Latest Diabetes Care issue highlights CGM, including consensus statements from JDRF, ATTD; major win for Beyond A1c movement – December 12, 2017

We were elated to see the December issue of Diabetes Care jam-packed with CGM news – a whopping eight out of 27 articles fell under “Continuous Glucose Monitoring and Risk of Hypoglycemia,” the very first category listed in the table of contents. Several items in the journal dealt with establishing clinically meaningful outcomes beyond A1c, aligning perfectly with the goals laid out at ADA 2017 last June and the Outcomes Beyond A1c workshop in July. This report focuses on two papers in particular – CGM consensus statements from JDRF’s T1D Outcomes Program and ATTD.

Included in this issue was an enthusiastic editorial on the Maturation of CGM and Glycemic Measurements Beyond HbA1c written by luminaries Drs. Matthew Riddle, Hertzel Gerstein, and William Cefalu. As quoted in the editorial: “Periodically, a new idea, method, or tool leads to a turning point in the management of diabetes. We believe such a moment is now upon us, brought by development of reliable devices for continuous glucose monitoring (CGM).” This is the strongest CGM endorsement yet from ADA/Diabetes Care and should bode very well for prescribing and reimbursement. The editorial highlights how CGM can help solve previously unanswered questions, including the link between complications vs. time-in-range, hypoglycemia, and glycemic variability. The authors expressed particular excitement about using CGM data as an outcome in clinical trials – we hope to see this become standard in the next few years! Above all, the patient experience with CGM must remain the focus, and the authors note that optimal CGM use (for now) may require more rather than less HCP support. Next steps include benchmarking outcomes beyond A1c (What should time-in-range goals be?), stronger research on who should use CGM, and how best to provide individualized advice.

The long-awaited JDRF T1D Outcomes program publication is also out – sharing consensus on outcomes beyond A1c from AACE, AADE, ADA, the Endocrine Society, JDRF, the Helmsley Charitable Trust, the Pediatric Endocrine Society, and the T1D Exchange. Likewise, ATTD has its own consensus statement in this issue following the February 2017 meeting. The papers reinforce each other brilliantly (see tables below), with clear alignment on CGM thresholds for reporting hypoglycemia (<70, <54 mg/dl), time in range (70-180 mg/dl), and hyperglycemia (>180, >250 mg/dl). The ATTD paper also makes recommendations on variability (CV <36%), recommended CGM data sufficiency (two weeks) and time blocks, how to define episodes, and more.

An important distinction is JDRF’s focus on type 1, whereas ATTD provides guidance for type 1 and type 2. The ATTD paper is also a bit more comprehensive, focusing more broadly with sections on limitations of A1c, use of CGM and SMBG, minimum performance requirements for CGM, glycemic variability, and data reporting. Both papers are very well cited and have an incredible author lineup. Read on to learn more , or read each in full here (JDRF et al.) and here (ATTD).

Consensus Statements

JDRF

Summary of Consensus Definitions (adapted from table 1)

ATTD

Key Metrics for CGM Data Analysis and Reporting (adapted slightly for readability from table 1)

* Severe hypoglycemia (level 3) and diabetic ketoacidosis (level 3) are not key CGM metrics per se. However, these conditions are included in the table because they are important clinical categories that must be assessed and documented.

“The electronic AGP report visualizes the key CGM metrics: 1) mean glucose, 2) hypoglycemia: clinically significant/very low/immediate action required, 3) hypoglycemia: alert/low/monitor, 4) target range, 5) hyperglycemia: alert/elevated/monitor, 6) hyperglycemia: clinically significant/very elevated/immediate action required, 7) glycemic variability, 8) eA1C, 9) time blocks, 10) collection period, 11) percentage of expected readings, 12) hypoglycemia/hyperglycemia episodes, 13) area under the curve, 14) hypoglycemia/hyperglycemia risk, and 15) standardized rtCGM/iCGM visualization. AUC, area under the curve; Avg; average; IQR, interquartile range; MAGE, mean amplitude of glucose excursions; MODD, mean of daily differences.”

