Local failure remains frequent after metastasis-directed RT in patients also receiving Lu-PSMA-617

Among 202 irradiated prostate cancer metastases, 54% developed local failure, while severe late radiation-attributed toxicity occurred after 2% of lesions.

KEY POINTS

  • This retrospective single-center study from Dana-Farber Cancer Institute and Brigham and Women's Hospital included 88 men with metastatic prostate cancer who received both ablative-intent external-beam radiotherapy and Lu-PSMA-617 during their disease course. A total of 202 metastatic lesions were irradiated, with median lesion-level imaging follow-up of 30.5 months. At the time of EBRT, 78.2% of lesions were treated in patients with castration-resistant disease, indicating a substantially more advanced population than typical oligorecurrent SBRT trials.
  • The treatment cohort was heterogeneous. SBRT accounted for 70.3% of irradiated lesions, while SRS/SRT, IMRT, and 3D conformal RT comprised the remainder. Five-fraction regimens were most common (46.1%), followed by single-fraction (25.3%) and three-fraction (14.4%) treatments; median BED10 was 59.5 Gy. Most metastases were osseous (82.2%), and approximately 59.9% of EBRT courses were accompanied by concurrent systemic prostate cancer treatment.
  • EBRT and Lu-PSMA-617 were usually delivered as separate components of longitudinal treatment, rather than as a planned combination. Lu-PSMA-617 preceded EBRT for only 35/202 lesions (17.3%); in the remaining cases, intervals between irradiation and subsequent radioligand therapy ranged from months to more than four years. Patients received a median four cycles of Lu-PSMA-617, and 39.8% underwent multiple distinct courses of metastasis-directed EBRT.
  • Local failure occurred in 109/202 treated lesions (54.0%), with median time to local failure of 21.1 months (95% CI 17.3–40.5). Among the 202 lesions, 194 (96%) had evaluable post-treatment imaging, and eight without evaluable assessment were censored at time zero. Median overall survival was 20.1 months, while median time to the next systemic treatment after EBRT was 10.2 months. The observed local failure rate was substantially higher than expected in selected hormone-sensitive oligometastatic SBRT cohorts.
  • Castration-resistant disease and nonosseous metastatic location were independently associated with shorter local control. Adjusted hazard ratios were 1.82 (95% CI 1.08–3.07; p=0.02) for castration-resistant versus hormone-sensitive disease and 1.84 (95% CI 1.02–3.31; p=0.04) for nonosseous versus osseous lesions. No statistically significant association with local failure was detected for EBRT sequencing relative to Lu-PSMA-617 (HR 0.62; p=0.27) or BED10 ≥60 versus lower doses (HR 0.99; p=0.97).
  • Grade ≥3 late toxicity attributed to RT occurred following four lesions (2.0%), including three vertebral compression fractures requiring vertebroplasty and one severe bladder injury requiring urinary diversion. Vertebral compression fractures of any grade developed after 13/76 irradiated vertebral lesions (17.1%), with 9/13 patients (69.2%) having documented osteopenia or osteoporosis. All reported vertebral fractures occurred following EBRT delivered before Lu-PSMA-617, often separated from radioligand treatment by several years.
  • Selection and immortal-time bias are major limitations. Patients entered the registry because they eventually received Lu-PSMA-617, so those irradiated years earlier had to survive and remain eligible for subsequent radioligand therapy. Radiotherapy dose, fractionation, disease state, imaging surveillance, and systemic treatment were not standardized, while toxicity was collected retrospectively. These data therefore cannot establish the efficacy of a specific SBRT/Lu-PSMA-617 combination, optimal treatment sequencing, or comparative safety.

CLINICAL TAKEAWAY

Metastasis-directed EBRT was associated with relatively uncommon documented severe late toxicity in patients who also received Lu-PSMA-617, but local failure was frequent in this heavily pretreated population. These findings inform counseling and future trial design, while the heterogeneous treatment timelines and strong selection bias prevent conclusions about synergy, optimal sequencing, or whether radioligand therapy alters radiotherapy outcomes.

SOURCE

Clinical and Translational Radiation Oncology

Browse more research Suggest a correction