Classification of results
The study’s findings suggest that individuals with breast cancer risk genes do not differ significantly from breast cancer patients with unknown variant status. However, non-oncological patients with risk genes are more sensitive to radiation than healthy individuals without known variant status, which can certainly be explained by the prevalence of risk gene variants. Overall, patients with cancer risk genes exhibit slightly increased mean sensitivity to radiation, a characteristic that differs from that observed in the healthy control group. Significant differences in radiosensitivity exist among various cancer risk genes. Patients with CHEK2, BRIP1, PALB2, or ATMhet have relatively average radiosensitivity, similar to healthy individuals. In contrast, variants in BRCA1/2 and TP53 lead to a limited increase in radiosensitivity, comparable to that observed in breast cancer patients. Patients with variants in MSH, RAD51C/D, and BARD1 lead to a significantly higher mean radiosensitivity than in control groups. It must be considered that the interpretation of these findings is limited by the absence of confirmed non-carrier status in the control groups and results for some genes should be interpreted with caution due to the small size of the cohort.
However, it should be noted that radiosensitivity can vary greatly within a gene variant. Even among genes with an average low radiosensitivity, there are individuals with high radiosensitivity. Similarly, among the three genes with very high radiosensitivity, there are individuals with average sensitivity. Thus, the correlation between measured radiosensitivity should be viewed in this context. An increase in chromosomal aberrations starting from about 0.4 B/M results in an ever-increasing risk of an undesirable therapeutic outcome up to a value of approximately 0.8 B/M. This range must be regarded as stochastic. Beyond that point, a more deterministic range begins, where the risk of an undesirable outcome is very high27.
Even genes with diverse functions may contribute to radiation sensitivity due to their role in DNA damage response pathways. CHEK2 and TP53 are checkpoint control proteins; ATM and BRCA1 are DNA repair sensor proteins; and BRCA2, PALB2, and BARD1 form heteromers that participate in DNA double-strand break (DSB) repair through homologous recombination (HR). BRCA1 and RAD51 are key enzymes in HR7. MSH, on the other hand, is a protein involved in mismatch repair28.
A specific mutation probably cannot be used to predict whether there is an increased sensitivity to radiation. This becomes clear when considering variant c.5266dupC (p.Gln1756fs) of BRCA1. This variant was studied in 18 patients, and the measured radiosensitivity ranged from 0.35 to 0.90 B/M. This shows a large variation in radiosensitivity within a specific mutation. Radiosensitivity is a complex, multifactorial process requiring many modifying factors to produce such a wide range of variation, as illustrated here29.
More mutations are found in the RING and BRCT binding domains of BRCA1, yet much less in the serine cluster domain (SCD)12. These two overrepresented domains are associated with increased radiosensitivity, but not the SCD domain. It is unclear why carriers of variant genes have these mutations more frequently in the binding domains. However, we believe that these mutations likely cause increased radiosensitivity because the loss of the ability to bind to other proteins results in a lack of repair activity30.
The main reason for measuring the background level of chromosomal aberrations is to subtract it from the level induced by radiation to obtain only the aberrations caused by radiation12. However, it is also exciting because different groups have different numbers of chromosomal aberrations as a background. Since ionizing radiation is rare in everyday life, other toxins are more likely responsible for these background mutations. BRCA1/2 plays a role in repairing DNA double-strand breaks via homologous recombination. More importantly, it is involved in the Fanconi pathway for repairing DNA interstrand crosslinks. These crosslinks are induced by chemicals31. The higher proportion of aberrations in patients with oncological disease, especially those with multiple tumors and recurrences, likely reflects their higher exposure to carcinogens by diet and the environment. Furthermore, genetic variants in enzymes that metabolise carcinogens, such as cytochrome P450 (CYP), glutathione S-transferases (GST) and N-acetyltransferases (NAT), may alter the metabolism of environmental carcinogens, affecting cancer risk as a result32. Additionally, different mean age could explain this difference as background aberrations increase in older people33, and risk gene carriers with tumor disease were on average about 10 years older than non-oncological patients. In the risk gene group, we found slightly increasing radiosensitivity with age for the non-oncological patients and no association with age in the oncological patients13. Maybe a selection bias leads to that result, because among the limited number of individuals especially young patients were studied for their radiosensitivity.
