Targeting the Replication Stress Response through Synthetic Lethal Strategies in Cancer Medicine

Abstract

The replication stress response (RSR) involves a downstream kinase cascade comprising ataxia telangiectasia-mutated (ATM), ATM and rad3-related (ATR), checkpoint kinases 1 and 2 (CHK1/2) and WEE1-like protein kinase (WEE1), which cooperate to arrest the cell cycle, protect stalled forks and allow time for replication fork repair. In the presence of elevated replicative stress, cancers are increasingly dependent on RSR to maintain genomic integrity. An increasing number of drug candidates targeting key RSR nodes, as monotherapy through synthetic lethality, or through rational combinations with immune checkpoint inhibitors and targeted therapies- are demonstrating promising efficacy in early phase trials. RSR-targeting is also showing potential in reversing PARP inhibitor resistance, an important area of unmet clinical need. In this review, we introduce the concept of targeting the RSR, detail the current landscape of monotherapy and combination strategies, and discuss emerging therapeutic approaches, such as targeting Polθ.

Replicative Stress and Other Relevant Genotoxic Sensors

Cells rely on coordinated pathways to repair endogenous and exogenous DNA damage, maintain genomic integrity and promote cell survival[1]. When DNA replication occurs aberrantly, DNA single strand breaks (SSBs) or double strand breaks (DSBs) accumulate. Subsequent fusion of DSBs and shortened telomeres may result in clinically significant translocations and gene amplifications[2–4]. Such DNA lesions distort the DNA double helix, leading to alterations in the transcription and replication of genetic material. DNA damage response (DDR) pathways provide a mechanism for the recognition of such damage and for the halting of the cell cycle in order to repair and preserve normal genomic architecture[5–8]. SSBs are generally repaired through three different pathways: (1) nucleotide excision repair, (2) base excision repair and (3) mismatch repair[5], while DSBs are repaired by either the error-free homologous recombination (HR) pathway, which replaces damaged DNA using the sister chromatid as a template, or the less reliable mechanisms of polymerase theta-mediated end joining (TMEJ) and non-homologous end joining (NHEJ) (Figure 1). Defects in these DDR mechanisms allow for high levels of DNA damage and genomic instability that not only alter gene function, but also lead to uncontrolled cellular growth. The activation of oncogenes or inactivation of tumor suppressor genes may ultimately lead to the development of cancer[5, 6, 8].

Figure 1: The Replication Stress Response.

The replication stress response (RSR). DNA damage or stalled replication forks occurring due to endogenous or exogenous factors trigger the RSR and downstream kinase cascade. ATM and rad3-related (ATR) and ataxia telangiectasia-mutated (ATM) primarily phosphorylate checkpoint kinase 1 (CHK1) and 2 (CHK2), respectively, which suppress CDC25C/A, preventing smooth transition through cell cycle checkpoints. WEE1-like protein kinase (WEE1) negatively regulates CDK2/ Cyclin E and CDK1/Cyclin B and is itself activated by CHK1/2. These kinases represent key nodes for therapeutic targeting of the RSR. Created with Biorender.com. Abbreviations: ATRi, ATR inhibitor; CCL5, chemokine ligand 5; CXCL10, C-X-C motif chemokine ligand 10; cGAS, cyclic GMP-AMP synthase; G2, gap 1; G2, gap 2; IFN, interferon; IRF3, interferon regulatory factor 3; M, mitosis; S, synthesis; STING, stimulator of interferon genes; TBK1, tank binding kinase 1.

Endogenous and exogenous genotoxic stressors leading to DNA damage and replicative errors in cells culminate in the activation of the DDR system in order to maintain the fidelity of DNA replication and genomic integrity. DDR is commonly considered a kinase cascade, orchestrated by the DNA damage sensors ataxia telangiectasia-mutated (ATM) and ATM and rad3-related (ATR) kinase, which serve as master regulators of DNA repair, replication and cell cycle progression. In response to genotoxic stressors that hamper DNA replication and stall replication forks, cells enter a state of ‘replication stress’ (RS), which leads to the activation of the DDR pathways and subsequent inhibition of cell cycle progression. This is known as the replicative stress response (RSR), with the ultimate aim of slowing DNA synthesis and replication to allow time for DNA repair. Perturbation of the RSR may thus lead to the induction of apoptosis or cellular senescence[9]. DSBs primarily trigger activation of ATM and DNA-dependent protein kinase (DNA-PK), while ATR and downstream checkpoint kinase 1 (CHK1) and WEE1-like protein kinase (WEE1) induce cell cycle arrest and work to restore stalled replication forks in response to exposed single-stranded DNA (ssDNA) or SSBs (Figure 1). During RSR, continued unwinding of the DNA helix by the MCM helicases leads to exposed ssDNA at stalled replication forks, which are coated with replication protein A (RPA)[10]. Proliferating cell nuclear antigen (PCNA) is displaced and a fork protection complex consisting of Timeless and Tipin heterodimers binds to RPA to stabilize the stalled fork. Together with ATR-interacting protein (ATRIP), ATR is recruited to RPA and is activated upon associating with DNA topoisomerase 2-binding protein 1 (TOPB1) or ETAA1 activator of ATR kinase (ETAA1). At that point, ATR phosphorylates several downstream targets including histone H2AX (H2AX) and CHK1[11]. The activated ATR-CHK1 pathway suppresses cell cycle progression by inactivating the cell division cycle 25 (CDC25) phosphatase family, which usually serve as cyclin dependent kinase (CDK) activators[12]. CHK1 activates WEE1, which in turn activates the intra- Synthesis (S) checkpoint and Gap 2 (G2)/ Mitosis (M) checkpoints by negatively regulating the CDK2/cyclin and the CDK1/cyclin complex-mediated cell cycle progression[13, 14]. ATR and CHK1 promote fork stabilization and cell cycle checkpoint activation as well as prevent firing of local and global replication origins until stalled forks are resolved[15]. This is important in order to shield active forks against irreversible breakage by preventing exhaustion of nuclear RPA[16]. Finally, ATR modulates fork reversal and restart by phosphorylating the fork remodeler SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A like 1 (SMARCAL1)[17] as well as by controlling nucleotide supply[18]. H2AX, phosphorylated by ATR, recruits RAD51 at reversed forks to protect newly synthesized DNA tracts from nuclease digestion and assists with fork restart[19]. These steps ultimately prevent DNA damage from being carried forward toward mitosis[20].

In cancer cells, upregulated oncogenic drivers, loss of critical tumor suppressors, and underlying defects in DDR mechanisms lead to unchecked proliferative signals, accumulation of genomic error, increased cellular RS, provoking carcinogenesis and tumor progression[21]. In this context, the upregulation of the RSR becomes crucial for tumor viability; paradoxically, it also becomes a targetable vulnerability for exploitation in cancer therapy because inhibition of ATM, DNA-PK or the ATR-CHK1-WEE1 axis can disrupt the RSR in tumors. Such disruption further enhances the RS state and allows cell cycle progression, despite high levels of unrepaired DNA damage and accumulation of stalled replication forks in the genome, thus triggering mitotic catastrophe and apoptosis. Recently, an increasing number of drug candidates targeting key RSR nodes have entered developmental pipelines. In this review, we discuss the current landscape of synthetic lethal RSR-targeting and we highlight the most promising emerging therapeutic approaches.

Targeting the RSR

Small molecule inhibitors targeting components of the RSR, such as ATM, ATR, DNA-PK, CHK1 and WEE1, are under development and have shown antitumor activity in preclinical and clinical studies (Table 1). ATR inhibitors (ATRi) under development [berzosertib (M6620, VX-970, Merck Serono), ceralasertib (AZD6738, AstraZeneca), elimusertib (BAY1895344, Bayer), M4344 (Merck Serono), ART0380 (Artios Pharma), RP-3500 (Repare Therapeutics)], have demonstrated anti-cancer activity, especially in cells known to have high RS levels and increased ATR-CHK1 dependency, such as from oncogenic RAS activation, CCNE1 or c-MYC amplification[22]. Similarly, inhibition of the ATR-CHK1 pathway has been shown to induce synthetic lethality in cells with inactivated tumor suppressors such as TP53, ARID1A or ATM, [23, 24]. While early phase clinical trials showed limited monotherapy activity in unselected patients with berzosertib and ceralasertib, durable objective responses were observed in selected patients harboring ATM mutations or ATM loss (determined via immunohistochemistry) that were treated with berzosertib, ceralasertib and elimusertib[25–29]. Clinical trials in biomarker-selected patient populations are underway (Table 1).

