Why this matters
For men eligible for either prostatectomy or prostate SBRT, sexual function is a central part of treatment selection. The relevant question is not simply how treatment affects early recovery, but what patients experience several years later.
The five-year sexual-function analysis from PACE-A provides a direct comparison within a randomized trial. Patient-reported erectile function favored SBRT, but the result requires careful separation of three measures: erectile function, broader sexual quality of life, and clinician-graded erectile dysfunction.
These measures did not produce identical statistical conclusions.
Study design
PACE-A was a multicenter, open-label randomized comparison of radical prostatectomy and SBRT in men with early localized prostate cancer who were suitable candidates for surgery.
A total of 123 participants were randomized: 60 to prostatectomy and 63 to SBRT. Treatment consisted of prostatectomy using an open, laparoscopic, or robotic approach, or 36.25 Gy in 5 fractions over 1-2 weeks. ADT was not permitted as part of the protocol treatment.
The original co-primary endpoints were patient-reported urinary and bowel symptoms at two years. This report concerns secondary sexual-function outcomes through five years, including EPIC-26 sexual domain scores, clinically important deterioration from baseline, IIEF-5 scores, and clinician-reported CTCAE grade 2 or higher erectile dysfunction.
The presentation reported a median overall follow-up of eight years. However, the sexual-function results presented here concern the five-year assessment.
Importantly, these analyses were performed according to treatment received, rather than preserving the original randomized allocation. The stated statistical significance threshold was p<0.01.
Key results
Patient-reported erectile function
The clearest statistically significant finding favored SBRT on the International Index of Erectile Function-5.
At five years, the presentation reported moderate-to-severe erectile dysfunction in:
- 44% after SBRT: 14 of 32 respondents
- 89% after prostatectomy: 23 of 26 respondents
- p<0.001
The abstract also reported median IIEF-5 scores of 19 after SBRT versus 5 after prostatectomy, p<0.001, with higher scores indicating better erectile function.
These are results among patients with available questionnaires, not percentages of the entire randomized population.
Broader sexual function
Median EPIC-26 sexual domain scores declined in both groups. After SBRT, the median score decreased from 79 at baseline to 58 at five years. After prostatectomy, it decreased from 62 to 17.
The reported median within-patient change was -17 points after SBRT versus -42 after prostatectomy, p=0.02.
A clinically important deterioration, defined using an 11-point reduction, occurred in 54% of evaluable SBRT patients (19/35) versus 82% after prostatectomy (18/22). The presentation reported p=0.02 for this comparison.
Both EPIC comparisons numerically favored SBRT, but neither met the study's stated p<0.01 significance threshold. The lower baseline EPIC score in the prostatectomy group also matters when interpreting the raw five-year scores.
Clinician-reported erectile dysfunction
Clinician grading showed a different pattern.
At five years, CTCAE grade 2 or higher erectile dysfunction was reported in 27.1% after SBRT (13/48) versus 35.1% after prostatectomy (13/37), p=0.48.
This comparison did not demonstrate a statistically significant difference. It also does not establish equivalence between treatments.
Interpretation
The central finding is that a substantial patient-reported erectile-function difference remained visible five years after treatment. It was not merely an early postoperative effect.
However, describing SBRT as preserving normal sexual function would overstate the result. Sexual function declined after both treatments. Moderate-to-severe erectile dysfunction still affected 44% of SBRT respondents, and more than half of patients with evaluable paired EPIC data experienced a clinically important deterioration.
The meaningful comparison is therefore better reported erectile function after SBRT, not freedom from sexual side effects.
The discrepancy between questionnaires and clinician grading is also clinically relevant. IIEF-5 identified a marked difference, while clinician-reported grade 2 or higher erectile dysfunction did not. These instruments measure different aspects of sexual health and used different evaluable populations. The data do not establish why the results diverged, but clinician grading alone would provide an incomplete account of the patient-reported findings.
The statistical distinction between the questionnaires must also remain explicit. The IIEF-5 comparison met the stated significance threshold. The EPIC change and deterioration comparisons did not, despite clinically notable numerical differences.
Finally, this is a secondary, as-treated analysis from a randomized trial. Baseline sexual function was already better in the SBRT group, and substantial questionnaire attrition occurred. The observed five-year differences should not be treated as fully baseline-adjusted estimates of the causal effect of treatment.
For treatment discussions, the findings support presenting SBRT as an option for men who prioritize long-term sexual outcomes, while acknowledging that neither treatment guarantees preservation of erectile function.
Limitations
Sexual function was a secondary outcome, and analysis according to treatment received weakens the protection against selection bias provided by randomization.
Five-year patient-reported data were available for 60% of the reported prostatectomy group and 63% of the SBRT group. Individual analyses involved even smaller samples: only 26 and 32 patients contributed to the five-year IIEF-5 categorical comparison. Outcomes among patients with missing questionnaires remain uncertain.
Baseline EPIC sexual scores differed between groups. The reported raw follow-up scores and change scores do not remove all concerns about baseline imbalance.
Several EPIC comparisons did not meet the stated p<0.01 threshold. They should not be described as statistically significant solely because their p-values were below 0.05.
The findings concern selected men with early localized disease treated without protocol ADT. They should not automatically be extended to higher-risk populations or radiotherapy regimens incorporating hormonal treatment.