2. UroToday Home
  3. Journal
  4. Prostate Cancer and Prostatic Diseases

PSA Response to Antiandrogen Withdrawal: A Systematic Review and Meta-Analysis - Full-Text Article

Background: Antiandrogen withdrawal (AAW) response is the paradoxical decrease in prostate-specific antigen (PSA) following the withdrawal of antiandrogen in patients with advanced prostate cancer. Currently, the reported literature on the proportion of patients exhibiting AAW response and the differences in PSA response between the types of antiandrogens is unclear.

Methods: This review aimed to explore the PSA response to AAW and to identify if the response depends on the type of antiandrogens. A literature search was performed using databases PubMed, Cochrane and EMBASE with a cut-off date of 23rd of November 2020. Studies reporting on outcomes of AAW and prostate cancer were included. Studies were screened by two reviewers and relevant data extracted. Meta-analysis of outcomes was reported using random-effects and fixed-effects model. A subgroup analysis was performed for type of antiandrogen.

Results: From 450 studies, 23 were included with a total of 1474 patients with advanced prostate cancer were available for further analysis. Overall, 395 (26%) patients had any reduction in PSA levels (95% CI: 20–32%) and 183 (11%) patients had a ≥50% reduction in PSA levels (95% CI: 6–16%). Among the 1212 patients on first-generation antiandrogens, 30% (95% CI: 23–38%) had any PSA decline with 15% patients having a ≥50% PSA decline (95% CI: 8–22%). In contrast, among the 108 patients on second-generation antiandrogens, 7% (95% CI: 0–13%) had any PSA decline and only 1% (95% CI: 0–5%) had a ≥50% PSA decline. Also, among the 154 patients on androgen synthesis inhibitors, 26% (95% CI: 19–33%) had any PSA decline and only 4% (95% CI: 0–13%) had a ≥50% PSA decline.

Conclusions: One-fourth of patients treated with AAW show a PSA response. However, PSA response to AAW is uncommon with second-generation antiandrogens and androgen synthesis inhibitors. Further research is required to understand the differences in response between the types of antiandrogen.

INTRODUCTION

Prostate cancer is one of the most common male cancers globally.1 Currently, the mainstay of management of advanced prostate cancer where the disease has progressed beyond the prostate, surrounding tissue and lymph nodes is androgen deprivation therapy (ADT).2 ADT involves castration by surgical methods such as orchiectomy or chemical methods using a gonadotrophin releasing hormone (GnRH) agonist or an antagonist with or without an antiandrogen. Antiandrogens can be classified as classical receptor blocking antiandrogens and androgen synthesis inhibitors based on their mechanism of actions. First-generation antiandrogens include bicalutamide, cyproterone, flutamide and nilutamide. More recently, potent second-generation antiandrogens such as apalutamide, darolutamide and enzalutamide were developed. In addition, androgen synthesis inhibitors such as abiraterone acetate are regularly used in clinical practice. Second-generation antiandrogens achieve androgen blockade by binding to the androgen receptors (AR) and preventing them from binding to DNA.3 On the other hand, androgen synthesis inhibitors bind to and inhibit the enzymes involved in the androgen synthesis pathway, causing a reduction in androgen production without affecting the AR.3 While the second-generation antiandrogens and androgen synthesis inhibitors were originally developed as second-line therapies after failure of first-generation antiandrogens for the treatment of castrate resistant prostate cancer, recently, these novel agents were available for the treatment of hormone sensitive and castrate resistant prostate cancer.

In general, administering a combination of a GnRH agent with or without an antiandrogen is known to cause an initial regression in the prostate cancer, followed by disease progression as the cancer becomes resistant to androgen deprivation. On progression, some patients undergo withdrawal of the antiandrogen as a treatment strategy, which results in disease regression in a proportion of patients. This regression response in prostate cancer following the withdrawal of antiandrogens is known as antiandrogen withdrawal (AAW) response.4,5

In some patients, the withdrawal of antiandrogen agents after being on maximal androgen blockade is accompanied by a regression in prostate-specific antigen (PSA) levels and clinical improvement.6 The AAW response may be observed within 6 weeks from cessation of antiandrogens. However, the duration of response is usually short term and the initial response is followed by disease progression in most patients.6 The molecular mechanisms behind the AAW response are not clear. However, it is suggested that after a period of treatment with antiandrogens, acquired mutations in AR may lead to agonistic/stimulatory downstream response instead of antagonistic response to these medications leading to the AAW response.4-7

While the AAW response was first described following flutamide withdrawal, it has since been observed with other antiandrogens.8 AAW has been employed as treatment strategy both after failure of either first- or second-generation antiandrogens. The reported literature indicates a wide range of varied PSA responses to AAW from 0 to 75%9,10 von Klot et al.9 postulated that such a varied response may depend on the generation of antiandrogens with first-generation antiandrogens being more likely to elicit an AAW response than second-generation antiandrogens.

Here, we report a systematic review and meta-analysis of AAW and the occurrence of AAW response as measured by PSA response (any PSA response and ≥50% PSA decline). This review was performed to determine if there is a difference in AAW response in patients with prostate cancer if first- or second-generation antiandrogens or androgen synthesis inhibitors are used. We hypothesised that AAW response would be more likely with first-generation antiandrogens based on the published literature. Using data extracted from identified studies, a meta-analysis was conducted to provide a pooled estimate of the PSA responses. A subgroup analysis was also performed to assess the occurrence of AAW response in participants treated with first- or second-generation antiandrogen.

METHODS

The methods in this review were conducted in accordance with PRISMA guidelines.11 We searched an international pharmacopeia and identified all the androgen signalling pathway inhibitors that were in widespread clinical use in the treatment of advanced prostate cancer.12 These are abiraterone acetate, apalutamide, bicalutamide, cyproterone acetate, darolutamide, enzalutamide, flutamide and nilutamide. As apalutamide and darolutamide have only recently been approved and thus unlikely to have any publications about their withdrawal, this review excluded them from the analysis.3 We then divided the remaining drugs into three groups. The first group was first-generation antiandrogens, which included bicalutamide, cyproterone acetate, flutamide and nilutamide. The second group was second-generation antiandrogens with enzalutamide being the only drug in the group. Finally, the third group was androgen synthesis inhibitors of which abiraterone acetate was the only drug.3

Sources of data A comprehensive literature search was performed via PubMed (searched up to 23rd of November 2020 from inception), EMBASE (1980 to August 2011) and Cochrane controlled-trials registry databases using the search terms “(anti-androgen withdraw* or anti androgen withdraw* or enzalutamide withdraw* or abiraterone acetate withdraw* or bicalutamide withdraw* or nilutamide withdraw*) AND (prostate cancer or prostate neoplasm* or prostate tumour)”. Date limits in EMBASE were set as the earliest possible date, which was 1980. Grey literature was also taken into consideration.

Study selection and eligibility criteria Studies that reported both androgen withdrawal and prostate cancer were included in the study. However, studies that were case reports, review articles, meta-analyses, animal studies, editorials or opinion pieces were excluded studies involving new or novel antiandrogens and were the subject of experimental reports and not yet in widespread clinical practice were excluded.

Search strategy The articles found in the search of the relevant databases were retrieved and had their titles and abstracts screened by two reviewers using the Covidence platform.13 Subsequently, relevant identified studies were reviewed at a full text level and those meeting the inclusion criteria proceeded to data extracted for meta-analysis. If there was a disagreement between the two reviewers, a third reviewer was consulted.

Data extraction From each selected study, we extracted the following outcomes of interest: age of patients, number of prior lines of therapy, name of antiandrogen immediately prior to AAW, total number of patients treated with AAW, number of patients with any fall in PSA, number of patients with ≥50% PSA decline, duration of PSA response, overall survival (OS), time between AAW and first PSA measurement and progression free survival time (PFS), if available.

Data synthesis The outcomes of data extraction were collated onto a summary table that listed the studies and extracted data. Individual studies were rated for the quality of evidence from 1 to 5 as modified from the Oxford Centre of Evidence-based Medicine for the ratings of individual studies.

Data analysis and statistics Primary outcome measures were (a) the proportion of patients who experience any PSA decline and (b) the proportion of patients who experience a ≥50% PSA decline. Both PSA response measures were reported as % of patients with PSA response among those who underwent AAW. The duration of response, OS and PFS were described whenever available. A subgroup analysis was performed to assess the proportion of patients who experience a PSA decline and the proportion of patients who experience a ≥50% PSA decline based on the generations of antiandrogens. A meta-analysis of the proportion of patients experiencing each outcome was performed using a random-effects and fixed-effects model.14 Subgroup analyses were also performed between first- and second-generation antiandrogens using chi-squared tests. Results were graphically reported as forest plots with the I 2 and p values. A p-value < 0.05 was considered statistically significant. Publication of bias was assessed using a visual inspection of funnel plots. All analyses were conducted using the statistical package Metafor in R.15

RESULTS

Study selection A PRISMA flow diagram depicting the search process and study selection is shown in Fig. 1. A total of 450 studies were identified using the search criteria and from review papers addressing the issue of AAW. One additional study was added from the grey literature.16 Thirty-six studies were found to be duplicates and removed. An examination of the titles and abstracts showed 391 studies did not meet the inclusion criteria and were removed. Finally, the full texts of the remaining 57 studies were reviewed and 34 studies were excluded.

Fig._1_PRISMA_diagram.png
Fig. 1 PRISMA diagram. The Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) flow chart. n = number of studies.

Study characteristics Of the 23 studies available for data extraction, 10 of these were retrospective studies. The others were case series, observational studies, cohort studies and one whose study design was unclear (Table 1). The selected studies were published between 1993 and 2020. The quality of all studies was rated as 3 or 4.

Population characteristics A pooled total of 1474 participants (median ages as reported in the individual studies ranged 59–78 years) were identified. First-generation antiandrogens were used in 1212 (82.2%) and second-generation antiandrogens in 262 (17.7%) patients. All the patients who received first-generation antiandrogens as initial therapy, AAW was attempted at disease progression on reaching castrate resistant state. However, those treated with enzalutamide or abiraterone already had castrate resistant disease when these agents were started; hence, AAW occurred as an intervention after being treated with more than two lines of systemic therapies.

Mean duration of response Among the three studies involving androgen synthesis inhibitors (abiraterone acetate), only Albiges et al.17 reported a mean duration of response of 5.5 weeks. Similarly, among the three studies examining second-generation antiandrogens (enzalutamide), only Poole et al.18 reported a mean duration of response of 3.3 months. In contrast, among the other studies that examined first-generation antiandrogens, the mean duration of responses ranges from 3.5 to 111.5 months19,20 These are summarised in Table 1.

Mean overall survival and progression-free survival Among the three studies examining androgen synthesis inhibitors (abiraterone acetate), none of them reported OS and PFS. Among the three studies examining second-generation antiandrogens (enzalutamide), only RodriguezVida et al.21 reported an OS of 16.7 months and PFS of 6.8 months. However, among the other studies that examine first-generation antiandrogens only five report OS ranging from 42.5 weeks (~10.6 months) to 37 months.22,23 Also, two studies report a PFS of 1.9 and 3 months, respectively.23,24 These are summarised in Table 1.

PCAN_Dec_table_1.png
PCAN_Dev_table_1_cont.png

Radiological and symptomatic changes Of the studies included in this analysis, seven studies that reported radiological and clinical improvements associated with the AAW response, albeit inconsistently. Caffo et al.,25 Murakami et al.,23 Herrada et al.26 and Albiges et al.17 reported radiological improvement, Sartor et al.24 reported no change and Matsumoto et al.27 report symptomatic worsening. In addition, Herrada et al.26 also reported symptomatic improvement on top of the radiological improvement.

PSA response in all participants The overall proportion of patients that experience any PSA decline was 26% (95% CI: 20–32) with a significant heterogeneity (I 2 = 85%, p < 0.01) (Fig. 2). The overall proportion of patients that experience a ≥50% PSA decline was 11% (95% CI: 6–16), significant heterogeneity (I 2 = 89%, p < 0.01) (Fig. 3). 

Fig._2.png
Fig. 2 Forest plot of any decline in PSA. AA Antiandrogen, 95% CI 95% Confidence Interval, NA Not Applicable, df degree of freedom.

