Abstract

Background: Androgens play a role in the development of both androgenic alopecia, commonly known as male pattern baldness, and prostate cancer. We set out to study if early-onset androgenic alopecia was associated with an increased risk of prostate cancer later in life.

Patients and methods: A total of 669 subjects (388 with a history of prostate cancer and 281 without) were enrolled in this study. All subjects were asked to score their balding pattern at ages 20, 30 and 40. Statistical comparison was subsequently done between both groups of patients.

Results: Our study revealed that patients with prostate cancer were twice as likely to have androgenic alopecia at age 20 [odds ratio (OR) 2.01, P = 0.0285]. The pattern of hair loss was not a predictive factor for the development of cancer. There was no association between early-onset alopecia and an earlier diagnosis of prostate cancer or with the development of more aggressive tumors.

Conclusions: This study shows an association between early-onset androgenic alopecia and the development of prostate cancer. Whether this population can benefit from routine prostate cancer screening or systematic use of 5-alpha reductase inhibitors as primary prevention remains to be determined.

introduction

Androgenic alopecia, commonly known as male pattern baldness, is a common disorder affecting almost 50% of men throughout their lifetime. It is usually seen in older men, but it may arise precociously. A link between male pattern baldness and androgens has previously been documented [1–3]. Androgens also play a role in the development and growth of prostate cancer. Finasteride, a type II 5-alpha reductase inhibitor that blocks the conversion of testosterone to dihydrotestosterone, is used for the treatment of androgenic alopecia and has been shown to decrease the incidence of prostate cancer [4, 5].

Despite this, a direct link between male pattern baldness and prostate cancer has not been consistently shown [6–12]. Authors have suggested that inconsistencies in results may be due to differences in the patterns of hair loss and in the time frame of development of androgenic alopecia. Alopecia occurring in younger males (before the age of 30), as well as hair loss implicating the vertex, may be a precursor to developing prostate cancer later in life [9, 10, 13]. To help answer these questions, we undertook a case–control study to evaluate the impact of early-onset androgenic alopecia on the incidence of prostate cancer.

patients and methods

Case subjects were all successive patients with a diagnosis of prostate cancer seen in a radiation oncology follow-up clinic at one of the participating institutions in France (Hôpital Européen Georges Pompidou, Paris; Institut Curie, Paris; and Institut Claudius Regaud, Toulouse). Controls were chosen from the same hospital databases and were patients with no history of prostate cancer or hormonal pathologies. Cases and controls groups were frequency matched according to date of birth.

Starting in September 2004, after approval by respective hospitals’ ethics boards, and for a period of 28 months, participants were contacted by mail and asked to complete a short questionnaire. All subjects were asked if there was a personal history of prostate cancer (yes/no checkbox), as well as a history of balding or prostate cancer in the father. Participants then scored their balding pattern at ages 20, 30 and 40, according to a set of four pictures (Figure 1), adapted from the Hamilton–Norwood scale [11, 14]. The first picture (stage I) represented no balding. The second picture (stage II) showed frontal hair loss, while the third picture (stage III) showed vertex hair loss. The fourth picture (stage IV) represented both concurrent frontal and vertex balding. This type of self-reporting questionnaire has been validated for retrospective assessment of alopecia [15, 16].

Figure 1.

Hair loss pattern according to a modified Hamilton–Norwood scale, used by patients to score alopecia at ages 20, 30 and 40.

Figure 1.

Hair loss pattern according to a modified Hamilton–Norwood scale, used by patients to score alopecia at ages 20, 30 and 40.

