Introduction: Progress in the treatment of patients with classical Hodgkin lymphoma (cHL) depends on identifying methods to better risk-stratify patients and assess prognosis and treatment response. While prognostic scores based on clinical characteristics have utility, inflammatory markers and signaling proteins may better reflect tumor biology, the microenvironment, or host response and might serve as prognostic factors. Increased expression of tumor-associated macrophage markers in tumor tissue, such as CD68 or CD163, has been shown to be associated with inferior survival outcomes in cHL patients. Blood-based markers are more practical in that specimens are readily and serially obtainable. Such markers, if shown to track with tumor response and/or prognosticate outcomes, can be used throughout treatment and follow-up. Some soluble markers, such as TARC (chemokine (C-C motif) ligand (CCL)-17) and soluble CD163 (sCD163), have been shown to reflect tumor burden and active disease, and markers such as soluble CD30 (sCD30), interleukin (IL)-6, chemokine (C-X-C motif) ligand (CXCL)-10/IP10, and IL-2 receptor have been associated with failure-free survival (FFS).Methods: Newly diagnosed cHL patients with locally extensive or advanced stage disease were prospectively enrolled in the Intergroup E2496 randomized controlled trial, which compared ABVD with Stanford V chemotherapy. A panel of serum cytokines, chemokines, and other soluble markers including TARC, CCL22, CCL24, sCD30, sCD163, CXCL10/IP10, CXCL13, soluble CD14 (sCD14), IL-6, IL-10, IL-12, and IL-13, IL1-receptor antagonist (IL1RA)) were measured in pretreatment serum specimens from 301 cHL patients (out of 854 on study) using multiplex (Luminex) bead array immunoassay (R&D Systems). Serum marker values were log-transformed for all analyses. Epstein-Barr virus (EBV) tumor status was determined by EBER in situ hybridization. A linear regression model was used to assess the association between pre-treatment serum marker levels and baseline clinical characteristics. A stratified Cox proportional hazards regression model was used to evaluate the association between serum marker levels as continuous variables and survival outcomes, including FFS (time from registration to disease progression/relapse or death) and overall survival (OS). Patients were divided into quartile groups based on serum marker levels and Kaplan-Meier curves were constructed with comparison using the stratified log-rank test. Three stratification factors were used in all modeling/testing include: stage I-II bulky vs. stage III-IV; IPS 0-2 vs 3-7; and treatment arms (Stanford V vs. ABVD). Two-sided p-values were reported.Results: Increased pre-treatment levels of CCL24, sCD30, sCD163, IP10, sCD14, IL-6, and IL-10 were associated with the presence of B symptoms, independent of age, tumor histology, and disease stage, in multivariate analysis. Higher sCD163 and IP10 levels were associated with EBV positive tumors and higher TARC, CCL22, and CXCL13 levels were associated with EBV negative tumors, after adjusting for age and histology. Several markers were associated with factors in the International Prognostic Score (Table 1). Adjusting for IPS, stage and treatment arms, high levels of IL1RA, sCD30, sCD163, IP10, and IL-10, were significantly (p<0.05) associated with inferior FFS. While high levels of TARC (p=0.02, hazard ratio (HR) 0.88, 95% confidence interval (CI) 0.79-0.98), and CXCL13 (p=0.04, HR 0.84, 95% CI 0.72-0.99) were associated with better OS, high levels of CD14 (p=0.02, HR 2.2, 95% CI 1.11-4.38), IP10 (p<0.0001, HR 2.22, 95% CI 1.64-3.00) and sCD163 (p=0.0002, HR 2.89, 95% CI 1.65-5.06) were associated with significantly inferior OS (Figure 1). Conclusion: These findings support prior work demonstrating the prognostic significance of markers such as CD163 in cHL tumor tissue with similar findings using a more readily available (blood-based) method of assessment. The study expands upon the number of blood-based markers shown to be prognostic in cHL, and associates high pre-treatment levels of several markers with IPS factors and other baseline clinical characteristics. Further, and most importantly, IP10 and sCD163 predict inferior FFS and OS independent of IPS and treatment. [Display omitted] [Display omitted] DisclosuresCheson:Pharmacyclics: Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Research Funding; Acerta: Membership on an entity's Board of Directors or advisory committees, Research Funding. Winter:Pharmacyclics: Research Funding; Medivation: Other: Provision of investigational agent for clinical trial; GSK: Research Funding; Seattle Genetics: Research Funding. Friedberg:Bayer: Honoraria, Other: Data Safety Monitoring Board. Kahl:This study was coordinated by the ECOG-ACRIN Cancer Research Group (Robert L. Comis, MD and Mitchell D. Schnall, MD, PhD, Group Co-Chairs) and supported by the National Cancer Institute of the National Institutes of Health under the following award number: Research Funding.