Background
Phase II RCT comparing proton beam therapy (PBT) vs intensity-modulated photon radiotherapy (IMRT) for neoadjuvant CRT followed by surgery in resectable esophageal cancer. 145 patients randomized at MD Anderson Cancer Center. All patients received concurrent carboplatin/paclitaxel (CROSS regimen) with 50.4 Gy concurrently, followed by esophagectomy. The trial was designed to assess whether proton therapy reduces postoperative complications and total toxicity score compared to IMRT.
Interventions and follow up
Arm A: Proton beam therapy (passive scatter or pencil beam scanning) + carboplatin/paclitaxel → esophagectomy
Arm B: IMRT + carboplatin/paclitaxel → esophagectomy
Primary endpoint: Total toxicity burden (TTB — composite of 20 organ-specific toxicities, 90-day postoperative)
mFollow up: 44.1 months (median)
Arm B: IMRT + carboplatin/paclitaxel → esophagectomy
Primary endpoint: Total toxicity burden (TTB — composite of 20 organ-specific toxicities, 90-day postoperative)
mFollow up: 44.1 months (median)
Results
Total toxicity burden (TTB): 2.05 (proton) vs 5.37 (IMRT) — significantly lower with proton (P=.033)
Grade ≥3 postoperative complications: 26.5% (proton) vs 39.8% (IMRT), P=.078 — trending
pCR rate: 28.3% (proton) vs 34.2% (IMRT), P=.40 — not significantly different
3-year OS: 45.6% (proton) vs 44.6% (IMRT), P=.99 — not significantly different
3-year PFS: Similar between arms
Grade ≥3 postoperative complications: 26.5% (proton) vs 39.8% (IMRT), P=.078 — trending
pCR rate: 28.3% (proton) vs 34.2% (IMRT), P=.40 — not significantly different
3-year OS: 45.6% (proton) vs 44.6% (IMRT), P=.99 — not significantly different
3-year PFS: Similar between arms
Adverse events
Main adverse events: Proton arm: lower total toxicity burden and trending lower postoperative complications. IMRT arm: higher acute respiratory and cardiac toxicities. Esophageal toxicity rates similar. Grade ≥3 radiation-related toxicities: lower with proton but statistical significance varied by endpoint.
Conclusions
Proton beam therapy significantly reduced the total toxicity burden during and after neoadjuvant CRT for esophageal cancer compared to IMRT, without compromising pCR rates or survival outcomes. Proton therapy may have a clinically meaningful advantage in reducing morbidity in the perioperative period, supporting its use when available.
Key Limitations
Key Limitations: Single-institution Phase II trial — not powered for OS or pCR endpoints. TTB is a composite endpoint weighting multiple toxicities; its clinical interpretability is limited. Significant equipment and expertise variation exists between proton centers. Cost is substantially higher for proton vs photon therapy; cost-effectiveness has not been established. Pencil-beam scanning and passive-scatter proton techniques were both used, introducing heterogeneity in the proton arm. Generalizability to centers without high-volume proton programs is limited.
Clinical Context
This is the first prospective RCT comparing proton to photon therapy for esophageal cancer. The TTB advantage supports proton therapy in patients at high risk for treatment-related morbidity, particularly those with poor pulmonary reserve or cardiac comorbidities. NCCN considers proton therapy an appropriate alternative for esophageal cancer where dosimetric advantages can reduce organ-at-risk exposure. However, access limitations and cost restrict widespread adoption.
References