Physicians of all respondents were asked to complete a separate questionnaire, confirming the presence or not of a history of prostate cancer in the patient. For case subjects, the physicians detailed the history of the prostate cancer: age at diagnosis, initial stage of disease [tumor–node–metastasis stage, Gleason score and initial prostate-specific antigen (PSA)], primary treatment received, failure of treatment, delay between treatment and failure and last medical impression of disease (asked to check between remission, failure or metastatic disease).

statistical analysis

Sample size was calculated based on computer simulation using prevalence found in previous studies of early-onset alopecia at 40 years of age. The power was to be 77% with an accrual of 1500 cases. Data was compiled at a single institution (Institut Curie). Ages between cases and controls were compared using the Student's t-test. We used a chi-square test to analyze the relationship between alopecia at various ages (ages 20, 30 and 40) and the prevalence of prostate cancer, as well as the relationship with aggressive prostate cancers (T3–4 tumors, PSA > 20, Gleason score ≥7). Odd ratios and confidence intervals (CIs) were obtained from 2 × 2 tables. Results were adjusted for potential cofounders (age and family history) by logistic regression. All tests were two tailed at 5%. The R 2.5.0 software was used for all analysis.

results

The response rate for cases, controls and physicians was 88%, 76% and 100%, respectively. A total of 669 subjects (388 cases, 281 controls) were included in the study. Breakdown of case subjects is presented in Table 1. The two groups were well balanced for age and family history of prostate cancer. The mean age (and standard deviation) for the case subjects was 67.2 (7.2) years and for the controls was 66.4 (9.1) years. Patients in the case group were diagnosed with prostate cancer between the ages of 46 and 84 [mean age 64.4 (7.0)]. There was no difference in the prevalence of alopecia or the incidence of prostate cancer in the father of cases and controls with early-onset alopecia (stage II–IV) (Table 2). Adjusted results are not presented because these were not significantly different.

Table 1.

Breakdown of case subjects

 Cases (N = 388)
 
 n 
T stage of PC 
    T1 138 35.6 
    T2 112 28.9 
    T3 121 31.2 
    T4 11 2.8 
    Tx 1.5 
N stage of PC 
    N0 283 72.9 
    N1 17 4.4 
    Nx 88 22.7 
M stage of PC 
    M0 357 92.0 
    M1 0.5 
    Mx 29 7.5 
Gleason score 
    <7 196 50.5 
    ≥7 178 45.9 
    N/A 14 3.6 
PSA 
    <10 217 55.9 
    10–20 91 23.5 
    >20 72 18.6 
    N/A 2.0 
 Cases (N = 388)
 
 n 
T stage of PC 
    T1 138 35.6 
    T2 112 28.9 
    T3 121 31.2 
    T4 11 2.8 
    Tx 1.5 
N stage of PC 
    N0 283 72.9 
    N1 17 4.4 
    Nx 88 22.7 
M stage of PC 
    M0 357 92.0 
    M1 0.5 
    Mx 29 7.5 
Gleason score 
    <7 196 50.5 
    ≥7 178 45.9 
    N/A 14 3.6 
PSA 
    <10 217 55.9 
    10–20 91 23.5 
    >20 72 18.6 
    N/A 2.0 

PC, prostate cancer; N/A, not available; PSA, prostate-specific antigen.

Table 2.

Prevalence of alopecia or incidence of prostate cancer in the father of subjects (cases or controls) with early-onset alopecia at age 20, 30 and 40

  Cases (%) Controls (%) P value 
Subject with stage II–IV alopecia at age 20 And father with alopecia 0.8 0.4 0.64 
And father with PC 2.6 3.2 0.63 
Subject with stage II–IV alopecia at age 30 And father with alopecia 3.9 3.6 0.97 
And father with PC 13.4 13.5 0.96 
Subject with stage II–IV alopecia at age 40 And father with alopecia 4.6 5.7 0.54 
And father  with PC 18.8 23.6 0.14 
  Cases (%) Controls (%) P value 
Subject with stage II–IV alopecia at age 20 And father with alopecia 0.8 0.4 0.64 
And father with PC 2.6 3.2 0.63 
Subject with stage II–IV alopecia at age 30 And father with alopecia 3.9 3.6 0.97 
And father with PC 13.4 13.5 0.96 
Subject with stage II–IV alopecia at age 40 And father with alopecia 4.6 5.7 0.54 
And father  with PC 18.8 23.6 0.14 

PC, prostate cancer.

hair loss and prostate cancer

Prevalence and pattern of hair loss by age for both case and control group subjects are presented in Table 3. Data revealed that any balding present at age 20 (stage II–IV) was associated with an increased incidence of prostate cancer later in life. When compared with control group subjects, those with prostate cancer were twice as likely to have signs of alopecia at age 20 [odds ratio (OR) 2.01, P = 0.0285, 95% CI 1.07–3.79]). This trend was lost at ages 30 or 40.

