Furthermore the baseline eGFR and the yearly eGFR changes seen in CRISP and SUISSE ADPKD were similar to the Chinese patients in the respective age categories. Taken together this suggests that the impact of the Chinese ethnicity on ADPKD is negligible. As a consequence, clinical trials in Chinese ADPKD patients might rely on similar assumptions regarding the definition of clinical endpoints. Earlier prospective studies in patients with ADPKD have shown that GFR and kidney volume correlate. Although the follow-up time in our cohort was short we could compute the yearly changes in TKV and eGFR in an extended number of age categories. An important finding was that many patients displayed a pronounced and unpredictable creatinine rise and eGFR decline, and this was apparent in all age categories. This suggests that it remains difficult to predict the course of the renal function in individual patients. Furthermore, steep increases in TKV were also found in all age categories and especially in patients with baseline TKV greater than 1500 cm3. For reasons not yet well understood it appears that the predictability of the change in TKV is as difficult as the predictability of changes in eGFR. This is also illustrated by the poor correlation between the yearly eGFR and TKV changes in our cohort, and by the data in Figure S2 which show that there are large excursions of TKV growth which do not correlate well with increasing baseline TKV. As longer follow-up times Bortezomib become available in our cohort the sudden volume changes – which could be due to hemorrhage or cyst rupture – might become less obvious, and the average TKV growth rate should become more reliable. The linear regression analysis revealed that the yearly decrease of eGFR was significantly associated with higher log10 protein/ creatinine ratio, log10 baseline TKV and age. On the other hand the linear regression analysis with percental yearly TKV change as dependent variable revealed that the intake of antihypertensive drugs, male sex, lower thrombocyte count and higher log10 protein/creatinine were associated with this variable. The fact that the TKV change was dependent on observation time implicates that the variance is larger for patients with short observation periods and decreases with longer observation periods. This might possibly be explained by the fact that bursting cysts lead to lower TKV at shorter observation periods but less so when the observation time is longer. With increasing observation time, the trend towards a steady TKV growth might get clearer. Consequentially and as stated above, it is essential for the study of TKV growth to maximize follow-up times in order to reduce the “noise” caused by ruptured cysts. An important finding which has not been described in other cohorts was the identification of a reduced thrombocyte count in older age groups. Although there is a small decrease of the thrombocyte count with increasing age in the normal population, we found a 32% lower thrombocyte count in the oldest age group when compared with the youngest cohort patients.