Abstract P4-08-06: Clock Genes in Breast Cancer

Abstract

Abstract Background: Disruption of circadian processes has been linked to cancer initiation, progression, metastasis, resistance, and mortality. Clock proteins are an emerging target for therapy in breast cancer. Circadian rhythms are controlled by a network of transcription/translation feedback loops primarily driven by BMAL and CLOCK and the transcriptional repressors period (PER1-3) and cryptochrome (CRY1-2). We investigated the molecular and clinical associations of clock genes in breast cancer. Methods: A total of 9563 breast tumors underwent molecular profiling (Caris Life Sciences). Analyses included next-generation sequencing of DNA (592 genes-NextSeq, WES-NovaSeq) and RNA (NovaSeq). Clock gene Score (CS) was determined using expression of clock pathway gene Z scores (positives of BMAL, CLOCK and negatives of PER1/2 and CRY1/2) and then stratified into quartiles. xCell was used to quantify immune cell infiltration in the tumor microenvironment (TME). ER/PR was tested by IHC and HER2 was tested by either IHC or CISH. Significance was determined as P values adjusted for multiple comparison (Q) of < 0.05. Real-world survival information was obtained from insurance claims data and was calculated from either tissue collection to last contact or time on treatment (TOT); comparison was done by Kaplan-Meier test. Results: TNBC had the highest median CS score, while HR+/HER2- had the lowest CS (0.96 vs 0.26 q<.001). In TNBC, PDL1 was significantly associated with higher CS (56% Q4 vs 28% Q1, q<.05). TP53 mutation was associated with higher CS (88% Q4 vs 75% Q1), while CDH1 and STK11 mutations were associated with lower CS (3.4% Q4 vs 8.7% Q1 and 0.1% Q4 vs 2.2% Q1). For the TME (xCell) in TNBC, CD8+ T cells, B cells, monocytes and NK cells were positively associated with CS, whereas CD4+ central and effector memory T cells, eosinophils, and endothelial cells were associated with lower CS (Q1 vs Q4 all q<.05). In HR+/HER2- tumors, PDL1 was also associated with higher CS (24% Q4 vs 14% Q1, q<.05). TP53 mutations (39% Q4 vs 23% Q1), HMGA2 (2% Q4 vs 0.4% Q1) and LGR5 amplifications (3% Q4 vs 0.4% Q1), and LOH (WES) (35% Q4 vs 21% Q1) were associated with higher CS. CDH1(12% Q4 vs 23% Q1), KMT2C (6% Q4 vs 10% Q1) and PIK3CA mutations (37% Q4 vs 45% Q1) were associated with lower CS (all q<.05). In HR+/HER2- tumors there was a decrease in CD4+ central memory cells, common myeloid progenitor cells, endothelial cells, and eosinophils in high CS tumors. Activated myeloid dendritic cells, B cells, CD4+ memory T cells, CD8+ naïve T cells, M1 macrophages, and Tregs all had higher abundance in high CS tumors. In HR+/HER2+ tumors, PDL1 trended positively with CS but did not reach significance likely due to sample size. For all tumors, PDL1 expression positively correlated with CS (17% Q1 vs 37% Q4) while ER (75% Q1 vs 55% Q4) and PR (49% Q1 vs 38% Q4, all q<.05) were negatively associated; no association was seen with HER2. Expression of TIMELESS (HR:0.7, CI: 0.65-0.77) and CLOCK (HR: 0.8, CI: 0.72-0.86) below median were associated with longer OS, while expression of CRY2 (HR: 1.4, CI: 1.3-1.6); PER2/3 (HR: 1.1, CI:1.0-1.2; HR:1.3, CI:1.2-1.4) above median was associated with longer OS. In TNBC, TOT IO therapy was prolonged with higher expression of CLOCK (HR: 0.5, CI: 0.41-0.72), TIMELESS (HR: 0.7 CI: 0.53-0.91), ARNTL (HR: 0.7 CI:0.54-0.92) and CRY1/2 (HR: 0.6, CI: 0.46-0.80; HR: 0.75 CI: 0.57-0.98). Conclusions: Dysregulation of clock genes is strongly associated with breast cancer subtype and survival. Higher CS is associated with TNBC and PDL1 expression and supports the use of checkpoint inhibitors. Prognosis is better with low expression of TIMELESS and CLOCK and high expression of CRY2 and PER2/3, suggesting a role in tumor development and maintenance. HR+/HER2- tumors have lowest CS, fitting with less aggressive phenotype and better prognosis. Clock genes are novel predictive and prognostic molecular markers and emerging targets for the development of new treatments in breast cancer. Citation Format: Priya Jayachandran, Yasmine Baca, Joanne Xiu, Yuanzhong Pan, Phil Walker, Francesca Battaglin, Hiroyuki Arai, Moh’d Khushman, Janice Lu, Darcy Spicer, Shannon Mumenthaler, Richard Goldberg, Benjamin Weinberg, Emil Lou, Michael Hall, Arielle L. Heeke, W. Michael Korn, Steve A. Kay, Heinz-Josef Lenz, Evanthia T. Roussos Torres. Clock Genes in Breast Cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-08-06.