD-1 or PD-L1 inhibitors in cisplatin-ineligible patients with advanced urothelial carcinoma (EV­201): a multicentre, single-arm, phase 2 trial.

BACKGROUND: Locally advanced or metastatic urothelial carcinoma is generally incurable and has scarce treatment options, especially for cisplatin-ineligible patients previously treated with PD-1 or PD-L1 therapy. Enfortumab vedotin is an antibody-drug conjugate directed at Nectin-4, a protein highly expressed in urothelial carcinoma. We aimed to evaluate the efficacy and safety of enfortumab vedotin in the post-immunotherapy setting in cisplatin-ineligible patients. METHODS: EV-201 is a multicentre, single-arm, phase 2 study of enfortumab vedotin in patients with locally advanced or metastatic urothelial carcinoma previously treated with PD-1 or PD-L1 inhibitors. Cohort 2 included adults (aged ≥18 years) with an Eastern Cooperative Oncology Group performance status score of 2 or less who were considered ineligible for cisplatin at enrolment and who had not received platinum-containing chemotherapy in the locally advanced or metastatic setting. Enfortumab vedotin was given intravenously at a dose of 1·25 mg/kg on days 1, 8, and 15 of every 28-day cycle. The primary endpoint was confirmed objective response rate per Response Evaluation Criteria in Solid Tumours version 1.1 assessed by blinded independent central review. Efficacy and safety were analysed in all patients who received at least one dose of enfortumab vedotin. EV-201 is an ongoing study and the primary analysis is complete. This study is registered with Clinicaltrials.gov, NCT03219333. FINDINGS: Between Oct 8, 2017, and Feb 11, 2020, 91 patients were enrolled at 40 sites globally, of whom 89 received treatment. Median follow-up was 13·4 months (IQR 11·3-18·9). At data cutoff (Sept 8, 2020), the confirmed objective response rate was 52% (46 of 89 patients; 95% CI 41-62), with 18 (20%) of 89 patients achieving a complete response and 28 (31%) achieving a partial response. 49 (55%) of 89 patients had grade 3 or worse treatment-related adverse events. The most common grade 3 or 4 treatment-related adverse events were neutropenia (eight [9%] patients), maculopapular rash (seven [8%] patients), and fatigue (six [7%] patients). Treatment-related serious adverse events occurred in 15 (17%) patients. Three (3%) patients died due to acute kidney injury, metabolic acidosis, and multiple organ dysfunction syndrome (one [1%] each) within 30 days of first dose and these deaths were considered by the investigator to be related to treatment; a fourth death from pneumonitis occurred more than 30 days after the last dose and was also considered to be related to treatment. INTERPRETATION: Treatment with enfortumab vedotin was tolerable and confirmed responses were seen in 52% of cisplatin-ineligible patients with locally advanced or metastatic urothelial carcinoma who were previously treated with PD-1 or PD-L1 inhibitors. These patients have few treatment options, and enfortumab vedotin could be a promising new therapy for a patient population with a high unmet need. FUNDING: Astellas Pharma Global Development and Seagen.

Pivotal Trial of Enfortumab Vedotin in Urothelial Carcinoma After Platinum and Anti-Programmed Death 1/Programmed Death Ligand 1 Therapy.

PURPOSE: Locally advanced or metastatic urothelial carcinoma is an incurable disease with limited treatment options, especially for patients who were previously treated with platinum and anti-programmed death 1 or anti-programmed death ligand 1 (PD-1/L1) therapy. Enfortumab vedotin is an antibody-drug conjugate that targets Nectin-4, which is highly expressed in urothelial carcinoma. METHODS: EV-201 is a global, phase II, single-arm study of enfortumab vedotin 1.25 mg/kg (intravenously on days 1, 8, and 15 of every 28-day cycle) in patients with locally advanced or metastatic urothelial carcinoma who were previously treated with platinum chemotherapy and anti-PD-1/L1 therapy. The primary end point was objective response rate per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 by blinded independent central review. Key secondary end points were duration of response, progression-free survival, overall survival, safety, and tolerability. RESULTS: Enfortumab vedotin was administered to 125 patients with metastatic urothelial carcinoma. Median follow-up was 10.2 months (range, 0.5 to 16.5 months). Confirmed objective response rate was 44% (95% CI, 35.1% to 53.2%), including 12% complete responses. Similar responses were observed in prespecified subgroups, such as those patients with liver metastases and those with no response to prior anti-PD-1/L1 therapy. Median duration of response was 7.6 months (range, 0.95 to 11.30+ months). The most common treatment-related adverse events were fatigue (50%), any peripheral neuropathy (50%), alopecia (49%), any rash (48%), decreased appetite (44%), and dysgeusia (40%). No single treatment-related adverse events grade 3 or greater occurred in 10% or more of patients. CONCLUSION: Enfortumab vedotin demonstrated a clinically meaningful response rate with a manageable and tolerable safety profile in patients with locally advanced or metastatic urothelial carcinoma who were previously treated with platinum and anti-PD-1/L1 therapies.