• The JDRF steering committee defined time in range as time spent between 70 mg/dl and 180 mg/dl in an effort to keep the range wide enough to account for variation and what is reasonable with current tools (see the recent Tidepool data on how time-in-range varies by age, even in a pretty engaged cohort). The ATTD paper similarly aligns on time in 70-180 mg/dl, but adds that the tighter 70-140 mg/dl can be a “secondary” range. As with all things in diabetes, we expect individuals with diabetes and their HCPs with choose an appropriate range; however research studies need to report things in the same way, which makes these statements a tremendous step.
• Both the JDRF and ATTD consensus statements defined three levels of hypoglycemia: (i) Glucose <70 mg/dl and ≥54 mg/dl; (ii) Glucose <54 mg/dl; and (iii) A severe event characterized by altered mental and/or physical status requiring assistance. At level one hypoglycemia, people with diabetes are at risk of dropping below 70 mg/dl and becoming hypoglycemic, thus establishing this range as a warning sign to take action. Even in the absence of severe acute symptoms, repeated episodes of hypoglycemia can lead to hypoglycemia unawareness, justifying glucose levels <70 mg/dl as clinically significant. At level two hypoglycemia, neurogenic and neuroglycopenic hypoglycemic symptoms arise, warranting immediate action. Brain dysfunction occurs at levels <50 mg/dl, so it is critical that steps be taken immediately. Level three hypoglycemia is not mutually exclusive from level one or two, as certain individuals can function independently at a blood glucose < 54 mg/dl – the defining characteristic of this level is required assistance from others. The ATTD panel emphasized that for clinical study CGM outcomes, hypoglycemia values <54 mg/dl should be more strongly considered than those in the level one range    
  
  • What about event duration? The JDRF paper noted that consensus on timing (how long an individual must remain in level one or two to be classified as a hypoglycemic event) is lacking. However, the ATTD panel has taken steps to resolve this issue, defining a hypoglycemic event as spending at least 15 minutes below the specified threshold. Prolonged hypoglycemia was defined as a consecutive 120 minutes or more spent below 54 mg/dl.
• Both papers split hyperglycemia into two levels: (i) Glucose >180 mg/dl and ≤250 mg/dl; and (ii) Glucose >250 mg/dl. While glucose profiles and postprandial blood glucose data indicate 140 mg/dl as the appropriate threshold for defining hyperglycemia, studies show that not only do the majority of people with diabetes exceed this threshold daily, they spend more than 60% of their day above it. Thus, the steering committee agreed to align with current clinical guidelines identifying 180 mg/dl as peak prandial glucose. The threshold of 250 mg/dl was chosen, as blood glucose levels surpassing this value are associated with increased risk for DKA and other complications. Still, more research is needed to elucidate the link between hyperglycemia and macrovascular disease.

Notable Quotes and Other Takeaways from Both Consensus Papers

JDRF

• “In the U.S., approximately one in five children and one in three adults meet hemoglobin A1c targets and the average patient spends 7 h a day hyperglycemic and over 90 min hypoglycemic.”
  
  • This is no surprise to those who work in the field, but this single sentence really frames the tremendous challenges and barriers type 1s still face – and these are at the best centers in the country. Though a T2D Exchange does not exist, we’d guess the stats are the same or perhaps worse.
• The JDRF paper emphasizes that time in range is a bar for effective interventions: “It is also evident that if the intent of a given intervention is to safely manage blood glucose but the intervention does not reliably maintain blood glucose within safe levels, then the intervention should not be considered effective.” Hear, hear!
• “To date, there is limited research correlating time in range with positive short-term and long-term type 1 diabetes outcomes, as opposed to the extensive research demonstrating the negative consequences of excursions into hyper- or hypoglycemia. More substantial evidence demonstrating a correlation or a direct causative relationship between time in range for patients with type 1 diabetes and positive health outcomes is needed.”
  
  • We hope this becomes a key research focus moving ahead – with an estimated 0.7-1.0 million CGM users worldwide (Close Concerns’ estimate, per Adam’s IDF presentation), there is now a lot of CGM data that can be correlated to short- and long-term outcomes. We hope the academic and industry communities work together on this front!
• “The use of validated PROs in type 1 diabetes clinical research is not currently widespread, and challenges to effectively measuring some PROs, such as quality of life, continue to confront researchers and developers. While many studies of type 1 diabetes treatments, including devices, in some way assess PROs, further work is needed to develop standard PROs for type 1 diabetes, including assessments of burden to patients. Such measures would need to be applicable across and between age ranges, settings, and over multiple years to evaluate trends in order to be relevant at the clinical trial level.”
  
  • This is also one of the clear next frontiers for the Beyond-A1c consensus movement: standardizing PROs. With the right commitment, we have no doubt this can happen over time; however, we’d guess the path might be longer than that for standardizing CGM reporting. With CGM, there was some baseline consensus on the thresholds, and only nuances had to be agreed on (e.g., less than vs. less than/equal to). With PROs, there are a wider variety of measurement tools (vs. CGM), and there seems to be generally less consensus at this stage. That said, we hope the challenge here motivates research and consensus alignment – this IS a critical area!