Consequences of radiotherapy
The most common side effects after breast irradiation are dermatitis, fatigue or pain. The design of radiotherapy plays an important role. Patients who received a 40 Gy/15 fractions regimen (three-week RT) experienced less toxicity than those who received 50 Gy/25 fractions (five-week RT)34. A study found that partial breast irradiation in lower-risk patients reduced symptoms such as pain, systemic side effects and breast symptoms35. In addition, techniques such as deep inspiration are being employed in studies to minimize radiation exposure to the heart. Cardiotoxicity is a late effect of radiation therapy, particularly in cases of left-sided breast cancer36. Serious side effects after breast irradiation are rare due to the lower total dose (~ 50 Gy)37 compared to, for example, head and neck radiotherapy (~ 68 Gy)38. In the ABPI2 study, only one patient out of 170 who received accelerated partial breast irradiation had grade 2 pneumonitis. But this patient had also an elevated radiosensitivity value of 0.79 B/M probably due to vitiligo, which has already been discussed to cause higher radiosensitivity39. Dose reduction could be considered in patients ≥ 0.55 B/M because of the increased risk of side effects, and we recommend strict follow-up for these patients. In the total risk gene group, we identified 67 (24.5%) patients with radiosensitivity values ≥ 0.55 B/M, for whom we would recommend these adjustments27. Since radiation side effects are rare in breast cancer patients, we could only find two patients with radiodermatitis (0.510 B/M, 0.745 B/M) and one patient with radio colitis (0.450 B/M) documented in the risk gene group.
Reasons for radiosensitivity in heterozygosity and application of radiosensitivity knowledge
The question arises as to why heterozygous variants can lead to increased radiosensitivity since one allele is unaffected. For cancer development, the second hit theory involving loss of heterozygosity (LOH) is an important aspect. In addition. studies suggest other causes without LOH. One explanation for a non-functional protein is haploinsufficiency, which could result in a reduced amount of protein that is unable to fully perform its function. This is accentuated under stress by metabolic disorders, DNA damage and in tissues with high proliferation and cell division, such as breast or ovarian tissue40,41,42. Whereas in other genes, like TP53, the dominant negative effect of a mutated subunit in the protein causes cancer43. This may explain the inter-individual differences in radiosensitivity found in risk gene carriers due to different reasons for the dysfunctional protein.
With a total of 273 high-risk individuals, the study represents a comparatively large overall cohort. The BRCA1/2 subgroup is particularly large, allowing for more robust conclusions to be drawn. In contrast, the numbers for other gene subgroups remain limited (e.g. BARD1, RAD51C/D). The comparison groups included 147 breast cancer patients and 211 healthy individuals. However, as the controls are not confirmed non-carrier groups, the validity of causal conclusions regarding gene-specific effects is limited. Therefore, while the dataset enables meaningful analyses of general radiation sensitivity and BRCA1/2-associated effects, conclusions regarding rarer gene variants should be considered exploratory and preliminary. DNA damage, as measured by the FiSH method, correlates with radiation exposure. It is also a suitable long-term biomarker for mapping radiation damage44. Chromosomal aberrations by FiSH are suitable for predicting radiosensitivity because the value of B/M corresponds to the clinical side effects of radiotherapy45,46. Even if the effects of radiation in breast cancer therapy are minimal, the knowledge gained about the radiosensitivity of breast cancer risk gene carriers could be therapeutically useful. In theory, radiosensitivity values found in lymphocytes can be extrapolated to the tumor. This could make it possible to predict the tumor-reducing effect of radiotherapy. Radiotherapy could be individually optimized to a lower, less harmful dose, trying to balance efficacy with minimized side effects27. For the genes with increased radiation sensitivity identified in this study (BARD1, RAD51C/D and MSH), current recommendations do not indicate any contraindications for radiation therapy. Radiation therapy is not recommended in cases of TP53 mutations. In addition, there are no restrictions for BRCA1/2, though mastectomy is often performed for therapeutic reasons due to the risk of recurrence47.
Limitations
There are some limitations in this study. For some rare variants, including BARD1, BRIP1, CHD1, NF1, PALB2, PMS2, RAD51C/D, TP53 and double heterozygote variants, relatively few patients could be analyzed. The results of these subgroups should be considered exploratory rather than definitive and require confirmation or refutation in larger studies. The comparison groups are limited due to the unknown proportion of risk carriers and the absence of confirmed non-carrier status, which results from the historical collection of data for a study with a different focus. Therefore, gene-specific conclusions are limited in this study, as it cannot be excluded that a small proportion of mutation carriers may be present in the comparison group. It would be valuable for future studies to include a control cohort with confirmed non-carrier status to address this limitation. The breast cancer group has many outlying values > 0.1 B/M in background aberrations compared to other groups. This could have been caused by previous cancer treatment. Background levels of heterogeneity may introduce bias through subtraction, which could potentially influence the final radiosensitivity results. Several factors can influence radiosensitivity. In addition to variant status, which affects DNA repair mechanisms, other genetic modifying variables are present. Other factors that influence radiosensitivity should be considered, such as age, which increases slightly with advancing age, and inflammatory diseases, such as rheumatoid arthritis48, vitiligo39 or lupus erythematosus27 and drugs like chloroquine and kinase inhibitors49. However, no gender-related differences were found50. As several variables can influence radiation sensitivity, it would have been preferable to conduct multivariate regression analyses to capture these interactions more comprehensively and arrive at more nuanced conclusions by taking multiple influencing factors into account simultaneously. This should be considered in future studies involving larger subgroups. Additionally, the FiSH radiosensitivity test should be performed prior to radiotherapy, as high background rates can complicate data interpretation51.
