Abstract
Background
The prognostic role of androgen receptor variant 7 (AR-V7) in non-metastatic prostate cancer remains unclear. We conducted a systematic review and meta-analysis to evaluate its impact on primary treatment outcomes and disease progression.
Methods
We systematically searched PubMed, Embase, and Cochrane Central for observational studies and randomized controlled trials (RCTs) comparing AR-V7-positive and AR-V7-negative patients with non-metastatic prostate cancer. The primary outcomes were overall survival (OS) and biochemical recurrenc (BCR). Secondary outcomes, including progression-free survival (PFS), cancer-specific survival (CSS), and the incidence of castration-resistant prostate cancer (CRPC), were assessed as non-meta-analyzable endpoints. Hazard ratios (HRs) and risk ratios (RRs) with 95 % confidence intervals (CIs) were calculated using a random effects model. Statistical analyses were conducted in Review Manager software (version).
Results
Seven studies were included, comprising 1,383 patients. AR-V7 expression was associated with significantly worse OS (HR 5.75; 95 % CI 2.24–14.78; P < 0.05) but showed no significant difference in BCR between groups (RR 4.81; 95 % CI 0.88–26.35; P = 0.07). Among the non-meta-analyzable secondary outcomes, individual studies suggested that AR-V7-positive patients had a higher risk of CRPC progression, shorter CSS, and reduced PFS compared to AR-V7-negative patients, but heterogeneity in study design and reporting precluded quantitative synthesis.
Conclusion
AR-V7 expression is a negative prognostic marker in non-metastatic prostate cancer, associated with significantly reduced OS and increased CRPC progression. While no significant difference was observed in BCR, the trend suggests potential implications for disease monitoring. Further studies with standardized detection methods are needed to clarify AR-V7's role in risk stratification and treatment decisions.
Keywords: AR-V7, Biochemical recurrence, Castration-resistant prostate cancer, Meta-analysis, Overall survival, Prostate cancer
1. Introduction
Currently, prostate cancer is the most diagnosed malignant disease in men worldwide, being responsible for 1.4 million new cases in 2020, equivalent to 15.2 % of all types of cancer in the male population.1 Prostate cancer risk stratification is performed into groups with a progression risk profile, using criteria based on T stage, Gleason score, prostate-specific antigen (PSA), PSA density, and number and percentage of positive fragments in the biopsy.2 However, this prostate cancer stratification is challenging due to a significant rate of misclassification (approximately 36 %) observed in biopsies compared to the definitive pathological findings from radical prostatectomy.3
The role of variants of androgen receptor expression, particularly androgen receptor variant 7 (AR-V7), in various physiological processes, such as prostate development and growth,4 remains a source of controversy and debate. Several robust cohorts have been conducted to establish if AR-V7 may play a pivotal role in worse clinicopathological outcomes in metastatic prostate tumors and higher prostate cancer-specific mortality given its endocrine resistance.5, 6, 7 However, there is a gap in the literature regarding a comprehensive synthesis of whether AR-V7's role in non-metastatic prostate cancer patients, particularly in relation to primary treatment outcomes and disease progression after initial therapy.8
In light of this, we aimed to compare AR-V7-positive and AR-V7-negative non-metastatic prostate cancer patients to synthesize current evidence on the role of AR-V7 in primary treatment outcomes and its association with disease progression after initial therapy.
2. Methods
This systematic review with meta-analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under protocol CRD42025642461. It was designed and conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines.9
2.1. Study eligibility
Inclusion in this meta-analysis was restricted to studies that met all the following eligibility criteria:1 observational or randomized controlled trial (RCT);2 patients with non-metastatic prostate cancer;3 studies comparing cytoplasmic V7+ and V7- non-metastatic prostate cancer;4 studies available in English and full-text; and10 studies that reported any of the clinical outcomes of interest. We excluded from this analysis studies:1 with no outcome of interest and2 overlapping study populations.
2.2. Search strategy and data extraction
We systematically searched PubMed, Embase, and Cochrane Central Register of Controlled Trials from inception to December 2024 with a predefined search strategy (Supplementary Table 1). In addition, the references of included studies, reviews, and meta-analyses were evaluated for additional studies. Two authors independently extracted the data (MQA and HSF) following predefined search criteria and quality assessment. Disagreements were resolved by consensus.
2.3. Endpoints
Primary and meta-analyzed outcomes were1 overall survival (OS) and2 biochemical-free recurrence (BCR). Secondary and non-meta-analyzed outcomes were1 biochemical and radiological progression after radical prostatectomy;2 progression-free survival (PFS);3 cancer-specific survival (CSS);4 time to progression to castration-resistant prostate cancer;10 incidence of castration-resistant prostate cancer progression;11 proportion of prostate cancer-related deaths, and12 determination of antibody specificity.
