ALTERED LEVELS OF INTERLEUKIN-1-BETA AND INTERLEUKIN-1 RECEPTOR ANTAGONIST IN CHRONIC MYELOGENOUS LEUKEMIA - CLINICAL AND PROGNOSTIC CORRELATES

We have recently demonstrated that interleukin (IL)-1 beta levels are elevated in advanced chronic myelogenous leukemia (CML) and that IL-1 inhibitors can suppress CML clonogenic growth. To further assess the c linical Implications of increased IL-1 beta expression in CML, we anal yzed IL-1 beta and IL-1 receptor antagonist (IL-1RA) levels in leukocy te lysates from a series of CML patients and from normal volunteers. B oth IL-1 beta and IL-1RA were measured by enzyme-linked immunosorbent assays (ELISAs), with the lower limits of sensitivity of the assays be ing 20 pg/mL and 6.5 pg/mL, respectively. The median IL-1 beta level i n the 81 CML patients tested was higher (115.8 pg/2.4 x 10(7) cells; r ange, 0 to 2,000 pg/2.4 x 10(7) cells) than the median level in 25 con trol samples (10.8 pg/2.4 x 10(7) cells; range, 0 to 95.5 pg/2.4 x 10( 7) cells) (P <.01). IL-1 beta was bioactive, as demonstrated with a bi oassay based on cytotoxicity to a melanoma cell line (A375). For survi val analysis, elevated IL-1 beta levels were defined as those exceedin g the mean + 2 SD of normal levels (83 pg/2.4 x 10(7) cells). The surv ival of the 44 patients with elevated IL-1 beta levels was significant ly shorter than that of those who had low IL-1 beta levels (median, 44 v 58 months; P =.049 by Wilcoxon-Gehan method). An association betwee n IL-1 beta and CML prognostic criteria shows that IL-1 beta levels we re significantly higher in patients in accelerated/blastic crisis phas es of the disease (364.0 pg/2.4 x 10(7) cells) compared with patients in chronic phase (102.0 pg/2.4 x 10(7) cells) (P <.01), and that high IL-1 beta levels correlated with increased blasts in the marrow and pe ripheral blood (P <.01). In contrast, while IL-1RA levels did not diff er between chronicphase CML patients (median, 471.7 pg/2.4 x 10(5)) an d healthy volunteers (median, 454.4 pg/2.4 x 10(5)), patients with acc elerated/blast crisis disease had significantly lower levels of IL-1RA (median. 218.7 pg/2.4 x 10(5); P =.03) Finally, although IL-1 beta ha s been previously shown to increase IL-1RA levels, there was no correl ation between IL-1 beta and IL-1RA levels in our CML patients. In summ ary, our results suggest that (1) both IL-1 beta and IL-1RA may be dys regulated in CML; (2) high IL-1 beta and low IL-1RA protein levels are associated with advanced disease; and (3) IL-1 beta and IL-1RA are in dependently altered in CML. (C) 1994 by The American Society of Hemato logy.