We investigated the dynamics of PSA in patients with prostate cancer who were treated with CIRT and neoadjuvant ADT in the present study. Both PSA bounces and PSA failure were correlated with younger age. To the best of our knowledge, this is the first report of PSA dynamics after CIRT with neoadjuvant ADT.
Multiple definitions of the PSA bounce have been reported, and no consensus has been established. Several studies used the definition of an increase of >0.2 ng/ml in PSA levels followed by a spontaneous decrease to the pre-bounce level or lower [14, 17, 21, 28–29]. In the present study, no patient met this definition, as the nadir PSA level was extremely low because of the use of neoadjuvant ADT. Thus, we defined the PSA bounce as an increase of at least 0.4 ng/ml followed by any decrease, in line with previous studies [25, 26]. In a study of PSA bounces in patients treated with conventional external radiotherapy, the bounce was defined as an increase of 0.5 ng/ml [30]. Conversely, the PSA bounce was defined as an increase of 0.1 ng/ml followed by two consecutive decreases after IMRT [31].
PSA bounces have been mainly reported after brachytherapy. PSA bounces were observed in 28–49% of patients after LDR-BT [15]. In two other studies, PSA bounces were observed in 43 and 48% of patients treated with HDR-BT, respectively [32, 33]. In a study of PSA dynamics after HDR-BT combined with conventional external beam radiotherapy, PSA bounces were detected in 31% patients [34].
PSA bounces have also been observed in patients treated with external beam radiotherapy alone. After IMRT, the occurrence rate of PSA bounces ranged 30%–32% [25, 31]. Recently, SRT has been performed for the definitive treatment of prostate cancer, and PSA bounces were also observed after SRT. In a multi-institutional analysis of PSA dynamics, PSA bounces were noted in 26% of patients [19]. Only one study of PSA dynamics after particle beam radiotherapy has been reported [21]. In that study, PSA bounces were observed in 55.7% of patients treated with CIRT alone for prostate cancer.
Age was one of the first and most frequently described predictive factors for PSA bounces after brachytherapy [15].Age was a significant consistent predictor of PSA bounces after IMRT [25, 31]. Regarding other treatment modalities, namely, SRT or HDR-BT combined with external beam radiotherapy, younger age was a significant predictor for PSA bounces [19, 34]. In addition, age was detected as a predictive factor for PSA bounces after CIRT [21]. Similar results were observed in the present study, as younger age was a predictive factor for PSA bounces and PSA failure. Therefore, it is suggested that age is a predictor for PSA bounce regardless of the radiotherapy modality.
Despite the accumulation of data on post-radiotherapy PSA dynamics, its relevance to clinical outcomes remains unclear. One study suggested that PSA bounces did not predict biochemical recurrence or clinical disease recurrence [31]. Another study reported that PSA bounces after external beam radiotherapy were correlated with PSA failure [26]. By contrast, some reports stated that the PSA bounce was a good predictive factor for PSA failure [20].Hinnen et al. found that PSA bounces after LDR-BT were predictive of better outcomes [35]. A long-term analysis suggested that the PSA bounce was a significant factor for better overall survival [36]. In CIRT, PSA bounce positivity was a significant predictor of favorable 5-year PSA failure-free survival [21]. In the present study, we found a correlation between PSA bounces and PSA failure only in univariate analysis. Longer follow-up is warranted to further explain this issue.
Some patients who exhibited PSA bounces experienced increases in PSA levels of 2 ng/ml or more, which met the Phoenix criteria. The PSA bounce exceeds the 2 ng/ml limit in approximately 10% of patients after brachytherapy [14].Approximately 1% of patients treated with SRT experienced a PSA increase of >2 ng/ml above the nadir [19]. However, PSA levels spontaneously decreased without any treatment in those patients. A similar clinical course was observed in the present study, as most patients experienced spontaneous decreases of PSA levels. Therefore, even among patients with PSA increases exceeding 2 ng/ml, which met the PSA failure criteria, continuous close PSA surveillance should be considered to confirm the PSA bounce without immediate treatment such as ADT. These findings may provide important information for both patients and physicians to understand PSA dynamics after CIRT.
The present study had several limitations, such as its single-institutional nature, small number of patients, short observation period, and lack of cases of clinical recurrence. Although the correlation between PSA bounces and androgen production in younger age patients was suggested [37], serum androgen levels were not measured in the present study.