Table 1:

Key completed and ongoing trials investigating inhibitors of ATM, ATR, DNA-PK, CHK1 and WEE1 as monotherapy or in combination.^a

Drug	Phase	Study population	Selective biomarker	Treatment	Efficacy	Safety	Reference
ATR inhibitors
Berzosertib (M6620; VX-970)	I	Advanced solid tumors; N= 40	N/A	Berzosertib once or twice weekly Berzosertib (D2, D9) + carboplatin AUC 5 (D1) q21 days	RP2D for once- or twice-weekly administration: 240mg/m2 RP2D for combination with carboplatin: 90mg/m2 CR and PFS of 29 months in a patient with metastatic colorectal cancer harboring ATM loss and ARID1A mutation PR and GCIG CA125 response in a patient with BRCA1-mut ovarian cancer	No DLT for monotherapy Grade ≥3 AEs of neutropenia (21.7%), anemia (4.3%) and thrombocytopenia (4.3%) were observed for combination therapy	NCT02157792
	II	Platinum-resistant ovarian cancer; N=88	N/A	Gemcitabine 1000 mg/m2 (D1, D8) +/− berzosertib 210mg/m2 (D2, D9) q21 days	PFS: 22.9 weeks vs 14.7 weeks (HR 0.57; 90% CI 0.33–0.98; p = 0.044)	Grade ≥3 AEs for gemcitabine + berzosertib were neutropenia (47%), thrombocytopenia (24%) 1 treatment related death in gemcitabine + berzosertib arm due to pneumonitis SAEs observed in 28% of patients in gemcitabine alone versus 26% of patients in gemcitabine + berzosertib arms	NCT02595892
	I	Esophageal cancers; N= 65	N/A	Stage A1: berzosertib + palliative RT Stage A2: berzosertib + cisplatin + capecitabine Stage B: berzosertib + radical chemo-RT	Recruitment ongoing	N/A	NCT03641547
	I	Advanced, unresectable solid tumors; N= 51	N/A	Irinotecan + berzosertib (D1, D15) q28 days for 12 cycles	Recruitment ongoing	N/A	NCT02595931
	I	Brain metastases from NSCLC or SCLC; N = 46	N/A	WBRT 15 fractions concurrent with berzosertib twice-weekly, 18–30 hours after first radiation therapy, for 3 weeks	Recruitment ongoing	N/A	NCT02589522
	I/II	SCLC and extrapulmonary small cell cancers; N = 70	N/A	Topotecan (D1–5) + berzosertib (D5 or D2 and D5 at escalating doses) q21 days	Recruitment ongoing	N/A	NCT02487095
	I/II	Platinum-sensitive ovarian cancer; N= 31	Platinum-sensitive disease	Gemcitabine 480–800 mg/m2 (D1, D8) + carboplatin AUC 4 (D1) + berzosertib 90–120mg/m2 (D2, D9) q21 days	Recruitment ongoing	N/A	NCT02627443
	II	Metastatic castration-resistant prostate cancer; N= 130	N/A	Docetaxel 60 mg/m2 (D1) + carboplatin AUC 4 (D1) q21 days versus Berzosertib 90mg/m2 (D2, D9) + carboplatin AUC 5 (D1) q21 days	Recruitment ongoing	N/A	NCT03517969
	II	Metastatic or unresectable gastric or gastroesophageal junction cancer; N=28	TP53 hot-spot mutation falling within exon 2 or exons 4–11	Irinotecan + berzosertib (D1 and D15) q28 days	Recruitment ongoing	N/A	NCT03641313
Ceralasertib (AZD6738)	I	Advanced solid tumors; N = 46	N/A	Escalating doses of ceralasertib followed by dose expansion	PR in 7%; SD in 48% of patients 21% of patients in dose-escalation and 25% in dose-expansion had SD ≥ 16 weeks	Grade ≥3 TRAEs occurred in 67% and 20% of patients receiving continuous and intermittent dosing (2 weeks on, 2 weeks off), respectively Frequent TRAEs were fatigue, anemia, nausea and thrombocytopenia	NCT02223923
	I	Advanced solid tumors; N=58	N/A	Paclitaxel (80mg/m2, D1, D8,D15) + ceralasertib at escalating doses 40mg OD – 240 mg BD q28 days	MTD: ceralasertib 240mg BD (D1–14) Of 51 evaluable patients, 1 CR (1.9%), 12 PR (23.5%), 18 SD (35.3%) were observed. ORR was 25.5%. Enrichment of NF1 somatic mutations and activating NRAS mutations were observed amongst melanoma patients enrolled Cyclical changes in IL-12 levels were observed in 3 patients who had clinical benefit	1 DLT of neutropenic fever occurred in each cohort of patients dosed at 160mg BD and 240 mg BD D1-D14.	NCT02630199
	I	Advanced solid tumors; N=45	N/A	Ceralasertib in escalating doses 60mg OD to 240 mg BD (D5–14) + Olaparib 100mg BD to 300mg BD (continuously)	RP2D: ceralasertib 160mg OD (D1–7) + olaparib 300mg BD (daily continuously) q28 days Of 39 evaluable patients, 1 CR (2.6%), 5 PR (12.8%) and 1 unconfirmed PR (2.6%) were observed in ovarian, breast, prostate, pancreas and ampullary cancer patients Responses were independent of status	Thrombocytopenia and neutropenia DLTs were noted Frequent TRAEs included thrombocytopenia, anemia, neutropenia, fatigue, decreased appetite, nausea, vomiting, constipation, diarrhea, cough	NCT02264678
	II	Advanced solid tumors; N = 24	Germline or somatic mutations in HRR genes or mutations in ATM, CHK2, MRN(MRE11/NBS1/RAD50, APOBEC	Ceralasertib 160mg OD (D1–7) + olaparib 300mg BD (D1–28) q28 days	Of 20 evaluable patients, 1/5 patients with ATM mutation achieved CR while 2/5 had durable clinical benefit >12 months PR was observed in another patient with BRCA1/2-mutated HGSOC who was PARPi-resistant	Myelosuppression was the most frequent toxicity, but no patients required discontinuation	NCT02576444
	I	Advanced cancer; N= 55	N/A	Gemcitabine at escalating doses (500mg/m2–1000mg/m2 D 3, 10, 17) + ceralasertib at escalating doses (40–120mg, D1–21), q28 days Gemcitabine at escalating doses (500mg/m2–1000mg/m2 D 3, 10, 17) + ceralasertib at escalating doses (40–120mg, for up to 12 days), q28 days	Recruitment ongoing	N/A	NCT03669601
	II	Metastatic triple negative breast cancer; N=268	Tumor mutations in HRR genes	Olaparib 300mg BD (D1–28) q28 days versus Ceralasertib 160mg OD + olaparib 300mg BD (D1–28) q28 days versus Adavosertib 150mg BD + olaparib 200mg BD (D1–21) q21 days	Active, not recruiting	N/A	NCT03330847
	II	Advanced pancreas, renal cell, urothelial cancer; N = 68	BAF250a positivity; ATM mutation	Ceralasertib BD (D1–14) q28 days Ceralasertib OD (D1–7) + olaparib BD (D1–28) q28 days	Recruitment ongoing	N/A	NCT03682289
	II	Advanced solid tumors; N=50	IDH1/2 mutation	Ceralasertib OD (D1–7) + olaparib BD (D1–28) q28 days	Recruitment ongoing	N/A	NCT03878095
	II	Metastatic castration-resistant prostate cancer; N=57	Mutations in DNA repair genes	Ceralasertib 160mg OD (D1–7) + olaparib 300mg BD (D1–28) q28 days	Recruitment ongoing	N/A	NCT03787680
	II	Platinum-sensitive or platinum-resistant high grade serous ovarian cancer; N = 86	Germline or somatic BRCA mutation or other HRD mutation or HRD-positive (Cohort C and D part II)	Cohorts A-C: ceralasertib 160mg OD (D1–7) + olaparib tablets 300mg BD (D1–28), q28days Cohort D: ceralasertib 160–320mg OD, (14 days) + olaparib 100–200mg OD, q28 days	Recruitment ongoing	N/A	NCT03462342
	II	Advanced or metastatic Her2-negative breast cancer; N=60	Germline BRCA mutation	Ceralasertib OD (D1–7) + olaparib BD (D1–28), q28 days	Recruitment ongoing	N/A	NCT04090567
	II	Recurrent gynecological cancers; N=40	ARID1A loss	Ceralasertib + olaparib	Recruitment ongoing	N/A	NCT04065269
	II	Advanced solid tumors; N=52	Target 60% of participants to have ATM expression (by IHC) ≤ 5%	Ceralasertib monotherapy	Recruitment ongoing	N/A	NCT04564027
	II	Recurrent osteosarcoma; N= 63	N/A	Ceralasertib OD (D1–7) + olaparib BD (D1–28), q28 days	Recruitment ongoing	N/A	NCT04417062
	II	Platinum-sensitive relapsed ovarian cancer who are PARPi-resistant; N= 192	Platinum-sensitive disease	Maintenance therapy after CR/PR/SD to platinum-based chemotherapy. Ceralasertib 160mg OD (D1–7) + olaparib 300mg BD (D1–28) versus Olaparib 300mg BD (D1–28) versus Placebo, q28 days	Recruitment ongoing	N/A	NCT04239014
M4344	I/II	PARPi-resistant recurrent ovarian cancer; N = 40	N/A	4-week lead in of niraparib monotherapy followed by combination of niraparib + escalating doses of M4344 100–200mg OD q28days	Not yet recruiting	N/A	NCT04149145
Elimusertib (BAY1895344)	I/Ib	Advanced solid tumors; N=241	ATM deleterious mutation or ATM loss for expansion cohorts	Escalating doses of elimusertib monotherapy	ORR: 30.7% in patients treated at ≥ 40 mg twice daily All responders had ATM loss of expression and/or ATM mutation Durable response in 1 patient who had BRCA1-mutated and olaparib- and chemotherapy-resistant HGSOC	Grade ≥4 TEAE: Neutropenia (33% in 80mg cohort, 17% in 60mg cohort), Thrombocytopenia (17% in 80mg cohort)	NCT03188965
	I	Advanced solid tumors and PARPi-resistant ovarian cancer; N=56	DDR deficiency	Elimusertib BD + niraparib OD q28 days	Recruitment ongoing	N/A	NCT04267939
	I	Advanced gastrointestinal cancer; N=90	N/A	Elimusertib BD (D1,2,15,16) + FOLFIRI (D1, 15) q28 days	Not yet recruiting	N/A	NCT04535401
	I	Advanced solid tumors, pancreas and ovarian cancer; N=54	N/A	Elimusertib BD (D1–3, 8–10) + gemcitabine (D1, 8) q21 days	Not yet recruiting	N/A	NCT04616534
	I	Advanced solid tumors, urothelial cancer; N = 68	N/A	Elimusertib BD (D2, 9) + cisplatin (D1) q21 days Elimusertib BD (D2,9) + gemcitabine (D1, 8) + cisplatin (D1) q21 days	Not yet recruiting	N/A	NCT04491942
	I	Advanced solid tumors, SCLC, neuroendocrine tumors; N = 87	N/A	Elimusertib BD (D1–2) + irinotecan (D1) q14 days Elimusertib BD (D2, 5) + topotecan (D1–5) q21 days	Not yet recruiting	N/A	NCT04514497
ART0380	I/II	Advanced solid tumors; N = 180	Target 40 participants to have loss of ATM expression	ART0380 intermittently OD (3 days on, 4 days off) or continuously +/− gemcitabine 1000mg/m2 (D1, D8) q21 days	Recruitment ongoing	N/A	NCT04657068
RP-3500	I/II	Advanced solid tumors; N = 239	N/A	RP3500 +/− talazoparib	Recruitment ongoing	N/A	NCT04497116
WEE1 inhibitors
Adavosertib (AZD1775; MK-1775)	I	Newly diagnosed GBM; N=51	N/A	Adavosertib D1–5 each week during concurrent RT/TMZ Adavosertib D1–5 combined with adjuvant TMZ D1–5 q28 days	MTD in combination with concurrent RT/TMZ: adavosertib 200mg MTD with adjuvant TMZ: adavosertib 425 mg	Concurrent phase: 275 mg dose level – 1 patient with G3 fatigue, 1 patient with G4 thrombocytopenia and neutropenia 200 mg dose level – 2/6 patients with DLTs Adjuvant phase: 500mg dose level – 2/3 patients with intolerable diarrhea	NCT01849146
	I	Locally advanced pancreatic cancer; N=34	N/A	Adavosertib in escalating doses OD (D1,2,8,9) + gemcitabine 1000 mg/m2 (D1,8) q21 days for 4 cycles Cycles 2 and 3 administered concurrently with radiation Optional continuation for cycle 5–8	RP2D of adavosertib: 150mg OD Median OS: 21.7 months (90% CI 16.7–24.8 months) Median PFS: 9.4 months (90% CI 8.0–9.9 months)	DLT occurred in 24% of patients, commonly anorexia, nausea, fatigue	NCT02037230
	II	Recurrent platinum-resistant high grade serous ovarian cancer; N = 124	N/A	Gemcitabine 1000mg/m2 (D1, D8, D15) +/− adavosertib 175mg OD (D1–2, D8–9, D15–16) q28 days.	PFS: 4.6 months versus 3.0 months (HR 0.56, 95% CI:0.35–0.90, p=0.015) OS: 11.5 months versus 7.2 months (HR 0.56, 95% CI: 0.34–0.92, p=0.022) PR rate: 21% versus 3% (p=0.02)	Grade ≥3 TEAE: Anemia (31% versus 18%), thrombocytopenia (31% versus 6%), neutropenia (62% versus 30%)	NCT02151292
