ESTRO 2024 - Abstract Book

S1010

Clinical - Gynaecology

ESTRO 2024

We collected data from 1416 patients with locally advanced cervical cancer who were treated at 24 different centres between 2008 and 2015. The treatment regimen consisted of EBRT to the pelvic region, with or without paraaortic and/or inguinal lymph node irradiation, delivering a dose of 45-50 Gy in 1.8 – 2 Gy fractions, followed by MRI-guided adaptive brachytherapy. Concurrent chemotherapy included up to 7 weekly cycles of cisplatin at doses of either 40 mg/m² of body surface area (full), 30 mg/m² (reduced), or 0 mg/m² (paused). Cumulative cisplatin dose for each patient was calculated by summing all administered cycles. The study was registered with ClinicalTrials.gov, NCT00920920. Patients were followed up at 3-month intervals during year one, 6-month intervals during the second and third years, and annually thereafter. Distant metastasis was defined as any recurrence outside the pelvic and paraaortic region. Recurrences within the pelvic and paraaortic region were defined as local or regional event. Only the first instance of recurrence was recorded. To compare incidence rates of distant metastasis between patients with different cumulative administered cisplatin doses, the cohort was divided into groups with dose levels ≤80, 90 and ≤160, 170 and ≤200, or >200 mg/m2 body surface area to investigate statistically significant differences. For each group, the cause-specific hazard rate for a distant metastasis was calculated from the observed number of events and person years. In addition, this analysis was also performed separately for risk groups based on statistically significant risk factors for distant metastasis identified in a previous analysis [1]. Only significant risk factors available at baseline were selected: Low-risk was defined as squamous cell carcinoma (SQ) and node negative (N0), high-risk as non-SQ (adeno- and adenosquamous carcinoma) and/or node positive (N1). A cumulative incidence analysis was performed to calculate the probabilities that after 60 months a patient had experienced (i) distant metastasis, (ii) a local or regional event without distant metastasis, or (iii) had died without recurrence. Based on the Fine-Gray competing risks regression model, the subdistributional hazard ratios for distant metastasis were calculated between the lowest dose and all further dose levels. A total of 1318 patients were eligible for overall analysis. The distribution of FIGO2009 stages was: IB1 (8%), IB2 (9%), IIA (5%), IIB (52%), IIIA (1%), IIIB (14%), IVA (3%), and IVB (Para-aortic nodes only, 7%). Among the population, 48% were N0, and 52% were N1. Histology was distributed as 82% SQ, and 18% non-SQ. 77 patients received chemotherapy other than cisplatin and were also excluded. The estimated incidence rates of distant metastasis (and standard errors) at 60 months for the four dose level groups from low to high dose were 0.28 (0.07), 0.23 (0.03), 0.15 (0.02), 0.21 (0.03). In low-risk patients the corresponding rates were 0.11 (0.06), 0.15 (0.04), 0.09 (0.02) and 0.10 (0.04), suggesting a similar outcome for all dose levels in this subgroup. In high-risk patients, the corresponding rates were 0.43 (0.11), 0.29 (0.05), 0.20 (0.03), 0.26 (0.04), indicating a lower incidence of distant metastasis at doses of 170-200 m². Figure 1 and 2 illustrate the respective dose relationships. Results: