Immunoglobulin genes in chronic lymphocytic leukemia: key to understanding the disease and improving risk stratification

While triggering through the B-cell receptor (BcR) facilitates B-cell development and maintenance, it also carries intertwined risks for the emergence of lymphoid malignancies, since malignant B cells can exploit BcR signaling pathways in order to initiate and fuel clonal expansion. Indeed, substantial research into chronic lymphocytic leukemia (CLL), largely based on immunogenetic data, supports the notion that the clonotypic BcR immunoglobulin (IG) engages in the recognition of and selection by putative (auto)antigen.1 This highlights the critical role of the BcR IG in the pathophysiology of CLL and implies that disease development is functionally driven and dynamic, rather than being a simple stochastic process. From a clinical perspective, the remarkable therapeutic efficacy of novel drugs such as ibrutinib and idelalisib which target effectors of the BcR signaling pathway (BTK and PI3K™, respectively), further vouch for this idea, and herald a major paradigm shift which may ultimately lead to changes in the natural history of the disease.2

The IG molecule is an essential component of the multimeric BcR complex and forms a unique genetic identity that is the perfect contender for a clonal marker since it is present from the birth of every B cell onwards, and thus also includes CLL tumor cells, as they derive from activated B cells. Moreover, in contrast to other markers, most notably genomic aberrations, the clonal BcR IG remains stable and unchanged as the disease evolves.3 From the inception of immunogenetics analyses in CLL (Figure 1), reports began to emerge indicating pronounced skewing in IG gene usage and differences from the repertoire of normal B cells, alluding to (super)antigen selection.4 Soon thereafter it was realized that a significant fraction of patients with CLL, approximately 50%, carried somatic hypermutations (SHM) within their BcR IG.4 Additionally, a varying imprint of SHM was seen in clonal BcR IG utilizing different IG heavy variable (IGHV) genes, pointing to functional selection. A further twist in the CLL immunogenetics story was provided by the discovery that the SHM status of the rearranged IGHV gene segregates CLL cases into two broad categories with markedly different outcomes. Cases with no or a limited SHM burden (germline identity (GI) ≥ 98%) constituted “unmutated CLL” (U-CLL), wherein patients generally follow an aggressive disease course with short time-to-first-treatment (TTFT), poor response to chemoimmunotherapy and inferior overall survival, thus starkly contrasting cases with GI < 98% (“mutated CLL”, M-CLL) wherein patients usually have a more indolent form of the disease.65 Within the clinical arena, IGHV mutation burden allowed us to predict the clinical course of the disease based on the number of SHMs within the expressed IG genes. In more recent years, the strongest molecular evidence for antigen selection in CLL emerged from the finding that unrelated patients can carry identical or almost identical BcR IGs, a phenomenon that cannot be attributed to chance alone and is now aptly termed “stereotypy”.74

Figure 1.Historical timeline of immunogenetic studies in chronic lymphocytic leukemia (CLL). IG: immunoglobulin; BcR: B cell receptor; SHM: somatic hypermutation; PFS: progression-free survival; M-CLL: mutated CLL; U-CLL: unmutated CLL; FCR: fludarabine, cyclophosphamide and rituximab; OR: overall response.

The aforementioned stratification of patients based on the SHM status of the clonotypic BcR IG has proved to be one of the most robust prognosticators in CLL,8 superseding the clinical impact of other prognostic markers that may fluctuate or change over time. This division reflects fundamental biologic differences alluding to a different ontogeny for the two mutational groups and, as a consequence, the BcR IG holds much promise and may be central for developing a biologically-oriented prognostication scheme for CLL.108 However, it should be kept in mind that within both M-CLL and U-CLL, a sizeable proportion of cases exhibit clinicobiological behavior that deviates from that associated with its mutational categorization. This highlights the fact that the renowned heterogeneity of CLL persists even after categorization based on the level of SHMs within the IG molecule.119 A classical example is offered by stereotyped CLL subset #2, defined by the expression of a distinctive BcR IG utilizing IGHV3-21/IGLV3-21, which has emerged as a prototype of aggressive disease independently of the SHM load.12109 Indeed, evidence suggests that the immunogenetic subclassification of CLL based on BcR IG stereotypy has clinical potential beyond subset #2, with individual subsets differing significantly in terms of demographics, clinical presentation and the presence or absence of prognostically relevant mutations or cytogenetic aberrations.9

