Therapeutic vulnerability to ATR inhibition in concurrent NF1 and ATRX- deficient high-grade solid tumors

This is the authors' abstract. We don't add key points for this paper.

Subsets of Neurofibromatosis Type 1 (NF1)-associated solid tumors have been shown to display high frequencies of ATRX mutations and presence of alternative lengthening of telomeres (ALT). In addition, a potential biologic vulnerability of ALT-positive cancer cells is ATR inhibition, a regulator of homologous recombination. We studied the phenotype of combined NF1 and ATRX deficiency in malignant solid tumors. For these experiments, cell lines derived from NF1-deficient sporadic glioblastomas (U251, SF188), a NF1-associated ATRX mutant glioblastoma cell line (JHH-NF1-GBM1), a NF1-derived sarcoma cell line (JHH-CRC65), and two NF1-deficient MPNST cell lines (ST88-14, NF90.8) were utilized. Cancer cells were treated with ATR inhibitors (AZD6738 and VE-822), in vitro, with or without a MEK inhibitor (AZD6244, selumetinib) or temozolomide. In contrast to the glioma cell line SF188, combined ATRX knockout (KO) and TERC KO led to ALT-like properties and sensitized U251 glioma cells to ATR inhibition (AZD6738 and VE-822) in vitro and in vivo. In addition, ATR inhibitors sensitized U251 cells to temozolomide, but not MEK inhibition (AZD6244), irrespective of ATRX level manipulation; whereas, the JHH-NF1-GBM1 cell line (ATRX loss/ALT-positive) demonstrated sensitivity to ATR inhibition (AZD6738), but not temozolomide. Similar effects were noted using the MPNST cell line NF90.8 after combined ATRX knockdown and TERC KO; however, not in the MPNST cell line ST88-14. Taken together, our study supports the feasibility of targeting the ATR pathway in subsets of NF1-deficient and associated tumors. Tumors with pre-existing ALT, or that subsequently develop ALT after ATRX downregulation, are particularly vulnerable to this therapeutic approach.