Abstract
Background
We conducted a study assess whether testosterone therapy (TT) alters prostate cancer risk using a large US commercial insurance research database.
Methods
From the HealthCore Integrated Research Database (HIRDSM), we selected men aged 30 years or greater who were new users of TT during 2007–2015. We selected two comparison groups: 1) unexposed (matched 10:1); 2) new users of phosphodiesterase type 5 inhibitor (PDE5i). Incident prostate cancer was defined as diagnosis of prostate cancer within four-weeks following prostate biopsy. Propensity scores and inverse probability of treatment weights were used in Poisson regression models to estimate adjusted incidence rates, incidence rate ratios (IRRs) and 95% confidence intervals (CI). Subgroup analyses included stratification by prostate cancer screening, hypogonadism, and follow-up time.
Results
The adjusted prostate cancer IRR was 0.77 (95%CI: 0.68, 0.86) when comparing TT with the unexposed group and 0.85 (95%CI: 0.79, 0.91) in comparison with the PDE5i group. Inverse associations between TT and prostate cancer were observed in a majority of subgroup analyses, although in both comparisons estimates generally attenuated with increasing time following initial exposure. Amongst TT users, duration of exposure was not associated with prostate cancer.
Conclusions
Men who received TT did not have a higher rate of prostate cancer compared with the unexposed or PDE5i comparison groups. The inverse association between TT and prostate cancer could be the result of residual confounding, contraindication bias, or undefined biologic effect.
Impact
This study suggests that limited TT exposure does not increase risk of prostate cancer in the short-term.
Keywords: Androgens, Epidemiology, Incidence, Prostatic Neoplasms, Risk, Testosterone
Introduction
In recent decades, testosterone therapy (TT) has dramatically increased in the United States1, though trends have recently plateaued and slightly decreased.2 TT is approved by the United States Food and Drug Administration for men with hypogonadism—confirmed morning serum testosterone of <300 ng/dL—due to disorders of the testicles, pituitary gland, or brain. There are few specific signs or symptoms of hypogonadism (incomplete/delayed sexual development, body hair loss, very small testes) and the majority of TT is prescribed for non-specific signs or symptoms (e.g., fatigue, reduced muscle bulk, increased body fat) that are age-related.3 Although testosterone trials have provided some evidence that TT in men older than 65 years may aid some of the maladies associated with hypogonadism (sexual function, physical function, mood, and depressive symptoms),4 no trial was designed to evaluate risk of prostate cancer, and most observational studies of this relation have been underpowered.
In 1941, Huggins and Hodges showed that prostate cancer was androgen-dependent5 and this led to the development of androgen deprivation therapy as well as a plethora of basic, animal and epidemiologic studies of sex steroid hormones and prostate cancer. These studies have clearly demonstrated the importance of the androgen pathway in prostate cancer progression, and the largest epidemiologic study to date of prediagnostic circulating hormones has recently found that men with very low endogenous testosterone may have a reduced risk of developing prostate cancer.6 Whether exogeneous testosterone alters risk of prostate cancer is largely unknown. Exogeneous supplementation of a single metabolite within a complex biochemical pathway with a multitude of phenotypic effects deserves careful scientific study that cannot be substituted with studies focused on endogenous androgens.7,8 Therefore, we conducted a study using a large U.S. commercial insurance research database to assess whether TT was associated with risk of prostate cancer.
Material and Methods
The HealthCore Integrated Research Database (HIRDSM) is a repository of administrative claims beginning in 2006 with linked medical, pharmacy and eligibility data for approximately 59 million researchable covered lives (at the time of this study) with a median continuous membership of 3 years enrolled in 14 commercial health plans across the US.9,10 From the HIRD, we selected men aged 30 years or greater with medical and pharmacy coverage who were newly dispensed a TT prescription (see Supplemental Table 1 for codes) during 1/1/ 2007–7/31/2015. We selected two control groups to which we compared the TT exposed group: 1) unexposed men; and 2) men dispensed one or more phosphodiesterase type 5 inhibitor (PDE5i) prescriptions. Our study design included the PDE5i group as to provide a comparison with a group of men who were willing, able, and motivated to seek medical care when symptomatic and fill prescriptions. In addition, PDE5i medications are not associated with risk of prostate cancer.11 The unexposed comparison group was matched with a target ratio of 10:1 using date of birth (+/− 183 days), outpatient physician visit (+/− 60 days of TT subject’s dispensed prescription date), and US region of residence. For the PDE5i comparison group, we selected all men newly dispensed a PDE5i during the study period. Men of comparison groups who were dispensed TT during follow-up were right-censored and entered, at that time, into the TT exposed group. Once exposed to TT, men remained in the TT group until event date or right-censoring. Male sex was determined by both self-report and a lack of ICD-9 codes indicating transgenderism or GID.
Index dates were: date of first dispensing of TT for exposed subjects; date of the matched outpatient physician visit for unexposed comparison group subjects; and date of first dispensing of PDE5i for the PDE5i comparison group subjects. We required a minimum 12-month continuous enrollment prior to index date. Men were excluded if, in the pre-index period, there was evidence suggesting they may have had prevalent prostate cancer. Thus we excluding men that had any code indicating prostate cancer, prostate cancer-specific treatment, elevated PSA (or lab test showing >4ng/ml), prostate ultrasound guidance, prostate biopsy, medications containing estrogen, transgenderism or GID (any timepoint), prostatectomy, congenital absence of prostate, TT dispensing, or PDE5i dispensing or erectile dysfunction (TT vs. unexposed comparison only) (see Supplemental Tables 1 and 2 for codes). Matching of unexposed men occurred after exclusions had been applied to maximize algorithm efficiency. All subjects were followed from index date until the earliest of: end of study (7/31/2015), health plan disenrollment, estrogen use, or prostate cancer (outcome).
TT exposure durations were calculated as days supplied plus a bridge rule to account for non-adherence and differences in dispensing and use. Based on exploratory analysis of missing days’ supply (for oral, topical, patch, implant) and recommended usage (for injection), the bridge rule for all TT formulations was 30 days, except for mail-order TT formulations with a bridge rule of 90 days.
Incident prostate cancer was defined as diagnosis of prostate cancer (ICD-9: 185.xx) within four-weeks following a prostate biopsy (see Supplemental Table 3 for codes) using biopsy date as date of diagnosis. To assess the validity of this definition, study eligible men who had ever lived in Georgia were submitted to the Georgia Comprehensive Cancer Registry (GCCR) for probabilistic linkage.9 We also assessed metastatic prostate cancer, using the Dolan algorithm.12 This study was approved by the New England IRB and the Georgia Department of Public Health IRB. HIRDSM data are accessible under contract with HealthCore Inc.
