Patterns of Failure and Toxicity Following Proton Beam Therapy for Pediatric Bladder and Prostate Rhabdomyosarcoma

Abstract

In the Children’s Oncology Group (COG), bladder and prostate are considered unfavorable sites for rhabdomyosarcoma (RMS). We report our institutional experience treating patients with bladder or prostate RMS (B/P-RMS) with proton beam therapy (PBT). We hypothesize that disease-related outcomes using PBT are comparable to previously-reported photon RT series, while the toxicity profile with PBT compares favorably to prior photon studies. Lastly, we provide details regarding simulation and daily image guidance for these young patients with pelvic lesions. Nineteen patients with newly-diagnosed B/P-RMS were enrolled on a prospective registry protocol between 2008 and 2017. All were treated with multimodality therapy, including chemotherapy and PBT; some were treated with surgical resection (n=8, 42%). All patients required sedation for both CT simulation and daily PBT treatments, owing to young age. All patients underwent daily kV image guidance, and 8 patients, all with bladder primary tumors, underwent daily bladder ultrasonography to assess bladder volume. Disease-related outcomes including overall survival (OS), progression-free survival (PFS) were calculated from date of diagnosis. Local control (LC) was calculated from date of PBT completion. The median age at diagnosis was 1.8 years (range 0.5-4.9). The majority of patients had bladder RMS (74%), Group III disease (68%), and intermediate-risk disease by COG risk stratification (89%). Median primary tumor size was 3.2cm (range 1.0-13.0cm), and 7 patients (37%) had primary tumors >5cm in size. Chemotherapy was VAC (ARST0331/0431/0531) in 68.4%, VI/VAC (ARST0431/0531) in 26.3%, and VA (ARST0331) in 5.3% of patients. Median RT dose was 50.4Gy(RBE) (range 36.0-50.5Gy[RBE]). With a median follow-up of 66.2 months, 5-year OS for the full cohort was 81%, and 5-year PFS was 82%. Three patients (16%) experienced disease relapse, all presenting with local failure. The 5-year LC rate was 82%. Tumor size predicted LC, with 5-year LC for patients with >5cm tumors being 54% versus 100% for those with ≤5cm tumors (p=0.02). Univariate analysis demonstrated an effect of tumor size on OS (tumor >5cm, HR 20.8, p=0.01) and PFS (HR 19.0, p=0.02). Acute grade 2+ toxicity was observed in 2 patients (11%), both with grade 2 proctitis that resolved within 2 weeks of PBT completion. Late grade 2+ toxicity was observed in 4 patients (21%), one with a grade 2 skeletal deformity, and 3 with transient grade 2 urinary incontinence. No grade 3+ toxicities were observed. There were no cases of urethral stricture, chronic cystitis, or secondary malignancy. PBT for B/P-RMS affords promising disease-related outcomes with a favorable toxicity profile. Higher local failure rates were observed for larger tumors, supporting the dose-escalation component of the ongoing COG ARST1431 protocol.