Journal of Clinical Oncology (ASCO Annual Meeting LBA5012) June 01, 2026

Early results of COMPPARE, a prospective comparison of outcomes with proton and photon radiation in prostate cancer

Nancy P. Mendenhall et al.

Bottom Line

Early results from the prospective COMPPARE study demonstrate no significant differences in patient-reported bowel function, severe toxicity, or 3-year biochemical control between proton therapy and conventional photon radiation for localized prostate cancer.

Key Findings

1. Over 2 years of follow-up, there were no statistically significant differences in EPIC bowel urgency (p=0.324) or bowel frequency (p=0.514) scores between the proton therapy (PT) and intensity-modulated radiation therapy (IMRT) cohorts.

2. The 2-year cumulative incidence of grade 2 or higher gastrointestinal (GI) toxicity was similar between groups: 5.2% for PT versus 5.6% for IMRT (Hazard Ratio 0.91, 95% CI 0.65-1.28, p=0.6).

3. At 3 years, biochemical progression-free survival (bPFS) was exceptional and nearly identical in both arms: 97.9% for PT and 98.0% for IMRT (p=0.90).

4. The use of temporary rectal spacers significantly reduced 2-year grade 2+ GI toxicity in both treatment groups (4.7% PT vs. 4.4% IMRT with a spacer; compared to 8.7% PT vs. 7.2% IMRT without a spacer).

Study Design

Design

Prospective Cohort Study

Open-Label

Sample

2,524

Patients

Duration

4.0 yr

Median

Setting

51 US centers

Population Men with de novo, localized prostate cancer (excluding very high-risk disease per NCCN guidelines).

Intervention Proton therapy (PT) utilizing standard or moderate hypofractionation.

Comparator Photon radiation (IMRT) utilizing standard or moderate hypofractionation.

Outcome Patient-reported bowel urgency and bowel frequency measured by the Expanded Prostate Cancer Index Composite (EPIC) at 2 years.

Study Limitations

• The study utilizes a non-randomized, prospective observational design, which introduces inherent risks of selection bias and unmeasured confounding, despite the use of inverse probability of treatment weighting.

• Median follow-up is only 4.0 years, which is insufficient to evaluate late-onset toxicities or definitive long-term oncologic endpoints like metastasis-free survival or overall survival.

• Overall observed GI toxicity rates were much lower than originally hypothesized during trial design, reducing the statistical power to detect a clinically meaningful difference between the two modalities.

• The pragmatic design permitted variations in daily dose fractionation and the selective use of androgen deprivation therapy and rectal spacers, adding heterogeneity to the study arms.

Clinical Significance

The initial COMPPARE data establish that both proton therapy and conventional photon IMRT deliver superb early disease control and low rates of toxicity for men with localized prostate cancer. Crucially, these early results do not support the hypothesis that the dosimetric advantages of proton therapy translate into superior early clinical outcomes or better quality-of-life preservation in this population. The data also strongly validate the protective benefit of rectal spacers, regardless of the radiation modality chosen.

Historical Context

Proton beam therapy has been heavily marketed for localized prostate cancer due to its unique physical property (the Bragg peak), which allows for minimal radiation exit dose and theoretical sparing of adjacent normal organs like the rectum and bladder. However, because proton facilities are highly expensive to build and operate, the treatment carries a significantly higher financial cost than conventional IMRT. For years, the oncology community debated the value of proton therapy due to a glaring lack of Level I prospective comparative evidence. Funded by the Patient-Centered Outcomes Research Institute (PCORI), the COMPPARE study was launched across 51 US centers to definitively investigate whether proton therapy genuinely offers superior quality-of-life and toxicity outcomes compared to contemporary IMRT.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the fundamental physical difference between proton and photon radiation therapies, and why was proton therapy hypothesized to reduce bowel toxicity in prostate cancer patients?

Key Response

Protons deliver radiation with a characteristic 'Bragg peak'—depositing maximum energy at a specific depth with almost no exit dose, theoretically sparing posterior structures like the anterior rectum. Photons (X-rays) have both entrance and exit doses. Understanding this physics concept is essential for grasping why studies like COMPPARE were initiated to see if this theoretical dosimetric advantage translates into actual clinical benefit.

Resident

A 65-year-old man with intermediate-risk prostate cancer asks if he should travel out of state and pay out-of-pocket for proton therapy to avoid bowel incontinence. Based on the early results of the COMPPARE trial, how should you counsel him?

Key Response

The early COMPPARE results showed no significant difference in patient-reported bowel function or severe toxicity between proton and photon therapies at 3 years. Residents must use this prospective evidence to counsel patients that the theoretical physical benefits of protons do not currently translate into significantly better early clinical or quality-of-life outcomes, advising against significant financial toxicity or logistical burden for an unproven superiority.

