Comparison of Health-related Quality of Life 5 Years after Brachytherapy (BT) or Radical Prostatectomy (RP): The SPIRIT (Surgical Prostatectomy vs. Interstitial Radiotherapy Intervention Trial) Cohort (ACOSOG Z0070)

Reviewer: Geoffrey Geiger, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 2, 2010

Authors: J. M. Crook, A. Gomez-Iturriaga, K. Wallace, S. Fung, S. Alibhai, M. Jewett, N. Fleshner
Institutions: University of British Columbia, Kelowna, BC, Canada; Cruces Hospital, Barakaldo, Spain; Princess Margaret Hospital, Toronto, ON, Canada; University Health Network, Toronto, ON, Canada


  • Worldwide, prostate cancer is the second most commonly diagnosed cancer in men after lung cancer, and in the United States, approximately one quarter million men are diagnosed with prostate cancer each year, nearly 2/3 of whom have early-stage disease.
  • Myriad treatment options exist for the treatment of this patient population, and the two most utilized definitive treatments are radiation therapy and radical prostatectomy.
    • Radiation can be delivered via an external beam technique, typically given via an intensity-modulated plan (IMRT) or through brachytherapy, where radioactive seeds are placed directly into the prostate.
  • No randomized trials exist directly comparing the efficacy of radical prostatectomy (RP) against either brachytherapy (BT) or external beam radiation (EBRT), although available data reveals clinical equipoise between radical prostatectomy and brachytherapy in many clinical settings.
  • One attempt at such a trial was the American College of Surgeons Oncology Group phase III trial (SPIRIT, ACOSOG Z0070) comparing RP and BT for low-risk prostate cancer, which opened in 2002 but closed in 2004 due to poor accrual.
  • The authors in this study report health related quality of life (HRQOL) at a mean of 5.3 years for 168 trial-eligible men who either chose or were randomized to RP or BT following a multi-disciplinary educational session, which explained the trial and their choices.

Materials and Methods

  • ACOSOG Z0070 was a phase III trial designed to randomize men aged 75 or younger to radical prostatectomy versus brachytherapy for early-stage, low-risk disease, which opened in 2002 across 31 centers in the United States and Canada.
  • Targeted accrual was 1980 patients.
  • Men with histologically confirmed prostate adenocarcinoma diagnosed within the previous 120 days were eligible for this study if they had an ECOG performance status of 0-2, T1c-T2aN0M0 disease with a PSA <10 ng/mL, a Gleason score of 6 or less, and a gland volume of 60 cc or less. Patients were allowed to enroll if they had undergone prior neoadjuvant hormonal therapy (NHT), provided that their pre-hormone PSA was less than 10 ng/mL.
  • After an initial lack of success with accrual for this trial, the authors began a multi-disciplinary educational session for eligible patients during which a urologist and a radiation oncologist together compared and contrasted the 2 treatment options and explained the trial rationale.
  • Forty-seven sessions were attended by 263 men, of whom 34 consented to randomization; of the remaining patients examined in this study, 62 chose RP, and 94 BT.
  • A single operator performed all brachytherapy whereas 30% of RPs were performed in the community setting.
  • 5 years later, after university and host hospital IRB approval, these 190 men underwent HRQOL evaluation using the cancer-specific 50-item EPIC, the SF-12 PCS (physical component score) and MCS (mental component score).
  • The Response rate was 88.4% (6 were lost and 16 declined).
  • Descriptive statistics were used to summarize demographics and the Wilcoxon Rank Sum Test to compare summary scores between the two interventions.


  • Unfortunately, SPIRIT closed in 2004 after enrolling only 56 patients in 2 years.
  • Of the 168 men who were either randomized or chose a treatment methodology after an educational session, 60.1% had BT (8.9% on trial) and 39.9% had RP (8.9% on trial).
  • Median age was 61.5 years (range: 45- 73) for BT and 59.5 (range 47-71) for RP.
  • Median follow-up was 5.2 years (range 3.2-6.5).
  • For BT vs. RP, there was no difference in bowel (93.0 vs. 94.3; p = 0.43), or hormonal (93.5 vs. 90.0; p = 0.09) domains, or in the SF-12 PCS score (55.8 vs. 55.5; p = 0.37).
  • However, men treated with BT scored better in the urinary (91.9 vs. 88.2; p = 0.02) and genderual (52.3 vs. 39.7; p = 0.002) domains, and in the satisfaction score (93.6 vs. 77.2; p <0.0001).
    • Examination of questions within the urinary domain revealed that urinary control, frequency of leakage and the degree of problem with leaking all showed a p value <0.0001 in favor of BT whereas none of the questions on irritative/obstructive symptoms favored either modality.
    • In the genderual domain, many questions showed a highly significant difference in favor of BT, including those on the ability to have an erection (p = 0.0007), quality of erections (p = 0.0002), and frequency of erections (p = 0.003). Erections firm enough for genderual activity were reported by 79% in the BT cohort compared to 48% for RP (p = 0.0002).
  • Median current PSA in these patients at the time of this analysis was <0.05 for the RP group (mean 0.03, SD 0.19) and 0.05 for the BT group (mean 0.15, SD 0.36).

Author's Conclusions

  • Side effects appear to stabilize by 3-5 years after treatment.
  • Although treatment allocation was only random in 17% of cases, all subjects received identical and unbiased information in a multi-disciplinary setting prior to selecting RP, BT or randomization.
  • HRQOL evaluated 3.2 - 6.3 years after treatment shows an advantage in the urinary and genderual domains, and in patient satisfaction for men treated with BT.

Clinical Implications

  • Although no large, randomized trials have directly compared outcomes between the two of the leading treatment options for patients with low-risk prostate cancer, historical comparisons have found no difference in overall survival between patients undergoing radical prostatectomy and brachytherapy.
    • Therefore, treatment decisions are often based on side effects and quality of life, which is a very subjective and personal experience. This study used a questionnaire to look at long term health related quality of life (QOL).
  • At a median follow-up of 5.2 years, the authors report that while there was no difference in bowel or hormonal domains, men treated with BT reported statistically significant better urinary outcomes and erectile outcomes compared to those treated with RP. Furthermore, these patients were more satisfied with the treatment received compared to RP patients, which was also statistically significant.
  • Unfortunately, the differences between patients receiving RP and BT were confounded by the low number of patients that were ultimately randomized.
    • Overall, only 34 patients agreed to randomization whereas 62 chose RP, and 94 BT, which potentially introduced significant selection bias.
    • The method of accrual in this trial makes direct comparisons between arms very difficult. The authors counter this point by stating that all patients received identical unbiased information in a multidisciplinary setting and all patients were equally appropriate for either modality with respect to body habitus, prostate size, voiding function, and medical comorbidities. They also reported that if either specialist felt that a patient was more suited to a particular modality, the patient was informed and they were not included in the analysis.
  • In this study, there were a higher number of non-responders to RP as compared to BT, which has the potential to skew the data in favor of BT. This is corroborated by the low satisfaction scores in the surgical group as compared to historical controls.
  • An important point in this study is that all the BT was performed by a single expert operator whereas many different surgeons performed the RPs. This introduces significant variability in outcomes and makes generalizability to patient populations at large more difficult.