Abstract
Background
Germline pathogenic variants (PVs) in the ataxia-telangiectasia mutated (ATM) gene are established moderate-risk factors for breast cancer (BC), however, population-specific variant spectra and the clinical significance of many missense variants remain incompletely characterized.
Aims
To evaluate the prevalence of ATM variants in a large cohort of patients with BC from North Macedonia and compare it with that in the general population, with a particular focus on the frequency of the p.(Leu2492Arg) variant and its distribution relative to global genomic datasets.
Study Design
Retrospective case–control study.
Methods
ATM variants were analyzed in 1,211 patients with BC from North Macedonia using a targeted hereditary cancer gene panel. These findings were compared with those from 1,303 population-based controls analyzed by clinical exome or whole-exome sequencing.
Results
Pathogenic ATM variants were identified in 1.9% of BC cases and 0.4% of controls, indicating a significantly increased risk of BC [odds ratio (OR) = 5.02, p = 0.0006]. Most PVs were protein-truncating, with six recurrent variants accounting for over 70% of detections, suggesting regional enrichment. Carriers showed a significantly higher prevalence of human epidermal growth factor receptor 2-positive tumors (OR = 2.92, p = 0.0189). Variants of uncertain significance were observed at comparable frequencies in cases and controls. The p.(Leu2492Arg) missense variant was more frequently detected in cases than in controls (1.9% vs. 1.1%; OR = 1.78, p = 0.086) and exhibited a markedly higher allele frequency in this population than in global databases.
Conclusion
These findings confirm ATM as a clinically relevant BC susceptibility gene in North Macedonia and highlight the population-specific enrichment of both PVs and the p.(Leu2492Arg) missense variant. The results emphasize the importance of using population-matched controls and regional genomic data for accurate risk assessment and variant interpretation.
INTRODUCTION
The ataxia-telangiectasia mutated (ATM) gene, located on chromosome 11q22.3, encodes a serine/threonine protein kinase that is essential for maintaining genomic stability.1 Comprising 63 exons, ATM acts as a central regulator of cell cycle checkpoint signaling pathways. It is primarily activated in response to DNA double-strand breaks, which represent one of the most cytotoxic forms of DNA damage.2 ATM activation is mediated by the MRE11–RAD50–NBS1 complex, after which it phosphorylates multiple downstream targets, including p53, BRCA1, CHEK2, and H2AX. This signaling cascade coordinates cell cycle arrest, DNA repair, or apoptosis, thereby preventing the accumulation and propagation of genomic damage.3 Pathogenic variants (PVs) in these genes impair DNA repair mechanisms and increase cancer susceptibility.4 Beyond its role in genome maintenance, ATM also influences several physiological processes, including neurological function, immune response, and aging. Biallelic PVs in ATM cause ataxia-telangiectasia (A–T), a rare autosomal recessive disorder characterized by progressive neurodegeneration, immunodeficiency, radiosensitivity, premature aging, and a markedly increased risk of cancer, particularly lymphoid malignancies.5 Although A–T is inherited in a recessive manner, heterozygous carriers of pathogenic ATM variants also have an elevated risk of cancer, particularly breast, prostate, and pancreatic cancers.6 Importantly, cancer risk varies considerably depending on the specific ATM variant. Protein-truncating variants (PTVs) are generally associated with a two- to four-fold increased risk of breast cancer (BC),7 whereas the clinical significance of many missense variants remains uncertain.8 BCs associated with ATM PVs often exhibit distinct clinical features, including hormone receptor positivity, poor differentiation, and increased lymph node involvement.7ATM variants may also influence treatment response, particularly in the context of radiotherapy, where variant-specific effects on contralateral BC risk have been reported.9 Despite growing evidence linking ATM variants to BC, the clinical interpretation of many ATM missense variants remains challenging. Furthermore, although ATM-associated BC risk has been extensively studied, populations from Southeastern Europe remain underrepresented in global genomic datasets. In this study, we analyzed ATM germline variants in a large cohort of patients with BC from North Macedonia and compared them with population controls. We estimated the prevalence of PVs and evaluated the frequency of the c.7475T > G, p.(Leu2492Arg) variant relative to global datasets. These findings provide insight into regional variant distribution and contribute to the improved interpretation of ATM variants in an underrepresented population.
MATERIALS AND METHODS
The case cohort included 1,211 patients diagnosed with invasive BC between 2009 and 2025. Clinical and demographic data were obtained for all patients. To estimate the carrier frequency of ATM PVs in the general population, we analyzed sequencing data from 1,303 individuals referred for genetic testing for pediatric and non-cancer conditions. The control cohort had a median age of 19 years and included 67.2% Macedonians, 25.3% Albanians, and 6.2% individuals of other ethnicities, with 52.5% males. In comparison, the BC cohort (median age, 47 years) consisted entirely of females and was predominantly Macedonian (83%). The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Subcommittee of the Macedonian Academy of Science and Arts for Medicine, Pharmacy, Veterinary Medicine and Dentistry (approval number: 03-203/4, date: 10.12.2024), and written informed consent was obtained from all participants. Genomic DNA was extracted from peripheral blood samples using standard protocols. Targeted sequencing of 94 cancer-associated genes, including ATM in the BC cohort, was performed using the TruSight Hereditary Cancer Panel (Illumina Inc., San Diego, CA, USA). Sequencing was conducted on the MiSeq platform using paired-end 150-bp reads. Bioinformatic analysis was performed in-house and included read alignment using BWA v0.7.15,10 variant calling with GATK v3.8,10 VCF processing with bcftools v1.9,11 and variant annotation using Ensembl Variant Effect Predictor v112,12 all based on the hg19 reference genome. Control samples were processed using either the TruSight One clinical exome panel (n = 290) or the Twist Human Core + RefSeq + Mitochondrial WES panel (n = 1,013), with sequencing performed on the MiSeq or NovaSeq 6000 platforms, respectively. All control data were analyzed using an identical in-house pipeline aligned to the hg38 “no_alt” reference genome, as previously described.13 Variant coordinates derived from hg19 were converted to hg38 using the UCSC LiftOver tool prior to allele frequency comparison. Variants within coding regions and ± 25 bp of flanking intronic sequences were evaluated, and ATM was fully covered in both library enrichment approaches. Variant visualization was performed using Integrative Genomics Viewer (IGV), and variant classification followed American College of Medical Genetics and Genomics guidelines.14 Variant nomenclature followed Human Genome Variation Society recommendations and was based on the ATM reference transcript NM_000051.4, with genomic coordinates corresponding to the GRCh38/hg38 reference genome assembly. Sanger sequencing was used to confirm variants in cases with inconclusive or low-confidence NGS results. Copy number variant (CNV) analysis (exon-level deletions/duplications) was not performed in this study.
