In order to assess inhibitor development in previously untreated patients (PUPs) with severe (factor VIII [FVIII]<1%) and moderate (FVIII 1 to 5%) hemophilia A, a prospective study was initiated in 1976. During the 23-year study period, 72 hemophiliacs were frequently exposed prophylactically or on demand to plasma-derived (pd) (n = 51) or recombinant FVIII (rFVIII) (n = 21) concentrates (median 270 exposure days [ED]). Inhibitor testing was performed before the first exposure and at regular intervals thereafter. Of the 72 hemophilia A patients, 22 (32%) developed an inhibitor after 15 ED in median (range 4 to 195); 17 (77%) were high responders (>5 Bethesda Units [BU]), and the remaining 5 patients (23%) were low responders (>0.6 to 5 BU). The severely affected patients (n = 46) showed a significantly higher frequency of inhibitor formation (43%) than did the moderate ones (8%). Comparing the severely affected patients receiving pd products exclusively (n = 35) with those treated with recombinant concentrate (n = 11), 37% of the pd group developed a high-titer inhibitor (>5 BU, median 290 ED in noninhibitor patients) and 36% of the recombinant group (median 49 ED in the noninhibitor patients). However, the exposure status of the recombinant noninhibitor patients is rather low and therefore remains a high risk of developing further inhibitors in the future. The mutation type profile revealed no difference between the pd- and the recombinant-treated patients.