Individualized Glycemic Goals for Older Adults Are a Moving Target

Treatment of type 2 diabetes (T2D) has several objectives: 1) decreasing risks of microvascular complications (retinopathy, nephropathy, and neuropathy), 2) decreasing risks of cardiovascular complications, and 3) avoiding hypoglycemia and symptomatic hyperglycemia. The American Diabetes Association (ADA) and European Association for the Study of Diabetes advocate a patient-centered approach to achieve these goals (1–3). This general guidance implicitly raises questions. Which patient-related factors should be considered in designing a therapeutic regimen? How should the treatment regimen be modified to take these factors into account? The ADA’s guidelines incorporate an approach to select an individualized hemoglobin A_1c (HbA_1c) target for older adults based on their comorbidities and function, which act as proxies for life expectancy (3). In this issue of Diabetes Care, Le et al. (4) focus on how this specific guidance was implemented in older Americans (aged ≥65 years) between 2001 and 2018. They conclude that “in accordance with the ADA’s recommendation, glycemic targets for older patients have been relaxed.” In other words, as the guidelines transitioned to less aggressive individualized targets, the average HbA_1c for older Americans with T2D receiving diabetes medications increased from 6.9% to 7.2%. Further, older Americans in this cohort achieved similar mean HbA_1c levels regardless of whether they were judged to be in “good,” “intermediate,” or “poor” health.

Le et al. (4) interpret their findings as evidence that clinicians are failing to practice individualized care in accordance with the ADA’s recommendations. This may be partly true, and reports such as that of substantial numbers of older veterans experiencing severe hypoglycemia while receiving hospice care are sobering examples of failures of individualization (5). However, for many older patients, clinicians may reasonably be considering factors for individualization beyond their life expectancy. An important factor listed in guidelines is patient preference. Current evidence suggests that many older adults will be reluctant to accept higher glucose levels and do not want longer-term life expectancy dictating their treatment (6,7). Clinicians may also be concerned that focusing on life expectancy will be perceived as giving up on patients and jeopardize the therapeutic relationship (8). Further, there may be external barriers to achieving an individualized glycemic target. Notably, insurance coverage gaps for Medicare beneficiaries often result in high out-of-pocket costs for diabetes medications, restricting access to newer medication classes that have better safety profiles (9).

Although we strongly advocate for individualized approaches to glycemic targets, the classification system in the ADA guidelines may often be insufficient to support clinical decision-making in this area. Older patients classified as having “complex/intermediate” health are extremely heterogeneous, comprising the majority of the population in the study by Le et al. (4). This category encompasses a 72-year-old woman with hypertension and emphysema who, based on the best available prognostic tools, would have a 10-year mortality as low as 20% and life expectancy of up to 20 years (10,11). It would also include other 72-year-old individuals with more serious comorbidities and life expectancies of <5 years (10,11). Clinicians know that a patient with well-controlled emphysema is not equivalent to one who has daily dyspnea and struggles to leave the home. Nevertheless, each of these patients could have a recommended HbA_1c target of <8.0% according to guidelines (3).

While the data of Le et al. (4) describe how glycemic control has evolved over time, recent therapeutic advances have important implications for pharmacological approaches to treat hyperglycemia. Three important new classes of HbA_1c-lowering drugs were approved between 2005 and 2013: dipeptidyl peptidase 4 (DPP-4) inhibitors, glucagon-like peptide 1 (GLP-1) receptor agonists, and sodium–glucose cotransporter 2 (SGLT2) inhibitors. Drugs from two of these classes (SGLT2 inhibitors and GLP-1 receptor agonists) have been demonstrated to improve clinical outcomes by decreasing risks of major adverse cardiovascular events, heart failure hospitalizations, and/or progression of diabetic kidney disease (12–18). Even in the most recent time period in the analysis of Le et al. (2013–2018) (4), fewer than 5% of patients received SGLT2 inhibitors or GLP-1 receptor agonists. Because of the favorable impact on important clinical outcomes, availability of these drugs has led the ADA and European Association for the Study of Diabetes to revise their guidelines and is likely causing physicians and patients to modify their therapeutic choices in important ways (2).