PSA response in patients on first-generation antiandrogens A subgroup analysis was performed based on the type of antiandrogen. This analysis included data from 16 studies, which reported that 1212 patients were treated with withdrawal of a first-generation antiandrogen. Overall, 30% (95% CI: 23–38%; I 2 = 85%, p < 0.01) of patients had a PSA decline after withdrawal of first-generation antiandrogens (Fig. 2). A similar subgroup analysis was performed to determine the proportion of patients experiencing ≥ 50% PSA decline. Overall, 15% (95% CI: 8–22%; I 2 = 90%, p < 0.01) of patients had ≥50% PSA decline after withdrawal of first-generation antiandrogens (Fig. 3).

Fig._3_Forest_plot.png
Fig. 3 Forest plot of ≥50% decline in PSA. AA Antiandrogen, 95% CI 95% Confidence Interval, NA Not Applicable, df degree of freedom

PSA response in patients on second-generation antiandrogens Three studies with 108 patients were treated with withdrawal of a second-generation antiandrogen (enzalutamide). Overall, 7% (95% CI: 3–13; I 2 = 76%, p = 0.02) of patients had any PSA decline after withdrawal of a second-generation antiandrogen (Fig. 2). However, only 1% (95% CI: 0–5%; I 2 = 0%, p = 0.58) of patients experience a ≥50% PSA decline after withdrawal of a second-generation antiandrogen (Fig. 3).

PSA response in patients on androgen synthesis inhibitor Three studies with 154 patients were treated with withdrawal of an androgen synthesis inhibitor (abiraterone). Overall, 26% (95% CI: 19–33; I 2 = 12%, p = 0.32) of patients had any PSA decline after withdrawal of an androgen synthesis inhibitor (Fig. 2). However, only 4% (95% CI: 0–13%; I 2 = 81%, p < 0.01) of patients experience a ≥50% PSA decline after withdrawal of an androgen synthesis inhibitor (Fig. 3).

Subgroup comparison A chi-squared analysis was performed to further examine the differences between first-generation, second-generation antiandrogens and androgen synthesis inhibitors. In the data for any PSA decline after withdrawal of the antiandrogen, the difference in the proportion of patients experiencing an AAW response was significant (chi2 = 27.41, df = 2, p < 0.05) as shown in Fig. 2. Similarly, the proportion of patients on the different types of antiandrogens experiencing ≥ 50% PSA decline was also significantly different (chi2 = 11.82, df = 2, p < 0.01) as shown in Fig. 3.

Publication bias A funnel plot was plotted with pseudo-95% confidence intervals and overall effect line as shown in Fig. 4. A visual inspection suggested no apparent asymmetry and there are an equal number of studies on either half of the overall effect line. Four studies on either side lie outside the pseudo-95% confidence interval. Funnel plot symmetry indicated publication bias was unlikely.28

Fig._4_Funnel_plot.png
Fig. 4 Funnel plot of studies for publication bias

DISCUSSION

Twenty-three studies involving 1474 men with advanced prostate cancer were included in this systematic review and meta-analysis that evaluated the PSA response to AAW. The pooled estimate of proportion with any PSA decline was 26% of patients and a ≥50% PSA decline was 11% from AAW. 

For the first time, we have demonstrated that the PSA response may depend on the type of antiandrogen. The PSA response to withdrawal of enzalutamide or abiraterone is much lower than the first-generation antiandrogen. The mechanistic reasons behind the differences in AAW response to first and second generation of antiandrogens are unclear. von Klot et al.6 have previously reported that the occurrence of AAW response in patients treated with enzalutamide was low. They hypothesised that such a low response to enzalutamide withdrawal may be due to the differences in the mechanisms of action between first- and second-generation antiandrogens. A first-generation antiandrogen has direct antagonistic effect through the blockade of AR with a contradictory agonistic effect when AR are overexpressed. On the contrary, a second-generation antiandrogen such as enzalutamide inhibits nuclear translocation of AR, inhibits AR binding to DNA and coactivator recruitment without any agonistic effect in addition to AR blockade.3 In addition, Tran et al.29 demonstrated that enzalutamide lacks agonistic activity to AR unlike their first-generation counterparts.3

Abiraterone acetate has a unique AAW response. It appears to be almost as likely to elicit an AAW response as first-generation antiandrogens. However, the magnitude of the ≥50% PSA decline to its withdrawal was lower than first-generation antiandrogens. It is likely that this is due to the difference in effect on both the levels of androgen and the AR. Abiraterone acetate is a CYP17 inhibitor and works by blocking the synthesis of androgens.3 However, some studies on prostate cancer cell lines find that abiraterone does indeed bind to both mutant and wild-type AR, altering the cytoplasmic and nuclear kinetics of AR.30,31 It is very likely that this unique action of abiraterone acetate on both the AR and androgen levels would account for the different AAW response and further research would provide clarity on this observation.

