Homologous recombination (HR) is one of the major pathways for the repair of DNA double-strand breaks (DSBs) in eukaryotic cells. Pathogenic mutations in genes encoding HR-related proteins are associated with the development of certain malignancies, including breast, ovarian, and other cancers (1). Normal and cancer cells rely on multiple DNA damage response pathways that specifically repair different forms of DNA damage. Key pathways include homologous recombination repair (HRR), base excision repair (BER), nucleotide excision repair (NER), mismatch repair (MMR), nonhomologous end-joining (NHEJ), translesion synthesis (TLS), and interstrand crosslink (ICL) repair (2, 3). However, these repair pathways are not equally effective in DNA repair, and some mechanisms are error-prone. For instance, non-canonical DNA repair systems such as NHEJ, single-strand annealing, and TLS are activated when the canonical pathways are deficient (4, 5). In response to DSBs, cells activate the HRR pathway, which relies on the undamaged sister chromatid as a template for repair. Because of this reliance on sister chromatids, HRR is active during the S and G2 phases of the cell cycle and is a high fidelity and error-free DNA repair pathway (3). By contrast, in NHEJ, the break ends are directly ligated without a homologous template, resulting in an error-prone repair pathway that can predispose to genetic instability (6, 7).

Genomic scars in relation to HR are caused by chromosomal and sub chromosomal aberrations. Genomic aberrations arise from mutation, structural copy number aberrations, or structural rearrangements. Mutations are substitutions (transversion & transition) or indel (insertion & deletion) mutations, and can inactivate tumor suppressor genes (TSGs). Structural copy number alterations can be copy number gain (leading to allelic imbalance) or copy number loss [leading to deletion, loss of heterozygosity (LOH), or haplo-insufficiency (one copy of a gene is deleted or contain loss of function mutation leading to insufficient level of proteins)]. Structural rearrangements can be inversion (paracentric), translocation (reciprocal), or recombination leading to copy neutral LOH events (8). Frequent copy number alterations are the hallmarks of homologous recombination deficiency (HRD) and can occur at the regional or whole chromosome level. Quantification of large-scale structural variants is used as an indicator of the HRD phenotype (presence of HRD in sporadic cancers other than BRCA1 and 2 inactivation), including telomeric allelic imbalance (TAI: large allelic imbalances extending to the telomere), large scale transition (LST: number of transitions between large regions of different allelic states or chromosomal breaks between adjacent regions of >10 MB), and LOH (large regions displaying somatic loss of one haplotype, which can be copy variable as in deletion or copy neutral LOH). Genes involved in HR, including tumor suppressor genes, can also be repressed by aberrant promoter hypermethylation, an epigenetic mechanism that contributes to HRD (9).

HR is a DNA repair pathway of clinical interest because of the sensitivity of HRD cells to poly-ADP-ribose polymerase (PARP) inhibitors (PARPIs) (10). DNA repair targeting therapies exploit DNA repair defects in cancer cells to generate synthetic lethality, and DNA repair defects vary according to cancer type. For example, approximately 50% of ovarian carcinomas exhibit dysfunctional HRR, whereas the rate of HRR dysfunction is lower in other cancer types such as colorectal cancer (CRC) (<5%) (11, 12). Hereditary mutations in one copy of the BRCA gene predispose patients to female breast cancer (85% lifetime risk), ovarian cancer (10%–40%), male breast cancer, pancreatic cancer (PC), prostate cancer, non-small cell lung cancer (NSCLC), CRC, and other cancer types. Precancerous cells deficient in BRCA1 and 2 cannot repair DSBs properly, resulting in genomic instability that eventually leads to cancer (8). These tumors are intrinsically sensitive to DNA damage response inhibitors (PARPIs), which induce synthetic lethality. Synthetic lethality can arise from the combined inactivation of HRR genes by (mutation, LOH and promoter hypermethylation) and PARP inhibition (13). The use of PARPIs in non-BRCA mutation carrier patients can be expanded to sporadic cancers that display “BRCAness” (cancers that have defective HR without germline BRCA1 and 2 mutations). Findings showed that TSGs with BRCAness phenotypes are often inactivated, for example, ATM, ATR, PALB2, RAD51, RAD51B, RAD51C, RAD51D, BARD1, and FANCM among others (14–17).

