Estrogen therapy linked to doubled odds of porphyria cutanea tarda

By Sophie Lindqvist

In a nationwide Danish matched case-control study of 438 female patients with a first-time hospital diagnosis of porphyria cutanea tarda (PCT) and 43,800 age-matched female controls, estrogen-containing therapy carried an adjusted odds ratio of 2.03 for incident PCT (95% CI, 1.66–2.49).

The investigators used nationwide registry data collected from January 1995 through November 2024. They defined exposure as at least one redeemed prescription for an estrogen-containing therapy during the 12 months preceding the index date, then adjusted for PCT-related comorbidities.

Until this analysis, evidence about estrogen and PCT came mainly from clinical case series and case reports. The registry approach addressed that gap by testing the association across a large national population.

Exposure was more frequent in the PCT group: 154 patients (35.16%) had redeemed at least one estrogen-containing prescription, compared with 9,433 controls (21.54%). After adjustment, the overall association remained approximately twofold.

Treatment categories

Associations appeared in every estrogen category the researchers assessed, although the point estimates differed.

• Estrogen-containing contraceptives showed the strongest association: adjusted OR 2.52; 95% CI, 1.67–3.82.
• Estrogen-only therapy was also associated with incident PCT: adjusted OR 1.58; 95% CI, 1.23–2.03.
• Combined estrogen and progestogen therapy was associated with incident PCT: adjusted OR 2.18; 95% CI, 1.53–3.11.

Estrogen-containing contraceptives carried the largest point estimate. All three lower confidence limits exceeded 1, supporting an association for each category in this dataset. The results do not prove that a particular product caused PCT.

Erichsen and colleagues concluded: “Estrogen-containing therapy was associated with incident PCT in this nationwide population-based study.”

Route of administration

Oral estrogen-containing therapy was associated with increased odds of PCT (adjusted OR, 2.44; 95% CI, 1.66–3.59). Non-oral administration routes were not associated with incident PCT in the analysis.

The non-oral result came without a numerical estimate, so it should be read as no detected association rather than demonstrated protection. The route finding nevertheless narrows the prescribing question: clinicians evaluating suspected PCT need to establish whether the patient used a contraceptive, estrogen-only therapy, or combined estrogen-progestogen treatment, and whether administration was oral or non-oral.

PCT is associated with chronic blistering and skin fragility. For clinicians evaluating that pattern, route and formulation become concrete parts of the exposure history.

The registry analysis estimates associations; it does not establish that oral estrogen caused an individual case. Nor does the route split show that changing administration prevents PCT. The report supports risk assessment, not a causal prescribing rule.

What does this change for estrogen prescribing?

For clinicians, the practical change is to document both formulation and route when taking an estrogen history in a patient with PCT or suspected PCT. The data support that focused review across all three treatment categories, not a blanket label attached to every estrogen prescription.

• Document the preparation: estrogen-containing contraceptive, estrogen-only therapy, or combined estrogen-progestogen treatment.
• Record the route: oral or non-oral. Only oral administration showed an association in this analysis.
• Review prescriptions redeemed during the 12 months before the index date, matching the study’s exposure window.

That is where the study changes practice. Route belongs in the risk discussion. Formulation still matters.

Erichsen PA et al. reported the analysis in JAMA Dermatology in 2026 (DOI: 10.1001/jamadermatol.2026.4329).

Confirmation now needs route- and formulation-specific data to test whether the oral signal persists across populations and whether the contraceptive estimate survives direct comparisons. Until then, the study supports targeted risk assessment rather than a causal verdict.