While the exact optimal glycemic target is not always well defined, published evidence strongly supports the conclusion that aiming for lower levels of HbA_1c can lead to improved clinical outcomes for patients across the age spectrum:

• Landmark clinical studies demonstrate that intensive glycemic control decreases risks for microvascular complications (19–21).

• Although considerable complexity surrounds the question of how intensive glycemic control affects the risk of macrovascular complications, long-term follow-up studies provide strong evidence that, in the fullness of time, lowering HbA_1c decreases the risk of adverse cardiovascular events (22,23).

• Placebo-controlled trials have demonstrated that SGLT2 inhibitors decrease risks of major adverse cardiovascular outcomes, hospitalization for heart failure, and progression of diabetic kidney disease (12–15).

• Placebo-controlled trials have demonstrated that GLP-1 receptor agonists decrease risks of major adverse cardiovascular outcomes, a clinical benefit that may be mediated by improved glycemic control (16–18).

• Older patients derive substantial cardiovascular benefits when treated with SGLT2 inhibitors or GLP-1 receptor agonists, possibly even greater than those for younger patients in the case of SGLT2 inhibitors (12).

Of course, the data on therapeutic benefits must be balanced against data on risks. At least three classes of diabetes drugs have major liabilities. Insulins and insulin secretagogues (including sulfonylureas) can cause life-threatening hypoglycemia and weight gain (2,24). Thiazolidinediones predispose to sodium retention, heart failure, and bone fractures (25). SGLT2 inhibitors increase risks of ketoacidosis, genitourinary infections, and possibly bone fractures (26). GLP-1 receptor agonists have tolerability issues, including nausea and vomiting, and cause weight loss, which may be undesirable in patients with frailty (27).

When developing a therapeutic approach for older patients, clinicians should consider this evidence in the context of a patient’s preferences and broader health needs. Competing medical demands may take priority over improving glycemic control, and the presence of geriatric syndromes may impede safe diabetes medication administration and monitoring (28). Furthermore, the time dimension must be considered. Just as it takes time for diabetes complications to develop, it takes time (at least 3–5 years) for patients to accrue benefit from intensive glycemic control (29). Nevertheless, available data suggest several principles to guide a modern approach to HbA_1c-lowering therapy:

• So long as it can be achieved safely, a therapeutic target HbA_1c <7.0% will provide the best protection from complications of diabetes.

• Patients with life expectancy of less than 3–5 years are unlikely to benefit from tight glycemic control, although prognosis may be difficult to determine. Deintensification of drugs with high risk of hypoglycemia (i.e., insulins and insulin secretagogues) should be strongly considered in patients with limited life expectancy.

• Some coexisting medical conditions (e.g., major adverse cardiovascular disease, congestive heart failure, or diabetic kidney disease) represent indications for using an SGLT2 inhibitor or a GLP-1 receptor agonist, regardless of HbA_1c, if the condition can be addressed by these medications, which can provide meaningful clinical benefit within time periods as short as 3–5 years (12–18).

• Insulin secretagogues (e.g., sulfonylureas) are associated with increased risk of life-threatening hypoglycemia and should be avoided whenever possible. This will become easier to accomplish as generic versions of new diabetes drugs become available (e.g., SGLT2 inhibitors and DPP-4 inhibitors).

• Insulins should be reserved for patients with late-stage T2D with insulin-dependent physiology due to advanced β-cell failure. When insulin is prescribed, it may be appropriate to relax HbA_1c targets (e.g., HbA_1c <8.0% rather than <7.0%).

Overall, the study by Le et al. (4) highlights how approaches to HbA_1c lowering have changed for older Americans with T2D over the past two decades and will continue to change in the future (Fig. 1). Accordingly, clinical guidelines are valuable documents that must evolve to keep pace with therapeutic advances. The availability of three newer classes of HbA_1c-lowering drugs moves us toward an era where patients with T2D can safely achieve lower levels of HbA_1c. As we move into this era, individualization is likely to focus less on HbA_1c level and more on tailoring therapy to patients’ comorbidities and avoiding drugs with unfavorable risk–benefit profiles. More work remains to be done to achieve broader and equitable access to newer drugs and to understand how to use them wisely for older patients across the diversity of their health and function.