It is also possible that the molecular mechanisms that are involved in the AAW response may be different in patients who had first- or second-generation AAW. Historically, the withdrawal of first generation of antiandrogens occurred after one line of therapy, while until recently, the second-generation antiandrogens were initiated after failure of two or more lines therapies for advanced prostate cancer (usually a first-generation antiandrogen in hormone sensitive setting and a chemotherapy drug in the castrate resistant setting). The current review included studies that attempted AAW in patients who had received a median of two lines of therapies prior to the initiation of a second-generation antiandrogen. As the molecular evolution of prostate cancer in such late stages indicate the existence of AR mutations that drive resistance to antiandrogens thereby reducing the odds of response to AAW of a second-generation antiandrogen.32,33

In addition, the second generation of antiandrogens are not uniform in their mechanisms of action with enzalutamide inhibiting AR and AR responsive genes, abiraterone acetate suppresses testosterone synthesis through CYP17a inhibition.3 It is unclear if there is any difference in PSA response to the withdrawal of either of these two drugs. Direct comparison between abiraterone and enzalutamide was not possible due to limited number of studies. Further research in this area is needed.

The mechanism of CYP17 inhibitors such as abiraterone acetate adds another layer of complexity to the analysis and understanding of the AAW response. The CYP17 inhibition reduces the levels of both testosterone and cortisol. The reduction in the level of cortisol causes a rise in ACTH levels due to the feedback mechanism and ultimately causes high levels of deoxycorticosterone, leading to hypokalaemia, hypertension and fluid retention.34 To avoid this, a low-dose glucocorticoid such as prednisolone is given together with abiraterone acetate. Historically, glucocorticoids have been used to treat prostate cancer effectively, leading to a decline in PSA levels.35 In addition, prednisolone has been shown to modify the ligand-binding domain of the AR.36,37 However, as there is no evidence for a pure “glucocorticoid withdrawal response” similar to the AAW response [35], it is unlikely that prednisolone withdrawal contributed to AAW response in abiraterone treated patients. Nevertheless, with an increasing interest in the AAW response in abiraterone acetate, it is possible that more evidence will emerge in the future that will improve our understanding of the effect.

Another area of variation in the reported literature on AAW was the definition of PSA response.21,24,38,39 In the current study, the proportion of patients who had any PSA decline and ≥50% PSA decline to AAW were considered as outcome measures. As radiological and clinical responses were inconsistently reported in the included studies, these outcomes could not be evaluated here. Moreover, there is no international consensus on AAW outcomes, which could facilitate uniform assessment and reporting.

With the rapid change in the treatment paradigm for hormone sensitive prostate cancer such as early use of docetaxel, early use of second-generation antiandrogens and the availability of other therapies such as cabazitaxel, the clinical relevance of AAW and PSA response to first-generation antiandrogens require elucidation. Future studies should explore response to withdrawal of second-generation antiandrogens when administered in the hormone sensitive setting of advanced prostate cancer.

A significant heterogeneity was found in studies that treated their patients with first-generation antiandrogens with a wide 95% CI between the studies included in the review. Such heterogeneity between studies could arise from the inclusion of patients with different disease characteristics, duration of therapy, differences in patient compliance, number of prior therapies and other confounding factors such as levels of serum testosterone. Moreover, inconsistent reporting between the studies precludes retrieving enough data to adequately control for these factors in our analysis.

Some of the studies in this analysis have provided additional information on the duration of PSA response, OS and PFS. They were few in number and their inconsistent reporting among the studies has made it difficult to analyse the data in a meaningful fashion. Nevertheless, in the data extracted from these studies some emerging trends can be observed. The sole study on abiraterone acetate reported a mean duration of response that is much lower than the duration of response for these studies in first-generation antiandrogens. In comparison, the study on enzalutamide reported a mean duration of response that is slightly lower than that of their first-generation counterparts. While this does reflect the finding in this analysis that the AAW response of abiraterone acetate is of a lower magnitude than other first-generation antiandrogens, it was reported only in a single study making it difficult to draw meaningful conclusions.