	II	Recurrent platinum-resistant recurrent ovarian cancer; N = 94	N/A	Adavosertib 175–225 mg BD (various schedules) + gemcitabine 800mg/m2 (D1, D8 and D15), paclitaxel 80mg/m2 (D1, D8 and D15) or carboplatin AUC 5 (D1) or pegylated liposomal doxorubicin 40mg/m2 (D1)	ORR: 31.9% Highest ORR (66.7%) noted in cohort receiving adavosertib (225mg BD D1–3, 8–10 and 15–17) with carboplatin (AUC 5, D1), every 21 days PFS (median, overall): 5.5 months Possible positive relationship between CCNE1 amplification and response	Grade ≥3 TEAE: Neutropenia (22–78%), thrombocytopenia (11–75%), anemia (11–57%) TEAEs led to adavosertib dose interruptions (63%), reductions (30%) and discontinuations (13%)	NCT02272790
	II	Recurrent platinum-sensitive recurrent ovarian cancer; N = 121	TP53-mutant	Paclitaxel 175mg/m2 (D1) + carboplatin AUC 5 (D1) + adavosertib 225 mg BD (for 2.5 days)/matched placebo, q21 days	Median PFS: 7.9 versus 7.3 months (HR 0.63, 95% CI 0.38–1.06, p = 0.080) Clinical benefit was observed for patients with different TP53 mutation subtypes	Grade ≥3 TEAE: 78% versus 65% for patients receiving adavosertib Greatest increase in Grade ≥3 diarrhea (75% versus 37%), vomiting (63% vs 27%), anemia (53% vs 32%)	NCT01357161
	II	Recurrent or persistent serous uterine cancer; N=35	N/A	Adavosertib 300mg OD (D1–5, 8–12) q21 days	Of 34 evaluable patients, ORR was 29.4% Median PFS: 6.1 months Median duration of response: 9.0 months	Grade ≥3 TEAE seen in 61.8% of patients Common grade ≥3 TEAE: neutropenia (32.4%), anemia (23.5%), thrombocytopenia (17.6%), fatigue (23.5%)	NCT03668340
	I	Incurable esophageal or gastroesophageal cancer; N=33	N/A	External beam RT OD (5 days per week) + adavosertib OD (2–5 days) during week 1 and 3 of RT	Recruitment ongoing	N/A	NCT04460937
	I	Advanced PARPi-resistant solid tumors; N=54	Cohort A: germline or somatic BRCA1/2 Cohort B: germline or somatic mutations in DDR genes including BRCA1/2, BRIP, FANCA, PALB2, ATM or CCNE1 amplification	Olaparib BD (D1–5, 15–19) + adavosertib OD (D8–12, 22–26), q28 days	Recruitment ongoing	N/A	NCT04197713
	I	Locally advanced uterine, cervical or vaginal cancer; N=33	N/A	Adavosertib in combination with cisplatin and external beam RT	Active, not recruiting	N/A	NCT03345784
	II	Advanced refractory solid tumors; N =32	CCNE1 amplification, defined by CCNE1 amplification > 7, or found on approved NGS panels	Adavosertib OD (D1–5, 8–12), q21 days	Recruitment ongoing	N/A	NCT03253679
	II	Advanced solid tumors and kidney cancer with SETD2-deficiency; N=60	SETD2 loss on next-generation sequencing panel	Adavosertib OD (D1–5, 8–12), q21 days	Recruitment ongoing	N/A	NCT03284385
	II	Recurrent PARPi-resistant ovarian, primary peritoneal, or fallopian tube cancer; N=88	N/A	Adavosertib OD (D1–5, 8–12), q21 days Adavosertib OD (D1–3, 8–10) + olaparib BD (D1–21) q21 days	Recruitment ongoing	N/A	NCT03579316
	IIb	Recurrent uterine serous carcinoma; N=120	N/A	Adavosertib 300 mg OD (D1–5, D8–12) q21 days	Recruitment ongoing	N/A	NCT04590248
Debio 0123	I	Advanced solid tumors; N=24	N/A	Debio 0123 OM (Day −3 to D-1) for cycle 1 Debio 0123 + carboplatin from cycle 2 onwards	Recruitment ongoing	N/A	NCT03968653
ZN-c3	I	Platinum-resistant ovarian cancer; N = 100	N/A	ZN-c3 OD + carboplatin AUC 5 (D1) q21 days ZN-c3 OD + PLD 50mg/m2 (D1) q28 days	Recruitment ongoing	N/A	NCT04516447
CHK1/2 inhibitors
Prexasertib (LY2606368)	I	Advanced solid tumors, Japanese patients; N = 12	N/A	Prexasertib 80–105 mg/m2 (D1, D15) q28 days	SD: 66.7%	1 patient experienced DLTs (neutropenia) at each dose level G4 related TRAEs: neutropenia (50%), leukopenia (33.3%), anemia (8.3%), febrile neutropenia (8.3%), thrombocytopenia (8.3%)	NCT02514603
	I	Advanced SCC; N=101	N/A	Prexasertib 105 mg/m2 (D1, D15) q28 days	RP2D: 105mg/m2 (D1, D15) q28 days Median PFS: 2.8 months (90% CI 1.9–4.2) for anal SCC; 1.6 months (90% CI 1.4–2.8) for HNSCC; 3.0 months (90%CI 1.4– 3.9) for NSCLC ORR: 15% for anal SCC, ORR 5% for HNSCC	G4 neutropenia (71%), febrile neutropenia (12%)	NCT01115790
	II	Recurrent high grade serous or endometrioid ovarian cancer; N= 24	BRCA1/2-wildtype	Prexasertib 105 mg/m2 (D1, D15) q28 days	PR rate: 33% (8/24) 12/19(63%) women whose tumors harbored pre-therapy CCNE1 amplification or overexpression experienced durable PFS >6 months	Grade ≥3 TEAE: Neutropenia (93%), Leukopenia (82%), Thrombocytopenia (25%), Anemia (11%)	NCT02203513
	II	Recurrent BRCA1/2-mutant high grade serous ovarian cancer; N=22	Germline or somatic BRCA1/2 mutation	Prexasertib 105 mg/m2 (D1, D15) q28 days	ORR: 11% (1 CR, 1 PR) No response was seen in platinum-resistant patients with prior PARPi Median duration of treatment: 4 months (range 1–9 months)	Most common grade 3/4 TRAEs: neutropenia (82%), leukopenia (64%), thrombocytopenia (14%)	NCT02203513
	Ib	Locally advanced HNSCC; N=70	N/A	Prexasertib q14 days + cisplatin q7 days + IMRT Prexasertib q14 days + cetuximab q7 days + IMRT	Completed	N/A	NCT02555644
	I	Advanced solid tumors, including PARPi-resistant; N=29	BRCA1/2 mutation will be studied in dose expansion	Prexasertib (D1, D15) + olaparib intermittent schedule	Active, not recruiting	N/A	NCT03057145
	II	Advanced solid tumors; N=50	Mutation in HRR genes; MYC amplification; Rb loss; FBXW7 mutation; CCNE1 amplification	Prexasertib 105 mg/m2 (D1, D15) q28 days	Active, not recruiting	N/A	NCT02873975
	II	Platinum-resistant or refractory recurrent ovarian cancer; N= 169	BRCA1/2 mutation will be studied in one cohort	Prexasertib 105mg/m2 (D1, D15) q28days	Active, not recruiting	N/A	NCT03414047
	I	Refractory metastatic TNBC, N=10	N/A	LY3023414 200mg BD until PD followed by prexasertib 105 mg/m2 until PD	Recruitment ongoing	N/A	NCT04032080
SRA737	I/II	Advanced cancer; N=135	Presence of genomic alterations in TP53, RAD50, CCNE1, MYC, ATR, CHK2, BRCA1, FANCA	SRA737 40–600mg (D2) + gemcitabine 50–300 mg/m2 (D1, 8, 15) q28days	Overall ORR: 4% Fanconi anemia/ BRCA network mutations were associated with the most favorable outcomes ORR = 25%, DCR = 81%	Grade ≥3 TEAE: Neutropenia (63.4%), Anemia (5.8%), Thrombocytopenia (3.6%), ALT increase (5.8%), AST increase (5.0%)	NCT02797977
	I/II	Advanced solid tumors; N = 111	Presence of genomic alterations in TP53, RAD50, CCNE1, MYC, ATR, CHK2, BRCA1, FANCA	SRA737 in escalating doses 160–1300 mg OD (D1–28) q28 days	MTD: SRA737 1000mg OD or 500mg BD RP2D: 800mg OD	Grade ≥3 TEAE: 68.2% Most common TEAEs: diarrhea (70%), nausea (64%), vomiting (51%), fatigue (47%)	NCT02797964
DNA-PK inhibitors
AZD7648	I/II	Advanced solid tumors; N=230	N/A	AZD7648 monotherapy AZD7648 + olaparib 300mg BD (D1–28) AZD7648 + PLD 40mg/m2 (D1) q28days	Recruitment ongoing	N/A	NCT03907969
Peposertib (Nedisertib; M3814)	I	Adult patients with advanced solid tumors; N = 31	N/A	Peposertib 100–200mg OD or 150–400 mg BD, q21 days	MTD not reached RP2D: 400mg BD SD rate: 38.7%	Grade 3 AEs: maculopapular rash and nausea 55% of patients had a serious treatment-emergent AE, 13% of which were peposertib-related All peposertib-related serious treatment emergent AEs occurred at the highest dose level of 400mg BD	NCT02316197
	I/Ib	Recurrent ovarian cancer; N = 49	N/A	Peposertib OD + PLD (D1) q28days	Recruitment ongoing	N/A	NCT04092270
	I/II	Locally advanced pancreatic cancer; N = 90	N/A	Peposertib OD for 14 days + hypo fractionated RT for 5 fractions	Recruitment ongoing	N/A	NCT04172532
	I/II	Locally advanced rectal cancer; N= 160	N/A	Peposertib in escalating doses starting from 50 mg OD (D1-D5) q7 days + capecitabine 825mg/m2 BD (D1-D5) q7 days + 50.4Gy RT in 25–28 fractions	Recruitment ongoing	N/A	NCT03770689
CC-115	I	Advanced solid tumors; N=115	N/A	Escalating doses of CC-115 0.5–40mg OD, q28 days	RP2D: 10mg BD SD achieved in 53%, 22%, 21%, and 64% of patients with HNSCC, Ewing sarcoma, GBM, and castration-resistant prostate cancer, respectively In CLL/SLL, PR: 38% SD: 25% 1 patient with endometrial cancer remained in CR for 4 years	DLTs were thrombocytopenia, stomatitis, hyperglycemia, asthenia/fatigue, and transaminitis	NCT01353625
	I	Metastatic castration-resistant prostate cancer; N=16	N/A	Enzalutamide 160 mg OD + CC-115 in escalating doses 5–10mg BD	RP2D: 10mg PO BD All evaluable patients had a > 50% PSA response, 60% of patients achieved a ≥90% decline Median time on study for patients with PTEN deletions or DDR alterations was 25 weeks and 29 weeks, respectively	G3 rash (46%) at RP2D	NCT02833883
Samotolisib (LY3023414)	I	Advanced solid tumors in Japanese patients; N=12	N/A	Samotolisib 150mg–200mg BD in escalating doses	DCR: 55.6%	DLT reported at 200mg for 2 patients with G3 stomatitis Related AEs Grade ≥ 3 included anemia, stomatitis, hypophosphatemia, hyperglycemia	NCT02536586
	I	Advanced solid tumors; N=47	N/A	Samotolisib 20–450mg OD or 150–250mg BD in escalating doses	Durable PR in a patient with endometrial cancer harboring PIK3R1 and PTEN truncating mutations Overall PR rate: 28%	At 450mg OD, DLTs occurred 3/3 patients (thrombocytopenia, hypotension, hyperkalemia) At 250mg BD, DLTs occurred in ¾ patients (hypophosphatemia, fatigue, mucositis) At 200mg BD, DLTs occurred in 1/15 patients (nausea)	NCT01655225
	Ib/II	Metastatic castration resistant prostate cancer; N =142	N/A	Samotolisib 200mg BD run-in 7 days then enzalutamide 160mg OD + samotolisib 200mg BD or placebo q28days	Median PFS: 7.5 months (samotolisib + enzalutamide) versus 5.3 months (placebo + enzalutamide) (HR 0.68, 95% CI 0.41–1.14, p =0.069) Increased PFS benefit observed for patients who were AR-V7 negative, median PFS: 13.2 versus 5.3 (HR 0.52, 95% Ci 0.28–0.95, p-0.028)	Grade ≥3 TRAE: fatigue (11%), nausea (8%), diarrhea (6%)	NCT02407054
	II	Recurrent endometrial cancer; N=31	N/A	Samotolisib 200mg BD q28 days	Active, not recruiting	N/A	NCT02549989
ATM inhibitors
AZD1390	I	GBM or CNS metastases; N = 132	N/A	AZD1390 intermittent or continuously concurrent with the following RT schedules: IMRT 3.5Gy over 10 fractions WBRT 30Gy or PBRT 3 Gy over 10 fractions IMRT 2Gy over 30 fractions	Recruitment ongoing	N/A	NCT03423628
AZD0156	I	Advanced cancers; N = 84	N/A	AZD0156 + olaparib AZD0156 + FOLFIRI or irinotecan	Active, not recruiting	Minor nausea, vomiting and anemia occurred in around 40% of all cohorts of AZD0156 + olaparib Grade 3 and 4 hematologic toxicities were observed when AZD0156 120 mg BD was combined with olaparib 200mg BD, this was considered intolerable	NCT02588105

^aAbbreviations: AE, adverse event; AUC, area under curve; CI, confidence interval; CLL, chronic lymphocytic leukemia; CR, complete response; DCR, disease control rate; DDR, DNA damage response; DLT, dose-limiting toxicity; GBM, glioblastoma multiforme; GCIG, gynecologic intergroup; HGSOC, high-grade serous ovarian cancer; HNSCC, head and neck squamous cell carcinoma; HR, hazard ratio; HRR, homologous recombination repair; IMRT, intensity-modulated radiotherapy; MTD, maximum tolerated dose; NSCLC, non-small cell lung cancer; ORR, objective response rate; OS, overall survival; PD, progression of disease; PFS, progression-free survival; PLD, pegylated liposomal doxorubicin; PR, partial response; RP2D, recommended phase II dose; RT, radiotherapy; SAE, serious adverse events; SCLC, small cell lung cancer; SD, stable disease; SLL, small lymphocytic leukemia; TEAE, treatment-emergent adverse events; TNBC, triple-negative breast cancer; TRAE, treatment-related adverse events; WBRT, whole brain radiotherapy.

Combinatorial strategies of ATRi with cytotoxic chemotherapies, such as the nucleoside analog gemcitabine, known to increase RS, are actively being investigated. Interestingly, a randomized Phase II study of gemcitabine plus berzosertib, or gemcitabine alone, in platinum-resistant ovarian cancer met its primary endpoint of improved progression-free survival (PFS) for the combination compared with chemotherapy alone (median PFS 22.9 weeks versus 14.7 weeks, hazard ratio 0.57, 90% confidence interval 0.33–0.98, P = 0.04). Serious adverse events (SAEs) were observed in 28% patients randomized to gemcitabine alone, compared to 26% of patients in the gemcitabine plus berzosertib arm[30]. Hence, the results from the phase 2 trial show that the gemcitabine and berzosertib combination seems to have an acceptable safety profile and provides evidence of feasibility.

Additionally, a recent Phase I trial studying the safety profile of ceralasertib plus paclitaxel, in a rolling six design, used a fixed dose paclitaxel 80mg/m² (days (D) 1, 8, 15 on a 28-day cycle) and escalating doses of ceralasertib until the maximum tolerated dose (MTD) of 240 mg ceralasertib twice-daily (BD) (D1–14 of a 28-day cycle) was reached. An objective response rate (ORR) of 22.6% amongst 57 patients with advanced solid tumors was reported from this trial[31]. One melanoma patient achieved complete response (CR), while partial responses (PR) were observed amongst ten melanoma patients, all of whom had progressed on prior immune check-point blockade (ICB), and durable responses were noted[32](Table 1).