RESULTS
Pathogenic ATM variants were identified in 23 BC cases (23/1211, 1.9%) and 5 controls (5/1303, 0.4%). ATM PVs were significantly associated with an increased risk of BC [odds ratio (OR) = 5.02, 95% confidence interval (CI): 1.90-13.26, p = 0.0006]. In total, 14 distinct ATM PVs were detected, the majority of which were PTVs. Six PVs were recurrent and detected in multiple individuals, collectively accounting for 71.4% (20/28) of all PV carriers in the study population (Table 1). Two PVs were observed exclusively in controls: c.2250G>A, p.(Lys750=), and c.8147T>C, p.(Val2716Ala). All identified PVs have been previously reported in ClinVar and classified as pathogenic or likely pathogenic. Patient-level clinicopathological data for BC patients carrying ATM PVs are summarized in Table 2.
TABLE 1. Frequency of Pathogenic ATM (NM_000051.4) Variants Among Breast Cancer Patients and Population Controls.
|
HGVSc |
HGVSp |
dbSNP |
ClinVar ID |
Effect |
AF_gnomAD |
BC cases |
Controls |
|
c.67C > T |
p.(Arg23*) |
rs746235533 |
232248 |
Stop gained |
3.72E-06 |
6 |
/ |
|
c.495_496 + 16del |
p.? |
rs1555059522 |
487452 |
Splice donor |
6.20E-07 |
2 |
/ |
|
c.1066-1G > A |
p.? |
rs876660038 |
232870 |
Splice acceptor |
/ |
3 |
1 |
|
c.1564_1565del |
p.(Glu522Ilefs*43) |
rs587779817 |
127340 |
Frameshift |
5.45E-05 |
1 |
2 |
|
c.2250G > A |
p.(Lys750=) |
rs1137887 |
3044 |
Splice, synonymous |
4.23E-05 |
/ |
1 |
|
c.3576G > A |
p.(Lys1192=) |
rs587776551 |
3035 |
Splice, synonymous |
2.48E-06 |
3 |
/ |
|
c.3603del |
p.(Phe1201Leufs*6) |
rs1057517129 |
371256 |
Frameshift |
/ |
1 |
/ |
|
c.3866del |
p.(Lys1289Argfs*4) |
/ |
2111936 |
Frameshift |
/ |
1 |
/ |
|
c.5005G > A |
p.(Glu1669Lys) |
rs1591693095 |
800343 |
Missense, splice |
/ |
1 |
/ |
|
c.6115G > A |
p.(Glu2039Lys) |
rs864622251 |
219787 |
Missense |
2.48E-06 |
1 |
/ |
|
c.6889dup |
p.(Gln2297Profs*76) |
/ |
2431274 |
Frameshift |
/ |
1 |
/ |
|
c.8147T > C |
p.(Val2716Ala) |
rs587782652 |
142700 |
Missense |
2.73E-05 |
/ |
1 |
|
c.8283_8284del |
p.(Gln2762Alafs*6) |
rs775899653 |
482713 |
Frameshift |
/ |
2 |
/ |
|
c.9139C > T |
p.(Arg3047*) |
rs121434219 |
3029 |
Stop gained |
1.61E-05 |
1 |
/ |
AF, allele frequency; BC, breast cancer; SNP, single nucleotide polymorphism.
TABLE 2. Patient-Level Clinicopathological Data of Breast Cancer Patients Carrying Pathogenic ATM Variants.
|
No. |
Patient ID |
HGVSc |
HGVSp |
Effect |
Ethnicity |
Age at diagnosis |
Family history for BC/OC |
Localization |
BC type |
T |
N |
G |
Stage |
ER pos. |
PR pos. |
HER2 pos. |
TN pos. |
|
1 |
BC-255 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
34 |
Yes |
Unilateral |
Ductal |
1c |
1a |
3 |
II A |
No |
No |
No |
Yes |
|
2 |
BC-1767 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
40 |
No |
Unilateral |
Ductal |
2 |
0 |
NA |
IIA |
Yes |
Yes |
Yes |
No |
|
3 |
BC-1997 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
54 |
No |
Unilateral |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
|
4 |
BC-2230 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
51 |
Yes |
Unilateral |
Lobular |
2 |
1a |
2 |
IIB |
Yes |
Yes |
No |
No |
|
5 |
BC-2416 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
48 |
Yes |
Unilateral |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
|
6 |
BC-2871 |
c.67C > T |
p.Arg23* |
Stop gained |
MK |
54 |
Yes |
Unilateral |
Ductal |
2 |
1 |
3 |
IIB |
Yes |
Yes |
NA |
No |
|
7 |
BC-577 |
c.495_496 + 16del |
p.? |
Splice donor |
AL |
33 |
No |
Bilateral |
Ductal |
4b |
0 |
3 |
IIIC |
Yes |
Yes |
Yes |
No |
|
8 |
BC-1396 |
c.495_496 + 16del |
p.? |
Splice donor |
AL |
61 |
Yes |
Unilateral |
Ductal |
2 |
1 |
2 |
IIB |
Yes |
Yes |
No |
No |
|
9 |
BC-656 |
c.1066-1G > A |
p.? |
Splice acceptor |
MK |
28 |
No |
Unilateral |
Lobular |
2 |
3 |
3 |
IIIC |
Yes |
Yes |
No |
No |
|
10 |
BC-1643 |
c.1066-1G > A |
p.? |
Splice acceptor |
MK |
32 |
Yes |
Unilateral |
Ductal |
1c |
2a |
3 |
IIIA |
Yes |
Yes |
No |
No |
|
11 |
BC-2415 |
c.1066-1G > A |
p.? |
Splice acceptor |
MK |
45 |
Yes |
Unilateral |
Ductal |
2 |
2 |
3 |
IIIA |
Yes |
No |
Yes |
No |
|
12 |
BC-1492 |
c.1564_1565del |
p.Glu522Ilefs*43 |
Frameshift |
MK |
61 |
Yes |
Unilateral |
Ductal |
2 |
1 |
2 |
IIB |
No |
No |
Yes |
No |
|
13 |
BC-91 |
c.3576G > T |
p.Lys1192= |
Splice, synonymous |
MK |
37 |
Yes |
Unilateral |
Ductal |
2 |
0 |
NA |
NA |
Yes |
Yes |
No |
No |
|
14 |
BC-2315 |
c.3576G > T |
p.Lys1192= |
Splice, synonymous |
MK |
37 |
Yes |
Unilateral |
Ductal |
1c |
x |
2 |
IA |
No |
No |
No |
Yes |
|
15 |
BC-2784 |
c.3576G > T |
p.Lys1192= |
Splice, synonymous |
MK |
34 |
Yes |
Unilateral |
Ductal |
2 |
3 |
3 |
IIIC |
Yes |
Yes |
No |
No |
|
16 |
BC-105 |
c.3603del |
p.Phe1201Leufs*6 |
Frameshift |
MK |
40 |
Yes |
Unilateral |
Ductal |
2 |
1 |
NA |
NA |
Yes |
Yes |
No |
No |
|
17 |
BC-160 |
c.3866delA |
p.Lys1289Argfs*4 |
Frameshift |
MK |
39 |
Yes |
Unilateral |
Lobular |
2 |
1 |
2 |
IIB |
Yes |
No |
Yes |
No |
|
18 |
BC-1048 |
c.5005G > A |
p.Glu1669Lys |
Missense, splice |
MK |
37 |
No |
Unilateral |
Ductal |
2 |
2a |
3 |
IIIA |
Yes |
Yes |
Yes |
No |
|
19 |
BC-1477 |
c.6115G > A |
p.Glu2039Lys |
Missense, splice |
MK |
41 |
Yes |
Unilateral |
Ductal |
1c |
1mi |
2 |
IB |
Yes |
Yes |
No |
No |
|
20 |
BC-1516 |
c.6889dup |
p.Gln2297Profs*76 |
Frameshift |
MK |
52 |
Yes |
Unilateral |
Ductal |
mi |
0 |
3 |
IA |
Yes |
Yes |
Yes |
No |
|
21 |
BC-865 |
c.8283_8284del |
p.Gln2762Alafs*6 |
Frameshift |
MK |
34 |
Yes |
Unilateral |
Ductal |
2 |
1a |
3 |
IIB |
Yes |
Yes |
Yes |
No |
|
22 |
BC-2490 |
c.8283_8284del |
p.Gln2762Alafs*6 |
Frameshift |
MK |
38 |
No |
Unilateral |
Ductal |
NA |
NA |
NA |
NA |
Yes |
Yes |
Yes |
No |
|
23 |
BC-1305 |
c.9139C > T |
p.Arg3047* |
Stop gained |
MK |
34 |
No |
Unilateral |
Ductal |
2 |
2 |
2 |
IIIA |
Yes |
Yes |
No |
No |
MK, Macedonian; AL, Albanian; PV, pathogenic variants; BC, breast cancer; OC, ovarian cancer; ER, estrogen receptor; PR, progesterone receptor; HER2/neu, human epidermal growth factor receptor 2; TN, triple negative; pos., positive; NA, not applicable; ATM, ataxia-telangiectasia mutated.
Co-occurrence of ATM PVs with PVs in BRCA1 or BRCA2 was observed in four BC patients. Two patients carried BRCA1 PVs, p.(Ala1453Glnfs*3) and p.(Tyr978*), and presented with triple-negative BC, both with a positive family history. Two additional patients carried BRCA2 PVs, p.(Ser1064Leufs*12) and p.(Ala938Profs*21), and had estrogen receptor (ER)-positive tumors, with only one reporting a family history of cancer. All cases were diagnosed at a young age and exhibited early-onset or aggressive disease phenotypes. These findings illustrate the co-occurrence of PVs in multiple BC susceptibility genes.
Associations between ATM PVs and clinicopathological characteristics are summarized in Table 3. Although ATM PVs were more frequently observed in patients of Macedonian ethnicity, in women diagnosed before the age of 50 years, and in those with a positive family history, these associations did not reach statistical significance. However, a statistically significant association was observed between ATM PV carrier status and human epidermal growth factor receptor 2 (HER2)-positive tumors (p = 0.0189).
TABLE 3. Association of Clinicopathological Characteristics with Pathogenic Variant Carrier Status in Breast Cancer Patients.
|
- |
All cases n=1211 |
PV carriers (n) |
PV carriers/ all BC cases (%) |
PV non-carriers (n ) |
BC non-carriers/all BC cases (%) |
OR (95% CI) |
p value |
|
Ethnicity |
- |
- |
- |
- |
- |
- |
- |
|
Macedonian |
1006 |
21 |
2.1 |
985 |
97.9 |
1.74 (0.4-7.48) |
0.76 |
|
Albanian |
165 |
2 |
1.2 |
163 |
98.8 |
- |
- |
|
Others |
40 |
0 |
0 |
40 |
100 |
- |
- |
|
Age of onset |
- |
- |
- |
- |
- |
- |
- |
|
≤ 50 |
724 |
17 |
2.3 |
707 |
97.7 |
1.90 (0.75-4.86) |
0.18 |
|
> 50 |
481 |
6 |
1.2 |
475 |
98.8 |
- |
- |
|
No data |
6 |
- |
- |
- |
- |
- |
- |
|
Bilateral BC |
- |
- |
- |
- |
- |
- |
- |
|
Yes |
71 |
1 |
1.4 |
70 |
98.6 |
- |
0.73 |
|
No |
1101 |
22 |
2 |
1079 |
98 |
1.43 (0.19-10.74) |
- |
|
No data |
39 |
- |
- |
- |
- |
- |
- |
|
Cancer FH |
- |
- |
- |
- |
- |
- |
- |
|
BC/OC |
605 |
16 |
2.6 |
589 |
97.4 |
1.91 (0.63-5.81) |
0.36 |
|
Other |
275 |
3 |
1.1 |
272 |
98.9 |
0.78 (0.17-3.64) |
- |
|
No FH |
286 |
4 |
1.4 |
282 |
98.6 |
- |
- |
|
No data |
45 |
- |
- |
- |
- |
- |
- |
|
ER |
- |
- |
- |
- |
- |
- |
- |
|
Positive |
767 |
18 |
2.3 |
749 |
97.7 |
2.62 (0.77-8.98) |
0.12 |
|
Negative |
331 |
3 |
0.9 |
328 |
99.1 |
- |
- |
|
No data |
113 |
2 |
- |
- |
- |
- |
- |
|
PR |
- |
- |
- |
- |
- |
- |
- |
|
Positive |
674 |
16 |
2.4 |
658 |
97.6 |
1.69 (0.62-4.67) |
0.31 |
|
Negative |
354 |
5 |
1.4 |
349 |
98.6 |
- |
- |
|
No data |
183 |
2 |
- |
- |
- |
- |
- |
|
HER2 |
- |
- |
- |
- |
- |
- |
- |
|
Positive |
225 |
9 |
4 |
216 |
96 |
2.92 (1.19-7.13) |
0.0189 |
|
Negative |
781 |
11 |
1.4 |
770 |
98.6 |
- |
- |
|
No data |
205 |
2 |
- |
- |
- |
- |
- |
|
TN |
- |
- |
- |
- |
- |
- |
- |
|
Yes |
161 |
2 |
1.2 |
159 |
98.8 |
- |
- |
|
No |
836 |
19 |
2.3 |
817 |
97.7 |
1.85 (0.43-8.02) |
0.41 |
|
No data |
214 |
2 |
- |
- |
- |
- |
- |
PV, pathogenic variant; BC, breast cancer; OC, ovarian cancer; OR, odds ratio; FH, family history; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2, TN, triple negative.
Variants of uncertain significance (VUS) were frequently identified in both cohorts. In BC patients, 19 distinct VUS were detected in 27 individuals (2.2%), whereas 22 VUS were identified in 24 controls (1.8%). All VUS were missense variants. Additionally, 45 variants with conflicting pathogenicity interpretations were observed in cases and 44 in controls. The c.7475T > G, p.(Leu2492Arg) variant showed substantial classification discordance in ClinVar, with submissions ranging from likely pathogenic to benign. This variant was more frequent in BC cases (1.9%) than in controls (1.1%); however, the association did not reach statistical significance (p = 0.086). The OR for this variant was 1.78 (95% CI: 0.91-3.48). The variant appeared more frequent among Albanian than Macedonian BC patients, whereas similar frequencies were observed between ethnic groups in controls. The variant affects a conserved residue within the ATM FAT domain and is predicted to be deleterious by multiple in silico tools, including AlphaMissense (0.91), REVEL (0.82), CADD (27), PROVEAN (-4.5), and SIFT (0.003). Strong evolutionary conservation at this position was also observed (PhyloP100 = 7.5). Visual inspection of sequencing reads using IGV confirmed the presence of the p.(Leu2492Arg) variant in all carriers. Analysis of seven single nucleotide polymorphisms (SNPs) located within ~50 kb of the ATM gene revealed homozygosity for the ancestral haplotype in 26 of 37 (70.3%) carriers of the p.(Leu2492Arg) variant. The remaining 11 carriers harbored the same haplotype in combination with four different haplotypes. The distribution of c.7475T > G variant carriers across ethnic groups and clinicopathological characteristics is presented in Table 4.