Statistical analysis
We first calculated unadjusted matched/crude prostate cancer incidence rates and 95% confidence intervals (CI) for each of the exposure groups. We then used logistic regression to compute propensity scores13 and estimated standardized differences to assess covariate balance.14 Propensity score models included age, region, index date, pre-index time, and the following pre-index period covariates: Deyo-Charlson co-morbidity index, obesity, benign prostatic hyperplasia, family history of prostate cancer, inflammatory diseases of the prostate, other prostate disorders, urinary symptoms, osteoarthritis, biologic treatment, antineoplastics treatment, anti-TNF treatment, alpha-reductase treatments, HIV therapy, presence of any oncologist visit, prostate cancer screening, annual examinations, ER utilization during the past year, and inpatient hospitalization utilization during the past year. Inverse probability of treatment weights were used in Poisson regression models to estimate adjusted prostate cancer incidence rates, incidence rate ratios (IRRs), and their respective 95% CIs using doubly robust estimation.15
We conducted subgroup analyses by pre-index prostate cancer screening, hypogonadism, and or benign prostatic hyperplasia. We assessed associations by amount of pre-index enrollment time, calendar year of index date and follow-up time since index date.
Amongst the TT group, we estimated IRRs by time on TT, number of TT refills, route of administration, and change in circulating testosterone concentration during TT.
Using the cancer registry data from GCCR as a gold standard, we estimated the positive predictive value (PPV) and sensitivity of the prostate cancer case definition in each group, and corrected effect estimates for outcome misclassification.16
Results
Cohort characteristics
There were 76,159 men in the TT group who were matched with 721,326 unexposed men (Table 1, Supplemental Table 4). For the PDE5i comparison, there were 113,041 TT men available for comparison with 147,620 PDE5i users. Median TT exposure time was 65 days. The TT vs unexposed groups were closely matched on age and region (Table 1). Propensity score weighting achieved good comparability (d <|0.20|) for both comparison groups (Table 1). Fatigue, hypogonadism, and psychosexual dysfunction were not included in the PS due to convergence issues. We conducted sensitivity analyses in which we additionally controlled for these factors. Our prostate cancer case definition had high sensitivity (91.2%, 95%CI: 87.7, 94.0%) and PPV (81.7%, 95%CI: 77.9, 85.0%) that did not differ substantially by analytic group.9
Table 1.
Characteristics of members in each treatment group before propensity score weighting
| Characteristic | TT | Unexposed | Standardized Difference | Standardized Difference after PS | TT | PDE5i | Standardized Difference | Standardized Difference after PS |
|---|---|---|---|---|---|---|---|---|
| n (%) / median (IQR) | n (%) / median (IQR) | n (%) / median (IQR) | n (%) / median (IQR) | |||||
| Number of men | 76,159 (100.0) | 721,326 (100.0) | – | – | 113,041 (100.0) | 147,620 (100.0) | – | – |
| Age at index date | 50.0 (15.0) | 50.0 (15.0) | −0.005 | 0.004 | 51.0 (14.0) | 53.0 (13.0) | −0.096 | 0.005 |
| 30–44 | 22,751 (29.9) | 214,074 (29.7) | 0.004 | −0.014 | 30,354 (26.9) | 30,753 (20.8) | 0.142 | 0.055 |
| 45–64 | 46,550 (61.1) | 442,171 (61.3) | −0.004 | 0.013 | 71,752 (63.5) | 105,509 (71.5) | −0.171 | −0.122 |
| 65+ | 6,858 (9.0) | 65,081 (9.0) | −0.001 | −0.001 | 10,935 (9.7) | 11,358 (7.7) | 0.070 | 0.117 |
| Region | ||||||||
| MidWest | 20,263 (26.6) | 191,169 (26.5) | 0.002 | 0.002 | 29,103 (25.7) | 20,649 (14.0) | 0.298 | 0.001 |
| Northeast | 6,204 (8.1) | 59,619 (8.3) | −0.004 | −0.008 | 10,382 (9.2) | 34,967 (23.7) | −0.399 | 0.004 |
| South | 31,166 (40.9) | 292,337 (40.5) | 0.008 | −0.001 | 46,577 (41.2) | 49,856 (33.8) | 0.154 | −0.007 |
| West | 17,246 (22.6) | 166,035 (23.0) | −0.009 | 0.005 | 24,743 (21.9) | 35,068 (23.8) | −0.044 | 0.008 |
| Other | 1,280 (1.7) | 12,166 (1.7) | −0.000 | −0.001 | 2,236 (2.0) | 7,080 (4.8) | −0.156 | −0.011 |
| Calendar year of index date | ||||||||
| 2007 – 2009 | 21,625 (28.4) | 215,048 (29.8) | −0.031 | −0.003 | 33,080 (29.3) | 68,183 (46.2) | −0.355 | 0.014 |
| 2010 – 2012 | 33,662 (44.2) | 318,694 (44.2) | 0.000 | 0.002 | 49,842 (44.1) | 54,855 (37.2) | 0.142 | −0.005 |
| 2013 – 2015 | 20,872 (27.4) | 187,584 (26.0) | 0.032 | 0.001 | 30,119 (26.6) | 24,582 (16.7) | 0.244 | −0.010 |
| Duration of pre-index period | ||||||||
| 5 or more years | 18,885 (24.8) | 209,139 (29.0) | −0.095 | 0.041 | 28,870 (25.5) | 26,077 (17.7) | 0.192 | −0.018 |
| 4–4.99 years | 9,323 (12.2) | 91,690 (12.7) | −0.014 | 0.002 | 14,170 (12.5) | 15,816 (10.7) | 0.057 | 0.011 |
| 3–3.99 years | 11,966 (15.7) | 115,617 (16.0) | −0.009 | −0.024 | 18,004 (15.9) | 22,193 (15.0) | 0.025 | 0.004 |
| 2–2.99 years | 17,510 (23.0) | 150,982 (20.9) | 0.050 | −0.018 | 25,658 (22.7) | 37,559 (25.4) | −0.064 | 0.004 |
| 1–1.99 years | 18,475 (24.3) | 153,898 (21.3) | 0.070 | −0.009 | 26,339 (23.3) | 45,975 (31.1) | −0.177 | 0.002 |
| In the year prior to index date | ||||||||
| History of medical diagnoses | ||||||||
| Family history of prostate cancer | 462 (0.6) | 4,300 (0.6) | 0.001 | 0.001 | 797 (0.7) | 870 (0.6) | 0.014 | −0.003 |
| Fatigue | 42,639 (56.0) | 172,488 (23.9) | 0.693 | 0.669 | 59,380 (52.5) | 31,501 (21.3) | 0.683 | 0.517 |
