Background: While androgen deprivation therapy (ADT) is a common treatment for prostate cancer, little is known regarding its long-term health effects, particularly as it relates to the development of second primary malignancies. Therefore, the goal of this study was to assess the association between ADT use and second primary malignancies among men diagnosed with localized prostate cancer.

Methods: We assessed whether use of ADT (specifically, gonadotropin-releasing hormone agonists) was associated with the development of second primary malignancies in a retrospective cohort of 24,038 men ages more than 18 years who were diagnosed with localized prostate cancer between 1998 and 2007, and followed through 2009. We used proportional hazards regression to estimate the risk of developing a second primary cancer among men who were treated with ADT compared with men who were not.

Results: Men who were treated with ADT were not more likely to develop any second primary malignancy compared with those who were not treated with ADT after adjustment for age, race, date of diagnosis, utilization, clinical stage, Gleason score, and radiation therapy [HR, 1.10; 95% confidence interval (CI), 0.98–1.22)]. Radiotherapy, diabetes, and obesity did not modify the association between ADT use and second primary cancer risk.

Conclusion: Our results suggest that among men with localized prostate cancer, ADT is not associated with an increased risk of second primary malignancies.

Impact: When evaluating the risks and benefits of using ADT as a treatment for localized prostate cancer, considering the risk of second primary malignancies may not be clinically important. Cancer Epidemiol Biomarkers Prev; 22(2); 313–6. ©2013 AACR.

While androgen deprivation therapy (ADT) is a common treatment for prostate cancer, little is known regarding its long-term health effects. However, the use of ADT remains a common, albeit controversial, treatment option for localized prostate cancer (1). Previous studies have linked ADT use to increased risk of osteoporosis, fractures, and cardiovascular outcomes (2–4). Gillessen and colleagues found a positive association between long-term use of ADT and the development of colorectal cancer (5). However, it remains unclear whether the use of ADT increases the risk of other second primary malignancies. Because ADT is associated with other known risk factors for cancer such as diabetes and obesity (2, 3, 6), its use may potentially influence second cancer development through these pathways. Also, it has been hypothesized that androgens may protect against carcinogenesis (7, 8). It is plausible that the suppression of androgen may then promote the progression of carcinogenesis. Therefore, the goal of this study was to assess the association between ADT use and second primary malignancies among a cohort of men diagnosed with localized prostate cancer.

A retrospective cohort of 24,038 men ages 18 years and older enrolled in Kaiser Permanente Southern California (Pasadena, CA) health plan who were diagnosed with localized prostate cancer between 1998 and 2007 were followed through 2009 for the development of second primary cancers as identified through the National Cancer Institute-Surveillance Epidemiology and End Results (NCI-SEER)–affiliated cancer registry. Men diagnosed with a second primary cancer or who died from it within 6 months of prostate cancer diagnosis (N = 173), had a history of chemotherapy treatment (n = 29), or who underwent orchiectomy (N = 82) were excluded. ADT use was defined as any use of a gonadotropin-releasing hormone (GnRH) agonist and was treated as a time-dependent exposure so that men could contribute person-time to the ADT group after treatment initiation and to the no ADT group either before ADT initiation, or if they never received ADT during follow-up. Proportional hazards regression was used to estimate the risk of a second primary cancer among men who were treated with ADT compared with men who were not after adjustment for age, race, date of prostate cancer diagnosis, utilization, prostate cancer clinical stage, Gleason score, and radiation therapy. Censoring occurred at development of second cancer, death, health plan disenrollment, or end of study, whichever occurred first. This study had 90% power to detect a difference in effect of 1.25 or more (0.8 or smaller). All analyses were conducted using SAS 9.3 with a α level of 0.05.

Of the 24,038 men with localized prostate cancer, 1,359 (5.7%) were diagnosed with a second primary cancer during follow-up. The most common site was colorectal (0.78%), followed by lung/bronchus (0.75%), bladder (0.55%), and kidney (0.32%). Men who were diagnosed with a second primary cancer were older, had lower income and education, and were more likely to have a history of type 2 diabetes and hypertension. (Table 1) Men who were treated with ADT were equally as likely to develop any second primary cancer (regardless of site) as men who were not treated with ADT after adjustment for age, race, date of prostate cancer diagnosis, utilization, clinical stage, Gleason score, and radiotherapy (HR, 1.10; 95% confidence interval (CI), 0.98–1.22; Table 2). Radiotherapy, diabetes, and obesity did not modify the association between ADT use and second primary cancer risk. (Data not shown)

Table 1.

