Multiparameter flow cytometry (MFC) has been demonstrated capable of identifying aberrant antigen expression related to myelodysplastic syndromes (MDS). The exact role and place of MFC in the diagnostic work-up of patients with suspected MDS, however, remains to be defined. Aim: Evaluation of the diagnostic impact of MFC in relation to cytomorphology (CM) and cytogenetic (CG) by determining the association of MFC results to overall survival (OS). Patients and Methods: In 1,013 patients with suspected MDS bone marrow samples had been analyzed in parallel by MFC, CM, and CG. CM confirmed MDS in 511 patients, excluded MDS in 277 patients, and showed dysplastic features but not sufficient to unequivocally diagnose MDS by CM in 225. The MFC diagnostic result was in agreement with MDS (“MDS by MFC”) in 446 patients including 382/511 patients with MDS proven by CM. CG revealed an aberrant karyotype in 245/1,013 patients. The median follow-up time amounted to 14.8 months, a total of 156 deaths was recorded. Results: The first set of analyses was performed on the cohort of 511 patients with MDS confirmed by CM. The median total number of aberrantly expressed antigens amounted to 3 (range, 0–11) and included expression of mature antigens in myeloid progenitors; abnormal CD13-CD16- and CD11b-CD16-expression patterns, aberrant expression of myeloid markers and reduced side scatter signal (SSC) in granulocytes; reduced expression of myelomonocytic markers in monocytes; aberrant expression of CD71 in erythroid cells; as well as expression of lymphoid markers in all myeloid cell lines. A higher total number of aberrantly expressed antigens as a continuous variable correlated with a shorter OS (Cox analysis, p=0.008). Next, patients were categorized based on the three parameters i) at least 3 aberrantly expressed antigens, ii) significantly reduced SSC in granulocytes, and iii) >5% myeloid progenitor cells in MFC. Patients with at least one of these criteria had a significantly shorter OS than those without (median 48.5 months vs. not reached (n.r.), p<0.001). Overall, the global diagnostic rating of “MDS by MFC” was the strongest MFC parameter: Patients with “MDS by MFC” had a shorter OS as compared to patients without (median 56.8 months vs. n.r., p=0.001). Non-MFC parameters related to OS in univariable Cox analysis included WBC count, thrombocyte count, CG (grouped according to IPSS), % blasts by CM (p<0.001 each), Hb level (p=0.001), and age (p=0.002). In order to determine the clinical relevance of “MDS by MFC” a multivariable analysis for OS was performed on this parameter together with non-MFC parameters (blood counts excluded due to incomplete data sets). It revealed an independent relation between “MDS by MFC” and OS (p=0.045). This was also true for relation of OS to the other parameters (CG, p<0.001; age, p=0.001, % blasts by CM p=0.014). Given this strong prognostic value of “MDS by MFC” in cases with MDS proven by CM a second set of analyses on the relation between MFC findings and OS were performed for the complete cohort of 1,013 patients, i.e. additionally including all cases with a diagnostic result by CM of “no MDS” or “dysplastic features not sufficient to diagnose MDS”. Again, significant relations to OS was found for the total number of aberrantly expressed antigens as a continuous variable (Cox analysis, p<0.001), for at least one of the above mentioned criteria i), ii) or iii) (median 75.6 months vs. n.r., p<0.001), as well as for “MDS by MFC” (median 60.5 months vs. n.r., p<0.001). Again, “MDS by MFC” proved to be the most relevant MFC parameter. Multivariable Cox analysis for OS including “MDS by MFC” and non-MFC parameters revealed a trend only for “MDS by MFC” (p=0.135) and significance for the other parameters (age, p<0.001; CG, p<0.001; blasts by CM, p=0.045). Conclusions: 1) The present data indicates the diagnostic use of MFC for MDS results in independent prognostic information for cases with MDS as proven by CM. 2) Furthermore, the diagnosis of MDS by MFC has a strong prognostic impact even without prove of MDS by CM which strengthens the diagnostic value of MFC even more. 3) This analysis therefore argues in favour of diagnosing MDS not only based on a combination of CM and CG but of adding also MFC for better classification and even prognostication in the future.
Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Alpermann:MLL Munich Leukemia Laboratory: Employment. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.
Asterisk with author names denotes non-ASH members.