COPEPTIN AS A POTENTIAL BIOMARKER IN PATIENTS WITH AUTOSOMAL-DOMINANT POLYCYSTIC KIDNEY DISEASE (ADPKD)-UPDATE FROM THE AD(H)PKD STUDY

Abstract BACKGROUND Prediction of disease progression is one of the main challenges in the treatment of Autosomal-Dominant Polycystic Kidney Disease (ADPKD). In addition to already established biomarkers such as total kidney volume (TKV), further biomarkers that are feasible and easy to collect would be important to allow conclusions on prognosis as well as, in the best case, on long-term treatment response. Post-hoc analyses of the TEMPO 3:4 study showed that copeptin could serve as such a biomarker [1, 2]. In this study, we investigated copeptin over a multi-year course in participants of the AD(H)PKD study. METHOD Copeptin was measure from serum samples of 486 patients in the AD(H)PKD study at both initial and follow-up presentations. A total of 95 patients were on tolvaptan therapy at least at one of the measurement time points. We examined the measured values with respect to their distribution in different patient groups (e.g. age, Mayo class) and their changes on tolvaptan administration. Baseline eGFR values were developed using historic creatinine measurements up to 8000 days prior and 1500 days after copeptin measurements. For each patient with at least 5 creatinine measurements an eGFR slope was calculated. The models were developed using a robust linear modelling approach. Slopes were extracted to be the parameter associated to date and multiplied by 365 to retrieve an annual eGFR slope. eGFRs predicted for the specific dates at which copeptin has been measured were then calculated and used in subsequent modelling approaches. To develop a model predictive of eGFR, we used a stepwise regression approach, removing insignificant coefficients at each step. RESULTS As expected, copeptin was significantly lower in 391 tolvaptan naive patients than in 95 patients on therapy (9.95 pmol/L versus 19.78 pmol/L; P < 0.0001). More advanced CKD stages showed higher copeptin levels, both with and without tolvaptan, than at lower stages. Independent from kidney function, copeptin levels were consistently higher on tolvaptan than without therapy. Also, larger TKV showed higher copeptin levels in both groups. In 40 patients, copeptin measurements before the start of tolvaptan therapy and follow-up measurements under tolvaptan at the target dose (90/30 mg) were available, and a significant increase in copeptin levels on tolvaptan (copeptin before tolvaptan therapy: 9.798 pmol/L, copeptin on tolvaptan 23.83 pmol/L; P < 0.0001) could be traced during the course. Linear regression models showed that copeptin is an independent variable for the prediction of eGFR slope when added to age, height-adjusted total kidney volume (htTKV) and baseline eGFR. Besides, correlation of predicted future eGFR based on Mayo Classification is slightly but significantly improved by adding copeptin to the model (R2 Mayo 0.9244 and R2 Mayo + copeptin 0.9446). CONCLUSION Our data obtained from the real-life setting should again be reconciled with the results from the TEMPO 3:4 cohort. Our findings show that copeptin levels are both associated with kidney function and with future eGFR loss. Adding copeptin to the model of the Mayo Classification indeed improves the prediction of future eGFR. However, it is not clear yet whether this mild effect will have an impact in routine clinical care. Copeptin has the potential to facilitate prognostic assessment as a biomarker and thus could be helpful in deciding on therapy initiation and possibly more patient-specific dosing. Future analyses of outcome in relation to copeptin will provide further data and might enable implementation in clinical practice.