Actionable co-alterations in breast tumors with pathogenic mutations in the homologous recombination DNA damage repair pathway
BREAST CANCER RESEARCH AND TREATMENT
Authors: Heeke, Arielle L.; Xiu, Joanne; Elliott, Andrew; Korn, W. Michael; Lynce, Filipa; Pohlmann, Paula R.; Isaacs, Claudine; Swain, Sandra M.; Vidal, Gregory; Schwartzberg, Lee S.; Tan, Antoinette R.
Abstract
Purpose Homologous recombination (HR)-deficient breast tumors may have genomic alterations that predict response to treatment with PARP inhibitors and other targeted therapies. Methods Comprehensive molecular profiles of 4647 breast tumors performed at Caris Life Sciences using 592-gene NGS were reviewed to identify somatic pathogenic mutations in HR genesARID1A, ATM, ATRX, BAP1, BARD1, BLM, BRCA1/2, BRIP1, CHEK1/2, FANCA/C/D2/E/F/G/L, KMT2D, MRE11, NBN, PALB2, RAD50/51/51B,andWRN,as well as 41 markers that may be associated with treatment response to targeted anticancer therapies. Results 17.9% of breast tumors had HR mutations (HR-MT, 831/4647) [ER/PR+ , HER2- 18.3%,n = 2183; TNBC 18.2%,n = 1568; ER/PR+ , HER2+ 15.6%,n = 237; ER/PR-, HER2+ 12.9%,n = 217; unknownn = 442]. Mean TMB was higher for HR-MT tumors across subtypes (9.2 mut/Mb vs 7.6 h-wild type (HR-WT),p <= 0.0001) and independent of microsatellite status. MSI-H/dMMR was more frequent among HR-MT tumors (2.1% HR-MT vs 0.2% HR-WT,p <= 0.0001), as was tumor PD-L1 overexpression (13.2% HR-MT vs 11.0% HR-WT,p = 0.08). Additional co-alterations were similar between HR-MT and HR-WT, with the exception ofPIK3CA(30.3% HR-WT vs 26.4% HR-MT,p = 0.024) andAKT1(3.7% HR-WT vs 2.1% HR-MT,p = 0.021). AR overexpression andPIK3CAmutations were more common among ER/PR+ tumors. ERBB2 mutations were seen in both HER2+ and HER2- tumors. Conclusions HR-MT was common across breast cancer subtypes and co-occurred more frequently with markers of response to immunotherapy (MSI-H/dMMR, TMB) compared to HR-WT tumors. Mutations were identified in both HR-MT and HR-WT tumors that suggest other targets for treatment. Clinical trials combining HRD-targeted agents and immunotherapy are underway and could be enriched through comprehensive molecular profiling.
DNA damage signaling in response to 5-fluorouracil in three colorectal cancer cell lines with different mismatch repair and TP53 status
INTERNATIONAL JOURNAL OF ONCOLOGY
Authors: Adamsen, Birgitte L.; Kravik, Katherine L.; De Angelis, Paula M.
Abstract
We studied patterns of DNA damage signaling and cell cycle response to clinically-relevant (bolus) and high doses of 5-fluorouracil (5-FU) in three colorectal cancer cell lines with differing MMR and TP53 status in an attempt to better understand how 5-FU exerts its cytotoxicity. The ATM/CHEK2/CHEK1 signaling pathway was not activated in response to bolus 5-FU in the MMR-deficient cell lines HCT116 (TP53-proficient or TP53-depleted) and HCT15 (TP53-deficient), consistent with negligible/reparable DNA damage and no cell death. The pattern of DNA damage checkpoint activation in bolus 5-FU-treated HT29 (TP53-deficient/MMR-proficient) cultures suggested SSB formation (CHEK1 activation) followed by DSB formation (CHEK2 activation and increased phospho-H2AX levels), but no cell death suggested that DNA repair capacity was not overwhelmed. High-dose 5-FU treatment led to activation of ATM/CHEK2/TP53 (not CHEK1) in TP53-proficient and TP53-depleted HCT116 (later CHEK2 activation relative to TP53-proficient) cultures; HCT15 cultures had ATM activation only. These data and increased phospho-H2AX levels indicated DSB formation; apoptosis was induced in both cell lines indicating irreparable DNA damage. TP53-depleted HCT116 cultures also had DSBs after high-dose 5-FU treatment but experienced a (transient) G(1)/S cell cycle arrest that protected them from apoptosis. TP53 phosphorylation at Ser20/33/37 was seen in TP53-proficient HCT116 cultures regardless of 5-FU concentration at >= 4 h following treatment, indicating TP53 stabilization/transcriptional activation. Overall, activation of ATM, CHEK1 and/or CHEK2 and phospho-H2AX levels reflected the nature of 5-FU-induced DNA damage and indicated when DNA damage was significant (5-FU-dose-dependent). DNA repair and cell cycle responses to 5-FU-induced DNA damage were distinctly affected by MMR and TP53 (role,in BER/NER) functionalities, but MMR deficiency especially seemed to confer less overall sensitivity to 5-FU.