2.4. Definitions
We defined OS as the time from surgery to death due to any cause. Deaths, whether disease-specific or not, were considered as events. Patients alive at their last follow-up visit during the observation period were categorized as alive (no event). One study defined BCR as PSA greater than 0.2 ng/mL,13 whereas another did not define BCR.14
2.5. Quality assessment
Two authors (MQA and HSF) independently evaluated the methodological quality of all included studies. We assessed the risk-of-bias in all studies using the Robins-I for observational studies.15 Disagreements were resolved by consensus. Publication bias was not assessed with funnel-plot analysis and Egger's test of the primary endpoints to evaluate the symmetric distribution of trials with similar weights since we had less than 10 studies.16
2.6. Statistical analysis
Comparative meta-analyses were done using hazard ratios (HR) and treatment effects for binary endpoints were compared using pooled risk ratios (RRs), both with 95 % CI and under a random effects model. Heterogeneity was assessed using I2 statistics, where I2 > 40 % was considered significant;16 p values inferior to 0.05 were considered statistically significant. The statistical analysis was performed using R 4.3.0 (R Core Team, 2023) with the meta package, employing the DerSimonian-Laird and the Generic Inverse methods.
3. Results
3.1. Study selection and baseline characteristics
As detailed in Fig. 1, our research yielded 1,383 studies from the Pubmed, Embase, and Cochrane Central databases. After excluding duplicates and studies not meeting the eligibility criteria, 80 studies were assessed in full text. Seven observational studies were included in the meta-analysis, comprising 1,383 patients with non-metastatic prostate cancer.13,14,17, 18, 19, 20, 21, 22 The mean patient age was 65.4 years. The median follow-up period varied across studies, with most ranging from 12 to 82 months. Among the included studies, AR-V7 positivity was associated with higher Gleason scores, advanced pathological stage, and increased nodal involvement (Table 1) (see Table 2).
Fig. 1.
PRISMA 2020 flow diagram for included studies.
Table 1.
Baseline characteristics of included studies
| Study | Patients, n, I/C | Age, y, I/C | GS, n, (%), I/C | Pathological tumor stage (n, %), I/C | Pathological lymph node stage (n,%), I/C | Metastatic pathological stage (n, %), I/C |
|---|---|---|---|---|---|---|
| Ouyang, et al. 2021 | 21/89 | 65.38 (5.35)/66.17 (6.31)a | GS, n, (%) ≤7 14 (66.7)/56 (62.9) GS, n, (%) ≥8 7 (33.3)/33 (37) |
Pathological stage n, (%) 2 14(66.67)/51(57.30) Pathological stage n, (%) 3 5(23.81)/32(35.96) Pathological stage n, (%) 4 2(9.52)/6(6.74) |
Pathological lymph node stage n, (%) 0 17(80.95)/62(69.66) Pathological lymph node stage n, (%) 1 4(19.05)/27(30.34) |
NA |
| Heng Li, et al. 2018 | 32/136 | 71(58-79)/72 (53-96)b | GS, n, (%) ≤7 6 (18.8)/54 (39.7) GS, n, (%) ≥8 26 (81.3)/82 (60.3) |
Pathological stage n, (%) ≤2 2(6.3)/14(10.3) Pathological stage n, (%) 3 8(25)/41(30.1) Pathological stage n, (%) >4 16(50)/49(36) Pathological stage n, (%) TX 6(18.8)/32(23.5) |
Pathological lymph node stage n, (%) 0 12(37.5)/32(23.5) Pathological lymph node stage n, (%) 1 10(31.3)/66(48.5) Pathological lymph node stage n, (%) X 10(31.3)/38(27.9) |
Pathological lymph node stage n, (%) 0 4(12.5)/32(23.5) Pathological lymph node stage n, (%) 1 26(81.3)/92(67.6) Pathological lymph node stage n, (%) X 2(6.3)/12(8.8) |
| Kaczorowskia, et al. 2020 | 34 to 40/123 to 129c | NA | NA | NA | NA | NA |
| Paul König, et al. 2020 | 347/59 | 65 (45 83) (61 69)d | GS, n, (%) 6 213 (52.0) GS, n, (%) 7a 78 (19.0) GS, n, (%) 7b 30 (7.3) GS, n, (%) 8 27 (6.6) GS, n, (%) 9-10 34 (8.3) GS 28 (6.8) |
Pathological stage n, (%) 2 236 (57.6) Pathological stage n, (%) 3 144 (35.1) Pathological stage n, (%) 4 30 (7.3) |
NA | NA |
| Xin Chen, et al. 2018 | 163 predominantly high-risk patients (cohort I) Primary tumors cohort (238, cohort II) |
NA | Cohort I consisted of predominantly a GS of ≥8/≥4 in over 70 % of patients. In cohort II, over 70 % of patients had a GS of 6 or 7 (3 þ 4). | Cohort I consisted of predominantly high-risk patients with over 80 % showing a tumor pathological stage of ≥3. Cohort II represents an unselected consecutive prostatectomy cohort with mostly low- and intermediate-risk patients. | Cohort I consisted of predominantly over 70 % with lymph node metastases. Cohort II only 2.5 % presented with lymph node metastasis. | NA |
| Sciarra, et al. 2021 | 30/26 | NA | NA | NA | NA | NA |
| Zhao, et al. 2012 | Mixed grade cohort of 53 men with prostate cancer | AR-V7 was positively correlated with age | 30 % to 70 % of patients had GS 4/5 | NA | NA | NA |
| Study | Surgical Margin (n,%) I/C |
Preoperative TPSA (ng/ml) I/C |
Prostate volume (cm3) I/C |
PSAD (ng/ml/cm3) I/C |
Follow-up (mean,SD,months) | Country |
|---|---|---|---|---|---|---|
| Ouyang, et al. 2021 | 17(80.95)/74(83.15) | 58.32.44(81.76)/41.19(52.02)a | 60.90(18.15)/74.30(36.39)a | 1.29(2.29)/0.60(0.63)a | 57.67(9.24)/57.55(10.90) | China |
| Li1, et al. 2018 | NA | 221.04(34.09–2507.10)/113.40(2.61–4003.40)b | 100.42(40.26–172.80)/108.51(31.45–719.71)b | 3.36(0.38–23.90)/0.97(0.02–62.63)b | NA | China |
| Kaczorowskia, et al. 2020 | NA | NA | NA | NA | NA | Germany |
| König, et al. 2020 | NA | 3.86 (0.99-7.59) | NA | NA | NA | United States of America |
| Chen, et al. 2018 | NA | NA | NA | NA | NA | Germany |
| Sciarra, et al. 2021 | NA | NA | NA | NA | NA | Italia |
| Zhao, et al. 2012 | NA | NA | NA | AR-V7 was inversely correlated with serum prostate-specific antigen | NA | United States of America |
Mean (standard deviation).
Median (range); cm, centimeter; GS, Gleason Score; I/C, intervention/control groups; ml, millimeter; NA, not available; ng, nanogram; PSDA, Prostate-Specific Antigen Density; TPSA, Total Prostate-Specific Antigen.
The overall AR-V7 positivity rate varied depending on the antibody used by the study: 24.9 % (AG10008) and 21 % (RM7).
Overall median (IQR) age since the study did not provide separate baseline data according to V7 expression.
Table 2.
Non-meta analyzed outcomes
| Study | Patients, n, I/C | PFS | CSS | OS | Time to progression to CRPC (mo, I/C) | Risk of bias |
|---|---|---|---|---|---|---|
| Ouyang, et al. 2021 | 21/89 | HR: 4.26; 95 % CI: 1.55–11.68; P = 0.003. AR-V7-positive patients had significantly worse progression-free survival compared to AR-V7-negative patients. | HR: 22.47; 95 % CI: 2.91–173.4; P = 0.003. Kaplan–Meier analysis indicated significantly lower CSS in AR-V7-positive patients, particularly in metastatic cases (P < 0.0001). | HR: 6.61; 95 % CI: 1.40–31.20; P = 0.017. Kaplan–Meier analysis showed significantly lower OS in AR-V7-positive patients, especially in metastatic cases (P < 0.0001). | NA | Moderate risk of bias |
| Li1, et al. 2018 | 32/136 | NA | Kaplan–Meier analyses indicated that cancer-specific was dramatically lower in AR-V7-positive patients, as in the stratified patients with metastases (both P < 0.0001) | Kaplan–Meier analyses indicated that overall survival was dramatically lower in AR-V7-positive patients, as in the stratified patients with metastases (both P < 0.0001) | AR-V7-positive: 12 months (range: 6–26); AR-V7-negative: 25 months (range: 5–82); P < 0.0001. AR-V7-positive patients progressed significantly faster than AR-V7-negative patients. | Serious risk of bias |
| Kaczorowskia, et al. 2020 | Primary tumors of 163 prostate cancer patients. | NA | NA | NA | NA | Moderate risk of bias |
| König, et al. 2020 | Prostate tumor sections from a cohort of 410 patients | NA | NA | The survival analysis showed that AR staining in the nucleus or in the cytoplasm was not associated with OSS. In addition, AR-V7 staining in both the cytoplasm and the cytoplasmic granules was not associated with OSS. However, positive AR-V7 staining in the cytoplasm was a negative Prognostic marker for RFS in univariate analysis, |
NA | Serious risk of bias |
| Chen et al. 2018 | 163 predominantly high-risk patients (cohort I) 238 patients with primary tumors (cohort II) |
NA | NA | NA | NA | Moderate risk of bias |
| Sciarra, et al. 2021 | 30/26 | NA | NA | NA | NA | Serious risk of bias |
| Zhao, et al. 2012 | Mixed grade cohort of 53 men | NA | NA | NA | NA | Serious risk of bias |
| Study | Incidence of castration-resistant prostate cancer progression | Proportion of prostate cancer-related deaths | Determination of antibody specificity | Risk of bias |
|---|---|---|---|---|
| Ouyang, et al. 2021 | NA | NA | NA | Moderate risk of bias |
| Li1, et al. 2018 | Among AR-V7-positive patients, 88 % (95 % CI: 75–100 %) experienced progression to castration-resistant prostate cancer, compared to 28 % (95 % CI: 20–36 %) in AR-V7-negative patients. | A total of 44 out of 168 patients (26 %) died of prostate cancer. Among AR-V7-positive patients, 14 out of 32 (44 %) experienced prostate cancer-related deaths. In contrast, 30 out of 136 AR-V7-negative patients (22 %) died due to prostate cancer | NA | Serious risk of bias |
| Kaczorowskia, et al. 2020 | NA | NA | The overall rate of positivity (i.e., weak, moderate or strong) was comparable between AG10008 (24.9 %) and RM7 (21 %). Identical positive staining intensity in only 34 of 484 patients (7 %) | Moderate risk of bias |
| König, et al. 2020 | NA | NA | NA | Serious risk of bias |
| Chen et al. 2018 | NA | NA | NA | Moderate risk of bias |
| Sciarra, et al. 2021 | NA | NA | NA | Serious risk of bias |
| Zhao, et al. 2012 | NA | NA | NA | Serious risk of bias |
CRPC, castration-resistant prostate cancer; CSS, cancer-specific survival; HR, hazard ratios; I/C, intervention/control groups; NA, not available; OS, overall survival; P, P-value; PFS, progression-free survival.
3.1.1. BCR
Our pooled analysis showed that there was no difference between groups in BCR (RR 4.81; 95 % CI 0.88-26.35; I2 = 69 %; Fig. 2) (see Fig. 3). The results regarding BCR were heterogeneous across studies. Ouyang et al. 2021 reported that AR-V7-positive patients had a significantly higher risk of BCR compared to AR-V7-negative patients, with a hazard ratio (HR) of 5.75 (95 % CI: 2.24–14.78; P < 0.05).17 Similarly, Kaczorowskia et al. 2020 found that AR-V7-positive status was associated with an unfavorable BCR-free survival (P = 0.047, log-rank test).20 In contrast, some studies observed trends without statistical significance. For example, Sciarra et al. 2021 reported that 17 out of 30 AR-V7-positive patients experienced BCR compared to only 1 out of 26 AR-V7-negative patients, though the small sample size limited the robustness of these findings.13
Fig. 2.
OS was significantly increased in AR-V7+ patients compared to AR-V7-.
Fig. 3.
There was no significant difference between groups in biochemical recurrence (BCR).
3.1.2. OS
Our analysis showed that OS was significantly increased in AR-V7+ patients compared to AR-V7- patients (HR 5.75; 95 % CI 2.24-14.78; P < 0.05; I2 = 0 %; Fig. 1). Several studies reported on OS, with a consistent trend of worse outcomes in AR-V7-positive patients compared to AR-V7-negative patients. For example, Li et al. 2018 observed significantly lower overall survival in AR-V7-positive patients, with a hazard ratio (HR) of 6.61 (95 % CI: 1.40–31.20; P = 0.017).20 Similarly, Ouyang et al. 2021 reported an HR of 4.26 (95 % CI: 1.55–11.68; P = 0.003), indicating that AR-V7 expression was associated with a markedly increased risk of mortality.17 Kaplan–Meier analyses in other studies reinforced these findings. For instance, König et al. 2020 found that 83.4 % of AR-V7-negative patients were alive at follow-up, compared to only 16.6 % of AR-V7-positive patients.18
3.2. Secondary and non-meta-analyzed endpoints
3.2.1. PFS
PFS was reported heterogeneously across studies. Ouyang et. al 2021 indicated a hazard ratio (HR) of 4.26 (95 % CI, 1.55 to 11.68; P = 0.003) favoring worse outcomes in AR-V7-positive patients compared to AR-V7-negative patients.17 However, other studies did not provide sufficient detail to allow for pooled analysis.
3.2.2. CSS
CCS was significantly lower in AR-V7-positive patients among two included studies.17,21 Ouyang et. al 2021 reported an HR of 22.47 (95 % CI, 2.91 to 173.4; P = 0.003), highlighting a marked reduction in survival in AR-V7-positive patients compared to their counterparts.17 Similarly, Li et. al 2018 demonstrated that CCS was dramatically lower in AR-V7-positive patients, as in the stratified patients with metastases (both P < 0.0001).21
3.2.3. Time to Progression to Castration-Resistant Prostate Cancer
The median time to progression to castration-resistant prostate cancer varied between studies. Li et. al 2018 showed that AR-V7-positive patients had a median time of 12 months (range: 6–26 months), compared to 25 months (range: 5–82 months) in AR-V7-negative patients (P < 0.0001).21 However, other studies did not provide sufficient detail to allow for pooled analysis.
3.2.4. Incidence of Castration-Resistant Prostate Cancer Progression
Li et. al 2018 showed that the incidence of progression to castration-resistant prostate cancer was substantially higher in AR-V7-positive patients. In this study, 88 % of AR-V7-positive patients experienced progression (95 % CI: 75–100 %), compared to only 28 % of AR-V7-negative patients (95 % CI: 20–36 %).21 The remaining studies did not provide sufficient detail to allow for pooled analysis.
3.2.5. Incidence of Prostate Cancer-Related Deaths
Li et. al 2018 showed that the proportion of prostate cancer-related deaths was also higher among AR-V7-positive patients. This study reported that 44 % of AR-V7-positive patients died from prostate cancer, compared to 22 % of AR-V7-negative patients.21
3.2.6. Determination of Antibody Specificity
Kaczorowskia et al. 2020 reported comparable positivity rates between antibodies AG10008 (24.9 %) and RM7 (21 %), but the agreement between these antibodies was limited, with identical staining observed in only 7 % of cases. Differences were attributed to factors such as epitope recognition and the use of monoclonal antibodies from rabbits versus mice. The variability underscores the absence of a standardized method for AR-V7 detection and its potential impact on study findings.20
3.3. Quality assessment
The ROBINS-I tool was applied to assess seven scientific articles across seven bias domains. In Domain 1, five studies presented a moderate risk,13,14,17,20,22 while three had a serious risk of bias.18,19,21 Domain 2 showed a lower risk profile, with four studies categorized as low risk,13,14,20,22 three as moderate,17,19,21 and one as serious.18 In Domain 3, most studies (five) were classified as moderate risk,14,17,20, 21, 22 with two rated as low risk13,19 and one as serious.18 Similarly, Domain 4 had five studies with a moderate risk, 13, 14, 17, 18, 20) two with low risk,17,19 and one with serious risk.21 For Domain 5, three studies were deemed moderate risk,17,21,22 three low risk,13,14,17 and two serious risk.18,19 In Domain 6, five studies had a moderate risk,13,14,17,20,21 two a serious risk,18,19 and one a low risk.22 Finally, in Domain 7, five studies were categorized as low risk,14,17,19,20,22 while three presented a serious risk.13,18,21 Following the ROBINS-I methodology, four studies were classified as having a serious risk of bias, three as moderate, and one as low risk. A detailed breakdown of the individual risk assessment is provided in Supplementary Table 2.
4. Discussion
In this systematic review with a meta-analysis of seven studies and 1,383 patients involved, we analyzed the role of expression of AR-V7 in primary PC as a prognostic marker in this population. The main findings from the combined analyses were:1 there is a trend in biochemical recurrence in primary prostate cancer patients with expression of AR-V7 compared with those without expression of AR-V7;2 overall survival was significantly higher in primary prostate cancer AR-V7- patients compared with primary prostate cancer AR-V7+ patients (HR 5.75 2.24-14.78 P = 0.0003).
BCR of prostate cancer (PC) occurs when the PSA level increases after treatment, and it is a widely used measure for monitoring patients with PC undergoing treatment.23 Although our study showed no significant differences between groups for BCR, there was a trend toward AR-V7+ patients with PC. (RR 4.81; P = 0.07). Sciarra et al. showed a significant correlation between AR-V7+ and BCR after radical prostatectomy.13 However, it is important to note that, from the 30 AR-V7 positive patients, 24 of these cases were intermediate or high risk, which could increase the incidence of BCR in this group of patients.13 On the other hand, Xin Chen et al. divided the patients in two cohorts: one with predominantly high-risk patients (cohort I) and the other one mostly with low- and intermediate-risk patients (cohort II).14 This study showed that AR-V7 expression was associated with an unfavorable BCR-free survival in a predominantly high-risk patient cohort (I) but failed to show prognostic significance in an unselected patient cohort (II). Therefore, these results underscore that AR-V7 may have an independent prognostic impact on the BCR-free survival of high-risk prostate cancer patients.
OS is a crucial indicator for evaluating cancer treatment efficacy, as it is accurate, easy to measure, reliable, and unambiguous, with no need for investigator interpretation.24 Studies have demonstrated the prognostic significance of AR-V7 expression in prostate cancer tissues.25 Similarly, previous literature has analyzed samples from men with metastatic castration-resistant PC and reported that AR-V7-positive patients had shorter OS than AR-V7-negative patients.26 Additionally, a previous meta-analysis of eight trials analyzing the role of AR-V7 expression in patients with metastatic prostate cancer concluded that OS was significantly better in AR-V7-negative patients compared to AR-V7-positive patients27 In line with these findings, our study also demonstrated that OS was significantly reduced among AR-V7-positive patients compared to AR-V7-negative patients with primary non-metastatic PC.
Heng Li et al. and Adam Kaczorowski et al. demonstrated that most AR-V7-positive patients had a Gleason score of 8 or higher, a pT stage of 3 or higher, and nodal involvement, which may explain the reduced OS observed in this group.20,21 Additionally, Konig et al. reviewed studies on AR-V7 expression in PC and noted that most reports focused on nuclear staining, with little emphasis on cytoplasmic AR-V7 staining.18 This was likely because these studies analyzed PC patients who had undergone neoadjuvant hormone therapy or had highly aggressive disease.28 A key limitation in the current literature is the lack of clear stratification of AR-V7-positive tumors by Gleason score, tumor stage, or nodal involvement. This makes direct comparisons across studies challenging and may contribute to discrepancies in reported outcomes.
AR-V7 is more commonly expressed in high-grade tumors, which are independently associated with poor prognosis.29 This confounding factor can affect the observed outcomes without proper characterization of the study population according to tumor risk groups.30 Zhao et al. analyzed radical prostatectomy specimens, providing a more detailed characterization of PC patients than previous studies.19 By focusing exclusively on patients with high-grade PC —who inherently have a poorer prognosis—the authors did not find an association between AR-V7 expression and Gleason grade or PSA recurrence, contrasting with prior studies.13,14,17,18,20, 21, 22 For instance, Li et al. and Kaczorowski et al. reported that AR-V7 is more frequently expressed in tumors with Gleason scores ≥8, advanced stages (pT3 or higher), and nodal involvement, all of which are linked to worse survival outcomes.20,21 Additionally, Konig et al. highlighted that most studies investigating AR-V7 in primary prostate cancer emphasize nuclear staining, often overlooking cytoplasmic expression, which may influence prognostic interpretations.18 The lack of clear stratification by tumor stage and Gleason grade in some analyses further complicates cross-study comparisons, potentially contributing to inconsistencies in the reported outcomes. While Zhao et al. conducted a rigorous evaluation, methodological heterogeneity across studies highlights the need for standardized approaches to assess AR splice variants and their relationship with established prognostic factors.
Across all non-meta-analyzed endpoints— PFS, CSS, time to progression to CRPC, incidence of CRPC, and prostate cancer-related mortality—individual studies suggested that outcomes were worse in AR-V7-positive patients. A growing body of evidence supports AR-V7 as a prognostic biomarker in metastatic CRPC31 and our findings align with this perspective in non-metastatic prostate tumors (referências na tabela). Similarly, Li et al. demonstrated that PFS was significantly better in AR-V7-negative patients. Furthermore, some studies indicated a tendency toward, compared to AR-V7-negative cases, prostate CSS and cancer-related mortality worsened in AR-V7-positive patients treated with enzalutamide.32
The lack of a standardized method for AR-V7 detection is a key factor contributing to the heterogeneity observed across studies. Kaczorowskia et al. highlighted that identifying AR-V7 protein using antibody-based techniques, such as immunohistochemistry, is highly dependent on antibody specificity. Additionally, these methods do not provide direct insights into AR-V7 functionality, which is influenced by cofactors.20 Their study detected nuclear AR-V7 expression in a subset of patients with primary prostate cancer before long-term hormonal manipulation and castration resistance. However, only one antibody stain showed prognostic significance, being associated with shorter BCR. Differences in detection rates may stem from the fact that rabbit-derived antibodies generally recognize more epitopes per antigen than their murine counterparts, leading to variability in protein localization and detection. Moreover, they observed a trend toward greater staining heterogeneity with increasing AR-V7 transcript levels relative to full-length AR, suggesting that AR-V7 expression and activity may not follow a linear pattern and could be modulated by cofactors.20
Most studies on AR-V7 have focused on metastatic prostate cancer,33 demonstrating its association with reduced cancer-specific and overall survival, as well as increased resistance to androgen receptor signaling inhibitors like abiraterone.34 However, relatively few studies have investigated the prognostic role of AR-V7 in non-metastatic prostate cancer, and those available exhibit significant methodological heterogeneity regarding detection techniques (antibody specificity, staining methods, and cellular localization).21 Furthermore, many studies lack a clear stratification of patients based on established risk factors, such as Gleason score and TNM stage, leading to inconsistent findings and divergent conclusions.31 Our meta-analysis significantly updates the literature on AR-V7 as a potential prognostic biomarker in primary prostate cancer, underscoring the need for standardized detection methods—particularly regarding antibody selection and cellular localization—and improved risk stratification in future research. While biochemical recurrence showed no significant difference in our analysis, overall survival was significantly worse in AR-V7-positive patients, suggesting that stronger prognostic biomarkers may emerge with further refinement in detection methods. Given the biological significance of AR splice variants, this pioneering meta-analysis highlights the importance of integrating AR-V7 into the broader androgen signaling axis in treatment-naïve patients. Our findings pave the way for further investigation into the predictive role of AR-V7, which could ultimately aid in treatment selection and risk assessment in non-metastatic prostate cancer.
In summary, intra- and inter-tumor heterogeneity makes PCa management a challenging task. It has been reported, for example, that androgen receptor-axis-targeted (ARAT) agents are less effective when AR-V7 is expressed.35 Androgen-deprivation therapy has historically been recommended as the primary therapeutic approach for metastatic hormone-sensitive prostate cancer.36 However, despite often initially responding positively to ADT, metastatic hormone-sensitive prostate cancer often progresses to a more aggressive subtype within 2–3 years.36
AR-V7, appears to be an intrinsic resistance factor to androgen deprivation therapy and androgen receptor inhibitors. It could act as a negative predictive marker for response to these therapies, but, paradoxically, its presence may signal a more aggressive disease requiring the possibility of immediate intervention. In light of this, we emphasize the need for larger prospective studies with standardized AR-V7 detection methods to validate this potential therapeutic implication.
4.1. Limitations
This meta-analysis has important limitations. First, there is notable heterogeneity among the included studies, with most focusing on patients with high-risk tumors, which limits the applicability of findings to broader patient populations. Second, many studies lacked stratification based on Gleason score or TNM staging, complicating comparisons of AR-V7 expression across similar risk groups and hindering assessment of AR-V7 as an independent prognostic factor. Additionally, the absence of a gold standard for AR-V7 detection methods introduces variability in sensitivity and specificity, affecting the consistency of findings. Finally, a limitation of our meta-analysis is that only BCR and OS were meta-analyzable. Other relevant endpoints, such as PFS, CSS, CRPC, and the incidence of CRPC could not be analyzed due to inconsistent or incomplete reporting across studies. Additionally, given the potential of AR-V7 in patient diagnosis and management, further research is needed to better define its prognostic role, particularly through standardized detection methods and larger, well-stratified cohorts.
5. Conclusion
This meta-analysis highlights the potential of AR-V7 expression as a prognostic biomarker in primary prostate cancer. While it does not significantly affect BCR, AR-V7-positive patients experience significantly reduced OS, suggesting its role as an adverse prognostic marker. Additional outcomes, including PFS and CSS also showed worse outcomes in AR-V7-positive patients. Variability in detection methods and patient stratification remain challenges in current research. Further studies are needed to refine detection techniques and explore AR-V7's clinical utility in treatment decisions for non-metastatic prostate cancer.
Sources of fundings
This study received no funding.
Conflicts of interest
All authors report no relationships that could be construed as a conflict of interest. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.prnil.2025.12.002.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.World health organization (WHO) Cancer today: data visualization tools for exploring the global cancer burden in 2020. Disponível em. www.who.int Acesso em: [data de acesso]
- 2.Matoso A.M., Epstein J.I. Defining clinically significant prostate cancer on the basis of pathological findings. Histopathology. 2019;74(1):134–135. doi: 10.1111/his.13712. [DOI] [PubMed] [Google Scholar]
- 3.Epstein J.I., Feng Z., Trock B.J., Pierorazio P.M. Upgrading and downgrading of prostate cancer from biopsy to radical prostatectomy: incidence and predictive factors using the modified Gleason grading system and factoring in tertiary grades. Eur Urol. 2012;61:1019–1024. doi: 10.1016/j.eururo.2012.01.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gobinet J.N., Poujol N., Sultan C.H. Molecular action of androgens. Mol Cell Endocrinol. 2002;198(1–2):15–24. doi: 10.1016/s0303-7207(02)00364-7. [DOI] [PubMed] [Google Scholar]
- 5.Watson P.A., Arora V.K., Buchbinder M. Androgen receptor signaling in castration-resistant prostate cancer: resistance mechanisms and therapeutic approaches. Nat Rev Cancer. 2015;15(12):701–711. doi: 10.1038/nrc4016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Antonarakis E.S., Lu C., Wang H., Lardas M., Matzuk A., Cheng C., et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371(11):1028–1038. doi: 10.1056/NEJMoa1315815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hornberger J., Shim J., Liu G., Aldape A. AR-V7 in advanced prostate cancer: implications for treatment choices. Oncology. 2019;33(4):124–130. [Google Scholar]
- 8.Zhang X., Morris B., Nguyen L. Androgen receptor variants and prostate cancer: new insights into molecular mechanisms and clinical implications. Oncogene. 2020;39:7135–7150. [Google Scholar]
- 9.Page M.J., McKenzie J.E., Bossuyt P.M., Boutron I., Hoffmann T.C., Mulrow C.D., et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Br Med J. 2021;372(71) doi: 10.1186/s13643-021-01626-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lander E.S., Linton L.M., et al. Initial sequencing and analysis of the human genome. Nature. 2001;409(6822):860–921. doi: 10.1038/35057062. [DOI] [PubMed] [Google Scholar]
- 11.Shafi A.A., Putluri V., Arnold J.M., Tsouko E., Maity S., Roberts J.M., et al. Differential regulation of metabolic pathways by androgen receptor (AR) and its constitutively active splice variant, AR-V7, in prostate cancer cells. Oncotarget. 2015;6(31):31997–32012. doi: 10.18632/oncotarget.5585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Smith A., Jones B. Role of androgens in the development and progression of prostate cancer. Urol Res. 2018;46(6):387–402. [Google Scholar]
- 13.Sciarra A., Maggi M., Salciccia S., Nicolai A., Tortorella E., Giantulli S., et al. Tissue expression of androgen receptor splice variant 7 at radical prostatectomy predicts risk of progression in untreated nonmetastatic prostate cancer. Urol Oncol: Semin Original Invest. 2018;36:161.e19–161.e30. doi: 10.1159/000512445. [DOI] [PubMed] [Google Scholar]
- 14.Chen X., Bernemann C., Tolkach Y., Heller M., Nientiedt C., Falkenstein M., et al. Overexpression of nuclear AR-V7 protein in primary prostate cancer is an independent negative prognostic marker in men with high-risk disease receiving adjuvant therapy. Oncology. 2021;99:251–255. doi: 10.1016/j.urolonc.2017.11.001. [DOI] [PubMed] [Google Scholar]
- 15.Ac Sterne J., Hernán M.A., Reeves B.C., Savović J., Berkman N.D., Viswanathan M., et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. The BMJ. 2016;355 doi: 10.1136/bmj.i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Higgins Julian P.T. Vol. 2. John Wiley & Sons; Chichester: 2019. (Cochrane handbook for systematic reviews of interventions). [Google Scholar]
- 17.Ouyang W., Zhang Y., Long G., Sun G., Liu M., Li F., et al. Androgen receptor splice variant 7 detected by immunohistochemistry is an independent poor prognostic marker in men receiving adjuvant androgen-deprivation therapy after radical prostatectomy. Biomark Res. 2021;9:23. doi: 10.1186/s40364-021-00276-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.König P., Eckstein M., Jung R., Abdulrahman A., Guzman J., Weigelt K., et al. Expression of AR-V7 (Androgen Receptor Variant 7) protein in granular cytoplasmic structures is an independent prognostic factor in prostate cancer patients. Biomark Res. 2021;9 doi: 10.3390/cancers12092639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhao H., Coram M.A., Nolley R., Reese S.W., Young S.R., Peehl D.M. Transcript levels of androgen receptor variant AR-V1 or AR-V7 do not predict recurrence in patients with prostate cancer at indeterminate risk for progression. J Urol. dez. 2012;188:2158–2164. doi: 10.1016/j.juro.2012.08.014. [DOI] [PubMed] [Google Scholar]