The use of PARPIs was recently expanded to other cancers in addition to breast and ovarian cancer, such as prostate and PC (18). Although PARPIs have shown beneficial effects in many other cancer types, the frequent development of resistance is challenging. For instance, in a phase II clinical trial, secondary resistance mutations were detected in circulating free tumor DNA in two patients with a germline BRCA2 mutation. These mutations were predicted to lead to the reversal of a somatic mutation (19). A comprehensive investigation of the underlying mechanisms is necessary to design strategies for overcoming PARPI resistance.

Another challenging issue is the development of effective biomarkers to identify patients who are more likely to respond to specific targeted therapy by using companion diagnosis (CDx). CDx is an in vitro medical device that uses biomarkers to provide information on the safe and effective use of drugs or biologicals. FDA-approved CDx includes BRACAnalysis CDx^® and Myriad myChoice^® CDx developed by Myriad Genetic Laboratories, and FoundationOne^® CDx [F1CDx] and FoundationOne^® Liquid CDx developed by Foundation medicine. The HRDetect test utilizes machine learning algorithm (20).

BRACAnalysis CDx^® is an in vitro diagnostic method used for the detection and classification of DNA sequence variants in the protein-coding regions, intron or exon boundaries of the germline BRCA1 and 2 genes from whole blood sample. PCR and Sanger sequencing are used to detect small insertions and deletions (indels), and single nucleotide variants (SNVs). Large deletions and duplications are detected by multiplex PCR. The test results used as an aid to identify eligible patients for PARPIs in breast, ovarian, pancreatic and prostate cancers treatment (21). Myriad myChoice^® CDx is NGS-based in vitro diagnostic test that evaluates the qualitative detection and classification of SNVs, indels and large rearrangements (LRs) in protein-coding regions and intron/exon boundaries of the BRCA1 and 2 genes, and determine Genomic Instability Score (GIS) by measuring [LOH, TAI, and LST] using DNA isolated from formalin-fixed paraffin-embedded (FFPE) tumor tissue. The test used to select eligible patients for ovarian cancer with positive HRD status for the treatment with Zejula^® (niraparib) (22).

FoundationOne^®CDx (F1CDx) is a qualitative NGS- and high throughput hybridization-based capture test for the detection of indels, substitutions and copy number alterations (CNAs) in 324 genes. It detects gene rearrangements, genomic signatures including microsatellite instability (MSI), tumor mutational burden (TMB) and positive HRD status (somatic BRCA-positive and/or LOH high) using DNA isolated from FFPE tumor tissue. It provides definite information for the identification of eligible patients for specific treatments of different class using specific biomarkers for many solid tumors (23).

FoundationOne^® Liquid CDx is a qualitative NGS based test, which can identify indels, and substitutions in 311 genes, rearrangements in 4 genes and CNAs in 3 genes. It utilizes circulating cell-free DNA (cfDNA) isolated from plasma-driven peripheral whole blood collected in anti-coagulants. The test identifies patients that can benefit from different targeted treatments for NSCLC, breast, ovarian and prostate cancers based on specific biomarkers detected in each cancer. Negative result does not rule out the presence of an alteration in the patient’s tumor, in this case patients can opt for another tumor tissue-based CDx. The test analytical accuracy is not well demonstrated in all genes e.g., the test does not detect heterozygous deletions, and copy number losses/homozygous deletion in ATM (24).

HRDetect is a whole genome sequencing (WGS)-based classifier of HRD that can predict BRCA1 and 2 deficiency based on six mutational signatures (the HRD index [LOH + TAI + LST], microhomology-mediated indels, base-substitution signature 3 and 8, and rearrangement signature 3 and 5). It can also identify HRD in sporadic cancers (BRCAness) with and without any single detectable defect in HR genes (20, 25). HRDetect was shown highly sensitive method as compared to other HRD detection CDxs (20), but require clinical validation in independent set to avoid overfitting issue.

