Since the hypothetical mechanisms of hydroperoxydation of arachidonic acid by, respectively, 5-lipoxygenase (5-LPO) and cyclooxygenase (CO) involve a redox cycle, a compound which reduces 5-LPO and CO to their inactive state would give a nonselective inhibitor of both enzymes. Structural modifications of such a compound could be expected to give improved potency and selectivity for 5-LPO and oral activity. Such an approach has led to the discovery of 1,2-dihydroindazol-3-ones which are potent 5-LPO inhibitors with various degrees of selectivity. Structure-activity relationship studies indicated that while N-1,N-2-unsubstituted and N-1-substituted derivatives are orally inactive, N-2-alkyl derivatives are orally active and inhibit both 5-LPO and CO. In contrast, N-2-benzyl derivatives are selective for 5-LPO but possess only weak oral activity. Further structural modifications have identified ICI 207968 [1,2-dihydro-2-(3-pyridylmethyl)-3H-indazol-3-one, 21a] which combines potent oral activity and high selectivity. Methemoglobin (MHb) induction by 21a in dog blood precluded its development for clinical use. Attempts at dissociating 5-LPO inhibitory properties and MHb formation showed that MHb formation in vitro seemed to be related to the redox potential of the compounds whereas 5-LPO inhibition was not. This study led to a series of 4-(N-n-pentylcarbamoyl)indazolinones which maintained in vitro 5-LPO potency but did not induce MHb.