Interleukin-2 (IL-2) targets cells bearing IL-2 receptors and induces different degrees of lymphocytosis. This study retrospectively evaluated whether lymphocytosis, in addition to clinical characteristics at baseline and to tumor objective response, may predict overall survival in metastatic renal cell carcinoma patients who received IL-2 subcutaneously (s.c.). Overall survival, clinical characteristics, tumor response, and total lymphocyte count at baseline and during the first treatment cycle of 266 advanced renal cell cancer patients, treated with 1 of 4 different first-line s.c. IL-2-based protocols, were studied using the Cox multivariate analysis. Median IL-2 cumulative dose and length of treatment (+/-SD) were 232 +/- 282 x 10(6)/m(2) in 7 +/- 5.9 weeks, respectively. Median overall survival (os) was 13.1 months (range 0.7-86.9+) in all. Tumor outcome consisted of: 9 CR (3%) (os = NR); 35 PR (13%) (os = 19.7 months.); 117 SD (44%) (os = 15.1 months); 105 PD (39%) (os = 6.4 months). Median lymphocyte counts were 1400/mm(3) at baseline (25th-75th, 900-1900/mm(3)) and 3600/mm(3) as a maximum value (25th-75th, 2600-4800/mm(3)). Death risk significantly decreased by 11% for each 1,000 lymphocytes/mm(3) (RR 0.89; 95% CI 0.82-0.97), after correcting for clinical characteristics (PS ECOG 0 versus > or =1, time from primary diagnosis > or =2 years versus <2 years, number of metastatic sites 1 versus >1) and tumor response (CR, PR). A two-step bootstrapping procedure confirmed such predictive performance. Lymphocyte count monitoring represents a biomarker of the host response to subcutaneous IL-2 treatment useful for multimodal clinical assessment, as it predicts overall survival in advanced cancer patients independently from tumor response and from main clinical characteristics.