| Hyperplasia of prostate | 10,126 (13.3) | 78,307 (10.9) | 0.075 | 0.013 | 17,596 (15.6) | 19,340 (13.1) | 0.070 | −0.005 |
| Hypogonadism | 32,246 (42.3) | 32,928 (4.6) | 0.996 | 0.948 | 49,047 (43.4) | 8,318 (5.6) | 0.977 | 0.890 |
| Inflammatory disease of prostate | 3,519 (4.6) | 26,187 (3.6) | 0.050 | 0.012 | 5,831 (5.2) | 5,832 (4.0) | 0.058 | −0.003 |
| Obesity | 15,670 (20.6) | 88,245 (12.2) | 0.227 | 0.014 | 23,427 (20.7) | 16,305 (11.0) | 0.267 | −0.007 |
| Osteoporosis | 1,770 (2.3) | 8,892 (1.2) | 0.083 | 0.009 | 2,553 (2.3) | 1,434 (1.0) | 0.102 | −0.019 |
| Other diseases of prostate | 1,528 (2.0) | 12,162 (1.7) | 0.024 | 0.006 | 2,505 (2.2) | 2,765 (1.9) | 0.024 | −0.001 |
| Psychosexual dysfunction | 5,394 (7.1) | 13,104 (1.8) | 0.257 | 0.258 | 10,012 (8.9) | 12,525 (8.5) | 0.013 | −0.003 |
| Pulmonary hypertension | 483 (0.6) | 3,903 (0.5) | 0.012 | 0.001 | 731 (0.6) | 598 (0.4) | 0.033 | 0.017 |
| Urinary symptoms | 10,078 (13.2) | 77,553 (10.8) | 0.076 | 0.015 | 16,351 (14.5) | 16,191 (11.0) | 0.105 | −0.006 |
| Prescription medications | ||||||||
| 5–alpha reductase inhibitors | 4,980 (6.5) | 35,075 (4.9) | 0.072 | 0.013 | 8,040 (7.1) | 8,343 (5.7) | 0.060 | −0.003 |
| Antineoplastics and adjunctive therapies | 610 (0.8) | 5,663 (0.8) | 0.002 | 0.002 | 893 (0.8) | 881 (0.6) | 0.023 | −0.003 |
| AntiTNFs and adjunctive therapies | 594 (0.8) | 5,649 (0.8) | −0.000 | 0.002 | 896 (0.8) | 876 (0.6) | 0.024 | −0.001 |
| Biologics | 917 (1.2) | 8,224 (1.1) | 0.006 | 0.007 | 1,347 (1.2) | 1,536 (1.0) | 0.014 | −0.006 |
| HIV antiretroviral therapy | 1,052 (1.4) | 3,588 (0.5) | 0.092 | 0.001 | 1,385 (1.2) | 912 (0.6) | 0.064 | −0.013 |
| Statins | 27,933 (36.7) | 217,569 (30.2) | 0.138 | 0.016 | 43,525 (38.5) | 51,344 (34.8) | 0.077 | −0.000 |
| Comorbidity measures | ||||||||
| Deyo Charlson comorbidity index | 0.0 (1.0) | 0.0 (1.0) | 0.130 | 0.009 | 0.0 (1.0) | 0.0 (1.0) | 0.182 | −0.014 |
| 0 | 41,673 (54.7) | 446,069 (61.8) | −0.145 | −0.065 | 59,775 (52.9) | 89,517 (60.6) | −0.157 | −0.018 |
| 1 | 18,483 (24.3) | 154,736 (21.5) | 0.067 | 0.057 | 28,148 (24.9) | 34,072 (23.1) | 0.043 | 0.028 |
| 2+ | 16,003 (21.0) | 120,521 (16.7) | 0.110 | 0.020 | 25,118 (22.2) | 24,031 (16.3) | 0.151 | −0.008 |
| Enhanced Elixhauser index | 2.0 (2.0) | 1.0 (3.0) | 0.250 | 0.055 | 2.0 (3.0) | 1.0 (3.0) | 0.307 | 0.038 |
| 0 | 13,717 (18.0) | 211,929 (29.4) | −0.270 | −0.199 | 18,994 (16.8) | 37,788 (25.6) | −0.216 | −0.097 |
| 1 | 18,289 (24.0) | 185,671 (25.7) | −0.040 | 0.007 | 26,015 (23.0) | 40,004 (27.1) | −0.094 | −0.010 |
| 2+ | 44,153 (58.0) | 323,726 (44.9) | 0.264 | 0.166 | 68,032 (60.2) | 69,828 (47.3) | 0.261 | 0.088 |
| Enhanced Charlson comorbidity index | 0.0 (1.0) | 0.0 (1.0) | 0.134 | 0.011 | 1.0 (2.0) | 0.0 (1.0) | 0.191 | −0.011 |
| 0 | 38,953 (51.1) | 422,280 (58.5) | −0.149 | −0.071 | 55,809 (49.4) | 84,852 (57.5) | −0.163 | −0.025 |
| 1 | 18,733 (24.6) | 158,990 (22.0) | 0.060 | 0.057 | 28,292 (25.0) | 34,732 (23.5) | 0.035 | 0.030 |
| 2+ | 18,473 (24.3) | 140,056 (19.4) | 0.117 | 0.028 | 28,940 (25.6) | 28,036 (19.0) | 0.159 | −0.001 |
| Service utilization | ||||||||
| Had a specialist visit (urologist, oncologist, pathologist) | 13,989 (18.4) | 97,261 (13.5) | 0.134 | 0.051 | 24,401 (21.6) | 24,080 (16.3) | 0.135 | 0.040 |
| Had an annual wellness visit | 10,083 (13.2) | 78,909 (10.9) | 0.071 | 0.045 | 11,466 (10.1) | 17,807 (12.1) | −0.061 | −0.055 |
| Had a prostate cancer screen | 5,625 (7.4) | 40,197 (5.6) | 0.074 | 0.046 | 6,660 (5.9) | 8,663 (5.9) | 0.001 | −0.039 |
| Outpatient visits | 6.0 (12.0) | 3.0 (7.0) | 0.349 | 0.093 | 5.0 (9.0) | 3.0 (7.0) | 0.161 | 0.002 |
| 0 | 10,661 (14.0) | 147,900 (20.5) | −0.173 | −0.105 | 18,565 (16.4) | 31,101 (21.1) | −0.119 | −0.037 |
| 1 | 5,610 (7.4) | 83,078 (11.5) | −0.142 | −0.107 | 10,590 (9.4) | 15,924 (10.8) | −0.047 | −0.012 |
| 2+ | 59,888 (78.6) | 490,348 (68.0) | 0.243 | 0.161 | 83,886 (74.2) | 100,595 (68.1) | 0.134 | 0.040 |
| ER visits | 0.0 (0.0) | 0.0 (0.0) | 0.133 | 0.039 | 0.0 (0.0) | 0.0 (0.0) | 0.055 | 0.002 |
| 0 | 68,116 (89.4) | 673,332 (93.3) | −0.140 | −0.107 | 104,044 (92.0) | 137,897 (93.4) | −0.053 | −0.005 |
| 1 | 6,654 (8.7) | 41,657 (5.8) | 0.114 | 0.089 | 7,729 (6.8) | 8,529 (5.8) | 0.044 | 0.006 |
| 2+ | 1,389 (1.8) | 6,337 (0.9) | 0.082 | 0.060 | 1,268 (1.1) | 1,194 (0.8) | 0.032 | −0.003 |
| Inpatient visits | 0.0 (0.0) | 0.0 (0.0) | 0.102 | 0.030 | 0.0 (0.0) | 0.0 (0.0) | 0.049 | 0.005 |
| 0 | 71,734 (94.2) | 696,286 (96.5) | −0.111 | −0.081 | 108,037 (95.6) | 142,556 (96.6) | −0.051 | −0.006 |
| 1 | 3,554 (4.7) | 20,812 (2.9) | 0.094 | 0.066 | 4,156 (3.7) | 4,329 (2.9) | 0.042 | 0.004 |
| 2+ | 871 (1.1) | 4,228 (0.6) | 0.060 | 0.049 | 848 (0.8) | 735 (0.5) | 0.032 | 0.006 |
| Office visits | 3.0 (5.0) | 1.0 (3.0) | 0.431 | 0.114 | 2.0 (3.0) | 2.0 (3.0) | 0.196 | −0.003 |
| 0 | 13,959 (18.3) | 195,665 (27.1) | −0.211 | −0.133 | 24,617 (21.8) | 40,126 (27.2) | −0.126 | −0.026 |
| 1 | 11,643 (15.3) | 165,353 (22.9) | −0.195 | −0.154 | 22,226 (19.7) | 32,492 (22.0) | −0.058 | −0.020 |
| 2+ | 50,557 (66.4) | 360,308 (50.0) | 0.338 | 0.239 | 66,198 (58.6) | 75,002 (50.8) | 0.156 | 0.038 |
Abbreviations: ER, emergency room; HIV, human immunodeficiency virus; IQR, inter-quartile range; PDE5i, phosphodiesterase type 5 inhibitor; PS, propensity score; PSA, prostate-specific antogen; TNF, tumor necrosis factor; TT, testosterone therapy.
