Prognostic Utility of Serum TARC, MDC, IL10, and Soluble CD163 Levels with Positron Emission Tomography (PET) Response-Adapted Therapy for Advanced-Stage Hodgkin Lymphoma (HL): a SWOG S0816 US Intergroup Correlative Study

Introduction: A network of cytokines and chemokines exists in HL that maintains a favorable tissue microenvironment. Serum levels of cytokines and chemokines have been examined as potential biomarkers of response and prognosis. TARC (CCL17) and MDC (CCL22) are produced by Hodgkin cells, attract Treg and Th2 cells, and have been associated with advanced stage of disease and inferior response. IL10, a pleiotropic cytokine produced by Tregs and Hodgkin cells fosters an immune suppressive environment by inhibiting Th1 cells and has been associated with unfavorable outcome in HL. CD163 is present in anti-inflammatory M2 tumor associated macrophages, and soluble CD163 (sCD163) levels have also been associated with tumor burden and response during HL therapy. We sought to identify prognostic serum markers alone and in combination with PET in a recent large prospective study of untreated HL.

Methods: S0816 was a US intergroup phase 2, risk adapted trial for stage III/IV untreated HL where patients (pts) received doxorubicin, bleomycin, vinblastine, dacarbazine (ABVD) chemotherapy for 2 cycles. Based on interim PET after cycle 2 (PET2), pts received escalated bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone (eBEACOPP) x 6 (if PET2 positive) or ABVD x 4 (total 6, if PET2-negative). Pts also received a PET scan at the end of therapy (PET3). We examined serum levels of TARC, MDC, IL10 and sCD163 and correlated these with PET results and progression free survival (PFS) and overall survival (OS). Pts who provided informed consent to participate had planned serum samples drawn at baseline (pretherapy), after cycle 2, and at completion of therapy. Serum samples were frozen locally and tested centrally (ALLIANCE Molecular Reference Laboratory) by enzyme-linked immunosorbent assay (R&D systems, Minneapolis MN). Statistical analysis was performed at the SWOG statistical center. All statistical analyses report unadjusted p-value across the four serum markers.

Results: 559 samples from an initial 236 pts were available for study (236 baseline, 166 post cycle 2, and 157 post-therapy samples). The clinical features (age, stage, presence bulk disease, presence B symptoms, and international prognostic scores (IPS)) were similar between the 236 pts and the overall trial cohort. Elevated levels of the 4 markers were seen at diagnosis with rapid declines after cycle 2 and at completion of therapy. For PET results, univariate landmark Cox regression showed that PET2 correlated with OS (HR 3.4, 95% confidence interval (CI) 1.14 - 10.27, P=.028) and PET3 strongly correlated with both PFS and OS (HR 9.0 and 7.7, 95% CI 5.10-15.98 and 2.17-27.40, P<.0001 and P=.0016, respectively). Univariable analysis of baseline levels of the 4 markers showed no association with PFS or OS. Post-cycle 2 levels of sCD163, adjusting for PET2 result, showed association with PFS (HR 0.5, 95% CI 0.34 - 0.90, P=0.017), but other markers at this time point showed no association with survival. A landmark analysis of elevated post therapy levels of IL10, adjusting for PET3 result, were associated with both inferior PFS and OS (HR 1.6 and 2.3, 95% CI 1.25-2.06 and 1.50-3.53, P=.0002 and P=.0001, respectively, Figure 1). Post therapy TARC was also associated with PFS (HR 1.4, 95% CI 1.07-1.84, P=0.015). We also examined the change in serum markers over time in conjunction with PET response (baseline to post cycle2 and post-therapy) for association with survival. Only sCD163 impacted PFS in multivariable modeling adjusting for PET3 results (HR 0.6, 95% CI 0.37-89, P=0.012).

Conclusions: We evaluated serum levels for IL10, sCD30, TARC and MDC in the setting of a PET-directed regimen in advanced-stage HL. Baseline serum levels were not prognostic, but IL10 levels after therapy were associated with PFS and OS. sCD163 may provide additional prognostic information. Collectively, IL10 and sCD163 are worthy of further investigation. Given the promise of immune checkpoint inhibitors, IL10 and its immunosuppressive properties may be of particular interest in regimens containing these therapeutic agents.

Support: NIH/NCI grants CA180888, CA180819, CA180821, and CA180820.

View largeDownload slide

View largeDownload slide

Close modal