Elevated Serum Levels of sCD30 and IL6 and Detectable IL10 Precede Classical Hodgkin Lymphoma Diagnosis

Abstract

Background: We investigated whether an immune system environment characterized by elevated serum levels of B-cell activation molecules was associated with the subsequent development of classical Hodgkin lymphoma (cHL).Methods: We measured serum levels of B-cell-stimulatory cytokines, IL6 and IL10, soluble CD30 (sCD30), and total IgE prior to cHL diagnosis in 103 cases and 206 matched controls with archived specimens in the DoD Serum Repository.Results: Prediagnosis serum sCD30 and IL6 levels had strong positive associations with risk of a cHL diagnosis 0 to 1 year prior to diagnosis [sCD30 OR = 5.5; 95% confidence interval (CI), 3.4-9.0; IL6 OR = 4.6; 95% CI, 2.9-7.5] and >1 year to 2 years pre-cHL diagnosis (sCD30 OR = 3.3; 95% CI, 1.6-6.7; IL6 OR = 2.9; 95% CI, 1.3-6.5). We observed similar, albeit not consistently significant positive associations, over 4 or more years preceding diagnosis. We did not observe a clear association with IgE levels. Of note, detectable IL10 levels were significantly associated with Epstein-Barr virus (EBV)-positive cHL cases compared with EBV-negative cases.Conclusion: In this prospective analysis, elevated sCD30 and IL6 levels and detectable IL10 preceded cHL diagnosis.Impact: The associations of these cytokines with cHL risk may reflect the production of these molecules by proliferating nascent cHL tumor cells, or by immune cells responding to their presence, prior to clinical detection. The stable elevation in cHL risk, 4 or more years prediagnosis, also suggests that a B-cell-stimulatory immune system milieu precedes, and may promote, lymphomagenesis. Cancer Epidemiol Biomarkers Prev; 26(7); 1114-23. ©2017 AACR.

Figure 1. Distribution of serum samples from 103 classical Hodgkin lymphoma cases and 206 controls across five time windows preceding cHL diagnosis

Numbers shown are the number of cHL cases with at least one sample within that time window; for every case at each time point, controls were matched 2:1 based on serum collection dates.

Figure 2. Pre-cHL serum levels of sCD30, IL-6, and IgE in five time windows by case-control status

Serum samples from cases (triangles) and matched controls (circles) were analyzed in five designated time windows as shown in Figure 1. Serum levels of (A) sCD30, (B) IL-6, and (C) IgE are plotted on the Y axis with log base 2 scale; the lower limit of detection for each analyte is indicated by a horizontal dotted line parallel to the X axis. Box plots display the median and interquartile range (IQR). P values comparing median values of cases and controls in each time window are computed using Wilcoxon signed-rank test.

Figure 3. Odds ratios (OR) and 95% confidence intervals (CI) for the association of cHL with elevated serum levels of sCD30, IL-6, and IgE prior to diagnosis

Elevated levels of (A) sCD30, (B) IL-6, and (C) IgE were defined as falling above the median levels among controls. OR (diamonds) and 95% CI (whiskers) were calculated by GEE analysis, modeling time before diagnosis in one-year increments and adjusted for age, race-ethnicity, sex, and serum draw date.

Figure 4. Within-person change per year in serum levels of sCD30, IL-6, and IgE by case-control status

Change per year in serum levels of (A) sCD30, (B) IL-6, or (C) IgE was calculated using natural log-transformed biomarker levels for the time period ranging from the earliest available sample to the sample closest to but preceding cHL diagnosis. Cases (n=64) are shown as open circles, matched controls (n=128) as filled circles; the dotted horizontal line at zero indicates no change in serum level over time.