Evaluating Therapeutic Strategies in MDS

点击翻译按钮获取中文摘要

Use of high intensity, AML-type chemotherapy (HI) in patients age > 60 with untreated high-risk MDS (International Prognostic Scoring System categories intermediate-2 or high) has yet to improve survival relative to supportive care, consequent to both a 35% rate of treatment -related mortality (TRM) and a 60% rate of "resistance" to therapy. New HI {e.g. clofarabine + ara-C) may decrease resistance, thereby increasing survival in older patients. Another approach emphasizes use of lower intensity, "targeted", therapies (LI) in such patients. As single agents, LIs likely produce lower short-term TRM rates than HI, but, to date, have also produced lower CR rates. This is noteworthy because, at least with HI, only patients who achieve CR achieve a potential long-term survival advantage. Thus, it is unclear whether untreated older patients with high-risk MDS should receive HI first followed at failure by LI (the two:course strategy HI, LI), or, alternatively, receive HI only after failing LI (the two-course strategy LI, HI). Our first specific aim is to compare these strategies in such patients. We will employ our recently published statistical design thai not only permits comparison of the individual treatments (Hl= clofarabine + ara-C, L11 = PKC412, LI 2= R115777, LI 3= decitabine), used either as initial or as salvage therapy, but also allows selection of the best strategy, L11, LI 2, or LI 3, followed at failure by HI, or HI followed at failure by L11, LI 2, or LI 3. Selection is based on probabilities of both response and death allowing for "trade-offs" between these outcomes such that that several potential rates of response, on the one hand, and death, on the other, are considered equal. To aid in evaluation of LIs in MDS, an NCI Working Group has recently published criteria for response. These include several categories of "minor response"(MR, less than CR). With HI, we have demonstrated that CR prolongs survival to a much greater extent than MR. Therefore; the second specific aim of the Project is to evaluate the significance of Ll-induced MR, for both survival and "quality of life". The "results" of any therapy are an average of results from different patients, some of whom the therapy may have helped or harmed more than a hypothetical" average" patient. Our previous attempts at identifying such subgroups have used statistical models to assess interactions between treatment and clinical prognostic factors, e.g. cytogenetics. In contrast, the LIs used here are directed at targets that are quantifiable in the laboratory. Thus our third specific aim is to identify patient subgroups responsive to a particular LI by testing for correlations between clinical outcome and the pre-treatment status of the target or the Li's ability to modulate the target. This will be done in collaboration with Projects 2-4 and Cores A-C.