RPS 2407 - Female pelvic cancer imaging: what's new?

Author Block: R. Chianura, P. P. Azzaro, C. Panico, B. Gui, E. Sala, G. Avesani, A. Fagotti, A. Rosati, D. Arrigo; Rome/IT
Purpose: To assess the usefulness of three-dimensional reconstructions (3DR) and virtual reality (VR) derived from CT/MRI datasets in complex gynecologic malignancies, with a focus on multidisciplinary decision-making and surgical planning.
Methods or Background: In this single-center pilot study, 40 patients undergoing highly challenging surgery (e.g., pelvic exenteration; recurrences involving muscle, vessels, or bone; tumors at difficult sites such as the hepatic hilum) were included. CT (n = 24) and multiparametric MRI (n = 16) datasets were segmented and converted into interactive 3D models; VR simulations were generated from the same meshes. Four gynecologic oncology surgeons, assisted by a urogenital radiologist, evaluated each case first with standard imaging and then with 3DR/VR. After review, they completed a 5-point Likert questionnaire (1 = no benefit, 5 = maximum benefit) covering patient counseling, tumor board decisions (surgery vs. alternative therapy) and surgical planning. Pre-processing time was recorded.
Results or Findings: Mean usefulness scores increased from 3.0 ± 0.5 with standard imaging to 3.8 ± 0.3 for 3DR (p < 0.01) and 4.6 ± 0.5 for VR (p < 0.001). Consensus on surgical strategy during multidisciplinary discussions was achieved in 34/40 cases, with VR prompting management-related questions in 14 cases. Surgeons reported improved spatial understanding in 63% of cases and predicted safer resection planes. Median preparation time was 120 minutes (IQR 100–130), representing the main barrier to daily practice.
Conclusion: Immersive 3D visualization significantly improves surgical planning and decision-making in complex gynecologic oncology. Streamlining segmentation and workflow processes will be essential for clinical implementation. In future phases, intraoperative augmented reality (AR) will be tested, with early results showing stable hologram registration and promising integration into surgical practice.
Limitations: The small sample size and reliance on simulated MDT limit the generalizability of the findings
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: ID prot.7640

Author Block: T. J. Vogl, A-I. Nica, L. S. Alizadeh, I. Yel, C. Wolfram, H. Adwan; Frankfurt/DE
Purpose: To retrospectively evaluate the efficacy of transarterial chemoperfusion (TACP) and transarterial chemoembolization (TACE) in the treatment of locally recurrent, unresectable gynecologic cancer regarding local tumor response and survival outcome.
Methods or Background: Between January 2000 and October 2023, 75 female patients diagnosed with locally recurrent, unresectable gynecologic tumors (cervical cancer, ovarian cancer) underwent 213 TACP and 154 TACE procedures. Among these, 33 underwent TACP, 20 received TACE, and 22 a combination of both therapies. Local tumor response was evaluated using the RECIST criteria and survival rates were determined with the Kaplan-Meier estimator.
Results or Findings: 50 patients (66.67%) maintained stable disease until completing therapy, while 10 patients (13.33%) exhibited partial response, 2 patients (2.67%) achieved complete response and 13 (17.33%) experienced progressive disease. Thus, 62 of 75 patients (87%) were classified as therapy responders. Overall, we observed a decrease of 5.53% in tumor size and 7.71% in tumor volume compared to baseline measurements. The median overall survival for all patients was 16.05 months, the median progression-free survival 13.5 months. The patients treated with TACP had the highest therapy response (90.91%) vs 81.92% of patients treated with both therapies and 70% of patients treated with TACE.
Conclusion: TACP and TACE are minimally invasive treatment options for patients with recurrent unresectable cervical and ovarian cancer. These therapies are particularly beneficial for patients who do not tolerate, refuse or do not respond to conventional treatment like chemo- or radiotherapy. However, further studies are still necessary
Limitations: -Retrospective and single-center study design
-Varying follow-up intervals and, in many cases, short follow-up period.
-Heterogeneous cohort, encompassing numerous tumor types with varying numbers of patients.
-Heterogeneous treatment strategy with transarterial therapies for both palliative and neoadjuvant purposes.
Funding for this study: No funding
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approved by the ethics committee of the University Hospital Frankfurt

