Detection of Immunoglobulin Heavy Chain Gene Clonality By High-Throughput Sequencing for Minimal Residual Disease Monitoring in Chronic Lymphocytic Leukaemia

Introduction: The negative minimal residual disease (MRD) after treatment has been recently accepted as endpoint for Chronic Lymphocytic Leukaemia (CLL) clinical trials. Conventionally, MRD can be detected by using multi-color Flow Cytometry (FC) with high sensitivity. Determination of the clonal immunoglobulin gene rearrangement can be a useful monitoring marker in a broad range of B-cell lymphoproliferative neoplasms. Moreover, the mutational status of immunoglobulin heavy chain variable (IgHV) rearrangement is considered one of the most important prognostic factors in CLL. Therefore, the identification of the IgHV rearrangement can be a useful marker both at diagnostic and as monitoring marker for MRD. Nowadays, high-throughput sequencing (HTS) technologies has enabled highly sensitive cancer genomic testing in clinical laboratories. There are same initiatives based on HTS to use IgHV rearrangement as marker for MRD monitoring in Acute lymphoblastic leukemia or multiple myeloma, but it remains unharmonized for application on CLL in the clinical laboratory. Objective: We evaluated the performance and clinical applicability of HTS assay for IgHV rearrangement in CLL MRD monitoring in 69 samples from 19 CLL patients treated. Methods: The libraries including IGH locus were performed using the Sequencing Multiplex Kit on IGH consensus primers. To simplify and make automatic the analysis of the data obtained, we developed a specific bioinformatic pipeline that covers from preprocessing to final data summarization and interpretation. The backbone of the analysis includes read preprocessing, mapping against IMGT reference sequences, consensus IgHV reads pairwise alignment to determine mutational status and read classification into rearrangements. Assessment of IgHV mutational status by Sseq, genomic DNA (gDNA; 50-100 ng), were used for IgHV analysis. gDNA was amplified using locus-specific primer sets for IgHV designed to allow for the amplification of all known alleles of the germline IgH sequence, as described previously. Inmunophenotypic studies were performed on erythrocyte-lysed whole PB samples according to Euroflow procedures. PB white blood cells (WBC) was systematically stained with the eight color combination panel recently proposed by the ERIC group for MRD detection (Rawstron AC et al. 2016). Data acquisition was performed on a FACSCanto II flow cytometer Becton-Dickinson Biociences using the FACSDiva software (V8.0; BD). For data analysis, the Infinicyt softwareTM (Cytognos SL, Salamanca, Spain) was used. The MRD levels were reported as fraction of CLL cells of all nucleated cells. MRD negativity was define as a fraction <10 -4 However, for treatments that preferentially clear the blood (for example monoclonal antibodies) the MRD was confirm in the bone marrow as has been recommended. Results: Patient demographics and the results of IGH clonality tests are summarized in Table 1. Interpretable results were obtained with higher sensitivity compared with Sseq at diagnosis stage (19/19 samples), and we are able to determine the same clone at subsequent samples. Among 50 follow-up samples, 44 MRD were positive in both techniques (HTS and FC); 3 follow-up samples, were negative in both determinations (HTS and FC); in contrast, 3 follow-up samples, were negative by FC but positive by HTS (8.3, 17.2, 18.2 samples). This is the case of Patient 8, in which the last follow-up sample, the detection of the primary clone is positive by HTS, while by FC is undetected (Figure 1). In addition, it was possible to detect IgHV clone in all Patient 7 samples. It was under long-term monitoring by HTS and FC, and 17 months after initial diagnosis and first line of treatment (7.5 sample), it was detected a loss of response (Figure 2.). Conclusions: We evaluated an HTS IgHV assay using initial and follow-up samples of 19 CLL patients. Using our pre-designed primer set in library preparation, and developed our specific bioinformatic pipeline that covers from preprocessing to final data summarization and interpretation, we were able to demonstrate that our method was more sensitive than FC in detecting positive follow-up samples and could be able to propose it for MRD monitoring in CLL. However, the prognostic impact of these low-level MRD detected by HTS should be validated with further investigations. Samples were provided by the INCLIVA Biobank. Funded by Gilead Felowship 257/17 Disclosures Ortiz: GILEAD SCIENCES: Research Funding. Terol:Abbvie: Consultancy; Gilead: Research Funding; Janssen: Consultancy, Research Funding; Astra Zeneca: Consultancy; Roche: Consultancy.