Statistical power calculation and number of subjects needed

The anticipated frequency of CRR at month 24 in patients of the “no repeat biopsy” arm is 40%. The investigators consider that a difference of 20% in favour of the “repeat biopsy” arm would be clinically meaningful (i.e. a frequency of 60% CRR at month 24). To detect such a difference, with an α level of 0.05 and a power of 0.80, a number of 98 patients per arm is required. In order to take an anticipated 5% drop-out proportion into account, 206 patients with incident proliferative [2003 ISN/RPS class III/IV (A or A/C)] lupus nephritis will need to be enrolled. Patients with incident active membranous LN (2003 ISN/RPS class V) will also be enrolled in ReBioLup; however, the target number of patients per arm will be based on proliferative cases only, since power calculation has been based on data from studies of proliferative LN.

Based on an anticipated yearly inclusion rate of 3–5 patients per year per centre, i.e. 6–10 patients during a 2-year period, approximately 20–30 centres of LN expertise will need to participate in the study.