WEE1 inhibition leads to increased CDK1/2 activity, and in turn, bypass of the G2/M checkpoint and unscheduled mitotic entry (Figure 1). Uncontrolled surges in replication initiation and continued cell cycle progression exhaust nucleotide supplies, deplete rate-limiting replication factors, and further increases RS[20]. These observations have supported the rationale for WEE1 targeting in p53-deficient cells, which are highly dependent on the G2/M checkpoint. Currently, three WEE1 inhibitors (WEE1i) are in clinical development [adavosertib (AZD1775, MK1775; AstraZeneca); ZN-c3 (Zentalis) and debio0123 (Debiopharm)]. Preclinically, adavosertib has been shown to sensitize TP53-mutant cells to DNA damaging chemotherapy through premature mitotic-entry[33]. Recent data also suggests that WEE1i have activity in TP53-wildtype cells harboring elevated RS, such as through co-deficiencies in DDR pathways, or in the setting of nucleotide starvation and other genotoxic stressors [34]. Uterine serous carcinoma (USC), a subtype of uterine cancer associated with a poor prognosis, is characterized by the nearly ubiquitous presence of a TP53 mutation that frequently occurs concomitantly with other oncogenic driver mutations or amplifications, such as CCNE1 or HER2 over-expression. On the single arm two-stage Phase II trial of adavosertib monotherapy in recurrent USC, patients who had progressed on prior platinum-based chemotherapy were treated with adavosertib 300mg daily (D1–5, 8–12, on a 21-day cycle). The trial co-primary endpoints were ORR and PFS at 6 months (PFS6). Amongst 35 patients enrolled, the reported ORR was 29.4%, PFS6 was 47.1% and median duration of response was 9.0 months, which compares favorably with historical controls. Grade 3 or higher toxicities reported were fatigue (23.5%) and myelosuppression characterized by neutropenia (32.4%), anemia (23.5%) and thrombocytopenia (17.6%). Discontinuation of treatment due to adverse events was required in 5.9% of patients, and dose reductions occurred in 76.5% of patients[35] (Table 1). In preclinical models, adavosertib and debio0123 in combination with RS-inducing chemotherapies, such as carboplatin, paclitaxel, gemcitabine and pegylated liposomal doxorubicin (PLD), have shown encouraging efficacy[36–39]. Two separate randomized Phase II clinical trials studying the efficacy of adavosertib and chemotherapy in patients with platinum-sensitive[40] and platinum-resistant[41] advanced ovarian cancer met their primary endpoint of improved PFS compared to chemotherapy alone. However, frequent gastrointes tinal toxicity and myelosuppression was observed: in the adavosertib plus carboplatin/paclitaxel combination, grade ≥3 adverse events were observed in 78% of patients, compared with 65% in patients receiving chemotherapy alone. Diarrhea (75%), vomiting (63%), and anemia (53%) were more frequent in patients randomized to the adavosertib-containing arm [40], while in the adavosertib plus gemcitabine combination, myelosuppression was observed at higher rate compared with patients receiving gemcitabine alone (neutropenia 62% vs. 30%, thrombocytopenia 31% vs. 6%) [41] (Table 1).

DNA-PK is a member of the phosphatidylinositol 3-kinases-related kinases (PIKK) family with pleiotropic roles in multiple cellular pathways[42]. Aside from the described roles in transcription regulation, innate immunity, and telomere capping, the most widely recognized role of DNA-PK lies in DSB repair within the DDR cascade, playing a key role in NHEJ DNA repair. Here, it serves as the protein scaffold that brings together repair proteins at sites of DNA damage and phosphorylates downstream X-ray repair cross-complementing protein 4 (XRCC4) and γH2AX [43]. Additionally, DNA-PK serves to stabilize p53 and RPA, allowing effective cell cycle arrest. Several DNA-PK inhibitors are in clinical development (Table 1). Preclinical studies of the highly specific and potent DNA-PK inhibitor, AZD7648 (AstraZeneca), showed efficacy when combined with olaparib in ATM-deficient cells, xenografts, and when combined with radiotherapy or doxorubicin [44]. AZD7648 is now in Phase I clinical trial testing as monotherapy and in different rational combinations (Table 1). Peposertib (M3814; Merck), another orally bioavailable DNA-PK inhibitor, has been shown to both sensitize tumors to radiotherapy in TP53-mutated cancer cell lines compared to TP53-wild type cells[45], and to enhance doxorubicin activity in cancer xenografts[46]. A first-in-human Phase I study has reported modest activity of peposertib monotherapy, with 22.6% (7/31) of patients achieving stable disease (SD) lasting ≥12 weeks[47]. While common grade ≥3 toxicities included rash and nausea, 55% of patients had a serious treatment-emergent adverse event, of which, 13% were deemed to be related to peposertib (maculopapular rash, pyrexia and nausea)[47](Table 1). Early phase clinical trials investigating DNA-PK inhibitors as monotherapy and in combination with chemo-radiotherapies are ongoing (Table 1). Finally, CC-115 (Celgene) and samotolisib (LY3023414, Eli Lilly), small molecule inhibitors with dual activity against DNA-PK and mechanistic target of rapamycin (mTOR) pathways, have been investigated in Phase I trials showing preliminary anti-tumor activity[48, 49](Table 1). Common adverse events related to LY3023414 and CC-115 were nausea, vomiting, fatigue and poor appetite. Toxicities consistent with those expected from mTOR inhibitors were described, including stomatitis, hyperglycemia, transaminitis and thrombocytopenia[48, 49](Table 1).

In response to DSBs, ATM plays a crucial role in activating key cell cycle checkpoints, namely the Gap 1 (G1)/S phase checkpoint and intra-S-phase checkpoint, preventing cells with damaged DNA from proceeding or completing S-phase, respectively[50] (Figure 1). Current ATM inhibitors (ATMi) in development include AZD0156 (AstraZeneca) and AZD1390 (AstraZeneca). Both ATMi have been shown to be potent radio-sensitizers in pre-clinical studies[51]. AZD1390 has been optimized to cross the blood-brain barrier, and it is being evaluated as a radiosensitizer for the treatment of radio-resistant glioblastoma[52] (Table 1). When combined with olaparib, AZD0156 was also able to potentiate the effects of PARP inhibitors (PARPi) across lung, gastric and breast cancer cell lines[51]. Mechanistically, the combination of AZD0156 with PARPi was shown to lead to transient G2 cell cycle-arrest, but eventual abrogation of the G2 DNA damage checkpoint, inappropriate mitotic entry and mitotic/post-mitotic cell death was also demostrated[53]. In colorectal cancer models, AZD0156 displayed in-vitro and in-vivo synergy with SN-38, the active metabolite of irinotecan[54]. Taken together, such preclinical data become the rationale for a Phase I trial of AZD0156 combined with olaparib or irinotecan-based chemotherapy in patients with advanced solid tumors. In a preliminary report of the AZD0156 plus olaparib arm of this trial, 46 patients were recruited across 8 cohorts. Around 40% of all patients experienced minor toxicities such as nausea, vomiting and anemia, while grade ≥3 hematological toxicities were only observed in the cohort that received AZD0156 120mg BD plus olaparib 200mg BD, which was considered intolerable[55] (Table 1).

CHK1/2 are serine/threonine kinases that lie downstream to ATR and ATM in the DDR cascade and serve as critical controllers of cell cycle checkpoints, in part by stabilizing replication forks to facilitate DNA repair[9] (Figure 1). Inhibition of CHK1/2 leads to increase RS, while allowing for the G2/M transition. Currently, two CHK1 inhibitors (CHK1i) are in early phase clinical trials [prexasertib (LY2606368; Eli Lilly), SRA737 (Sierra Oncology)]. A small Phase II trial of prexasertib in high-grade serous ovarian cancer has reported preliminary results in BRCA1/2-wildtype[56] and mutant populations[57] with encouraging efficacy observed (PR rate 33%), even in BRCA1/2-wildtype patients[56]. Neutropenia-predominant myelosuppression was observed in both cohorts[56, 57]. Analysis of BRCA1/2-wildtype patients led to the observation that 63% of patients with tumors harboring CCNE1 amplification or overexpression derived durable PFS benefit from prexasertib, exceeding 6 months[56](Table 1). Conversely, on a Phase I trial of SRA737 in advanced solid tumors, high-grade serous ovarian cancer patients had a high disease control rate of 54%, while patients with mutations in BRCA/Fanconi anemia genes appeared to benefit the most, and no association to clinical benefit was made in the group of patients with CCNE1 amplification[58, 59] (Table 1).

Despite the encouraging early efficacy signals, challenges remain in determining the optimal dose schedules for each of these agents to balance dose-dependent myelosuppression with antitumor activity, both as monotherapy and in combination with other agents. A promising combinatorial partner to CHK1i are immunotherapeutic agents, which do not share overlapping toxicity profiles with RSR-targeting drugs.

RS and the immune response

Immunity and DNA repair pathways, including the pathways involved in replicative stress, have evolved in parallel leading to inextricable links between both networks[60]. For instance, recent studies have described how key RSR members such as DNA-PK, meiotic recombination 11 (MRE11) and RAD50 double strand break repair protein (RAD50) are not only integral to replicative stress, but also serve as pattern recognition receptors (PRR) and sensors of cytoplasmic double-stranded DNA (dsDNA) that initiate stimulator of interferon genes (STING)-dependent signalling[61–63] and T cell activation. DSBs also lead to upregulation of PD-L1 on cancer cells through the signal transducer and activator of transcription 1 (STAT1) and STAT3 mediated pathways, which require ATR/ATM/CHK1 and are further upregulated in the context of BRCA2 or XRCC5/6 depletion[64]. Similarly, it has also been demonstrated that ATR/ATM/CHK1 is required for upregulation of natural killer group 2D (NKG2D) cell surface ligands on natural killer (NK) cells[65]. The dawn of cancer immuno-oncology has given rise to a growing interest in cellular links that may lead to the development of synergistic combinations of ICB with RSR-targeting drugs. Several mechanisms surrounding the immune-modulatory effects of RSR-targeting have been previously described.

