The role of MGMT is to remove alkyl adducts from the O⁶ position of guanine. Thus, the protective effect of MGMT could diminish the cytotoxic effects of alkylating agents (Middleton and Margison, 2003), suggesting that MGMT activity is likely to be a useful marker of the sensitivity of cancer cells to alkylating agents. It has been reported that high MGMT expression in tumor cell is associated with the resistance to 1,3- bis- (2-chloroethyl) -1- nitrosourea (BCNU) and temozolomide (TMZ) (Happold et al., 2018; Hsu et al., 2018), which target the O⁶-position of guanine, resulting in cytotoxic and mutagenic DNA adducts (Rabik et al., 2006). Recently, researchers found that MGMT-mediated the resistance to DNA alkylating agents in cancer cell is profoundly dependent on the DNA repair enzyme PARP. Combination of temozolomide with PARP inhibitors (PARPi) in MGMT-positive cancer cells enhanced the anticancer effects (Erice et al., 2015; Jue et al., 2017).

The inactivation of MGMT in tumor cells has been appreciated as a therapeutic target for sensitizing cells to O⁶-alkylating agents (Maki et al., 2005). In vitro and in vivo studies demonstrated that O⁶-Benzylguanine (O⁶-BG), a typical pseudo-substrate that was developed to inactivate MGMT, in combination with O⁶-alkylating agents increased the therapeutic efficacy of chemotherapeutic alkylating agents (Maki, Murakami, 2005). Lomeguatrib (called O⁶-(4-bromothenyl) guanine, as well as PaTrin-2), another pseudo-substrate tested in clinical trials, has been shown to increase the therapeutic index of methylating agent temozolomide in nude mice bearing A375M human melanoma xenografts and patients with advanced solid tumors (Middleton et al., 2002; Ranson et al., 2006). Bobustuc GC et al. demonstrated that inhibition of MGMT suppressed the expression of survivin and enhanced the cytotoxicity of gemcitabine in pancreatic cancer (Bobustuc et al., 2015). Another approach to MGMT inactivation is to silence the MGMT gene expression through its promoter methylation. Several studies in animal models have suggested that the therapy of MGMT gene silence was able to overcome TMZ resistance and increase tumor cell death (Viel et al., 2013). Clinical study indicated that patients with glioblastoma containing a methylated MGMT promoter obtained more benefits from TMZ than those who did not have a methylated MGMT promoter (Hegi et al., 2005). Lately, it has been confirmed that MGMT gene methylation can be a biomarker for temozolomide (TMZ) treatment and a potent prognostic factor in patients with GBM (Kim et al., 2012; Iaccarino et al., 2015; Zhao et al., 2016; Binabaj et al., 2018). However, according to the data from National Cancer database (NCDB) indicated that only 4.9% of GBM patients have MGMT promoter methylation. Even though MGMT promoter methylation status has prognostic value, it is ignored in the United States (Lee et al., 2018). More researches need to conduct to identify the prognostic value of MGMT promoter methylation in tumor patients responding to alkylating agents.

A number of investigations have shown that inhibition of BER pathway can enhance the sensitivity of cancer cells to alkylating agents and radiotherapy (Neijenhuis et al., 2005; Gao et al., 2019). The primary methods to prevent the activity of BER pathway focus on the development of AP endonuclease 1 (APE1) or Poly (ADP-ribose) polymerase (PARP) inhibitors.

