Circulating tumor DNA (ctDNA) can be used as an early predictor of treatment outcome in patients with gynecologic cancers receiving radiotherapy

Summary: Assessing treatment outcomes of radiotherapy in patients with GYN malignancies requires careful follow up with imaging and clinical exams. Early detection of ctDNA in blood draw may provide a non-invasive way to predict response to treatment and progression. In this study, we used ctDNA to assess response to RT in patients with GYN cancers. After IRB approval, patients with vulvar, cervical, primary,and recurrent and metastatic endometrial cancer were treated with RT at our institution between 2022-2024. ctDNA was obtained using Signatera™ test (Natera Inc.) pre-RT, mid-way through RT, pre-boost with brachytherapy or SBRT, at the end and in follow-up at 1, 3, 6 and every 6 months -post RT, respectively. During and after RT, ctDNA of 0.00MTM/ml was defined as complete metabolic response (cMR) but a reduction without achieving the value 0.00MTM/ml in ctDNA was deemed partial metabolic response (pMR). Correlation between ctDNA levels and imaging (PET-CT, MRI, CT) was also assessed. Statistical analyses used were descriptive statistics, Wilcoxon rank-sum test, Chi-square test and a spearman-rank correlation coefficient (ρ). A total of 179 serial ctDNA blood draws were obtained from 32 patients with 8 (25%) vulvar, 2 (6%) vaginal, 10 (31%) cervical, 2 (6%) neuroendocrine, 4 (13%) primary endometrial, 2 (6%) metastatic endometrial and 4 (13%) recurrent endometrial cancers (reEMCa). Median age was 65 years (range, 35-90 years). Median follow-up was 9 months (range, 1-25 months). Median number of ctDNA draws per patient was 6 (range, 1-10). Median radiation dose was 5900 cGy (range, 4500-7000 cGy), brachy boost of 28Gy/4fx T&O and SBRT 30Gy/5fx for cervix cases, and SBRT boost 2750cGy/5fx for reEMCa. There was significant reduction in ctDNA values (metabolic response) from pre-to post-RT (median 1.2 vs. 0, p=0.0004): 73% cMR and 27% pMR. A strong correlation was observed between elevated ctDNA and SUV/measurable disease on imaging pre-treatment (ρ=0.67, p = 0.001), as well as undetectable ctDNA and decline of SUV/complete resolution of SUV at 3-6 months following RT (p=0.01). Among 4 patients (1 vulvar, 2 cervical and 1 neuroendocrine) with radiographical progression in follow-up, 100% had elevated ctDNA preceding radiographical progression. The median lead time between the elevated ctDNA and radiographic appearance was 63 days (range 5-177 days). ctDNA dynamics correlated with the response to radiotherapy/chemoradiotherapy in patient with gynecological malignancies. The detection of post-treatment elevated ctDNA may serve as an earlier predictive biomarker of progression than radiography. A larger prospective evaluation is warranted. Bin Gui, MD (Presenting Author) - Northwell; Lily Ottensoser, MD (Co-Author) - Northwell; Bhupesh Parashar, MD, DrPH (Co-Author) - Northwell; Louis potters, MD (Co-Author) - Northwell; Gabriella Wernicke, MD, MS (Co-Author) - Northwell