Quantitative Dynamics of Flow Cytometric Aberrancies during Treatment with Erythropoietin/G-CSF Are Predictive for Responses in LOW/INT-I Risk Myelodysplastic Syndromes.

Flow cytometry (FCM) in myelodysplastic syndromes (MDS) is recognized as a useful tool in identifying aberrancies in expression of differentiation antigens on immature and mature bone marrow cells of erythroid and myelomonocytic lineages. FCM in MDS correlates significantly to morphology according to WHO and the WHO-based Prognostic Scoring System (WPSS). Moreover, certain aberrancies in pure RA+/−RS patients, i.e. CD7 on myeloid blasts, seemed to predict transfusion dependency and disease progression independent of classical prognostic parameters such as WHO, cytogenetics and IPSS (Van de Loosdrecht et al., Blood 2008, 111). Flow cytometric aberrancies in bone marrow of 29 patients with low or int-I risk MDS were analyzed at diagnosis and after approximately 1 year of treatment with a standardized regimen of Epo (NeoRecormon®) and G-CSF (median: 12 months (range: 4–24) in order to evaluate the value of FCM in disease monitoring. Patients were classified as RA+/−RS (n=11), RCMD+/−RS (n=15), 1 hypoplastic MDS, 1 MDS-U and 1 MDS/MPD. Progression was defined as an increase in WHO subgroup to at least RAEB-1 within 18 months after diagnosis of MDS. Before treatment, median number of aberrancies in the myelomonocytic lineage, as assessed by FCM, was 4 in RA+/−RS (range 1–8) and 5 in RCMD+/−RS (range 3–10); other MDS subgroups were too small. In age-matched normal bone marrow samples the median number of aberrancies was 1 (range 0–3, n=18). Aberrant expression of CD5, CD7 or CD56 was observed on myeloid blasts at diagnosis in 13 out of 18 patients that were transfusion-dependent or suffered from progressive disease during follow-up (4 RA+/−RS, 7 RCMD+/−RS, 1 MDS-U and 1MDS/MPD). Interestingly, expression of these markers was only detected in 1 out of 11 patients that were not or low transfusion dependent. This particular RCMD-RS patient had short response duration. The number of blast aberrancies correlated significantly to transfusion dependency and disease progression (r=0.459, p=0.012). Nine patients showed hematological improvement upon Epo/G-CSF treatment. In most of these patients (4/6 RA+/−RS and 2/3 RCMD+/−RS) the number of flow cytometric aberrancies decreased (median 4 at diagnosis to 2.5 at follow-up in all responders). This decrease was caused by a diminished number of monocytic aberrancies (p=0.028); a stable blast aberrancy was seen in only one patient. In 15 patients with stable disease, no significant changes in the number of flow cytometric aberrancies could be detected (median 5 and 5, at diagnosis and follow-up, respectively); whereas in 4 of the 5 patients with progressive disease during Epo/G-CSF (2/2 RA+/−RS, 1/2 RCMD+/−RS and 1 MDS/MPD) the number of aberrancies increased (median 7 and 9, at diagnosis and follow-up, respectively). This increase was mainly caused by changes in the number of aberrancies in blast and granulocyte subpopulations. The response to therapy seems to depend on the number of blast aberrancies (r=0.351, p=0.062). From these data we hypothesize that FCM can be used as an objective sensitive disease monitoring parameter in low/int-I risk MDS. Furthermore, in a subset of these patients treatment with Epo/G-CSF might have altered disease progression reflected by normalization or stabilization of aberrancies on myelomonocytic cells. This underscores observations that Epo/G-CSF improves overall survival and/or leukemia free survival. Prospective studies are being conducted to validate flow cytometric analysis in the diagnosis, prognostication and disease monitoring of low/int-I risk MDS during Epo/G-CSF and lenalidomide. To conclude, FCM can contribute to the management of low/int-I risk MDS patients as changes in the number of flow cytometric aberrancies reflect response to treatment.