While the hallmark of the AAW response is a fall in PSA levels, there have been some indications that a decline in PSA may be associated with more clinically relevant end points such as radiological and symptomatic improvement. While the reporting of the radiological and clinical effects was inconsistent among the studies, they do indicate a possibility of radiological improvement associated with the AAW response, though the symptomatic effects associated with the AAW response tend to be less consistent

These varied clinical manifestations described give the AAW response a unique place in the management of advanced prostate cancer. The AAW response is observed mainly in patients with advanced prostate cancer who have experienced treatment failure after prolonged use of an antiandrogen. In this population, treatment options are rapidly evolving and the AAW response has the potential to become a reasonable and inexpensive treatment option to provide disease control in a proportion of men with advanced prostate cancer.

To our knowledge, this is the first systematic review and meta-analysis to examine AAW in both first-generation, second-generation antiandrogens and androgen synthesis inhibitors. We acknowledge the limitations of this review: data on baseline characteristics were not uniformly available; imaging-based outcome measures and duration of response and progression free and OS outcomes were not included due to inconsistency in reporting. Limited information on prior therapies was available. Despite these limitations, the results from our study for the first time indicate that further research is warranted to examine the influence of the type of antiandrogen and prior therapy on the occurrence of AAW.

Compliance with ethical standards

Conflict of interest: The authors declare no competing interests.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Authors: Alwin Soo1 Michael E. O’Callaghan2,3,4 Tina Kopsaftis2,3 Sina Vatandoust1,5 Kim Moretti2,4,6,7 Ganessan Kichenadasse1,5

  1. School of Medicine, Flinders University, Bedford Park, SA, Australia
  2. South Australian Prostate Cancer Clinical Outcomes Collaborative, Flinders Medical Centre, Bedford Park, SA, Australia
  3. Department of Urology, Flinders Medical Centre, Bedford Park, SA, Australia
  4. Discipline of Surgery, University of Adelaide, Adelaide, SA, Australia
  5. Department of Medical Oncology, Flinders Centre for Innovation in Cancer, Flinders Medical Centre/Flinders University, Bedford Park, SA, Australia
  6. School of Population Health, University of South Australia, Adelaide, SA, Australia
  7. Monash University, Clayton, VIC, Australia

References:

