Long-term oncological and reproductive outcomes following treatment with the levonorgestrel intra-uterine device for endometrial hyperplasia and adenocarcinoma: results from the feMMe phase 2 randomized clinical trial

OBJECTIVE: The feMMe phase 2 randomized trial demonstrated that approximately 60% of participants with endometrial hyperplasia with atypia or clinical stage 1, International Federation of Gynecology and Obstetrics grade 1 endometrioid adenocarcinoma had a pathological complete response after 6 months of treatment with the levonorgestrel intra-uterine device (NCT01686126). The objective of this study was to determine the long-term oncological and reproductive outcomes of a sub-group of feMMe trial participants in Queensland. METHODS: A total of 96 feMMe trial participants from 6 sites in Queensland were contacted to participate in this study, and 52 (54%) consented. Medical records were retrieved for 51 participants, and 44 participants completed questionnaires. RESULTS: Median follow-up time after trial completion was 18.9 months (range; 0.2-45.1). Of the 33 participants who had a pathological complete response at trial completion, 21 of 33 (64%) had a sustained complete response throughout follow-up (range; 6.7-43.5). Recurrence developed in 8 of 33 participants (24%). The probability of recurrence-free survival was 93% (95% confidence interval 84% to 100%) at 12 months and 74% (95% confidence interval 59% to 93%) at 24 months. Four of 33 participants (12%) did not have endometrial sampling beyond trial completion. Of the 19 participants who did not have a pathological complete response at trial completion, 4 of 19 (21%) went on to achieve a complete response. Persistent disease occurred in 5 of 19 participants (26%), and 10 of 19 participants (53%) did not have endometrial sampling beyond trial completion. Overall, 21 of 52 participants (40%) had a hysterectomy. Diagnostic concordance between the latest biopsy and the hysterectomy specimen was 62% (13 of 21; κ = 0.44). Four participants attempted to conceive, and the live birth rate was 50%. CONCLUSIONS: Long-term follow-up of feMMe trial participants demonstrated that most participants who developed a pathological complete response at trial completion were likely to sustain this response, while those who did not respond at trial completion were likely to require a hysterectomy for disease control.