3202 - Luminal Subtypes in Metastatic Prostate Cancer Circulating Tumor Cells Mirror Tissue and Are Associated with Prognosis and Response to 177Lu-PSMA-617
Presenter(s)
S. C. Callahan1, M. Sharifi2, J. Sperger3, A. K. Taylor4, K. E. Tippins4, S. R. Reese3, V. Carreno4, K. R. Kaufmann3, A. H. Chang4, L. A. Nunamaker3, C. Linebarger3, L. Mora-Rodriguez3, J. Schehr5, H. M. Krause3, K. T. Helzer6, M. L. Bootsma6, G. C. Blitzer7, J. M. Floberg6, J. Lang5, and S. G. Zhao8; 1UW Carbone Cancer Center, Madison, WI, 2Department of Human Oncology, University of Wisconsin Hospitals and Clinics, Madison, WI, 3Carbone Cancer Center, University of Wisconsin-Madison, Madison, WI, 4University of Wisconsin-Madison, Madison, WI, 5University of Wisconsin School of Medicine and Public Health, Madison, WI, 6Department of Human Oncology, University of Wisconsin-Madison, Madison, WI, 7Department of Human Oncology, University of Wisconsin Hospital and Clinics, Madison, WI, 8University of Wisconsin, Madison, WI
Purpose/Objective(s):
Luminal subtypes have been described in localized and metastatic prostate cancer. These subtypes have both prognostic and treatment response implications, primarily with regards to AR-directed therapies. However, the association of these subtypes with newer therapies like 177Lu-PSMA-617 is not known. A major limitation in measuring real-time treatment response is the inherent challenge of collecting serial tissue biopsies on treatment; however, this can be overcome using blood-based liquid biopsies.Materials/Methods:
Utilizing a novel circulating tumor cell (CTC) isolation approach, we collected over 270 CTC samples from patients with metastatic prostate cancer for RNA sequencing. We then classified the CTCs into transcriptional subtypes mirroring those identified in tissue. In addition to our CTC-based approach, we also examined 203 publicly available mCRPC tissue biopsy RNA sequencing samples with available overall survival outcomes. In order to investigate radiation response, we utilized a prostate cancer radiation response signature, PORTOS, which has been validated in multiple randomized trials. To calculate PORTOS scores, the CTC and metastatic tissue gene expression were first normalized to the original training dataset of PORTOS (GSE46691) using COMBAT. Scores were then calculated on the normalized gene expression values in the CTCs and metastatic tissue datasets using the original PORTOS model.Results:
We identified CTC transcriptional subtypes mirroring those previously identified in tissue studies. Patients with a luminal-B-like CTC phenotype (LumB), defined by persistent AR signaling and high proliferation, had significantly shorter survival than patients with luminal-A-like phenotypes (LumA) in both CTC and tissue cohorts. Additionally, pre-treatment CTC LumB was associated with early progression on 177Lu-PSMA-617. Interestingly, we observed that PORTOS scores were lower in LumB vs. LumA samples in both the CTC and tissue cohorts, consistent with less benefit from radioligand therapy in the LumB phenotype.Conclusion:
We report the largest CTC RNA sequencing cohort of patients with metastatic prostate cancer and demonstrate that CTC RNA-seq can identify clinically important prostate cancer transcriptional subtypes. LumB patients have shorter overall survival, appear to derive less benefit from 177Lu-PSMA-617, and have lower PORTOS scores than LumA patients. While exploratory in nature, these findings suggest the LumB subtype may predict poor response to radiopharmaceutical therapy.