Acne and Cancer, What the Research Shows

Written by Dr Gerard Ee, MBBS (UK), MRCS (Edinburgh), PGDip Practical Dermatology (Cardiff), sole author. First published November 2018. Last medically reviewed and updated 1 September 2026.

Acne and cancer, the short answer

There is no convincing evidence that acne causes cancer, and acne is not a recognised sign of most internal cancers. Several observational studies have reported associations between severe adolescent acne and later prostate cancer, breast cancer or melanoma, although the findings are inconsistent and do not establish causation. Certain cancer medicines, particularly epidermal growth factor receptor (EGFR) inhibitors, produce an acne-like eruption in a high proportion of patients. Very rarely, a hormone-secreting tumour produces abrupt acne alongside other features of androgen excess, and in that setting it is the combination rather than the acne that warrants investigation. Acne occurring on its own is not evidence of cancer.

The sections below set out what each study measured, where the results disagree, and which findings are too unstable to act on. This article sits within my what is acne series.

The androgen hypothesis linking acne and prostate cancer

Both the sebaceous gland and the prostate are androgen-responsive. Androgens rise at puberty, increase sebum production in the skin and stimulate growth of the prostate. The hypothesis follows from that shared dependence, because if a man's lifetime androgen exposure were unusually high both tissues would be affected, and acne visible on the face would then serve as an external marker of an androgen environment acting unseen on the prostate. The reasoning is coherent, which is why it has been tested repeatedly in cohorts assembled for other purposes.

The hypothesis is weakened at its first step, because the premise it rests on does not hold for most patients. Most people with acne do not have raised circulating androgen levels. The mechanism is end-organ sensitivity rather than systemic excess, meaning that the gland responds unusually strongly to hormone levels that are themselves normal. Acne-prone skin shows increased local 5-alpha-reductase activity, which converts testosterone to the more potent dihydrotestosterone (DHT) within the gland itself, together with a higher density of androgen receptors on which that DHT acts. The gland is also a site of active steroidogenesis, producing and metabolising androgens largely independently of blood levels, with insulin and insulin-like growth factor 1 (IGF-1) amplifying the response [1, 2]. Acne is therefore not reliable evidence of a high-androgen state but evidence of a gland that responds strongly to ordinary hormone levels, which is a different claim and one that weakens the reasoning behind every study that follows.

What the prostate cancer studies found

Four studies carry most of the weight in this question, and they do not point in the same direction. They are set out below with the design of each stated first, because the design accounts for most of the disagreement between them.

Swedish conscript cohort, 243,187 men followed for a median of 36.7 years [3]

This was a prospective cohort study. Men who had acne recorded at their conscription examination went on to develop prostate cancer at a somewhat higher rate than men who did not. The difference was larger for advanced disease, and larger again within the small group classified as having severe acne. The study is large, the follow-up is long, and the exposure was recorded prospectively rather than remembered. Its principal limitation is that acne was assessed on a single occasion, at the age of 18, so a man whose acne cleared that year and a man whose acne persisted into his thirties are recorded identically.

Meta-analysis of eight studies [4]

This was a pooled analysis of eight studies. Taken together, the eight studies show no association at all. Separating them by design changes the answer. The studies that followed men forward in time found a modest increase, while those that asked men already diagnosed with cancer to recall their teenage skin found nothing. The eight also disagreed with one another to a considerable degree. It remains the best single summary of the evidence, and the divergence between the two designs, rather than the pooled figure, is what the analysis demonstrates.

Mexican case-control study [5]

This was a case-control study. Later puberty and the absence of adolescent acne were both associated with a lower rate of high-grade prostate cancer, which is consistent in direction with the Swedish finding. Two features of its design limit the weight it can carry. Exposure depended on men recalling their teenage skin decades afterwards, and the population studied differs from the Swedish and North American cohorts, so the estimate does not necessarily transfer between them.

Mendelian randomisation meta-analysis, 2024 [6]

This was a genetic causal-inference study. It found no evidence of a causal relationship. Mendelian randomisation uses inherited genetic variants as a proxy for lifetime exposure, and because those variants are fixed at conception it is less vulnerable to the confounding that affects observational designs. Two qualifications apply. It assesses genetically predicted acne rather than a clinical diagnosis of severe adolescent acne, and the method rests on assumptions about the variants used that cannot be fully verified.

The figure most often repeated outside the literature is the sixfold one. The Swedish cohort did report a roughly sixfold rate among men classified as having severe acne, but that figure rested on a very small number of prostate cancer cases within an already small group. When a result depends on a handful of cases, the range of values the data are consistent with becomes enormous, and this study could rule out neither an effect close to nothing nor one far larger than sixfold. It detected a possible signal without being able to measure it. Quoting the point estimate without its interval presents a fragile number as a settled one.

The gap between the cohort and case-control results has several possible explanations, none of which the present evidence can distinguish. Retrospective studies depend on men recalling teenage acne accurately, and misremembering would tend to blur a real association towards the null. Cohort studies remain vulnerable to differences in how acne was defined at the point of recording, to residual confounding, and to chance, particularly where only a small number of exposed cancer cases are available. The 2024 genetic analysis does not resolve the question either, because finding no causal signal is consistent both with there being no association and with an association that is real but not causal.

Melanoma and breast cancer, the conflicting cohort findings

The Nurses' Health Study II followed 99,128 women for 20 years and tested severe teenage acne against eight separate cancers. It reported nominal associations with both breast cancer and melanoma. Neither survived adjustment for the fact that eight cancers had been tested at once, which is the correction that prevents a chance finding among many comparisons being read as a result. The melanoma finding was nonetheless reproduced afterwards in a separate study of 930 people with melanoma and 1,026 without, and participants with teenage acne were also more likely to have moles [7].

The same Swedish group then examined melanoma in men, using the cohort that produced the prostate finding. Among 242,096 men and 1,058 melanomas there was no association at all [8].

Two large and well-conducted cohorts examined the same exposure and reached opposite conclusions. Real differences separate them, namely sex, self-reported severe acne against acne recorded at a conscription examination, and different underlying populations, and the discrepancy has never been resolved. Findings of this magnitude are unstable enough that a single cohort result should not be treated as established until it has been reproduced.

Breast cancer and severe acne, the Sister Study finding

A later prospective study, the Sister Study, followed 50,884 women and found a positive but statistically inconclusive association between ever having been diagnosed with severe acne and breast cancer, with a stronger signal among women whose severe acne was diagnosed before the age of 18 [9]. The finding did not reach the threshold at which a result is treated as reliable, and it remains compatible with there being no association at all. These findings remain observational, so they do not establish that acne causes breast cancer, and they do not make acne an indication for additional breast screening.

Cutibacterium acnes and prostate cancer, what the laboratory evidence shows

The second hypothesis does not involve hormones at all. Cutibacterium acnes, still widely called Propionibacterium acnes and the organism at the centre of acne, is a skin commensal that has also been recovered from prostate tissue. The laboratory evidence for its role there is considerably more developed than the epidemiological evidence, and it has accumulated in five steps, each of which establishes something the previous one did not.

