Drs. Scott and Estey provide an excellent, concise review of current diagnostic and therapeutic approaches for patients with myelodysplastic syndromes (MDS). Both physicians are recognized world leaders in this area. They start their article by reviewing current diagnostic and prognostic classifications of MDS from the initial French-American-British schema, the World Health Organization (WHO) classification, the International Prognostic Scoring System (IPSS), and the most recent WHO Classification-Based Prognostic Scoring System (WPSS) proposal. We agree that most of these classifications have limitations, including the need for subjective manual quantification of blast percentage and morphologic categorization of disease subtype.
Prognostic Issues
In our experience, very few patients are referred to our center with an initial diagnosis based on WHO criteria. Currently at M.D. Anderson Cancer Center (MDACC), we make our therapeutic decisions based on IPSS score,[1] dividing patients into lower-risk (low and intermediate-1) and higher-risk disease (intermediate-2 and high). We favor that classification because it provides information on the natural history of the disease from a group of patients who had not received any prior meaningful therapy. That said, the IPSS score has several major limitations, the most important of which is probably the fact that is not a good tool for predicting prognosis of patients with lower-risk disease. This is important as approximately two-thirds of MDS patients are in that category.
To study this issue, we analyzed the outcomes of a large cohort of patients with lower-risk disease who were referred to MDACC. We developed a simple model that includes age, hemoglobin levels, platelet count, abnormal cytogenetics, and percentage of blasts to predict the overall survival of patients with lower-risk MDS.[2] Survival could range from no median survival in a group of patents with very low risk (mainly younger patients who are transfusion-independent) to a subset of patients with a survival of less than 24 months. Close to 80% of patients were in a higher risk category.
The second finding from this analysis was the fact that a large majority of patients died without having progressed to acute myelogenous leukemia (AML), with the most frequent causes of death being infections and bleeding episodes. This strengthens the notion of the need to develop therapeutic interventions for patients with “early-stage” MDS, the same way we initiate therapy in early chronic phase chronic myelogenous leukemia (CML) without waiting for transformation to accelerated phases.
The main limitation of our analysis is that the study was restricted to patients referred to MDACC and that survival was computed from time to referral and not from initial diagnosis. Nevertheless, this tool is important for tertiary care centers such as the Fred Hutchinson Cancer Research Center. It also should be noted that this new low-risk model was recently validated, in part, by German investigators.[3]
Treatment Issues
In the second section of their article, the authors review current supportive and therapeutic measures in MDS. Although we agree with most of the data presented, some points warrant further discussion. First, we do not completely agree that asymptomatic patients with stable blood counts should simply be observed. Data discussed above indicate that a significant fraction of patients with lower-risk disease would benefit from earlier treatment of their disease if a nontoxic effective therapeutic approach existed.
Such intervention does not currently exist but needs to be developed in clinical trials. Examples at our center could include the use of oral or very low-dose schedules of hypomethylating agents and single-agent use of histone deacetylase inhibitors such as vorinostat (Zolinza) or LBH589. We recommend participation on clinical trials to all patients with MDS who are referred to MDACC. (A complete list of MDS trials at MDACC can be obtained at http://www.mdanderson.org/diseases/mds/.)
We agree that data on monitoring, transfusion thresholds, and use of chelating agents is controversial. However, data on both combination growth factors and iron chelation therapy suggest that these strategies may have an impact on overall survival in patients with very low risk.[4,5] We fully agree that randomized clinical trials are needed to prove the role of both interventions, particularly the growth factor combination for patients early in their disease course and chelation for patients with anticipated long survivals who are heavily red cell transfusion-dependent. Indeed, only a minority of these patients will be candidates for supportive care only. In view of the expected low toxicity from these agents, we think that is generally reasonable to administer them.
Other Therapeutic Options
Two other alternatives exist for patients with lower-risk disease. One includes the use of immunomodulatory approaches such as ATG (Atgam, Thymoglobulin) and cyclosporine combinations, and the other, the use of thalidomide(Drug information on thalidomide) derivates such as lenalidomide (Revlimid). In terms of our experience with ATG, our data are consistent with findings from Mufti et al in London, where most patients who benefited had hypoplastic MDS independent of DR15 positivity.[6] In an ongoing MDACC study of rabbit ATG with cyclosporine, most responses have been observed in patients with aplastic anemia. Therefore, we only recommend ATG-based therapy for older patients with hypoplastic MDS or younger patients without a donor for allogeneic stem cell transplantation (alloSCT).
The role of lenalidomide is clearer. In our opinion, this drug should be used in patients with anemia and a chromosome 5 alteration without significant thrombocytopenia early in the course of their disease. Lenalidomide has remarkable activity in this setting. Outside of this, we still consider its use for lower-risk non-del5q MDS or higher-risk del5q to be investigational. New schedules, agents (ie, CC11006), and combinations (with growth factors or azacitidine [Vidaza]), are already being explored in this population.
