Abstract 6256: Longitudinal transcriptional profiling of CTCs in metastatic breast cancer patients receiving CDK4/6 inhibitors to predict response

Abstract

Abstract Background: CDK4/6 inhibitors recently became the first choice for treatment of metastatic (M), hormone receptor-positive/HER2-negative (HR+/HER2-) breast cancer (BC) patients (pts). However, predictive markers are missing. Circulating tumor cells (CTCs) represent the heterogeneous disease in real time. Here, we aim to identify resistance markers to CDK4/6 inhibitors by mRNA profiling of CTCs. Methods: Blood of 51 HR+/HER2-MBC pts drawn at baseline, before endocrine and Palbociclib treatment (n=22 pts first line; n=29 pts second line or more), blood of 19 HR+/HER2-MBC pts drawn at baseline of endocrine monotherapy (control group; n=6 pts first line; n=13 pts second line or more), as well as blood samples of 34 of these 70 pts after six months under treatment were analyzed. Pts with stable disease within the first six months of treatment were defined as responders. Isolation of CTCs was conducted using AdnaTest EMT2/StemCell Select. Preamplified cDNA was analyzed by a new multimarker qPCR panel utilizing QuantiNova LNA Probe assays targeting 25 genes. qPCR data were normalized to CD45 and data of 20 healthy female donors. Consumables: QIAGEN, Germany. Statistical analysis was conducted via log-rank test and fisher’s exact test. Results: EpCAM, PCNA, STAT1 and YAP1 signals at baseline showed a higher prevalence in non-responders compared to responders in the Palbociclib treated cohort, while CXCR4 signals were only detected in responders. WWTR1 signals significantly correlated with reduced PFS (p=0.001) and OS (p=0.043) in all first line treated pts. YAP1 signals significantly correlated with decreased PFS (p=0.018) and OS (p=0.000) in the first line Palbociclib treated cohort and significantly correlated with decreased PFS in all first line treated pts, too (p=0.044). After six months of treatment, CDK2 signals were significantly more common in non-responders compared to responders in the Palbociclib treated cohort (p<0.05) while TEAD2 (p=0.014) and CDK2 (p=0.046) signals significantly correlated with worse OS in the Palbociclib treated group, but not in the control group. We identified a higher percentage of disappearing signals in responders than in non-responders and the percentage of newly appearing signals was higher in non-responders compared to responders. The dynamics of TEAD2 (p=0.036) and MLH3 (p=0.000) signals significantly correlated with OS in the Palbociclib treated cohort, but not in the control cohort. Conclusion: Preliminary results of the transcriptional profiling of CTCs indicate that WWTR1 and YAP1 signals at baseline might be predictive markers that do not favor Palbociclib treatment. CDK2 signals after six months or the dynamics of TEAD2 and MLH3 signals might be suitable as monitoring markers. To validate the findings, we plan to analyze additional blood samples drawn at disease progression under CDK4/6 inhibition and expand our cohort. Citation Format: Corinna Keup, Charlotte Gruber, Oliver Hoffmann, Rainer Kimmig, Sabine Kasimir-Bauer. Longitudinal transcriptional profiling of CTCs in metastatic breast cancer patients receiving CDK4/6 inhibitors to predict response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6256.