• The organism has been located inside prostate cells. Immunohistochemistry on radical prostatectomy specimens found the bacterium within glandular epithelium and stromal macrophages, more often in cancerous than in control tissue, alongside higher nuclear NF-κB expression and in proportion to the grade of chronic inflammation present [10].
• The organism has been cultured from prostate tissue and typed. In one surgical study, P. acnes was cultured from 60% of prostates containing cancer and 26% of control prostates, a difference large enough to be unlikely to have arisen by chance. Three features limit how confidently that difference can be read. The controls were men undergoing surgery for bladder cancer rather than healthy men, a previous transrectal biopsy could itself have introduced the organism, and a common skin bacterium is difficult to study reliably in tissue that carries very little bacterial material [11].
• The organism causes prostatitis in animals. Live P. acnes introduced into the prostates of 98 rats produced chronic infection and sustained inflammation, which establishes the direction of causation within an animal model although not within human disease [12].
• The organism alters the local immune response. Macrophages stimulated with C. acnes increase expression of immunosuppressive genes, and prostate cancer tissue from which the organism can be cultured carries more regulatory T-cells, the cell type that dampens an immune attack rather than mounting one [13].
• A possible genomic mechanism was proposed in 2024. In cultured prostate epithelial cells, C. acnes infection suppressed BRCA2 and other DNA-repair pathways, producing a state resembling the BRCA-deficient phenotype in which a cell loses part of its ability to repair damaged DNA. The finding is biologically important, and it remains preclinical evidence that does not show the same process causing prostate cancer in patients [14].

The counter-evidence carries equal weight and is set out here for that reason. Higher circulating antibody levels against P. acnes have been associated with a reduced risk of prostate cancer, particularly advanced disease, which is the opposite direction to the one the infection hypothesis predicts. Detection rates also vary widely between studies according to the method used, and reviews describe the overall findings as inconsistent [15]. No study has shown that treating acne, or eradicating the organism, alters prostate cancer risk in any patient.

The result is an increasingly well-characterised mechanism attached to an association that may not exist. That combination is unusual and is stated plainly here, because a detailed mechanism is easily mistaken for evidence that the association it would explain has been established.

Does cancer cause acne?

Cancer does not cause acne in any ordinary sense. Acne is not a paraneoplastic sign, meaning a skin change produced at a distance by an internal tumour. Reviews of the skin manifestations of internal malignancy catalogue a specific list of such signs, namely acanthosis nigricans, the Leser-Trélat sign, dermatomyositis, Sweet syndrome and necrolytic migratory erythema, and acne does not appear on it [16]. The rare exception is a hormone-secreting tumour of the ovary or adrenal gland, which can produce abrupt acne as one feature of rapid-onset androgen excess, alongside new hair growth, deepening of the voice or menstrual disturbance. Acne appearing suddenly in an adult together with those other signs warrants investigation. Acne appearing on its own does not.

The reverse is common, in that cancer treatment frequently causes an acne-like rash. Papulopustular eruptions occur in roughly 90% of patients taking EGFR inhibitors such as cetuximab and panitumumab, usually with dryness, itch and paronychia, which is inflammation of the skin around the nail [17]. Rates and severity differ between agents, in that first-generation and second-generation tyrosine kinase inhibitors, namely gefitinib, erlotinib and afatinib, commonly produce it [18], whereas the third-generation agent osimertinib has a milder skin profile [19].

The eruption is not acne. It arises from blockade of the epidermal growth factor receptor within the hair follicle, which injures the keratinocytes lining it and provokes inflammation without infection, rather than from the androgen-driven excess of sebum that produces acne, and it is therefore managed differently. A pre-emptive skin regimen reduces clinically significant toxicity by a measurable margin. In the STEPP trial, patients receiving panitumumab who were given moisturiser, sunscreen, a topical corticosteroid and doxycycline had grade 2 or worse skin toxicity reduced from 62% to 29% over six weeks [20]. The regimen itself should be chosen by the oncology team according to the drug, the patient and the severity of the eruption. In several anti-EGFR studies the development or severity of the eruption has correlated with treatment response or survival. That correlation is not a reason to leave the eruption untreated, but a reason to treat the skin so that the patient can remain on the cancer drug wherever possible.

What this evidence means in practice

In my view no individual man should be worried by this literature. The association concerns acne recorded at the age of 18 in men followed for nearly four decades, it weakens or disappears once all the evidence is pooled, and a genetic analysis designed specifically to test causality found none. Taken at its strongest, it describes a small shift in the rate across a population rather than a prediction about any one person.

I do not use it in any clinical decision. Prostate screening decisions rest on age, family history and ethnicity, and they belong with a family doctor or a urologist. A history of teenage acne is not a screening indication, and raising it in consultation would generate anxiety without generating information.

