TY - JOUR
T1 - Personalized prediction of local control after stereotactic radiosurgery for craniopharyngioma
T2 - a multicenter machine learning survival model
AU - Reyes, Jheremy S.
AU - Hadjipanayis, Constantinos G.
AU - Bernstein, Kenneth
AU - Speckter, Herwin
AU - Gonzalez, Ivan
AU - Chytka, Tomas
AU - Liscak, Roman
AU - Bowden, Greg N.
AU - Sumi, Takuma
AU - Narita, Kentaro
AU - Kano, Hideyuki
AU - Martínez-Moreno, Nuria
AU - Martínez-Álvarez, Roberto
AU - Picozzi, Piero
AU - Franzini, Andrea
AU - Tripathi, Manjul
AU - Rai, Ashutosh
AU - Kumar, Narendra
AU - Douri, Keiss
AU - Mathieu, David
AU - Dono, Antonio
AU - Amezquita-Contreras, Christian
AU - Blanco, Angel I.
AU - Esquenazi, Yoshua
AU - Tos, Salem M.
AU - Mantziaris, Georgios
AU - Peker, Selcuk
AU - Samanci, Yavuz
AU - Duzkalir, Ali Haluk
AU - Meng, Ying
AU - Sheehan, Jason P.
AU - Kondziolka, Douglas
AU - Lunsford, L. Dade
AU - Niranjan, Ajay
N1 - Publisher Copyright:
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2026.
PY - 2026/8
Y1 - 2026/8
N2 - Background: Stereotactic radiosurgery (SRS) is used in selected patients with craniopharyngioma, yet counseling and follow-up planning often rely on population-level local control rates rather than individualized expectations over time. Objective: To develop and internally validate a multicenter survival model to predict imaging-defined time to progression after SRS for craniopharyngioma. Methods: We analyzed a multicenter IRRF registry of SRS-treated craniopharyngioma patients. Imaging progression was defined by the overall last imaging response (PD vs. non-PD), with censoring at last imaging follow-up when progression was not observed. A Random Survival Forest (RSF) model was evaluated using 5-fold out-of-fold cross-validation. Performance was assessed using the concordance index, time-dependent AUC at 12, 24, and 60 months with bootstrap 95% confidence intervals, integrated Brier score (IBS) over 0–60 months, and risk-stratified calibration. Benchmarks included a penalized Cox model and a Kaplan–Meier baseline. Results: Among 277 patients (event rate 13.0%; median imaging follow-up 57.0 months by reverse Kaplan–Meier), RSF achieved an out-of-fold C-index of 0.905. Time-dependent AUC was 0.895 (95% CI 0.828–0.959) at 12 months, 0.897 (95% CI 0.833–0.952) at 24 months, and 0.934 (95% CI 0.889–0.969) at 60 months. IBS (0–60 months) was 0.050 with favorable calibration. Conclusions: A multicenter machine learning survival model can provide individualized, well-calibrated estimates of local control over time after SRS for craniopharyngioma to support non-prescriptive decision support. Clinical trial number: Not applicable.
AB - Background: Stereotactic radiosurgery (SRS) is used in selected patients with craniopharyngioma, yet counseling and follow-up planning often rely on population-level local control rates rather than individualized expectations over time. Objective: To develop and internally validate a multicenter survival model to predict imaging-defined time to progression after SRS for craniopharyngioma. Methods: We analyzed a multicenter IRRF registry of SRS-treated craniopharyngioma patients. Imaging progression was defined by the overall last imaging response (PD vs. non-PD), with censoring at last imaging follow-up when progression was not observed. A Random Survival Forest (RSF) model was evaluated using 5-fold out-of-fold cross-validation. Performance was assessed using the concordance index, time-dependent AUC at 12, 24, and 60 months with bootstrap 95% confidence intervals, integrated Brier score (IBS) over 0–60 months, and risk-stratified calibration. Benchmarks included a penalized Cox model and a Kaplan–Meier baseline. Results: Among 277 patients (event rate 13.0%; median imaging follow-up 57.0 months by reverse Kaplan–Meier), RSF achieved an out-of-fold C-index of 0.905. Time-dependent AUC was 0.895 (95% CI 0.828–0.959) at 12 months, 0.897 (95% CI 0.833–0.952) at 24 months, and 0.934 (95% CI 0.889–0.969) at 60 months. IBS (0–60 months) was 0.050 with favorable calibration. Conclusions: A multicenter machine learning survival model can provide individualized, well-calibrated estimates of local control over time after SRS for craniopharyngioma to support non-prescriptive decision support. Clinical trial number: Not applicable.
KW - Craniopharyngioma
KW - Local control
KW - Machine learning
KW - Stereotactic radiosurgery
KW - Survival prediction
UR - https://www.scopus.com/pages/publications/105046316008
U2 - 10.1007/s11060-026-05736-8
DO - 10.1007/s11060-026-05736-8
M3 - Article
C2 - 42536204
AN - SCOPUS:105046316008
SN - 0167-594X
VL - 179
JO - Journal of Neuro-Oncology
JF - Journal of Neuro-Oncology
IS - 1
M1 - 22
ER -