Next-Generation Deep Sequencing Reveals Multiple Ighv Clones in One Third of CLL Patients Defining New Prognostic Subgroups and Improving Previous Classification

Introduction: Chronic Lymphocytic Leukemia (CLL) is characterized by a heterogeneous clinical course. Currently, the mutational status of the immunoglobulin heavy chain variable (IgHV) region defines 2 risk groups: patients with ≥ 2% difference from the germline are considered as unmutated (UM) and have a poor prognosis while the opposite is observed for mutated (M) patients (Hamblin et al, 1999). Until now, clonality has been determined using PCR/gel electrophoresis and the percentage of IgHV mutations by Sanger sequencing using the Biomed-2 method (van Dongen, Leukemia 2003) which only allows for analysis of the major clone and assumes that only one clone per patient should be considered.

Methods: in the present study, we sequenced the IgHV gene for 200 CLL patients with a median follow-up of 70 months (range, 1-309) by next-generation (NGS) and Sanger sequencing (SS) and investigated the impact on prognosis for the different subgroups. Briefly, 100ng of cDNA generated from RNA extracted from highly purified CD19+ cells obtained at diagnosis was amplified using IGH LEADER master mixes. For SS, clonality of PCR product was determined and monoclonal samples were directly sequenced. For NGS, PCR products were purified using magnetic beads, normalized and pooled to create a library for sequencing using the MiSeq v3 Reagent kit (600 cycles). Sequencing data was analyzed using LymphoTrack™ bioinformatics software and percentage of clone was based on the first 200 most abundant clones detected. Clones with an abundancy of < 2.5% were not considered. The IMGT database was used to calculate the percentage of mutation compared to germline.

Results: Using SS, 49% (98/200) of patients were IgHV M, 38% (76/200) UM, while 13% (26/200) were polyclonal and further Sanger sequencing was inconclusive. Concordance between SS and NGS was 98.3%: for 90.2%, the major clone detected by NGS was the same by SS, for 8.5% the clone was the same but the percentage mutated compared to germ-line differed slightly with a median error of 0.6%, while for 1.7%, SS detected another clone which was less abundant by NGS. Interestingly, among patients considered as monoclonal by SS, 25.3% (44/174) were found to be polyclonal by NGS. NGS revealed that 35% (70/200) of patients display different IgHV re-arrangements in the same patient: the median frequency of the first two most abundant clones were 78.2% (range, 28.5-94.9) and 16.4% (range, 2.5-49.0) respectively. This implies that at least in these patients, leukemia-initiating events must occur prior to IgHV re-arrangement in a pro-B cell. Further in-depth studies of the HSC and early B cell progenitor compartments in these patients will be required to confirm these observations. In terms of prognosis, M, UM and polyclonal patients determined by SS had a median treatment-free survival (TFS) of 178, 29 and 129 months respectively (P<0.0001) and a median overall survival (OS) of >309, 183 and >309 months (P<0.0001). Since NGS was able to highlight different clones, we were able to create 5 different categories: patients with (a) multiple M clones, (b) 1 M clone, (c) a mix of M-UM clones, (d) 1 UM clone, (e) multiple UM clones. Using this new NGS classification, we found patients with different prognosis among patients already classified by SS. For example, SS-M patients were classified in 3 subgroups with a median TFS of >251 (a), 178 (b), 56 (c) months (P=0.0014). Similar results were observed for SS-UM patients who were divided in 3 subgroups with a median TFS of 94 (c), 24 (d), and 19 (e) month (P=0.0296). When all patients where considered, IgHV-NGS classification stratified patients in 5 different subgroups with median TFS of >251 (a), 178 (b), 57 (c), 24 (d), 19 (e) months (P<0.0001) and a median OS of 270, >309, >309, 183, 88 months (P<0.0001).

Conclusions: determination of IgHV mutational status by NGS has excellent concordance with classical SS and in addition enables the detection of small clones which has a significant impact on prognosis. This allows the analysis of all clones without labor intensive manipulations especially for polyclonal patients. Here we showed for the first time that one third of CLL patients present with multiple IgHV subclones that impact on prognosis and refine the previous SS-IgHV classification.