Demographic and medical characteristics by second primary cancer status (N = 24,038)

Any second primary cancer N = 1,359No second primary cancer N = 22,679
DemographicsN (%)N (%)Pa
Age at baseline    
 40–50 18 (1.3%) 1,092 (4.8%) <0.0001 
 50–60 190 (14%) 5,551 (24.5%)  
 60–70 532 (39.1%) 9,139 (40.3%)  
 70–80 506 (37.2%) 5,498 (24.2%)  
 80+ 113 (8.3%) 1,388 (6.1%)  
Race/ethnicity    
 White 943 (69.4%) 13,882 (61.2%) <0.0001 
 Black 249 (18.3%) 3,927 (17.3%)  
 Asian/PI 49 (3.6%) 1,307 (5.8%)  
 Hispanic 112 (8.2%) 2,969 (13.1%)  
 Other/Unknown 6 (0.4%) 594 (2.6%)  
Household median income    
 0%–25% 166 (30.5%) 3,602 (26.5%) 0.0201 
 25%–50% 148 (27.2%) 3,390 (24.9%)  
 50%–75% 126 (23.2%) 3,358 (24.7%)  
 75%–100% 104 (19.1%) 3,239 (23.8%)  
Education, y    
 Median (q1,q3) 38.6 (24.6,55.0) 40.8 (26.2,57.3) 0.0196 
Medical History    
BMI    
 Underweight (<18.5) 8 (0.8%) 75 (0.4%) 0.0218 
 Normal (18.5–24.9) 250 (25.2%) 4,124 (22%)  
 Overweight (25–29.9) 448 (45.1%) 8,733 (46.6%)  
 Obese (30 or greater) 288 (29%) 5,814 (31%)  
History of type 2 diabetes    
 No 900 (66.2%) 16,714 (73.7%) <0.0001 
 Yes 459 (33.8%) 5,965 (26.3%)  
History of hypertension    
 No 143 (10.5%) 4,500 (19.8%) <0.0001 
 Yes 1,216 (89.5%) 18,179 (80.2%)  
Charlson index    
 0 548 (62.6%) 12,373 (68.4%) 0.0003 
 1–2 289 (33%) 4,848 (26.8%)  
 3+ 39 (4.5%) 856 (4.7%)  
Utilization (No. of visits)    
 Median (q1, q3) 19.0 (10.0,31.0) 15.0 (8.0,26.0) <0.0001 
Prostate cancer characteristics    
Stage at diagnosis    
 I 14 (1%) 87 (0.4%) 0.0016 
 II 1,345 (99%) 22,591 (99.6%)  
Radiation therapy    
 No 1,010 (74.3%) 18,122 (79.9%) <0.0001 
 Yes 349 (25.7%) 4,557 (20.1%)  
Gleason score    
 ≤7 1,153 (84.8%) 19,866 (87.6%) 0.0029 
 >7 206 (15.2%) 2,813 (12.4%)  
Any second primary cancer N = 1,359No second primary cancer N = 22,679
DemographicsN (%)N (%)Pa
Age at baseline    
 40–50 18 (1.3%) 1,092 (4.8%) <0.0001 
 50–60 190 (14%) 5,551 (24.5%)  
 60–70 532 (39.1%) 9,139 (40.3%)  
 70–80 506 (37.2%) 5,498 (24.2%)  
 80+ 113 (8.3%) 1,388 (6.1%)  
Race/ethnicity    
 White 943 (69.4%) 13,882 (61.2%) <0.0001 
 Black 249 (18.3%) 3,927 (17.3%)  
 Asian/PI 49 (3.6%) 1,307 (5.8%)  
 Hispanic 112 (8.2%) 2,969 (13.1%)  
 Other/Unknown 6 (0.4%) 594 (2.6%)  
Household median income    
 0%–25% 166 (30.5%) 3,602 (26.5%) 0.0201 
 25%–50% 148 (27.2%) 3,390 (24.9%)  
 50%–75% 126 (23.2%) 3,358 (24.7%)  
 75%–100% 104 (19.1%) 3,239 (23.8%)  
Education, y    
 Median (q1,q3) 38.6 (24.6,55.0) 40.8 (26.2,57.3) 0.0196 
Medical History    
BMI    
 Underweight (<18.5) 8 (0.8%) 75 (0.4%) 0.0218 
 Normal (18.5–24.9) 250 (25.2%) 4,124 (22%)  
 Overweight (25–29.9) 448 (45.1%) 8,733 (46.6%)  
 Obese (30 or greater) 288 (29%) 5,814 (31%)  
History of type 2 diabetes    
 No 900 (66.2%) 16,714 (73.7%) <0.0001 
 Yes 459 (33.8%) 5,965 (26.3%)  
History of hypertension    
 No 143 (10.5%) 4,500 (19.8%) <0.0001 
 Yes 1,216 (89.5%) 18,179 (80.2%)  
Charlson index    
 0 548 (62.6%) 12,373 (68.4%) 0.0003 
 1–2 289 (33%) 4,848 (26.8%)  
 3+ 39 (4.5%) 856 (4.7%)  
Utilization (No. of visits)    
 Median (q1, q3) 19.0 (10.0,31.0) 15.0 (8.0,26.0) <0.0001 
Prostate cancer characteristics    
Stage at diagnosis    
 I 14 (1%) 87 (0.4%) 0.0016 
 II 1,345 (99%) 22,591 (99.6%)  
Radiation therapy    
 No 1,010 (74.3%) 18,122 (79.9%) <0.0001 
 Yes 349 (25.7%) 4,557 (20.1%)  
Gleason score    
 ≤7 1,153 (84.8%) 19,866 (87.6%) 0.0029 
 >7 206 (15.2%) 2,813 (12.4%)  

