FDA Diabetes Outcome Measures Beyond Hemoglobin A1c Workshop

Executive Summary

With more than 900 registered individuals attending in-person or watching live via FDA’s webcast, patients, regulators, industry, and professional organizations came together today for an 18-months-in-the-making, landmark FDA CDER Workshop on diabetes outcomes beyond A1c. A compelling speaker lineup supported time-in-range, hypoglycemia, and patient-reported outcomes (e.g., quality of life, burden) to supplement A1c in regulatory decision making. The Agency was far more positive than we had dared to expect today, and we are hopeful that proposed next steps will come to fruition: at least two more FDA workshops in the next nine months to more specifically discuss what metrics (e.g., <55, <70, 70-180 mg/dl, etc.) and patient-reported-outcome measures (which QOL instruments) to use in clinical trials and ultimately drug labels. Dr. Bob Ratner noted in the final roundtable that “This is the end of the beginning,” and FDA’s historically tough Dr. Jean-Marc Guettier optimistically added, “We’re all talking about roughly the same outcomes. This is where we start, and now we figure out how to make sure these outcomes eventually make it into decision-making.” In other words, the next step is driving to consensus on what to specifically measure.

Time-in-range had a strong showing today, and more FDA education will be valuable – the Agency still didn’t seem to grasp how far CGM has come, or why non-severe hypoglycemia is a critical outcome Consensus papers are coming in the next 12 months – on hypoglycemia and quality of life from ADA, on outcomes beyond A1c from the T1D Outcomes Program (led by JDRF, working with a range of professional groups including ADA, AACE, AADE, Endo Society, Helmsley Charitable Trust, Pediatric Endocrine Society, and the T1D Exchange) and on time in zone and PROs from dQ&A and The diaTribe Foundation – and the key will be securing FDA buy-in that these should be measured in trials, will be considered in regulatory reviews, and included in drug labeling. Industry was extremely positive on this front, as a wider set of outcomes will equip clinicians and patients with more comprehensive data to individualize therapy – particularly novel therapies that may dramatically improve quality of life, but might not impact A1c (or even raise it slightly). We believe that such outcomes would enable smarter and more directed use of more recently approved drug classes (GLP-1, SGLT-2 inhibitors, DPP-4 inhibitors) and presumably if these outcomes became widely used, perhaps more use of them could also be made in existing compounds.

The agenda was headlined by an enthusiastic opening talk from FDA Commissioner Dr. Robert Califf, followed by regulatory perspectives (FDA, EMA, Health Canada), patient advocacy perspectives (AACE ADA, JDRF,), and industry views (Novo Nordisk, Sanofi, Lilly, Janssen). Dr. Francis Kalush gave a compelling opening welcome and introduction – we are inspired that she was able to do so much to bring together so many stakeholders and plan such a strategic day of discussions. We were particularly impressed with how she choreographed so many views (national, international, young, older, industry, patient) – she and her team at FDA did a phenomenal job and we were especially impressed with the moderation of the roundtables by Dr. Kalush’s colleagues at FDA including Dr. Helene Clayton-Jeter and Rea Blakey. Below, we bring you our top highlights from the packed day, followed by The diaTribe Foundation/dQ&A’s slides. You can watch the entire day on diaTribe’s Facebook page here and can see the patient video here.

Top 10 Highlights

1. FDA representatives, particularly Dr. Francis Kalush and Commissioner Dr. Robert Califf (!) and Dr. Robert Temple, opened with quite thoughtful and positive perspectives on the prospects of outcomes beyond A1c. FDA understands the need to supplement A1c, and posed tangible questions on what to measure, how to measure it, and in whom. This is a solvable problem, and we’re leaving today pretty encouraged.

2. What are the next steps? Dr. Ratner, Dr. George Grunberger, Ms. Kelly Close, and others called for more dedicated FDA workshops in the coming year on time-in-range/hypoglycemia and quality of life. FDA is definitely open to more discussion, but did not firmly commit to future workshops yet. The most concrete step was Dr. Bob Ratner asking for two workshops in the next nine months. Our fingers are crossed!

3. Time in range, a metric that can now be measured with accurate glucose sensors, was a frequently suggested supplement to A1c. The diaTribe Foundation and dQ&A presented the most-quoted data of the day on this (n=3,455), highlighting that time-in-range has the biggest impact on patients’ daily lives (type 1 and type 2); that current therapies are falling short on time-in-range; and improved time-in-range positively impacts quality of life. See slides below and here.

