Agent Orange Exposure Does Not Predict for Shorter Overall Survival in Patients with Chronic Lymphocytic Leukemia: A National Veteran Affairs Tumor Registry Study

BACKGROUND: Agent Orange (AO), a 1:1 mixture of herbicides + TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin), was used during the Vietnam War to destroy dense jungle and enemy crops. In 2002 the Department of Veteran Affairs (VA) determined that chronic lymphocytic leukemia (CLL) was associated with AO exposure. Case-control studies suggest an increased risk of death from CLL in areas where herbicide use was highest. There is also an increased incidence of other cancers (prostate, melanoma) in AO-exposed veterans. Limited data exists as to the specific impact of AO exposure on CLL disease presentation and outcome.

METHODS: Patients (pts) diagnosed with CLL from 2009-2013 were identified in the National VAMC Tumor Registry. Baseline demographic and laboratory parameters were obtained, including Rai stage, marrow cytogenetics (when available), and lymphocyte doubling time (LDT). AO exposure was identified according to the medical record. The VA Benefits and Compensation officers determine AO exposure based on whether a person served on land and in the brown waters in Vietnam during the appropriate timeframe. Timing and types of CLL therapies were identified to determine if AO exposure influenced CLL treatment.

RESULTS: 2052 CLL pts were identified, of which 418 had AO exposure. AO-exposed pts presented at a younger age (63.2 versus [vs] 70.5 years (yrs), p <0.0001), had a higher hemoglobin (14.3 vs 13.8 g/dl, p<0.001) and lower lactate dehydrogenase (LDH) (203 vs 227 IU/L, p = 0.01) compared to those without AO exposure. There were no differences in white cell, platelet, or absolute lymphocyte counts, Rai stage or LDT among the groups. Cytogenetic data was available for 1167 pts. There was no difference in the incidence of 17p-, 11q-, or 13q- between the two groups. Median overall survival (OS) was significantly better in patients with AO exposure, even when adjusted for age and Rai stage (median not reached vs 91.2 months, p <0.0001. OS benefit was primarily seen in pts age 60-69 yrs (p = 0.002), and those with 11q- (p = 0.001). No OS differences were found in pts with 17p- or 13q-. Among all pts, regardless of AO exposure status, OS decreased with higher Rai stage. There was a trend towards AO-exposed pts to be more likely to receive CLL-directed therapies (37% vs 32%, p = 0.07). AO exposed pts were more likely, than unexposed pts, to receive therapies as follows: fludarabine, chlorambucil, rituximab (FCR) first-line (38% vs 21%) and second-line (11.6 vs 5%); bendamustine + rituximab (BR) first-line (25% vs 18%), second-line (35 vs 26%), and third-line (31 vs 23%). Pts with no AO exposure were more likely to receive single agent chlorambucil or cyclophosphamide as first-line therapy (17 vs 10%).

CONCLUSION: Pts with AO exposure, compared to unexposed pts, had an OS benefit independent of age and Rai stage, with this benefit seen primarily in younger pts (age 60-69 yrs) and in those with 11q-. AO-exposed pts were also more likely to receive disease-specific therapy. This unexpected OS finding will require further analyses for confounding variables, but could potentially be related to earlier treatment with regimens as FCR or BR.