Patterns of recurrence after low-dose-rate prostate brachytherapy: a population-based study of 2223 consecutive low- and intermediate-risk patients – Beyond the Abstract

Radical prostatectomy (RP) and low-dose-rate brachytherapy (LDR-PB) are two of the most common modalities for prostate cancer treatment, but studies comparing these two treatment modalities are scarce. While local control is a logical metric to compare treatment effectiveness, previous comparison studies have not examined this endpoint.

We examined patterns of recurrence and estimated the local recurrence rate after LDR-PB using a prospective population-based database maintained at our institution, with clinical, dosimetric and outcome data for all LDR-PB procedures performed in British Columbia (BC)1. At a median follow-up of 5 years, 108 of 2223 patients treated with LDR-PB developed biochemical relapse. An additional patient was found to have local recurrence on transurethral resection of the prostate, but this was not preceded by biochemical failure. Thus, the crude rate of any disease relapse was 4.9% (109/2223). Of these 109 patients with relapse, the site of first recurrence was established in 48 cases, in which 18 of 2223 (0.8%) were local, and 30 of 2223 (1.3%) were distant. Of the remaining 61 of 2223 patients, 93% had digital rectal exams, 30% had post-treatment biopsies, 74% had bone scans, and 56% had CT imaging of the abdomen and pelvis, all of which failed to identify a site of recurrence.

To determine the local recurrence rate after LDR-PB, we first established that the widest range within which this local failure rate could fall was between 0.8% (those with proven local recurrences) and 4.9% (if we assumed that every relapse was local). A more realistic upper limit approximation of 2.7% (61 of 2223) was achieved by eliminating the patients with proven distant failure, and those with both a negative biopsy and negative digital rectal exam. The lower limit of local recurrence rate was adjusted by assuming that patients with an unknown site of failure had at least the same proportion of local recurrences (38%) as those whose site of failure was known; with this assumption, 23 of 61 patients were added to the 18 known local failures, yielding a lower limit of 1.8% (44 of 2223). Therefore, we concluded that the local recurrence rate after LDR-PB in BC could be as low as 0.8% or as high as 4.9%, but most likely ranged from 1.8-2.7%.

For comparison, we estimated the rate of local relapse after RP in BC, based on surgical pathology and data from randomized control trials. In a population-based study, Tyldesley et al showed that 30% of RP patients with organ-confined (pT2 disease) (N=1614) in BC had positive surgical margins2. According to the European Organization for Research and Treatment of Cancer (EORTC) randomized controlled trial 229113, half of all pT2 patients with positive surgical margins will develop biochemical relapse, and approximately half of these recurrences can be prevented by the use of adjuvant RT. Combining this information with that of Tyldesley et al2, about one-half of the 30% of pT2 patients with positive margins will develop biochemical relapse, and one-half of these recurrence events can be prevented by using local RT and are therefore local. This suggests that the local recurrence rate after RP for pT2 disease in BC is approximately 7.5%. In all likelihood, this 7.5% is an underestimate of the true local recurrence rate after RP because it does not account for local relapses that would have been salvaged in the observation arm, nor for the isolated local relapses that postoperative RT would not have cured.

The limitations in our estimates of local recurrence rates after LDR-PB and after RP need to be considered. In calculating local recurrence rates, we did not include the possibility of non-secreting PSA relapses, but these events were equally unaccounted for in both treatment modalities. Another limitation is that by combining data from Tyldesley’s study with that of EORTC 22911, we are assuming that the results of EORTC 22911 are directly applicable to our post-prostatectomy population, an assumption for which we have no direct evidence. Furthermore, we lack data for pretreatment disease characteristics in the RP patients, which limits our ability to ensure balanced risk cohorts between the 2 treatment modalities for a more rigorous comparison of their efficacy. We attempted to balance the cohorts by eliminating pathologic T3 patients treated with RP from our analysis.

In the context of the limitations of our study design, this population-based analysis indicates that the local recurrence rate after LDR-PB appears to be as low or lower than that following RP in our jurisdiction. Further research is required to determine whether this finding can be generalized to other institutions and populations.

1. Lo AC, Morris WJ, Pickles T, et al. Patterns of Recurrence After Low-Dose-Rate Prostate Brachytherapy: A Population-Based Study of 2223 Consecutive Low- and Intermediate-Risk Patients. Int. J. Radiat. Oncol. Biol. Phys. 2015;91(4):745–751.
2. Tyldesley S, Peacock M, Morris JW, et al. The need for, and utilization of prostate-bed radiotherapy after radical prostatectomy for patients with prostate cancer in British Columbia. Can Urol Assoc J. 2012;6(2):89–94.
3. Bolla M, van Poppel H, Collette L, et al. Postoperative radiotherapy after radical prostatectomy: a randomised controlled trial (EORTC trial 22911). The Lancet. 2005;366(9485):572–578.
Int J Radiat Oncol Biol Phys. 2015 Mar 15;91(4):745-51. doi: 10.1016/j.ijrobp.2014.12.014.

Written by:
Andrea C. Lo, MD and W. James Morris, MD
Department of Radiation Oncology, BC Cancer Agency Vancouver Centre, 600 West 10th Avenue, Vancouver, BC, Canada.

Abstract: Patterns of recurrence after low-dose-rate prostate brachytherapy: a population-based study of 2223 consecutive low- and intermediate-risk patients