5. Speakers from Novo Nordisk, Sanofi, Lilly, and Janssen provided strong and unanimous support for outcomes beyond A1c, especially composite metrics (e.g., % achieving A1c <7% + no hypoglycemia + no weight gain) and patient-reported outcomes and all suggested tangible ways of moving forward. Notably, all these speakers supported pre-competitive collaboration to standardize definitions. Industry badly needs FDA guidance on what metrics to use – there is no standardization right now and our sense was that previous attempts to offer related metrics had fallen flat. Indeed, to date, there is only one drug globally, Trulicity, that has a patient-reported outcome on its EMA label (buried in the Clinical Efficacy and Safety question, below weight results and above blood pressure – but there!): “Dulaglutide significantly improved total treatment satisfaction compared to exenatide twice daily. In addition, there was significantly lower perceived frequency of hyperglycaemia and hypoglycaemia compared to exenatide twice daily.” These results emanated from the Diabetes Treatment Satisfaction Questionnaire (DTSQ). Presumably, perceived frequency of hypoglcyemia and hyperglycemia could be matched to actual frequency if CGM had been used (though it was not used to support this specific claim).

6. University of Arizona’s Dr. David Marrero shared that an ADA group will soon publish a paper in Diabetes Care summarizing and recommending quality of life scales. Like many speakers, he conveyed that quality of life is very difficult to measure, and a wide array of general and diabetes-specific survey instruments exist – consensus is desperately needed and we look forward to the experts gathering and to learning from other therapeutic areas (whose job is probably easier due to great homogeneity of patients).

7. EMA’s Dr. Bart Van der Schueren voiced strong support for using CGM in drug trials and called for global, harmonized standardization of patient-reported outcome measures (PROs). Notably, he said that the EMA is in the midst of revising its 2012 Guidance to add even more about tools like CGM and FGM.

9. We have many questions leaving today: When will further workshops on time-in-range/hypoglycemia and quality of life happen? Will CDER accept CGM data to measure glycemic excursions? Will CGM data be incorporated in labels? Is there potential for diabetes drugs to make hypoglycemia or quality of life label claims in the future? How will FDA contribute to discussion and buy-in on which quality of life instruments or glycemic metrics to use? More below!

10. We include some of the most quotable quotes from today’s discussion below.

Table of Contents

Executive Summary
Top 10 Highlights
- - Quotable Quotes
The diaTribe Foundation/dQ&A Selected Slides

Top 10 Highlights

1. FDA representatives, particularly Commissioner Dr. Robert Califf and Dr. Robert Temple, were much more positive on the prospect of outcomes beyond A1c than we expected. Following today’s discussion, FDA understands and supports the need to supplement A1c, but seems unsure of what to measure and how to measure it. This is a solvable problem and will be the clear focus of next steps. Dr. Califf stated that we are currently “caught between A1c and death,” two important outcomes, and he characterized the biggest questions as:

What else should be measured?
How should it be measured and in whom?
How do we involve more people?

Dr. Califf was particularly positive in his compelling opening remarks on the value of social media as a democratizing tool that can elucidate what patients need/want. We were thrilled by how committed he was to helping people with diabetes and how he championed the importance of patient perspectives.

In a later talk, DMEP division director Dr. Jean-Marc Guettier acknowledged that A1c is not the only clinical benefit that is important to people with diabetes. He then echoed Dr. Califf’s musings, wondering what clinical outcomes beyond A1c the FDA should focus on, in whom, and why? These were most valuable questions, and have evolved from 2014’s FDA Advisory Committee for MannKind’s Afrezza. With guidance from the EMA about what the FDA would like to see on this front, we know the community can go a very far way.

In a panel discussion, Dr. Guettier raised challenges when it comes to patient reported outcomes and hypoglycemia (e.g., false positives, glucometer errors), but he did not suggest these are insurmountable. We believe that the FDA’s CDER can work closely with others at FDA – what has been learned from the improved accuracy of CGM and the comprehensive hypoglycemic and hyperglycemic metrics collected in automated insulin delivery trials at CDRH, for example? FDA’s Dr. Robert Temple was in attendance for most of the day and noted he has diabetes himself – this was not widely known in our view and it is valuable to know there is some patient perspective within the drug side of FDA. His leadership could lead to even more patient-friendly outcomes and aggressive follow-up to this workshop.

2. What are the next steps? ADA’s Dr. Robert Ratner, AACE’s Dr. George Grunberger, diaTribe’s Ms. Kelly Close, and others called for more dedicated FDA workshops in the coming year on time-in-range/hypoglycemia and quality of life; Some suggested driving toward revision of the FDA’s 2008 Guidance and it will be interesting to see if global regulatory agencies can collaborate, as recommended by the EMA’s Dr. Bart Van der Schueren. FDA is definitely open to more discussion, but did not firmly commit to future workshops. Our fingers are crossed! Panelists agreed that this meeting was only the tip of the iceberg, and that further collaborative conversations should be scheduled for the year ahead. During the final roundtable discussion, Dr. Ratner suggested two more meetings “within the next nine months” (before he retires in June 2017), and Ms. Kelly Close suggested two more workshops within roughly the same time period focused on hypoglycemia/time-in-range and quality of life.

