Toxicities and outcome in elderly patients treated with targeted therapies for metastatic renal cell carcinoma

Abstract

with a median age of diagnosis of 65 years. Data from randomized controlled trials suggest similar choice of chemotherapy drugs, toxicities and feasibility in young and older patients. There was a lack of observational data from real-life setting. Objectives: To assess 1) associated factors of prescription of the combination of gemcitabine and cisplatin (GC), 2) chemotherapy feasibility in older patients aged 70 years or more with metastatic bladder cancer and 3) prognostic value of feasibility of chemotherapy. Methods: Between1999and2011, consecutive patients aged70 years ormorewithmetastatic bladder cancer referred to four teaching hospitals in France (Creteil, Lyon, Montpellier and Villejuif) were included. At baseline, we recorded clinical and biological parameters (nutrition, functional status assessed by Performance Status PS, creatinine clearance, hemoglobin level) and tumor characteristics (histologic subtype, grade and number of metastases). Non-feasibility was defined by a number of cycles less than 3. We used a propensity score method to assess probability of initial prescription of GC. Multivariate logistic regression and Cox Proportional Hazard Model were used to estimate predictors of chemotherapy non-feasibility and overall 1-year mortality. Results: 193 patients were included (mean age 76 ± 4.3 years, 85% males, 92.2% urothelial carcinoma, 96.3% grade 3) with two metastases in median [1–3]. Among them 25.3% had PS of ≥2, 42.1% had poor nutritional status and 43% had renal dysfunction. For first cycle prescription, 43.5% had GC, 36.3% had a combination of gemcitabin and carboplatin and 20.2% had gemcitabin alone. Probability of GC prescription decreased with age independent of functional status and creatinine clearance, respectively 0.59 for [70– 75] years, 0.32 for [75–80] years and 0.13 for ≥70 years (p b 0.0001). In multivariate analysis, factors associated with chemotherapy nonfeasibility were age (adjusted odds ratio (aOR)1 year increase = 1.11; CI 95% = 1.02–1.20; p = 0.01) and PS ≥2 (aOR = 2.10; 1.36–6.06; p = 0.005). Median survival was 9.6 months [5.2–15.4]. In a multivariate analysis, factors associated with overall 1-year mortality were chemotherapy non-feasibility (adjusted hazard ratio (aHR) = 4.52; 2.60–7.85 for patients with PS b2 and aHR= 24.87; 11.40–54.54 for patients with PS ≥2; p b 0,0001), age (aHR= 1.06; 1.01–1.13; p= 0.02), number of metastases (aHR= 1.2; 1.01–1.42; p= 0.03) and hemoglobin decrease (aHR1 point decrease = 1.19; 1.03–1.37; p= 0.01). Conclusion: The prescription of a combination of gemcitabine and cisplatin decreased with age independent of functional status and creatinine clearance in patients aged 70 years or more with metastatic bladder cancer. Increasing age was an independent predictor of chemotherapy non-feasibility. The association between non-feasibility and overall 1-year mortalitywasmodified by baseline functional status. Disclosure of interest: None declared.