WHAT THE STUDY ACTUALLY SAYS

Five-day prostate radiotherapy was not superior to the longer course in NRG-GU005

Five-fraction SBRT improved several quality-of-life measures but was not superior to longer IMRT for prostate cancer. Three-year disease-free survival was 88.6% with SBRT and 92.1% with IMRT.

Disease-free survival at 3 yearsMH-IMRT (longer course): 92.1%; SBRT (5 fractions): 88.6%0%50%100%MH-IMRT (longer course)92.1%SBRT (5 fractions)88.6%
Disease-free survival at 3 years
GroupValue (%)
MH-IMRT (longer course)92.1 (88.9 to 95.2)
SBRT (5 fractions)88.6 (85.2 to 92.1)
Disease-free survival at 3 years Intermediate-risk localized prostate cancer; 698 patients randomised 1:1. One-sided log-rank P < .001 in the non-superiority direction. Source: JAMA

Radiotherapy for prostate cancer has been getting shorter. Where a course once took eight or nine weeks, newer schedules pack the dose into fewer, larger treatments. The most compressed version, stereotactic body radiotherapy (SBRT), can be delivered in five sessions [s1]. The appeal is obvious — fewer trips to the clinic, less time, lower cost. NRG-GU005 asked the harder question: is the fast version actually better?

The trial, published in JAMA, compared SBRT against a longer, moderately hypofractionated course and tested whether the shorter schedule was superior — not merely as good, but better — on both side effects and cancer control [s1].

What the trial did

NRG-GU005 was a phase-3, international, open-label, randomised clinical trial [s1]. It was activated on November 16, 2017, closed to accrual on June 8, 2022, and had its last follow-up on October 13, 2024, across 136 centres in Asia, Canada, Europe and the United States [s1].

Eligible patients had localized, intermediate-risk prostate cancer [s1]. They were randomised 1:1 to SBRT — 36.25 Gy in 5 fractions (n = 353) — or to moderately hypofractionated intensity-modulated radiation therapy, MH-IMRT, given as 70 Gy in 28 fractions or 60 Gy in 20 fractions (n = 345) [s1]. A total of 698 patients were randomised [s1]. Their median age was 68 years, and the population was 3% Asian, 13% Black and 80% White [s1]. The intended accrual of 692 patients was set to give greater than 80% power to detect a hazard ratio of 0.62 in disease-free survival and to catch reductions of 8% and 10% in the frequency of a clinically important decline in the urinary-irritative/obstructive and bowel domains [s1]. The trial is registered as NCT03367702 [s2].

There were two primary outcomes, and both were framed as superiority tests: the frequency of a minimal clinically important decline (MCID) in the urinary irritative/obstructive and bowel domains of the EPIC-26 patient questionnaire at 2 years, and disease-free survival (DFS) at 3 years [s1]. The design was powered to detect a hazard ratio of 0.62 in DFS [s1].

What happened

On the primary cancer-control endpoint, SBRT was not superior. At 3 years, DFS was 92.1% with MH-IMRT (95% CI 88.9% to 95.2%) versus 88.6% with SBRT (95% CI 85.2% to 92.1%), with a one-sided log-rank P < .001 — a result pointing away from SBRT superiority, not toward it [s1]. The rate of PSA failure was not significantly improved with SBRT either [s1].

On the primary quality-of-life endpoint, the picture was mixed. At 2 years there was no significant difference in the urinary-irritative domain (35.4% versus 33.7%, P = .68) [s1]. But there were significantly fewer clinically important declines with SBRT in the bowel domain (34.9% versus 43.8%, P = .03) [s1].

Other measures leaned toward the shorter course. Urinary incontinence at 1 and 2 years, and sexual function at 1 year, favoured SBRT [s1]. Fewer grade 3 and 4 genitourinary adverse events occurred with SBRT than with MH-IMRT (0.6% versus 2.5%, P = .04) [s1].

How to read a "not superior" trial

This is the kind of result that is easy to misreport in either direction. SBRT did not prove better for keeping the cancer at bay — on the numbers, DFS was a few points lower, and the formal superiority test went the other way [s1]. But it did ease several side-effect and quality-of-life measures [s1].

What the trial does not establish is that SBRT is worse for cancer control. NRG-GU005 was designed as a superiority trial, not a non-inferiority trial [s1]. A superiority test that fails cannot, by itself, certify that the two treatments are equivalent — that requires a differently powered design. The honest summary is the authors' own: SBRT "improved multiple quality-of-life domains but was not superior to MH-IMRT in terms of DFS" [s1].

Who funded it

This was a publicly funded cooperative-group trial run by NRG Oncology with support from the US National Cancer Institute, not an industry study of a proprietary product [s1] [s2]. That independence is part of why the result is worth taking at face value: there was no commercial incentive to favour either schedule.

What to watch

The median follow-up was 3.2 years [s1]. Prostate cancer recurs slowly, so the gap between the two DFS curves could widen, narrow or hold as longer data mature [s1]. For now, the trial reframes the pitch for five-fraction radiotherapy: its case rests on convenience and a better side-effect profile in some domains, not on superior cancer control [s1].

This article describes trial results. It is not medical advice, and nothing here should be used to choose or change a cancer treatment.

Sources

Sources

  1. Stereotactic Body Radiotherapy vs Moderately Hypofractionated IMRT for Localized Intermediate-Risk Prostate Cancer: A Randomized Clinical Trial — JAMA , August 13, 2026
  2. Phase III IGRT and SBRT vs IGRT and Hypofractionated IMRT for Localized Prostate Cancer (NRG-GU005, NCT03367702) — ClinicalTrials.gov , December 11, 2017

More on

Related coverage