Biological diversity and risk-adapted treatment of chronic lymphocytic leukemia

Chronic lymphocytic leukemia (CLL) is the most common adult leukemia in the western world. Unlike other forms of leukemia, the proliferative component of CLL is relatively small, but recent evidence has pointed to the existence of a proliferative pool, which can be surprisingly sizeable.1 There are several additional features of CLL that set it apart from other cancers. The most prominent pathogenic mechanisms include: (i) genomic aberrations targeting critical genes (e.g. miRNA, TP53, ATM); (ii) antigen drive and stereotyped B-cell receptors (BCR); and (iii) microenvironmental stimulation.2,3 While the precise sequence of events is currently unclear, our growing understanding of CLL biology is enabling translation into clinical practice.

The papers by Marincevic et al.4 and Giné et al.5 in this issue of Haematologica touch on two important issues in CLL biology as well as clinical care and thus significantly expand our conception of aspects of this disease.

Table 1.Summary of current prognostic factors and potential treatment options in patients with chronic lymphocytic leukemia and an indication for first-line treatment.

Biological diversity of chronic lymphocytic leukemia

Different levels of somatic hypermutation of the IGHV (and IGKV) divide patients with CLL into two clinical subgroups with different prognoses (IGHV mutated and unmutated CLL).6,7 In addition to variable mutations in the IG variable regions, CLL display a remarkably biased IGHV gene repertoire (e.g. IGHV1-69, IGHV4-34, IGHV3-23 and IGHV3-21)8 and several groups have reported multiple CLL subsets with similar BCR often arising from the use of common H and L chain V region gene segments that share CDR3 structural features. These ‘stereotyped’ BCR occur in up to 30% of patients.2 The striking degree of structural restriction of the entire BCR in CLL suggests that common antigens are recognized by CLL cells and support the contribution of an antigen-driven process.

Certain stereotyped BCR have also been indicated to have distinct clinical features. For instance, IGHV4-34/IGKV2-30 patients with stereotyped BCR appear to have an indolent disease course compared to that of CLL patients with non-stereotyped IGHV4-34. Patients with IGHV3-21 with either stereotyped or non-stereotyped BCR have been suggested to have an inferior overall survival, independently of IGHV mutational status.9

In the work presented in this issue of the Journal, Marincevic et al.4 assessed genomic aberrations by 250k single nucleotide polymorphism arrays to discover imbalances of genetic aberrations between the different subsets (V3-21; V4-34). While the numbers of cases is too low to draw firm conclusions, a pattern is emerging, which continues on the theme of earlier work showing a high incidence of high-risk genetics (11q-) in cases with unmutated IGHV and low-risk aberrations (13q- single) in cases with mutated IGHV.10 A similar scenario is described in the analysis by Marincevic et al., in which low-risk aberrations were present in IGHV4-34/IGKV2-30 patients with stereotyped BCR (subset #4), while higher risk aberrations (11q deletion and potentially increased numbers of aberrations) were more common in IGHV3-21 patients. While the findings are interesting and raise a number of challenging questions concerning the sequence of events in CLL pathogenesis, the different groups with stereotype subsets (e.g. IGHV3-21) were quite small. In a recent analysis from the UK LRF CLL4 trial, 40/532 patients had IGHV3-21 usage (0.75%). This suggests that particular emphasis should be paid to the identification of surrogate markers for particular IGHV usage groups and stereotyped BCR, which may be a prerequisite for broader clinical application of the findings.

The biological diversity of CLL is also exemplified in a clinically very relevant study presented in this issue of the journal by Giné et al.,5 who analyzed tissue biopsies from 100 patients with CLL. The biopsies were taken because of the suspicion of disease transformation. High grade transformation of CLL remains one of the most difficult to treat “complications” of CLL and we are only slowly identifying predisposing factors and underlying mechanisms. One pattern emerging is that cases in which the malignant clones are unrelated based on IGHV usage appear to behave more like de novo diffuse large B-cell lymphoma (DLBCL), while clonally related cases may have a dismal prognosis with survival in the range of months.

In the analysis by Giné et al., the biopsies showed histological transformation to DLBCL in 22% of cases. In the remaining 78% patients, the presence of expanded proliferation centers (judged by their size and Ki-67) predicted poor outcome. Based on this assessment, 23% of patients were considered to have “accelerated” CLL defined based on expanded proliferation centers, a mitosis count of greater than 2.4 or Ki67 greater than 40% per proliferation center. The median survival from biopsy of patients with “non-accelerated” CLL, “accelerated” CLL and transformed DLBCL were 76, 34, and 4.3 months, respectively. While the pathogenic mechanism underlying the evolution to accelerated CLL remains to be identified, the study by Giné et al.5 is of practical relevance because it clearly points to the clinical importance of the newly described category of “accelerated” CLL.

