Although the small number of positive biopsies, low Gleason scores (< 7) and low stages (cT1-2) may be associated with pT0 (8), only ADT duration was indicative of conversion into pT0 in the present study. The duration of ADT, particularly after PSA reached < 0.2 ng/ml, was significantly associated with pT0. Of the six patients who received ≥10 months of ADT after PSA reached < 0.2 ng/ml, five (83.3%) were classified as pT0. Whether viable cancer cells persist cannot be easily determined even by using immunohistochemistry or thin sections of specimens. Thus, pT0 may not be interpreted as the complete elimination of cancer. Kollermann et al. [8
] did not observe a significant difference in PSA progression-free survival between patients classified as pT0 and non-pT0. However, we observed PSA progression in only one of 23 pT0 patients (4.3%) with a median follow-up of 35 months, and patients in the pT0 group had a tendency for longer PSA progression-free survival. Kitagawa et al. [9
] showed that PSA progression was not observed in patients with a grade 3 pathological effect, regarded as pT0, with a median follow-up of 34.2 months. These observations indicate that pT0 can be regarded as an evidence of considerable cancer cell damage. Even when a small number of viable cancer cells remain, those cells will not develop clinically significant cancer in all cases. Thus, durations of ADT that induce pT0 could cause serious damage to cancer cells and achieve long-term control of localized prostate cancer.
Accumulated data illustrated the effect of ADT in combination with radiation therapy or prostatectomy. RTOG 92-02 [10
] and EORTC 22961 [11
] demonstrated the survival advantage of long-term ADT (24-36 months) compared with short-term ADT (4-6 months) in combination with radiation therapy. Because local treatment was identical in both groups, the systemic effect of 2-3 years of ADT could reduce or eliminate micrometastases, as well as any residual primary cancer cells, resulting in a better outcome. Most studies on neoadjuvant ADT before prostatectomy did not show a survival advantage [12
], suggesting that neoadjuvant ADT does not completely eliminate cancer cells. In those studies, however, the duration of neoadjuvant ADT was 3-8 months. In intermittent androgen suppression (IAS), usually performed for localized, metastatic, or recurrent prostate cancer, the median on-therapy time is 3-9 months and PSA increases after 3-16 months of off-therapy [13
]. These observations suggest that duration of ≤ 9 months may be insufficient, but 2-3 years of ADT may considerably eliminate cancer. Although whether 2-3 years of ADT is optimal remains unclear, the efficacy of limited-time, not lifelong, ADT may have the potential to obtain long-term control or cure for prostate cancer.
Labrie et al. [14
] demonstrated the possibility of a cure with MAB of limited duration for localized prostate cancer. With a median follow-up of 4.9 years (for stage T2) and 5.6 years (for stage T3), non-PSA failure rates were 36%, 87.5% and 91.7% for patients previously treated with MAB for 3.5-6.5 years, 6.5-10 years and 10-11.7 years, respectively. No biochemical or clinical progression occurred in patients with localized cancer (T2) treated with MAB for ≥6.5 years. These results indicate that long term (> 6.5 years) continuous MAB offers the possibility of long-term control or a possible cure for localized prostate cancer. Although long-term (over 6.5 years) ADT may cure localized prostate cancer, whether such long-term ADT is really required remains unclear. In the study by Labrie et al., PSA failure was defined as an elevation of PSA > 1.0 ng/ml, which is too low to define cancer recurrence, because benign prostate cells can also produce PSA. Following radiation therapy with or without neoadjuvant ADT, PSA recurrence was defined as a PSA value of ≥2.0 ng/ml above the lowest value [15
]. Then, duration ≤ 6.5 years of ADT may be sufficient to control localized cancer.
Cessation of ADT may not completely cure localized prostate cancer that may be obtained by long-term ADT, because pT0 does not always indicate complete elimination of cancer cells. Prostate cancer may recur following cessation of ADT in some patients. When PSA increased after cessation of ADT, Labrie et al. [14
] observed a decrease of PSA in all cases by re-introduction of ADT. Re-introducing ADT can decrease the PSA level in most patients treated with IAS, which is at least as effective as continuous ADT [13
]. These observations indicate that a vast majority of cancer cells remain androgen-sensitive after approximately 2 years of ADT. Rapid progression is unlikely, even in patients whose PSA levels rise after cessation of ADT, and patients could be treated successfully by re-introducing ADT at the time of recurrence. Therefore, cessation of ADT may not result in a poorer outcome compared with long-term continuous ADT in patients with localized prostate cancer.
The present study had some limitations. First, as all patients received MAB, the duration of ADT with mono-therapy, either castration or solely anti-androgen, is yet to be determined. Second, all patients in this study were Japanese. The response to ADT may differ among races, and Japanese males respond better to ADT than Caucasian males [16
]. The appropriate duration may differ among races and should be defined by an investigation in other races. Third, the prevalence of pT0 was higher than that in previous reports [8
]. We performed pathological evaluation using only hematoxylin and eosin (H & E) staining; the prevalence of pT0 may decrease if immunohistochemistry was employed. Relatively long durations of neoadjuvant ADT (58.8% were ≥8 months) and the race (Japanese) in our patients may also have been associated with the higher prevalence of pT0.