Differential Impact of Signaling Inhibitor Treatment Versus Standard Chemoimmunotherapy on Chronic Lymphocytic Leukemia T Cells: Clinical Implications

Studies by us and others have documented the clonal architecture of the T cell repertoire in chronic lymphocytic leukemia (CLL), with T cell clones which persist and expand over time in untreated patients, appear disease-specific and may be shared among different patients, thereby indicating selection by restricted antigens. Here, we investigated T cell repertoire dynamics in CLL after different types of treatment by employing high-throughput next-generation sequencing in order to comprehensively assess changes in relation to the type of treatment as well as the clinical response. We analyzed 42 pre- and 3-month post-treatment blood samples from 16 CLL patients who received (i) standard chemoimmunotherapy (FCR regimen, n=5), (ii) ibrutinib (IB, n=9), and/or (iii) rituximab-idelalisib (R-ID, n=7). At the post-treatment sampling, all patients who received FCR had achieved complete remission (CR), and all patients on IB or R-ID had achieved partial remission (PR). Two patients, one on IB and one on R-ID, who later achieved CR were analyzed at that timepoint, as well. We excluded patients who developed rituximab-related late-onset neutropenia after FCR, since it is known to be mediated by cytotoxic T-cell clones. Starting material was PB mononuclear cells. TRBV-TRBD-TRBJ gene rearrangements were RT-PCR amplified and subjected to paired-end NGS. Raw NGS reads were processed through a previously published, purpose-built bioinformatics pipeline designed for paired-read stitching, sequence curation, clonotype computation and repertoire analysis. Only productive TRBV-TRBD-TRBJ rearrangements were included in the analysis (n= 7,494,236, 85.3% of filtered-in sequences, median 178,063/sample). For repertoire analysis, clonotypes (i.e. TRB rearrangements with identical TRBV gene usage and amino acid complementarity-determining region 3 sequence) were considered (median 11,290 distinct clonotypes/sample). The TRBV gene repertoire remained relatively stable after treatment across all treatment groups. All cases displayed significant clonal T cell expansions both pre- and post-treatment (median cumulative frequency of the 10 most expanded T cell clonotypes/sample 30.3% versus 38.6%, respectively). However, differences were noted between the three treatment subgroups. More specifically, clonality (measured as the median cumulative frequency of the 10 most expanded clonotypes/sample) significantly increased after FCR (20.8% to 39.0%, p=0.03) and R-ID (33.0% to 41.1%, p=0.001), whereas it tended to decrease after IB (36.1% to 31.0%, p=0.34). Notably, achieving deeper clinical response (from PR to CR) correlated with a significant increase of clonality in the R-ID case (41.0% to 56.2%), but had no impact in this respect for the IB case (48.3% to 48.1%). The increased clonality in the R-ID case culminated from further expansion of 5 major clonotypes which also dominated the pre-treatment repertoire, suggesting that they may correspond to T-cell clones with anti-tumor properties activated by treatment. Overall, FCR resulted in reconstitution of the T-cell repertoire (median of 2/10 most frequent T-cell clones pre-treatment remained significantly expanded post-treatment), whereas pre-treatment major clones persisted after both IB and R-ID (8/10 and 7/10, respectively). In conclusion, NGS immunoprofiling in CLL documents the differential impact of various treatments on T cells. Particularly, chemoimmunotherapy increases T cell clonality likely through an ablative mechanism, in contrast to signaling inhibitors which retain T cell clones that may have developed in response to tumor antigens and possibly activate them, at least as indicated for the R-ID case, with obvious clinical implications.