FDA’s Dr. Jean-Marc Guettier also agreed that the day’s discussion was a good collaborative start: “We are grateful to have everyone in one room discussing common concerns across all stakeholders.” Dr. Guettier further summarized that many experts in attendance spoke about “roughly the same outcomes,” including time-in-range, hypoglycemia, and quality of life measures such as daily burden of the disease, fear of going low, and emotional well-being. In answering the question “what’s next?”, he explained that through continued collaboration, stakeholders in diabetes care can figure out how to incorporate these outcomes into decision-making in drug development and approval. We’d be thrilled to see rapid and robust follow-up meetings by August 2017, hopefully with even more collaboration and more specificity on what metrics and instruments to use.

3. Time in range, a metric that can now be measured with accurate glucose sensors, was a frequently suggested supplement to A1c. The diaTribe Foundation and dQ&A presented the most-quoted data of the day on this (n=3,455), highlighting that time-in-range has the biggest impact on patients’ daily lives (type 1 and type 2); that current therapies are falling short on time-in-range; and improved time-in-range positively impacts quality of life. dQ&A surveyed an impressive 3,455 people with diabetes (n=1,025 type 1s; n=1,150 type 2s on insulin; n=1,280 type 2s not on insulin) this month, asking about success of their current therapies, their priorities for diabetes care improvements, and the impact of diabetes on quality of life.

Wood said that 3,455 patients took this online survey, and mentioned that this adds up to approximately 900 patient hours committed to this project in just a few days. The response rate was high (62%): “We had over 1,000 respondents from each of three groups: people with type 1 diabetes, people with type 2 on insulin, and people with type 2 diabetes not on insulin.” Wood thanked the FDA for working on expanding diversity in clinical trials and emphasized that this was an important goal for dQ&A as well.

View select slides below (or click here to get the full deck), which told a fascinating story in the context of this meeting:

Few respondents said they feel “very successful” on current therapies, regardless of the metric: blood glucose, complications, emotional well-being, burden of care, social life. As the traffic light slide shows below, there was a lot of red across all six categories of measurement, and emotional well-being was unsuccessful across the board. It is our view that patients may feel more successful if labels were allowed to use such metrics today – this would enable better individualization and personalization of medicine. Many doctors appeared to agree with this, especially noted hypoglycemia expert Dr. Simon Heller – he remarked that it is better when patients and doctors decide on goals together and acknowledged this is not always possible.
Time-in-range had the biggest impact on daily life in the dQ&A survey with type 1 and 2 diabetes in the dQ&A survey. It was a shared - and unmet - need. Participants ranked a series of outcomes by whether they had a big impact, slight impact, or no impact on their daily life. For both type 1 and type 2 diabetes (on insulin and not on insulin), time-in-range had the biggest or equal to biggest impact on daily life. A1c was quite low in type 1 and more important in type 2 and Ms. Close emphasized, as did others throughout the day, that A1c was a very useful measure and that additional measures would supplement A1c for patients.

Rank Order: “Big Impact” on Daily Life	Type 1 (n=1,016)	Type 2 on Insulin (n=1,141)	Type 2 Not on Insulin (n=1,266)
#1	Time in Range	Time in Range	Time in Range
#2	Unexpected BG Numbers	A1c	A1c
#3	Dosing insulin	Non-Diabetes Health Issues	Non-Diabetes Health Issues
#4	Hypoglycemia	Dosing Insulin	Unexpected BG Numbers
#5	A1c	Unexpected BG Numbers	Symptoms of Diabetes Complications

