Equivalence and noninferiority trials – are they viable alternatives for registration of new drugs? (III)