Prolonging the treatment time is generally not advisable in radiation oncology, because of the theoretical risk of accelerated tumor cell repopulation. For squamous cell carcinoma of the head and neck, this has been a well-documented paradigm. For prostate adenocarcinoma, the results have been less clear, with conflicting studies. The University of Florida analyzed local control using time–dose scatter distributions, as well as absolute treatment time divided by stage (Stage A-C) (9
). The cohort analyzed was in the pre-PSA era using non-Gleason score grading and 45% of patients were diagnosed by transurethral resection of the prostate. In addition, some patients were treated with split-course RT. Although this group is clearly not comparable to the patient in the present report, they did show worse local control for Stage B2 patients whose total treatment time was >8 weeks instead of ≤8 weeks. Stage B2 was defined as palpable tumor confined to the prostate and involving at least an entire lobe. Of these Stage B2 patients, 77% had their tumor graded as well or moderately differentiated, which might have been comparable to some of the low-risk cohort in the present report. In the Florida series, when the radiation dose was limited to ≥65 Gy, the overall treatment time was significant for well, moderate, and poorly differentiated grades.
Lai et al.
) analyzed overall treatment time on prostate cancer outcome using patients enrolled in two prospective Radiation Therapy Oncology Group studies (17
). The patients were required to have received a minimal dose of 65 Gy and no androgen deprivation therapy. The patients were stratified into three total treatment time groups: ≤7, 8–9, and >9 weeks. No differences were found in local control, disease-free survival, or overall survival according to treatment time for this cohort, even when stratified by Gleason score or T stage. Another retrospective, single-institution experience likewise showed no difference in outcome according to overall treatment time (5
These previous studies described the results of a much different group of patients than in the present study. To our knowledge, the only other series addressing overall treatment time in the PSA era was from William Beaumont Hospital (19
). This series addressed the effect of clinical- and treatment-related prognostic factors on biochemical outcome. Treatment time was not predictive of the FFBF on univariate or multivariate analysis, including pretreatment PSA level, T stage, or Gleason score. The additional strengths of the present study include higher radiation doses and longer follow-up. Dose-escalated RT has been shown to improve FFBF through better local control (20
). This benefit with higher doses might allow differences in local control related to secondary factors, such as the NTDR, to be more apparent. Also, extended follow-up enables better detection of late BF resulting from a slowly increasing PSA level, a pattern consistent with local disease persistence (23
The present study is the largest reported to address this question and the first to stratify by contemporary risk groups. To account for the varying radiation doses, which corresponded to the varying durations of treatment, using conventional fractionation, we analyzed the treatment time using a ratio (NTDR) of the nontreatment days to the total number of elapsed days. The NTDR analyzes nontreatment days as a percentage of the total treatment duration. An NTDR of <33% corresponds to ≤3 nontreatment days (including holidays) during a prostate RT regimen totaling 40 fractions delivered 5 d/wk. With a greater NTDR, a statistically significant detriment in FFBF was found in the low-risk group.
The relationship of NTDR to BF was only seen in the low-risk group. We postulated that the importance of treatment time for local RT is inversely related to the risk of metastatic disease. Patients most likely to be affected by differences in local treatment are those most likely to have local-only disease (i.e.
, low risk). Conversely, one would expect intermediate- and high-risk groups to be more likely to present with subclinical metastatic disease that was untreated in this study of RT alone (no androgen deprivation therapy). Subsequent progression of metastatic disease contributes to BF and might have diluted the effect of an increasing NTDR in the intermediate- and high-risk patients (23
). These intermediate- and high-risk patients might require radiation doses greater than those used in our study and/or adjuvant androgen deprivation therapy to see an effect of prolonged treatment time.
This study was a retrospective, single-institution study and was therefore subject to biases. These included patient selection for allowing treatment breaks, patient selection for RT alone, preference for the day of the week patients began treatment, and the use of a 2-Gy fraction size. However, these data are thought provoking and should be investigated in other data sets, as well as prospectively. This study supports minimizing treatment breaks to <3 days when 38 fractions (2-Gy fractions, 5 d/wk beginning on a Monday, 76 Gy total) are used and to <4 days for 40 fractions (total dose 80 Gy). Individual institutions are encouraged to calculate the NTDR for a more precise evaluation of treatment protraction according to the RT regimen used.