Background
Based on the pioneering work by Huggins [
1], androgen deprivation therapy (ADT) has been the primary treatment for advanced prostate cancer (PC). Unfortunately, most advanced cases of PC eventually become castration-resistant (CRPC), despite the continued use of ADT [
2]. Novel therapies such as docetaxel, enzaltamide, abiraterone, cabazitaxel and sipuleucel-T [
2‐
5] have been developed to treat CRPC. However, the development of agents that inhibit progression to CRPC may represent alternative therapeutic options for PC.
The results of recent studies have revealed growth regulation of PC via steroid nuclear receptors, which included not only the androgen receptor (AR) [
6,
7] but also members of the estrogen receptor (ER) family [
8,
9]. ERα and ERβcx (ERβ2) in particular have been implicated in PC progression and PC-related mortality, whereas ERβ inhibits tumor growth [
8,
9]. In this regard, selective estrogen receptor modulators (SERMs) are expected to change the clinical course of PC. For example, toremifene, an ERα antagonist in the prostate [
10], decreased the incidence of PC in men with high-grade prostatic intraepithelial neoplasia (HGPIN) [
11,
12]. Furthermoere, raloxifene inhibited androgen-independent PC growth in 5 (28 %) of 13 patients [
13]. However, SERMs have not been fully investigated for use in those with treatment-naïve PC. We hypothesized that additional SERMs may prolong the durability of ADT, because androgen and estrogen signaling drive PC progression. In the present study, we conducted a prospective randomized clinical phase IIA trial to investigate the effects of SERMs (toremifene and raloxifene) when added to ADT in treatment-naïve bone metastatic PC.
Discussion
The most common initial therapy for metastatic PC is ADT; however, the durability of ADT is limited and affected by various factors including pretreatment PSA level, GS, tumor stage and PSA nadir [
19]. The durability of ADT is also influenced by the ER status of the tumor [
18,
20].
In fact, estrogens were initially used as one of the earliest forms of treatment agents; however, they were associated with thromboembolic and cardiovascular side effects [
21]. SERMs are synthetic estrogen ligands that can exhibit either estrogenic or anti-estrogenic effects depending on tissue types [
22]. Toremifene significantly reduced the incidence of PC in a transgenic adenocarcinoma mouse prostate model [
23], as well as in men with HGPIN [
11]. In addition, toremifene increased the bone mineral density of the hip and spine [
24] and improved lipid profiles in men receiving ADT for PC [
25]. Raloxifene, which acts as an ER agonist in the bone tissue [
26], has been developed for the treatment of osteoporosis in women [
27] and showed some tumor-inhibitory effects in CRPC in a pilot study [
13]. To date, the anti-cancer effects of these SERMs have not yet been fully investigated in treatment-naïve PC patients. We hypothesized that concurrent use of SERMs would prolong the duration of efficacy of ADT in men with bone metastatic PC.
Currently, we have demonstrated that TOPADT significantly improved the biochemical recurrence rate in men with bone metastatic PC compared with men treated with ADT alone. The results of a recent study showed that the 5-year BCR-free rate was 30 % in men who received ADT plus docetaxel with median serum PSA levels of 26.7 (range, 5.0–106) [
5]. In our study, similar rates were noted for men treated with ADT alone (30 %). Surprisingly, the 5-year BCR-free rate in the TOPADT group was 100 %.
Theoretically, the tumor inhibitory effects of toremifene would be mediated via the suppression of ERα-related signals [
28,
10]. ERα expression in PC cells was confirmed by immunohistochemistry and quantitative reverse transcription polymerase chain reaction analyses [
8,
9,
18]. The mRNA expression of ERα was much lower than AR in PC cells (1:100 ratio); however, ERα expression in cancer-associated stromal cells was significantly related to cancer-specific survival in men with bone metastatic PC [
18]. ERα expression was negatively correlated with survival after radical prostatectomy in locally advanced PC [
29]. Additionally, ERα promoted proliferation by regulating
MYC expression and glucose sensitivity in phosphatase and tensin homolog (PTEN)-deficient mouse PC cells [
8]. Conversely, depletion of
ERα inhibited growth in PTEN-deficient mice via a reduction in
MYC protein and alteration of glucose sensitivity [
8]. The results of present study demonstrated that toremifene significantly improved the durability of ADT, suggesting blockade of ERα signaling as a potential target for advanced PC.
ERβ signaling has been associated with a tumor-inhibitory effect in PC through both the classical (ERβ and estrogen-response element complex) and non-classical pathways (ERβ, Krüppel-like zinc finger transcription factor 5, and adenosine 3′,5′-monophosphate response element-binding protein-binding protein complex) [
30]. ERβ modulators are expected to inhibit PC growth. Raloxifene exhibits diverse activities via ER depending on whether ERα or ERβ is expressed in the target organ [
26]. The results of the present study did not prove a distinct tumor-inhibitory effect mediated by RAPADT as compared to ADT alone. The difference in the reason tumor inhibitory effect between TOPADT and RAPADT may have been attributed to the potency of the drugs and the pattern of ER expression in PC cells. The tumor-inhibitory effect of fulvestrant, another ERβ modulator, was limited because the median time to progression was only 4.3 months in men with CRPC treated with fulvestrant [
31]. Further investigations of additional ERβ modulators are warranted with respect to their potential role in the inhibition of human PC.
The known adverse events associated with the use of SERMs include hot flushes, sweating, nausea, dizziness, edema, vomiting and thrombosis [
27]. In the present study, two men in the ADT group, two men in the TOPADT group, and three men in the RAPADT group complained of mild hot flushes; however, no medical intervention was deemed necessary. Only one man in the TOPADT group discontinued toremifene administration because of a headache. No events of liver dysfunction or thrombosis were observed.
The present study was not without limitations. The sample size was small, and the cohort was limited to a single institution in an all Asian population. A multicenter external validation study would be necessary to further elucidate the additional effect of toremifene on ADT that we found in patients with advanced PC.
Competing interests
The authors declare no conflict of interest.
Authors’ contributions
TF conceived of the study and wrote the draft. ST, SI and YH revised the draft. HK, YY, MS, HF and TN contributed to data collection and analysis. TU and KT contributed to data interpretation and analysis. All authors approved the final version.