DNA-PK enforces the S/G2 checkpoint in ATM-deficient cells to limit ATR inhibitor efficacy

The long-standing consensus in the DNA damage response field is that ATR and its downstream effector kinase CHK1 prevent cells from entering mitosis with damaged or incompletely replicated DNA when encountering replication stress or DNA damage, and play central roles in suppressing excessive replication origin activation and alleviating replication stress ^[1]. ATR forms an interdependent complex with ATRIP, is recruited to damage sites by recognizing RPA-coated single-stranded DNA, and initiates checkpoint signaling ^[2][3]. Because tumor cells commonly harbor p53 mutations causing G1 checkpoint defects, they rely more heavily on S and G2/M checkpoints, making ATR an attractive therapeutic target, and multiple ATR inhibitors have entered clinical development ^[1][4]; ATM-deficient tumors are considered a synthetic lethal indication for ATR inhibition ^[4]. However, preclinical and clinical observations suggest that ATM and DNA-PK can compensatorily maintain checkpoint function, limiting the efficacy of ATR inhibitor monotherapy, yet the respective roles of the two kinases during ATR inhibition were previously unclear.

This study systematically dissected the division of labor between DNA-PK and ATM under ATR inhibition exposure. The authors found that both kinases are activated at the S/G2 boundary in a cell cycle-dependent manner after ATR inhibition, but with distinct functions: DNA-PK maintains the S/G2 checkpoint and promotes completion of DNA replication; ATM recognizes persistent abnormal replication structures and their accompanying DNA damage in late S phase and G2/M, establishing the G2/M checkpoint to facilitate repair. In ATM-deficient cells and patient tumors, the situation changes—DNA-PK is activated after ATR inhibition and displays greater multifunctionality: it not only continues to maintain the S/G2 checkpoint but also replaces ATM in recognizing abnormal replication structures during S phase, enabling cells to tolerate ATR inhibition. This finding echoes previous reports of modular integration and functional interplay among ATM, ATR, and DNA-PKcs in the DNA damage response ^[5][6].

Based on this mechanism, the authors demonstrated that DNA-PK inhibition synergizes with ATR inhibition in ATM-deficient cells and enhanced the efficacy of ATR inhibitors in ATM-deficient patient-derived xenograft (PDX) models. This result provides direct translational rationale for combined inhibition of ATR and DNA-PK in ATM-deficient tumors, and also explains why ATR inhibitor monotherapy strategies using ATM loss alone as a biomarker have shown limited efficacy. Consistent with prior studies, dual inhibition of ATR and DNA-PKcs in ATM-deficient prostate cancer models blocked downstream DDR signaling and achieved radiosensitization ^[7][8], indicating that combined targeting of these two pathways has cross-model reproducibility in ATM-deficient tumors. Related drug delivery and toxicity management issues remain to be answered at the clinical level, particularly since DNA-PK inhibition may impair DNA repair in normal tissues.

It should be noted that this study was mainly based on cell models and PDX models; the tolerability of combined inhibition in patients, dosing regimens, and whether there are other predictive biomarkers beyond ATM loss still need to be tested in clinical trials. The authors also only explicitly proposed clinical evaluation of combined ATR and DNA-PK inhibition in ATM-deficient cancers, and its applicability outside this population remains unknown.