In this review, we did not classify mutation and hypermethylation data as bi-allelic or mono-allelic inactivation. Oftentimes, cases of pathogenic mutation in tumor suppressor genes lead to bi-allelic inactivation. Whole-exome sequencing analysis of breast cancer cases by Mutter et al. (26) revealed that 89% of bi-allelic inactivation results in HRD, whereas in cases of mono-allelic inactivation significant association existed between RAD51 functional status and LST. In a study by Li et al. (27) mono-allelic germline pathogenic mutation of PALB2 had predisposed to a high-risk breast cancer development, underscoring the role of PALB2 in HR repair. Moreover, protein-truncating variants and rare missense variants of DNA repair genes were significantly associated with the risk of breast cancer (16). Many findings confirmed the importance of haplo-insufficiency in tissue and gene specific manner; for instance, PTEN hypermorphic mice expressing 80% normal levels of PTEN protein was sufficient to predispose for different cancers development (28). Mono-allelic inactivation of TSGs (e.g., p53 and PTEN) leads to the inability to perform normal cellular functions which contributed to cancer development (29).

Here, we investigated the potential implications of pathogenic mutations, LOH, and promoter hypermethylation of HR-related genes using recent data in ovarian, breast, colorectal, pancreatic, non-small cell lung, and prostate cancers. Additionally, the mechanisms underlying PARPI resistance and possible strategies to overcome PARPI resistance are discussed.

The International Agency for Research on Cancer (IARC) (38) reports that ovarian cancer is the 8^th and 9^th cause of incidence and mortality worldwide in females, respectively. In 2020, the incidence and mortality of ovarian cancer were estimated at 3.4% (313,959 of all cancer types) and 4.68% (207,252 of all cancer types), respectively. Approximately 75% of epithelial ovarian cancer patients are diagnosed with advanced disease, which is curable in only a minority of cases, resulting in a modest 5-year survival rate of 20–30% (39). According to the cancer genome atlas (TCGA), high grade serous ovarian cancers (HGSOCs) are characterized by frequent genetic and epigenetic alterations of HR pathway genes, most commonly BRCA1 and BRCA2. In addition, approximately 50% of patients with HGSOC exhibit genetic and epigenetic alterations in the FANC-BRCA ( Figure 1 ) pathway (11). Germline mutations in the BRCA1 and BRCA2 genes are well-known mechanisms of HRD, and loss of BRCA1 or BRCA2 thus poses a significant risk to genome integrity, leading not only to cancer predisposition, but also affecting the sensitivity to DNA-damaging agents and thus therapeutic approaches (40). Pathogenic variants of BRCA1 and BRCA2 only explain the genetic cause of approximately 10% of hereditary breast and ovarian cancers (transmitted to offspring), underscoring the clinical importance of testing other DNA repair genes (41). Findings showed that BRCA1 and 2 inactivation frequently led to higher HRD score in ovarian and breast cancers. This high HRD score has positive prognostic significance for platinum and PARPI therapy. In unclassified ovarian cancer patients who undertook germline BRCA1 and 2 test, 19% (44/235) were carriers of germline mutations, and somatic mutation test was done on 28 specimens, 42.9% (9/21) and 28.6% (2/7) were found to be BRCA1 and 2 positives, respectively (42). In another study by Pennington et al. (43), among 367 ovarian carcinomas tested for somatic mutation, 2.5% (19/367) and 1.63% (6/367) were positive for BRCA1 and 2, respectively. These carriers of somatic mutation have shown a positive impact on overall survival and platinum responsiveness as germline BRCA1 and 2 mutation carriers. Other factors such as germline and somatic mutations in HR genes ( Table 1 ) and epigenetic alterations (promoter hypermethylation) are implicated in HRD (84).