- 20.Kaczorowski A., Chen X., Herpel E., Merseburger A.S., Kristiansen G., Bernemann C., et al. Antibody selection influences the detection of AR-V7 in primary prostate cancer. Cancer Treat Res Commun. 2020;24 doi: 10.1016/j.ctarc.2020.100186. [DOI] [PubMed] [Google Scholar]
- 21.Li H., Wang Z., Xiao W., Yan L., Guan W., Hu Z., et al. Androgen-receptor splice variant-7-positive prostate cancer: a novel molecular subtype with markedly worse androgen-deprivation therapy outcomes in newly diagnosed patients. Mod Pathol. 2018;31:198–208. doi: 10.1038/modpathol.2017.74. [DOI] [PubMed] [Google Scholar]
- 22.Lieb V., Abdulrahman A., Weigelt K., Hauch S., Gombert M., Guzman J., et al. Cell-free DNA variant sequencing using plasma and AR-V7 testing of circulating tumor cells in prostate cancer patients. Cells. 2021;10:3223. doi: 10.3390/cells10113223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.National cancer institute Biochemical recurrence. Dictionary of cancer terms. https://www.cancer.gov/publications/dictionaries/cancer-terms/def/biochemical-recurrence Available at:
- 24.Fiteni F., Westeel V., Pivot X., Borg C., Vernerey D., Bonnetain F. Title of the article. J Visc Surg. Feb. 2014;151(1):17–22. doi: 10.1016/j.jviscsurg.2013.10.001. [DOI] [PubMed] [Google Scholar]
- 25.Yuanyuan Qu, Dingwei Ye, Bo Dai, Yunyi Kong, Kun Chang, Chengyuan Gu, et al. Impact of AR-V7 expression on overall survival for patients with metastatic prostate cancer. Zhonghua Wai Ke Za Zhi. Aug. 2014;52(8):622–626. [PubMed] [Google Scholar]
- 26.Scher H.I., Lu D., Schreiber N.A., Louw J., Graf R.P., Vargas H.A., et al. Association of AR-V7 on circulating tumor cells as a treatment-specific biomarker with outcomes and survival in castration-resistant prostate cancer. JAMA Oncol. nov. 2016;2(11):1. doi: 10.1001/jamaoncol.2016.1951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Li H., Wang Z., Tang K., Zhou H., Liu H., Yan L., et al. Prognostic value of androgen receptor splice variant 7 in the treatment of castration-resistant prostate cancer with next generation androgen receptor signal inhibition: a systematic review and meta-analysis. Eur Urol Focus. jul. 2018;4(4):529–539. doi: 10.1016/j.euf.2017.01.004. [DOI] [PubMed] [Google Scholar]
- 28.Lu C., Brown L.C., Antonarakis E.S., Armstrong A.J., Luo J. Androgen receptor variant-driven prostate cancer II: advances in laboratory investigations. Prostate Cancer Prostatic Dis. 2020;23(3):381–397. doi: 10.1038/s41391-020-0217-3. set. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Antonarakis E.S., Lu C., Wang H., Luber B., Nakazawa M., Roeser J.C., et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371(11):1028–1038. doi: 10.1056/NEJMoa1315815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Scher H.I., Graf R.P., Schreiber N.A., McLaughlin B., Lu D., Louw J., et al. Nuclear-specific AR-V7 protein localization is necessary to guide treatment selection in metastatic castration-resistant prostate cancer. Eur Urol. 2018;74(5):531–539. doi: 10.1016/j.eururo.2018.07.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bastos Diogo A., Antonarakis Emmanuel S. CTC-derived AR-V7 detection as a prognostic and predictive biomarker in advanced prostate cancer. Expert Rev Mol Diagn. fev. 2018;18(2):155–163. doi: 10.1080/14737159.2018.1427068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Antonarakis E.S., Lu C., Luber B., Wang H., Chen Y., Nakazawa M., et al. Androgen receptor splice variant 7 and efficacy of taxane chemotherapy in patients with metastatic castration-resistant prostate cancer. JAMA Oncol. 2015;1(5):582–591. doi: 10.1001/jamaoncol.2015.1341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Scher H.I., Lu D., Schreiber N.A., Louw J., Graf R.P., Vargas H.A., et al. Association of AR-V7 on circulating tumor cells as a treatment-specific biomarker with outcomes and survival in castration-resistant prostate cancer. JAMA Oncol. 2016;2(11):1. doi: 10.1001/jamaoncol.2016.1951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Antonarakis E.S., Lu C., Wang H., Luber B., Nakazawa M., Roeser J.C., et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371(11):1028–1038. doi: 10.1056/NEJMoa1315815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Fujita N., Hatakeyama S., Tabata R., Okita K., Kido K., Hamano I., et al. Real-world effects of novel androgen receptor axis-targeted agents on oncological outcomes in non-metastatic castration-resistant prostate cancer: a multi-institutional retrospective study. Prostate Int. 2024;12:46–51. doi: 10.1016/j.prnil.2023.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Joung J.Y., Jeong I.G., Kang S.G., Ko Y.H., Koo K.C., Kim K.H., et al. Establishing consensus recommendations for metastatic hormone-sensitive prostate cancer in South Korea: a modified Delphi study. Investig Clin Urol. 2025 Sep;66(5):416–430. doi: 10.4111/icu.20250147. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.