Comparison of TT with unexposed group
We identified 335 prostate cancers in 178,704 person-years in the TT group and 4,133 cases in 1.6 million person-years in the unexposed group (Table 2). Unadjusted age- and region-matched prostate cancer incidence rates were 187.5 per 100,000 person-years in the TT group and 245.5 in the unexposed group, resulting in an IRR of 0.76 (95%CI: 0.68, 0.85, Figure 1). This estimate was similar to the propensity score adjusted IRR of 0.77 (95%CI: 0.68, 0.86). An unmeasured confounder would have to have an IRR of 1.92 with both TT and prostate cancer to explain away this observed association.17,18 In analyses further adjusted for fatigue, hypogonadism, and psychosexual dysfunction, results were slightly attenuated with an adjusted IRR of 0.86 (95%CI: 0.74, 1.00), while results were similar when further adjusting for outcome misclassification (adjusted IRR=0.75, 95%CI: 0.67, 0.84). Inverse associations between TT and prostate cancer were also observed when stratified by prostate cancer screening, hypogonadism, or benign prostatic hyperplasia in the pre-index period (Table 2). Effect estimates did not differ appreciably by duration of pre-index time, although incidence rates did decrease in both TT and unexposed groups which appeared to be related to decreasing prostate cancer incidence by calendar year. Effect estimates attenuated with time from index date, from stronger estimates of 0.51 (<6 months), to 0.72 (6–12 months), to 1.03 (12–24 months). The last period analyzed, of 24 months and greater post index, had an IRR of 0.74.
Table 2.
Incidence rates and incidence rate ratios for prostate cancer for TT and unexposed groups and by pre-index characteristics
| Category | TT |
Unexposed |
Age- and region-matched Incidence Rate Ratio (95%CI) | PS Adjusted Incidence Rate Ratio (95%CI) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| N | # PCa Cases | Person-years | Incidence Rate* (95%CI) | N | # PCa Cases | Person-years | Incidence Rate* (95%CI) | |||
| Overall | 76,159 | 335 | 178,704 | 187.5 (168.4, 208.6) | 721,326 | 4,133 | 1,683,470 | 245.5 (238.1, 253.1) | 0.76 (0.68, 0.85) | 0.77 (0.68, 0.86) |
| Subgroups | ||||||||||
| PSA test in pre-index period | ||||||||||
| Yes | 14,546 | 55 | 32,875 | 167.3 (128.4, 217.9) | 59,069 | 407 | 130,239 | 312.5 (283.6, 344.4) | 0.54 (0.41, 0.71) | 0.54 (0.40, 0.72) |
| No | 61,613 | 280 | 145,829 | 192.0 (170.8, 215.9) | 662,257 | 3,726 | 1,553,231 | 239.9 (232.3, 247.7) | 0.80 (0.70, 0.90) | 0.80 (0.70, 0.91) |
| Prostate cancer screen in the pre-index period | ||||||||||
| Yes | 20,713 | 100 | 47,903 | 208.8 (171.6, 254.0) | 162,966 | 1,222 | 350,320 | 348.8 (329.8, 368.9) | 0.60 (0.49, 0.73) | 0.60 (0.48, 0.74) |
| No | 55,446 | 235 | 130,801 | 179.7 (158.1, 204.2) | 558,360 | 2,911 | 1,333,150 | 218.4 (210.6, 226.4) | 0.82 (0.72, 0.94) | 0.82 (0.71, 0.95) |
| No PSA test and no PCa screen in the pre-index period | 45,541 | 197 | 107,982 | 182.4 (158.7, 209.8) | 521,186 | 2,655 | 1,248,121 | 212.7 (204.8, 221.0) | 0.85 (0.74, 0.99) | 0.86 (0.73, 1.01) |
| Hypogonadism diagnosis in the pre-index period | ||||||||||
| Yes | 32,246 | 123 | 72,994 | 168.5 (141.2, 201.1) | 32,928 | 124 | 45,322 | 273.6 (229.4, 326.3) | 0.62 (0.48, 0.79) | 0.63 (0.49, 0.82) |
| No | 43,913 | 212 | 105,710 | 200.5 (175.3, 229.4) | 688,398 | 4,009 | 1,638,148 | 244.7 (237.3, 252.4) | 0.81 (0.71, 0.94) | 0.84 (0.73, 0.98) |
| Benign prostatic hyperplasia in the pre-index period | ||||||||||
| Yes | 10,126 | 74 | 25,676 | 288.2 (229.5, 362.0) | 78,307 | 891 | 185,506 | 480.3 (449.8, 512.9) | 0.60 (0.47, 0.76) | 0.55 (0.43, 0.71) |
| No | 66,033 | 261 | 153,028 | 170.6 (151.1, 192.6) | 643,019 | 3,242 | 1,497,965 | 216.4 (209.1, 224.0) | 0.79 (0.69, 0.89) | 0.82 (0.71, 0.94) |
| Pre-index enrollment time | ||||||||||
| ≥2 years | 51,876 | 212 | 121,050 | 175.1 (153.1, 200.4) | 525,550 | 2,702 | 1,129,304 | 239.3 (230.4, 248.5) | 0.73 (0.63, 0.84) | 0.72 (0.62, 0.84) |
| ≥3 years | 36,818 | 146 | 83,444 | 175.0 (148.8, 205.8) | 386,530 | 1,771 | 772,056 | 229.4 (218.9, 240.3) | 0.76 (0.64, 0.90) | 0.74 (0.62, 0.89) |
| ≥4 years | 25,934 | 87 | 55,541 | 156.6 (127.0, 193.3) | 279,871 | 1,145 | 516,564 | 221.7 (209.2, 234.9) | 0.70 (0.56, 0.88) | 0.68 (0.54, 0.86) |
| ≥5 years | 17,418 | 47 | 34,197 | 137.4 (103.3, 182.9) | 194,522 | 608 | 326,212 | 186.4 (172.1, 201.8) | 0.74 (0.54, 1.00) | 0.71 (0.51, 0.97) |
| Calandar year of index date | ||||||||||
| 2007–2009 | 21,625 | 173 | 72,756 | 237.8 (204.9, 276.0) | 215,048 | 2,229 | 778,501 | 286.3 (274.7, 298.5) | 0.83 (0.70, 0.97) | 0.84 (0.71, 1.00) |
| 2010–2012 | 33,662 | 143 | 80,793 | 177.0 (150.2, 208.5) | 318,694 | 1,642 | 684,425 | 239.9 (228.6, 251.8) | 0.73 (0.62, 0.88) | 0.72 (0.60, 0.87) |
| 2013–2015 | 20,872 | 19 | 25,155 | 75.5 (48.2, 118.4) | 187,584 | 262 | 220,545 | 118.8 (105.2, 134.1) | 0.64 (0.40, 1.01) | 0.59 (0.35, 0.99) |
| Follow-up period relative to index date | ||||||||||
| <6 months | 76,159 | 27 | 35,069 | 77.0 (52.8, 112.3) | 721,326 | 528 | 323,782 | 163.1 (149.7, 177.6) | 0.47 (0.32, 0.70) | 0.51 (0.34, 0.77) |
| ≥6 & <12 months | 64,164 | 45 | 29,479 | 152.7 (114.0, 204.5) | 589,975 | 487 | 270,125 | 180.3 (165.0, 197.0) | 0.82 (0.60, 1.13) | 0.72 (0.51, 1.01) |
| ≥12 & <24 months | 53,878 | 93 | 45,116 | 206.1 (168.2, 252.6) | 492,355 | 936 | 411,903 | 227.2 (213.1, 242.3) | 0.95 (0.77, 1.18) | 1.03 (0.82, 1.30) |
| ≥24 months | 36,539 | 170 | 69,041 | 246.2 (211.9, 286.2) | 334,378 | 2,182 | 677,660 | 322.0 (308.8, 335.8) | 0.78 (0.66, 0.92) | 0.74 (0.62, 0.88) |
per 100,000 person-years
Abbreviations: CI, confidence interval; PCa, prostate cancer; PS, propensity score; PSA, prostate-specific antogen; TT, testosterone therapy.
Figure 1: Crude cumulative incidence of prostate cancer comparing TT exposed group with unexposed group.
The x-axis shows the years of follow-up and the y-axis shows the crude cumulative incidence of prostate cancer as a percentage of the denominator. The crude cumulative incidence for the TT exposed group is shown as a dashed line, with adjacent smaller-width, dashed lines representing 95 percent confidence intervals. The crude cumulative incidence for the unexposed group is shown as a solid line, with adjacent smaller-width, solid lines representing 95 percent confidence intervals.
Comparison of TT with PDE5i group
Similar results were observed when comparing TT with PDE5i users (Table 3). The crude IRR was 0.71 (95%CI: 0.64, 0.76, Figure 2) and the propensity score adjusted IRR was 0.85 (95%CI: 0.79, 0.91). Results were similar to the overall analyses, when further adjusting for unbalanced factors not included in the propensity score (adjusted IRR=0.87, 95%CI: 0.80, 0.94), and when adjusting for outcome misclassification (adjusted IRR=0.91, 95%CI: 0.85, 0.98). Subgroup analyses using the PDE5i comparison group were mostly similar to those using the unexposed comparison group, including attenuation to the null with increased time since index date with sequential 6-month period IRRs of 0.53, 0.83, 0.88 and 0.96. One difference in the PDE5i analyses, to that of the unexposed group analyses, was an attenuated association with increased pre-index time, with an adjusted IRR of 1.06 (95%CI=0.85, 1.33) among individuals with at least 5 years.
Table 3.
Incidence rates and incidence rate ratios for prostate cancer for TT and PDE5i analytic groups and by pre-index characteristics
| Category | TT |
PDE5i |
Unadjusted Incidence Rate Ratio (95%CI) | PS Adjusted Incidence Rate Ratio (95%CI) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| N | # PCa Cases | Person-years | Incidence Rate* (95%CI) | N | # PCa Cases | Person-years | Incidence Rate* (95%CI) | |||
| Overall | 113,041 | 563 | 267,795 | 210.2 (193.6, 228.3) | 147,620 | 1,091 | 370,507 | 294.5 (277.5, 312.5) | 0.71 (0.64, 0.79) | 0.85 (0.79, 0.91) |
| Subgroups | ||||||||||
| PSA test in pre-index period | ||||||||||
| Yes | 22,022 | 95 | 49,665 | 191.3 (156.4, 233.9) | 18,328 | 135 | 43,989 | 306.9 (259.3, 363.3) | 0.62 (0.48, 0.81) | 0.73 (0.60, 0.87) |
| No | 91,019 | 468 | 218,130 | 214.6 (196.0, 234.9) | 129,292 | 956 | 326,518 | 292.8 (274.8, 312.0) | 0.73 (0.66, 0.82) | 0.87 (0.81, 0.94) |
| Prostate cancer screen in the pre-index period | ||||||||||
| Yes | 33,083 | 181 | 77,206 | 234.4 (202.7, 271.2) | 35,143 | 293 | 87,886 | 333.4 (297.3, 373.8) | 0.70 (0.58, 0.85) | 0.85 (0.74, 0.97) |
| No | 79,958 | 382 | 190,589 | 200.4 (181.3, 221.6) | 112,477 | 798 | 282,622 | 282.4 (263.4, 302.6) | 0.71 (0.63, 0.80) | 0.85 (0.78, 0.92) |
| No PSA test and no PCa screen in the pre-index period | 65,288 | 323 | 156,778 | 206.0 (184.7, 229.8) | 99,367 | 703 | 250,853 | 280.2 (260.3, 301.8) | 0.74 (0.64, 0.84) | 0.89 (0.81, 0.97) |
| Hypogonadism diagnosis in the pre-index period | ||||||||||
| Yes | 49,047 | 223 | 111,853 | 199.4 (174.9, 227.3) | 8,318 | 46 | 16,034 | 286.9 (214.9, 383.0) | 0.70 (0.51, 0.95) | 0.72 (0.57, 0.91) |
| No | 63,994 | 340 | 155,942 | 218.0 (196.0, 242.5) | 139,302 | 1,045 | 354,473 | 294.8 (277.5, 313.2) | 0.74 (0.65, 0.84) | 0.87 (0.80, 0.94) |
| Benign prostatic hyperplasia in the pre-index period | ||||||||||
| Yes | 17,596 | 148 | 44,547 | 332.2 (282.8, 390.3) | 19,340 | 195 | 47,111 | 413.9 (359.7, 476.3) | 0.80 (0.65, 0.99) | 0.86 (0.74, 1.00) |
| No | 95,445 | 415 | 223,249 | 185.9 (168.8, 204.7) | 128,280 | 896 | 323,396 | 277.1 (259.5, 295.8) | 0.67 (0.60, 0.75) | 0.83 (0.77, 0.90) |
| Pre-index enrollment time | ||||||||||
| ≥2 years | 78,598 | 372 | 184,835 | 201.3 (181.8, 222.8) | 91,827 | 625 | 224,866 | 277.9 (257.0, 300.6) | 0.72 (0.64, 0.82) | 0.85 (0.78, 0.93) |
| ≥3 years | 56,239 | 249 | 128,035 | 194.5 (171.8, 220.2) | 59,092 | 340 | 138,275 | 245.9 (221.1, 273.5) | 0.79 (0.67, 0.93) | 0.91 (0.81, 1.02) |
| ≥4 years | 39,785 | 154 | 85,328 | 180.5 (154.1, 211.4) | 39,125 | 185 | 84,928 | 217.8 (188.6, 251.6) | 0.83 (0.67, 1.03) | 0.95 (0.81, 1.10) |
| ≥5 years | 26,704 | 83 | 52,415 | 158.4 (127.7, 196.4) | 24,432 | 80 | 47,526 | 168.3 (135.2, 209.6) | 0.94 (0.69, 1.28) | 1.06 (0.85, 1.33) |
| Calendar year of index date | ||||||||||
| 2007–2009 | 33,080 | 290 | 111,245 | 260.7 (232.3, 292.5) | 68,183 | 744 | 216,131 | 334.2 (320.4, 369.9) | 0.76 (0.66, 0.87) | 0.83 (0.70, 0.97) |
| 2010–2012 | 49,842 | 242 | 120,078 | 201.5 (177.7, 228.6) | 54,855 | 318 | 129,315 | 245.9 (220.3, 274.5) | 0.82 (0.69, 0.97) | 0.73 (0.62, 0.88) |
| 2013–2015 | 30,119 | 31 | 36,472 | 85.0 (59.8, 120.9) | 24,582 | 29 | 25,061 | 115.8 (80.4, 166.5) | 0.73 (0.44, 1.22) | 0.64 (0.40, 1.01) |
| Follow-up period relative to index date | ||||||||||
| <6 months | 113,041 | 47 | 52,104 | 90.2 (67.8, 120.1) | 147,620 | 153 | 67,150 | 227.8 (194.5, 267.0) | 0.40 (0.29, 0.55) | 0.53 (0.43, 0.65) |
| ≥6 & <12 months | 95,454 | 81 | 43,860 | 184.7 (148.5, 229.6) | 122,605 | 132 | 56,164 | 235.0 (198.2, 278.7) | 0.76 (0.58, 1.01) | 0.83 (0.68, 1.01) |
| ≥12 & <24 months | 80,191 | 139 | 67,363 | 206.3 (174.7, 243.7) | 102,326 | 233 | 86,795 | 268.4 (236.1, 305.2) | 0.80 (0.65, 0.99) | 0.88 (0.75, 1.02) |
| ≥24 months | 54,774 | 296 | 104,468 | 283.3 (252.8, 317.5) | 72,432 | 573 | 160,398 | 357.2 (329.2, 387.7) | 0.81 (0.70, 0.94) | 0.96 (0.87, 1.07) |
per 100,000 person-years
Abbreviations: CI, confidence interval; PCa, prostate cancer; PDE5i, phosphodiesterase type 5 inhibitor; PS, propensity score; PSA, prostate-specific antogen; TT, testosterone therapy.
Figure 2: Crude cumulative incidence of prostate cancer comparing TT exposed group with PDE5i comparison group.
The x-axis shows the years of follow-up and the y-axis shows the crude cumulative incidence of prostate cancer as a percentage of the denominator. The crude cumulative incidence for the TT exposed group is shown as a dashed line, with adjacent smaller-width, dashed lines representing 95 percent confidence intervals. The crude cumulative incidence for the PDE5i comparison group is shown as a solid line, with adjacent smaller-width, solid lines representing 95 percent confidence intervals.
Assessments within TT users
Among men who received TT, duration of therapy, number of prescription fills, and route of administration were not associated with prostate cancer (Table 4). Within the TT group, men who experienced more extreme changes in circulating testosterone levels (increased or decreased) had inverse associations with prostate cancer, relative to men closer to the average change.
Table 4.
Among TT exposed group, adjusted incidence rate ratio of prostate cancer by TT therapy characteristics.
| Characteristic | Cases | N | Person-Years | Unadjusted IRR (95%CI) | Adjusted IRR (95%CI) |
|---|---|---|---|---|---|
| Time on testosterone - all medications1 | |||||
| <3 months | 172 | 39,184 | 88,374 | Referent | Referent |
| 3–6 months | 67 | 16,529 | 36,970 | 0.93 (0.70, 1.23) | 0.97 (0.73, 1.29) |
| 6 or more months | 91 | 19,080 | 51,571 | 0.91 (0.70, 1.17) | 0.97 (0.75, 1.26) |
| Number of testosterone fills of first medication2 | |||||
| 1 to 2 | 211 | 46,602 | 106,212 | Referent | Referent |
| 3 to 7 | 79 | 18,321 | 42,918 | 1.17 (0.90, 1.52) | 1.14 (0.69, 1.49) |
| 8 or more | 46 | 9,869 | 27,784 | 1.20 (0.87, 1.65) | 1.14 (0.83, 1.56) |
| Route of Administration3 | |||||
| Patch | 205 | 41,037 | 102,531 | 0.85 (0.68 – 1.06) | 0.86 (0.69 – 1.08) |
| Injection | 125 | 33,400 | 73,408 | Referent | Referent |
| Change in testosterone level on therapy4 | |||||
| Decrease | 6 | 2,268 | 6,417 | Referent | Referent |
| Increase | 27 | 6,480 | 18,877 | 1.53 (0.63, 3.71) | 1.40 (0.56, 3.49) |
| Change in testosterone level on therapy (based on quartiles)5 | |||||
| Decrease | 6 | 2,268 | 6,417 | 0.60 (0.22, 1.66) | 0.64 (0.23, 1.80) |
| 0 – 91 | 12 | 2,101 | 6,131 | 1.26 (0.55, 2.92) | 1.13 (0.48, 2.68) |
| 92–286 | 10 | 2,192 | 6,446 | Referent | Referent |
| 287 + | 5 | 2,187 | 6,300 | 0.51 (0.17, 1.50) | 0.50 (0.17, 1.49) |
Adjusted for: age, pre-index time, DCI, pre-index annual flag, pre-index T test, pre-index PSA test, index date year.
Adjusted for: age, route of administration, prindex time, DCI, pre-index annual flag, pre-index T test, pre-index PSA test, index date year.
Other administrations excluded (N=356); Adjusted for: time on T, age, pre-index time, DCI, pre-index annual flag (1yr), pre-index PSA test, pre-index T test, index date year
Adjusted for: pre-index T result, age, time between tests, DCI, pre-index PSA test, index date year, urinary symptoms, other prostate disease, inflammatory prostate, BPH, hypogonadism, osteoporosis, diabetes (complicated and uncomplicated), liver failure.
Adjusted for: pre-index T result, age, time between tests, pre-index annual flag (1yr), DCI, pre-index PSA test, index date year, pre-index specialist flag, urinary symptoms, other prostate disease, inflammatory prostate, BPH, hypogonadism, family history of prostate cancer, psychosexual dysfunction, osteoporosis, diabetes (complicated and uncomplicated), liver failure.
Associations with metastatic prostate cancer
There were 17 prostate cancer cases with metastatic disease at diagnoses during 178,704 person-years in the TT group and 195 cases during 1.7 million person-years in the unexposed group providing an adjusted IRR of 0.77 (95%CI:0.46, 1.29). For the PDE5i comparison, the TT group had 28 metastatic prostate cancer cases diagnosed during 267,795 person-years and the PDE5i group had 50 cases during 370,507 person-years providing an adjusted IRR of 1.07 (95%CI:0.77, 1.49).
Analyses of testosterone and prostate-specific antigen
In men who received TT for whom we had testosterone laboratory data, circulating testosterone concentrations increased from a pre-index median of 237 ng/dL to 351 ng/dL (Supplemental Table 5). This increase didn’t vary by age but did vary by pre-index circulating testosterone concentration, with greater increases observed for men with lower pre-index concentrations. Circulating prostate-specific antigen (PSA) concentrations, in the subset of men we had such data for, were not largely different between TT and comparison groups (Supplemental Tables 6 and 7). PSA concentrations in TT men increased by a median change of 0.1 ng/mL when comparing pre-index with post-index periods, while the median changes were zero in the comparison groups.
Discussion
In this study of a large healthcare claims database, men who received TT had a lower rate of prostate cancer compared with unexposed men or men receiving a PDE5i. The inverse association between TT and risk of prostate cancer was observed for a majority of subgroup analyses—including stratifications by prostate cancer screening, hypogonadism, and benign prostatic hyperplasia in the pre-index period—yet the association between TT and reduced prostate cancer risk generally attenuated with increased time following initial exposure.
The majority of prior studies to have assessed TT in relation to prostate cancer have had small populations with imprecise effect estimates.19–25 There has been only two previous studies that have had large numbers of topical TT users and that have had sufficient prostate cancer cases to provide precise estimates of association. The first was a study of cases from the National Prostate Cancer Register of Sweden compared with matched controls.26 This study included 284 prostate cancer cases who had previously received a TT prescription and found no association between TT and prostate cancer (odds ratio (OR)=1.03, 95% CI:0.90, 1.17). The second study was a retrospective cohort of 147,593 U.S. male Veterans that had one or more laboratory test-based flags for hypogonadism.27 Within this cohort, 58,617 received TT, and a total of 1,439 prostate cancers were diagnosed. The adjusted hazard ratio (HR) for the association between TT and prostate cancer was 0.90 (95% CI:0.81, 1.01). Associations in these two studies did not vary by route of TT administration, time between therapy and risk period, or duration of treatment.
The Swedish study26, discussed above, also observed that TT was positively-associated with more favorable-risk prostate cancer (OR=1.35, 95%CI:1.16, 1.56) and negatively-associated with aggressive cancer (OR=0.50, 95%CI:0.37, 0.67). The positive association with favorable-risk prostate cancer was already apparent and strongest within the first year of TT, leading the authors to suggest detection bias. The negative association with aggressive cancer only became apparent after the first year of TT, which the authors speculated may be the result of hypogonadal-genesis of poorly differentiated prostate cancer that is reversible with short-term TT. However, the study of U.S. Veterans found no association between TT use and aggressive prostate cancer (HR=0.89; 95% CI:0.70, 1.13). Although we could not assess prostate cancer stage and grade in this study, our findings of inverse associations between TT and the outcomes of metastatic prostate cancer and overall prostate cancer incidence contrast with the Swedish study and are more in line with the U.S. Veterans study.
There have been two prior studies of intravenous TT in older men of the SEER-Medicare database. The first found evidence for an inverse association with high grade prostate cancer (OR=0.84, 95%CI:0.67, 1.05) and no association with the high-risk disease proxy of receipt of primary androgen deprivation therapy (OR=0.97, 95% CI:0.74, 1.30)28. The second study found no increased risk of higher grade or higher stage prostate cancer.29 In fact, this latter SEER-Medicare study—which assessed TT exposure retrospectively amongst a cohort of confirmed prostate cancer cases—found that men who were TT-exposed were more likely to be diagnosed with moderately-differentiated than less-differentiated prostate cancer, and were more likely to be diagnosed with clinical stage T3 over T4 prostate cancer, each relative to prostate cancer cases who had not previously received TT. Combined with our findings herein, the results of these four large observational studies support a hypothesis of no increased risk of prostate cancer amongst men who receive TT and acknowledge the possibility that an inverse association may exist.26
An inverse association between TT and prostate cancer could be attributable to residual confounding. For example, very low endogenous free testosterone levels could decrease the risk of prostate cancer6 and increase the likelihood of receiving TT; although it is important to note that, amongst those diagnosed with hypogonadism, ~80% of TT prescriptions are based on tests of testosterone and ~15% are based on tests of free testosterone.27 Another example is diabetes has been inversely associated with prostate cancer30 and correlates with hypogonadism, increasing the likelihood of TT.31 Residual confounding would have to be strong, given our estimate that an unmeasured confounder would have to have an IRR of 1.92 with both TT and with prostate cancer to account for the TT-prostate cancer association. However, residual confounding is supported by attenuation to the null with increased pre-index time in the PDE5i comparison, which increases the likelihood of ascertaining confounding factors. Contraindication bias could also explain the observed association, whereby factors and symptoms perceived by the physician to be related with a higher risk of prostate cancer (e.g., family history, borderline PSA, urogenital symptoms) may reduce the likelihood of the physician prescribing TT,32 which would have the effect of causing a decreased prostate cancer incidence rate in the TT group and increased rate in the unexposed group. Lastly, the inverse association between TT and prostate cancer could be the result of a biological effect, such as causing an increase in pro-apoptotic signaling during early prostate carcinogenesis.33 Any biological effect would have to be able to explain the immediacy of the observed association and the general attenuation of effect over the 2-year period following initial exposure. It is true that the estimate with the unexposed comparison was similar in the last time period (adjusted IRR≥24 months=0.74, 95%CI:0.62, 0.88) as the overall association, but the equivalent estimate with the PDE5i comparison group was null (adjusted IRR≥24 months=0.96, 95%CI:0.87, 1.07), a pattern which mirrors the results of another recent study.25 Lastly, there was no evidence that length of time of TT exposure altered the rate of prostate cancer within the TT group.
Strengths of our study include use of a large database that enabled assessment of a younger population (unlike SEER-Medicare), and use of a validated prostate cancer definition with high sensitivity (91.2%) and positive predictive value (81.7%) that were similar between comparison groups, thus mitigating outcome misclassification bias.9 Limitations include lack of cancer stage and grade, lack of an ability to offer a precise estimate of TT in relation to metastatic or aggressive prostate cancer, reliance on claims/payer data to infer clinical variables, availability of PSA and testosterone concentrations for only a subset of subjects, data are not informative about long latency between exposure and outcome, and limited ability or inability to assess differences by age at initial TT exposure, race, specific forms/regimens of TT, or hypogonadal subtype.
This study provides evidence that men who receive TT do not have a higher rate of prostate cancer than unexposed men or men receiving PDE5i. The inverse association between TT and prostate cancer could be the result of residual confounding, contraindication bias, or undefined biologic effect.
Supplementary Material
Acknowledgements
We wish to thank Bola Ekezue,PhD, Nandini Selvam,PhD, Gayathri Sridhar,PhD, and Anna Wallace, PhD for their help and support in conducting this study.
FUNDING
Intramural Program of the National Cancer Institute, National Institutes of Health, Department of Health and Human Services.
Intramural Funding Project Number: Z01 CP010180
Footnotes
FINANCIAL DISCLOSURES
DCB, LEP, and SL are employees of HealthCore, Inc. a subsidiary of Anthem Inc. PTC is a former employee of HealthCore, Inc. a subsidiary of Anthem Inc. and holds stock in Anthem Inc.
