In these years of personalized medicine, the study of individual’s genotype is an important part of the determination of his specific susceptibility to several diseases, including cancer. The National Comprehensive Cancer Network Breast Cancer Risk Panel (NCCN) has for years been updating with publishing the indications for genetic testing of cancer patients and their family members (Sorscher 2019). The majority of Breast Cancer (BC), Ovarian Cancer (OC) and Pancreatic Cancer (PC) cases are sporadic (75%–80%), ∼15%–20% are considered familial types and 5%–10% are hereditary (Russo et al., 2009; Antonucci et al., 2017b; Incorvaia et al., 2020). Over the past 20–30 years, molecular diagnosis of hereditary BC, OC or PC has focused primarily on two high-penetrance genes, BRCA1 and BRCA2 (Antonucci et al., 2017a). The identification of germline deleterious variants in BRCA1/BRCA2 has a significant impact on clinical management of both affected individuals and their family members (Babore et al., 2019; Lombardi et al., 2019; 2022). Nevertheless, an increasing number of families with pedigrees suggestive of a Mendelian form of BC, OC or PC have not detectable mutations in BRCA1/BRCA2. The problem of “missing heritability” can be explained with the presence of pathogenic gene variants in other susceptibility genes involved with low frequency or with reduced penetrance, usually not included in the diagnostic flowchart of patients with hereditary cancer, mainly due to the costs and the time required for the analysis in the Sanger sequencing era. Therefore, it has become mandatory to study many genes in a brief time and in an economic way. In this scenario, advances in genetic technology and implementation of NGS in clinical oncology have accelerated the discovery of new cancer-related genes revolutionizing cancer research, diagnosis and therapies (Rossi et al., 2022). The advent of NGS allows the simultaneous sequencing of multiple samples and genes (Fountzilas et al., 2018). Because of the advantage from cost-benefit reduction, this approach provides a powerful enforcement for patients with LPVs and PVs in other genes, beyond BRCA1/2. Several germline PVs in susceptibility genes as CDH1, PALB2, PTEN, STK11, TP53, ATM, CHEK2, BARD1, BRLP1, RAD51C, and RAD51D (Shah et al., 2016; Fanale et al., 2020) can be associated with hereditary tumors. Most of these genes are involved in cell cycle checkpoint and DNA damage repair mechanism, and function together in these physiological pathways (Nielsen et al., 2016; Piombino et al., 2020; Neiger et al., 2021); therefore, a fundamental comprehension of the disease drivers in the cascades would facilitate the accurate evaluation of the genetic risk of cancer development (Yoshimura et al., 2022). In our study we used a multi-gene panel including 27 genes in the diagnostic iter of 546 patients with BC, OC or PC (Table 1). The aims of this work were: 1) to investigate the prevalence of PVs or LPVs in susceptibility genes implicated in hereditary cancer predisposition, and 2) to assess the utility of carrying out a multi-gene panel testing in BC, OC or PC individuals who fulfill specific criteria on their familiar and personal history of tumor.

In our study, 546 cases with BC, OC, or PC were enrolled from January 2020 to December 2021. PVs or LPVs on BRCA1/2 genes were detected in 44 patients (8%), specifically 32/423 (7%) with BC, 9/59 (15%) with OC and 3/64 (5%) with PC. On the other hand, 46 patients (8%), namely 33/423 (8%) with BC, 3/59 (5%) with OC and 10/64 (16%) with PC harbored germline PVs/LPVs in other cancer susceptibility genes, as follows: 17 (37%) in MUTYH, 11 (24%) in CHEK2, 4 (9%) in ATM, 3 (6%) in RAD51C and TP53, 2 (4%) PALB2, BRIP1, and NBN. In addition, a single PV in POLE, MSH2, and RAD51D was detected in two patients (Table 2).

Seven subjects enrolled showed two pathogenic variants in the genes analyzed.

