The maintenance of busulphan-induced remissions in chronic granulocytic leukaemia with recombinant interferon alpha-2b.

Interferon (IFN) shows no specificity in inhibiting the growth of colonies of myeloid leukaemia blasts in culture as compared to normal haemopoietic precursors, but does reduce the self-renewal capacity of myeloblasts. We have tested the ability of IFN to slow the leukocyte doubling time (Ldt) and to prolong remissions induced by busulphan in 14 patients with chronic granulocytic leukaemia (CGL). Patients served as their own controls; the Ldt during relapse from a busulphan-induced remission on no therapy was determined and compared with the Ldt on IFN maintenance therapy. The initial dose of IFN (2 x 10(6) U M-2 subcutaneously, three times per week) was adjusted up, or down, to prevent the leukocyte count from rising and the platelet count from falling below 75 x 10(9) l-1. The dose of IFN required to prevent relapse in seven patients ranged from 1 x 10(6) U M-2 three times per week to 5.2 x 10(6) U M-2 daily, with a median of 2 x 10(6) U M-2 three times per week. IFN maintenance therapy has prevented relapse in six patients for more than 22 months to more than 68 months. In five patients the Ldt was slowed initially but the disease later progressed in four patients to enter the accelerated (three patients) or blast phase (one patient). The Ldt during IFN therapy did not change from the Ldt on no therapy in one patient; this patient later progressed to the blast phase. In two additional patients the leukaemia progressed during the first course of IFN, with shortening of the Ldt; both of these patients entered the blast phase. In the four patients who have discontinued IFN following relapse in the chronic phase, the Ldt remained prolonged for at least one relapse after the IFN was stopped. IFN maintenance therapy failed to control the leukocyte count in the six patients with a control Ldt of less than 40 days and five of these have progressed to enter the accelerated or blast phase. The early survival of this group of patients resembles the survival of 'good risk' CGL patients reported by others. We conclude that IFN maintenance therapy does alter the relapse pattern of a subset of CGL patients, either slowing the Ldt or preventing relapse.