Aberrations in the RSR pathway have been shown to increase cytosolic DNA fragment accumulation, which in turn, activates the cyclic-GMP-AMP-synthase (cGAS)-STING-tank binding kinase 1 (TBK1)- interferon regulatory factor 3 (IRF3) pathway, and upregulates the transcription of type I interferon (IFN) genes. This effectively induces an innate immune response, downstream T cell activation and increased PD-L1 expression[66]. In breast cancer, a DDR-deficient molecular signature was associated with increased T cell infiltration: cGAS-STING activation and PD-L1 expression[67]. It has thus been hypothesized that targeting the RSR using specific inhibitors would amplify this response and sensitize tumors to ICB (Figure 2). In prostate cancer models, ATR inhibition led to increased S-phase DNA damage and upregulation of the chemokines C-X-C motif chemokine ligand 10 (CXCL10) and chemokine ligand 5 (CCL5), which are key in CD4+ and CD8+ T-cell chemotaxis[68]. CXCL10 and CCL5 are also known transcriptional targets of IRF3, providing evidence of an active type I IFN response. Similarly, in preclinical models of refractory small cell lung cancer, CHK1 inhibition with SRA737 induced increased expression of PD-L1, and evidence of type I IFN signaling included increased IFNβ, CXCL10 and CCL5 mRNA expression in-vitro. These immunomodulatory effects correlated with antitumoral immune responses observed when SRA737 was combined with an anti-PDL1 antibody in refractory small cell lung cancer (SCLC) murine models[69]. On a recently reported Phase I trial of ceralasertib with paclitaxel, durable responses in anti-PD1-resistant melanoma patients were described. Interestingly, translational analyses detected cyclical fluctuations in serum IL-12 levels of several clinically benefitting patients, which may point toward underlying immunological mechanisms of action of this combination, although this is still not completely understood[32].

Figure 2: RSR blockade stimulates innate anti-tumor immunity through the cGAS-STING pathway.

Inhibition of components of the RSR ultimately leads to cell cycle checkpoint abrogation and inappropriate mitotic entry of cells with unresolved DNA damage. This culminates in mitotic catastrophe and resultant micronuclei formation. Cytosolic DNA fragments released bind to and activate the cytosolic DNA sensor cGAS, which binds to the adaptor STING. When activated, STING translocates into Golgi body compartments and activates TBK1, which phosphorylates IRF3. IRF3 enters the nucleus to facilitate transcription of Type I interferon, TNFα, and interleukins. Type I interferon upregulates chemokines CCL5 and CXCL10, which encourage effector T cell infiltration into the tumor. Created with Biorender.com. Abbreviations: ATM, ataxia telangiectasia-mutated; ATR, ATM and rad3-related; CCL5, chemokine ligand 5; CDC25, cell division cycle 25; CHK, checkpoint kinase; CXCL10, C-X-C motif chemokine ligand 10; DSB, double strand break; cGAS, cyclic GMP-AMP synthase; G1, gap 1; G2, gap 2; IRF3, interferon regulatory factor 3; M, mitosis; RSR, replication stress response; S, synthesis; SSB, single strand break; STING, stimulator of interferon genes; TBK1, tank binding kinase 1; WEE1, WEE1-like protein kinase.

Preclinical studies have also shown synergistic immune effects of combining ATR, DNA-PK, CHK1, ATM or WEE1 inhibitors with radiotherapy or radio-immunotherapy[70–72]. In KRAS-mutated cancer murine models, the combination of the ATRi ceralasertib and radiotherapy attenuated radiation-induced CD8+ T cell exhaustion, reduced tumor-infiltrating regulatory T cells, and potentiated CD8+ T cell activity[70]. In a similar manner, evidence of immunologic memory was observed in complete responder mice when ceralasertib was combined with radiation therapy[70]. Similarly, in immuno-competent mouse xenografts of hepatocellular carcinoma, ceralasertib synergized with radiation and PD-L1 inhibition to increase effector CD8⁺ T cell infiltration, reduce T cell exhaustion and suppress infiltrating T regulatory cells through the activation of the cGAS/STING pathway. These data demonstrate that the combination of ceralasertib with radio-immunotherapy creates a more favorable tumor immune microenvironment and lead to improved survival compared to radio-immunotherapy alone[71]. Several ongoing trials are assessing combinations of ATRi, DNA-PKi, CHK1i, ATMi or WEE1i in combination with varying schedules of radiotherapy (Table 1), while two Phase I trials are evaluating the combinatorial synergy of the DNA-PKi peposertib, avelumab and palliative radiotherapy combination in advanced solid tumors (Table 2).

Table 2.

Combination strategies of immune checkpoint blockade with RSR-targeting drugs.^a

Drug	Phase	Study population	Biomarker	Treatment	Efficacy	Safety	Reference
Berzosertib (M6620; VX-970)	I	Advanced solid tumors that are DDR deficient; N=36	Mutations suggestive of DDR deficiency including ARID1A, ATM, ATR, ATRX, BAP1, BARD1, BRCA1/2, BRIP1, CDK12, CHEK2, FANCA, FANCC, FANCD2, FANCE, FANCF, FANCM, MRE11A, MSH2, NBN (NBS1), PALB2, RAD51, RAD51C, RAD51D, SMARCB1, VHL	Avelumab (D1, D15) + berzosertib (D1, D,8, D15, D22) q28 days	Recruitment ongoing	N/A	NCT04266912
	II	Recurrent platinum-sensitive ovarian cancer, that is resistant to PARPi; N= 90	N/A	Part A: Safety run-in of carboplatin AUC 5 (D1) + avelumab 1600mg (D1) + berzosertib 90mg/m2 (D2) q21 days Part B: chemotherapy (carboplatin + paclitaxel/gemcitabine/PLD) +/− bevacizumab → bevacizumab maintenance; versus carboplatin + avelumab + berzosertib for 6 cycles → avelumab maintenance	Recruitment ongoing	N/A	NCT03704467
Ceralasertib (AZD6738)	II	Advanced NSCLC who have progressed on immunotherapy; N=410	Subjects with ATM loss were assigned to ceralasertib + durvalumab	Novel durvalumab combinations including ceralasertib 240 mg BD (D1–14) + durvalumab 1500mg (D1), q28 days	ORR: 8.3% Median PFS: 7.43 months (95% CI, 3.45–9.46) Median OS:15.80 months (95% CI, 11.01-NR)	N/A	NCT03334617
	II	Advanced biliary tract cancers who have previously received immunotherapy; N=26	N/A	Ceralasertib 240 mg BD (D15–28) + durvalumab 1500mg (D1), q28days	Recruitment ongoing	N/A	NCT04298008
	II	Advanced NSCLC who have progressed on immunotherapy; N=120	N/A	Novel durvalumab combinations compared against docetaxel chemotherapy. Including ceralasertib 240 mg BD (D1–7 for cycle 1, D22–28 for subsequent cycles) + durvalumab 1500mg (D1), q28 days	Recruitment ongoing	N/A	NCT03833440
Elimusertib (BAY1895344)	Ib/II	Advanced solid tumors; N = 110	DDR deficiency biomarker positive, including ATM deleterious mutation	Elimusertib + pembrolizumab	Recruitment ongoing	N/A	NCT04095273
	I	Recurrent HNSCC; N= 38	N/A	Elimusertib + pembrolizumab + SBRT	Not yet recruiting	N/A	NCT04576091
Prexasertib	I	Advanced solid tumors; N=17	N/A	Prexasertib + LY3300054 (novel PD-L1 inhibitor).	Active, not recruiting	N/A	NCT03495323
Peposertib (Nedisertib; M3814)	I	Advanced solid tumors; N=47	N/A	Peposertib + avelumab +/− palliative RT	Recruitment ongoing	N/A	NCT03724890
	I/II	Metastatic castration resistant prostate cancer; N =24	N/A	Radium-223 (D1) Radium-223 (D1) + peposertib OD/BD (D3–26) q28 days × 6 cycles Radium-223 (D1) + peposertib OD/BD (D3–26) + avelumab (D1, D15) q28 days × 6 cycles	Recruitment ongoing	N/A	NCT04071236
	I/II	Advanced solid tumors and hepatobiliary cancers; N=92	N/A	Hypo fractionated RT (D-17 to D-7) + avelumab (D1,D15) q28 days Hypo fractionated RT (D-17 to D-7) + peposertib OD (D1–28) + avelumab (D1,D15) q28 days	Recruitment ongoing	N/A	NCT04068194
ZN-c3	I/II	Advanced solid tumors; N=360	N/A	ZN-c3 monotherapy in escalating doses followed by dose-expansion ZN-c3 + talazoparib ZN-c3 + pembrolizumab	Recruitment ongoing	N/A	NCT04158336
Adavosertib	I	Advanced solid tumors; N=54	N/A	Cohort A: adavosertib 125 mg BD (D1–5, 15–19) + durvalumab 1500mg (D1) q28days Cohort B: adavosertib 125–175 mg BD (D15–17, 22–24) + durvalumab 1500mg (D1) q28days Cohort C: adavosertib 125 mg BD (D8–10, 15–17, 22–24) + durvalumab 1500mg (D1) q28days Cohort D: adavosertib 200–300 mg (D15–19, D22–26) ) + durvalumab 1500mg (D1) q28days	MTD/RP2D: adavosertib 150mg BD (D15–17, 22–24) + durvalumab 1500mg (D1) q28 days Overall DCR: 36%	Grade≥3 AEs 43–86%, most commonly fatigue (15%), diarrhea (11%), nausea (9%) 13% of patients had adavosertib-related SAEs including drug-induced liver injury	NCT02617277

^aAbbreviations: AE, adverse events; AUC, area under curve; CI, confidence interval; DCR, disease control rate; DDR, DNA damage response; HNSCC, head and neck squamous cell carcinoma; MTD, maximum tolerated dose; NR, not reached; NSCLC, non-small cell lung cancer; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; PLD, pegylated liposomal doxorubicin; RP2D, recommended phase II dose; RT, radiotherapy; SAE, serious adverse events.

