Prognostic Role of the CD68+ Cell Count and Early Response Evaluation with Interim-PET and Levels of Plasma Thymus and Activation-Regulated Chemokine (TARC) in ABVD-Treated Classical Hodgkin Lymphoma

Prognostication in classical Hodgkin lymphoma (cHL) has mainly explored three areas of interest beyond the use of clinical risk factors in recent years: I. Early response evaluation with positron emission tomography scan (“interim PET”) after 2 cycles of treatment with doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) has become the most interesting functional imaging technique to accurately predict treatment response in cHL. II. An increased number of tumor-infiltrating macrophages stained with the CD68 antigen in the microenvironment of cHL has been shown to be a simple tissue biomarker with an adverse impact on outcome. III. Among circulating biomarkers, early changes in plasma levels of the chemokine TARC during therapy have been proposed as alternative biomarker for response evaluation.

The aim of our study was to evaluate plasma levels of TARC at diagnosis and at interim PET in combination with the tissue biomarker CD68 and interim PET as predictors of outcome in cHL. We included 84 patients diagnosed with cHL (45 limited stage, 39 advanced stage) between 2007 and 2013 who were treated with ABVD in our department. Median age was 34 years (range 15-73), 43 patients were males and 41 females. PET-CT scan was performed at baseline and after two courses of chemotherapy. Interim PET results were assessed using the Deauville five-point scale. Treatment changed according to the PET result in 2 patients only. Plasma TARC levels were measured using ELISA (Human CCL-17/TARC DuoSet, R&D Systems Inc), and CD68+ cells were stained using the PGM1 antibody.

When compared to normal individuals? (n=63), plasma TARC levels were elevated above the normal range (upper limit 162 U/ml) at diagnosis in 89% (73/82) patients with cHL. Significantly higher TARC levels were observed in patients with advanced stage disease (p=0.05) and patients with bulky disease (p<0.001). There was no significant association to the CD68+ cell count in the tumor tissue (>5% vs <5%, p=0.09). After 2 cycles of ABVD, interim PET according to Deauville criteria was negative (score 0-3) in 67 (85%) patients and positive (score 4-5) in 12 patients (15%). At interim PET, TARC levels were elevated (>162 U/ml) in 18% of patients and were significantly associated with a positive PET result (p=0.007). Progression-free survival (PFS) at a median observation of 28 months was 76% (95% CI, 65-84%). Strong predictors for PFS were a negative interim PET (86% vs 25% , p<0.0001) and CD68+ cell count <5% (88% vs 66%, p<0.03). Limited stage of disease (p=0.005) and age < 45 years (p<0.05) were favorable prognostic factors. TARC levels (with different cut-off levels and as continuous variable) both at diagnosis and at interim evaluation, were not associated to prognosis. The multivariate analysis including interim PET, CD68 count , stage and age, showed that only a positive interim PET and CD68+ cell count >5% remained independent predictive parameters for PFS (Hazard ratio 11.3, 95% C.I, 3.3-38.4, p<0.001; and hazard ratio 5.8, 95% C.I. 1.4-25, p<0.02, respectively).

In conclusion, baseline TARC levels are elevated in the vast majority of patients with classical HL in relation to tumor burden, and turn to normal in most patients after 2 courses of ABVD. Persistent elevated levels of TARC associate with a positive interim PET, but cannot substitute for it as outcome predictor. Interim PET and CD68+ cell counts in the microenvironment maintain their prognostic significance both in univariate and multivariate analysis.