Antoine Gaudet-Chardonnet

Platinum free interval and clinical benefit of the second-line chemotherapy in recurrent uterine and ovarian carcinosarcoma: a retrospective cohort analysis

Uterine and ovarian carcinosarcomas (OCSs) are rare and aggressive neoplasms. We assessed whether progression free survival after initial treatment (PFS1) was associated with the clinical benefit of chemotherapy after progression, estimated as overall survival (OS) after progression/relapse. All consecutive patients treated with chemotherapy for stage I-IV uterine/OCS in Cochin University Hospital between 2010 and 2022 were included in this retrospective cohort. Association between PFS1 and OS after progressive disease (PD) was determined by Cox regression. Optimal PFS1 threshold for OS after PD prediction was determined by a time-dependent receiver operating characteristic-curve analysis. Forty patients treated for endometrial (n=32) or OCS (n=8) were included. Median PFS1 and OS after PD were 16 months 95% confidence interval (95% CI=11-not available [NA]) and 6 months (95% CI=2-15). In patients who relapsed/progressed (n=20), OS after PD was anticipated by PFS1 (Pearson r=0.61; area under the curve=0.79; 95% CI=0.6-1). At the threshold of PFS1 ≤/>9 months (n=6/n=7), median OS post PD were 2 months (0.1-NA) and 15 months (6-NA), for patients treated with platinum/anthracycline based chemotherapy in second line. Patients receiving best supportive care alone (n=7) had a median OS post PD of 8 months (1.3-NA). Our results highlight that a subgroup of carcinosarcomas patients exhibits a durable benefit from chemotherapy in the relapse settings, and suggest the use of PFS1, as a proxy of platinum-sensitivity, to select patients who might derive higher clinical benefit of a 2nd line of chemotherapy.

Identification of a very-high risk subgroup of localized endometrial carcinoma before surgery using circulating tumor DNA: a proof-of-concept study

We aimed to study whether the detection of circulating tumor DNA (ctDNA) may predict the risk of early relapse for patients with localized endometrial carcinoma. Patients who underwent surgical resection at Cochin University Hospital (2021-2023) for International Federation of Gynecology and Obstetrics 2018 stage I to III endometrial carcinoma were prospectively included in a prospective biocollection cohort study. All patients had a plasma sample before surgery (EDTA collection tubes, 4-5 mL). After extraction and bisulfite-conversion of cell-free DNA, ctDNA was evaluated using a droplet-digital polymerase chain reaction assay targeting universally-hypermethylated positions in endometrial carcinoma (OXT, ZSCAN12 genes), and defined as significantly detected above the limit of detection. Patients were classified as high-risk based on 2022 European Society for Medical Oncology/European Society of Gynaecological Oncology/European Society of Pathology guidelines, or preoperative features (non-endometrioid histology, p53-abnormal tumors, or stage III). Events of interest were tumor progression or relapse (event-free survival). Adjusted-HR (aHR) was estimated using Cox regression. Among 128 patients included with median follow-up of 26 months (interquartile range; 15-35), ctDNA was detected in 18 patients (14%). Patients with ctDNA had a 1-year event-free rate of 67% (95% CI [48% to 92%]), vs 91% [82% to 100%] among patients without ctDNA. The ctDNA was detected in 10 (29%) patients among those with preoperative high-risk features (N = 34, 1-year event-free rate = 60% [36%-100%]). ctDNA was associated with event-free survival independently of stage (aHR = 4.26 [1.68-10.8]), 2022 guidelines high-risk (aHR = 3.72 [1.57-8.87]), or preoperative high-risk features (aHR = 3.98 [1.65-9.60]). Elevated ctDNA before surgery identifies a very high-risk subgroup of newly diagnosed endometrial carcinoma, suggestive of occult metastasis. Further studies are warranted to validate this finding and investigate the window of opportunity for neoadjuvant approaches.