Despite these associations, a conflicting study by Sato and colleagues demonstrated that although PD-L1 upregulation in DNA repair-proficient tumor cells was dependent on the ATM/ATR/CHK1 pathway after treatment with ionizing radiation or DNA damaging agents, such as etoposide or olaparib, the introduction of specific inhibitors of ATM, ATR or CHK1 in this context suppressed PD-L1 upregulation[64]. Other recent data have shed light on the parallel STING-independent DNA sensing pathway (SIDSP), which relies on DNA-PK as its primary DNA sensor[73]. Unlike the cGAS-STING pathway, SIDSP appears to be activated by DNA ends, compared with cGAS-STING which detects the sugar phosphate DNA backbone. DNA transfection of STING-knockout U937 lymphoma cells led to a broad and potent type I IFN response, with upregulation of over a thousand genes, mediated by the SIDSP[73]. When a DNA-PK inhibitor was introduced, the SIDSP was completely inhibited. This raises additional considerations to the design of trials involving DNA-PK inhibitors, whereby blunting of the SIDSP by DNA-PK inhibition may be counterproductive to the triggering of innate immunity in the tumor microenvironment. Taken together, these findings reflect that the mechanisms underlying the crosslinks between DNA damage and tumor immunogenicity have not yet been completely elucidated, and raise an urgent need for more studies to clearly delineate contexts in which combining RSR inhibitors with immune checkpoint blockade (ICB) are likely to be synergistic rather than antagonistic[74].

At present, ongoing clinical trials are actively investigating varying combinations of ICB with RSR-targeting drugs (Table 2). In colorectal cancer murine models, the addition of avelumab to the ATRi berzosertib and cisplatin or carboplatin not only led to improved tumor reduction and overall survival compared with berzosertib/platinum chemotherapy alone, but it was also well tolerated.[75]. Interestingly, re-inoculation of complete responder mice with further doses of MC38 colorectal cancer cells failed to re-establish tumor growth, suggesting durable immunogenic memory[75]. Results are awaited from another Phase Ib study of the ATRi berzosertib in combination with carboplatin and avelumab in patients with PARPi-resistant ovarian cancer, as well as berzosertib plus avelumab in DDR-deficient advanced solid tumors (Table 2). The ATRi ceralasertib was investigated in combination with durvalumab in a Phase Ib trial of advanced non-small cell lung cancer (NSCLC) or head and neck squamous cell carcinoma (HNSCC), with an initial report of the first 25 patients treated describing two cases of dose-limiting toxicity with an otherwise tolerable side effect profile and no obvious overlapping toxicities [76]. Further trials of durvalumab plus ceralasertib in immunotherapy-refractory biliary tract cancers and NSCLC are ongoing (Table 2). A third ATRi, elimusertib, is being studied in combination with pembrolizumab in a Phase I trial of advanced solid tumors, and a dose-expansion cohort in patients with biomarkers of DDR-deficiency is planned (Table 2). Several trials are testing the selective WEE1i adavosertib (AZD1775; AstraZeneca), in combination with ICB. For example, a Phase I trial investigated varying doses and schedules of adavosertib with durvalumab[77], with common grade ≥3 toxicities including fatigue (15%), diarrhea (11%) and nausea (9%). Also, 13% of patients developed reversible drug-induced liver injury related to adavosertib as a SAE. The recommended Phase II dose of adavosertib was determined to be 150 mg BD (D15–17, D22–24, 3 days on, 4 days off), when combined with durvalumab 1500 mg every 28 days. Encouraging preliminary efficacy was reported, with a disease control rate of 36% [77] (Table 2).

Predictive genomic biomarkers through synthetic lethality

The recognition of RSR-targeting as a novel strategy for therapeutic exploitation in cancer has allowed significant strides in unraveling the biological significance of key RSR players and has launched promising drug candidates into the developmental pipelines. Despite these important advances, significant gaps remain in discerning the optimal therapeutic context for RSR-targeting. Priority must be given to determining specific patient populations which are susceptible to these different novel agents by developing relevant validated predictive biomarkers of response.

Loss of ATM function is thought to increase cellular dependency on ATR pathways, and it may induce synthetic lethality in the presence of ATRi[50]. Preclinically, siRNA screens have showed increased sensitivity to ATRi in mantle cell lymphoma cells with induced loss of ATM function[78]. The ATRi ceralasertib was noted to be more active in-vitro and in-vivo, in ATM-deficient cell lines and xenografts, compared with ATM-wildtype[79, 80]. ATM loss induced via clustered regularly interspaced short palindromic repeat (CRISPR) knockout or drug inhibition sensitized pancreatic cancer cell lines to the combination of ceralasertib and gemcitabine by augmenting replication catastrophe-mediated cell death [81]. Furthermore, early phase clinical trials investigating ATRi have reported anecdotal responses in patients with ATM loss and/or ATM aberrations[26, 29]. Such responses are inducing the use of ATM loss as a selective biomarker for expansion cohorts of ongoing trials (Table 1).

Other genomic aberrations that have been identified as potential synthetic lethal partners for ATRi targeting include ARID1A mutations [23] and deficiencies in HRR [82]. Clinical trials utilizing these as criteria for patient selection are in progress (Table 1). In preclinical studies, inhibition of RAD51 has been shown to increase dependency on ATR-CHK1 RSR signaling and the subsequent inhibition of ATR or CHK1 led to preferential killing of homologous recombination deficient (HRD) cells[83]. Silencing of FA/HRR genes has also been demonstrated to result in increased RS and exhaustion of nucleotide pools after WEE1i introduction, culminating in inappropriate mitotic entry [84] and suggesting the relevance of synthetic lethality with WEE1i in this genomic context. Several ongoing trials are using pathogenic mutations in FA and HRR genes as a selective biomarker for patient inclusion (Table 1). Broader genomic mutational signatures of HRD, such as signature-3[85] have been shown to correlate with PARPi response as well as response to the niraparib plus pembrolizumab combination,[86] but no data regarding their utility as predictors of response to inhibitors of the ATR-CHK1-WEE1 axis is available. Recently, Dreyer and colleagues reported on a novel signature of replication stress in pancreatic cancer patient-derived cell lines (PDCL) and organoids[87]. Increased sensitivity to ATRi and WEEi was observed in PDCL and organoids within the top quintile of the replication stress signature, and in-vitro responses were observed independent of known DDR deficiency status or tumor molecular subtype[87]. Such replication stress signatures may provide a broader means of patient selection and could become relevant as a potential biomarker if validated in the clinic[87].

Increased cyclin E-CDK2 complexes via CCNE1 amplification may promote transition from G1 to S phase, thus increasing RS. However, TP53 mutations lead to a defective G1/S checkpoint and increased reliance on the G2/M checkpoint, offering a rationale for WEE1 targeting. To demonstrate this, Kok and colleagues showed that induced cyclin E1 or CDC25A expression led to slowing of the replication fork and aberrant mitotic progression[88]. The genomic instability observed was further exploited by concurrent ATR or WEE1 inhibition, triggering mitotic aberrancy and leading to cell death[88]. Interestingly, p53 inactivation further exacerbated these findings[88]. Another potential biomarker for WEE1 inhibition is the SET Domain Containing 2 (SETD2) loss of function that causes histone H3 trimethylated at lysine 36 (H3K36me3) deficiency[89]. SETD2 is the only H3K36me3 methyltransferase, and it functions as a tumor suppressor. Inactivating SETD2 mutations were synthetically lethal with WEE1i in-vitro and in-vivo[89]. A Phase II trial is currently investigating adavosertib monotherapy in advanced solid tumors and renal cell cancer with SETD2 deficiency (Table 1).

Combination strategies beyond immuno-oncology

Increasing data now points to the potential for additional synergy by targeting more than one RSR protein. PARPi trap PARP1 at the sites of DNA damage with varying potency, blocking downstream DNA replication and generating increased RS. Reported synergy between PARPi and RSR-targeting drugs such as ATRi[90, 91], WEE1i[92–94], ATMi and CHK1i[95] has been described, albeit with varying purported mechanisms of action. PARPi-resistant BRCA1-deficient cells have been shown to be highly dependent on ATR for survival[96], and the addition of ATRi or CHK1i to PARPi led to increased G2 transition, genomic instability and apoptosis[91]. The synergy between ATRi or CHKi and PARPi is described to have several possible mechanistic underpinnings. Increased replication fork collapse may occur due to loss of independent fork stabilization mechanisms controlled by ATR-CHK1 and PARP, respectively[91]. Furthermore, cells harboring increased DNA damage after PARPi treatment are allowed premature and inappropriate mitotic entry after ATRi/CHKi exposure through abrogation of the G2/M phase checkpoint, represented phenotypically by an increase in M-phase chromatid breaks[91, 97].