TABLE 4. Patient-Level Clinicopathological Data of Breast Cancer Patients Carrying ATM c.7475T > G. p.(Leu2492Arg) Variant.
|
Patient ID |
Ethnicity |
Age at diagnosis |
FH for BC/OC |
Localisation |
BC type |
T |
N |
G |
Stage |
ER pos. |
PR pos. |
HER2 pos. |
TN pos. |
|
BC-132 |
MK |
83 |
Yes |
Unilateral |
Ductal |
1mi |
0 |
3 |
IA |
No |
No |
No |
Yes |
|
BC-284 |
MK |
55 |
Yes |
Unilateral |
Ductal |
3 |
3a |
3 |
IIIC |
No |
No |
No |
Yes |
|
BC-345 |
MK |
67 |
No |
Unilateral |
Ductal |
2 |
0 |
3 |
IIA |
No |
No |
No |
No |
|
BC-634 |
MK |
57 |
No |
Unilateral |
Lobular |
1 |
0 |
- |
IA |
No |
NA |
NA |
NA |
|
BC-754 |
MK |
61 |
Yes |
Unilateral |
Ductal |
1 |
0 |
3 |
IA |
Yes |
Yes |
No |
No |
|
BC-773 |
MK |
26 |
Yes |
Unilateral |
Ductal |
1 |
3a |
3 |
IIIC |
Yes |
Yes |
Yes |
No |
|
BC-783 |
MK |
56 |
Yes |
Bilateral |
Lobular |
2 (m) |
2a |
2 |
IIIA |
No |
No |
No |
Yes |
|
BC-897 |
AL |
68 |
Yes |
Unilateral |
Ductal |
4b |
3a |
3 |
IIIC |
Yes |
Yes |
No |
No |
|
BC-995 |
MK |
62 |
No |
Unilateral |
Ductal |
2 |
0 |
3 |
IIA |
Yes |
Yes |
No |
No |
|
BC-1173 |
AL |
29 |
No |
Unilateral |
Ductal |
2 (m) |
1a |
3 |
IIB |
Yes |
Yes |
No |
No |
|
BC-1181 |
AL |
64 |
Yes |
Unilateral |
Ductal |
1c (m) |
1a |
2 |
IIA |
Yes |
Yes |
No |
No |
|
BC-1361 |
AL |
38 |
No |
Unilateral |
Ductal |
4b (m) |
3a |
3 |
IIIC |
Yes |
Yes |
No |
No |
|
BC-1380 |
MK |
58 |
Yes |
Unilateral |
Lobular |
1c |
2a |
2 |
IIIA |
Yes |
Yes |
No |
No |
|
BC-1567 |
MK |
44 |
Yes |
Unilateral |
Ductal |
1c |
0 |
3 |
IA |
Yes |
Yes |
Yes |
No |
|
BC-1651 |
MK |
40 |
Yes |
Unilateral |
Ductal |
2 |
0 |
3 |
NA |
Yes |
Yes |
NA |
No |
|
BC-1868 |
MK |
60 |
No |
Unilateral |
Lobular |
NA |
NA |
NA |
NA |
No |
No |
NA |
NA |
|
BC-1869 |
MK |
47 |
No |
Unilateral |
Na |
NA |
NA |
NA |
NA |
Yes |
Yes |
No |
No |
|
BC-2188 |
AL |
33 |
Yes |
Unilateral |
Ductal |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
NA |
|
BC-2383 |
AL |
63 |
Yes |
Unilateral |
Ductal |
2 |
3a |
3 |
IIIC |
NA |
NA |
NA |
NA |
|
BC-2472 |
MK |
64 |
Yes |
Unilateral |
Ductal |
2 |
0 |
2 |
IIA |
Yes |
Yes |
No |
No |
|
BC-2625 |
AL |
48 |
No |
Unilateral |
Lobular |
2 |
1 |
2 |
IIB |
Yes |
Yes |
No |
No |
|
BC-2631 |
MK |
45 |
No |
Unilateral |
Ductal |
1c |
1c |
3 |
IIA |
Yes |
No |
No |
No |
|
BC-2713 |
MK |
61 |
Yes |
Unilateral |
Ductal |
1 |
0 |
2 |
IIA |
Yes |
Yes |
No |
No |
MK, Macedonian; AL, Albanian; FH, family history; BC, breast cancer; OC, ovarian cancer; T, tumor size; N, nodes; G, grade; ER, estrogen receptor; PR, progesterone receptor; HER2/neu, human epidermal growth factor receptor 2; TN, triple negative; pos., positive; NA, not applicable; ATM, ataxia-telangiectasia mutated.
DISCUSSION
Heterozygous PVs in ATM are well-established moderate risk factors for BC, with meta-analyses and large population-based sequencing studies reporting an approximately 2–5-fold increase in lifetime risk compared with the general population.15, 16, 17, 18, 19 Several studies estimate the lifetime BC risk in heterozygous ATM carriers to be 21-24%, whereas the population risk is 12.5% (National Comprehensive Cancer Network Guidelines, Version 2.2026). In the present study, pathogenic ATM variants were identified in 1.9% of BC cases and 0.4% of controls, supporting the role of ATM as a moderate-penetrance BC susceptibility gene. The observed (OR = 5.02) lies at the higher end of published estimates and approaches the risk reported for certain high-penetrance founder mutations, such as c.7271T > G, p.(Val2424Gly).20 This relatively high estimate may be influenced by several factors, including the recurrence of specific PVs in our cohort (suggesting possible regional enrichment), differences in case–control structure (e.g., age distribution and sex composition), and the relatively small number of ATM PV carriers.
