The Prognostic Value Of Dynamic Monitoring C-Reactive Protein (CRP) Serum Levels In NK/T-Cell Lymphoma

Abstract The prognostic value of dynamic monitoring C-reactive protein (CRP) serum levels in NK/T-cell lymphoma Background and Objective C-reactive protein (CRP) is a kind of acute phase protein against inflammatory reaction. CRP has been proved to be associated with prognosis in various malignancies. Yet, the prognostic value of CRP in natural killer/T-cell lymphoma (NK/TCL) remains to be discussed. In this study, we aimed to observe the dynamic change of CRP serum levels during anti-lymphoma treatment, and to evaluate the prognostic value of CRP as a simple and economical biomarker during the early stage of treatment in patients with NK/T-cell lymphoma. Patients and Methods Between January 2003 and December 2011, 161 patients with newly diagnosed NK/TCL at the Sun Yat-sen University Cancer Center (SYSUCC) were reviewed retrospectively. Clinical and laboratory information was collected and analyzed. The following CRP serum levels were evaluated: pretreatment CRP (CRP0), early-treatment CRP (CRP1, after one cycle of chemotherapy, or two or three weeks since beginning of radiation), and post-treatment CRP (CRP2, after first-line treatment, or failure of first-line treatment). Results Overall, 161 patients were reviewed and analyzed. The median age was 44 years (range, 11¨C74). The majority had localized disease (stage I/ II; 75.5%). The CRP serum levels (mean ± standard deviation) were as follows: CRP0, 17.2±23.1 mg/L; CRP1, 14.0±30.0 mg/L; CRP2, 14.1±27.0 mg/L. The pretreatment CRP serum levels correlated with others unfavorable factors, including serum lactate dehydrogenase (LDH) (P = 0.005), presence of bulky disease (P = 0.002), presence of B symptoms (P = 0.017), the International Prognostic Index (IPI) score (P = 0.001) and the Korean Prognostic Index (KPI) score (P = 0.001). By the time of the final follow up assessment (May 2013), the median follow-up time was 23.0 months (range, 1.0-106.0 months). The median overall survival (OS) and progression-free survival (PFS) were 45.0 months (range, 1.0-106.0 months) and 32.0 months (range, 1.0-74.0 months), respectively. The independent unfavorable prognostic factors for OS (P < 0.05) in multivariate analysis included: elevated CRP1 (P < 0.001), presence of bulky disease (P = 0.007), and elevated β2-microglobulin (β2-mg) serum levels (P = 0.004). The receiver operating characteristic analysis revealed a similar cut-off value of CRP (8.16 mg/L) to the default value of the biochemical analyzer (8.2 mg/L). In addition, the result suggested a satisfying capacity of CRP1 in predicting response to initial treatment (sensitivity 94%, specificity 74%) and 5-year OS (sensitivity 75%, specificity 90%), among different serum biomarkers, including EBV-DNA, LDH, β2-mg, CRP0, CRP1, and CRP2. In all groups of patients classified according to dynamic CRP values, the patients with CRP0 (+) /CRP1(-) presented the most favorable prognosis (5-year estimated OS: 72.5%). Conclusions Our study suggests that elevated early-treatment CRP is an independent unfavorable prognostic factor for patients with NK/TCL. Compared with the baseline CRP, the dynamic change of CRP levels may provide more important information for prognosis during treatment. Disclosures: No relevant conflicts of interest to declare.