Author Block: G. Imbemba, S. Bottazzi, V. Celli, E. S. Trogu, F. L. Lia, M. Dolciami, L. D'Erme, E. Sala, B. Gui; Rome/IT
Purpose: To systematically evaluate the adherence of external pelvic MRI studies to ESUR guidelines in patients referred to the uterine MDTs at our tertiary centre, identifying the most frequent protocol deviations, assessing their impact on patient’s management.
Methods or Background: This retrospective, observational, single-centre study includes patients with cervical cancer (CC) and endometrial cancer (EC) discussed at the uterine MDTs between September 2024 and September 2025, whose external pelvic MRI scans are reviewed by a dedicated GYN radiologist.
Results or Findings: Between September 2024 and September 2025, 1820 cases were discussed and 607 required imaging revision. Of these, 277/607 were acquired at our Institution, while 330/607 were not and 110/330 (33%) were deemed non-compliant to ESUR guidelines (84 CC and 26 EC cases).
Among CC MRIs, the main causes of non-adherence were absence of axial-oblique T2-WI in 68/84, absence of axial-oblique DWI in 68/84, absence of DWI images in 14/84, slice thickness >4mm in 19/84 cases. Also, contrast media was administered in 72/84 cases (86%). 24/84 patients required repeating MRI. Among those who did not: in 38 cases the necessary information was obtained through PET/CT and in 22 cases MRI was deemed sufficient to exclude locally-advanced CC and proceed directly to surgery.
Among EC MRIs, the main non-adherence causes were: absence of axial-oblique DWI in 24/26 (92%), absence of axial-oblique T2-WI in 22/26 (85%), absence of axial-oblique post-contrast images in 25/26 (96%). Five patients required repeat MRI to assess the suitability for fertility-sparing procedures.
Conclusion: Adherence to ESUR guidelines for CC and EC MRI varies across institutions, missing recommended sequences and including non-recommended ones. This may reduce diagnostic accuracy, delay treatment initiation, and increase economic and environmental burden due to the unnecessary use of contrast agents.
Limitations: Na
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study has been registered to the Ethical committee with the ID: ID8033.

Author Block: F. Abubacker Sulaiman, R. Praveenkumar, S. Zainamb Begum, M. Nivitha, J. Lydia, A. Raashid Ibrahim; Chennai/IN
Purpose: To assess the diagnostic performance of diffusion-weighted imaging (DWI), dynamic contrast-enhanced MRI (DCE-MRI), and radiomics features in differentiating benign and malignant ovarian masses, and to correlate imaging biomarkers with histopathological findings.
Methods or Background: A prospective study was conducted on 80 patients presenting with ovarian masses who underwent multiparametric MRI prior to surgery or biopsy. MRI protocol included T1- and T2-weighted sequences, DWI (b-values 0, 800 s/mm²), and DCE-MRI with quantitative perfusion analysis. Apparent diffusion coefficient (ADC) values, enhancement curve types, and radiomics features (first-order, texture, and shape-based metrics) were extracted from tumor regions using semi-automated segmentation. Histopathological diagnosis served as the reference standard. Statistical analysis included ROC curve evaluation and logistic regression to identify the best predictive model.
Results or Findings: Malignant lesions demonstrated significantly lower mean ADC values (0.89 ± 0.17 × 10⁻³ mm²/s) and higher K^trans values on DCE compared to benign lesions (1.45 ± 0.24 × 10⁻³ mm²/s, p < 0.001). Radiomics parameters such as gray-level entropy and surface irregularity showed strong correlation with tumor grade (r = 0.71, p < 0.001). The combined model integrating ADC, DCE kinetics, and key radiomics features achieved an AUC of 0.94 for malignancy prediction, outperforming any single parameter. Imaging patterns corresponded well with histopathologic features such as cellular density, necrosis, and stromal vascularity.
Conclusion: Multiparametric MRI incorporating diffusion, perfusion, and radiomics analysis enhances non-invasive characterization of ovarian masses and correlates strongly with histopathologic aggressiveness. This integrated approach may improve preoperative risk stratification and treatment planning.
Limitations: Single-center design, modest sample size, and absence of external radiomics validation restrict generalizability. Larger multicentric datasets and standardized acquisition protocols are needed for clinical translation.
Funding for this study: No external funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Institutional ethics commitee approval obtained.

Author Block: F. Masino¹, C. Locorotondo², A. Torrente², M. Montatore¹, M. Balbino¹, V. Buonocore², M. Morgillo², G. Guglielmi¹, E. Muscogiuri²; ¹Foggia (FG)/IT, ²Brindisi/IT
Purpose: To evaluate the added diagnostic value of integrating ADC with conventional MRI features in the differentiation of uterine smooth muscle tumors.
Methods or Background: Preoperative distinction between leiomyomas, STUMP, and leiomyosarcomas is critical due to radically different therapeutic implications. While MRI represents the reference imaging technique, morphology alone is insufficient to ensure accuracy. In this retrospective single-center cohort of 27 surgically confirmed cases, both qualitative MRI features (shape, margins, necrosis, T1/T2 signal, contrast enhancement, DWI) and quantitative ADC values were systematically assessed. Logistic regression and ROC analysis tested the incremental value of ADC.
Results or Findings: Among 27 patients, 22 had benign lesions and 5 had non-benign tumors. Morphological features such as irregular margins, necrosis, and hemorrhagic T1 foci correlated significantly with malignancy (p<0.05), but showed partial overlap with atypical leiomyomas. The ADC alone yielded excellent discrimination (AUC 0.995). Importantly, when ADC was combined with morphological parameters, diagnostic performance improved from an AUC of 0.982 (morphology alone) to 1.000 (integrated model). This integration enhanced risk stratification, supporting accurate identification of leiomyosarcomas while avoiding unnecessary radical surgery in benign cases.
Conclusion: A multiparametric MRI approach integrating ADC with morphological features achieves nearly ideal accuracy in differentiating uterine smooth muscle tumors, providing radiologists with a powerful tool to optimize patient management.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: