Previous Agent Orange Exposure Is Correlated with Worse Outcome in Patients with Multiple Myeloma (MM)

BACKGROUND: Agent Orange (AO) is an herbicide that was previously used by the United States armed forces during the Vietnam conflict during the years 1961-1971.  AO was contaminated with the potent carcinogen, 2, 3, 7, 8-tetrachlorodibenzodioxin (dioxin) , during the manufacturing process.  Exposure to AO has previously been defined as a risk factor for the development of several malignancies including multiple myeloma (MM).   While numerous studies have been evaluated the association between AO exposure and MM incidence, none have evaluated the association between AO exposure and survival.

METHODS:Patients with MM were identified in the Veterans Health Administration (VHA) central cancer registry using International Classification of Diseases for Oncology, version 3, code 9732/3.  Patients who did not receive treatment within 6 months of diagnosis were excluded in order to remove patients with monoclonal gammopathy or smoldering myeloma miscoded as MM (n=1,002).  Data was linked to administrative data on AO exposure, through patient self-report and clinician report, for a total of 308 AO exposed patients.  Data was also obtained on age, race, serum creatinine, co-morbid conditions calculated into a modified Charlson co-morbidity score, bisphosphonate use, year of diagnosis (before or after 2005), treatment with autologous stem cell transplantation, and body mass index (n=2968).  Overall survival (OS) was calculated from date of MM diagnosis to death.   Cox proportional hazards modeling was used to estimate the association between reported AO exposure and OS while controlling for potential confounding variables.

RESULTS: Variables significantly associated with reduced hazard of death included obese BMI (HR = 0.79, 95% CI 0.71-0.88), overweight BMI (0.85, CI 0.77-0.93), bisphosphonate use (HR 0.850, CI 0 .779-0.927), use of autologous transplantation (HR 0, 52, CI 0.44-0.67), and diagnosis after 2005 (HR 0.84, CI 0.78-0.92).  Race was associated with a non-significant trend towards reduced hazard of death (HR – 0.92 CI 0.84 – 1.01) Variables significantly associated with increased hazard of death included underweight BMI (HR 1.61 CI 1.28-2.04), advanced age (HR 1.02, CI 1.02-1.-3); higher co-morbidity score (HR 1.07 CI 1.06-1.09), creatinine greater than 2 mg/dl (HR 1.40 CI 1.28-1.56), and reported AO exposure (HR 1.29 CI 1.07-1.40). 

CONCLUSIONS: This is the largest reported study of the association between AO exposure and OS in patients diagnosed with MM, and it suggests a statistically significant worsening of OS in those exposed.  Whether this is due to the development of higher risk cytogenetic profiles in those exposed to AO could not be determined for this study, but is worthy of further research.  A major limitation is the inability to objectively confirm the history of AO exposure.  However, it is unlikely that a systematic bias would drive patients with poorer prognosis MM to aberrantly claim AO exposure.  It is important to highlight that regardless of the detriment in estimated OS after AO exposure, MM remains an incurable disease that results in a disease-related death in most patients.