Breast cancer is the most common cancer among women and ranks first and fifth among causes of incidence and mortality, respectively, compared with all other cancers in both sexes in 2020 worldwide. According to the IARC (38), an estimated 2,261,419 (11.72% of all cancer types) new incidences and 684,996 (6.88% of all cancer types) deaths from breast cancer were recorded worldwide in 2020. Similar to ovarian cancer, DNA repair pathways are frequently anomalous in breast cancer, leading to the accumulation of DNA damage and genomic instability. The BRCA1 and BRCA2 genes are associated with hereditary breast and ovarian cancer (85). BRCA1 and BRCA2 play a significant role in DNA repair, especially as components of the FANC/BRCA DNA damage response pathway ( Figure 1 ). This DNA repair pathway is a highly conserved system involved in the DSB response via HR and in the BER pathway for the repair of DNA single-strand breaks (SSBs) (9). The PARP enzyme plays a decisive role in this pathway and is critical for resolving the stalled replication forks. Inhibition of PARP during the base excision process requires BRCA-dependent HRR to resolve it. Targeting HR-related genes ( Table 2 ) has potential for destabilizing tumor genomic integrity (13). Clinical trials confirmed that HRD is necessary for the sensitivity to DNA-damaging agents (e.g., cisplatin) and PARPIs (106).

Although BRCA1 and BRCA2 have been studied extensively, other genes are also involved in the occurrence of breast cancer. In 3,388 breast cancer patients who underwent genetic testing for 25 genes, nearly half of the pathogenic variants were in the BRCA1 (24%) and BRCA2 (24.5%) genes. The remaining 51.5% of pathogenic variants were detected in other genes tested including CHEK2 (11.7%), ATM (9.7%), PALB2 (9.3%), and Lynch syndrome genes (7%); other genes accounted for the remaining 13.8%. The same study showed that pathogenic mutations in BRCA1, PALB2, BARD1, BRIP1, and RAD51C are significantly more prevalent than those of other genes in triple negative breast cancer (TNBC) (85). In a study by Lang et al. (144) using NGS based sequencing, the prevalence of somatic BRCA1 and 2 mutation in sporadic breast cancer cases which carries germline-BRCA (gBRCA) mutations was 3.5% (15/416). Among these, 1.9% (8/416) and 1.7% (7/416) were BRCA1 and 2, respectively. In the same study, somatic BRCA mutation in gBRCA-negative cases was not detected; indicating somatic BRCA mutation in gBRCA-negative cases is rare. The prevalence of pathogenic mutations, LOH, and promoter hypermethylation in breast cancer is summarized in Table 2 based on recently published data.

CRC is the third most common cancer worldwide. In 2020 alone, the incidence of CRC was estimated at 1,931,590 (10% of all other cancers) worldwide, and it is the second most common cause of cancer-related deaths 935,173 (9.39% of all other cancers) in both sexes after lung cancer (38). Most CRC occurrences are sporadic and are not related to genetic predisposition or family history; however, 20–30% of patients with CRC have a positive family history, and 5% of these tumors arise from genetic predisposition (145). Mutation is a frequent event in CRC. According to TCGA (146), 16% of colorectal carcinomas are hyper-mutated; of these, 75% have high microsatellite instability, typically with hypermethylation and MLH1 silencing, and 25% have somatic mismatch repair gene and polymerase ϵ (POLE) mutations. The prevalence of mutations in APC, TP53, SMAD4, and PIK3CA was also reported, as well as KRAS mutations. Recurrent copy number alterations (e.g., ERBB2 amplification), chromosomal translocations, such as the fusion of NAV2 and the WNT pathway member TCF7L1, and biallelic inactivation of APC were among the common features leading to LOH in CRC. These and other findings confirmed the differences in both oncogenes and tumor suppressor genes involved in CRC compared with those in breast and ovarian cancers. However, a recent study from Japan revealed moderate genomic alterations in DNA repair genes including BRCA2, ATM, and NBN in CRC patients (147). High penetrance genes in HBOC differ from those in CRC; high penetrance genes in CRC are APC, MLH1, MSH2, MSH2/MUTYH, SMAD4, MAP, and APC/PMS2; moderate penetrance genes are MSH6 and PMS2; and low penetrance genes are APC p.I1307K and MUTYH mono-allelic (148).