The part of this research I consider most valuable is the C. acnes work, because it shows that an organism regarded as a harmless resident of facial skin behaves quite differently within other tissue. The BRCA2 finding is a preclinical result of the kind that changes how a bacterium is understood. It has not been connected to any clinical decision and may never be, and it is nonetheless a better reason to keep following this literature than the risk figures are.

None of it changes what I tell men with significant acne, which is that acne should be treated because it scars. The 2024 American Academy of Dermatology (AAD) guideline strongly recommends isotretinoin for severe acne, for acne that is causing scarring or psychosocial burden, and for acne that has not responded adequately to standard oral or topical treatment [21]. How it works is set out on my isotretinoin page. For men who would rather avoid oral medication, the device options are compared on AGNES RF vs AviClear vs Gold PTT, and where the acne follows the persistent adult pattern, my adult acne page sets out how I sequence treatment.

Frequently asked questions about acne

Does acne cause cancer?

There is no convincing evidence that it does. Several studies have reported an association between severe adolescent acne and later prostate cancer, although pooling all eight gives no overall association [4], and a 2024 genetic analysis designed to test causality found none [6]. An association that appears in some study designs and not in others is not evidence of causation.

Does cancer cause acne?

Cancer does not cause acne in any ordinary sense. Acne does not appear on the recognised list of skin signs of internal malignancy [16]. The rare exception is a hormone-secreting ovarian or adrenal tumour, which can cause acne to appear abruptly alongside other signs of androgen excess such as new hair growth, voice change or menstrual disturbance, and it is that combination rather than the acne alone that warrants investigation. Some cancer treatments separately cause an acne-like rash, and around 90% of patients taking EGFR inhibitors develop one [17], which the oncology team manages together with a dermatologist.

Is there such a thing as malignant acne?

"Malignant acne" is an old and potentially misleading name for acne fulminans, a rare, abrupt and exceptionally severe inflammatory form of acne. The word "malignant" does not mean that the lesions are cancerous, and acne lesions do not transform into cancer. Acne fulminans presents with a sudden outbreak of painful haemorrhagic pustules and ulceration, typically in male teenagers with pre-existing acne, and may be accompanied by systemic features such as fever and joint pain [22]. It requires prompt medical treatment, because it ulcerates and scars extensively.

Do men with acne have high testosterone?

Most men with acne do not, because circulating androgen levels are normal in the majority of people with acne. The difference is local rather than systemic, in that acne-prone skin shows increased 5-alpha-reductase activity within the gland, which converts testosterone to the more potent DHT, a higher density of androgen receptors for that DHT to act on, and a capacity to produce and metabolise androgens independently of blood levels, with insulin and IGF-1 amplifying the response [1, 2].

Does finasteride cause acne, or treat it?

Neither effect is established. Finasteride and dutasteride inhibit 5-alpha-reductase, which makes the question a reasonable one, but neither drug is an approved acne treatment and reliable evidence that they routinely cause acne is lacking. The anti-androgen treatments with a clearer evidence base in acne are combined oral contraceptives and spironolactone, both used in women [23]. Spironolactone acts partly by blocking the androgen receptor, is conditionally recommended in the 2024 AAD guideline [21], and outperformed placebo in a large 2023 randomised trial [24].

Can treating my acne reduce my cancer risk?

There is no evidence that it does, and I would not offer treatment on that basis. No study has tested whether treating acne, or clearing C. acnes, changes prostate cancer risk. The reasons to treat acne are that it scars and that it affects how people feel [21].

Should men with severe acne be screened for prostate cancer?

Severe acne is not itself a reason to screen. Screening decisions rest on age, family history and ethnicity, and they are a conversation for a family doctor or a urologist.

Is acne linked to melanoma or breast cancer?

