S441
Clinical - Gynaecological
ESTRO 2026
Chisato Tonoiso 1 , Ikushima Hitoshi 1 , Akiko Kubo 1 , Takashi Kawanaka 1 , Masafumi Harada 2 1 Radiotherapy, Tokushima University Hospital, Tokushima, Japan. 2 Radiology, Tokushima University Hospital, Tokushima, Japan Purpose/Objective: Cervical cancer is the fourth most common malignancy among women worldwide. Although the incidence and mortality have decreased with improved screening, approximately 45% of patients present with locally advanced disease. Local recurrence occurs in 10–20% of cases, mostly within two years after initial treatment, making early prediction essential for improving outcomes. For treatment response evaluation after chemoradiotherapy (CRT), the NCCN guidelines recommend post-treatment 18FDG-PET/CT; however, in Japan, limited high-quality evidence has prevented similar recommendations. This study aimed to evaluate the utility of 18FDG-PET/CT for assessing treatment response at the primary tumor site after CRT for cervical cancer. Material/Methods: Seventy-three patients with cervical cancer who received radical CRT at Tokushima University Hospital and underwent both pre- and post-treatment MRI and 18FDG-PET/CT were included. Primary tumor size was measured on MRI. From PET imaging, MTV, SUVmax, and TLG of the primary tumor were collected. Treatment response was assessed using RECIST and PERCIST. Optimal cutoff values for each PET metric were determined using the Youden index, and patients were stratified accordingly. ROC analysis was
performed using Cox proportional hazards models; due to multicollinearity, each PET indicator (post-treatment MTV, SUVmax, TLG, PERCIST) was analyzed in separate models. Results: ROC analysis showed the following sensitivity/specificity/PPV/NPV for predicting local recurrence: 0.70/0.84/0.41/0.96 for post- treatment MTV, 0.50/0.95/0.63/0.92 for post- treatment SUVmax, and 0.56/0.95/0.63/0.94 for post-treatment TLG. Univariate analysis showed significant associations for metabolic response (p=0.035), pretreatment MTV (p=0.031), pretreatment TLG (p=0.014), post- treatment MTV, SUVmax, and TLG (all p<0.001), and PERCIST (p=0.013). RECIST showed no significant association (p=0.549). Multivariate analysis showed significant associations for post-treatment MTV (p=0.008), SUVmax (p=0.001), and TLG (p=0.001), whereas PERCIST was not significant (p=0.152). Conclusion: While RECIST did not correlate with local control, post-treatment PET metabolic parameters (MTV, SUVmax, TLG) were strong predictive factors. Post-treatment 18FDG- PET/CT appears to complement morphological assessment and may improve accuracy in evaluating treatment response
after CRT for cervical cancer. Keywords: cervical cancer, chemoradiotherapy
Digital Poster 2111
Therapeutic results and impact of dose escalation and interstitial image-guided brachytherapy in locally advanced cervical cancer in a Tunisian cohort Omar Nouri, Eya Moalla, Oumaima Bouabidi, Fatma Dhouib, Wafa Mnejja, Hanene Ben Salah, Nejla Fourati, Henda Boucetta, Tarek Sahnoun, Leila Farhat, Wicem Siala, Jamel Daoud Radiotherapy and Oncology Department, Habib Bourguiba Hospital, Sfax, Tunisia
performed for post-treatment PET parameters. Local control rates were
estimated using the Kaplan–Meier method, and differences were analyzed with the log- rank test using age, tumor size, disease stage, histopathology, metabolic response, MTV,
SUVmax, TLG, RECIST, and PERCIST as variables. Multivariate analysis was
Made with FlippingBook - Share PDF online