aMissing numbers were excluded in the estimation of P values (2-sided).

Abbreviations: BMI, body mass index.

Table 2.

Crude and adjusted HR and 95% CIs of the associations between ADT use and second primary cancers among 24,038 men with localized prostate cancer

No. of casesCrude HR (95% CI)Adjusted HR (95% CIb)
Any second primary cancer 
 No ADT 719 1.0 (ref) 1.0 (ref) 
 ADTa 640 0.94 (0.79–1.12) 0.96 (0.79–1.17) 
Colorectal    
 No ADT 113 1.0 (ref) 1.0 (ref) 
 ADT 74 0.77 (0.44–1.35) 0.91 (0.47–1.74) 
Lung    
 No ADT 78 1.0 (ref) 1.0 (ref) 
 ADT 100 1.27 (0.84–1.91) 1.21 (0.71–2.06) 
Bladder    
 No ADT 61 1.0 (ref) 1.0 (ref) 
 ADT 71 0.67 (0.38–1.18) 0.53 (0.26–1.06) 
Kidney    
 No ADT 49 1.0 (ref) 1.0 (ref) 
 ADT 27 1.92 (0.76–4.82) 2.16 (0.51–9.23) 
Other    
 No ADT 418 1.0 (ref) 1.0 (ref) 
 ADT 368 0.86 (0.68–1.08) 0.92 (0.71–1.20) 
No. of casesCrude HR (95% CI)Adjusted HR (95% CIb)
Any second primary cancer 
 No ADT 719 1.0 (ref) 1.0 (ref) 
 ADTa 640 0.94 (0.79–1.12) 0.96 (0.79–1.17) 
Colorectal    
 No ADT 113 1.0 (ref) 1.0 (ref) 
 ADT 74 0.77 (0.44–1.35) 0.91 (0.47–1.74) 
Lung    
 No ADT 78 1.0 (ref) 1.0 (ref) 
 ADT 100 1.27 (0.84–1.91) 1.21 (0.71–2.06) 
Bladder    
 No ADT 61 1.0 (ref) 1.0 (ref) 
 ADT 71 0.67 (0.38–1.18) 0.53 (0.26–1.06) 
Kidney    
 No ADT 49 1.0 (ref) 1.0 (ref) 
 ADT 27 1.92 (0.76–4.82) 2.16 (0.51–9.23) 
Other    
 No ADT 418 1.0 (ref) 1.0 (ref) 
 ADT 368 0.86 (0.68–1.08) 0.92 (0.71–1.20) 

aADT defined as use of GnRH agonist only.

bAdjusted for age, race, year of prostate cancer diagnosis, utilization (healthcare visits), stage, Gleason score, and radiotherapy.