The methylation profile of CRC differs from that of other cancers; it is characterized by global hypomethylation and promoter-specific DNA hypermethylation. At the whole genome level, CRC has 10–40% lower levels of absolute methylation than normal colonic tissue (149, 150). Global DNA hypomethylation, which is accompanied by genomic instability and tumor initiation, is primarily due to loss of methylation within repetitive elements such as long interspersed nuclear element-1 (LINE-1) and Arthrobacter luteus restriction endonucleases (Alu), and it is supposed to contribute to CRC initiation by enhancing genomic instability. On the other hand, genes methylated in CRC are established tumor suppressor genes, and 50% of these are also methylated in the normal colonic epithelium. As a result, methylation in CRC can have three phenotypes according to the CpG-island methylator phenotype (CIMP), i.e., CIMP-high, CIMP-low, and non-CIMP tumors (151). Similar to breast and ovarian cancers, the prevalence of mutation, LOH, and promoter hypermethylation in 35 HR-related genes is described based on recently published data ( Table 3 ).

Pancreatic cancer (PC) is one of the most devastating types of cancer. PC has a low 5-year survival rate of 9%, and the development of new therapeutics is urgent. The global incidence and mortality of PC in 2020 were 2.6% (495,773 of all cancer types) and 4.68% (466,003 of all cancer types), respectively. PC is the 12th and 6th leading cause of cancer-related incidence and deaths in the world, respectively (38). Although most PC cases occur sporadically, familial (individual having two or more first-degree relatives diagnosed with PC) and hereditary syndrome PC account for 10% of cases. Hereditary cancer syndromes associated with increased risk of PC include Lynch syndrome, familial adenomatous polyposis, Peutz-Jeghers syndrome, atypical multiple mole melanoma, and familial and hereditary breast and ovarian cancer syndrome. Hereditary cancer syndrome accounts for 3% of PC cases, whereas familial PC accounts for 4–10% of cases (171, 172). Familial PC includes individuals with two or more affected first-degree relatives with PC excluding patients with hereditary syndrome. The risk of PC increases with the number of affected first-degree relatives. A study of 838 families including 5,179 individuals showed that the relative PC risk was 4.5 [95% confidence interval (CI): 0.5–16.3) among 1,253 cases with one affected first-degree relative, 6.4 (95% CI: 1.8–16.4) among 634 cases with two affected first-degree relatives, and 32 (95% CI: 10.4–74.7) in 106 cases with three or more affected first-degree relatives (173).

The specific hereditary pancreatic susceptibility genes are PRSS1, SPINK1, GGT1, CTRC, and CFTR, and mutations in these genes cause early onset PC. However, mutations in these genes are rare and account for a small proportion of PC cases, although the cumulative risk at 70 years age reaches 7–40% with early onset PC (174). Familial PC susceptibility genes include BRCA2, ATM, PALB2, CDKN2A, PRSS1, STK11, MLH1, and MSH2 (175). Recent findings showed the association of PC with genetic alterations in BRCA2, BRCA1, ATM, CHECK2, PALB2, FANCC, and CDKN2A genes (176–178). Mutations in BRCA2 are among the most common genetic mutations involved in familial pancreatic ductal carcinoma. BRCA mutation predisposes to PC, and PC more frequently affects BRCA2 mutation carriers than BRCA1 carriers. Among 204 BRCA mutation carriers with PC, 42.7% (87/204) had BRCA1 mutations and 57.3% (117/204) had BRCA2 mutations (179). The prevalence of mutation, LOH, and promoter hypermethylation of 35 HR genes is described according to recently published data in PC ( Table 4 ).