The evidence is inconsistent. A cohort of 99,128 women in the United States reported associations with both melanoma and breast cancer, neither of which survived adjustment for multiple testing, although the melanoma finding was replicated separately [7]. A cohort of 242,096 Swedish men found no melanoma association at all [8]. A later study of 50,884 women found a possible breast cancer association that did not reach the threshold for a reliable result [9]. None of this establishes causation, and none of it changes screening.

Does the acne bacterium cause prostate cancer?

It has not been shown to. C. acnes is cultured more often from cancerous than from control prostates, at 60% against 26% [11], causes prostatitis in animal models [12], alters the local immune response [13], and suppresses BRCA2 in cultured prostate cells [14]. Higher antibody levels against it have nonetheless been associated with lower risk, and reviews describe the findings as inconsistent [15]. A plausible mechanism is not proof that the mechanism operates in patients.

My teenager has bad acne. Is this something to worry about long-term?

There is no cancer-related reason for concern. What is worth acting on is scarring, which is permanent and far easier to prevent than to correct. Acne that is scarring, or that is affecting mood and confidence, is an indication for effective treatment under the 2024 guideline regardless of how many lesions are counted [21].

Where should I start if I want my acne treated?

The starting point is a diagnosis of which type of acne is present and how much scarring risk it carries, rather than a product. My what is acne page covers the basics, adult acne covers the persistent pattern in older patients, and patients who would like an assessment can book an acne consultation at The Clifford Clinic.

References (24)