In this study of men with localized prostate cancer, ADT therapy was not associated with an increased risk of second primary malignancies. In contrast to a previous study (5), ADT use was also not associated with an increased risk of colorectal cancer. It is possible that the inclusion of younger men in this study may have resulted in a lower risk of secondary cancers overall, thus resulting in more conservative estimates of the risk of second primary cancers than reported previously.

While this study is one of the first to evaluate the risk of developing any type of second primary malignancy after ADT treatment, there are potential limitations to consider. Because of the complexities of determining injectable drug dosing in our data, ADT dose and duration were not available. Therefore, whether ADT dose influences second primary cancer development could not be assessed. Confounding by factors not measured in this study, such as the use of screening for secondary cancers, might have influenced the results. However, the use of screening is not thought to differ by ADT use, thus limiting the effect of this bias. Finally, because this study included only men with localized prostate cancer, generalizing these findings to men with advanced disease is limited.

In conclusion, ADT treatment for localized prostate cancer was not associated with an increased risk of second primary cancers. When evaluating the risks and benefits of using ADT as a treatment for localized prostate cancer, considering the risk of secondary cancers may not be clinically important.

No potential conflicts of interest were disclosed.

Conception and design: L.P. Wallner, R. Haque

Development of methodology: L.P. Wallner, S.J. Jacobsen, R. Haque

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R. Wang, R. Haque

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L.P. Wallner, R. Wang, R. Haque

Writing, review, and/or revision of the manuscript: L.P. Wallner, S.J. Jacobsen, R. Haque

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S.J. Jacobsen, R. Haque

Study supervision: R. Haque

This study was supported by Research & Evaluation, Kaiser Permanente Southern California with supplemental funding through NCI R01 CA142934-01 (Potosky)

1.
Kawakami
J
,
Cowan
JE
,
Elkin
EP
,
Latini
DM
,
DuChane
J
,
Carroll
PR
, et al
Androgen-deprivation therapy as primary treatment for localized prostate cancer: data from Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE)
.
Cancer
2006
;
106
:
1708
14
.
2.
Keating
NL
,
O'Malley
AJ
,
Freedland
SJ
,
Smith
MR
. 
Diabetes and cardiovascular disease during androgen deprivation therapy: observational study of veterans with prostate cancer
.
J Natl Cancer Inst
2010
;
102
:
39
46
.
3.
Kintzel
PE
,
Chase
SL
,
Schultz
LM
,
O'Rourke
TJ
. 
Increased risk of metabolic syndrome, diabetes mellitus, and cardiovascular disease in men receiving androgen deprivation therapy for prostate cancer
.
Pharmacotherapy
2008
;
28
:
1511
22
.
4.
Shahinian
VB
,
Kuo
YF
,
Freeman
JL
,
Goodwin
JS
. 
Risk of fracture after androgen deprivation for prostate cancer
.
N Engl J Med
2005
;
352
:
154
64
.
5.
Gillessen
S
,
Templeton
A
,
Marra
G
,
Kuo
YF
,
Valtorta
E
,
Shahinian
VB
. 
Risk of colorectal cancer in men on long-term androgen deprivation therapy for prostate cancer
.
J Natl Cancer Inst
2010
;
102
:
1760
70
.
6.
Collier
A
,
Ghosh
S
,
McGlynn
B
,
Hollins
G
. 
Prostate cancer, androgen deprivation therapy, obesity, the metabolic syndrome, type 2 diabetes, and cardiovascular disease: a review
.
Am J Clin Oncol
2012
;
35
:
504
9
.
7.
Aoki
K
,
Nakajima
A
,
Mukasa
K
,
Osawa
E
,
Mori
Y
,
Sekihara
H
. 
Prevention of diabetes, hepatic injury, and colon cancer with dehydroepiandrosterone
.
J Steroid Biochem Mol Biol
2003
;
85
:
469
72
.
8.
Izbicki
JR
,
Hamilton
SR
,
Wambach
G
,
Harnisch
E
,
Wilker
DK
,
Dornschneider
G
, et al
Effects of androgen manipulations on chemically induced colonic tumours and on macroscopically normal colonic mucosa in male Sprague-Dawley rats
.
Br J Cancer
1990
;
61
:
235
40
.