2077 - Definitive Proton Beam Therapy for Oligometastatic Disease from Colorectal Cancer: A Single-Institution Retrospective Study Assessing the ESTRO-EORTC Classification

03:00pm - 04:00pm ET

Poster Hall - Exhibit Hall A

Screen: 17

POSTER

Presenter(s)

Masaaki Goto, MD, PhD - University of Tsukuba, Tsukuba City 305-8575, Ibaraki

M. Goto¹, K. Baba¹, K. Nakai¹, M. Mizumoto¹, H. Numajiri¹, T. Saito¹, T. Iizumi¹, M. Nakamura¹, T. Ishida¹, T. Fujisawa¹, T. Sumiya¹, T. Sawada¹, M. Harada¹, T. Okumura^1,2, and H. Sakurai¹; ¹Departement of Radiation Oncology and Proton Medical Research Center, Faculty of Medicine, University of Tsukuba, Tsukuba, Japan, ²Department of Radiation Oncology, Ibaraki Prefectural Central Hospital, Kasama, Japan

Purpose/Objective(s):

Metastasis-directed radiotherapy is increasingly used for oligometastatic disease (OMD), where definitive local treatment may improve prognosis. Colorectal cancer (CRC) is often relatively radioresistant, requiring high-dose irradiation for durable local control. Although proton beam therapy (PBT) enables high-dose delivery while sparing normal tissue, evidence supporting PBT for CRC-derived OMD remains limited. Furthermore, OMD is a heterogeneous state, and optimal patient selection remains a clinical challenge. We hypothesized that PBT for CRC-derived OMD achieves favorable local control and that the ESTRO-EORTC classification effectively stratifies prognosis.

Materials/Methods:

We retrospectively reviewed patients with CRC metastases treated with definitive PBT at our institution between 2016 and 2023. Fifty-two patients with 80 lesions were included, with a median follow-up of 44 months. Treated sites included the liver, lung, and others (n=56/15/9). Prescribed doses were determined based on the standardized protocol, and the median prescribed dose was 72.6 Gy in 22 fractions. The primary endpoint was overall survival (OS). Secondary endpoints were local control (LC) and disease-free survival (DFS). Predictors of OS were assessed using multivariable Cox proportional hazards models with prespecified covariates, including induced OMD (history of polymetastatic disease), oligoprogression (progressive OMD during systemic therapy), concurrent chemotherapy, and biologically effective dose 10 (BED10) > 100Gy. OS was estimated from the start of initial PBT using the Kaplan–Meier method.

Results:

According to the ESTRO-EORTC classification, repeat OMD was most common, followed by de-novo OMD and induced OMD (n=41/28/11). Twenty lesions were treated with concurrent chemotherapy. The 3-year OS, LC, and DFS were 66%, 81%, and 15%, respectively, with a median OS of 58 months. In multivariable analysis, oligoprogression was associated with worse OS (HR: 2.6, 95% CI, 1.0-6.7), while concurrent chemotherapy was associated with improved OS (HR: 0.30, 95% CI, 0.10-0.97). No grade >= 3 adverse events were observed.

Conclusion:

PBT for CRC-derived OMD achieved favorable OS and LC. The high proportion of repeat OMD suggests that PBT is a feasible option for re-treatment. Oligoprogression was a poor prognostic factor, whereas concurrent chemotherapy was associated with improved survival. Even with high local control after PBT, subsequent disease progression is common, supporting the importance optimal patient selection and an integrated strategy with systemic therapy.