Factors predictive of survival among 337 patients with mesothelioma treated between 1984 and 1994 by the Cancer and Leukemia Group B

Purpose: To examine the individual and joint effect of various pretreatment clinical characteristics on the survival of patients with mesothelioma treated by the Cancer and Leukemia Group B (CALGB). Patients and methods: Between June 1984 and September 1994, 337 patients with malignant mesothelioma and no prior chemotherapy were accrued to seven phase II studies conducted by the CALGB which screened the efficacy of 10 treatment regimens or dose levels. The eligibility criteria for all studies were virtually identical. Patient characteristics include the following: age older than 60 years (63%); male (83%); performance status (PS) of 0 or 1 (81%); chest pain (60%); definite asbestos exposure (62%); >5% weight loss (41%); and pleural involvement (94%). Median survival time (MST) for the 10 treatment regimens ranged from 3.9 to 9.8 months (overall=7.2; 95% confidence interval [CI], 6.5 to 8.3), with I-year survival between 14% and 50% (overall=27%; 95% CI, 23 to 33%). Results: Cox survival models and exponential regression trees were used to examine the prognostic importance of pretreatment patient characteristics. Univariate analyses show that patients with poor Eastern Cooperative Oncology Group PS, chest pain, dyspnea, platelet count (PLT) >400,000/mu L, weight loss, serum lactate dehydrogenase (LDH) level >500 IU/L, pleural involvement, low hemoglobin (HGB) level, high WBC count, and increasing age over 75 years have a worse prognosis. With decreasing risk ratio, multivariate Cox analyses showed that pleural involvement, LDH >500 IU/L, Door PS, chest pain, PLT >400,000/mu L, nonepithelial histology, and increasing age older than 75 years jointly predict poor survival. PS was the most important prognostic split in the regression tree. Terminal nodes were amalgamated to form six distinct prognostic subgroups with MST (2-year survival) of 13.9 (38%) in 36 patients, 9.5 (21%) in 36 patients, 9.2 (10%) in 146 patients, 6.5 (3%) in 33 patients, 4.4 (0%) in 73 patients, and 1.4 (0%) in 13 patients (p<0.0001). Conclusions: The subgroup with the best survival (MST=13.9 months) included patients with PS=0 and age younger than 49 years, and patients with PS=0, age of 49 years or older, and HGB greater than or equal to 14.6. The worst survival (MST=1.4 months) occurred for patients with PS=1/2, and WBC greater than or equal to 15.6/mu L.