Nosrati N, Han J, Flores R, Sood R, Tholpady SS. The Effect of Agent Orange on Nonmelanoma Skin Cancer Regression Rates. JAMA Surg. 2014;149(11):1205-1206. doi:10.1001/jamasurg.2014.953
Decades ago, the US military used herbicides, including Agent Orange (2,3,7,8-tetrachlorodibenzodioxin), in conflicts such as the Vietnam War. Agent Orange has since been linked to multiple ailments, including prostate cancer, porphyria, chloracne, Hodgkin disease, and many more.1 Recently, a pilot study by Clemens et al2 suggested that exposure to Agent Orange may increase the incidence of nonmelanotic invasive skin cancer (NMSC) among veterans when compared with the general population. They also found a greater risk when the exposure was through more direct means such as spraying the herbicide.
Frequently, following a biopsy, NMSCs may regress to the point that no evidence of carcinoma may be found after excision. The observation was first described by Goldwyn and Kasdon in 1978.3 Since their initial report, several studies4- 6 have found regression rates varying from 24% to 72%. With the pilot study2 suggesting that Agent Orange increased the incidence of NSMC, we aimed to determine if these NSMC lesions still behaved similarly. We then proceeded to design a study comparing regression rates between veterans with NMSC who were exposed to Agent Orange and veterans with NMSC who were not.
Veterans Affairs (VA) institutional review board approval was obtained to conduct a retrospective study of the period from 2003 to 2013. Participants did not provide consent because our study was deemed to be exempt by the institutional review board owing to its retrospective nature and lack of any patient communication or contact. There was also no financial compensation.
A database was created of patients from the Richard L. Roudebush VA Medical Center using International Classification of Diseases, Ninth Revision codes for NMSC and Current Procedural Terminology codes for both biopsy and excision. Inclusion criteria consisted of patients who had a biopsy of the lesion performed without any other concurrent dermatologic treatment (eg, electrodessication and curettage, cryotherapy, or topical chemotherapy), positive margins after initial biopsy, and, finally, excision of the lesion. Data on patient demographics, Agent Orange exposure, and lesion characteristics were collected. Agent Orange exposure was identified by the VA Agent Orange registry. Regression was defined as no evidence of residual carcinoma in the excision following positive margins on the initial biopsy. Statistical analysis was then performed with SPSS software (IBM) using the Fisher exact t test.
There were 1499 lesions identified in 1024 patients. Demographic data are presented in Table 1. Of the 1499 lesions, 100 were the result of the patients being exposed to Agent Orange in Vietnam. There were 66 basal cell carcinomas (BCCs) and 34 squamous cell carcinomas (SCCs) in patients exposed to Agent Orange. For the entire population, a regression rate of 41.0% was observed. A differential regression rate was present with 30.7% regression for patients with BCC and 55.7% for patients with SCC. Differences between SCC and BCC regression were statistically significant (P < .05). For the patients exposed to Agent Orange, an overall regression rate of 43.0% was observed, with a rate of 36.4% for patients with BCC and 53.5% for patients with SCC. Once again, there was a significant difference between BCC and SCC, but no difference between exposure to Agent Orange and nonexposure (P < .05) (Table 2).
In the pilot study by Clemens et al,2 veterans exposed to Agent Orange were found to have an increased incidence of NMSCs compared with the general population. If borne out in a larger data set, the difference is critical because it mandates closer surveillance of these patients. A logical extension of this study2 is that NMSCs behave differently or are harder to treat. The present study analyzed if these NMSCs behave differently with respect to regression when compared with patients without Agent Orange exposure. We found similar rates of regression between both cohorts in all 3 categories: BCC, SCC, and overall NSMC. With the suggestion that exposure results in patients developing more invasive forms of NMSC, the expectation is that Agent Orange would result in lesions that are less likely to regress. However, these findings demonstrate that, in a large cohort, the NMSC lesions developing in patients exposed to Agent Orange do not regress any differently than the lesions developing in patients who were not exposed.
Corresponding Author: Sunil S. Tholpady, MD, PhD, Division of Plastic Surgery, Department of Surgery, Indiana University, and Richard L. Roudebush VA Medical Center, 705 Riley Hospital Dr, RI 2514, Indianapolis, IN 46202 (firstname.lastname@example.org).
Published Online: September 3, 2014. doi:10.1001/jamasurg.2014.953.
Author Contributions: Drs Nosrati and Tholpady had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Dr Nosrati and Ms Han contributed equally and would like to be considered as co–first authors.
Study concept and design: All authors.
Acquisition, analysis, or interpretation of data: Nosrati, Han, Tholpady.
Drafting of the manuscript: Nosrati, Han, Tholpady.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Nosrati, Tholpady.
Administrative, technical, or material support: Nosrati, Flores.
Study supervision: Nosrati, Flores, Sood, Tholpady.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This paper was presented at the Annual Meeting of the Association of VA Surgeons; April 7, 2014; New Haven, Connecticut.