In this study, we have evaluated BIRC3 and ATM deletions in CLL patients in order to analyze their frequency and association with prognostic factors of the disease. BIRC3 gene belongs to the IAP (Inhibitor of apoptosis) family; it is mainly expressed in lymphoid tissue, particularly in spleen and peripheral blood lymphocytes. It is a negative regulator of the non-canonical NF-kB (Nuclear factor kappa B) signaling pathway [27, 28]. BIRC3 protein cooperates in the same complex with TRAF2 and TRAF3 (tumor necrosis factor receptor-associated factors 2 and 3) that negatively regulates MAP3K14 (mitogenactivated protein kinase 14), a central activator of non-canonical NF-kB pathway. Furthermore, it is involved in the maintenance of TP53 levels, through MDM2 (Mouse Double Minute 2, Human Homolog) modulation and, its suppression contributes to neoplastic progression [29]. BIRC3 mutations/deletions are infrequently detected in CLL at diagnosis (4%) but they reached 24% in fludarabine-refractory patients [12, 30]. On the other hand, ATM gene, a member of the PI3K (phosphatidylinositol-3 kinase) family, has a key role in protect the genome integrity by regulating the cell-cycle, preventing the DNA damage, activating DNA-repair pathways and inducing apoptosis if the DNA damage cannot be repaired [31]. ATM deficiency allows the accumulation of genetic alterations related to genomic instability and is associated with a predisposition to lymphoid malignancies [32, 33]. In addition, ATM is a p53 regulator and its inactivation is associated to p53 dysfunction [34].
The analysis of our data showed, in agreement with previous results [11, 14, 33, 35], an increased presence of large deletions in 11q22, given that 84.8% of cases had BIRC3 and ATM losses. Among them, 3 patients showed higher frequency of ATM deletion than BIRC3 loss and 7 cases had higher frequency of BIRC3 deletion than ATM loss, indicating the presence of two separated deletion events, resulting in clonal evolution. Alhourani et al [14] also observed a similar finding in a small number of cases. It is also interesting to note the presence of 3 patients with only BIRC3 deletion and 2 cases with only ATM loss. In concordance with our data, Rose-Zerilli et al [13] found one case with BIRC3 deletion without concomitant loss of ATM. These findings show the heterogeneity of del(11q) and suggest the importance to consider the inclusion of the analysis of BIRC3 deletions in the routine laboratory screening of CLL patients. Besides, our cohort showed a high frequency of CK, confirming previous reports [36, 37] and reflecting the high complexity observed in CLL cases with del(11q) as well as their influence in progression of the disease and poor outcome [19, 35, 38].
Concerning FISH analysis, 48.5% of cases showed two or more abnormalities, being del(13q) the most frequent, followed by del(17p) and trisomy 12. Data of the literature support our findings for del(13q) and trisomy 12 [12, 13, 17, 32], but there are controversial results about del(17p). Some reports showed low frequency of del(17p) in cases with del(11q) [13, 14, 16], suggesting that both alterations cannot coexist [12, 34]. On the contrary, other authors found association of monoallelic deletions in both genes in 25.5%-43% of cases [15, 17], supporting the co-occurrence of these alterations in CLL patients, which was associated with more aggressive disease and poor outcome [16, 17]. Simultaneously, 54.5% of cases showed intratumor heterogeneity, phenomenon related to genomic instability and disease progression [23, 39]. Although it is not clear the exact mechanism involved in the generation of intratumor heterogeneity, different reports suggest that it could be originated by genetic disorders within the tumor cells and/or under the influence of changes in the tumor microenvironmental conditions [22, 40]. Interestingly, Yi et al [23] found that the integrated analysis of the number of cytogenetic alterations and the intratumor cytogenetic heterogeneity provide a better information for prognostic stratification of CLL patients than the conventional model [3, 4]. More studies will be necessary to confirm these results.
Regarding the association with prognostic factors, our series showed 72.7% of patients with U-IGHV [7, 20, 36], intermediate value between those found in previous studies (68-89%) [17, 33, 41, 42]. Simultaneously, a significantly shorter TTFT in patients with del(11q) compared to cases with del(13q) single and patients without cytogenetic and FISH alterations was observed [3, 4, 42, 43]. Even though the introduction of ibrutinib therapy has improved the clinical outcome of patients with del(11q) [44, 45], recent studies [30, 46] showed the lack of response to ibrutinib but not to venetoclax in CLL cells with BIRC3 disruption, making its detection relevant for treatment decisions. In addition, Asslaber et al [47] found that patients with low BIRC3 expression had rapid disease progression and short TTFT, associated to an altered NK-kB pathway, upregulation of antiapoptotic BCL-2 family members, and high sensitivity to venetoclax treatment in vitro. More recently, different authors reported the presence of BIRC3 gene inactivation (~5% of cases) [48, 49], suggesting that these patients would represent a subgroup with the worst outcome following initial chemotherapy treatment, reinforcing the importance to identify this abnormality in order to refine the risk stratification of CLL patients and highlighting the benefit of novel agents-based therapy.
Our study has some limitations, as the low number of our cohort and the retrospective nature of the analysis. Despite these limitations, our analysis shows concomitant ATM and BIRC3 deletions, with a number of patients with different clonal frequency indicating the presence of clonal evolution, as well as cases with only ATM or BIRC3 deletions, suggesting the importance to consider the inclusion of BIRC3 analysis in the routine FISH laboratory screening of CLL patients. In addition, a strong association between ATM and BIRC3 deletions with TP53 loss, U-IGHV, complex karyotypes, and intratumor heterogeneity was observed, highlighting the increased genomic instability present in this group of patients.