Key Points
Question
Does inheritance of the variant HSD3B1 (1245C) allele predict worse clinical outcomes in men treated with androgen-deprivation therapy (ADT) for biochemically recurrent prostate cancer after primary radiotherapy?
Findings
In this cohort of 213 men treated at a large academic center and retrospectively genotyped, inheritance of the HSD3B1 (1245C) allele was not associated with a shorter time to progression or overall survival, but it was associated with shorter times to metastasis according to the number of variant alleles inherited.
Meaning
The variant HSD3B1 (1245C) allele was associated with more rapid development of metastases in men receiving ADT for biochemical recurrence after primary radiotherapy.
Abstract
Importance
The variant HSD3B1 (1245C) allele enhances dihydrotestosterone synthesis and predicts resistance to androgen-deprivation therapy (ADT) for biochemically recurrent prostate cancer after prostatectomy and for metastatic disease. Whether this is true after radiotherapy is unknown.
Objective
To determine whether the HSD3B1 (1245C) allele predicts worse clinical outcomes from ADT for biochemical recurrence after radiotherapy.
Design, Setting, and Participants
The Prostate Clinical Research Information System at Dana-Farber Cancer Institute was used to identify the study cohort, which included men treated with ADT for biochemical recurrence after primary radiotherapy between 1996 and 2013. We retrospectively determined HSD3B1 genotype.
Main Outcomes and Measures
Time to progression, time to metastasis, and overall survival according to genotype. Demographic and treatment characteristics were evaluated for confounders. Multivariable analyses were performed to adjust for known prognostic factors.
Results
A total of 218 eligible men were identified, of whom 213 (98%) were successfully genotyped. Of these, 97 of 213 (46%), 96 of 213 (45%) and 20 of 213 (9%) carried 0, 1, and 2 variant alleles. Overall variant allele frequency was 136 of 426 alleles (32%). Median patient age (interquartile range) was 69 (63-74), 72 (65-78), and 69 (65-77) years for 0, 1, and 2 variant alleles (P = .03). Demographic and treatment factors were otherwise similar. During a median follow-up of 7.9 years, median time to progression was 2.3 years (95% CI, 1.6-3.1 years) with 0 variant alleles, 2.3 years (95% CI, 1.5-3.3 years) with 1 variant allele, and 1.4 years (95% CI, 0.7-3.3 years) with 2 variant alleles (P = .68). Median time to metastasis diminished with the number of variant alleles inherited: 7.4 (95% CI, 6.7-9.7), 5.8 (95% CI, 4.9-6.5), and 4.4 (95% CI, 3.0-5.7) years, with inheritance of 0, 1, and 2 variant alleles, respectively (P = .03). Median OS was 7.7 (95% CI, 6.7-10.3), 6.9 (95% CI, 5.8-8.4), and 7.2 (95% CI, 3.8-7.9) years with inheritance of 0, 1, and 2 variant alleles, respectively (P = .31). On multivariable analysis with 0 variant alleles as the reference, the adjusted hazard ratio for metastasis was 1.19 (95% CI, 0.74-1.92) (P = .48) for 1 variant allele and 2.01 (95% CI, 1.02-3.97) (P = .045) for 2 variant alleles. Multivariable analysis did not demonstrate significant differences in TTP or OS.
Conclusions and Relevance
In this study, the HSD3B1 (1245C) allele was associated with more rapid development of metastases in men treated with ADT for biochemical recurrence after primary radiation therapy for prostate cancer. Notably, 105 of 213 men (49%) had received prior ADT, and 119 of 213 (56%) received an androgen receptor antagonist during salvage treatment, both of which may attenuate the effect of the variant allele.
This cohort study evaluates whether the HSD3B1 (1245C) allele is associated with worse clinical outcomes from androgen-deprivation therapy for biochemical recurrence after radiotherapy.
Introduction
Since Huggins and Hodges discovered the therapeutic impact of castration, androgen-deprivation therapy (ADT) has been the cornerstone of systemic therapy for prostate cancer and has improved overall survival (OS) in multiple contexts. While nearly all men will respond to ADT, most will eventually develop castration-resistant disease. There is wide variation in the duration of response to ADT, ranging from immediate progression to responses of several years. A germline variant, 1245A>C, in HSD3B1 (OMIM 109715), encodes a more stable enzyme required for dihydrotestosterone synthesis from extragonadal precursor steroids. Inheritance of the variant HSD3B1 (1245C) allele has been shown to predict intrinsic prostate cancer resistance to ADT in 2 postprostatectomy biochemically recurrent cohorts and a metastatic cohort. These findings were recently independently validated. Whether the variant allele is also associated with inferior clinical outcomes for men treated with ADT in the context of biochemical recurrence after radiotherapy (RT) is unknown. We sought to investigate the impact of HSD3B1 genotype in this setting.
Methods
We used the previously characterized Prostate Clinical Research Information System at Dana-Farber Cancer Institute to identify men who were treated with ADT for biochemical recurrence after definitive RT. Key clinical data were extracted and systematically quality assured by 2 genitourinary oncologists. Biological specimens and clinical data were obtained in accordance with Dana-Farber institutional review board policy, and all patients provided written informed consent. The HSD3B1 genotype was determined retrospectively from buffy coat using previously published methods by investigators blind to the clinical data.
Kaplan-Meier methods were used to assess time to progression (TTP), time to metastasis (TTM), and OS from initiation of salvage ADT. We analyzed differences in these end points according to genotype using an additive genetic model with the log-rank trend test. Progression was defined as the first occurrence of (1) prostate-specific antigen (PSA) progression; (2) radiographic progression; or (3) initiation of second-line therapy. We defined PSA progression as a minimum of 2 increases in PSA, with the date of first increase (nadir plus >0.02 ng/mL) as the progression date. For patients who received intermittent ADT, PSA increases during off-ADT periods were not counted as progression. Both TTP and TTM were censored on the date of last disease assessment for those who did not develop progression or metastasis.
Demographic and treatment characteristics were compared across genotypes to assess for confounders using Kruskal-Wallis and Fisher exact tests. Multivariable analyses were performed with Cox regression to adjust for known prognostic factors, including biopsy Gleason score, PSA at ADT initiation, age, and use of prior ADT with RT. P < .05 was interpreted as statistically significant. Analyses were performed with SAS software, version 9.4 (SAS Institute Inc).
Results
We identified 218 men treated with ADT between 1996 and 2013 for biochemical recurrence after RT for localized prostate cancer, of whom 213 (98%) were successfully genotyped. Of these, 97 of 213 (46%), 96 of 213 (45%), and 20 of 213 (9%) carried 0, 1, and 2 variant alleles, respectively. The overall variant allele frequency was 136 of 426 alleles (32%). Median follow-up was 7.9 years. Demographic and treatment factors were similar across genotypes, although the median age at ADT initiation was slightly higher in heterozygotes than in the other 2 groups (Table).
Table. Demographic and Treatment Characteristics by HSD3B1 Genotypea.
| Characteristic | Patients, No.(%) | P Valueb | ||
|---|---|---|---|---|
| Homozygous Wild-type (n = 97) |
Heterozygous (n = 96) |
Homozygous Variant (n = 20) |
||
| Age at ADT initiation, median (IQR), y | 69 (63-74) | 72 (65-78) | 69 (65-77) | .03 |
| Clinical T stage at diagnosis | .73 | |||
| T1 | 48 (50) | 49 (51) | 10 (50) | |
| T2 | 24 (25) | 17 (18) | 2 (10) | |
| T3-T4 | 4 (4) | 3 (3) | 1 (5) | |
| Tx | 8 (8) | 12 (13) | 2 (10) | |
| Unknown | 13 (13) | 15 (16) | 5 (25) | |
| Clinical N stage at diagnosis | .76 | |||
| N0 | 37 (38) | 38 (40) | 7 (35) | |
| Nx | 47 (48) | 43 (45) | 8 (40) | |
| Unknown | 13 (13) | 15 (16) | 5 (25) | |
| Biopsy Gleason score | .43 | |||
| ≤6 | 29 (30) | 20 (21) | 2 (10) | |
| 7 | 33 (34) | 33 (34) | 10 (50) | |
| 8-10 | 28 (29) | 32 (33) | 6 (30) | |
| Unknown | 7 (7) | 11 (12) | 2 (10) | |
| ADT as part of local therapy | .84 | |||
| No | 49 (51) | 50 (52) | 9 (45) | |
| Yes | 48 (49) | 46 (48) | 11 (55) | |
| Year of ADT initiation | .66 | |||
| 1996-2000 | 25 (26) | 25 (26) | 8 (40) | |
| 2001-2005 | 36 (37) | 31 (32) | 4 (20) | |
| 2006-2010 | 31 (32) | 32 (33) | 6 (30) | |
| 2011-2013 | 5 (5) | 8 (8) | 2 (10) | |
| ADT schedule | .93 | |||
| Continuous | 63 (65) | 62 (65) | 12 (60) | |
| Intermittent | 34 (35) | 34 (35) | 8 (40) | |
| Received an antiandrogen during ADT | >.99 | |||
| No | 43 (44) | 42 (44) | 9 (45) | |
| Yes | 54 (56) | 54 (56) | 11 (55) | |
| PSA at ADT, median (IQR), ng/mL | 8.7 (4.6-18) | 11.4 (6.4-19) | 13.6 (6.0-27) | .13 |
| Follow-up after ADT initiation, median (IQR), y | 6.4 (4.1-10.9) | 8.3 (4.1-13.0) | 7.9 (5.9-13.6) | .56 |
Abbreviations: ADT, androgen-deprivation therapy; IQR, interquartile range; PSA, prostate-specific antigen.
Data are presented as number (percentage) of patients unless otherwise indicated.
Kruskal-Wallis test for continuous variables or Fisher exact test for categorical variables.
Kaplan-Meier plots of time to event outcomes are shown in the Figure. Median TTP was 2.3 years (95% CI, 1.6-3.1 years) in men who inherited 0 variant alleles, 2.3 years (95% CI: 1.5-3.3 years) with 1 variant allele, and 1.4 years (95% CI: 0.7-3.3 years) with 2 variant alleles (P = .68). Median TTM diminished with the number of variant alleles inherited: 7.4 (95% CI, 6.7-9.7), 5.8 (95% CI, 4.9-6.5), and 4.4 (95% CI, 3.0-5.7) years, with inheritance of 0, 1, and 2 variant alleles, respectively (P = .03). Median OS was 7.7 (95% CI, 6.7-10.3), 6.9 (95% CI, 5.8-8.4), and 7.2 (95% CI, 3.8-7.9) years with inheritance of 0, 1, and 2 variant alleles, respectively (P = .31). On multivariable analysis with 0 variant alleles as the reference, the adjusted hazard ratio (HR) for metastasis was 1.19 (95% CI, 0.74-1.92) (P = .48) for 1 variant allele and 2.01 (95% CI, 1.02-3.97) (P = .045) for 2 variant alleles. Multivariable analysis did not demonstrate significant differences in TTP or OS.
Figure. Time to Progression (A), Time to Metastasis (B), and Overall Survival (C) According to HSD3B1 Genotype.
Discussion
To our knowledge, this is the first study to analyze the association of HSD3B1 genotype with response to ADT after RT for localized prostate cancer. Previous analyses in men treated with ADT for biochemical failure after prostatectomy or for metastatic disease have shown a strong association between HSD3B1 genotype and response to ADT. In contrast to those studies, there was not a statistically significant association between genotype and composite TTP or OS in the current study. Nonetheless, we found a prominent difference in TTM, with differences between genotypes measured in years. These results are consistent with those published for the Cleveland Clinic postprostatectomy cohort, in whom there was also a large difference in TTM according to genotype.
There are a few notable differences between our cohort and the previously published post-prostatectomy cohorts. In the present study 105 of 213 men (49%) had received prior ADT as part of RT. In contrast, few men had previously received ADT in the published cohorts. This is relevant, since the selective pressure from ADT ultimately leads to emergence of multiple somatic resistance mechanisms, which could dilute the impact of germline HSD3B1 genotype. Moreover, 119 of 213 men (56%) in the present study received an androgen receptor (AR) antagonist during ADT, whereas the vast majority in the previously published cohorts did not. Although the HSD3B1 (1245C) allele increases the rate at which adrenal androgen precursors are converted to dihydrotestosterone within tumor cells, AR antagonists compete with intratumoral androgens and may attenuate the effect of the variant allele. Thus, it is possible that the high rates of prior ADT exposure and frequent use of AR antagonists during salvage ADT modified the impact of genotype with respect to composite TTP and OS. Nonetheless, the large impact on TTM is statistically and clinically significant. In addition, patients who receive definitive RT often differ from those who receive prostatectomy, and typically the former have more aggressive disease. Adverse genetics underlying more aggressive disease may thus be more prevalent in those treated with RT, which might also contribute to differences in the effect of HSD3B1 genotype between cohorts.
Limitations
Regarding OS, a landmark trial of ADT for biochemical recurrence after RT demonstrated a median OS of 8.8 to 9.1 years, and this may be an underestimate with modern systemic therapy. Thus, the follow-up length for this cohort is inadequate to reliably exclude a difference in OS by genotype. Given the degree of difference in TTM, one would expect that there may eventually also be a difference in survival, albeit delayed by several years. While speculative, it is noteworthy that the OS curves for our cohort are potentially starting to diverge by approximately 7.5 years. Prostate cancer–specific mortality could not be analyzed, since cause of death data were not available. Finally, OS data may be confounded by subsequent therapies. As reported by Almassi et al in the companion article in this issue of JAMA Oncology, the variant allele has been associated with delayed progression in men receiving nonsteroidal CYP17A1 inhibition.
Conclusions
In conclusion, the HSD3B1 (1245C) allele that enhances dihydrotestosterone synthesis from extragonadal precursor steroids is associated with more rapid development of metastases in men treated with ADT for biochemical recurrence after RT for prostate cancer.
References
- 1.Huggins C, Hodges CV. Studies on prostatic cancer, I: the effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res. 1941;1(4):293-297. [DOI] [PubMed] [Google Scholar]
- 2.D’Amico AV, Chen MH, Renshaw AA, Loffredo M, Kantoff PW. Androgen suppression and radiation vs radiation alone for prostate cancer: a randomized trial. JAMA. 2008;299(3):289-295. [DOI] [PubMed] [Google Scholar]
- 3.Messing EM, Manola J, Sarosdy M, Wilding G, Crawford ED, Trump D. Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in men with node-positive prostate cancer. N Engl J Med. 1999;341(24):1781-1788. [DOI] [PubMed] [Google Scholar]
- 4.Bolla M, Gonzalez D, Warde P, et al. . Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med. 1997;337(5):295-300. [DOI] [PubMed] [Google Scholar]
- 5.Duchesne GM, Woo HH, Bassett JK, et al. . Timing of androgen-deprivation therapy in patients with prostate cancer with a rising PSA (TROG 03.06 and VCOG PR 01-03 [TOAD]): a randomised, multicentre, non-blinded, phase 3 trial. Lancet Oncol. 2016;17(6):727-737. [DOI] [PubMed] [Google Scholar]
- 6.Watson PA, Arora VK, Sawyers CL. Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer. Nat Rev Cancer. 2015;15(12):701-711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dai C, Heemers H, Sharifi N. Androgen signaling in prostate cancer. Cold Spring Harb Perspect Med. 2017;a030452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sharifi N, Dahut WL, Steinberg SM, et al. . A retrospective study of the time to clinical endpoints for advanced prostate cancer. BJU Int. 2005;96(7):985-989. [DOI] [PubMed] [Google Scholar]
- 9.Chang KH, Li R, Kuri B, et al. . A gain-of-function mutation in DHT synthesis in castration-resistant prostate cancer. Cell. 2013;154(5):1074-1084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hearn JWD, AbuAli G, Reichard CA, et al. . HSD3B1 and resistance to androgen-deprivation therapy in prostate cancer: a retrospective, multicohort study. Lancet Oncol. 2016;17(10):1435-1444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Agarwal N, Hahn AW, Gill DM, Farnham JM, Poole AI, Cannon-Albright L. Independent validation of effect of HSD3B1 genotype on response to androgen-deprivation therapy in prostate cancer. JAMA Oncol. 2017;3(6):856-857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Oh WK, Hayes J, Evan C, et al. . Development of an integrated prostate cancer research information system. Clin Genitourin Cancer. 2006;5(1):61-66. [DOI] [PubMed] [Google Scholar]
- 13.Barocas DA, Alvarez J, Resnick MJ, et al. . Association between radiation therapy, surgery, or observation for localized prostate cancer and patient-reported outcomes after 3 years. JAMA. 2017;317(11):1126-1140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Crook JM, O’Callaghan CJ, Duncan G, et al. . Intermittent androgen suppression for rising PSA level after radiotherapy. N Engl J Med. 2012;367(10):895-903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Almassi N, Reichard C, Li J, et al. . HSD3B1 and response to a nonsteroidal CYP17A1 inhibitor in castration-resistant prostate cancer [published online October 12, 2017]. JAMA Oncol. doi: 10.1001/jamaoncol.2017.3159 [DOI] [PMC free article] [PubMed] [Google Scholar]