To treat pelvic nodes or not: could the greater testicular scatter dose from whole pelvic fields confound results of prostate cancer trials?

Whether or not the elective irradiation of pelvic nodes provides any benefit compared with treating the prostate only has been a longstanding therapeutic dilemma. The results of many retrospective studies were conflicting. Then, in 2003, the first results of Radiation Therapy Oncology Group (RTOG) 94-13 were published. This fourarm randomized trial for prostate cancer patients at high risk of occult nodal involvement compared whole pelvic (WP) versus prostate only (PO) fields with either neoadjuvant and concurrent hormone therapy (NHT) or adjuvant hormone therapy (AHT). It showed that when administered with NHT, there was improved progression-free survival (PFS) with WP compared with PO radiation fields (PFS: 60% v 44% at 4 years, respectively; P .008) and improved prostate-specific antigen (PSA) relapse-free survival (biochemically no evidence of disease [bNED]: 70% v 57% at 4 years, respectively; P .048). However, no overall survival (OS) differences were observed. What at first seemed to be a fairly clear-cut positive trial became complicated when, in 2007, a published update with 1.6 years longer follow-up demonstrated diminished differences in PFS (P .065), which were now interpreted as a trend in favor of WP with NHT. The bNED differences had now vanished (P .43) between these two treatment arms. In an unexpected twist, differences in OS were now emerging but were significant only when comparing WP with NHT versus WP with AHT (82% v 74%, respectively; P .019). However, the trial was not originally powered to test differences between these two treatment arms. These collective findings were interpreted as the result of an unexpected and unrecognized biologic interaction between the timing of hormone therapy and pelvic nodal radiotherapy, and the conclusion was still to recommend WP with NHT for high-risk prostate cancer. To complicate matters further, in 2007, the results of another randomized trial, Groupe d’Etude des Tumeurs Uro-Genitales (GETUG) -01, comparing WP versus PO fields with NHT for highrisk prostate cancer patients showed no difference in 5-year PFS (66% v 65%, respectively; P .34) or OS (86.5% v 88%, respectively; P .62). Although the results of RTOG 94-13 are provocative, they are puzzling and difficult to accept as evidence in support of WP radiotherapy given our lack of biologic understanding of why it is that these results would seem to change with longer follow-up. This leaves the issue of whether or not to electively irradiate pelvic nodes for high-risk patients with prostate cancer still without consensus. We propose a hypothesis for this unrecognized biologic interaction between the timing of hormone therapy and WP radiotherapy that could explain the curious results of RTOG 94-13. Our reasoning is as follows: there is a larger scatter dose to the testes with WP fields compared with PO fields; the negative effect on testosterone production is proportional to the dose to the testes (or more precisely the Leydig cells), and therefore, the time to testosterone recovery is proportional to testicular dose; and the increase in time to testosterone recovery leads to a delay in biochemical failure and therefore gives an apparent but temporary advantage for WP radiotherapy. With partial Leydig cell depletion from NHT, a synergistic effect leading to longer time to testosterone recovery from testicular dose would also be predicted for WP radiotherapy with NHT compared with WP radiotherapy with AHT. First, we will discuss higher testicular scattered dose with larger field size. We first look at the data that pertains to the scatter dose to the testes from megavoltage external-beam radiotherapy using threedimensional conformal four-field techniques for the treatment of prostate or rectal cancers, summarized in Table 1. None of the patients in these studies received hormone therapy. For a prostate-only field delivering 68 to 72 Gy, the mean scattered testicular dose is 2.1 Gy. For a course delivering 50 Gy to the pelvis, the mean scattered testicular dose is 3.6 to 8.4 Gy. From this data, the cumulative scatter dose from a course of radiotherapy with WP (45 to 50 Gy) and prostate boost (25 Gy) is calculated to be within the range of 4.3 to 9.1 Gy. Thus two to four times more scattered testicular dose is administered with WP fields compared with PO fields. Our second point concerns dose-response relationship on testicular function. The impact of testicular dose on hormonal function (including testosterone, luteinizing hormone, and follicle-stimulating hormone levels) is also summarized in Table 1. Significantly lower testosterone levels compared with baseline levels have been demonstrated after pelvic radiotherapy, with a mean decrease of 20% to 30% measurable within weeks of radiotherapy and persisting for months afterwards. In the largest and most detailed study of more than 600 patients receiving PO radiotherapy without hormones, the testosterone levels decreased by a mean of 17%. In 40% of patients, the testosterone recovery was never to their baseline levels. In addition, on multivariate analysis, the decrease in testosterone levels was greater among patients treated with larger field sizes and patients with lower JOURNAL OF CLINICAL ONCOLOGY COMMENTS AND CONTROVERSIES VOLUME 27 NUMBER 36 DECEMBER 2