  1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68:394–424.
  2. Small EJ, Halabi S, Dawson NA, Stadler WM, Rini BI, Picus J, et al. Antiandrogen withdrawal alone or in combination with ketoconazole in androgen-independent prostate cancer patients: a phase III trial (CALGB 9583). J Clin Oncol. 2004;22:1025–33.
  3. Rice MA, Malhotra SV, Stoyanova T. Second-generation antiandrogens: from discovery to standard of care in castration resistant prostate cancer. Front Oncol. 2019;9:801.
  4. Griend DJV, d’Antonio JM, Isaacs JT. Hormonal Regulation of the Prostate. In: Berges R, Tombal B, editors. Androgens and prostate cancer. Belgium: Ismar Healthcare Publishing; 2009. p. 11–34.
  5. Heinlein CA, Chang C. Androgen receptor in prostate cancer. Endocr Rev. 2004;25:276–308.
  6. Miyamoto H, Rahman MM, Chang C. Molecular basis for the antiandrogen withdrawal syndrome. J Cell Biochem. 2004;91:3–12.
  7. Feldman BJ, Feldman D. The development of androgenindependent prostate cancer. Nat Rev Cancer. 2001;1:34.
  8. Scher HI, Kelly WK. Flutamide withdrawal syndrome: its impact on clinical trials in hormone-refractory prostate cancer. J Clin Oncol. 1993;11:1566–72.
  9. von Klot CA, Kramer MW, Boker A, Herrmann TR, Peters I, Kuczyk MA, et al. Is there an anti-androgen withdrawal syndrome for enzalutamide? World J Urol. 2014;32:1171–6.
  10. Dupont A, Gomez J-L, Cusan L, Koutsilieris M, Labrie F. Response to flutamide withdrawal in advanced prostate cancer in progression under combination therapy. J Urol. 1993;150: 908–13.
  11. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg. 2010;8:336–41.
  12. MIMS Online [Internet]. MIMS Australia. 2020. Available from: http://www.mims.com.au. 
  13. Veritas Health Innovation. Covidence systematic review software. Melbourne, Australia: Veritas Health Innovation.
  14. Higgins JPT, Green S. Cochrane handbook for systematic review of interventions. The Cochrane Collaboration; 2011. Available from: www.cochrane-handbook.org.
  15. Viechtbauer W. Conducting meta-analyses in R with the metafor package. J Stat Softw. 2010;36:1–48.
  16. Vatandoust S, Kichenadasse G, O’Callaghan ME, Kopsaftis T, Walsh S, Borg M, et al. Antiandrogen withdrawal (AAWD) in patients (pts) with prostate cancer (PCa): Retrospective analysis of data from the South Australian Prostate Cancer Clinical Outcome Collaborative (SA-PCCOC). J Clin Oncol. 2015;33 7_suppl:240. PSA response to antiandrogen withdrawal: a systematic review and meta-analysis 835
  17. Albiges L, Auclin E, Rousseau B, Boughalem E, Levy A, Loriot Y, et al. Is there a withdrawal syndrome with abiraterone acetate (AA)? J Clin Oncol. 2013;31 (6_suppl):89.
  18. Poole A, Gill D, Hahn AW, Johnson E, Carroll E, Boucher K, et al. Incidence and characterization of antiandrogen withdrawal syndrome after discontinuation of treatment with enzalutamide in castration-resistant prostate cancer. Clin Genitourin Cancer. 2018;16:e169–72.
  19. Small EJ, Srinivas S. The antiandrogen withdrawal syndrome. Experience in a large cohort of unselected patients with advanced prostate cancer. Cancer. 1995;76:1428–34.
  20. Figg WD, Sartor O, Cooper MR, Thibault A, Bergan RC, Dawson N, et al. Prostate specific antigen decline following the discontinuation of flutamide in patients with stage D2 prostate cancer. Am J Med. 1995;98:412–4.
  21. Rodriguez-Vida A, Bianchini D, Van Hemelrijck M, Hughes S, Malik Z, Powles T, et al. Is there an antiandrogen withdrawal syndrome with enzalutamide? BJU Int. 2015;115:373–80.
  22. Sella A, Flex D, Sulkes A, Baniel J. Antiandrogen withdrawal syndrome with cyproterone acetate. Urology. 1998;52:1091–3.
  23. Murakami T, Obata H, Akitake N, Shiota M, Takeuchi A, Kashiwagi E, et al. Prognostic and predictive factors for antiandrogen withdrawal in castration-resistant prostate cancer. Anticancer Res. 2018;38:4115–21.
  24. Sartor AO, Tangen CM, Hussain MHA, Eisenberger MA, Parab M, Fontana JA, et al. Antiandrogen withdrawal in castrate refractory prostate cancer. Cancer. 2008;112:2393–400.
  25. Caffo O, Palermo A, Veccia A, Maines F, Chierichetti F, Berruti A, et al. Biochemical and objective response to abiraterone acetate withdrawal: incidence and clinical relevance of a new scenario for castration-resistant prostate cancer. Urology. 2013;82: 1090–3.
  26. Herrada J, Dieringer P, Logothetis CJ. Characterization of patients with androgen-independent prostatic carcinoma whose serum prostate specific antigen decreased following flutamide withdrawal. J Urol. 1996;155:620–3.
