Eight Colors Flow Cytometry Phenotyping for Blood Minimal Residual Disease Monitoring in Hairy Cell Leukaemia Patients.

Abstract 1609

Poster Board I-635

Purine analogues (2-chlorodeoxyadenosine: 2-CdA and 2'-deoxycoformycin: dCF) significantly improved the complete response (CR) rate and the overall survival of hairy cell leukaemia (HCL) but are usually not curative with about 40% relapse rate. We developed an 8 colors flow cytometry (8-FC) technique (CANTO II^TM, BD Biosciences) to follow the blood minimal residual disease (MRD) and evaluated its clinical relevance. 34 patients diagnosed with HCL at the Necker University Hospital (France) were retrospectively evaluated on frozen blood samples taken at diagnosis to select the relevant markers for MRD analysis. MRD follow-up (106 samples) using 8-FC was compared to IgH PCR clonality, using consensus qualitative IgH FR1 and FR2 PCR and Genscan revelation, performed on the same samples.

Several antibodies were tested. The phenotype of hairy cells (HC) appeared to be heterogeneous from one patient to another. After a first identification of HC on the basis of CD45^high, CD3 negativity, CD19/CD20/CD103/CD25 (moderate to high) positivity and light chain isotype restriction, we tested the expression of CD123 and CD305 (LAIR-1). They were expressed in all cases with high intensity as defined by the mean fluorescence intensity ratio (MFIR) calculated by dividing the MFI of each marker by that of the isotypic control. They could distinct HC from normal B cells in 82% and 78% whereas CD103 and CD25 were expressed with a lesser intensity and were not relevant for MRD analysis in 30% and 22%. Therefore, we have chosen the 8 antibodies combination as follows: CD103^FITC/CD305^PE/CD19^{Percp Cy5.5}/CD123^APC/CD25^PC7/CD3^Alexa700/CD45^Amcyan/CD20^{Pacific Blue} which discriminates HC in 100%. This combination was used to define in all 34 patients a “hairy cell associated phenotype” based on the expression of at least 2 HCL associated markers. The specificity was assessed on 10 normal blood samples and Our 8-FC combination appeared very specific with a robust sensitivity of at least 1.10^-4. The comparative analysis of blood CF versus molecular MRD analyses showed an overall concordance of 81%: 52 samples (49%) being negative and 35 samples (33%) being positive by both techniques. 19 samples (18%) were discordant. In all these cases, CF MRD was positive (levels ranging from 10^-4 to 10^-2) but IgH-PCR appeared polyclonal. These data demonstrate the higher sensibility of CF as already shown by others.

We then looked for a clinical relevance of FC MRD follow-up in a cohort of 18 patients with normal blood cell counts after 2-CdA therapy (5 in first line and 13 after 2 to 8 previous therapies including purine analogues, splenectomy, interferon-á and others). A median of 4 samples by patient were tested and the median follow-up time was 44 months [18-84]. Two groups could be distinguished. (A) Patients (n=13) with a sustained negative FC MRD (<10^-4). Nine patients were negative from the first sample taken at around 6 months post 2-CdA. A FC MRD relapse was observed in three of them at 32, 46 and 72 months after therapy. Four patients were positive for the first sample but became negative at around 1 year post 2-CdA and remained negative after a follow-up of at least 18 months. (B) Patients (n=5) remaining FC MRD positive (median of 3 samples). Three were treated for haematological relapse at 40, 48 and 49 months from 2-CdA therapy. Two kept normal blood cell counts after a follow-up of 38 and 60 months.

These data suggest that the risk of relapse after 2-CdA therapy is lower in case of negative blood FC MRD (<10^-4) obtained in the 12 months following therapy than in patients with sustained positive MRD (≥10^-4). This should lead, at least, to a closer haematological and FC follow-up of the positive patients. A prolonged follow-up is required to determine if such patients would benefit from a complementary therapy. Finally, we propose modified NCI criteria of response as follows in table 1.