In this large, prospective cohort of American male health professionals, a history of severe acne, as measured by self-reported tetracycline use for four or more years, was associated with a significantly increased risk of prostate cancer. This association persisted after adjustment for correlates and risk factors of acne and prostate cancer, and factors associated with long-term tetracycline use.
Our positive study findings are similar to those from two previous prospective studies, 5, 6
one of which observed a non-significant positive association between self-reported history of acne and prostate cancer mortality,6
and the other observed a non-significant positive association between detection of P. acnes
DNA in TURP specimens and incident prostate cancer.5
Our findings differ, however, from those from two case-control studies.7, 8
The first of these studies observed a non-significant inverse association between a history of acne and prostate cancer, and more pronounced inverse associations for interviewer-observed facial acne scarring, self-reported onset of acne after 15 years of age, and greater than 24 months treatment for acne,7
while the second observed no association between self-reported history of acne during participants’ mid-teens and prostate cancer.8
One possible explanation for our positive findings is general immune sensitivity to P. acnes
in men with a history of severe acne.24
Although most boys are colonized by P. acnes
, only a small proportion go on to develop severe acne vulgaris,4
possibly due to stronger humoral or cell-mediated immune responses to P. acnes.25, 26
By extension, these same individuals may also be more likely to develop stronger inflammatory immune responses to intraprostatic P. acnes
. Persistent low-grade P. acnes
-mediated inflammation may then lead to the development of proliferative inflammatory atrophy lesions,24
regenerative lesions that have been proposed to serve as either precursor lesions for prostate cancer or as markers of an intraprostatic environment conducive to the development of prostate cancer.27
Although all of the exposed men in our study used tetracycline, an anti-P. acnes
and anti-inflammatory drug,28, 29
it is unclear what effect use of this drug would have had on low-grade intraprostatic P. acnes
infection because tetracycline has not proven useful in treating other non-acute intraprostatic infections caused by tetracycline-sensitive bacteria, such as chronic bacterial prostatitis, 30
despite its use in treating recurrent urinary tract infections (UTIs) due to bacterial prostatitis.31
For this same reason, we believe that tetracycline is unlikely to have an independent effect on prostate cancer risk because of its questionable diffusion into the non-acutely inflamed prostate. Other possible biologic explanations for our positive findings include P. acnes
-mediated enhanced immune sensitivity to other stimuli,24
greater general immune sensitivity to infection regardless of the specific pathogen or site of infection, or greater general susceptibility to infection.
Although intriguing, our findings should be viewed as preliminary for several reasons. First, they rely on a small number of exposed cases and thus may be imprecise. Second, they are based on a surrogate measure of severe acne. Although participants were reminded of acne as an indication for tetracycline use on the follow-up questionnaire, the sensitivity of long-term tetracycline use for severe acne may be limited by its availability, cost, distribution under different names, or acceptability as an acne therapy over time. Its specificity may also be limited by its prescription for other medical indications, such as rosacea,9
periodontal disease 9
and recurrent UTIs due to chronic bacterial prostatitis.31
Although some men reported these and other indications in our supplemental study, they were less likely to report alternate indications if they used tetracycline for four or more years than if they used tetracycline for less then four years, thus increasing the specificity of tetracycline as a surrogate for severe acne among the long-term users. Additionally, although indirect, our measure of severe acne may have been as or more specific than more direct measures, such as self-reported history of severe acne, because these measures may be more influenced by individual perception of acne severity. Irrespective of the sensitivity and specificity of long-term tetracycline use for severe acne, if P. acnes
infection is truly associated with prostate cancer, then misclassification of some men with a history of acne who did not use tetracycline or some men without a history of acne who used tetracycline for other indications should only have resulted in an attenuation of even stronger findings because misclassification is unlikely to have been differential by later prostate cancer status. Indeed, when we restricted the analyses to men without histories of periodontal disease or clinical prostatitis, indications for which information was available in this cohort, slightly stronger associations were observed. Although long-term tetracycline use may also serve as a surrogate for exposure to possibly harmful acne therapies used prior to or in conjunction with tetracycline therapy, we believe this is unlikely to explain our findings because most of the men in our supplemental study who reported other acne treatments used facially-oriented therapies, such as topical preparations, and ultraviolet light and X-ray therapy.
Just as long-term tetracycline use may serve as a surrogate for severe acne, severe acne may also serve as a surrogate for other measures. In previous epidemiologic studies,6, 7
acne has been used as a marker of hormonal activity, particularly androgen activity, but also possibly growth hormone and insulin-like growth factor activity. We explored this hypothesis by investigating possible associations between long-term tetracycline use and factors believed to be associated with androgen activity, such as vertex baldness later in life, and body habitus, vigorous physical activity and ejaculation frequency earlier in life. However, none of these factors were associated with long-term tetracycline use, except for vertex baldness, which was inversely associated.
In summary, men with a history of severe acne, as measured by tetracycline use for four or more years, were significantly more likely to develop prostate cancer than men without a history of severe acne in this large prospective study of American health professionals. Although intriguing, these findings should be interpreted cautiously because of the small number of exposed cases, the indirect assessment of severe acne and the complex and incompletely defined etiology of acne, which is not limited to P. acnes infection. Therefore, additional biologic and epidemiologic studies are necessary to determine and elucidate the possible role of P. acnes infection in prostate carcinogenesis.