Assessment of success following treatment
Detection of recurrent disease Assessment of success following treatment with an implant is based on biochemical or PSA (bNED) control as well as clinical control. Defining bNED control following prostatectomy is more straightforward in cases in which the prostate is removed and PSA is undetectable. Assessing bNED control following treatment with either EBRT or implants is more complicated in cases in which the prostate is not removed and the normal tissue makes some PSA, even if the cancer is eradicated. Before 1997, various institutions across the United States and Europe defined bNED control differently, and comparing results was difficult [64]. In an effort to standardize reporting, the ASTRO convened a panel to develop a standardized definition of PSA success or failure. The consensus statement on PSA after radiation therapy (RT) [65] was published in 1997 and quickly adopted within the radiation oncology community, allowing for uniformed reporting of bNED control. This definition defined the point of biochemical failure as the time midway between the posttreatment PSA nadir and the first of the three consecutive rises in PSA level. This definition was to be applicable in clinical practice and in research trials; avoid the issue of the amount of baseline serum PSA that might be produced in the prostate gland following RT; be valid for comparing different methods of radiation delivery; and avoid requiring a specific single value for posttreatment nadir PSA, which can be fraught with statistical peril. However, this definition also has significant drawbacks. First, many of the articles that publish biochemical-free survival in patients treated with brachytherapy have used alternative methods for evaluating treatment failure. Most commonly, absolute PSA cutoffs have been used in the studies [22,66], making future comparisons with studies using the ASTRO criteria difficult. In a recent study from Fox Chase Cancer Center evaluating 1017 men treated with radiotherapy for localized prostate cancer, the ASTRO definition of biochemical failure was noted to artificially improve bNED rates. Several alternative definitions have been proposed [67]. Another study from the University of Virginia, Coblentz et al [10] showed that when a PSA level greater than 0.2 ng/mL was used to define biochemical progression, 88% (95% CI = 80–97) of patients remained free of disease at 24 months compared with 97% (CI = 92–100) and 90% (CI = 82–98) when a PSA level greater than 1.0 ng/mL and ASTRO criteria were used, respectively, demonstrating how different methods of defining bNED can lead to slightly different outcomes. This definition also is difficult to apply to patients treated with neoadjuvant hormonal manipulation in view of the fact that their PSA nadir is related to the duration of hormonal manipulation. 742 E.M. Horwitz et al / Urol Clin N Am 30 (2003) 737–750 Although there are recognized limits to the definition—including those described above and the fact that the definition was designed originally for EBRT patients only—the use of this definition has been valuable for reporting and comparing most results.