According to age of onset, we found PVs/LPVs in 20 early onset patients (≤45 for PC and OC, ≤40 for BC) and in 70 late onset patients (>45 for PC and OC, >40 for BC). Eleven early onset patients with BC (14%) had PVs or LPVs mutations in BRCA1 or BRCA2 genes, whereas 17 patients (11%) reported mutations in one of the other genes included in the multi-gene panel. On the other hand, 27 late onset patients with BC (36%) had PVs or LPVs mutations in BRCAor BRCA2 genes, whereas 30 patients (40%) reported mutations in one of the several genes included in the multi-gene panel. On the OC and PC patients groups 2 early onset subjects (18%) had a PV or LPV in BRCA1/2, while 2 patients (18%) had PV or LPV in other gene. In the late onset group 10 patients (9%) had a PV or LPV in BRCA1/2 and 11(10%) with pathogenic variant in other gene. The distribution of PVs/LPVs in BRCA1/2 or in other genes in the different groups of patients is reported in Table 3.

MUTYH resulted as the gene with the higher percentage of mutation within the group analyzed by the multigene panel (16 out of 46 detected mutations), with the second most recurrent involved genes represented by CHEK2 with 11 cases (Table 2; Figure 1; Figures 2A, B). All MUTYH variants reported in this study are in heterozygous, except only one subject that had two PVs/LPVs on MUTYH gene, respectively c.734G>A and c.884C>T.

As to PVs/LPVs, the most frequent PV was c.1187G>A p.(Gly396Asp) of MUTYH gene, located in coding exon 13 and causing the substitution of a Glycine with Aspartate in codon position 396. This alteration, found in seven patients (15.2%) with BC, PC and OC, is frequently reported as founder mutation in multiple populations. M. Nielsen et al. have shown that this missense variant change the function of MUTYH protein (M. Nielsen et al., 2009).

The second recurrent PV found on the MUTYH gene was c.734G>A p. (Arg245His), in coding exon 9, results from the substitution of a Guanine to Adenine, and consequently the replacement of the arginine with a histidine at codon 245. Literature’s data supports that this missense variant has a deleterious effect on protein structure/function (Viel et al., 2017). We found this mutation in five patients (10.9%).

Analyzing the second most gene mutated, CHEK2, the other most recurrent PVs were: c.1100delC p. (Thr367fs), c.470T>C p. (Ile157Thr) and c.499G>A p. (Gly167Arg). All subjects with CHEK2 variant, are carriers of only one PV/LPV.

Specifically, CHEK2 c.1100delC caused deletes of one Cytosine from exon 11 in position 1,100 causing a frameshift at codon 367, and a premature translational stop signal p. (Thr367fs). This is expected to result in an absent or disrupted protein product. (Weischer et al., 2008). This variant is linked to increased risk of BC and OC.

