Should PSA levels be “undetectable” at 5 years after successful brachytherapy?
Posted Apr 06 2010 12:00am
Most brachytherapy centers today use the Phoenix criteria to define clinical success and failure after brachytherapy. But a new paper suggests that brachytherapy can (and perhaps should) use a higher standard for long-term follow-up.
In an article in press in the on-line UroToday International Journal, Barrett and Herzfeld have carried out a retrospective review of data from 163 consecutive patients treated with radioactive iodine-125 seed implantation as definitive treatment between 1996 and 2003. None of these patients received external radiation or hormone therapy as adjunctive treatment. All patients were followed for a minimum of 3 years, and their PSA levels were drawn 1 month and 3 months following the implant, every 3 months for the rest of the first year, and every 4 to 6 months thereafter. A single PSA assay method was used to assess all PSA levels for these patients over this time period.
The authors carefully analyzed data on the post-brachytherapy PSA levels of all 163 patients, categorizing them as ≤ 0.1, ≤ 0.2, ≤ 0.3, and ≤ 0.5 ng/ml.
All 163 patients lacked clincial evidence of extraprostatic disease at the time of treatment. The majority of patients were clinical stage T1c, but ranged from T1a through stage T2c. The median Gleason score was 6 (range, 5 to 8). The median PSA level was 6 ng/ml (range, 1.6 to 17.7 ng/ml). Thus the series included patients with low, intermediate, and high risk for progression.
The authors also note that at the time of original treatment and follow-up for these patients, biochemical failure was based on the ASTRO consensus definition, which was three consecutive rises in PSA at least 3 months apart from each other. The Phoenix criteria used today define biochemical failure after radiation therapy as a PSA level equal to the lowest (nadir) PSA level achieved after therapy plus 2 ng/ml.
By comparison, the standard expectation for a PSA after radical prostatectomy today is an “undetectable” level, usually taken to mean a PSA level of < 0.1 ng/ml. (Although several different PSA levels have been used in the literature over time to define clinical success and failure after radical prostatectomy.)
Bartlett and Herzog were able to show the following results in this series of patients:
At a median follow-up of 85.2 months (about 7 years), 120/163 patients (73.6 percent) achieved and maintained an undetectable PSA level (< 0.1 ng/ml).
The median time to achievement of an undetectable PSA level was 40 months following the brachytherapy implant.
The mean and the median number of PSAs obtained per patient was 19.
91/140 patients who were biochemically free of disease (64 percent), based on the ASTRO consensus criteria, experienced a PSA bounce (transient elevation).
23/163 patients (14.1 percent) experienced biochemical recurrence by the ASTRO consensus statement definition.
None of the patients who obtained an undetectable PSA following prostate brachytherapy had a subsequent rising PSA.
131/163 patients (80.4 percent) patients achieved and maintained a PSA level ≤ 0.2 ng/ml.
136/163 patients (83.4 percent) patients achieved and maintained a PSA level ≤ 0.3 ng/ml.
139/163 patients (85.3 percent) achieved and maintained a PSA level ≤ 0.5 ng/ml.
Only 6/23 patients who had a biochemical recurrence according to the ASTRO consensus statement received salvage therapy of any type.
It is also worth noting the following results in the 26 patients with a prostate volume > 50 cm3:, none of whom received neoadjuvant hormonal therapy to decrease prostate volume:
15/26 patients (57 percent) had undetectable PSA levels (< 0.1 ng/ml) and 17/26 patients (65 percent) had PSAs ≤ 0.2 ng/ml.
Biochemical failure, according to the ASTRO consensus definition, occurred in 5/26 patients (19.2 percent).
What is interesting in looking at these data is (a) the extended time it takes for the patients to achieve a PSA level < 0.1 ng/ml as compared to the surgical patients and (b) the fact that it is, in fact, an apparently achievable endpoint. It is tempting to wonder whether 5-year and 10-year outcomes for brachytherapy patients should in fact be reported on this basis, permitting a direct “apples to apples” comparison of outcomes of brachytherapy patients to surgery patients. As the authors are careful to point out, this series “is relatively small and of intermediate follow-up duration.” However, they are still able to conclude that, “Using identical biochemical endpoints, brachytherapy appears to confer similar disease-free status to prostatectomy.”