Gm-Csf Improves Activity of Interferon as Primary Therapy for CML.

Not only does IFN prolong the survival of CML patients, but the 10-20% of patients who achieve a complete cytogenetic remission (CCyR) with IFN have a median survival >10 yrs and some of these patients may actually be cured. Imatinib mesylate (IM) has replaced IFN as front line therapy for newly diagnosed patients due to its favorable toxicity profile and superior initial response rate. Recent laboratory and clinical data suggest the impact of IM on CML stem cells may be limited, raising the question of durability of IM responses. Conversely, our data suggest that IFN is directly toxic to CML stem cells, at least in part through induction of CML stem cell terminal differentiation that also requires myeloid growth factors. IFN’s primary activity against CML stem cells rather than mature CML progenitors may explain the slow, but often durable, responses seen in IFN-treated patients, while the rapid responses induced by IM are likely to be a consequence of its impressive activity against mature CML progenitors. We report results of our clinical trial adding GM-CSF to IFN as upfront therapy to improve cytogenetic responses over IFN alone in patients with newly diagnosed CML. Patients received IFN with a goal of 5 x 10⁶ units/m²/day and dose reductions for cytopenias. Upon achieving a hematologic response, pts started GM-CSF at 125μg/m²/day for 6 months. GM-CSF was initially dose-reduced for WBC > 10,000/μL; this parameter was changed to 20,000/μL after 2 of the first 3 pts were removed from study because of persistently elevated WBC despite 2 dose reductions of GM-CSF. No further pts have been removed from the study due to increased WBC. Response was monitored every 3 months by peripheral blood FISH for bcr-abl and every 6 months by bone marrow cytogenetics and RT-PCR. A total of 58 pts were enrolled on the trial with 51 considered evaluable for response (2 unevaluable due to persistently elevated WBC and 5 pts withdrew consent before the initial 3 month evaluation) and all 58 pts for toxicity. The combination of IFN and GM-CSF was well tolerated and 12/58 (20%) of enrolled pts withdrew consent due to treatment-related toxicity. However, many pts withdrew consent prior to achieving landmark responses, including 15/58 before achieving a MCR and 15/58 prior to a CCyR. Unfortunately, these groups received a historically short course of IFN-based therapy (median 14, range 3–45, mos) in part due to the availability of IM during this trial. Interestingly, 3 pts withdrew consent despite achieving a CCyR on the combination. Only 2 pts came off study due to progression (>35% Ph+). 35 (69%) of 51 evaluable patients obtained a MCR (Ph+ < 35%) at a median of 3 (range 3 – 15) months with over 1/3 of them achieving a CCyR at a median of 9 (range 3–21) months. An additional 13 pts in a MCR on IFN + GM-CSF achieved a CCyR after switching to IM because of patient preference. Thus, 51% (26/51) of evaluable patients on this trial are currently in a CCyR at a median follow-up of 4.3 (range 3–6) years. IFN + GM-CSF appears to be more active in CML than IFN alone, inducing a MCR in nearly 70% of patients. The incidence of CCyR with this combination is difficult to determine because many of the patients switched to IM while still responding to IFN. Because IFN + GM-CSF appears to target CML stem cells, while IM may not, this combination should be considered as part of the treatment algorithm for CML.