Risk-adapted treatment in chronic lymphocytic leukemia

Over recent years our better understanding of genetic and biological groups in CLL has improved risk stratification of patients with this disease. The ultimate goal of risk-adapted approaches in cancer is genotype- or risk factor-adapted therapy in all patients. This depends on the identification of the prognostic and predictive factors of strongest clinical impact. It is important to point out that this is not necessarily identical to what is generally considered a “strong” prognostic marker based on statistical assessments. The most powerful prognostic factors in CLL include age, Binet or Rai stage, serum markers (β2-microglobulin and thymidine kinase), and genetic factors (genomic aberrations, IGHV mutation status, and TP53 mutations). Nonetheless, even the most reliable prognostic markers may not immediately translate into practice changes. While the IGHV mutation status reliably identifies two groups of patients of roughly similar size, this information is not currently changing treatment approaches outside clinical trials. However, future treatment strategies may target deregulated signaling pathways specifically in IGHV unmutated (or mutated) CLL.

In contrast, the first genotype-specific treatment for CLL patients has been developed for cases with 17p deletion who have a very poor prognosis (median survival of less than 2 years from first treatment indication) with alkylator and nucleoside-based chemo(immuno)therapy (chlorambucil, fludarabine, fludarabine + cyclophosphamide, or fludarabine + cyclophosphamide + rituximab).11 Since a number of agents have been shown to act independently of functional p53 in CLL, current treatment approaches in clinical trials use these agents upfront with early allogeneic stem cell transplantation as consolidation after remission has been achieved in eligible patients. In addition to CLL with 17p deletion, the group of patients with TP53 mutations (even in the absence of 17p deletion) is a subgroup gaining increasing attention.12 TP53 mutations are found in 8–12% of patients with an indication for first-line treatment. The incidence increases during the course of disease and was 37% in a cohort of fludarabine-refractory cases.13 In a recent study within the German CLL4 study (which compared fludarabine and fludarabine + cyclophosphamide treatment), the incidence of TP53 mutations in the absence of 17p deletion was similar to that of 17p deletions and the clinical course of patients with these forms of CLL was very similar.14 These data –as well as information from a number of retrospective cohort studies – suggest that TP53 mutation should be added to the diagnostic work-up of patients with CLL in need of treatment.

Table 1 summarizes different risk categories for CLL patients and potential treatment approaches. Patients with 17p deletion, TP53 mutation or fludarabine-refractory CLL have a very short overall survival once treatment is indicated.14,15 These patients are prime candidates for investigational approaches even if previously untreated. In addition, fit patients are candidates for allogeneic stem cell transplantation.16

Based on a number of studies including prospective trial data,17–19 patients with unmutated IGHV, 11q deletion, V3-21 usage and high levels of serum markers (e.g. β-2 microglobulin) form a high-risk group with a median survival in the range of 53 months following an indication for first-line treatment. These patients are now generally treated with the combination of fludarabine + cyclophosphamide + rituximab (if fit and younger), but future trials may consider maintenance strategies in such patients.20

The group of patients with none of the above aberrations and mutated IGHV status have a very favorable outcome and could even be considered for de-escalation studies.

In spite of this clinically relevant risk hierarchy, the decision to treat is currently not based on the risk profile but on symptomatic disease.21 This is important and further supported by the observation that in some subgroups of patients, such as those with 17p deletion (and mutated IGHV), the disease may have an indolent course.22

While risk-adapted treatment approaches are generally used interchangeably with different treatments for certain prognostic risk groups, future risk-adapted treatment strategies will undoubtedly also have to integrate the patient’s performance status, co-morbidities and age, which is of particular importance considering that the median age at diagnosis is over 70 years old.

Footnotes

• Thorsten Zenz is a Haematooncologist at the Department of Internal Medicine III, Ulm University, Germany. Daniel Mertens is a group leader at the Department of Internal Medicine III, Ulm University, Germany, and the German Cancer Research Center, Heidelberg, Germany. Stephan Stilgenbauer is the Vice-Chairman of the Department of Internal Medicine III, Ulm University, Germany.
• ( Related Original Article on pages 1519 and 1526)
• Financial and other disclosures provided by the authors using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org.