In this large series with long-term follow-up, we found excellent biochemical control rates and low and acceptable toxicity. PSA fell steadily after treatment and achieved very low levels (mean of 0.25 ng/ml) within 4–5 years, findings that portend good long-term disease control outcomes [29, 30]. These findings support an estimate of the α/β ratio of 1.5 Gy. A ratio of 1.5 Gy means that we have delivered an equivalent dose (EQD) of 90–96 Gy at 1.8 Gy per fraction, an EQD which accounts for the higher control rates than those seen with the use of 81 Gy . Although our results with high-risk patients are encouraging, it is important to note that our study included only 12 patients and more data is necessary to confirm these findings.
Our outcomes are consistent with those that have resulted from high dose rate brachytherapy (HDR BT), with or without EBRT [15, 16]. In a recent paper Demanes et al.  reported an 8-year recurrence-free survival of 97% in a mixed cohort of low and intermediate-risk patients. If additional follow-up confirms that this level of long-term disease control can be obtained with SBRT, SBRT’s advantages over HDR BT, primarily its non-invasiveness and ability to deliver treatment to patients without anesthesia or hospitalization may make it the preferred modality.
We employed two dose levels in our study. We initially treated patients with 35 Gy but escalated to 36.25 Gy six months into the study after observing low acute toxicity at 35 Gy and after reports from others of acceptable toxicity at a dose of 36.25 Gy. Based on current data, however, the higher dose does not appear to be necessary for low and low-intermediate patients. No difference in PSA control or nadirs were seen between the two doses, a finding which corroborated a recent matched-pairs study with 48 month follow-up . A trend to increased toxicity with the 36.25 Gy dose was observed. It is possible that these events did not rise to the level of statistical significance due to the small number of patients within the 35 Gy group. Due to these findings, we resumed treating low- and low-intermediate risk patients with 35 Gy soon after the present study was completed. With more patients and longer follow-up a significant improvement in toxicity at the lower dose may be observed, in which case 35 Gy may be the optimal dose to assure long-term disease control and low toxicity. Such a finding would imply a flattening of the biologically equivalent dose response curve from 90–96 Gy EQD (assuming an α/β ratio of 1.5 Gy).
Our results are supported by a recent study of 1101 patients in a pooled analysis from eight institutions , reported at the 2012 meeting of the American Society of Radiation Oncology (ASTRO). This analysis reported only on biochemical control outcomes and found 96%, 92%, and 80% control with five-year actuarial follow-up for low-, intermediate- and high-risk patients, respectively. These results excluded the PSA failures that subsequently resolved on their own (i.e., “bounces”). Importantly, the three-year median follow-up results were excellent (at 80% control) for more than 100 patients with high-risk disease. These outcomes approximate those obtained in the current study, as well as from a study by Katz et al.  that reported long-term follow-up results for high-risk patients that received a CyberKnife boost after EBRT. In this study a biochemical control rate of 77.7% at 3 years was obtained for high-risk patients who received 45 Gy to the pelvis followed by a CyberKnife boost of 18–21 Gy. No differences were found between patients who received 35 Gy or doses as high as 40 Gy. The use of ADT also did not affect outcomes. Longer follow-up with more patients is warranted before firm conclusions can be made about the efficacy of SBRT monotherapy or SBRT as a boost for these patients at a higher risk for disease outside the prostate.
Because surgery is often used instead of radiation to treat prostate cancer, patients need information on both disease control and QOL changes associated with either modality. To better gauge the impact of prostate cancer treatment on QOL, one study compared the QOL responses from a large group of patients who had recently received CyberKnife SBRT to those of a similar group who instead underwent open surgery . EPIC scores were used to assess QOL. For all time intervals up to 36 months, the patients who received SBRT had superior EPIC scores (in terms of urinary and sexual domains) than those who underwent surgery. Bowel domain was slightly worse in the short term for those in the SBRT group, but patients in both the SBRT and surgery groups had excellent preservation of bowel function after 12 months. It is important to note that surgical patients underwent open prostatectomy. It is possible that improvements in prostatectomies, including the use of laparoscopic techniques, will improve QOL post-surgery. On the other hand, SBRT patients in this study were older, on average, a factor that could have swayed QOL in favor of surgery. Although it is clearly not appropriate to assert forcefully the superiority of SBRT over surgery based on the data in this study, there is at least no evidence that SBRT results in poorer QOL outcomes for prostate cancer patients.