The carrier frequency in controls (0.4%) is consistent with the reported European prevalence of 0.2-0.5%.21, 22 Most detected ATM PVs were PTVs, accounting for 85.7% of unique variants, in line with previous reports showing that truncating variants predominate due to loss-of-function effects.23 Six recurrent PVs accounted for 71.4% of all detections, suggesting regional enrichment and highlighting the importance of population-specific risk assessment for ATM variants. Available data from Southeastern Europe remain limited and are primarily derived from multigene panel studies. These studies report the presence of ATM PVs but do not identify clear founder mutations, instead indicating a heterogeneous spectrum of rare variants.18, 24, 25, 26, 27 The co-occurrence of ATM PVs with BRCA1 or BRCA2 PVs in four early-onset BC cases highlights the value of multigene panel testing, as concurrent germline variants in DNA repair genes may modify cancer risk and phenotype. This is particularly relevant for young patients and those with a positive family history, consistent with current guidelines recommending comprehensive sequencing in high-risk individuals.
ATM PVs were more frequent among patients of Macedonian origin (2.1%) than among Albanian patients (1.2%) or individuals of other ethnic backgrounds. Of the 23 carriers, 21 were Macedonian and two were Albanian; both Albanian carriers harbored the same variant, c.495_496+16del, which was not detected in Macedonian patients. All ATM PVs identified in controls were found exclusively in individuals of Macedonian origin. Together with the high recurrence of specific PVs, these findings support the presence of regional enrichment. Identification of regionally enriched variants has important clinical implications, enabling targeted testing strategies and more efficient cascade screening.
Regarding clinical features, non-significant trends toward earlier age at diagnosis (before 50 years) and a positive family history of breast or ovarian cancer were observed. These effect sizes are consistent with published data, including a meta-analysis reporting a relative risk of 4.94 for ATM carriers diagnosed before age 50.28 The lack of statistical significance is likely due to the limited number of carriers rather than the absence of a true biological association.
Consistent with previous reports, ATM PV carriers predominantly presented with ER-positive tumors.16, 28 Notably, a significant association with HER2-positive status was observed (OR = 2.92, p = 0.0189). However, because of the relatively small number of carriers, multivariable adjustment for age and ethnicity was not performed; therefore, this finding should be interpreted with caution. In addition, the limited sample size resulted in wide confidence intervals, which may affect the precision of the estimated effect size. Recent meta-analyses have reported enrichment of germline ATM PVs in hormone receptor–positive/HER2-positive BC compared with triple-negative disease.29 Supporting this, analyses of HER2-positive tumors have identified ATM as one of the most frequently mutated genes, with potential implications for response to trastuzumab therapy.30 In our cohort, 39% of ATM PV carriers were HER2-positive, and most of these tumors were also ER-positive, corresponding to the Luminal B subtype, which is associated with a poorer prognosis.31 This observation may reflect biological heterogeneity within ATM-associated BCs, in which ATM dysfunction may interact with HER2-driven tumorigenesis. However, confirmation in larger, population-matched cohorts is required to determine whether a true association between ATM PVs and HER2-positive disease exists.
VUS were exclusively missense and were detected at comparable frequencies in BC cases (2.2%) and controls (1.8%). Of particular interest was the ATM c.7475T > G, p.(Leu2492Arg) (rs56399857) variant, which was more frequent in BC cases than in controls (1.9% vs. 1.1%), corresponding to a nearly twofold but not statistically significant increase in risk (OR = 1.78, p = 0.086). Its allele frequency in our control cohort (0.54%) was substantially higher than that reported in gnomAD (Genome Aggregation Database; 0.016% globally and 0.021% in non-Finnish Europeans). Balkan populations are underrepresented in global reference datasets such as gnomAD, which may result in underestimation of the true background frequency of certain variants in this region. The variant appeared more frequent among Albanian than Macedonian BC patients, whereas similar frequencies were observed between ethnic groups in controls; however, the small number of carriers limits definitive conclusions. Published data on the prevalence of this variant in Balkan populations remain limited, although it has been reported at lower frequencies in Greece, Slovenia, and Bulgaria.32, 33 In the GeneBass database, the variant shows a nominal association with BC (p = 2.43 × 10⁻³), although the reported allele frequencies remain substantially lower than those observed in our cohort. The elevated frequency observed in both cases and controls supports population-specific enrichment and suggests that p.(Leu2492Arg) may represent a regionally enriched variant in the Macedonian population rather than a clearly disease-associated allele. Analysis of seven SNPs within a ~50 kb region surrounding ATM suggested that the p.(Leu2492Arg) variant is linked to an ancestral haplotype. However, analysis of phased haplotypes and additional markers in larger cohorts will be required to more accurately assess the potential founder origin of this variant. This finding further highlights the importance of population-matched control datasets when interpreting missense variants in moderate risk genes such as ATM. The ATM p.(Leu2492Arg) variant has been reported in several cancer types, including colorectal, prostate, chronic lymphocytic leukemia, glioblastoma, breast, and ovarian cancers as well as in hereditary cancer syndromes.26, 34, 35 The variant is located within the highly conserved FAT domain of ATM and affects a residue with strong evolutionary conservation. Multiple in silico prediction tools consistently indicate a deleterious effect; however, such predictions alone are insufficient to establish pathogenicity in the absence of functional validation, which is currently lacking. Given its presence in healthy populations, repeated detection in cancer cohorts, and conflicting ClinVar classifications, the most appropriate classification remains a VUS. Further population-based and functional studies are required to clarify its clinical relevance.