Several studies indicated that methoxyamine (MX), a small alkoxyamine that can bind with the free aldehyde of AP site to prevent APE1 cleavage at AP sites, thereby inhibiting APE-1 endonuclease activity. Combined treatment with chemotherapeutic alkylating agent such as TMZ and BCNU could reinforce the cytotoxicity of alkylating agent by targeting BER pathway (Liu et al., 2003; Montaldi and Sakamoto-Hojo, 2013). Recently, based on preclinical studies, several clinical trials were conducted, for example combination therapy with MX and TMZ in patients with advanced solid tumors has completed (NCT00892385). Currently, phase Ⅰ clinical trials of MX in combination of TMZ is undergoing in patients with relapsed solid tumors and lymphomas (NCT01851369). MX combination with pemetrexed disodium, cisplatin, is now investigating in phase Ⅰ/II stage in patients with advanced malignant solid neoplasm (NCT02535312). Lucanthone, a topoisomerase II inhibitor as well as an APE1 endonuclease inhibitor, has been shown to reinforce the cell killing effect of alkylating agents in human breast cancer cell line MDA-MB-231 (Luo and Kelley, 2004). Lucanthone combination with radiation and TMZ in GBM patients was tested in phase Ⅱ clinical trial (NCT01587144). However, it was terminated in 2016. Another phase II clinical trial investigating lucanthone combination with radiation in patients with brain metastases from non-small cell lung cancer was withdrawn due to drug issues (NCT02014545).

PARP family is composed of 17 members, of which PARP1 and PARP2 are well-recognized DNA damage sensors, especially PARP1. PARP1 detect the region of damaged DNA and play a key role in several DNA repair pathway including BER, HHR and MMEJ (Konecny and Kristeleit, 2016). While PARP1 is best studied in BER and the mechanism of PARP inhibitor (PARPi) is based on trapping PARP1 on SSBs DNA site to inhibit BER repair. Finally, it converted SSBs into DSBs and impelled cell death in HR-deficiency tumor, for example BRCA1/2 mutations, RAD51 deficiency (Figure 2) (Konecny and Kristeleit, 2016; Brown et al., 2017; Lord and Ashworth, 2017; Oplustil O’Connor et al., 2016). In 2005, two pre-clinical researches published in nature indicated that BRCA1 or BRCA2 deficient cells highly sensitized to PARP inhibition (Farmer et al., 2005; Bryant et al., 2005). Based on the concept of “synthetic lethality”-targeting either gene alone in a synthetic lethal pair is tolerated, but simultaneous targeting both genes is lethal, researchers applied PARPi to BRCA mutation tumors (Dhillon et al., 2016). Several clinical trials using PARPi including Olaparib, Veliparib, Rucaparib (Table 1) as monotherapy for the treatment of patients with germline BRCA1/2 mutation tumors including advanced breast cancer, ovarian cancer, pancreatic cancer and prostate cancer presented significantly antitumor effect (Kaufman et al., 2015; Robson et al., 2017; Moore et al., 2018; Golan et al., 2019). Olaparib as maintenance therapy also significantly prolonged progression-free survival in advanced ovarian cancer patients with HRD-positive tumors who have achieved first-line standard therapy including bevacizumab. It has been approved by FDA for utilization of Olaparib in patients with advanced germline BRCA-mutated ovarian cancer following three or more prior lines of chemotherapy (Kim et al., 2015). On May 19, 2020, the FDA also approved Olaparib for patients with metastatic castration-resistant prostate cancer (mCRPC) carrying HRR gene-mutated based on NCT02987543. PAPR1 inhibitors in combination with IR or with other different anticancer agents are currently undergoing clinical trials for treatment of patients with BRCA1/2 mutation or HRR-deficiency advanced solid tumors, which shown promising clinical activity (Bang et al., 2017; Wilson et al., 2017; Loibl et al., 2018; Coleman et al., 2019; Farago et al., 2019; Konstantinopoulos et al., 2019; Liu et al., 2019).

