A minimal mathematical model of RA-9 antioxidant action indicates these increases in GSH are nonlinearly related to BETP glucose reduction and are characterized by a glucose threshold. The model therefore predicts that lowering glucose below threshold can be expected to result in a dramatic improvement in GSH levels. In 34 of 49 patients we were able to quantitatively model the trajectories of individual responses; in each of these cases we obtain estimates of maximal glutathione at low stress, a glucose threshold for halfmaximal glutathione, and a rate at which recovery progresses. We have concentrated on investigating OS remission during therapy because antioxidant defense is likely to be a primary protection mechanism underlying the clinical control of glucose. Major strengths of this study include the use of GSH to obtain a quantifiable, objective measure of recovery from GS. The model itself is robust, and hence it was possible to use it reliably to deconstruct pathophysiological differences between individuals. The mathematical model is deliberately kept minimal, for two reasons: to retain the essential antioxidant action of glutathione in the simplest, robust fashion, and to avoid over-fitting data. The modelling procedures require only three measurements over two months; it remains to be seen if more frequent measurements of GSH and glucose would improve model estimates. It will also be interesting to construct more elaborate network models, since much is known about glutathione biochemistry; however, complex models will, in all likelihood, have to be fundamentally consistent with the minimal model. Should OS profiling become mainstay in diabetes research, such models could shed further light on the intricacies of individual differences. It will also be necessary to better understand sources of variability in glutathione both across, and within individuals. Further, our findings suggest that serial studies longer than eight weeks need to be carried out as they can reveal important information regarding the extent to which it is possible to push glucose control.