Table 3.

Hair loss pattern by age in cases and controls at ages 20, 30 and 40

 Cases (N = 388) Controls (N = 281) OR P value 
20 years 
    No balding (stage I) 351 267   
    Stage II 36 12   
    Stage III   
    Stage IV   
    Any balding (stage II–IV) 37 (9.5%) 14 (5.0%) 2.01 0.03 
30 years 
    No balding (stage I) 266 200   
    Stage II 98 61   
    Stage III 22 16   
    Stage IV   
    Any balding (stage II–IV) 122 (31.4%) 81 (28.8%) 1.13 0.47 
40 years 
    No balding (stage I) 199 141   
    Stage II 91 64   
    Stage III 65 50   
    Stage IV 33 26   
    Any balding (stage II–IV) 189 (48.7%) 140 (49.8%) 0.96 0.78 
 Cases (N = 388) Controls (N = 281) OR P value 
20 years 
    No balding (stage I) 351 267   
    Stage II 36 12   
    Stage III   
    Stage IV   
    Any balding (stage II–IV) 37 (9.5%) 14 (5.0%) 2.01 0.03 
30 years 
    No balding (stage I) 266 200   
    Stage II 98 61   
    Stage III 22 16   
    Stage IV   
    Any balding (stage II–IV) 122 (31.4%) 81 (28.8%) 1.13 0.47 
40 years 
    No balding (stage I) 199 141   
    Stage II 91 64   
    Stage III 65 50   
    Stage IV 33 26   
    Any balding (stage II–IV) 189 (48.7%) 140 (49.8%) 0.96 0.78 

OR, odds ratio.

pattern of hair loss and prostate cancer

In our population, the pattern of hair loss (frontal versus vertex versus both) was not a predictive factor for the development of cancer. There were a very low number of subjects with stage III and no subject with stage IV hair loss at age 20. At age 30, compared with no hair loss (stage I), presence of vertex alopecia, whether alone (stage III) or with frontal alopecia (stage III and IV), was not associated with an increased risk of prostate cancer (stage I versus III: OR 1.03, P = 0.92, 95% CI 0.53–2.02; stage I versus III + IV: OR 0.9, P = 0.7454, 95% CI 0.48–1.68). At age 40, results were similar (stage I versus III: OR 0.91, P = 0.6821, 95% CI 0.6–1.4; stage I versus III + IV: OR 0.91, P = 0.6052, 95% CI 0.63–1.31).

hair loss and aggressiveness of prostate cancer

Subjects with early-onset alopecia were at no increased risk of developing tumors at a younger age. Patients with no balding by age 40 had a mean age of diagnosis of prostate cancer of 64.3 years, while those with any pattern of balding by age 20 and 40 had a mean age of diagnosis of 64.4 and 64.5 years (P = 0.80), respectively. Neither was early-onset alopecia (stage II–IV) linked to a diagnosis of more aggressive prostatic tumors as defined by T3–T4 tumors, a Gleason score ≥7 or PSA > 20 (Table 4).

Table 4.

Breakdown of aggressive tumors in case subjects with and without alopecia by age 40

 Stage I (n = 199), % Stage II–IV (n = 189), % P value 
T3–T4 21.2 18.5 0.62 
Gleason score ≥ 7 46.4 48.9 0.37 
PSA > 20 19.9 17.9 0.63 
 Stage I (n = 199), % Stage II–IV (n = 189), % P value 
T3–T4 21.2 18.5 0.62 
Gleason score ≥ 7 46.4 48.9 0.37 
PSA > 20 19.9 17.9 0.63 

PSA, prostate-specific antigen.

discussion

For prostate cancer, recent publications have not been able to show an absolute benefit for screening in the general population [17, 18]. An improved knowledge of risk factors, especially those that are easily identifiable in the patient, may allow us to target a population at high risk of developing prostate cancer and that may benefit from screening or chemoprevention.

