RADCOMP Trial — Proton vs Photon Therapy in Breast Cancer

RADCOMP is the largest randomised comparison of proton versus photon radiotherapy in breast cancer. It asked a deliberately long-term question: when the internal mammary nodes are treated and some dose to the heart is unavoidable, does the lower cardiac dose achieved with protons translate into fewer major cardiovascular events over 10 years? The early quality-of-life and dosimetry answers are now published — the primary cardiac endpoint is not.

RADCOMP at a glance

1,239
patients randomised (624 proton · 615 photon)
45.0–50.4 Gy
breast/chest wall + comprehensive nodal RT, including internal mammary nodes, in both arms
0.89 vs 2.99 Gy
mean heart dose, proton vs photon (published dosimetry analysis)
10 years
time horizon of the primary endpoint: major cardiovascular events

The question

  • Does proton therapy reduce major cardiovascular events (MCEs) compared with photon therapy in patients whose radiotherapy must include the internal mammary nodes?
  • Pragmatic design: treatment delivered to local standards (photons = 3D conformal or IMRT; protons = passive scatter or pencil-beam scanning), with blinded central adjudication of cardiac events.

Design

  • Multicentre, randomised phase III superiority trial (open label), run by the Radiotherapy Comparative Effectiveness (RADCOMP) Consortium as RTOG Foundation study 3510.
  • Sponsor: Abramson Cancer Center, University of Pennsylvania. Funded by the Patient-Centered Outcomes Research Institute (PCORI).
  • Dose in both arms: 45.0–50.4 Gy in 1.8–2.0 Gy fractions to breast/chest wall and comprehensive nodal volumes including the internal mammary nodes, ± tumour bed boost.
  • Stratification: age (<65 vs ≥65 years), cardiovascular risk factors (0–2 vs >2), surgery (mastectomy vs lumpectomy), laterality (left vs right; bilateral counted as left).

Who was enrolled

  • Adults ≥21 years with non-metastatic invasive breast cancer (AJCC 7th edition stage I–III, or locoregional recurrence; yp0–III after neoadjuvant chemotherapy) proceeding to breast/chest wall and nodal irradiation including internal mammary node treatment.
  • Bilateral disease permitted. Key exclusions: metastatic disease and prior ipsilateral breast/chest wall radiotherapy.
  • Accrual: February 2016 to 28 March 2024 (40 US sites). Target sample size was revised during the trial from 1,716 to 1,278, then 1,238; 1,239 patients were randomised (61.8% left-sided).

Endpoints

  • Primary: major cardiovascular events at 10 years — cardiovascular death (atherosclerotic coronary heart disease or other heart disease), myocardial infarction, coronary revascularisation, or hospitalisation for a major cardiovascular event (heart failure, valvular disease, arrhythmia, unstable angina or pericardial disease).
  • Secondary: local-regional recurrence and any recurrence (non-inferiority); health-related quality of life (body image, fatigue, anxiety, social roles, financial burden, satisfaction); dose–response modelling of cardiac dose against events and quality of life; longer-term breast cancer-specific and overall survival and second malignancies (15 years).

Reported so far

  • Quality of life (ASTRO 2025, late-breaking abstract): no statistically or clinically meaningful differences between arms in PROMIS fatigue, BREAST-Q, FACT-B Trial Outcome Index or satisfaction with breast cosmesis at end of treatment, 1 and 6 months. A shortness-of-breath signal favouring photons on unadjusted analysis was not significant after correction for multiple comparisons.
  • Cardiac dosimetry (published in full, 2026): among 1,160 patients with complete cardiac substructure data, protons gave significantly lower mean doses to the whole heart (0.89 vs 2.99 Gy), left ventricle, left atrium and left anterior descending artery (all p<0.001). No dose–volume metric was associated with cardiac-specific quality of life over the first 6 months.
  • Grade ≥3 toxicity was low in both arms (≤5.5%).

Still pending

  • The primary endpoint (10-year major cardiovascular events) has not been reported. The analysis is event-driven (70 events required); the trial is in follow-up, ‘pending primary outcome analysis’.
  • Local-regional control and survival outcomes have not been reported.
  • A 2026 review estimates primary-endpoint reporting between 2028 and 2032; the registry lists estimated study completion in November 2036, reflecting the 15-year follow-up.

Bottom line for practice

  • RADCOMP confirms the dosimetric premise — protons roughly quarter mean heart dose when the internal mammary nodes are treated — and shows no early quality-of-life penalty or benefit for either technique.
  • Whether the dose reduction prevents cardiac events a decade later remains unanswered. Until the primary endpoint reports, proton referral for nodal breast irradiation continues to rest on dosimetry, patient risk profile and judgement, not on randomised outcome evidence.

Key sources

  • ClinicalTrials.gov NCT02603341 (sponsor Abramson Cancer Center at Penn Medicine; collaborator PCORI; status active, not recruiting, verified November 2025).
  • Bekelman JE et al. Pragmatic randomised clinical trial of proton versus photon therapy for patients with non-metastatic breast cancer: the RadComp Consortium trial protocol. BMJ Open 2019;9:e025556.
  • MacDonald SM et al. Phase III randomized trial of proton vs photon therapy for patients with non-metastatic breast cancer receiving comprehensive nodal radiation: a RADCOMP Consortium trial — health-related quality of life outcomes. Int J Radiat Oncol Biol Phys 2025;123(4):1195 (ASTRO 2025, LBA01).
  • Wright JL et al. Associations between cardiac radiation dose-volume metrics and health-related quality of life for patients with breast cancer on the RadComp trial. Int J Radiat Oncol Biol Phys 2026;126(4):1158–1171.
  • RTOG Foundation study 3510, Data Monitoring Committee open-session report, November 2025.