Flow Cytometric Response Monitoring in MDS with Del(5q) Using a Simple 5-Parameter-Score - an Elnet Imds-Flow Experience

Recently, we have established a 5-parameter flow cytometry (FCM)-based score allowing for a precise prediction a deletion (5q) in therapy naïve MDS patients. The aim of this prospective study was to test, whether this FCM-based profiling is at least equal to the cytogenetics/FISH-based del(5q) detection for monitoring response to treatment.

Overall, 229 FCM investigations were performed in 57 patients with MDS and del(5q) (IPSS-R very low/low: n=24, int: n119, high/very high n=22) including 39 patients with isolated del(5q) or one additional cytogenetic abnormality. The majority of analyses were performed in patients receiving lenalidomide or azacitidine (n=28 and n=21 patients), or in patients receiving chemotherapy and/or allogeneic transplantation (n=3), or growth factors (n=5). The del(5q)-FCM-score includes the following 5 parameters: myeloid progenitors (myPC > 2%) = 3 points, CD45 MFI ratio (lymphocytes: myPC ≤ 7.0) = 10, SSC ratio (granulocytes : lymphocytes < 6.0) = 2, CD71 (≤ 20%) on granulocytes = 1.5, and female gender = 1.5; a score ≥ 15.0 indicates the presence of del(5q). A standardized FCM (lyse-stain-wash) and cytogenetics/FISH were performed according to ELN guidelines at the TU of Dresden, VUMC of Amsterdam, UH of Bristol, and UH of Guadalajara.

Before therapy, in 61 cytogenetic/FISH analyses in 40 MDS patients a del(5q) was detectable. In 53/561 (87%) FCM measurements, performed in parallel, a del(5q) could be predicted using the del(5q)-FCM-score. Monitoring response to treatment, a cytogenetic complete response (CCR) with the disappearance of del(5q) was accompanied by the disappearance of the typical del(5q)-FCM-profile in all of the 50 (19 patients) FCM measurements (sensitivity=100%). Otherwise, in patients without CCR the presence of del(5q) could be predicted by FCM in 96/118 measurements (specificity=81%).

Aiming at a still higher specificity, in the following analyses we included only patients who presented with a typical del(5q)-FCM-score before therapy (33 patients; 54 measurements; median score=16.5) resulting in the above described sensitivity of 100% (32 measurements; median score=13.0) as well as a remarkably high specificity of 97% (59/61 measurements; median score=16.5) for response prediction.

Next, we compared cytogenetics/FISH and del(5q)-FCM-score with further methods for response monitoring as cytomorphology, two well established diagnostic FCM scores: FCSS (flow cytometry scoring system; Wells et al. 2003) and the diagnostic score (Ogata et al. 2009), as well as the assessment of hematological improvement (HI). Thus, cytomorphology and FCSS, analyzing dyspoiesis of multiple cell lineages, showed a CR or disappearance of a MDS typical FCSS in less than the half of all investigations being in cytogenetic CR (sensitivity: 41% and 38%), but those two methods revealed a high specificity (98% and 97%). On the other side, the analysis of HI was high sensitive (91%) but not specific (49%). The use of the diagnostic Ogata score, considering almost only abnormalities of the myeloid progenitors, ended up with a slightly lower sensitivity (91%) and specificity (89%) as cytogenetics/FISH and the del(5q)-FCM-score.

Finally, response monitoring using the del(5q) FCM-score allowed for a better separation of overall survival compared to cytogenetics/FISH (p=0.039 vs. p=0.1725).

Flow cytometry-based detection of a del(5q)-specific immunophenotype is feasible and can serve as a rapid tool for response monitoring during treatment with disease-modifying drugs. At the moment, we test whether the better separation of the patients' survival curves applying the FCM-score holds true in a larger patient cohort to be analyzed within the ELNet iMDS-Flow working group.