Several limitations of this study should be acknowledged. The control cohort was not fully matched to the BC cohort in terms of age and sex, as controls were primarily individuals referred for genetic testing for non-oncological conditions and therefore had a younger median age and included both males and females. However, because ATM PVs are germline alterations present from birth, their carrier frequency is unlikely to be strongly influenced by age or sex distribution. In addition, the absence of CNV analysis may have led to a slight underestimation of the true prevalence of pathogenic ATM variants, as approximately 5-10% of such variants are deletions or duplications involving part or the entire ATM gene.
Overall, our findings highlight the importance of population-specific analyses for interpreting ATM variants and suggest that regional genetic architecture may significantly influence variant frequencies and risk estimates in BC.
Footnotes
Ethics Committee Approval: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Subcommittee of the Macedonian Academy of Science and Arts for Medicine, Pharmacy, Veterinary Medicine and Dentistry (approval number: 03-203/4, date: 10.12.2024).
Informed Consent: Written informed consent was obtained from each patient.
Data Sharing Statement: The datasets analyzed during the current study are available from the corresponding author upon reasonable request.
Authorship Contributions: Concept- I.M.K., D.P.K.; Design- I.M.K., D.P.K.; Supervision- D.P.K.; Funding- S.K., M.V., M.T.; Data Collection or Processing- S.K., P.N., M.V., M.T.; Analysis and/or Interpretation- I.M.K., S.K., D.P.K.; Literature Review- I.M.K., D.P.K.; Writing- I.M.K., D.P.K.; Critical Review- S.K., P.N., D.P.K.
Conflict of Interest: The authors declare that they have no conflict of interest.
Financial Disclosure: The authors declared that no financial support was received for this study.
References
- 1.Khanna KK, Chenevix-Trench G. ATM and genome maintenance: defining its role in breast cancer susceptibility. J Mammary Gland Biol Neoplasia. 2004;9(3):247–262. doi: 10.1023/B:JOMG.0000048772.92326.a1. [DOI] [PubMed] [Google Scholar]
- 2.Bakkenist CJ, Kastan MB. DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature. 2003;421(6922):499–506. doi: 10.1038/nature01368. [DOI] [PubMed] [Google Scholar]
- 3.Lee JH, Paull TT. ATM activation by DNA double-strand breaks through the Mre11-Rad50-Nbs1 complex. Science. 2005;308(5721):551–554. doi: 10.1126/science.1108297. [DOI] [PubMed] [Google Scholar]
- 4.Dombernowsky SL, Weischer M, Allin KH, Bojesen SE, Tybjaerg-Hansen A, Nordestgaard BG. Risk of cancer by ATM missense mutations in the general population. J Clin Oncol. 2008;26(18):3057–3062. doi: 10.1200/JCO.2007.14.6613. [DOI] [PubMed] [Google Scholar]
- 5.Shiloh Y. Ataxia-telangiectasia: closer to unraveling the mystery. Eur J Hum Genet. 1995;3(2):116–138. doi: 10.1159/000472285. [DOI] [PubMed] [Google Scholar]
- 6.Easton DF, Pharoah PD, Antoniou AC, et al. Gene-panel sequencing and the prediction of breast-cancer risk. N Engl J Med. 2015;372:2243–2257. doi: 10.1056/NEJMsr1501341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Stucci LS, Internò V, Tucci M, et al. The ATM gene in breast cancer: its relevance in clinical practice. Genes. 2021;12(5):727. doi: 10.3390/genes12050727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dorling L, Carvalho S, Allen J, et al. Breast cancer risks associated with missense variants in breast cancer susceptibility genes. Genome Med. 2022;14(1):51. doi: 10.1186/s13073-022-01052-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bernstein JL, Haile RW, Stovall M, et al. Radiation exposure, the ATM gene, and contralateral breast cancer in the women’s environmental cancer and radiation epidemiology study. J Natl Cancer Ins t. 2010;102:475–483. doi: 10.1093/jnci/djq055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McKenna A, Hanna M, Banks E, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–1303. doi: 10.1101/gr.107524.110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Danecek P, Bonfield JK, Liddle J, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):giab008. doi: 10.1093/gigascience/giab008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.McLaren W, Gil L, Hunt SE, et al. The ensembl variant effect predictor. Genome Biol. 2016;17(1):122. doi: 10.1186/s13059-016-0974-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Noveski P, Matevska-Geshkovska N, Plaseska-Karanfilska D, Dimovski A. Comprehensive pharmacogenetic allele landscape from whole exome sequencing: single-center cohort analysis in the population of North Macedonia. In: Samardzic J, editor. Pharmacokinetics and pharmacogenetics: principles, applications, and challenges. London: IntechOpen; 2025. [Google Scholar]
- 14.Richards S, Aziz N, Bale S, et al. ; ACMG Laboratory Quality Assurance Committee. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405–424. [Google Scholar]