While similar biological mechanisms seem to be involved in the development of both male pattern baldness and prostate cancer, no direct link has yet been identified. Androgens, however, seem to be implicated. Finasteride blocks the conversion of testosterone to dihydrotestosterone, the active metabolite of testosterone, slowing the progression of androgenic alopecia and decreasing the incidence of prostate cancer.

In this study, we have identified male pattern baldness arising at an early age, an easily identifiable and early-occurring trait, as a risk factor for developing prostate cancer. Data in the literature on the subject is, however, conflictual. Earlier studies failed to link male pattern baldness with prostate cancer, yet these evaluated balding status at the time of diagnosis of the prostate cancer [7, 8]. This is in contrast with more contemporary studies, which suggest that androgenic alopecia is associated with an increased, yet delayed, risk of prostate cancer. In a report by Hawk et al. [10], based on a cohort of men followed prospectively for nearly 20 years, men with male pattern baldness at baseline had a greater incidence of prostate cancer (relative risk 1.5). In another study conducted at Duke University [9], it was found that men who developed vertex baldness by age 30 had nearly a twofold increase in risk of developing prostate cancer. A more recent study, however, showed opposite results. Wright et al. [19] conducted a population-based case–control study where subjects recalled hair loss pattern at age 30 and at 1 year before diagnosis of prostate cancer using showcards. In this study, alopecia at age 30 had a 29% relative risk reduction for prostate cancer.

Our study, on the other hand, revealed that patients with prostate cancer were twice as likely to have androgenic alopecia at age 20. We were unable to find an association with the pattern of hair loss and the development of cancer. This may have been due to the very low prevalence of vertex (stage III) and combined frontal/vertex alopecia (stage IV) at the ages of 20 and 30 in our population. In addition, our study was not powered to detect a small difference between these groups. Finally, we did not detect a link between early-onset androgenic alopecia and an earlier age of diagnosis of prostate cancer or with the development of more aggressive prostate tumors. Possible limitations of our study include recall and selective recall bias; however, the retrospective self-reporting of male balding patterns has been validated [16, 17] and was used in other studies. Furthermore, we believe that because alopecia can impact self-perceptions [20], most men would remember developing baldness, especially if it occurs at an early age. Another limitation of our study may be due to the fact that we did not control for other risk factors of prostate cancer such as African heritage or dietary differences; nevertheless, family history of prostate cancer was comparable in cases and controls.

conclusions

Androgenic alopecia is a common disorder usually seen in older men, but it may arise precociously. Its link to prostate cancer is unclear and there is much disparity in the literature. While androgens may be implicated in both androgenic alopecia and prostate cancer, and the incidence of both increases with age, a direct relationship has yet to be proven. We encourage further work on the subject, whether molecular or genetic.

disclosure

The authors declare no conflict of interest.

References

1.
Sinclair
R
Male pattern androgenic alopecia
BMJ
 , 
1998
, vol. 
317
 (pg. 
865
-
869
)
2.
Vierhapper
H
Nowotny
P
Maier
H
Waldhausl
W
Production rates of dihydrotestosterone in healthy men and women and in men with male pattern baldness: determination by stable isotope/dilution and mass spectrometry
J Clin Endocrinol Metab
 , 
2001
, vol. 
86
 
12
(pg. 
5762
-
5764
)
3.
Ellis
JA
Harrap
SB
The genetics of androgenic alopecia
Clin Dermatol
 , 
2001
, vol. 
19
 (pg. 
149
-
154
)
4.
Thompson
IM
Goodman
PJ
Tangen
CM
, et al.  . 
The influence of finasteride on the development of prostate cancer
N Engl J Med
 , 
2003
, vol. 
349
 
