We were thrilled to see the highly-respected New England Journal of Medicine (NEJM) report on the JDRF continuous glucose monitoring trial results, almost as if the EASD (the major European meeting that took place in Rome all last week) had planned and sanctioned its release! The study was said by many experts to be extremely well designed and executed with results revealing much needed information about glucose sensor optimization. We see it as a positive for patients, doctors, educators, manufacturers of CGM, and payors - despite it being early in the life of the technology, there are some long-term very bright lights here. In the presentation, researchers went into the data in even more detail than the NEJM offered, and voiced some of their opinions and speculations regarding its consequences. The main news was that the A1c drop was statistically significant for adults, whereas children and teens didn’t see a significant A1c benefit – this from our view had to be tied to usage on the device, where adults used the CGM significantly more than kids and teens. Here’s our take:
What is this trial we keep hearing about? It’s a landmark study that for the first time demonstrates that CGM use can clearly lower A1C by a significant month in a six-month time frame. Health insurers and national health services, take note! The investigators anticipated results well and designed accordingly so that each age group was powered a priori for a 0.5% A1c drop. They also understood that CGM success has a significant behavioral component and gave a lot of attention and support to patients. Patients were either sorted into a SMBG group or real-time CGM usage; patients received education in glucose control and users of CGM had additional training on the devices. They could choose from any device (Abbott’s Freestyle Navigator, DexCom, or MiniMed Real-Time CGM) and were free to switch devices throughout the trial. Patients had several follow up clinic visits periodically in addition to phone support follow-ups.
What were the results? There was a significant improvement in what they call both primary and secondary outcomes with CGM use. After the 26 week study, there was a -0.5% decrease in the adult CGM group from baseline of 8.0% and a tiny 0.03% increase in the control group, combining in a highly significant between-group difference of 0.53%. Significant differences were seen for outcomes as well, with 35% of adults in the CGM group reaching a target of 7.0% compared to about 10% in the control group, 25% in CGM versus 5% in the control achieving a relative drop of ≥10%, and 47% versus 10% seeing an absolute drop of ≥0.5%. Results show that 83% of adults in the trial used CGM > 6 days/wk over the entire six months. We think what led to these results is that adults figured out how to use CGM to get less hyperglycemia and less hypoglycemia.
Side note: Patients were mostly recruited from some of the top diabetes practices in the USA. Dr. Aaron Kowalski of the JDRF said that although the patients were representative of these practices, they were not “otherwise special” and other doctors concur that patients in the trial weren’t ‘special’ since they all had A1c >7%. We believe excellent support given during the trial made a difference on the results – patients received six in person check-ins and scheduled telephone calls between visits. We do think it’s critical that funding increase for doctors and nurses so they can spend time with patients helping them learn this valuable technology.
Additionally, it was good to see that hypoglycemia did not increase in the CGM group – in fact, it decreased in adults while at the same time A1c also decreased. While the hypoglycemia reduction wasn’t “significant,” the trend toward less hypoglycemia was – this is certainly exciting as we at Close Concerns who had diabetes can attest – we love anything that will help us have less hypoglycemia and have certainly used CGM toward that end. Doctors who presented results at EASD said that hypoglycemia is the biggest barrier standing in the way of better A1cs – this has been said for years and finally there is something that can address these barriers.
Other “secondary” (in science lingo) outcomesincluded the percentage of patients achieving an A1c target of 7%, and those achieving a ≥10% relative or ≥0.5% absolute A1c drop, as well as CGM profiles obtained in both groups at three and six months, data about hypoglycemic events, and questionnaires about patients’ quality of life and the costs/benefits of CGM use. We’ll be finding out much more about secondary outcomes as the data is analyzed – and this trial is actually a one-year trial, so there will be key “one-year” results to see.
What are the next steps? There is still much more extensive work to come, including an analysis of hypoglycemia in 120 patients with A1c <7% (already in the target zone), what they call psychosocial factors (think little family-diabetes conflict, ability to access medical care, etc.) and healthcare economics, and the outcome of the follow-on six-month crossover study, when all the patients use CGM. After the first six months, patients in the CGM group will continue CGM therapy for another six months, and patients in the self-monitoring blood glucose (SMBG) group will be offered CGM therapy for the same period with a less intensive management schedule. This additional period was designed to allow researchers to examine how long CGM works (this is what they call “durability”) and how long positive effects seen with CGM last and to examine whether CGM use is beneficial even without intensive clinical follow-up.
What’s the future of CGM? The bottom line seems to be that some patients do respond to CGM better than others, and that patient motivation, support and improvements in the devices can make a big difference. It turns out that teens and children in the study used the devices much less than adults did – kids only used the devices 51% of the time and young adults 30%). We aren’t surprised not to see substantial A1c changes if people didn’t use the devices (!) and it just raises the question that now the technology is here, how can it be made easier to use.
The results seem to point out that there are ‘requirements’ for success of CGM therapy. Most doctors and interpreters of the trial recommend for patients to learn to react to trends (rather than static, exact numbers). Also, patients must be able to call upon their health care team, and members of these teams should start changing their preconceived notions about diabetes and invest the time necessary for optimal use of CGMs. Although both younger groups saw less of an effect than the adult group, there are several predictors of CGM success, mostly surrounding time using the device and psychosocial barriers, that should be addressed (psychosocial barriers to diabetes care should be addressed anyway!). We’re very positive about the future of this important technology and look forward to further trial results in 2009! Congratulations to everyone who worked on this trial and who participated in it – this is just the start !