- 15.Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet. 1998;62:676–689. doi: 10.1086/301749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hall MJ, Bernhisel R, Hughes E, et al. Germline pathogenic variants in the ataxia telangiectasia mutated ( ATM ) gene are associated with high and moderate risks for multiple cancers. Cancer Prev Res (Phila) 2021;14(4):433–440. doi: 10.1158/1940-6207.capr-20-0448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lesueur F, Easton DF, Renault AL, et al. ; CARRIERS and Ambry Groups; Stoppa-Lyonnet D, Andrieu N. First international workshop of the ATM and cancer risk group (4-5 December 2019). Fam Cancer. 2022;21:211–227. doi: 10.1007/s10689-021-00248-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Moslemi M, Moradi Y, Dehghanbanadaki H, et al. The association between ATM variants and risk of breast cancer: a systematic review and meta-analysis. BMC Cancer. 2021;21:27. doi: 10.1186/s12885-020-07749-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Marabelli M, Cheng SC, Parmigiani G. Penetrance of ATM gene mutations in breast cancer: a meta-analysis of different measures of risk. Genet Epidemiol. 2016;40(5):425–431. doi: 10.1002/gepi.21971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shaitelman SF, Anderson BM, Arthur DW, Bazan JG, Bellon JR, Bradfield L, Coles CE, Gerber NK, Kathpal M, Kim L, Laronga C, Meattini I, Nichols EM, Pierce LJ, Poppe MM, Spears PA, Vinayak S, Whelan T, Lyons JA. N Engl J Med. 2. Vol. 384. Dorling L, Carvalho S, Allen J, et al; 2021. Breast cancer risk genes - association analysis in more than 113,000 women. pp. 428–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Girard E, Eon-Marchais S, Olaso R, et al. Familial breast cancer and DNA repair genes: insights into known and novel susceptibility genes from the GENESIS study, and implications for multigene panel testing. Int J Cancer. 2019;144:1962–1974. doi: 10.1002/ijc.31921. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Karczewski KJ, Francioli LC, Tiao G, et al. ; Genome Aggregation Database Consortium; Neale BM, Daly MJ, MacArthur DG. The mutational constraint spectrum quantified from variation in 141,456 humans. Nature. 2020;581:434–443. doi: 10.1038/s41586-020-2308-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Choi M, Kipps T, Kurzrock R. ATM mutations in cancer: therapeutic implications. Mol Cancer Ther. 2016;15(8):1781–1791. doi: 10.1158/1535-7163.MCT-15-0945. [DOI] [PubMed] [Google Scholar]
- 24.Ak B, Akin H, Zekioğlu O, Serin G, Göktepe B, Solmaz AE. Breast cancer subtypes in CHEK2, ATM, and other non-BRCA germline carriers in Turkey. Academia Oncology. 2026;3. [Google Scholar]
- 25.Bozhanov SS, Angelova SG, Krasteva ME, et al. Alterations in p53, BRCA1, ATM, PIK3CA, and HER2 genes and their effect in modifying clinicopathological characteristics and overall survival of Bulgarian patients with breast cancer. J Cancer Res Clin Oncol. 2010;136(11):1657–1669. doi: 10.1007/s00432-010-0824-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Krivokuca A, Mihajlovic M, Susnjar S, et al. Mutational profile of hereditary breast and ovarian cancer–establishing genetic testing guidelines in a developing country. Curr Probl Cancer. 2022;46:100767. doi: 10.1016/j.currproblcancer.2021.100767. [DOI] [PubMed] [Google Scholar]
- 27.Fostira F, Kostantopoulou I, Apostolou P, et al. One in three highly selected Greek patients with breast cancer carries a loss-of-function variant in a cancer susceptibility gene. J Med Genet. 2020;57:53–61. doi: 10.1136/jmedgenet-2019-106189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hu C, Hart SN, Gnanaolivu R, et al. A population-based study of genes previously implicated in breast cancer. N Engl J Med. 2021;384:440–451. doi: 10.1056/NEJMoa2005936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.de Baumont AC, Cadore NA, Pedrotti LG, et al. Germline rare variants in HER2-positive breast cancer predisposition: a systematic review and meta-analysis. Front Oncol. 2024;14:1395970. doi: 10.3389/fonc.2024.1395970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zakaria NH, Hashad D, Saied MH, Hegazy N, Elkayal A, Tayae E. Genetic mutations in HER2-positive breast cancer: possible association with response to trastuzumab therapy. Human Genomics. 2023;17(1):43. doi: 10.1186/s40246-023-00493-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Stagni V, Manni I, Oropallo V, et al. ATM kinase sustains HER2 tumorigenicity in breast cancer. Nature Commun. 2015;6(1):6886. doi: 10.1038/ncomms7886. [DOI] [PubMed] [Google Scholar]
- 32.Maver A, Juvan P, Kotnik U, Lovrecic L, Bergant G, Peterlin B. Creating the Slovenian genome database and browser as a source of comprehensive variation of the Slovenian population. Sci Rep. 2025;15(1):41118. doi: 10.1038/s41598-025-24991-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kalfakakou D, Fostira F, Papathanasiou A, et al. CanVaS: documenting the genetic variation spectrum of Greek cancer patients. Human Mutat. 2021;42(9):1081–1093. doi: 10.1002/humu.24249. [DOI] [PubMed] [Google Scholar]
- 34.Yurgelun MB, Kulke MH, Fuchs CS, et al. Cancer susceptibility gene mutations in individuals with colorectal cancer. J Clin Oncol. 2017;35:1086–1095. doi: 10.1200/JCO.2016.71.0012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Mangolini A, Rocca C, Bassi C, et al. Detection of disease‐causing mutations in prostate cancer by NGS sequencing. Cell Biol Int. 2022;46(7):1047–1061. doi: 10.1002/cbin.11803. [DOI] [PMC free article] [PubMed] [Google Scholar]