A Randomized Trial Comparing Two Fractionation Schedules for Patients with Localized Prostate Cancer

Reviewer: John Wilson, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 20, 2003

Presenter: H. Lukka
Presenter's Affiliation: McMaster University
Type of Session: Plenary


  • Prostate cancer is the most commonly diagnosed cancer in men
  • External beam radiation therapy is a potentially curative treatment for men with localized prostate cancer
  • Most treatments involve a long course of radiation, with up to 8 weeks of treatment
  • Several studies suggest that prostate cancer may benefit from hypofractionation, or less fractions with higher doses per fraction. This belief is based on a low alpha/beta ratio measured in prostate cancer. (1.5 per Brenner and Hall 1999)
  • Several nonrandomized studies of hypofractionation for prostate cancer in the UK, Australia, and Canada have already shown comparable results compared to standard fractionation
  • This group set out to perform a randomized study in Canada to compare hypofractionation to standard fractionation

Materials and Methods

  • Men with T1 or T2 prostate cancer were randomized to a standard Long Arm (66 Gy in 33 fractions over 6.5 weeks) or a hypofractionated Short Arm (52.5 Gy in 20 fractions over 4 weeks)
  • The primary outcome was treatment failure, defined by any one of the following: 1. Biochemical failure by ASTRO criteria (3 successive rises in PSA) 2. Clinical local failure 3. Distant failure 4. Hormonal intervention 5. Prostate cancer death
  • Secondary outcomes measured included positive biopsy at 2 years, survival, and toxicity


  • 936 patients were randomized between 1994 and 1998, with 470 to the Long arm and 466 to the Short arm
  • The median follow-up was 59 months
  • The patients were similar according to stage, Gleason score, and pre-treatment PSA level
  • As of September 2002, there have been 460 failures, with 216 and 244 in the Long and Short arms respectively
  • The 5 year failure probability was 48.6 and 55.6% in the Long and Short arms, respectively, with a difference of 7.0% and a 90% confidence interval of 1.4% to 12.6%
  • 73% of patients underwent a biopsy at 2 yrs, with positive results in 53.2% of Long arm and 50.7% of Short arm patients
  • Overall survival at 5 years was 85.7% and 88.3% in the Long and Short arms, respectively
  • Acute grade 3 or 4 toxicity occurred in 8.1% of Long arm and 13.5% of Short arm patients, with rectal toxicity in 2.8 vs 4.3% and urinary toxicity in 5.1 vs 9.2% of Long and Short arm patients, respectively
  • Late toxicity that was grade 3 or higher was 3.2% in the Long arm and 2.6% in the Short arm

Author's Conclusions

  • This is the first randomized clinical trial studying the use of hypofractionation in localized prostate cancer
  • The failure rate was higher in the Short arm
  • Acute toxicity was higher in the Short arm, but late toxicity was similar between arms
  • Hypofractionation increases patient convenience, decreases cost, and has a potential radiobiologic advantage
  • This study will be useful for further studies in prostate hypofractionation using dose escalation in conjunction with conformal or intensity modulated radiotherapy

Clinical/Scientific Implications

    This is the first randomized clinical trial studying the use of hypofractionation in localized prostate cancer. The failure rate was higher in the Short arm along with higher toxicity. Although the radiation dose used in the long arm (66 Gy) was considered adequate at the time this study was designed, currently radiation is typically delivered with 3D conformal planning to doses of 74 Gy and higher. It has been shown that these higher doses result in a biochemical DFS benefit over the lower doses of radiation typically used in the past. Thus any future studies that consider comparing hypofractionated radiation should use a control arm at a higher radiation dose with a radiobiologically equivalent hypofractionated dose for clinical conclusions to be relavent. Although shortening the course of radiation may increase patient convenience and potentially decrease cost, it should not be considered standard at this time as it may be inferior to standard dose treatment.

Oncolink's ASTRO Coverage made possible by an unrestricted Educational Grant from Siemens Medical Solutions.