Clinical development of continuous glucose monitoring systems: considerations for the optimal strategy

Some glucose monitoring systems (GMS) are available nowadays and a number are in the approval/clinical development process. Up to now some clinical-experimental and clinical studies were performed evaluating the reliability of these GMS. Unfortunately, no standardized approach is used which practically prevent any comparison between the different GMS. At least in Europe it is relatively (too?) easy to get a CE-mark. Therefore, a standardized staged approach (comparable to that of the clinical development of new drugs) should be invented: Before the clinical development starts, clear specifications and targets should be defined in order to be able to critically review throughout the development program if such requirements are fulfilled or not (“achievement of mile stones”). In the first step, clinical-experimental studies with human beings should be performed. In glucose clamps studies the blood glucose levels should be kept constant at certain glucose levels for prolonged periods of time. In the second step, the evaluation should be performed under more clinical conditions with patients with diabetes under the conditions of an in-house setting. Patients are not confined to bed, are allowed to consume food ad libitum (or receive test meals, respectively) and maintain their usual diabetes therapy. Under such safe conditions any risk for the patient by erroneous glucose measurements of the GMS can be avoided. This also allows to evaluate the impact of different calibration procedures on the reliability of the measurements. In the third and last step, larger groups of patients should use the system on their own at home, making long-term use of the GMS. If the GMS shows a reliable performance in all such studies, one can assume that its use is safe and that it helps patients with diabetes to optimize their metabolic-control. Use of the proposed standardized clinical development procedure would allow comparing the results obtained with different GMS. The advantages and disadvantages will be discussed in detail. From my point of view there is a pressing need for a careful, critical and thorough evaluation of GMS before they come into the market.