The identification of PV or LPV in genes responsible for hereditary cancers plays a key role in the prognosis, prevention and therapy of these conditions. In fact, cancer patients carriers of such gene variants must undergo specific protocols for the prevention of additional cancers but they can also benefit from specific drug therapies, such as those based on PARP inhibitors (PARPi), which represent a successful example of precision medicine (Slade, 2020). On the other hand, unaffected family members of a cancer patient carrier of a PV/LPV should be tested for the presence of the same variant and, when positive, specific prevention protocols, different from the common cancer screening programs used for the general populations, should be offered. In this view, a critical issue is represented by the number of genes to analyze in each condition, mainly in order to maintain a balanced cost/benefit ratio. While in a first moment it was suggested that each different type of cancer was related to one or a few specific genes (e.g., BRCA1/2 for BC and OC, APC for familial adenomatous polyposis, etc.), our study revealed that often there is not correspondence between tumor type and the associated mutated gene, raising the question about the need for more genes to be analyzed in hereditary cancers. Interestingly, our study showed that 94% of MUTYH carriers had a heterozygous variant. PVs/LPVs in MUTYH are associated with colorectal adenomatous polyposis autosomal recessive, while recent literature data revealed the association between monoallelic MUTYH variants and several type of cancer (Dell’Elice et al., 2021). BC, PC and OC, together with colon and prostate cancer, are the major tumors linked to clinical familiar history, as well as the major BRCA-associated cancers (Daly et al., 2021). Nevertheless, many of these patients result negative to the genetic testing for BRCA1/2 genes PVs and LPVs, even in presence of an evident familiar and/or personal cancer’s background. This has been confirmed by data obtained in the present study, showing that no more that 8% of BC, OC or PC cancer show BRCA1/2 mutations even in the group of early onset cases.; that the use of multi-gene hereditary cancer panels increases the possibility to identify individuals with cancer predisposing gene variants (Shin et al., 2020; Hu et al., 2021). In an association analysis involving 113,000 women, the Breast Cancer Association Consortium, Dorling L, Carvalho S, et al. define the susceptibility genes that are most clinically useful for inclusion on panels for the prediction of breast cancer risk (Breast Cancer Association Consortium et al., 2021). By extending the test using a multi-gene panel, we found an additional 8% mutations in different susceptibility genes, such as MUTYH, CHEK2, ATM, NBN, BRIP1, and TP53 involved in several hereditary cancer syndromes (Desmond et al., 2015; Tsaousis et al., 2019; N; Tung et al., 2015). These results confirmed the studies already performed in 2021 by Bono et al., where a considerable percentage of PVs/LPVs have been lost without the use of multi-gene panel (Bono et al., 2021). Thus, our results evidenced that both in early and late onset cancer patients, using the classical approach of BRCA1/2 testing, we would have lost a large number of cases resulted BRCA1/2 negative, but actually carriers of a PV/LPV in other genes. In addition to the increased detection rate, the use of multigenic panel test allow the identification of specific prevention strategies based on the gene involved, in a precision medicine approach. For example, we diagnosed three patients with Li-Fraumeni syndrome (LFS) associated with PV/LPV in TP53 on chromosome 17p13.1. This syndrome represents a severe condition inherited in an autosomal dominant manner with very high penetrance. Prevention strategies of this condition are different from the one used for BRCA1/2 mutation carriers, since LFS component tumors also include soft tissue sarcomas, osteosarcoma, brain tumors, and adrenocortical carcinomas. Interestingly, in these patients no strong familiar history was found, but they all showed early onset cancer (≤35). In one case, a “de novo” origin of the mutation was demonstrated, allowing to suggest that the age of onset of the disease could be considered as a more reliable indicator of the presence of a genetic condition than the familiarity itself. Oncology therapy putting forth the concept of selective targeting of cancer cells thanks to precision medicine. According to our goal, one of the most interesting future perspectives is the therapy with poly-adenosine diphosphate-ribose polymerase (PARP). PARP inhibitors (PARPi) were a significant example of precision medicine (Slade, 2020). The identification of specific mutations in genes different from BRCA1/2 is relevant also for the therapeutical strategies. In fact, while the benefits of PARP inhibition have been well characterized for BRCA1/2 (Risdon et al., 2021), the efficiency of this therapy in carriers of other mutations is so far a question of debate. For the therapy of metastatic breast cancer (MBC), is in progress a phase II study that are showing the efficacy of PARPi’s Olaparib, in patients with germline/somatic (g/s) mutations in related genes (PALB2, ATM and CHEK2) other than BRCA1/2 (N. M. Tung et al., 2020). Responses were seen only with gPALB2 mutations, while there are not evidences for ATM or CHEK2 mutations respectively. For this reason, Olaparib could be used in patients with gPALB2 mutation beyond in gBRCA1/2 mutation carriers, significantly expanding the number of patients with MBC who would benefit from PARPi (Pommier et al., 2016; Lord and Ashworth., 2017; Cortesi et al., 2021). In conclusion, the multi-gene panel approach could be useful for targeting therapy in oncology patients that are carriers of mutations in susceptibility genes, beyond BRCA1/2.