Abstract
Elevated prediagnostic serum levels of the immune activation markers sCD27 and sCD30 have been associated with non-Hodgkin lymphoma (NHL). However, the use of a single sample per participant in these studies has limited etiologic inferences. We report findings, overall and by NHL subtype, from a case-control analysis (422 cases, 434 controls) within the Janus Serum Bank with two samples per subject collected on average 5 years apart. Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) was associated with elevated sCD27 in the later, but not earlier, prediagnostic sample (odds ratio [OR] 4.2, 95% confidence interval [CI] 1.5-11.6 and 1.7, 0.7-4.7 per log increase, respectively) in analyses adjusting for both analytes, while follicular lymphoma (FL) was associated with elevated sCD30 in both the later and earlier samples (OR 2.9, 95% CI 1.4-4.4 and 2.3, 1.2-4.4, respectively). CLL/SLL cases were significantly more likely than controls to have higher sCD27 in the later vs. earlier sample (OR 1.4, 95% CI 1.1-1.9 per standard deviation increase); no such difference in sCD30 was apparent for FL. In a joint analysis, NHL cases were more likely than controls to have below-median sCD27 in the earlier sample and above-median sCD27 in the later sample (OR 1.5, 95% CI 1.0-2.3). For sCD30, the association between sCD30 and FL was confined to subjects with above-median analyte levels in both samples (OR 2.5, 95% CI 1.1-5.9). Our findings are compatible with elevated sCD27 representing a disease-induced effect and sCD30 representing a marker of increased FL susceptibility.