  27. Matsumoto K, Tanaka N, Hayakawa N, Ezaki T, Suzuki K, Maeda T, et al. The type of patients who would benefit from antiandrogen withdrawal therapy: could it be performed safely for aggressive prostate cancer? Med Oncol. 2013;30:647.
  28. Higgins J, Altman D, Sterne JE. Chapter 8: Assessing risk of bias in included studies. In: Higgins J, Churchill R, Chandler J, Cumpston ME, editors. Cochrane handbook for systematic reviews of interventions. Cochrane; Chichester, UK, 2017.
  29. Tran C, Ouk S, Clegg NJ, Chen Y, Watson PA, Arora V, et al. Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science. 2009;324:787–90.
  30. Giatromanolaki A, Fasoulaki V, Kalamida D, Mitrakas A, Kakouratos C, Lialiaris T, et al. CYP17A1 and androgen-receptor expression in prostate carcinoma tissues and cancer cell lines. Cur Urol. 2019;13:157–65.
  31. Richards J, Lim AC, Hay CW, Taylor AE, Wingate A, Nowakowska K, et al. Interactions of abiraterone, eplerenone, and prednisolone with wild-type and mutant androgen receptor: a rationale for increasing abiraterone exposure or combining with MDV3100. Cancer Res. 2012;72:2176–82.
  32. Davies A, Conteduca V, Zoubeidi A, Beltran H. Biological evolution of castration-resistant prostate cancer. Eur Urol Focus. 2019;5:147–54.
  33. Lorenzin F, Demichelis F. Evolution of the prostate cancer genome towards resistance. J Transl Genet Genom. 2019;3:5. https:// doi.org/10.20517/jtgg.2019.01.
  34. Yamamoto Y, Akashi Y, Minami T, Nozawa M, Kiba K, Yoshikawa M, et al. Serious hypokalemia associated with abiraterone acetate in patients with castration-resistant prostate cancer. Case Rep Urol. 2018;2018:1414395.
  35. Sartor O, Parker CC, De Bono J. Reappraisal of glucocorticoids in castrate-resistant prostate cancer. Asian J Androl. 2014;16:666.
  36. Veldscholte J, Berrevoets C, Ris-Stalpers C, Kuiper G, Jenster G, Trapman J, et al. The androgen receptor in LNCaP cells contains a mutation in the ligand binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol. 1992;41:665–9.
  37. Azad AA, Eigl BJ. Evaluation of prostate-specific antigen response following cessation of abiraterone acetate: is there evidence for a withdrawal syndrome? Eur Urol. 2014;65:504–5.
  38. Lau YK, Chadha MK, Litwin A, Trump DL. A dramatic, objective antiandrogen withdrawal response: case report and review of the literature. J Hematol Oncol. 2008;1:21.
  39. Leone G, Tucci M, Buttigliero C, Zichi C, Pignataro D, Bironzo P, et al. Antiandrogen withdrawal syndrome (AAWS) in the treatment of patients with prostate cancer. Endocr Relat Cancer. 2018;25:R1–9.
  40. Bracarda S, Procopio G, Alesini D, Grillone F, Massari F, Zaniboni A, et al. Enzalutamide activity in patients with metastatic castration resistant prostate cancer (mCRPC) previously responding to antiandrogen withdrawal syndrome (AWN): a preliminary report. J Clin Oncol. 2014;32:e16508.
  41. El Geneidy MM, Lewis G, Dainer P, Terris MK, Brown J, Coleman T. Factors predicting a response to anti-androgen withdrawal maneuvers in prostate cancer patients failing combined androgen blockade. J Clin Oncol. 2008;26:e16116.
  42. Momozono H, Miyake H, Tei H, Harada KI, Fujisawa M. Clinical outcomes of anti-androgen withdrawal and subsequent alternative anti-androgen therapy for advanced prostate cancer following failure of initial maximum androgen blockade. Mol Clin Oncol. 2016;4:839–44.
  43. Morote J, Bellmunt J. Bone alkaline phosphatase serum level predicts the response to antiandrogen withdrawal. Eur Urol. 2002; 41:257–61.
  44. Schellhammer PF, Venner P, Haas GP, Small EJ, Nieh PT, Seabaugh DR, et al. Prostate specific antigen decreases after withdrawal of antiandrogen therapy with bicalutamide or flutamide in patients receiving combined androgen blockade. J Urol. 1997;157: 1731–5.
  45. Yoneyama S, Miyoshi Y, Yasui M, Uemura K, Kawahara T, Hattori Y, et al. The evaluation of antiandrogen withdrawal syndrome after discontinuation of bicalutamide in metastatic castration-resistant prostate cancer. Int J Urol. 2017;24:38.
  46. Figg WD, McCall NA, Reed E, Sartor O. The in-vitro response of 4 antisteroid receptor agents on the hormone-responsive prostate cancer cell-line lncap. Oncol Rep. 1995;2:295–8.

Prostate Cancer and Prostatic Diseases; Received: 6 September 2020 / Revised: 8 January 2021 / Accepted: 28 January 2021 / Published online: 18 February 2021

Read an Editorial on this Article: PSA Response to Antiandrogen Withdrawal To Identify the Response on Type of Antiandrogens - Editorial