Lung cancer is the leading cause of cancer-related morbidity and mortality in both genders worldwide. According to the IARC, the estimated number of incidences and deaths from lung cancer was 2,206,771 (11.44%) and 1,796,144 (18.04%) of all cancers worldwide in 2020, respectively (38). The majority of lung cancer cases are associated with smoking or the use of different tobacco products, although other factors such as asbestos, air pollution, radon gas exposure, and chronic infection also contribute to lung carcinogenesis (194). Inherited and acquired mechanisms of lung cancer susceptibility have been proposed, although they are rare. For instance, germline T790M mutation predisposes to a unique hereditary lung cancer syndrome that affects never-smokers and accounts for 31% of the estimated risks for lung cancer in never-smoker carriers (195, 196). Lung cancer is highly invasive, rapidly metastasizing, and broadly categorized into two histological groups, small-cell lung carcinomas (SCLCs) and NSCLCs, which grow and spread differently. NSCLCs account for 87% of cases and can be subdivided into three or four subtypes (adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and undifferentiated NSCLCs), whereas SCLCs account for 12% of lung cancer cases (194).

Generally, lung adenocarcinoma is characterized by recurrent aberrations in multiple key pathways, including activation of RTK/RAS/RAF; activation of PI3K-mTOR; alterations of p53, cell cycle regulators, and the oxidative stress pathway; and mutation of various chromatin and RNA splicing factors. The research network of TCGA demonstrated the activation of oncogenes including KRAS (32%), EGFR (11%), MET (7%), BRAF (7%), MDM2 (8%), CDK4 (7%), PIK3C4 (4%), and CCND1 (4%) and the inactivation of tumor suppressor genes such as TP53 (46%), CDKN2A (43%), KEAP1 (19%), STK11 (17%), NF1 (11%), ATM (9%), RBM10 (9%), ARID1 (7%), ARID2 (7%), and RB1 (7%) in lung cancer (197). Analysis of DNA repair genes associated with squamous cell carcinoma showed a correlation between pathogenic mutations of DNA repair genes and tumor mutation burden. Among DNA repair genes BRCA1and BRCA2 showed the greatest mutation frequency and the tumor burden increased in correlation with the number of affected DNA repair genes (198). A study analyzing mutations in the BRCA 1 and 2 genes in different cancers showed that BRCA mutation is associated with increased incidence of non-breast and ovarian cancers in first- and second-degree relatives of high-risk breast cancer patients. Among 337 BRCA mutation carriers, the second highest BRCA mutation rate was recorded in lung cancer [8.8% (33/337)] after stomach cancer [13.8% (52/337)] (199). In this review, we displayed the prevalence of mutation, LOH, and promoter hypermethylation of 35 HR genes in NSCLC ( Table 5 ).

Prostate cancer is the most prevalent cancer in men next to lung cancer, according to the IARC in 2020 (38), it ranked 2^nd and 5^th in its incidence and mortality among men, respectively. The estimated number of incidences and deaths from prostate cancer in 2020 were 1,414,259 (14.05%) and 375,304 (6.79%) worldwide, respectively. Prostate cancer is characterized by a high degree of heritability, and genetic components contribute significantly to disease incidence (210). A large cohort study conducted in the Nordic region that analyzed the different cancer heritability risks in monozygotic and dizygotic twins identified a risk of prostate cancer of 57% (95% CI: 51–63), which was higher than that of other cancer types such as ovarian cancer at 39% (95% CI: 23–55) and breast cancer at 31% (95% CI: 11–51) (210). Although the high rate of heritability of prostate cancer has been demonstrated in patients with a positive family history, candidate genes that contribute to prostate cancer heritability have not been identified except HOXB13 (211). Recurrent mutation of the HOXB13 gene at G84E was identified in many families. The HOXB13 G84E allele accounts for approximately 5% of hereditary prostate cancer (211). A study including 2,443 prostate cancer families of European descent detected at least one HOXB13 G84E mutation carrier, among 112 prostate cancer families (4.6%) (212). Moreover, a study conducted in unrelated subjects of European descent revealed, HOXB13 G84E mutation was detected in 1.4% (72/5083) and 0.07% (1/1401) of participants with- and without prostate cancer, respectively (P<0.05) (213). Another study comprising 9,012 men diagnosed with different cancers showed a rate of 0.54% (49/9012) of HOXB13 G84E mutation carriers, of whom 1.4% (19/1362) were positive for prostate cancer compared with 0.4%(23/5,898) of HOXB13 G84E mutation carriers without prostate cancer (p < 0.05) (214). Prostate cancer has a genetic origin in <5% of cases, and this risk becomes higher when high penetrance genes such as HOXB13 are involved (215). Recent gene linkage studies identified additional prostate cancer susceptibility genes such as HPC1, HPC2/ELAC2, MSR1, BRCA1, BRCA2, and BRIP1 (216).

