Rare germline ATM variants predispose to secondary cancer in chronic lymphocytic leukaemia patients

Abstract

Chronic lymphocytic leukaemia (CLL) has one of the strongest familial risks of all cancers, as evidenced by the eight-fold increased risk seen in relatives of CLL patients, yet much of the heritable risk remains unexplained [1]. Patients with CLL also have a high rate of secondary cancer, i.e., the development of second primary malignancy, compared to the general population, which may be explained by immune dysregulation due to CLL and/or its treatment, but also by environmental and genetic risk factors [2]. The development of second cancer in patients with CLL contributes to higher morbidity in these patients [2]. As the population of long-term CLL survivors expands due to novel agents used in therapy, the identification of patients with risk of developing the second cancer may help to improve longevity of CLL patients.

In this monocentric study, we aimed to assess whether patients with CLL carrying rare germline variants in ATM have a higher risk of familial CLL or a higher risk of cancer in first-degree relatives than patients without these variants. Next, we aimed to assess whether patients carrying rare ATM variants develop secondary cancer or develop CLL at an earlier age than patients without these variants, and whether these variants influence the time to first treatment (TTFT) and overall survival (OS) in patients with CLL. In this large cohort, we also sought to confirm our previous observation that rare variants in ATM are associated with the development of IGHV-unmutated CLL [3], which has a worse prognosis compared to IGHV-mutated disease.

A total of 629 patients with CLL were included in the study with a median follow-up of 6 years (range, 0.08-37.80; Supplementary Materials and Methods, Supplementary Table S1). Ten (1.6%) of these patients carried rare pathogenic (P) or likely pathogenic (LP) variants in ATM and 18 (2.9%) carried rare variant of uncertain significance (VUS) in ATM, all heterozygous (Supplementary Table S2, Figure 1). Patients harbouring P/LP ATM variants had a higher risk of secondary cancer (50.0% of patients; relative risk [RR], 2.88; 95% confidence interval [CI], 1.51-5.47; P = 0.001) than those without these variants (17.0%). Similarly, patients harbouring rare VUSes in ATM had a higher risk of secondary cancer (46.0% of patients; RR, 2.56; 95% CI, 1.48-4.41; P = 0.001) than those without these variants (17.0%). Half (7/14) of patients who carried these germline ATM variants and had secondary cancer were first diagnosed with CLL, while the other half (5/9) were first diagnosed with another malignancy. Secondary cancers were always of a different type, with the exception of prostate cancer, which occurred in 2 patients.

Regarding age at diagnosis of first malignancy in patients with secondary cancer, there was no significant difference between patients carrying rare P/LP ATM variants (median age 62 years) compared to patients without these variants (63 years), and the same was true for age at CLL diagnosis (median age 65 years versus 64). As for the group of patients carrying VUSes in ATM, there was similarly no significant difference in age at diagnosis of the first malignancy or at CLL diagnosis (Supplementary Table S1).

We then looked at whether rare P/LP ATM variants were associated with the familial CLL. In our cohort, 3% of patients (18/629) were found to have CLL in first-degree relatives as assessed from patient notes, but none were found to carry rare ATM variants. Regarding prevalence of cancer in first-degree relatives, patients harbouring P/LP ATM variants had a higher risk of cancer in their relatives (70.0% of patients; RR, 1.55; 95% CI, 1.02-2.34; P = 0.040) than those without these variants (45.0%). There was no significant difference in the prevalence of lymphoproliferative diseases between patients with P/LP ATM variants and patients without these variants.

We were further interested in whether rare ATM variants influence CLL clinical outcomes, specifically TTFT and OS. Neither median TTFT nor OS differed significantly between patients with and without P/LP rare ATM variants (Supplementary Figure S1). Regarding Richter transformation (RT), a higher risk of RT was observed in patients carrying rare P/LP ATM variants (20.0%; RR, 6.01; 95% CI, 1.62-122.33; P = 0.007) compared to patients without these variants (3.3%). We further confirmed previous observations by us and other authors [3, 4] that the presence of rare P/LP ATM variants was associated with the acquisition of del(11q) during the course of CLL (60.0%; RR, 2.73; 95% CI, 1.61-4.62; P = 0.002).

Finally, in this large cohort, we confirmed a higher risk of developing IGHV-unmutated disease in patients with P/LP rare ATM variants (90.0% of patients; RR, 1.62; 95% CI, 1.30-2.02; P < 0.001) compared to patients without these variants (56.0%).

Our study investigating the impact of rare P/LP ATM variants in CLL did not find an association of these variants with familial CLL, further confirming observations in individual families with CLL [5, 6]. This finding suggests that rare ATM variants are not responsible for the hereditary component of CLL, but rather increase the risk of cancer in general, as shown by the association with the higher prevalence of secondary cancers in our patients with CLL. An increased risk for secondary cancer has been associated with CLL [2] and our study identified rare P/LP/VUS ATM variants as a predisposing factor for secondary cancer in CLL and patients who carry rare ATM variants would benefit most from cancer prevention screening.

Regarding clinical outcomes of CLL patients, our study did not find an association of rare P/LP ATM germline variants with earlier age at CLL diagnosis, nor with reduced TTFT or OS. Similarly, previous studies did not observe an effect of rare ATM germline variants on OS [7] and TTFT [8] in CLL, whereas the latter study found that CLL patients with ATM germline variants were younger at diagnosis. Consistent with previous reports [4, 7], rare ATM variants were found to be associated with the acquisition of del(11q) during the course of CLL as a second hit in the remaining allele. A higher risk of RT was observed in patients carrying rare P/LP ATM variants and this finding deserves further research. Also, a higher risk of cancer was observed in first-degree relatives of patients carrying P/LP ATM variants.

Importantly, we confirmed in a large cohort our previous observation of the association of rare ATM variants with increased risk of IGHV-unmutated CLL [3]. ATM has a specific role in developing lymphocytes: it is required for the processes of V(D)J recombination and somatic hypermutation and later in immunoglobulin class switch recombination [9, 10]. How exactly P/LP germline ATM variants interfere with these processes deserves further investigation.

We acknowledge that our study has limitations. First, the sample size of patients with rare ATM variants is modest, similar to other studies on rare germline variants [4, 7, 8]. Of note, this study used the largest CLL cohort with complete clinical data to study rare ATM variants in CLL to date. Second, variant classification in this study is based on the current ClinVar evaluation and may be reclassified based on future research.

In conclusion, the significance of rare P/LP ATM variants in CLL appears to lie in the higher susceptibility to CLL in carriers of rare ATM variants, and specifically to the prognostically poor form of CLL with unmutated IGHV status, and in the threefold higher risk of secondary cancer, either before or after CLL diagnosis.

Anna Petrackova and Eva Kriegova designed research. Jirina Manakova and Romana Nesnadna performed experiments. Zuzana Kubova and Tomas Papajik collected patient samples and clinical characteristics. Anna Petrackova wrote the manuscript. Eva Kriegova critically revised the manuscript. All authors read and approved the final manuscript.

The authors declare that they have no competing interests.

This study was supported by the project JG_2024_035 implemented within the Palacký University Young Researcher Grant, by the Internal grant agency of Palacky University (IGA_LF_2025_014), and in part by Ministry of Health of Czech Republic (MH CZ–DRO (FNOL, 00098892)).

All patients provided written informed consent about the usage of biological materials for the purpose of this study, which was performed in accordance with the Helsinki Declaration and approved by the Ethics Committee of the University Hospital and Palacký University Olomouc (permit number: ID16-32339A).