Among various DNA lesions, DSBs is the leading lethal damage that leads to cell death and genetic mutations. DNA-dependent protein kinase (DNA-PK), a member of the PI3K-related protein kinase (PIKK) family, is involved in DSBs repair pathway via non-homologous end joining (NHEJ) (Jette and Lees-Miller, 2015). It has been reported that DNA-PK activity plays a role in chemo-radiotherapy resistance (Wang Y. et al., 2018; Stefanski et al., 2019; Alikarami et al., 2017; Liu et al., 2015). Selective DNA-PK inhibitor have been developed, including NU7026 (Dolman et al., 2015), NU7441 (Yang et al., 2016), IC87361 and SU11752 (Shinohara et al., 2005). They could inhibit DSBs repair pathway and enhance the sensitivity of cancer cells to ionizing radiation or/and chemo-potentiation such as doxorubicin (Ciszewski et al., 2014). The combination of DNA-PK inhibitor M3814 with type II topoisomerase inhibitors, including doxorubicin, etoposide and pegylated liposomal doxorubicin, enhanced the efficacy of type II topoisomerase inhibitors in ovarian cancer xenografts (Wise et al., 2019). Several novel DNA-PK inhibitors including MSC2490484A, VX-984 (M9831), M3814 are under clinical trial as single-agent or combination with Chemo-radiotherapy (Table 2). Alexander K. Tsai et al. recently found that NU7441 combination with a multikinase inhibitor regorafenib altered immune microenvironment of melanomas and enhanced the efficacy of various immunotherapies (Tsai et al., 2017).

Ataxia-teleangectasia mutated (ATM) and ATM-RAD3-related (ATR) protein, like DNA-PK protein, are the members of PIKK family. They work as a transducer of the DSB signal, and are involved in the repair of DNA DSBs (Weber and Ryan, 2015). A large of ATM inhibitors, including KU-55933, KU-60019, KU-59403, CP-466722, AZ31, AZ32, AZD0156, and AZD1390, have been developed and their antitumor effects have been investigated (Jin and Oh, 2019). It has been reported that human tumor cells treated with KU-55933, a specific inhibitor of the ATM kinase, could sensitize tumor cells to the cytotoxic effects of IR and DNA DSBs-inducing chemotherapeutic agents such as etoposide, doxorubicin, and camptothecin (Hickson et al., 2004; Hoey et al., 2018). KU-60019, an improved ATM kinase inhibitor, acts as a highly effective radio-sensitizer in human glioma cells (Biddlestone-Thorpe et al., 2013). AZD0156, a newly discovered ATM inhibitor, has the potential to promote the survival of leukemia-bearing mice and now is under clinical trial (Morgado-Palacin et al., 2016). Preclinical study demonstrated that ATM inhibitor AZD1390 enhanced the radiosensitivity of tumor cells and extended animal survival in preclinical brain tumor models (Durant et al., 2018). AZD1390, as a radiosensitizer, is now undergoing two clinical trials in patients with brain cancer (NCT03423628) or non small cell lung cancer (NCT04550104). Many inhibitors aiming at both ATM and DNA-PK have been reported to have great potential as a chemo- and radiotherapy sensitizing agents in cancer therapy (Powell and Bindra, 2009).

The cell cycle checkpoint kinases CHK1 and CHK2 are downstream substrates of ATM /ATR, which act as the “central transducers” of the DDR (Pilie et al., 2019). Activation of these pathways is essential for the proper regulation of checkpoint and DNA repair (Smith et al., 2010). The ATM–Chk2 and ATR–Chk1 pathways respond to different DNA damages, ATM is activated at DSBs, whereas ATR is recruited to tracts of ssDNA (Di Benedetto et al., 2017). Subsequently, CHK1 and CHK2 activated by ATR and ATM respectively upon their recruitment to DNA damage sites. Protein kinase WEE1 functioned as furthest downstream in ATR/CHK1 pathway, which is indirectly regulated by DNA damage (Cleary et al., 2020). WEE1 actives the G2/M cell cycle checkpoint by impeding cyclin-dependent kinase 1 and 2 (CDK1/2) activity, thereby inducing cell cycle arrest and promoting DNA damage repair. Inhibition of WEE1 causes aberrant DNA replication and replication-dependent DNA damage in cells by suppressing CDK2 (Guertin et al., 2013). Recently, compounds targeting CHK1 are currently in clinical trials (Table 1). The first-in-class WEE1 kinase inhibitor AZD1775 is also undergoing a series of clinical trials as monotherapy or in combination with other therapies (Table 1).