3
(pg. 
215
-
224
)
5.
Vickers
AJ
Savage
CJ
Lilja
H
Finasteride to prevent prostate cancer: should all men or only a high-risk subgroup be treated?
J Clin Oncol
 , 
2010
, vol. 
28
 
7
(pg. 
1112
-
1116
)
6.
Greenwald
P
Damon
A
Kirmss
V
Polan
AK
Physical and demographic features of men before developing cancer of the prostate
J Natl Cancer Inst
 , 
1974
, vol. 
53
 
2
(pg. 
341
-
346
)
7.
Demark-Wahnefried
W
Lesko
SM
Conaway
MR
, et al.  . 
Serum androgens: associations with prostate cancer risk and hair patterning
J Androl
 , 
1997
, vol. 
18
 
5
(pg. 
495
-
500
)
8.
Hsieh
CC
Thanos
A
Mitropoulos
D
, et al.  . 
Risk factors for prostate cancer: a case-control study in Greece
Int J Cancer
 , 
1999
, vol. 
80
 
5
(pg. 
699
-
703
)
9.
Demark-Wahnefried
W
Schildkraut
JM
Thompson
D
, et al.  . 
Early onset baldness and prostate cancer risk
Cancer Epidemiol Biomarkers Prev
 , 
2000
, vol. 
9
 
3
(pg. 
325
-
328
)
10.
Hawk
E
Breslow
RA
Graubard
BI
Male pattern baldness and clinical prostate cancer in the epidemiologic follow-up of the first National Health and Nutrition Examination Survey
Cancer Epidemiol Biomarkers Prev
 , 
2000
, vol. 
9
 
5
(pg. 
523
-
527
)
11.
Giles
GG
Severi
G
Sinclair
R
, et al.  . 
Androgenetic alopecia and prostate cancer: findings from an Australian case-control study
Cancer Epidemiol Biomarkers Prev
 , 
2002
, vol. 
11
 
6
(pg. 
549
-
553
)
12.
Faydaci
G
Bilal
E
Necmettin
P
, et al.  . 
Baldness, benign prostate hyperplasia, prostate cancer and androgen levels
Aging Male
 , 
2008
, vol. 
11
 
4
(pg. 
189
-
192
)
13.
Demark-Wahnefried
W
Schildkraut
JM
Correspondence re: E. Hawk, et al., Male pattern baldness and clinical prostate cancer in the epidemiologic follow-up of the First National Health and Nutrition Examination Survey
 
(Letter and Reply). Cancer Epidemiol Biomarkers Prev 2001; 10 (4): 415–416
14.
Norwood
OT
Male pattern baldness: classification and incidence
South Med J
 , 
1975
, vol. 
68
 (pg. 
1359
-
1365
)
15.
Taylor
R
Matassa
J
Leavy
JE
Fritschi
L
Validity of self reported male balding patterns in epidemiological studies
BMC Public Health
 , 
2004
, vol. 
4
 pg. 
60
 
16.
Littman
AJ
White
E
Reliability and validity of self-reported male balding patterns for use in epidemiologic studies
Ann Epidemiol
 , 
2005
, vol. 
15
 
10
(pg. 
771
-
772
)
17.
Andriole
GL
Grubb
RL
3rd
Buys
SS
, et al.  . 
Mortality results from a randomized prostate-cancer screening trial
N Engl J Med
 , 
2009
, vol. 
360
 (pg. 
1310
-
1319
)
18.
Schroder
FH
Hugosson
J
Roobol
MJ
, et al.  . 
Screening and prostate-cancer mortality in a randomized European study
N Engl J Med
 , 
2009
, vol. 
360
 (pg. 
1320
-
1328
)
19.
Wright
JL
Page
ST
Lin
DW
Stanford
JL
Male pattern baldness and prostate cancer risk in a population-based case-control study
Cancer Epidemiol
 , 
2010
, vol. 
34
 (pg. 
131
-
135
)
20.
Cash
TF
The psychological consequences of androgenetic alopecia: a review of the research literature
Br J Dermatol
 , 
1999
, vol. 
141
 
3
(pg. 
398
-
405
)