However, EV-related neuropathy is common and can be debilitating, leading to high rates of dose reduction and treatment discontinuation; for some patients, EV-related neuropathy may even be irreversible. For patients whose comorbidities may put them at increased risk of treatment-related toxicity, some clinicians elect for upfront EV dose reduction even prior to the onset of toxicity. To date, there has been limited evidence regarding the impact of upfront EV dose reduction on treatment tolerability or efficacy.
In this context, our study published in JAMA Oncology provides some of the first real-world data examining whether upfront EV dose reduction affects treatment continuity and overall survival in patients with advanced urothelial carcinoma.
Using the Flatiron Health electronic health record database, we identified nearly 500 patients with advanced urothelial cancer who initiated first-line EV+P. Approximately 1 in 4 patients were initiated on a dose-reduced EV (defined as starting at 1.0 or 0.75 mg/kg, compared to the standard dose of 1.25 mg/kg).
Our primary endpoint was time to first treatment interruption (≥35-day gap between EV doses), a pragmatic real-world signal of treatment tolerability, while our secondary endpoint was overall survival. To address confounding factors, we used inverse probability of treatment weighting.
Two findings stood out:
- Upfront EV dose reduction cuts the risk of treatment interruption in half.
Patients who started at a reduced dose had a 51% lower likelihood of EV interruption (aHR = 0.49, significant). This effect was conserved (aHR = 0.49, significant) in the subgroup of older and/or physiologically vulnerable patients (defined as ≥80 years, CrCl <30 mL/min, or ECOG ≥2). - No significant difference in overall survival with upfront EV dose reduction.
Upfront EV dose reduction did not appear to negatively affect survival (aHR 1.24; not significant). Results were similar in the subgroup of older and/or physiologically impaired patients (aHR = 1.15; not significant).
As with all retrospective analyses, residual confounding remains a possibility. Our dataset lacked details on metastatic disease sites and burden, two variables that would likely impact survival. Additionally, follow-up time was relatively short given the regimen’s recent approval. Prospective data are needed to validate the findings of this study.
Written by: Henry K. Litt, MD, Ronac Mamtani, MD, MSCE, and Ryan D. Chow, MD
- Division of Hematology and Oncology, Department of Medicine, University of Pennyslvania, Philadelphia, PA.