Recently, the exploration of novel anticancer strategies aiming at the differences in mitochondrial function and structure between normal cells and cancer cells has received intensive attention (Porporato et al., 2018). However, there are few studies that have discovered new anticancer approaches via targeting mtDNA repair pathway.

Like nDNA, efficient mtDNA repair pathway, especially BER pathway that mainly repairs ROS-induced lesion, may play an important role in cellular resistance to cancer therapeutic agents. MtDNA D-loop mutations were common in gastrointestinal cancer and correlated with carcinoma progression (Wang B. et al., 2018). It has been found that human breast cancer cells defective of mtDNA repair are more sensitive to oxidative damage than the control cells (Shokolenko et al., 2003). Grishko V I et al indicated that mtDNA repair pathways played an important role in protecting cells against ROS in normal HA1 Chinese hamster fibroblasts (Grishko et al., 2005). Another study clarified that mtDNA repair capacity was important for cellular resistance to oxidative damage by increasing their viability following exposure to oxidative stress (Shokolenko et al., 2003). Ueta E et al demonstrated that downregulation of the mtDNA repair-associated molecules, mitochondrial transcription factor A (mtTFA) and Polγ by using inhibitors of PI3K/Akt signaling in oral squamous cell carcinoma cells (OSC) increased the susceptibility of radio-sensitive OSC cells and radio-resistant OSC cells to gamma-rays (Ueta et al., 2008). This observation implied that PI3K/Akt signal inhibitors can suppress mtDNA repair capacity. Thus, these inhibitors combined with ionizing irradiation or chemotherapeutic drugs may be utilized as an effective strategy in cancer therapy.

DNA glycosylases are involved in the initiation step of BER that recognizes and removes the abnormal base (Anderson and Friedberg, 1980). 8-OxoG-recognizing DNA glycosylase 1 (OGG1) is an important DNA glycosylase for repair of 8-oxoguanine (8-oxoG), which is one of the major DNA lesions both of the nDNA and mtDNA, especially in mtDNA (Rachek et al., 2002). It has been found that tumor cells harboring overexpressed recombinant OGG1 were more proficient at repairing of oxidative damage to mtDNA, and had increased cellular survival under oxidative stress (Rachek et al., 2002; Yuzefovych et al., 2016). We previously found that Sirt3, a major mitochondrial NAD⁺-dependent deacetylase, physically associated with OGG1 and deacetylated this DNA glycosylase, and that deacetylation by Sirt3 prevented the degradation of the OGG1 protein and controlled its incision activity (Cheng et al., 2013). We further showed that regulation of the acetylation and turnover of OGG1 by Sirt3 played a critical role in repairing mitochondrial DNA (mtDNA) damage, protecting mitochondrial integrity, and preventing apoptotic cell death under oxidative stress. We observed that following ionizing radiation, human tumor cells with silencing of Sirt3 expression exhibited oxidative damage of mtDNA, as measured by the accumulation of 8-oxoG and 4,977 common deletion, showed more severe mitochondrial dysfunction, and underwent greater apoptosis, in comparison to the cells without silencing of Sirt3 expression. Our results not only reveal a new function and mechanism for Sirt3 in defending the mitochondrial genome against oxidative damage and in protecting from the genotoxic stress-induced apoptotic cell death, but also provide evidence supporting a new mtDNA repair pathway. Recently, researchers also proved that overexpression of mitochondrial OGG1 decreased breast cancer progression and metastasis (Yuzefovych et al., 2016).

In conclusion, combination of DNA repair pathway inhibitors with anticancer agents may enhance the tumor sensitivity to certain chemotherapeutic drugs and radiation. More effective and less toxic DNA-damaging agents have been developed and carried out in preclinical studies (Table 2). Based on the preclinical data, a number of clinical trials have been launched to test whether targeting DNA repair pathways can reinforce the efficacy of some anticancer drugs and benefit cancer patients (Table 1).