But current therapies have significant room to improve on time-in-range. Just 35% of patients (n=3,460) reported that their current therapy is “very successful” at bringing blood glucose numbers between 70-180 mg/dl; 26% of patients reported that their current therapy is “very successful” at bringing post-meal blood glucose numbers to <180 mg/dl; and 32% of patients reported that their current therapy is “very successful” at bringing overnight blood glucose numbers between 70-130 mg/dl. It is our sense that these numbers would have been higher had patients been prescribed therapy more appropriate for all of their needs, which presumably would’ve been possible if these more specific outcomes were included. The survey also asked what would be most likely to put patients in a positive frame of mind about their diabetes and general health. “My blood glucose is on target all day” (i.e., time-in-range) was #1 in insulin users (type 1 and 2) and #2 in type 2s not on insulin; #1 in type 2 non-insulin users and #2 in type 2 insulin users was “I take my medicine as prescribed”.
Patient advocate Ms. Anna McCollister-Slipp highlighted plans for a study to connect time-in-range, glycemic variability, and excursions to quality of life. This could help validate the negative patient impact of non-severe glycemic events. The protocol is still in draft form, but would take Dexcom users with an iPhone and pair it with an Apple ResearchKit study. Patients would be prompted with in-the-moment quality of life questions (“I am feeling ...”) to link a current CGM-measured blood glucose to quality of life. We love the idea, particularly the use of emoji faces as an easy indicator of “I am feeling.” CDER met with Ms. McCollister-Slipp and Dr. Ratner about this study, which is hopefully a good sign it can get off the ground and make a difference.
At the meeting, CDER committed to working with ADA and Ms. McCollister-Slipp to refine the protocol for a measure connecting CGM data with real-time feedback from patients. The plan discussed previously and reiterated at the “Outcomes” meeting is to schedule a Critical Path Innovation Meeting to further refine the concept of the protocol, after which the more refined concept will be crowdsourced on VitalCrowd for broad input from patients, caregivers, physicians and researchers. At that point, the plan is to run the protocol by them again, then develop the study tools and launch the study as a ResearchKit study, pending final funding. Dr. Ratner and Ms. McCollister-Slipp will lead the effort along with colleagues at Scripps/STSI and Sage, will input from the entire community. Ms. McCollister-Slipp’s compelling slide presentation can be found here.

4. Hypoglycemia was also presented as a critical outcome beyond A1c, and experts emphasized the need for a global, standard definition and more attention paid to non-severe episodes. The International Hypoglycemia Working Group will propose a new classification defining three levels of hypoglycemia: <70 mg/dl (Level 1), <55 mg/dl (Level 2, “serious”), and severe cognitive impairment requiring third party assistance (Level 3, “severe”). As Dr. Elektra Papadopoulos (FDA, Silver Spring, MD) put it, “we don’t want individual companies coming up with their own measures; we need a harmonized core measure of hypoglycemia.” Ultimately, establishing a consistent standard definition will be an important step in standardizing data collection for all cases of hypoglycemia – severe or non-severe – and in promoting hypoglycemia as an outcome to consider in designing, approving, and labeling diabetes drugs. Patients will be glad to hear this – the dQ&A data showed that patients are troubled by all hypoglycemia not just severe; this is unsurprising as patients often reference productivity problems stemming from moderate hypoglcyemia. When the impact is considered on a population level, of course, this is particularly troubling. Drs. Robert Ratner and Simon Heller mentioned that the paper from the international hypoglycemia working group is forthcoming.

But even with future consensus on hypoglycemia definitions, we believe more FDA education will be needed on the clinical value of reducing non-severe hypoglycemia, as well as the accuracy of using CGM to measure hypoglycemia. The FDA seemed unfamiliar with the value of CGM in measuring hypoglycemia, and the recent progress in sensor accuracy has clearly not trickled over to the drug division at this stage. Dr. Guettier and others kept mentioning SMBG and associated measurement problems for detecting hypoglycemia, particularly with PRO questionnaires. FDA reviewer Dr. Lisa Yanoff mentioned that CGM is still a new tool and research around it is currently ongoing – we disagree at this stage that it is a “new” tool and point to the 2012 approval of Dexcom’s G4 as an instrument that has been around for quite some time that is accurate in hypoglycemia. We think a significant educational effort is needed here, ideally from the device side of the FDA. Perhaps CDER reviewers could wear CGM for a couple weeks! FDA’s Dr. Guettier explained that patient-reported outcomes could be useful in getting at a “softer endpoint” than severe hypoglycemia in order to probe for how the condition really affects patients’ lives – since plenty of evidence backs up the dangers of moderate hypoglycemia, we don’t think this would be as useful as the actual measurements, though PROs could supplement the numerical data. FDA’s Dr. Temple noted that “Hypoglycemia is of particular interest” as it is “dangerous and encourages poor control.” Still, he said it is “Not easy to show,” since it requires “less hypoglycemia with at least equivalent control” in one study arm vs. another. Dr. Aaron Kowalski noted in his valuable presentation that some patients would prefer slightly worse A1c and better hypoglcyemia – this did not seem surprising at all to many patients discussing this and we believe discussions on this front will be rich and varied. We think CGM can help a lot here, and showing equivalent glycemic control is not as necessary – what would be key in our view would be showing an improvement in hypoglycemia combined with an improvement in time-in-range. A drug that cuts hypoglycemia and improves time-in-range will increase A1c, an acceptable tradeoff for many patients vs. one that maintains baseline time-in-range. Dr. Temple suggested using enrichment designs (e.g., hypoglycemia unaware populations) – we look forward to other experts weighing in on what would be the most scalable and expedient.
Speakers also sounded off on the pervasive issue of hypoglycemia unawareness – in the absence of symptoms, broader use of CGM in drug trials could help identify episodes of hypoglycemia more accurately and comprehensively. Moreover, panelists pointed out that we miss crucial information when we focus solely on severe hypoglycemia – something no patient would argue with, but an area that FDA does not seem to be sold on. EMA’s Dr. Bart Van der Schueren and Sheffield’s Dr. Simon Heller further underscored that mild hypoglycemia still affects daily life in terms of fatigue, concentration, and a host of other challenges; this was reinforced in dQ&A data though not shown due to time constraints.
People with diabetes want medications to lower hypoglycemia risk. Novo Nordisk’s Dr. Alan Moses highlighted the incredible DAWN 2 study (n=15,438), which found that 56% of people with diabetes felt “very worried” about the risk of hypoglycemia. As noted, JDRF’s Dr. Aaron Kowalski spoke about his brother’s experience with severe hypoglycemia, articulating how a slight increase in his A1c (0.3%) would be well worth a substantial reduction in hypoglycemia risk (“a slam dunk”).

