Risk of cancer in psoriasis: study of a nationally representative sample of the US population with comparison to a single-institution cohort

TY - JOUR

T1 - Risk of cancer in psoriasis

T2 - study of a nationally representative sample of the US population with comparison to a single-institution cohort

AU - Herbosa, C. M.

AU - Hodges, W.

AU - Mann, C.

AU - Demehri, S.

AU - Cornelius, L. A.

AU - Semenov, Y. R.

N1 - Funding Information: Psoriasis is a chronic inflammatory disease, affecting up to 4% of the population.1 Despite the mounting evidence of increased rate of malignancy in psoriatics, studies report conflicting results on the specific types of cancer seen in this population.1–4 To provide insight into the risk of malignancy among psoriatics, we performed a retrospective population-based cohort study using a nationally representative sample of the U.S. population. We further compared our results from the national database to a large cohort of patients seen at a tertiary level academic centre. The nationally representative sample was comprised of adults who participated in 2003–2006 and 2009–2014 National Health and Nutrition Examination Surveys (NHANES). History of psoriasis and malignancy was ascertained from medical history questionnaires. Risk ratios (RR) of malignancy among psoriasis participants were calculated for the NHANES cohort. The independent risk of malignancy was assessed using multivariable logistic regressions adjusted for age, sex, body mass index, smoking and alcohol use. Results were compared to a cross-sectional population of adults at Washington University School of Medicine (WUSM). The SlicerDicer feature of electronic medical record system, EPIC, allowed for the anonymous identification the WUSM cohort using ICD-10 codes for psoriasis and malignancies. A total of 24 044 participants were included from NHANES, of which 2.8% had a history of psoriasis. Participants with psoriasis had an overall 1.8-fold increased risk of malignancy compared to those without psoriasis (13.7 vs. 7.7%; P < 0.001). NHANES participants with psoriasis had significantly higher prevalence of specifically non-melanoma skin cancer (NMSC; 3.0 vs. 1.3%; RR = 2.29; P < 0.001), thyroid cancer (0.6 vs. 0.2%; RR = 3.00; P = 0.027) and gynecological malignancies (8.6 vs. 4.5%; RR = 1.91; P < 0.001) than the NHANES participants without psoriasis (Fig. 1). Adjusting for potential confounders, psoriasis was independently associated with 1.5-fold increased risk of malignancy, specifically NMSC [adjusted RR (aRR) = 2.06; P = 0.004), thyroid cancer (aRR = 3.59; P = 0.016) and gynecological cancers (aRR = 1.73; P = 0.006). Of the 1 620 412 participants included from the WUSM cohort, 0.5% had a history of psoriasis. Unadjusted analysis of the WUSM cohort revealed a similarly increased risk of malignancy with a 2.4-fold increased risk of malignancy among psoriatics (12.2 vs. 5.0%, P < 0.001; Fig. 2). Psoriasis patients in the WUSM cohort had comparably increased risks of NMSC (2.8 vs. 0.6%; RR = 4.53; P < 0.001), thyroid cancer (0.4 vs. 0.2%; RR = 2.49; P < 0.001) and gynecological malignancies (4.6 vs. 2.4%; RR = 2.06; P < 0.001) to those observed in the NHANES cohort. Adjusted analyses could not be performed for the WUSM cohort due to anonymized identification of cases in SlicerDicer. This study provides evidence for increased cancer risk in psoriasis using a nationally representative US population, specifically NMSC, thyroid cancer and gynecological malignancies. Our study was limited by our ability to investigate the association of psoriasis with certain solid organ malignancies (e.g. lung cancer) due to lack of cases reported in the NHANES psoriasis cohort, most likely a result of sample size limitations. In addition, we were unable to adjust for medical and demographic comorbidities in the WUSM cohort due to the anonymized identification of cases. Despite these limitations, our study demonstrates a consistently increased risk of malignancy among psoriatics from two separate populations. The observed increased risk is likely multifactorial, potentially related to the inflammatory nature of disease, systemic treatments or increased surveillance.1–4 These results suggest that more intensive cancer screening may be warranted for psoriasis patients. Future large population-based cohort studies are needed to further assess the independent risk of specific malignancies in this population. Funding sources: Washington University School of Medicine, Division of Dermatology.

PY - 2020/9/1

Y1 - 2020/9/1

UR - http://www.scopus.com/inward/record.url?scp=85084575167&partnerID=8YFLogxK

U2 - 10.1111/jdv.16447

DO - 10.1111/jdv.16447

M3 - Letter

C2 - 32277504

AN - SCOPUS:85084575167

SN - 0926-9959

VL - 34

SP - e529-e531

JO - Journal of the European Academy of Dermatology and Venereology

JF - Journal of the European Academy of Dermatology and Venereology

IS - 9

ER -