1. Briganti S, Flori E, Mastrofrancesco A, Ottaviani M. Acne as an altered dermato-endocrine response problem. Exp Dermatol. 2020;29(9):833-839. doi:10.1111/exd.14168.
2. Zaenglein AL. Acne vulgaris. N Engl J Med. 2018;379(14):1343-1352. doi:10.1056/NEJMcp1702493.
3. Ugge H, Udumyan R, Carlsson J, Andrén O, Montgomery S, Davidsson S, Fall K. Acne in late adolescence and risk of prostate cancer. Int J Cancer. 2018;142(8):1580-1585. doi:10.1002/ijc.31192.
4. Zhang X, Lin Y, Xie X, Shen M, Huang G, Yang Y. Is acne in adolescence associated with prostate cancer risk? Evidence from a meta-analysis. PLoS One. 2018;13(11):e0206249. doi:10.1371/journal.pone.0206249.
5. Hernández-Pérez JG, López DS, Rodríguez-Valentín R, Vázquez-Salas RA, Sierra-Santoyo A, Torres-Sánchez L. Late puberty onset and lack of acne during adolescence reduce high-grade prostate cancer at adulthood. Prostate. 2023;83(14):1342-1350. doi:10.1002/pros.24596.
6. He M, Wang X, Wang D, Wang Y. No genetic causal relationship between acne and prostate cancer through Mendelian randomization combined with meta-analysis. Front Oncol. 2024;14:1460467. doi:10.3389/fonc.2024.1460467.
7. Zhang M, Qureshi AA, Fortner RT, et al. Teenage acne and cancer risk in US women: a prospective cohort study. Cancer. 2015;121(10):1681-1687. doi:10.1002/cncr.29216.
8. Mota Garcia T, Hiyoshi A, Udumyan R, Sjöqvist H, Fall K, Montgomery S. Acne in late adolescence is not associated with a raised risk of subsequent malignant melanoma among men. Cancer Epidemiol. 2017;51:44-48. doi:10.1016/j.canep.2017.10.001.
9. Murphy JD, Sandler DP, White AJ, O'Brien KM. Severe acne and risk of breast cancer. Breast Cancer Res Treat. 2019;177(2):487-495. doi:10.1007/s10549-019-05302-z.
10. Bae Y, Ito T, Iida T, et al. Intracellular Propionibacterium acnes infection in glandular epithelium and stromal macrophages of the prostate with or without cancer. PLoS One. 2014;9(2):e90324. doi:10.1371/journal.pone.0090324.
11. Davidsson S, Mölling P, Rider JR, et al. Frequency and typing of Propionibacterium acnes in prostate tissue obtained from men with and without prostate cancer. Infect Agent Cancer. 2016;11:26. doi:10.1186/s13027-016-0074-9.
12. Olsson J, Drott JB, Laurantzon L, et al. Chronic prostatic infection and inflammation by Propionibacterium acnes in a rat prostate infection model. PLoS One. 2012;7(12):e51434. doi:10.1371/journal.pone.0051434.
13. Davidsson S, Carlsson J, Greenberg L, Wijkander J, Söderquist B, Erlandsson A. Cutibacterium acnes induces the expression of immunosuppressive genes in macrophages and is associated with an increase of regulatory T-cells in prostate cancer. Microbiol Spectr. 2021;9(3):e0149721. doi:10.1128/spectrum.01497-21.
14. Ashida S, Kawada C, Tanaka H, et al. Cutibacterium acnes invades prostate epithelial cells to induce BRCAness as a possible pathogen of prostate cancer. Prostate. 2024;84(11):1056-1066. doi:10.1002/pros.24723.
15. Fu F, Yu Y, Wang B, et al. Prostate and urinary microbiomes in prostate cancer development: focus on Cutibacterium acnes. Front Cell Infect Microbiol. 2025;15:1562729. doi:10.3389/fcimb.2025.1562729.
16. Wick MR, Patterson JW. Cutaneous paraneoplastic syndromes. Semin Diagn Pathol. 2019;36(4):211-228. doi:10.1053/j.semdp.2019.01.001.
17. Yu Z, Dee EC, Bach DQ, Mostaghimi A, LeBoeuf NR. Evaluation of a comprehensive skin toxicity program for patients treated with epidermal growth factor receptor inhibitors at a cancer treatment center. JAMA Dermatol. 2020;156(10):1079-1085. doi:10.1001/jamadermatol.2020.1795.
18. Chen KL, Lin CC, Cho YT, et al. Comparison of skin toxic effects associated with gefitinib, erlotinib, or afatinib treatment for non-small cell lung cancer. JAMA Dermatol. 2016;152(3):340-342. doi:10.1001/jamadermatol.2015.4448.
19. Chu CY, Choi J, Eaby-Sandy B, Langer CJ, Lacouture ME. Osimertinib: a novel dermatologic adverse event profile in patients with lung cancer. Oncologist. 2018;23(8):891-899. doi:10.1634/theoncologist.2017-0582.
20. Lacouture ME, Mitchell EP, Piperdi B, et al. Skin toxicity evaluation protocol with panitumumab (STEPP), a phase II, open-label, randomized trial evaluating the impact of a pre-emptive skin treatment regimen on skin toxicities and quality of life in patients with metastatic colorectal cancer. J Clin Oncol. 2010;28(8):1351-1357. doi:10.1200/JCO.2008.21.7828.
21. Reynolds RV, Yeung H, Cheng CE, et al. Guidelines of care for the management of acne vulgaris. J Am Acad Dermatol. 2024;90(5):1006.e1-1006.e30. doi:10.1016/j.jaad.2023.12.017.
22. Dall'Oglio F, Puglisi DF, Nasca MR, Micali G. Acne fulminans. G Ital Dermatol Venereol. 2020;155(6):711-718. doi:10.23736/S0392-0488.20.06711-5.
23. Costa CS, Bagatin E, Yang Z, et al. Systemic pharmacological treatments for acne: an overview of systematic reviews. Cochrane Database Syst Rev. 2021;(11):CD014917. doi:10.1002/14651858.CD014917.
24. Santer M, Lawrence M, Renz S, et al. Effectiveness of spironolactone for women with acne vulgaris (SAFA) in England and Wales: pragmatic, multicentre, phase 3, double blind, randomised controlled trial. BMJ. 2023;381:e074349. doi:10.1136/bmj-2022-074349.

Every reference above was verified against Crossref for authors, title, journal, volume, issue, pages and year. The evidence search was completed on 31 August 2026.