Initial transfusion frequency and survival in myelodysplastic syndromes: hopping onto a fast train to nowhere

“I hear the train a comin’. It's rolling round the bend. And I ain't seen the sunshine since I don't know when” [1].

Johnny Cash, Folsom Prison Blues

Anemia has long been one of the defining characteristics of the myelodysplastic syndromes (MDS), and is present in over 50% of patients at the time of diagnosis. This clinical finding informed the MDS nomenclature used in the French–American–British pathologic classification almost four decades ago; more recently, a molecular revolution has been rapidly illuminating the genetic underpinnings driving marrow failure, and specifically some of the causes of anemia. Exemplifying this point, advanced sequencing techniques have identified alterations in, for instance, SF3B1 and RPS14 – abnormalities involved in initiating dyserythropoiesis and the consequent need for red cell transfusions [2,3]. Despite these and other exciting revelations, a full understanding of marrow failure in MDS remains elusive. What has always been clear is that marrow dysfunction is strongly linked to survival outcomes in patients with MDS, and intuitively worsening cytopenias and transfusion burdens foreshadow a worse survival.

Over the past three decades a multitude of prognostic models in MDS have been developed, most of which include anemia as a clinical factor correlating with survival [4–8]. Emblematic of this is the revised International Prognostic Scoring System (IPSS-R), which, despite the increased emphasis on the import of cytogenetics, still includes hemoglobin level at diagnosis as a prognostic marker [7]. Going hand-in-hand with anemia, red cell transfusion dependence has also been linked to worse survival [9], and is one of the three variables included in the original version of the World Health Organization (WHO) Prognostic Scoring System (WPSS) [8]. The WPSS was developed as a dynamic model enabling prognostication at any point in a patient's disease course. It is easy to imagine clinical scenarios that could confound the accuracy of transfusion dependence as a marker of disease progression, though, including the development of splenomegaly, gastrointestinal bleeding, acquisition of alloantibodies or other factors that could reduce the effectiveness of red cell transfusions to mitigate anemia. Perhaps as a result, the WPSS has undergone revision to focus on anemia in place of transfusion dependence [10]. Furthermore, it is unclear whether the inferior survival associated with red blood cell transfusion dependence is a result of iron accumulation and end-organ damage from serial transfusions or just a marker of a more bellicose disease. While there is an inverse relationship between the marker of iron overload, serum ferritin, and survival [7], iron chelation has not convincingly demonstrated a disease-modifying effect in patients with MDS, thus leading to controversy regarding its place in the current treatment arsenal [11–14].

Chan and colleagues overcame these limitations in their report that accompanies this Commentary [15]. By focusing on red cell transfusion requirements in the 4 weeks following a patient's first transfusion, they limit the confounding effect of iron accumulation, as well as the development of splenomegaly and alloantibodies, on the relationship between transfusion burden and survival. Furthermore, by including only newly diagnosed patients followed at their center, they were able to obtain complete and accurate transfusion data in the setting of consistent transfusion parameters. Notably, few patients received active treatment during the study period. With a median follow-up of 33.5 months, patients who required their second transfusion in the 4 weeks subsequent to their first transfusion went on to have a higher transfusion burden and received more cumulative transfusions over time. These patients also had an inferior overall survival when compared to those who did not require their second transfusion within 4 weeks, with a median of 3.2 vs. 5.8 years (p < 0.01). In a multivariate analysis, the predictive value of initial transfusion intensity persisted, and was independent of IPSS, age and progression to acute myeloid leukemia (hazard ratio [HR] = 3.69, 95% confidence interval [CI] 1.41–9.64).

Initial transfusion intensity provides a simple yet powerful predictor of outcome that can be applied without access to sophisticated laboratories. In an age of next-generation sequencing and complex panels of acronymous prognostic markers, there is a quiet elegance in this study's simplicity, the results of which can be applied in any office or clinic. Analogous to the milestones for reductions in BCR–ABL transcripts after initiation of therapy for chronic myeloid leukemia [16], initial transfusion intensity can be used as a simple predictor of ultimate outcome, and can identify those who may need to intensify or even change therapy independent of IPSS.

The results of this study also add to the understanding of how transfusion burden and iron overload relate to prognosis in MDS. Although it has been suggested that transfusion dependence may mediate its effect on overall survival through iron accumulation, by focusing on patients not yet heavily transfused without opportunity to accumulate iron, Chan and colleagues were able to draw conclusions independent of iron overload, adding credence to the hypothesis that transfusion burden can be viewed as more of a marker of disease severity. What remains to be demonstrated is whether these findings can be repeated prospectively, and incorporated into prognostic scoring systems such as the IPSS-R to enhance their accuracy. Naturally, the most critical step is to use these findings to alter therapy in identified patients at high risk, and thus favorably modify the ultimate destination of their disease's train.

Potential conflict of interest

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.