cepia

Clinical Epidemiology and Ageing

Early chemotherapy discontinuation and mortality in older patients with metastatic bladder cancer: The AGEVIM multicenter cohort study.

Laurent M, Brureau L, Demery MEl, Fléchon A, Le Thuaut A, Carvahlo-Verlinde M, Bastuji-Garin S, Paillaud E, Canoui-Poitrine F, Culine S Urol Oncol. 2017;35(1):34.e9-34.e16.

<p><b>PURPOSE: </b>Median age for the diagnosis of metastatic bladder cancer (MBC) is 73 years. The feasibility of chemotherapy in older patients is controversial. Our objectives were to assess associations linking age to first line chemotherapy regimen selection, early chemotherapy discontinuation, and 1-year mortality in everyday practice.</p><p><b>MATERIALS AND METHODS: </b>Between 1999 and 2011, 197 consecutive patients aged≥70 years with MBC referred to 4 hospitals were included in the AGEVIM multicenter cohort. At baseline, we recorded performance status (PS); tumor characteristics; the Charlson Comorbidity Index; and plasma creatinine, hemoglobin, and albumin. Early discontinuation data were available for 193 patients, and overall 1-year mortality for 180 patients. We assessed the probabilities of initial cisplatin-based combination chemotherapy (CCC), early discontinuation (≤2 cycles), and 1-year mortality, using multivariate logistic regression and Cox proportional hazards modeling.</p><p><b>RESULTS: </b>Among the 193 patients (mean age: 76±4.3y), with 2 metastatic site in median 43.5% received CCC, 36.3% gemcitabine and carboplatin, and 20.2% gemcitabine alone. The probability of CCC decreased with age independently from sex, PS, creatinine clearance, and Charlson Comorbidity Index (P<0.0001), early discontinuation occurred in 24.9% of patients. Factors independently associated with global chemotherapy early discontinuation were age (adjusted odds ratio = 1.11; 95% CI: 1.02-1.20; P = 0.01) and higher metastatic-site number (adjusted odds ratio = 1.45; 95% CI: 1.08-1.95; P = 0.01). The number of patients was too small for a robust analysis of factors associated with early chemotherapy discontinuation in each chemotherapy regiment subgroup. Independent predictors of 1-year mortality (median = 9.6 mo) were early discontinuation (adjusted hazard ratio [aHR] = 4.77 [2.85-7.96] when PS<2 and 20.6 [9.43-44.82] when PS≥2; P<0.0001), albumin<35g/l (aHR = 3.06 [1.81-5.17], P = 0.0001), creatinine clearance<30ml/min (aHR = 2.96 [1.45-6.06], P = 0.009), and higher metastatic-site number (aHR = 1.34 [1.14-1.56], P<0.0001).</p><p><b>CONCLUSION: </b>Less than half of older patients with MBC received initial CCC and 25% had≤2 cycles of chemotherapy. Older age was associated with decreased CCC prescription, independently from known contraindications, and with global chemotherapy early discontinuation, but not with 1-year mortality.</p>

MeSH terms: Age Factors; Aged; Aged, 80 and over; Antineoplastic Combined Chemotherapy Protocols; Carboplatin; Cisplatin; Creatinine; Deoxycytidine; Female; Gemcitabine; Health Status; Humans; Male; Neoplasm Metastasis; Prospective Studies; Serum Albumin; Survival Rate; Time Factors; Tumor Burden; Urinary Bladder Neoplasms; Withholding Treatment
DOI: 10.1016/j.urolonc.2016.08.003