A study of tobacco abuse as a risk factor for adrenal cortical carcinoma.

Abstract

e16119 Background: Smoking has been suggested as a risk factor for adrenal cortical carcinoma (ACC), but, due to the rarity of this tumor, this hypothesis has only been inferred from single study using a mixed population of adrenal tumors that included pheochromocytoma and neuroblastoma as well as ACC. Methods: De-identified IRB exempted records were analyzed for age, gender, and smoking history in adult patients diagnosed with ACC in both the West Virginia University database containing 2,699,193 patients and, in the larger TriNetX database of 21,426,477 patients from 2008-2018. In addition, the statewide ratio of smoking to ACC prevalence was computed in all 50 states using data from SEER and the CDC. Results: Given the high rate of both tobacco use and ACC in the West Virginia Health System, we hypothesized that smoking contributed to increased prevalence of ACC in our state. We compared the rates of ACC and smoking in all 50 states using data from SEER. Among the 50 states, West Virginia had the highest smoking rate and the second highest rate of ACC. Utah had the lowest rate of smoking and the third lowest rate of ACC. The sample correlation between statewide smoking and ACC is .4177 indicating a significant association between the two (p = .0035). From 2008 to 2018, 20 patients with ICD10 code c74.0 (ACC) were abstracted from our EMR system. Examination of the individual charts determined 15 of the 20 patients had confirmed ACC diagnosis. This cohort was compared with matched age and gender matched controls. In the control group, there 13.3% were smokers while in the ACC group, 60% were smokers. The difference is significant (p = .0005) (OR) = 9.75 (95% CI; (2.54, 37.38). We next reviewed the larger TriNetX database and found that the prevalence of smoking in the ACC group to be 35.2% as compared to 9.0% in the TriNetX group(p-value < .0001). Conclusions: Our retrospective review strongly supports smoking as a major risk factor ACC. To our knowledge, this is the first study to demonstrate a strong association of smoking in a cohort of biopsy proven ACCs.