The objective of this study was to examine the correlation between serum interleukin-6 (IL-6) and IL-10 levels and outcome in chronic lymphocytic leukemia (CLL). Serum IL-6 and IL-10 levels were measured by enzyme-linked immunoabsorbent assays from 159 and 151 CLL patients, respectively, and from healthy control subjects (n = 55 [IL-6]; n = 37 [IL-10]). Cytokine levels were correlated with clinical features and survival. Serum IL-6 levels were higher in CLL patients (median, 1.45 pg/mL; range, undetectable to 110 pg/mL) than in control subjects (median, undetectable; range, undetectable to 4. 30 pg/mL) (P <.0001). Serum IL-10 levels were higher in CLL patients (median, 5.04 pg/mL; range, undetectable to 74 pg/mL) than in normal volunteers (median, undetectable; range, undetectable to 13.68 pg/mL) (P <.00001). Assays measuring both Epstein-Barr virus-derived and human IL-10 yielded higher values than assays measuring primarily human IL-10 (P <.05). Patients with elevation of serum IL-6 or IL-10 levels, or both, had worse median and 3-year survival (log rank P <.001) and unfavorable characteristics (prior treatment, elevated beta(2)-microglobulin or lactate dehydrogenase, or Rai stage III or IV). Elevated IL-6 and IL-10 levels were independent prognostic factors for survival when analyzed individually or in combination (Cox regression analysis). However, if beta(2)-microglobulin was incorporated into the analysis, it was selected as an independent prognostic feature, and IL-6/IL-10 were no longer selected. In patients with CLL, serum IL-6 and IL-10 (viral and human) levels are elevated and correlate with adverse disease features and short survival. In multivariate analysis, however, beta(2)-microglobulin is the most important prognostic factor.