3232 - Comparison of Circulating Tumor DNA and PSA after Prostate Cancer Treatment: A Retrospective Analysis
Presenter(s)
E. Garemanian1, S. M. Dufault2, O. Yazdanpanah3, S. Dwabe3, D. Kaakour3, S. Feinstein4, N. Mar5, A. Rezazadeh6, and S. N. Seyedin7; 1UC Irvine Health, Orange, CA, 2University of California, San Francisco, Department of Epidemiology and Biostatistics, San Francisco, CA, 3University of California, Irvine, Division of Hematology/Oncology, Orange, CA, 4University of California, Irvine, Department of Radiation Oncology, Orange, CA, 5UC Irvine School Of Medicine, Division of Hematology/Oncology, Department of Medicine, Orange, CA, 6Division of Hematology/Oncology, Department of Medicine, UC Irvine School Of Medicine, Orange, CA, 7University of California, San Francisco, Department of Radiation Oncology, San Francisco, CA
Purpose/Objective(s): The ability of circulating tumor DNA (ctDNA) assays to find minimal residual disease (MRD) in prostate cancer remains unclear. We hypothesize that ctDNA would be detectable for those with more advanced disease compared to their localized counterpart.
Materials/Methods: We retrospectively selected prostate cancer patients who underwent tumor-informed ctDNA assessment at our institution at any time. Standard clinical variables extracted include PSA, TNM stage, treatment after obtaining ctDNA, and extent of disease at the time of ctDNA collection. The primary endpoint of this study was the rate of ctDNA detection. Secondarily, we also searched for patient and tumor characteristics correlated with positive ctDNA. Finally for patients who completed ctDNA assessment after treatment, patterns of relapse were investigated. Wilcoxon-rank sum and Fisher’s exact tests were used to evaluate continuous and categorical associations, respectively.
Results: Among 31 patients selected, 24 presented with clinically localized disease at diagnosis. Five patients presented with pelvic nodal disease. Grade group 3 disease (19%) was the most prevalent. Ten patients (32%) exhibited a positive ctDNA. The median PSA of those with appreciable ctDNA was numerically higher but this was not significant (2.00 vs 0.3, p = 0.14). At the time of ctDNA assessment, patients with a positive result displayed different patterns of relapse than those with a negative result (Fisher’s exact p < 0.001); notably, higher rates of osseus metastasis (60% vs 14%) but not pelvic node positive disease (30% vs 29%). None of the patients with prostate confined disease or PSA relapse alone after prostatectomy demonstrated a positive ctDNA result. 23 patients then received androgen deprivation and radiation therapy in the form of either metastatic directed therapy, prostate bed or definitive upfront treatment. After treatment, only one patient remained ctDNA positive with residual PSA value of 0.3 ng/ml. He had completed SBRT to left iliac and sacrum, exhibited residual ctDNA disease two months later, and then developed another metastasis at C5 one month later. Among the 19 patients with undetectable ctDNA after treatment, only four displayed a PSA level at or above 0.1 with a median of 0.2.
Conclusion: ctDNA detection was significantly associated with osseous metastases but was not detected in patients with localized disease or biochemical recurrence alone. Nearly all patients converted to ctDNA-negative after treatment, suggesting PSA remains a more sensitive MRD marker. Future research will explore whole-genome sequencing-based tumor-informed ctDNA assays to improve MRD detection in localized prostate cancer.