ATRi treatment has also been described as a potential means to overcome resistance to PARPi caused by rewiring of HRR and increased fork protection[96, 98, 99]. For example, ATRi have been shown to disrupt RPA-mediated localization of partner and localizer of BRCA2 (PALB2)-BRCA2 to DSBs in PARPi resistant BRCA1-deficient cells[96]. In the setting of restored fork protection, ATRi suppress RAD51 loading at stalled forks and enhance fork degradation by nucleases, thus helping to reverse PARPi resistance [96]. Finally, ATRi has also been described to overcome PARPi-resistance mediated by Schlafen 11 (SLFN11) inactivation[100]. When compared to PARPi monotherapy, synergy related to increased mitotic catastrophe in cells harboring a large amount of DSBs was observed with the combination of olaparib with ceralasertib in BRCA1-mutant and BRCA2-reversion high-grade serous ovarian cancer cell lines[99]. While one clinical study investigating ceralasertib plus olaparib has reported a PR in one BRCA wild type ovarian cancer patient, other Phase II clinical trials using the same combination are recruiting patients with advanced solid tumors, advanced breast cancer, castration-resistant prostate cancer, high-grade serous ovarian cancer, ARID1A-deficient gynecological cancers, and osteosarcoma, with cohorts focusing on biomarker-selected patients such as those with germline or somatic mutations in HRR genes or mutations in ATM, CHK2, MRN (MRE11/NBS1/RAD50), APOBEC, amongst others (Table 1). Trials studying combinations of PARPi with higher PARP1-trapping potency, such as niraparib or talazoparib, with ATRi are also ongoing (Table 1).

PARPi lead to increased RS triggering G2 cell cycle arrest[91, 101], which adds to the rationale for synthetic lethality with combined WEE1 targeting. In preclinical models of SCLC, olaparib plus adavosertib has shown to be efficacious compared to standard chemotherapy[93]. The combination of PARPi plus WEE1i has been observed to be synergistic in PARPi-resistant ovarian cancer xenografts, where evidence of increased mitotic catastrophe and apoptosis indicate that this combination may be a mean for overcoming PARPi resistance[102]. Ongoing Phase II trials of adavosertib plus olaparib are focusing on PARPi-resistant populations of ovarian cancer and other solid tumors (Table 1). Combining ATRi with WEE1i has also been described to be a potential strategy. In triple negative breast cancer (TNBC) cell lines, adavosertib plus ceralasertib provoked increased DNA damage reflected by ϒH2AX accumulation, and also forced mitotic entry of cells harboring DNA damage, culminating in cell death[103, 104]. This combination therapy also further sensitized TNBC cells to cisplatin and PARPi by inactivation of RAD51-mediated HRR[103]. Multiple early phase trials of different PARPi, ATRi, WEE1i, CHK1i and DNA-PKi combinations are currently in progress (Table 1).

Novel combinations are aiming to exploit crosslinks between metabolic pathways with replicative stress. Pathogenic gain of function mutations in IDH1/2, a tricarboxylic acid cycle enzyme that catalyzes the rate-limiting conversion of isocitrate to α-ketoglutarate, lead to the production of the ‘oncometabolite’ 2-hydroxyglutarate (2-HG) instead[105]. Accumulation of 2-HG has been shown to suppress histone demethylation and tet methylcytosine dioxygenase 2 (TET2) function, leading to the accumulation of oncogenic epigenetic changes [106]. Unexpectedly, IDH1/2 mutations have been described to induce ‘BRCAness’ or HRD phenotype, leading to PARPi sensitivity[107]. An ongoing Phase II trial is investigating ceralasertib plus olaparib in IDH1/2-mutant tumors (Table 1).

Chemotherapies such as gemcitabine and topoisomerase I/II inhibitors have been well described to increase RS[9] and have been evaluated in combination with ATRi, WEE1, CHK1i and DNA-PKi (Table 1). Gemcitabine is a pyrimidine nucleoside analog that irreversibly inhibits ribonucleotide reductase and limits nucleotide supply leading to replication stress[108]. dFdCTP, the product of gemcitabine, may also be aberrantly incorporated into DNA strands, leading to termination of replication and increased RS [108]. Increased ATR-CHK1-WEE1 axis dependency because of increased stalled forks provides a rationale for sequential treatment of tumor cells with gemcitabine followed by ATR-CHK1-WEE1 inhibition[109]. Preclinically and clinically, gemcitabine efficacy has been shown to increase when combined with various checkpoint inhibitors, including CHK1i, ATRi, and WEE1i (Table 1). Topoisomerase I inhibitors such as topotecan or irinotecan exert cytotoxicity by topoisomerase enzyme trapping on DNA, producing irreversible DNA lesions and breaks, increasing reliance on the ATR-CHK1 pathway for their resolution[110]. This finding led to ATR depletion being identified as a synthetic lethal approach when combined with topoisomerase I inhibitors[111]. Similarly, topoisomerase II inhibitors such as doxorubicin or etoposide, respectively, bind to topoisomerase II and form topoisomerase-DNA cleavage complexes[112] and rapid generation of DNA damage. DSBs produced from these complexes are reliant on NHEJ for repair[46]. In-vitro and in-vivo, the combination of peposertib with various topoisomerase inhibitors has been studied in ovarian cancer cell lines[46]. While peposertib showed no efficacy as a monotherapy, it was able to prevent DNA repair in the presence of DNA damage induced by etoposide or doxorubicin, exacerbating cell death. Peposertib was also found to be ineffective in cell lines that were resistant to topoisomerase II inhibitors[46]. Therefore, differences in the DNA lesions generated by these cytotoxic drugs may affect how they interact with RSR-targeting inhibitors, reflecting a need for further studies to investigate mechanisms underlying optimal pairing of chemotherapeutic agent with RSR inhibitor[102].

In terms of toxicity, a major challenge in the development of RSR inhibitors relates to dose-dependent myelosuppression, particularly for combination strategies with chemotherapies or PARPi. The combination of ATRi with carboplatin was shown to be challenging due to dose-dependent myelosuppression in ovarian cancer mouse models[113], as well as on Phase I trials[114], necessitating dose reductions of ATRi that may compromise efficacy[113, 114]. Creative combinatorial schedules, such as through sequential dosing of different drug classes, may be required in order to maintain acceptable therapeutic window. In ovarian cancer models, sequential PARPi (talazoparib or veliparib) followed by WEE1i was compared with concurrent therapy of both inhibitors, and surprisingly showed similar induction of DNA damage, mitotic entry of cells and reduced replication fork speed, while ameliorating toxicity[102]. Prior clinical trials have described difficulty of combining high potency PARPi with cytotoxic chemotherapy due to overlapping myelosuppression from both drug classes, leading to sequential or maintenance strategies being adopted as the favored approach in the clinic[115] – which is a principle relevant to the future clinical development of RSR-targeting drug combinations.

Novel targets – Polymerase theta inhibitors

On the horizon, polymerase theta inhibitors (Polθi) are a novel and attractive class of drugs that are in preclinical development and are slated to enter clinical trials shortly. Polθ mediates the microhomology-mediated TMEJ pathway, which functions in error-prone DSB repair and is also involved in base excision, inter-strand crosslink repair pathways and translesion synthesis[116]. Not only has Polθ been found to be upregulated in a wide array of cancers, but it has also been linked to poorer prognosis[117, 118]. In particular, HRD cells use TMEJ as a compensatory mechanism for DSB repair, leading to increased dependency on Polθ activity and synthetic lethal interactions[116]. Polθ-mediated repair leads to specific mutational signatures, which have been shown to occur more frequently in HRD-related tumor types such as breast, ovarian and pancreatic cancer[119]. Feng and colleagues have recently discovered 140 Polθ synthetic lethal genes via CRISPR genetic screens and showed that associated TMEJ addiction was linked to increased RS. From The Cancer Genome Atlas breast cancer data set, 30% of tumors harbored alterations in Polθ synthetic lethal genes and displayed genomic scars of TMEJ addiction, suggesting a larger population of patients who may retrieve benefit from this class of therapy[120]. Several preclinical studies have combined Polθ inhibition with ATMi[121], ATRi, PARPi[122] and platinum-based chemotherapy[123], showing the potential for additive or synergistic combinatorial strategies. Given that PARPi-resistant BRCA1/2-mutant cells have been described to display TMEJ-specific mutational signatures, it remains plausible that Polθ-targeting may be yet another strategy to overcome PARPi resistance[124]. At present, three different Polθ inhibitors are in development, and phase I trials are being planned by IDEAYA Biosciences, REPARE Therapeutics and Artios Pharma.

Concluding Remarks

Increasing data in recent years has uncovered the therapeutic potential of synthetic lethality of agents targeting RSR in appropriate genomic contexts, raising hopes for the advancement of precision oncology. Important questions need to be urgently addressed regarding patient selection, therapeutic index and how to personalize and define patient-specific RS, real-time in the clinic (see Outstanding questions). Before the promise of RSR-targeting strategies can truly come of age, well-designed translation in ongoing and upcoming clinical trials is paramount to inform and guide the future development of the field.

Highlights.

Elevated replicative stress in cancer cells stemming from endogenous and exogenous genotoxic stressors is a targetable vulnerability for exploitation. Inhibiting key kinases of the replicate stress response (RSR) cascade —including ATM, ATR, CHK1/2, WEE1, DNA-dependent protein kinase (DNA-PK), amongst others— abrogates cell cycle checkpoints, allowing aberrant cell cycle progression and culminating in mitotic catastrophe.

Several drug candidates that inhibit key kinases of the RSR cascade have also shown antitumor activity in preclinical and clinical studies, as well as preliminary efficacy in early phase trials: a Phase I trial using the combination of an ATM inhibitor plus placlitaxel reported several partial and durable responses in melanoma patients, who had previously progressed on immune check-point blockade therapy, and two separate trials that combined a WEE1 inhibitor with chemotherapy in platinum-sensitive and platinum-resistant advanced ovarian cancer patients showed improved PFS for the combination compared to chemotherapy alone. Aside from these, a significant number of Phase I and Phase II clinical trials using RSR-targeting agents, as monotherapy and as combinations with other inhibitors and immunotherapy agents, are currently ongoing.

Recently, it has been recently shown that targeting RSR has the potential of reversing PARP-inhibition resistance; hence, rationale combination and patient selection through synthetic lethal strategies are emerging as a highly promising therapeutic approach.