ATTD

• “Although HbA1c remains the reference marker for assessing glycemic control and predicting the risk of development of long-term complications, it has several limitations: HbA1c 1) provides only an average of glucose levels over the previous past 2–3 months; 2) does not detect hypoglycemia or hyperglycemia on a daily basis; 3) is an unreliable measure in patients with anemia, hemoglobinopathies, or iron deficiency and during pregnancy; 4) does not reflect rapid changes in daily glucose control; and 5) does not provide data as to how to adjust the treatment regimen when HbA1c levels are elevated. In summary, although HbA1c has proved extremely valuable in patient management, is a valuable measure of population health, and remains a validated indicator of glycation as a risk factor for complications, it is not as helpful for personalized diabetes management.”
  
  • We felt this was one of the best, most succinct summaries we’ve ever seen of A1c’s limitations. The ATTD paper makes the point that A1c is well validated and a great population measure, but co-interpretation with CGM data offers significant additional value. Agreed!
• Interestingly, the ATTD statement emphasized that time in “ranges” (TIRs), rather than time in range, provides a more descriptive metric for clinical and research outcomes. Time in range alone cannot capture the complete picture either. Instead, the authors note that composite measures have value. For example, by combining time in target range with time in level two hypoglycemia (see above), one could determine whether a therapy achieves increased time in target range while also remaining non-inferior to the amount of time spent in hypoglycemia. We love this detailed, creative approach, as it has the potential to evaluate therapy efficacy for specific outcomes, allowing for improved individualization of treatment. That said, EMA’s Dr. Bart Van der Schueren has noted at recent conferences that composite measures are difficult for regulators, since they often conflate safety and efficacy endpoints. This is arguably most important for adjunctive therapies in type 1 diabetes like SGLT-2 inhibitors. 
• “CGM should be considered in conjunction with HbA1c for glycemic status assessment and therapy adjustment in all patients with type 1 diabetes and patients with type 2 diabetes treated with intensive insulin therapy who are not achieving glucose targets, especially if the patient is experiencing problematic hypoglycemia.”
  
  • It was excellent to see this focus on both type 1 AND type 2. We’ll be fascinated to see how the latter develops in the coming years as the major CGM players move into type 2 diabetes with easier to use, less expensive products. 
• The ATTD statement emphasized that glycemic variability derived from CGM data should be considered as contributing to the overall clinical representation of glycemic control. The authors recommended that coefficient of variation (CV) be the primary measure of variability, where stable glucose levels are defined as CV <36%. The JDRF paper did not address glycemic variability with this level of specificity. This recommendation is directly in line with a paper by Monnier et al. that appeared in Diabetes Care in 2016.
• The ATTD consensus statement noted that there are currently no internationally accepted standards for CGM performance. While MARD is the most commonly used metric, the authors advised against comparing MARD values from different clinical studies, recommending instead that head-to-head studies be performed. Additionally, with the exception of the FreeStyle Libre, all CGMs require calibration via SMBG, and given the DTS’ BGM Surveillance results released last July identifying just six of the 18 tested BGMs to achieve sufficient accuracy, the ATTD authors emphasize use of an accurate BGM for calibration. We’d STRONGLY echo this point, since it drives both the subjective experience and objective performance of using CGM.
• “ISO/IEEE FDIS 11073-10425 provides a normative definition of the communication between CGM devices and managers (such as cell phones, personal computers, personal health appliances, and set-top boxes) in a manner that enables plug-and-play interoperability.”
  
  • Though this was a small mention in the paper, we’re elated to see this focus on interoperability and use of communication standards for CGM. Hopefully combined with JDRF’s Open Protocol Initiative, all devices will move to a more plug-and-play model.
• “In adults with type 1 diabetes, severe hypoglycemia is more related to duration of diabetes and socioeconomic status than HbA1c… Similarly, in children aged 6–17 years old with type 1 diabetes or adults with type 2 diabetes (mostly receiving insulin or sulfonylureas, severe hypoglycemia was most common among those with the lowest and highest HbA1c levels.”
  
  • This has been a running theme at different conferences – the traditional relationship between tighter control driving severe hypoglycemia is not apparent in recent studies.

-- by Maeve Serino, Adam Brown, and Kelly Close