MCM2 and the origin licensing complex: A putative node of ER+/HER2- breast cancer therapy resistance

Background: The use of CDK4/6 inhibitors (CDK4/6i) in combination with endocrine therapy (ET) has revolutionized treatment practices for advanced and high-risk ER+/HER2- breast cancer (BC) via effective inhibition of cell cycle activation. However, the vast majority of patients eventually progress on these therapies, underscoring the need for new treatment strategies to select patients, monitor response, and target novel therapeutic vulnerabilities. We previously reported that high expression of MCM2 and its origin licensing complex (LC) is enriched in primary tumors refractory to CDK4/6i (SABCS22-P2-03-10). LC factors are transcriptionally regulated by E2F/MYC-driven programs, which are activated by diverse drivers of resistance. We propose that MCM2 /LC expression may be a biomarker reflecting multiple ET and CDK4/6i resistance mechanisms, due to its role in cell cycle activation. Methods: A retrospective review was conducted on a real-world, treatment-naive ER+/HER2- BC cohort of 1,960 largely metastatic patients with biopsies from breast or other sites (~50% each) that underwent NGS profiling (DNA: 592-gene panel/whole-exome; RNA: whole-transcriptome) at Caris Life Sciences. Tumors were stratified by MCM2 or LC gene set mRNA into Q4 (high) versus Q1 (low; n=490 each), balanced for ET/CDK4/6i exposure, MSI-H, and TMB-H. Histology, PAM50, mutations ( mut ), gene amplifications( amp ) and deletions( del ), and differential gene expression were analyzed. Chi-square, Fisher's Exact, or Mann-Whitney U tests were performed with Benjamini-Hochberg FDR for molecular variables (q<0.05). Results: MCM2 -high vs -low tumors were LumB-dominant (LumB 85.5% vs 39.8%; LumA 5.3% vs 56.3%, p<1e-63) with less lobular histology (6.1% vs 17.1%, p=0.0001). Transcriptomically, PI3K/AKT/mTOR, E2F/MYC targets, G2M checkpoint, and DNA repair hallmark pathways were enriched in MCM2 -high (all FDR q<0.05). Genomically, MCM2 -high harbored a higher frequency of TP53 mut (32.6% vs 16.6%, q=4e-6), RB1 mut (5.8% vs 1.2%, q=0.052), CCND1 amp (26% vs 6%, q=4.5e-11), FGF3 / 4 / 19 amp (~25% vs ~6%, q≤2.2e-10), MYC amp (5.4% vs 0.8%, q=0.0136), and CDKN2A / B del ( CDKN2A 29.3% vs 15.5%, q=6.9e-4; CDKN2B 17.5% vs 7.3%, q=1.38e-3). Notably, MCM2 and MKI67 gene expression were highly correlated (Spearman's ρ=0.77, p<0.05). Very similar results were found for LC-high vs -low tumors. Conclusions: In ER+/HER2- BC, high MCM2 /LC is associated with LumB subtype, E2F/MYC and cell cycle activation, selection of TP53 and RB1 mutations, and CCND1 / MYC / FGF amplifications. Our findings suggest that MCM2 or LC mRNA may act as a surrogate marker for a network of oncogenic pathways previously identified to drive endocrine and CDK4/6i resistance. Future studies will evaluate MCM2 /LC beyond proliferation, as a predictive biomarker for CDK4/6i-based regimens, as we have previously shown in the primary ET-treated setting (SABCS24-P1-03-20).