5. Speakers from Novo Nordisk, Sanofi, Lilly, and Janssen provided strong and unanimous support for outcomes beyond A1c, especially composite metrics (e.g., % achieving A1c <7% + no hypoglycemia + no weight gain) and patient-reported outcomes. Industry desperately needs FDA guidance on what metrics to use – there is no standardization right now. Fortunately, speakers supported pre-competitive collaboration to standardize definitions – that would be extremely valuable in our view. While it’s not particularly surprising that industry wants more data on the label (one keen investor noted that, in other therapeutic areas, this would be viewed as a negative in terms of introducing more bureaucracy and potentially time consuming requirements – very far from the case here!), Novo Nordisk’s Dr. Alan Moses outlined why this inclusion is especially critical in the US, where providers (and some patients) depend on drug labels to understand the value of treatment.

Building on this, Sanofi’s Dr. Rachele Berria explained that pharmaceutical reps are limited to discussing what’s detailed on a package insert, which means valuable educational information on health benefits beyond A1c doesn’t get communicated to providers (and consequently, to inform conversations with patients). Notably, she said that an ongoing Sanofi study has time-in-range as its primary endpoint – as we understand it, this is a Toujeo vs. Lantus trial in process now and we are very excited to see the results. “Will we see it on the label? I don’t know, but I think it was the right thing to do,” she emphasized. In response to a question from our very own Ms. Kelly Close about patient-reported labels in existence (she noted there was one in the EMA for Lilly’s Trulicity), Lilly’s Kristina Boye shared that this patient-reported outcome included on the European label for the GLP-1 agonist Trulicity (dulaglutide) was a positive, although the same information is not on the product’s FDA-approved US label. “We need to sit down, roll up our sleeves, and start working on validated patient-reported outcomes that will fit into FDA guidance,” she commented, arguing that it’s in patients’ best interest to have non-A1c benefits also listed clearly on drug labels. [While the Trulicity claim is only a few sentences long and fairly buried in the label, it is a starting point.] We hope various stakeholders and FDA can come together to decide on standardized measures. Right now, companies are measuring this in various ways, making it difficult to compare treatments and making regulators’ jobs harder.

We agree that an important end goal of this expansion to diabetes outcomes beyond A1c is inclusion of more data on labels, including data from CVOTs. While some were disappointed to hear earlier this month that the FDA has delayed a decision about a label update for Lilly/BI’s Jardiance (empagliflozin) to reflect the positive results from EMPA-REG OUTCOME, we feel confident at this point that the FDA is at work on this front and we await the outcome (so to speak). This serves as a good example of how a broader perspective on relevant diabetes outcomes (such as the all-important cardiovascular health) might be valuable in disseminating therapeutic information to busy patients and providers.

6. University of Arizona’s Dr. David Marrero shared that an ADA group will soon publish a paper in Diabetes Care summarizing and recommending quality of life scales in pediatrics and adults. Like many speakers, he conveyed that quality of life is very difficult to measure, and a wide array of general and diabetes-specific survey instruments exist – consensus is desperately needed. Still, he strongly encouraged the FDA to start examining the impact of therapies on quality of life. Such instruments are necessary because a treatment with excellent therapeutic efficacy but a poor quality of life impact may not be taken – the adherence to such medicines may be extremely poor, but that would not be captured in a randomized controlled trial. Conversely, a treatment with excellent quality of life and more moderate clinical efficacy may still be of value to patients. Inclusion of such assessment could therefore guide the clinical decision-making process and improve patient care. But what is a good measure? Dr. Marrero suggested that it would clearly identify the primary goal of an intervention, and then consider the effect on the participant if the goal is achieved. It would also (i) distinguish between patient-reported-outcomes in general and quality of life; (ii) avoid using solely composites of component measures, which risks reducing the chances of detecting specific changes; and (iii) consider the impact of an intervention on the quality of life of close family and caregivers as well as the patient. Dr. Marrero concluded by recommending that a panel re-convene to assess the state of available instruments, to determine if one should replace the existing options.