Furthermore, it appears that the particular clinical and cellular background of a genetic lesion, shaped by distinctive signaling through a particular clonotypic BcR IG, may mediate the prognostic or predictive value of recurrent genetic lesions. This idea is exemplified by the finding that within M-CLL, patients harboring trisomy 12 have a TTFT similar to that of patients carrying TP53 aberrations, whereas, in contrast, trisomy 12 is associated with a favorable outcome within U-CLL.13 These findings may explain why trisomy 12 emerges as an intermediate-risk aberration in prognostic indices when the SHM status of the CLL IG is not taken into consideration.14 A similar finding is evidenced when analyzing patients carrying del(17p), with M-CLL cases exhibiting a significantly longer overall survival and TTFT compared to U-CLL patients carrying the same genetic defect (Figure 2). Also worth mentioning is the finding of an asymmetric distribution of certain gene mutations amongst patients bearing distinct immunogenetic features e.g., MYD88 mutations are exclusively found within M-CLL while the vast majority of NOTCH1 mutations are detected within U-CLL.1615 Taken collectively, and bearing in mind that the clinical impact of several biological features is strongly influenced by the SHM status, it is increasingly apparent that definitive conclusions about the clinical implications of any given biomarker should be drawn only after also taking into consideration the SHM status of the rearranged BcR IG. This holds even for well-established prognosticators such as TP53 aberrations.

Figure 2.Kaplan-Meier curves for overall survival (OS) and time-to-first-treatment (TTFT) in CLL patients carrying del(17p). M-CLL harboring del(17p) exhibit significantly longer OS (A) and TTFT (B) compared to U-CLL carrying the same genetic defect. Cases included in this analysis are part of a multi-institutional cohort from our collaborative consortium comprising 8563 CLL patients. TTFT analysis was performed in early stage (Binet A) patients. M-CLL: mutated CLL; U-CLL: unmutated CLL.

Additional support for the pivotal role of immunogenetic analysis in CLL was recently provided by studies demonstrating that IGHV gene SHM status is a strong marker for predicting the response to chemoimmunotherapy, in particular the fludarabine, cyclophosphamide and rituximab (FCR) regimen which is the gold standard treatment for medically fit CLL patients lacking TP53 defects.17 More specifically, M-CLL cases treated with FCR in the context of clinical trials or general practice were independently reported to achieve prolonged responses, often with no detectable minimal residual disease, thus differing significantly from U-CLL cases.2018 Interestingly, upon treatment with newer therapeutic agents such as ibrutinib and idelalisib, CLL patients appear to benefit equally and experience similar overall responses irrespective of the IGHV mutational status;2321 however, the follow-up time is still limited thus precluding definitive conclusions. That said, differences have been noted between the two mutational groups regarding certain clinical parameters following the administration of novel drugs, for example, the initial ibrutinib-induced rise in lymphocyte count and also the duration of lymphocytosis is reported to be greater in M-CLL than that found in U-CLL patients.21 While further investigations into the patterns of lymphocytosis and their association with a clinical response are warranted, these observations may hold value for follow-up assessment.

Altogether, the aforementioned examples strongly indicate that determining the SHM status of the IG molecule is imperative not only for general assessment of the disease course in CLL, but also for guiding treatment decisions; put simply, IG gene analysis should no longer be viewed only as a prognostic test but also as a predictive test for the use of certain therapies. This idea of following an IG-centric model in order to better stratify CLL patients will likely continue to gain value in the near future due to the emergence of novel treatments and the growing concept of precision therapy, and will have a direct impact on the clinical management of patients with CLL.

When broaching the topic of immunogenetic analysis in CLL, what is irrefutable is that the accurate reporting of results obtained from such analyses is paramount, and rigorous standards and meticulous attention to detail are critically important. ERIC, the European Research Initiative on CLL, has been at the forefront of setting standards for immunogenetic research in CLL through pioneering the adoption of good practices by: (i) arranging dedicated educational workshops for the international community; (ii) formulating recommendations for determining the SHM status of IG genes in CLL aimed at harmonizing IG gene sequence analysis in CLL in order to ensure that results are reliable and comparable among different laboratories;2524 (iii) establishing the IG Network, which promotes and advances immunogenetic analysis across the medical community; and (iv) launching a certification system with external quality control, an asset for accreditation of laboratories performing IG analysis.

In conclusion, immunogenetic analysis has proved essential for understanding CLL pathophysiology. We argue that it is equally essential for predicting responses to therapies in this most unpredictable and clinically heterogeneous disease. Indeed, IG-centric risk stratification appears more appealing and relevant today now that signaling inhibition has emerged as a powerful, non-chemotherapeutic approach towards eventually curing CLL, a still incurable disease.

Acknowledgements

This work was supported in part by the Swedish Cancer Society, the Swedish Research Council, the Lion’s Cancer Research Foundation, the Marcus Borgström Foundation and the Selander’s Foundation, Uppsala; H2020 “AEGLE, An analytics framework for integrated and personalized healthcare services in Europe” by the EU; “MEDGENET, Medical Genomics and Epigenomics Network” (H2020, No.692298) by the EU; research grant AZV-MZ-CR 15-30015A-4/2015, Bloodwise, UK (grants 16003 and 14037), GCH-CLL: funded by the General Secretariat for Research and Technology (GSRT) of Greece and the Italian Ministry of Health (MoH) and IMI2 “HARMONY” funded by the EU. AWL receives unrestricted research grants from Roche-Genentech.