S1375
Interdisciplinary - Global health
ESTRO 2026
in resource-limited settings. In Sub-Saharan Africa (SSA), radiotherapy delivery is constrained by workforce shortages, limited equipment, overburdened infrastructure, and bottlenecks from prolonged manual planning processes. Whole-brain radiotherapy (WBRT) remains the mainstay of palliative treatment. Artificial intelligence (AI)–assisted automated systems offer an opportunity to enhance plan quality, streamline workflow, and improve access to timely radiotherapy treatment. This study evaluated the dosimetric and operational performance of AI- assisted WBRT planning in a high-volume SSA oncology center. Material/Methods: Fifty consecutively treated WBRT cases at the Medserve–LUTH Cancer Centre (MLCC), Lagos, were retrospectively replanned using the RADformation EZFluence automated planning platform. Automated plans were compared with original 3D-conformal (3D- CRT) plans using dosimetric parameters including D2%, D95%, Dmean, conformity index (CI), homogeneity index (HI), V95%, and organ-at-risk (OAR) doses. Monitor units (MUs), planning time per case, and workflow efficiency were also evaluated. Correlations between target coverage and OAR doses were analyzed using Pearson’s correlation. Results: Automated plans achieved mean D95% = 28.5 Gy ± 4.7, D2% = 30.2 Gy ± 4.9, and V95% = 97.9%, confirming optimal target coverage. Plan quality was consistently high with mean HI = 0.06 ± 0.01 and CI = 1.00 ± 0.01, indicating excellent uniformity and conformity. OAR doses consistently remained within tolerance limits (lens < 10 Gy; optic nerve Dmax < 30 Gy). A weak-to-moderate correlation was observed between target coverage (D95%) and optic nerve Dmax (r = 0.42, p < 0.05), with no significant association for lens dose; suggesting minimal compromise in OAR sparing despite strong PTV coverage. Average planning time was 468.4 seconds per plan, demonstrating high efficiency in a high- volume SSA oncology setting. AI-generated plans required higher MUs (mean = 554.4 vs. 133.9), implying marginally longer beam-on times. Conclusion: AI-assisted planning can significantly enhance workflow efficiency, plan uniformity, while maintaining OAR protection, a valuable combination in resource- limited environments. This experience from Sub- Saharan Africa demonstrates the feasibility and clinical value of automation for improving radiotherapy workflow efficiency, standardizing quality, and bridging technical and workforce gaps. By reducing planning bottlenecks and standardizing outputs, this approach enhances access to timely and effective radiotherapy, and supports more equitable oncology service delivery across sub-Saharan Africa. Broader
Conclusion: This first national experience demonstrates that single-fraction 20 Gy SRS provides effective and well- tolerated local control for brain metastases, achieving outcomes comparable to international series despite resource constraints. These findings support the feasibility of implementing advanced radiotherapy techniques in developing oncology programs and provide a foundation for future research, policy development, and capacity building in underserved regions. References: 1- Gruber I, Weidner K, Treutwein M, et al. Stereotactic radiosurgery of brain metastases: a retrospective study. Radiat Oncol. 2023;18:202. https://doi.org/10.1186/s13014-023- 02389-z2- Almeida ND, Kuo C, Schrand TV, et al. Stereotactic radiosurgery for intracranial breast metastases: a systematic review and meta-analysis. Cancers. 2024;16(20):3551. 3- Al - Wassia R, Iskanderani O. Stereotactic radiosurgery (SRS) experience on brain metastases: a 3 - year retrospective study at King Abdulaziz University Hospital. Saudi J Biol Sci. 2021;28(9):5042 - 5047. https://doi.org/10.1016/j.sjbs.2021.05.017 Keywords: SRS, Brain Metastases, Resource-Limited Setting Digital Poster 5111 Improving Radiotherapy Workflow Efficiency and Access in Sub-Saharan Africa: Insights from Automated Planning in a Nigerian Oncology Centre Nusirat Adedewe 1 , Inioluwa Ariyo 1 , Abdallah Kotkat 1 , Moses Ibekwe 1 , Gbenga Eruola 1 , Adedayo Joseph 1,2 , Francis A. Durosinmi-Etti 1 1 Medserve-LUTH Cancer Centre, Lagos University Teaching Hospital, Lagos, Nigeria. 2 College of Medicine, University of Lagos, Lagos, Nigeria Purpose/Objective: Brain metastases are the most common intracranial malignancy in adults, affecting up to 40% of patients with cancer, and present a major palliative challenge
Made with FlippingBook - Share PDF online