Dr. Marrero recommended a number of available quality of life evaluations, starting with pediatrics: Pediatric Quality of Life Inventory (PedsQL; includes a generic core scale and a diabetes module) and Parent and Family Health related quality of life (PedsQL Family Impact Module and Joslin’s Diabetes Family Impact Scale). He then moved on to discuss quality of life instruments for adults, which he characterized as “a little fuzzier” and more complex: Well-being index (WHO-5; looks at positive emotions and experience in general), Patient Health Questionnaire (PHQ-9; for screening, diagnosing, measuring, and monitoring the severity of depression), Diabetes Distress Scale (point prevalence of diabetes distress among type 2 adults is a striking ~42%, and 18-month incidence is ~54%), Type 1 Diabetes Distress Scale, Diabetes Symptom Checklist (DSC-R; correlates with outcomes in clinical trials). Dr. Marrero noted the lack of a partner/caregiver quality of life assessment – we are sure this is a gap that could be filled with the right support.

7. EMA’s Dr. Bart Van der Schueren voiced strong support for using CGM in drug trials and called for global, harmonized standardization of patient-reported outcome measures (PROs). Dr. Van der Schueren highlighted the improved accuracy of CGM, which gives much more information about glycemic control than the current consensus metrics. Indeed, the EMA’s 2012 clinical guidance for diabetes drugs recommends the use of CGM in in five instances, particularly for capturing overnight and post-meal glucose profiles. Dr. Van der Schueren argued for global standardization of patient-reported outcome measures (PROs) before these can be used as the basis of regulatory decision-making. Sanofi’s Dr. Rachele Berria provided the actionable suggestion of a collaborative task force to implement the FDA’s draft guidance for PROs, establishing evidentiary standards and a taxonomy of different PRO instruments. We would view such a meeting as valuable progress, particularly if it combined EMA and FDA together, as Dr. Van der Schueren suggested. We agree with Dr. Van der Schueren that early standardization efforts of this kind are paramount to ensuring consistency and future innovation – all companies sponsoring clinical trials and all regulatory bodies making decisions should ideally use the same group of validated PROs. Beyond streamlining the regulatory decision making process, Dr. Robert Ratner (ADA, Arlington, VA) pointed out that the development of standardized PROs would give regulatory agencies greater power to encourage company sponsors to incorporate these outcomes into their study designs, thus incentivizing therapies which perform well by these patient-centered measures.

The FDA’s Dr. Elektra Papadopoulos emphasized that in order for the FDA to make rulings on the basis of PROs, these endpoints must not only be rigorously statistically verified, but also validated for use in a broad range of people across diverse patient populations. For PROs to become globally standardized measures, it is essential that they are formulated to be easily translatable across both cultural and demographic boundaries. Dr. Robert Ratner echoed this sentiment, specifically highlighting the necessity of designing PROs that are adaptable across different age ranges. The distinction between pediatric and adult patients may be challenging to bridge under the same outcome measures, due to the vastly different priorities and daily functional demands these groups of patients experience.

8. One of the greatest challenges in outcomes beyond A1c is the heterogeneity of diabetes – what is important to one patient may not apply to another. An open question is what efficacy tradeoffs patients are willing to make (e.g., a 30% reduction in hypoglycemia is worth a 0.3% increase in A1c), and how those tradeoffs may differ in sub-populations. JDRF’s T1D Outcomes Program plans to assess this in type 1 diabetes, and presumably similar work can be done in type 2 diabetes. Lilly’s Dr. Kristina Boye eloquently pointed out, diabetes is unlike many other diseases insofar as there is not one agreed upon list of symptoms that all patients experience. Diabetes is heterogeneous and personalized disease, and one that is unique in its demand for constant, 24/7 attention and the invisibility of many of its symptoms. Whether an individual patient is most concerned with weight maintenance, time spent in range, or hypoglycemia avoidance, A1c is clearly far from the only issue as patients weigh different therapeutic plans. “It comes down to quality of life,” remarked immediate past AACE president Dr. George Grunberger, noting how well the notion of patient-reported outcomes aligns with the AACE’s (and ADA/EASD’s) guidelines advocating for a more individualized approach to diabetes therapy. Dr. Grunberger illustrated the heterogeneity of the diabetes experience point with the example of hypoglycemia: Minimization of hypoglycemia risk is an outcome of paramount importance to people with insulin-requiring diabetes and older drugs (that are in very common use) such as SFUs, but does not present much of a concern to those not taking insulin on newer drugs (SGLT-2s, GLP-1s) that are glycemic-dependent. Subjective measures of well-being and functionality are of course challenging to operationalize and standardize, but to ignore such factors in our discussion of diabetes outcomes would be to “diminish the patient voice,” as Janssen’s Dr. Shana Traina explained.

9. We have many questions leaving today: When will further workshops on time-in-range/hypoglycemia and quality of life happen? Will CDER accept CGM data to measure glycemic excursions? Will CGM and quality of life data be incorporated in labels? Is there potential for diabetes drugs to make hypoglycemia or quality of life label claims in the future? Will it be difficult to get FDA buy-in on which quality of life instruments or glycemic metrics to use? How far are we from consensus on both fronts? How will new measures impact innovation – will therapies be developed and reviewed faster? How will different patient groups and sub-populations prioritize different outcomes beyond A1c? What will JDRF’s T1D Outcomes Program and the next phase of the dQ&A data show? What will the commonalities of the work be in type 1 and type 2 diabetes? Will quality of life instruments be sensitive enough to detect changes with different therapies?

10. We include quotable quotes from today’s discussion below – this will be expanded in the coming days!

Quotable Quotes

What’s Next?

“We’re all talking about roughly the same outcomes. This is where we start, and now we figure out how to make sure these outcomes eventually make it into decision-making.” – Dr. Jean-Marc Guettier (FDA, Silver Spring, Maryland)
“Let’s schedule at least two more meetings in the next nine months ... This is the end of the beginning. We are in the pre-commercial, pre-competitive environment to define all of these parameters and put them forward prospectively.” – Dr. Bob Ratner (ADA, Alexandria, VA)
“I view defining outcomes beyond A1c as a palate of opportunity. What specific tests and what specific PROs should we use? If we use the ones that are pre-approved in an appropriate and well-executed trial, hopefully it will be included in the review and the regulatory language.” – Dr. Alan Moses (Novo Nordisk, Copenhagen, Denmark)
"The time is now, let's move ahead..." – Ms. Anna McCollister-Slipp (Patient Advocate, Washington DC)

On Moving Beyond A1c

“We are caught between A1c and death. Those are two pretty solid outcomes that are important. No one is arguing that these other outcomes are not important. We have an amazing opportunity as we work on new ways to study things... the question is what should we measure? How should we measure it and in whom? And how do we involve more people?” – Dr. Robert Califf (FDA, Silver Spring, MD)
“Hypoglycemia is of particular interest. It is dangerous, encourages poor control, compliance, and under-dosing.” – Dr. Robert Temple (FDA, Silver Spring, MD)
“Here were the key takeaways in the survey on moving beyond A1c: (i) First, few people with diabetes feel very successful on current therapies; (ii) second, time-in-range has the biggest impact on daily life for BOTH type 2 and type 1 diabetes; and (iii) third, people with diabetes would be very interested in a greater FDA focus on time-in-range and Quality of Life.” – Richard Wood (dQ&A, San Francisco, CA)
“It’s time in range …” – Christel Aprigliano (DPAC, Washington, DC)
“With the artificial pancreas guidance, we had this discussion with the device team at FDA. What is the trial design for labeling that would claim a reduction in hypoglycemia? In the draft guidance document, one of the debates we had was, “What if I caused a decrement in A1c?” There was some discussion on whether that would be an approvable device. I come with a personal perspective of having a brother with hypoglycemia unawareness, who has ended up in the hospital a number of times. That calculus, if you could have a device that reduces his risk of ending up in the hospital, but costs him 0.3% in A1c – that’s a slam dunk for him! That should be approvable if it’s safe. What we have consensus on is the physician and patients are making risk-benefit decisions based upon their situation.” – Dr. Aaron Kowalski (JDRF, New York, NY)
“For A1c to be the most important thing in the exam room is incomprehensible. We need a comprehensible global measure to capture hypoglycemia, hyperglycemia, and time-in-range. It comes down to quality of life. This perfectly aligns with the AACE guidelines advocating for more individualized therapy.” – Dr. George Grunberger (AACE, Michigan)
“Time-in-range defines the daily experience of living with diabetes. It’s a physical, biological, and psychological battle between two extremes. You feel terrible when you’re high or low; it’s all about staying in zone. People with diabetes spend so much time recovering from being out of zone. This is why we must move to outcomes beyond A1c – a three-month average cannot capture this …” – Kelly Close (The diaTribe Foundation, San Francisco, CA)
“Of course, weight is also a big issue. So what did we find out about it? First: It’s a challenge common to so many people with diabetes. The majority of these respondents have BMIs >30. A third are at 35 or higher. Second: despite new drugs, current diabetes care is not delivering success. Only 16% described their care as ‘very successful’ in terms of weight – that was our lowest score. Third, weight is stuck in the waiting room. Only 12% of respondents have a weight goal they have agreed with their doctor, compared to 42% who have an agreed A1c goal. 42% is not good for sure, but we feel that 12% is particularly alarming.” – Richard Wood (dQ&A, San Francisco, CA)
“It makes no sense. Unfortunately, A1c has become such a measure, it’s like a godlike thing out there.” – Dr. George Grunberger (AACE, Michigan)

On Patient-Reported Outcomes (PROs)

“We need to do more collective work in the pre-competitive space to define PROs.” – Dr. Alan Moses (Novo Nordisk, Copenhagen, Denmark)
“Let’s identify those PROs that industry, patient advocacy groups, and patients themselves can agree upon with the FDA moving forward. If the FDA says it wants certain parameters in the label, then all of industry will do it this way. I encourage us to come up with precise tools and precise definitions. You’ll never hear me say this again, but I’ll say it now: Give the agency more power to do this.” – Dr. Bob Ratner (ADA, Arlington, VA)
“What’s changed is that we now have new technology that generates data 24/7, and the ability to understand what all of us who have been living with the disease have been feeling. And now we can see patterns in data and how they relate to how we are feeling. If we can standardize those definitions, that’s the extraordinary opportunity we have for type 1 and type 2 diabetes… The time is now, we have the capability, we have to work together.” – Ms. Anna McCollister-Slipp (Patient Advocate, Washington, DC)
“No one is talking about throwing A1c out. We are talking about moving to the next level to have better interventions for people with diabetes.” – Dr. Bob Ratner (ADA, Alexandria, VA)
“We are open to all measures beyond A1c that have been suggested so far: glycemic variability, time-in-range, PROs, biomarkers, and others. At present, no consensus exists on the exact measures that patients feel would be most relevant.” – Dr. Francis Kalush (FDA, Silver Spring, MD)

On Quality of Life

“What you hear, resoundingly, is that it’s not just about dying. There’s life every day, and that’s what we’re getting at.” – Dr. Aaron Kowalski (JDRF, New York, NY)
“Some people with diabetes will go through 40, 50, even 60 years of treatment. We need to ensure that this treatment is beneficial and engaging. Clearly, A1c is not it.” – Dr. George Grunberger (Oakland University, Bloomfield Hills, MI)
“Regardless of whether you have type 1 or type 2, dealing with the ups and downs of this disease impacts not only complications, but how you function day to day. Time in range is enormously helpful.” – Ms. Anna McCollister-Slipp (Patient Advocate, Washington, DC)
“Mental health is incredibly under-funded in the US – we’d like more attention there as well as more attention on measures of diabetes distress, and measures of emotional well being. We’d love to see more manufacturers assessing these qualities in a standardized way … determined by the behavioral health experts and other specialists.” – Kelly Close (The diaTribe Foundation, San Francisco CA)
“One third of type 2s not on insulin said diabetes negatively impacts their energy for new challenges and opportunities. Nearly half of type 2s on insulin said diabetes prevents them from being physically able to do what they want. And around half of type 1s worry about money and don’t get enough sleep. How would you feel if if you had no energy for new challenges, couldn’t physically do what you want, worried too much about money, were stressed and anxious all the time, and didn’t get enough sleep?” – Richard Wood (dQ&A, San Francisco, CA)
“How can we make life better, not just longer, for people with diabetes? Now that we’ve done better in terms of glycemic control (and accompanying reductions in heart attack, stroke, amputation, and diabetic kidney disease) we have the opportunity to go to the next level. Quality of life, whether you want to define that in terms of hypoglycemia, weight maintenance, or time-in-range as it relates to functional capacity, these are all the things we are beginning to explore.” – Dr. Bob Ratner (ADA, Alexandria, VA)

On the Heterogeneity of Diabetes

“It’s not unusual for the experts to be surprised about what matters to patients.” – Dr. Robert Califf (FDA, Silver Spring, MD)
“The challenge in diabetes is that there’s not one agreed upon list of symptoms that all patients experience. Coming to one universal measure of quality of life for all trials and interventions might be possible for symptomatic diseases. But diabetes is a heterogeneous, personalized disease.” – Dr. Kristina Boye (Lilly Diabetes, Indianapolis, IN)
“You’d hope patients and physicians would sit together to discuss the goals of individuals. Patients are different, and you will make a different decision – a true joint decision. If it’s just the physicians, then you won’t get the outcomes you desire. I think we need to think about individualization, and I think we should’ve been doing that for a long time.” – Dr. Simon Heller (Sheffield University, UK)
“Most people don’t have insulin-requiring diabetes. What would be the most relevant thing for people that are controlled by non-insulin medications? Hypoglycemia is not really a risk for them. We should explore that.” – Dr. George Grunberger (AACE, Michigan)
“If we make strong attempts to objectify subjective measures, we’re diminishing the patient voice.” – Dr. Shana Traina (Janssen, New Brunswick, NJ)