Fellow

The COMPPARE study utilized patient-reported outcomes (PROs) for bowel function as a primary endpoint, but currently only reports 3-year data. In the context of modern photon techniques (IMRT/VMAT) and image guidance (e.g., hydrogel spacers), how might late toxicities (>5 years) or the adoption of extreme hypofractionation alter the long-term comparative efficacy of these two modalities?

Key Response

Fellows must recognize that 3 years is early for late radiation-induced toxicities in prostate cancer. Furthermore, modern advancements in photon therapy (perirectal spacers, precise IMRT, SBRT) have already drastically reduced rectal toxicity, narrowing the potential delta for proton therapy to show superiority. Evaluating this trial requires anticipating how longer follow-up and evolving standard-of-care photon techniques might impact final comparative effectiveness.

Attending

Given the equivalent 3-year toxicity and biochemical control outcomes between protons and photons, how should radiation oncology departments and health systems approach capital investments and value-based care pathways for localized prostate cancer?

Key Response

Attendings must weigh clinical outcomes against the immense capital cost of proton centers. Since COMPPARE demonstrates non-superiority in early clinical and PRO endpoints, the focus shifts to cost-effectiveness and equitable care. This informs institutional strategic planning, suggesting that resource allocation might be better directed toward improving access to high-quality photon IMRT/SBRT rather than building new proton facilities specifically to treat prostate cancer.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

COMPPARE is a prospective observational cohort study rather than a randomized controlled trial (RCT), largely due to strong patient preference for proton therapy. What methodological techniques are critical in this design to adjust for selection bias, and what unmeasured confounders might still threaten the internal validity of the null finding?

Key Response

Because patients self-selected their treatment, proton patients often have higher socioeconomic status, different baseline health literacy, and potentially different baseline PROs. PhD-level critique focuses on the use of propensity score matching or inverse probability weighting, and the limitation that these techniques only account for measured confounders (like age, PSA, Gleason score), leaving unmeasured variables (like diet, minor baseline functional differences, or spatial access to care) as potential threats to validity.

Journal Editor

As a reviewer evaluating the early COMPPARE results, how would you scrutinize the study's statistical power to detect a clinically meaningful difference in the EPIC bowel domain, and what concerns would you raise regarding informative missingness in patient-reported outcomes over the 3-year follow-up?

Key Response

An editor must question whether 'no significant difference' is due to true equivalence or a Type II error (underpowered study). If the non-inferiority/equivalence margin in EPIC scores was set too wide, or if missing data was not handled appropriately (e.g., informative missingness where patients with worse toxicity drop out of the survey), the conclusion of clinical equivalence could be fundamentally flawed.

Guideline Committee

Current ASTRO and NCCN guidelines state that proton therapy is an acceptable alternative to photon-based therapies but do not recommend it as superior for localized prostate cancer. Do the early COMPPARE results provide sufficient evidence to maintain or modify this stance, and how should this data influence formal insurance coverage guidelines?

Key Response

The committee must evaluate whether new evidence reinforces or challenges current guidelines. Since COMPPARE (a large, multi-center prospective trial providing strong Level 2 evidence) confirms the lack of superiority of protons for toxicity and efficacy at 3 years, it strongly solidifies the current ASTRO/NCCN stance. It provides robust justification against routine insurance mandates to cover the higher cost of protons for localized prostate cancer, emphasizing that it remains an option but lacks proven clinical superiority.

Clinical Landscape

Noteworthy Related Trials

2005

PROG 95-09

n = 393 · JAMA

Tested

High-dose radiation therapy (79.2 GyE) using a proton boost

Population

Men with clinically localized prostate cancer

Comparator

Conventional-dose radiation therapy (70.2 GyE)

Endpoint

Biochemical failure

Key result: High-dose radiation therapy significantly reduced the risk of biochemical failure compared to conventional doses without increasing high-grade toxicity.

2016

CHHiP Trial

n = 3,216 · Lancet Oncol

Tested

Hypofractionated radiotherapy (60 Gy in 20 fractions)

Population

Men with localized prostate cancer

Comparator

Conventionally fractionated radiotherapy (74 Gy in 37 fractions)

Endpoint

Biochemical or clinical failure

Key result: Hypofractionated radiotherapy was non-inferior to conventional fractionation in terms of efficacy, with similar long-term side effects.

2018

RTOG 0126

n = 1,532 · JAMA Oncol

Tested

High-dose 3D-CRT or IMRT (79.2 Gy)

Population

Men with intermediate-risk prostate cancer

Comparator

Standard-dose 3D-CRT or IMRT (70.2 Gy)

Endpoint

Overall survival

Key result: Dose escalation reduced biochemical and clinical failure but did not improve overall survival and was associated with increased late genitourinary and gastrointestinal toxicity.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis