ORIGIN: no demonstrable benefits, yet beneficial.
These days, many of us seem to feel that every new large trial brings new disappointment. Following VADT, ADVANCE and ACCORD, many may have anticipated the negative results of ORIGIN (N Engl J Med 2012; 367:319-328) and the question arises whether any of the currently running large ‘hard-endpoint’ trials of new drugs have any chance of finding improvements in mortality or cardiovascular (CV) events. However, there is still much to be learned and treasured in the wealth of data thus accumulated.
ORIGIN set out to demonstrate a difference in the rate of cardiovascular events in people with IFG/IGT or type 2 DM and prior CV damage when comparing the early use of insulin glargine versus control treatment (which consisted mainly of diet and 1 or 2 oral glucose lowering drugs). In a group of over 12,000 participants, after a median follow-up of 6.2 years, the incidence rates of main CV events (2.94 vs 2.85/ 100 patientyears) were exactly the same, as were microvascular event rates, cancer rates and overall death rates. However, the rate of severe hypoglycaemia was threefold increased in the glargine group (1.00 vs. 0.31 events/100 patientyears). Overall, the conclusion must be that early insulin initiation in this secondary prevention population has no benefits.
So far for the bad news. Looking at the data from a different point of view, there is a lot of positive news to be learned from ORIGIN that will help us in clinical practice.
First of all, ORIGIN confirmed that it is perfectly possible to maintain people in good glycemic control (median HbA1c 44 to 47 mmol/mol) with conventional treatment for 6 years in a row (and possibly longer). While the data may be somewhat flattered by the 11% admixture of IFG/IGT patients, this is a pattern that is also observed in long-running observational studies such as the Dutch Hoorn study. Secondly, ORIGIN may help us debunk the myth of (sulfonylurea induced) hypoglycaemia. One of the most striking findings of ORIGIN is not that severe hypoglycaemia increases with the use of insulin, but the fact that 75% of those in the conventional arm never experienced any hypoglycemic event at all during the whole 6 year follow-up; even in the insulin arm 43% remained free from hypoglycaemic events. For confirmed events these rates were even higher (86% and 58%). While the details have yet to be studied, it is noteworthy that in the conventional arm diet (35%), metformin (46%) and Sulfonylureaderivatives (25%) formed the mainstay of therapy. Even if we assume that all confirmed hypoglycemic episodes occurred in those using SU derivatives, 30% would have been event free for the whole duration of the study; likewise less than 10% would experience a severe hypoglycemic event. Thirdly, ORIGIN help us to put the idea of insulin-induced weight gain in some perspective. While individual responses may have varied, the difference in weight gain between insulin users and those on conventional treatment over 6 years was on average only 2kg , further corroborating previous evidence from DCCT and the UKPDS insulin arm.
More is yet to be learned from ORIGIN. For instance, we desperately need more data about diabetes in the elderly. When looking at detail at the age of participants, were there different patterns of disease progression, glycaemic control or hypoglycaemia rates? The combined data from trials such as ORIGIN and the other mega-trials currently under way, even though their principal endpoints may show no benefits, may ultimately provide us with a much clearer view on the ‘natural’ course of diabetes in the 21st century.