Introduction:

the association between eruptive seborrhoeic keratosis and some malignant neoplasms has been proposed. However, the evidence about this is poor: three European case-control studies whose results do not support it. The objective of the study is to evaluate the existence or not of this association in the Colombian population.

Material and methods:

study of 130 cases and 260 matched controls for age and sex in patients over 18 years of age among patients with gastrointestinal or pancreatobiliary cancer (cases) or hospitalized for other non-oncological reasons (controls) at the University Hospital of Santander, Bucaramanga (Colombia). All patients underwent meticulous skin examination including dermatoscopy, to identify the presence and time of evolution of seborrheic keratosis in areas covered by clothing.

Results:

the prevalence of seborrheic keratosis among the cases was 26.9% (95% CI 19.5-35.4) and 15.4% (95% CI 11.2-20.4) among the controls (OR 1.94, 95% CI 1.17-3.19). Seven patients with cancer presented seborrheic keratosis in the six months prior to the diagnosis of neoplasia (5.3%, 95% CI 2.2-10.8) compared to none among control patients (0.0%, 95% CI 0.0-1.4, p <0.001); three of them had 20 or more lesions less than six months of onset (2.3% of cases, 95% CI 0.5-6.6) compared to none among controls (p = 0.014).

Discussion:

this study tries to correct the weaknesses of the only three previous studies that explore the association between seborrheic keratosis and cancer by rigorous selection of patients, the inclusion of cancer patients from a single body system and a strict process to evaluate the skin lesions with dermatoscopy. This allows concluding that in Colombian patients with gastrointestinal cancer there is a greater frequency of seborrheic keratosis and eruptive seborrheic keratosis than in people of the same age and sex without cancer. (Acta Med Colomb 2017; 42: 224-229).