DATA ANALYSIS INTEGRITY
DCB had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
References
- 1.Layton JB, Kim Y, Alexander GC, Emery SL. Association Between Direct-to-Consumer Advertising and Testosterone Testing and Initiation in the United States, 2009–2013. JAMA. 2017;317(11):1159–1166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Baillargeon J, Kuo Y, Westra JR, Urban RJ, Goodwin JS. Testosterone prescribing in the united states, 2002–2016. JAMA. 2018;320(2):200–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018. [DOI] [PubMed] [Google Scholar]
- 4.Snyder PJ, Bhasin S, Cunningham GR, et al. Lessons from the Testosterone Trials. Endocr Rev. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.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]
- 6.Watts EL, Appleby PN, Perez-Cornago A, et al. Low Free Testosterone and Prostate Cancer Risk: A Collaborative Analysis of 20 Prospective Studies. Eur Urol 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dupree JM, Langille GM, Khera M, Lipshultz LI. The safety of testosterone supplementation therapy in prostate cancer. Nat Rev Urol. 2014;11(9):526–530. [DOI] [PubMed] [Google Scholar]
- 8.Klotz L Testosterone therapy and prostate cancer--safety concerns are well founded. Nat Rev Urol. 2015;12(1):48–54. [DOI] [PubMed] [Google Scholar]
- 9.Parlett LE, Beachler DC, Lanes S, Hoover RN, Cook MB. Validation of an Algorithm for Claims-based Incidence of Prostate Cancer. Epidemiology. 2019;30(3):466–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Falchook AD, Dusetzina SB, Tian F, Basak R, Selvam N, Chen RC. Aggressive End-of-Life Care for Metastatic Cancer Patients Younger Than Age 65 Years. J Natl Cancer Inst. 2017;109(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wu Y, Qu X, Wang Y, et al. Effect of phosphodiesterase type 5 inhibitors on prostate cancer risk and biochemical recurrence after prostate cancer treatment: A systematic review and meta-analysis. Andrologia. 2019;51(2):e13198. [DOI] [PubMed] [Google Scholar]
- 12.Dolan MT, Kim S, Shao YH, Lu-Yao GL. Authentication of Algorithm to Detect Metastases in Men with Prostate Cancer Using ICD-9 Codes. Epidemiol Res Int. 2012;2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.McCaffrey DF, Ridgeway G, Morral AR. Propensity score estimation with boosted regression for evaluating causal effects in observational studies. Psychol Methods. 2004;9(4):403–425. [DOI] [PubMed] [Google Scholar]
- 14.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28(25):3083–3107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bang H, Robins JM. Doubly robust estimation in missing data and causal inference models. Biometrics. 2005;61(4):962–973. [DOI] [PubMed] [Google Scholar]
- 16.Brenner H, Gefeller O. Use of the positive predictive value to correct for disease misclassification in epidemiologic studies. Am J Epidemiol. 1993;138(11):1007–1015. [DOI] [PubMed] [Google Scholar]
- 17.VanderWeele TJ, Ding P. Sensitivity Analysis in Observational Research: Introducing the E-Value. Ann Intern Med. 2017;167(4):268–274. [DOI] [PubMed] [Google Scholar]
- 18.Mathur MB, Ding P, Riddell CA, VanderWeele TJ. Website and R Package for Computing E-Values. Epidemiology.Publish Ahead of Print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Debruyne FM, Behre HM, Roehrborn CG, et al. Testosterone treatment is not associated with increased risk of prostate cancer or worsening of lower urinary tract symptoms: prostate health outcomes in the Registry of Hypogonadism in Men. BJU Int. 2017;119(2):216–224. [DOI] [PubMed] [Google Scholar]
- 20.Haider A, Zitzmann M, Doros G, Isbarn H, Hammerer P, Yassin A. Incidence of prostate cancer in hypogonadal men receiving testosterone therapy: observations from 5-year median followup of 3 registries. J Urol. 2015;193(1):80–86. [DOI] [PubMed] [Google Scholar]
- 21.Feneley MR, Carruthers M. Is testosterone treatment good for the prostate? Study of safety during long-term treatment. J Sex Med. 2012;9(8):2138–2149. [DOI] [PubMed] [Google Scholar]
- 22.Shabsigh R, Crawford ED, Nehra A, Slawin KM. Testosterone therapy in hypogonadal men and potential prostate cancer risk: a systematic review. Int J Impot Res. 2009;21(1):9–23. [DOI] [PubMed] [Google Scholar]
- 23.Calof OM, Singh AB, Lee ML, et al. Adverse events associated with testosterone replacement in middle-aged and older men: a meta-analysis of randomized, placebo-controlled trials. J Gerontol A Biol Sci Med Sci. 2005;60(11):1451–1457. [DOI] [PubMed] [Google Scholar]
- 24.Cui Y, Zong H, Yan H, Zhang Y. The effect of testosterone replacement therapy on prostate cancer: a systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 2014;17(2):132–143. [DOI] [PubMed] [Google Scholar]
- 25.Santella C, Renoux C, Yin H, Yu OHY, Azoulay L. Testosterone Replacement Therapy and the Risk of Prostate Cancer in Men with Late-Onset Hypogonadism. Am J Epidemiol. 2019. [DOI] [PubMed] [Google Scholar]
- 26.Loeb S, Folkvaljon Y, Damber JE, Alukal J, Lambe M, Stattin P. Testosterone Replacement Therapy and Risk of Favorable and Aggressive Prostate Cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2017;35(13):1430–1436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Walsh TJ, Shores MM, Krakauer CA, et al. Testosterone treatment and the risk of aggressive prostate cancer in men with low testosterone levels. PLOS ONE. 2018;13(6):e0199194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Baillargeon J, Kuo YF, Fang X, Shahinian VB. Long-term Exposure to Testosterone Therapy and the Risk of High Grade Prostate Cancer. J Urol. 2015;194(6):1612–1616. [DOI] [PubMed] [Google Scholar]
- 29.Kaplan AL, Hu JC. Use of testosterone replacement therapy in the United States and its effect on subsequent prostate cancer outcomes. Urology. 2013;82(2):321–326. [DOI] [PubMed] [Google Scholar]
- 30.Jian Gang P, Mo L, Lu Y, Runqi L, Xing Z. Diabetes mellitus and the risk of prostate cancer: an update and cumulative meta-analysis. Endocr Res. 2015;40(1):54–61. [DOI] [PubMed] [Google Scholar]
- 31.Holmboe SA, Jensen TK, Linneberg A, et al. Low Testosterone: A Risk Marker Rather Than a Risk Factor for Type 2 Diabetes. J Clin Endocrinol Metab. 2016;101(8):3180–3190. [DOI] [PubMed] [Google Scholar]
- 32.Gooren LJ, Behre HM, Saad F, Frank A, Schwerdt S. Diagnosing and treating testosterone deficiency in different parts of the world. Results from global market research. Aging Male. 2007;10(4):173–181. [DOI] [PubMed] [Google Scholar]
- 33.Friedman AE. Isbarn Re: Hendrik, Pinthus Jehonathan H., Marks Leonard S., et al. Testosterone and prostate cancer: revisiting old paradigms. Eur Urol 2009;56:48–56. Eur Urol. 2009;56(5):e23; author reply e24. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.