Erythropoietin (Epo) and G-CSF is standard treatment for patients with low and intermediate risk-I myelodysplastic syndromes (MDS) with low Epo levels and transfusion need. Recently, it was demonstrated that flow cytometry adds significantly in the distinction of clinically relevant subgroups in MDS with respect to transfusion dependency and progressive disease. (Van de Loosdrecht et al., Blood 2008, 111) We report a prospective clinical study of 48 patients with low and intermediate-I risk MDS treated with a standardized Epo/G-CSF regimen to address the question whether flow cytometric analysis was instrumental in predicting response. All patients started with Epo if symptomatic at a Hb of less than 6.2 mmol/l independent of endogenous erythropoietin level. Epo (NeoRecormon®) was started at a dose of 30.000U once weekly. In the absence of an increase in Hb of at least 0.6mmol/l within 6 weeks, Epo was escalated to 60.000U/weekly. If still no response was achieved, G-CSF was added (3 times weekly; dose dependent on weight). Hematological response was evaluated according to IWG2006 response criteria. (Cheson, et al. Blood,2006, 108) All patients were scheduled to Epo 60.000U/week and additional G-CSF was dosed in 46 patients. From these patients 17/48 (35%) responded to the standardized Epo/G-CSF regimen. Patients with a low IPSS (0, n=25) showed hematological improvement (HI) in 44% of the cases. Patients with intermediate-I risk MDS (0.5–1.0, n=22) showed a HI in 32% of the cases. When the WHO classification-based prognostic scoring system (WPSS) was used to classify patients into subgroups (Malcovati et al., J Clin Oncol 2007, 25), 43% of the patients with a very low WPSS score (0, n=7) responded to treatment, whereas 67% in the low WPSS (1, n=21) and none in the intermediate and high risk WPSS group (2 or 3, n=20). The levels of endogenous Epo significantly discriminated between responders and non responders (p=0.003). When a cut off of 100U/l was used, 14 of the 22 patients (64%) with an Epo level below 100U/l were responsive to Epo/G-CSF treatment with a median time to response of 11 months (SD 6 mo.); 23/26 patients with high Epo levels were non responsive/Epo levels correctly predicted response to treatment in 77% of the cases (37/48, p<0.001). Strikingly, aberrancies within the myeloid blasts compartment (defined by flow cytometry as CD45dimCD34+SSClow) were mainly seen in non-responding patients (22/31 non responders, 71%). Observed aberrancies were the expression of a lineage infidelity marker (CD5, CD7 or CD56) or a loss of the myeloid antigen CD33. Only 1 out of 17 patients that responded to treatment (6%) showed an aberrant phenotype (CD7); however, response duration in this particular patient was only 4 months as compared to a median time of 12 months in the other cases. The presence of flow cytometric aberrancies correctly predicted response to treatment in 79% of the cases (38/48 cases, p<0.001). None of the responders had both a high Epo level and flow cytometric aberrancies, whereas 3 responsive patients had a relatively high Epo level without flow cytometric aberrancies. Combining Epo levels and flow cytometric aberrancies increased correct prediction of response to therapy to 85%. In 5/8 (63%) of non responders with low Epo levels an aberrant phenotype was observed. Thus, when the presence of flow cytometric aberrancies is taken into account, it may prevent needless treatment with Epo in this group. To conclude, flow cytometry may add significantly to the well known validated predictive parameters for response to select patients who are likely to respond to Epo/G-CSF. The role of flow cytometry in prediction of response to new drugs such as lenalidomide in low/int-I risk MDS is currently under investigation.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author