The role of C-reactive protein as a biomarker in dogs with lymphosarcoma

C-reactive protein (CRP) is a highly sensitive but non-specific acute phase protein that is produced in the liver in response to injury, inflammation and disease. CRP has been by shown by several authors to be elevated in dogs with lymphoma (LSA) compared to healthy control dogs, however the findings of their small studies have suggested that CRP is too non-specific to be used in a clinical setting to monitor dogs with LSA that are receiving treatment with chemotherapy. Studies in the human literature have shown that CRP concentration is significantly affected by extent of disease and by the presence of clinical signs in patients with both non-Hodgkin’s and Hodgkin’s lymphoma, and it has been suggested that an elevated concentration of CRP at time of diagnosis carries a poorer prognosis. No veterinary studies to date have been large enough to investigate the effect of patient clinical parameters on CRP concentration, or to determine whether CRP concentration carries any prognostic significance. The aim of this study was to determine whether serum CRP was a clinically relevant biomarker in dogs with LSA using a large population. Specific aims included confirming if differences existed between the CRP concentration of healthy dogs and dogs with LSA, and determining if any patient variables had a relationship with CRP concentration. The effect on CRP concentration in dogs receiving chemotherapy treatment was investigated, with an aim to determine whether CRP could be used for categorisation of remission status. Finally, the role of CRP concentration as a prognostic indicator was investigated with respect to disease relapse and survival. This study included 59 control dogs and 73 dogs with LSA. Spare serum was prospectively harvested and stored at -70°C prior to batch analysis on an immunoturbidimetric assay (Pentra 400, Horiba ABX). Serial samples were obtained at time of diagnosis, throughout treatment with chemotherapy, at routine re-check appointments and at time of recrudescence of disease. Dogs with LSA were fully staged at time of diagnosis to determine the extent of disease and all dogs were assigned a WHO stage and substage. Of the dogs that received treatment with chemotherapy, a remission status (i.e. complete/partial remission, stable disease or progressive disease) was assigned at each visit based on the subjective response to treatment from the palpation of peripheral lymph nodes. Patient clinical data was retrospectively obtained from both paper and electronic records. CRP was investigated as both a continuous and categorical variable, and categorisation was based on the degree of elevation of concentration. Significance was set at a p-value of ≤0.05. CRP concentration was deemed normal if ≤10 mg/L. Results suggest that serum CRP concentration is significantly higher at time of diagnosis in dogs with LSA compared to the control dogs, however not all dogs with LSA exhibit an elevated CRP concentration. CRP concentration returns to within normal limits for the majority of dogs following 4 weeks of treatment with chemotherapy. Of the patient variables investigated, WHO stage and substage, and pre treatment albumin concentration were shown to be significant following univariate analysis. Significant differences between the median CRP concentration of the different remission statuses exists, however the overlap of ranges of CRP concentration in each group suggests that it could not be used to categorise remission status. The median CRP concentration at time of diagnosis is significantly higher than at time of relapse of disease; however CRP concentration does become elevated in most dogs at time of relapse. CRP concentration was not shown to be significantly elevated in the weeks leading up to relapse, indicating that this biomarker is not useful for predicting early relapse prior to recrudescence of disease becoming clinically apparent. Finally, results of the survival analysis revealed that pre treatment CRP concentration is not a significant variable, however immunophenotype and WHO substage are. This is the largest study to date investigating CRP concentration in dogs with LSA and is the first study to suggest a relationship between CRP concentration at time of diagnosis and pre-treatment albumin, WHO stage and substage. Findings from this study have confirmed those of previous authors; for example CRP concentration becomes elevated in dogs with LSA both at time of diagnosis and at time of relapse of disease, and that immunophenotype and WHO substage carry prognostic significance. Although CRP concentration did not show any prognostic significance, there was a trend for those dogs with a moderate and marked elevation in CRP concentration (>30 mg/L) to have a reduced overall survival time