An estimated 20% of patients with prostate cancer have a positive family history, which can be attributed not only to shared genes, but also to a shared pattern of exposure to environmental carcinogens and common lifestyle habits (216). Additional challenges in the management of prostate cancer include its genetic heterogeneity and a high rate of sporadic cases; many common genetic variants are associated with prostate cancer, explaining the familial clustering of the disease rather than hereditary causes (211). The importance of both germline and somatic alterations in DNA repair genes is suggested by the fact that carriers of mutations in these genes are at a high risk of developing aggressive or metastatic prostate cancer (211). Deleterious mutations of BRCA1 and BRCA2 are associated with increased risk of prostate cancer and experienced very aggressive course of the disease (215). However, studies focusing on families with only prostate cancer failed to identify a significant number of BRCA1 or BRCA2 mutations, indicating their minimal role in hereditary prostate cancer predisposition (211). A comprehensive genomic analysis of 1,013 prostate cancer patients revealed the indispensable role of alterations in DNA repair genes (78). Here, the prevalence of mutation, LOH, and promoter hypermethylation in 35 DNA repair genes in prostate cancer was described based on recently published data ( Table 6 ).

HRR is the guardian of the genome because of its role in repairing DSBs with high fidelity. Defects in HR due to mutation, LOH, and promoter hypermethylation of certain HR genes result in HRD, which confers sensitivity to DNA-damaging agents and PARPIs. This review demonstrated that HRD is higher in ovarian and breast cancers than in other cancer types such as CRC, PC, NSCLC, and prostate cancer. HRD is not limited to BRCA1 and 2, and comprises many DNA repair genes. The fundamental vulnerability of HRD has led to the design of a wide range of HRD-directed therapies. DNA repair targeted therapies exploit DNA repair defects because HR-deficient tumors are intrinsically sensitive to PARPIs. This highlights the concept of synthetic lethality associated with the concurrent inactivation of two or more HRR genes. The use of PARPIs in non-BRCA mutation carriers can be expanded to sporadic cancers that display DNA repair defects. PARPIs are essential for the treatment of ovarian and breast cancers. Recently, FDA approved olaparib for the treatment of prostate and pancreatic cancers characterized by HRD. However, the benefits of PARPIs are limited by the development of resistance, especially when used as monotherapy. Many mechanisms of resistance to PARPI have been identified in HR-deficient cancers, which are challenges to overcome. Numerous preclinical and clinical studies revealed that combination therapy of PARPI with targeted chemotherapy or ICIs improved the efficacy by overcoming PARPI resistance. Understanding the mechanisms of PARPIs resistance will be useful for designing strategies to overcome PARPI resistance as summarized in this review. Based on accumulated research, more potential PARPIs and more effective combined regimens targeting HR-deficient cancers would be developed in the future.

NM, collected data, sketched the initial version, and reviewed the final manuscript. HY, reviewed and drafted the final version of the manuscript. YS, conceived the idea, guided the write up, and reviewed the final manuscript. All authors contributed to the article and approved the submitted version.

This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant number: HI17C2196).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.