Date of Award

January 2014

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Deepak Narayan

Subject Area(s)

Medicine, Surgery


ANTIDEPRESSANTS AND MELANOMA: IS THERE A LINK? Stephanie R. Douglas, Deepak Narayan. Section of Plastic and Reconstructive Surgery, Department of Surgery, Yale University School of Medicine, New Haven, CT.

Cutaneous melanoma is the sixth most common cancer in the United States, and its incidence continues to rise steadily at a rate of 2-3% per year. Meanwhile nearly 11% of the general population is treated with antidepressant medication. Laboratory studies suggest a number of theoretical links between antidepressants and melanoma. Serotonin is a growth factor for melanocytes, and antidepressants upregulate components of the Wnt/beta-catenin, mTOR, and MAP kinase signaling pathways, all of which may be involved in melanoma transformation and carcinogenesis. To investigate the potential link between antidepressants and melanoma in vivo, localized tumors were induced in a conditional mouse model of BrafV600E-induced, Pten-deficient melanoma. Citalopram hydrobromide in 0.1% saccharin was administered at a dose of 20 mg/kg/day orally via drinking water beginning at the time of melanoma induction. Control animals received 0.1% saccharin. At weekly intervals, the overall tumor volume was assessed so that the tumor growth rate could be calculated for each mouse. At day 75 following tumor induction, necropsy was performed to assess for the presence of metastases. Antidepressant administration appeared to have no effect on tumor volume at any of the time points measured (day 29, p=0.997; day 36, p=0.761; day 44, p=0.612; day 50, p=0.682; day 57, p=0.797; day 66, p=0.691; day 75, p=0.736), or on metastasis. Secondly, the charts of 1271 patients treated for melanoma at Yale Cancer Center between 1997 and 2013 were reviewed, taking particular note of medication history. A health history questionnaire eliciting information about melanoma risk factors as well as medication histories was administered to patients seen at an outpatient surgery clinic at Yale-New Haven Hospital who had no history of melanoma. Age, sex, and group (cases or controls) were entered into a binary logistic regression model to calculate an adjusted odds ratio for antidepressant exposure as a function of group. Additionally, age, sex, and tumor stage were entered into a second model for the odds of antidepressant exposure as a function of tumor stage. Melanoma patients were less likely to have a history of antidepressant use relative to controls (OR 0.567, 95% CI 0.331-0.972, p=0.039). There was no association between antidepressants and melanoma stage at diagnosis. Further study is needed in order to clarify the nature of the association between antidepressants and melanoma risk.