Flow Cytometric Score Correlates with Clinical Response to Azacitidine In Intermediate-2 and High Risk Myelodysplastic Syndrome Patients

Abstract 441

New treatment strategies that potentially change the natural course of intermediate (int)-2 and high risk myelodysplastic syndromes (MDS), such as azacitidine, are emerging. Recently, we reported that flow cytometric analysis of bone marrow (BM) in low and int-1 risk MDS is instrumental to identify clinically relevant subgroups. (Westers et al, Blood 2010) Moreover, it was reported that a flow cytometric scoring system (FCSS) is predictive for worse outcome in MDS. (Wells et al, Blood 2003, van de Loosdrecht et al, Blood 2008) The FCSS is a scoring system that allows for a numerical display of immunophenotypic aberrancies in the (im)mature myelo-monocytic lineage. Scores are generated by enumerating abnormalities; e.g. high scores reflect a high number of aberrancies. The current study aimed to investigate the role of this flow cytometry-based scoring system to assess and monitor response to treatment in int-2 and high risk MDS patients treated with azacitidine. Bone marrow aspirates were analyzed by flow cytometry in 18 MDS patients who were treated with azacitidine. Aspirates were drawn before treatment and after every third cycle of azacitidine. Response to treatment was evaluated using IWG-2006 criteria. The median age was 71 (range 50–78). Distribution over WHO 2001 categories was RCMD-RS n=2, RAEB-2 n=7, AML with 20–30% blasts n=6 and MDS/MPD n=3. International prognostic scoring system (IPSS) categories comprised int-2 n=8 and high n=5. In 5 patients the IPSS score could not be assessed due to lack of cytogenetics, these patients were at least int-2 MDS patients. Flow cytometric follow up was available in 12 patients due to short follow up, i.e. in 4 responders, 4 progressive disease (PD) and 4 stable disease (SD) patients; 3 patients stopped due to non-hematologic toxicity, 4 patients died of PD. Median follow up was 7 months (range 3–12). Median pre-treatment Hb was 6.7 mmol/L, platelets 35.5*10e9/L and absolute neutrophil count (ANC) 0.82*10e9/L. Responders had a significant increase in Hb (median 7.8 mmol/L, p=0.04), platelets (291.5 *10e9/L, p=0.03) and ANC (1.4*10e9/L, p=n.s. compared with baseline values). SD and PD patients had a median Hb of 6.5 mmol/L, platelets 69*10e9/L and ANC 0.77*10e9/L. The median pre-treatment flow score was 6 (range 3–8). Interestingly, responders had a significant decrease in flow score from median 5 to median 2 (range 1–3, p=0.005) after 3 months of treatment. No change in flow scores was seen in PD and SD patients after 3 months of treatment (median=6, range 4–8 and pre-treatment FCSS median=6.5, range 3–8). A sustained decrease in flow score was seen in 4 responders after 6 months of treatment (median 2, range 1–3) parallel to a further increase in median Hb (9.0 mmol/L), platelets (278.5 *10e9/L) and ANC (1.7*10e9/L). After 9 cycles of azacitidine, 3 patients were still responsive to treatment (median Hb 9.9 mml/L, platelets 169, ANC 4.07, median flow score 5, range 2–4). The majority of SD and PD MDS patients had aberrant marker expression on myeloid progenitors, such as CD5, CD7 and/or CD56 compared with responsive MDS patients, (63% (5/8) vs 25% (1/4). Moreover, initial loss of aberrant marker expression on myeloid progenitors was detected in one patient who responded to azacitidine treatment. At 9 months, this initially responsive patient and a patient from the stable disease group developed progressive disease. Interestingly, the initially responsive patient showed an increase in the percentage of myeloid progenitors with an aberrant immunophenotype at 6 months; of note, the total percentage of CD34+ cells was less than 3% by flow cytometry. The initial loss of aberrant marker expression on myeloid progenitors might be caused by a relative increase of normal progenitors and decrease of malignant progenitors caused by azacitidine treatment. In conclusion, our data indicate that flow cytometry identifies MDS patients who may benefit from azacitidine treatment by detection of aberrant marker expression on myeloid progenitors. Moreover, the data indicate that the FCSS may be instrumental in selection of SD patients who may benefit from prolonged treatment with azacitidine.