Diabetes Care Volume 44, Supplement 1, January 2021 S111

9. Pharmacologic Approaches to American Association Glycemic Treatment: Standards of Medical Care in Diabetesd2021 Diabetes Care 2021;44(Suppl. 1):S111–S124 | https://doi.org/10.2337/dc21-S009 .PAMCLGCAPOCE OGYEI TREATMENT GLYCEMIC TO APPROACHES PHARMACOLOGIC 9.

The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes” includes the ADA’s current clinical practice recommendations and is intended to provide the components of diabetes care, general treatment goals and guidelines, and tools to evaluate quality of care. Members of the ADA Professional Practice Committee, a multidisciplinary expert committee (https://doi.org/10.2337/dc21- SPPC), are responsible for updating the Standards of Care annually, or more frequently as warranted. For a detailed description of ADA standards, statements, and reports, as well as the evidence-grading system for ADA’s clinical practice recommendations, please refer to the Standards of Care Introduction (https://doi .org/10.2337/dc21-SINT). Readers who wish to comment on the Standards of Care are invited to do so at professional.diabetes.org/SOC.

PHARMACOLOGIC THERAPY FOR

Recommendations 9.1 Most people with type 1 diabetes should be treated with multiple daily injections of prandial and basal , or continuous subcutaneous insulin infusion. A 9.2 Most individuals with type 1 diabetes should use rapid-acting insulin analogs to reduce risk. A 9.3 Patients with type 1 diabetes should receive education on how to match prandial insulin doses to intake, premeal , and anticipated physical activity. C

Insulin Therapy Because the hallmark of type 1 diabetes is absent or near-absent b- function, insulin treatment is essential for individuals with type 1 diabetes. In addition to , insulinopenia can contribute to other metabolic disturbances like and ketoacidosis as well as tissue catabolism that can be life threatening. Severe metabolic decompensation can be, and was, mostly prevented Suggested citation: American Diabetes Association. with once or twice daily injections for the six or seven decades after the discovery of 9. Pharmacologic approaches to glycemic treatment: insulin. However, over the past three decades, evidence has accumulated supporting Standards of Medical Care in Diabetesd2021. more intensive insulin replacement, using multiple daily injections of insulin or Diabetes Care 2021;44(Suppl. 1):S111–S124 continuous subcutaneous administration through an , as providing the © 2020 by the American Diabetes Association. best combination of effectiveness and safety for people with type 1 diabetes. The Readers may use this article as long as the work is properly cited, the use is educational and not for Diabetes Control and Complications Trial (DCCT) demonstrated that intensive therapy profit, and the work is not altered. More infor- with multiple daily injections or continuous subcutaneous insulin infusion (CSII) mation is available at https://www.diabetesjournals reduced A1C and was associated with improved long-term outcomes (1–3). The study .org/content/license. S112 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

was carried out with short-acting (regu- effort made to reach the patient’s gly- See Section 7 “Diabetes Technology” lar) and intermediate-acting (NPH) hu- cemic targets. (https://doi.org/10.2337/dc21-S007) for a man . In this landmark trial, Most studies comparing multiple daily full discussion of insulin delivery devices. lower A1C with intensive control (7%) injections with CSII have been relatively In general, patients with type 1 di- ledto;50%reductionsinmicrovascular small and of short duration. However, a abetes require 50% of their daily insulin complications over 6 years of treatment. recent systematic review and meta- as basal and 50% as prandial. Total daily However, intensive therapy was asso- analysis concluded that pump therapy insulin requirements can be estimated ciated with a higher rate of severe has modest advantages for lowering A1C based on weight, with typical doses hypoglycemia than conventional treat- (20.30% [95% CI 20.58 to 20.02]) and ranging from 0.4 to 1.0 units/kg/day. ment (62 compared with 19 episodes for reducing severe hypoglycemia rates Higher amounts are required during pu- per 100 patient-years of therapy). Follow- in children and adults (12). However, berty, , and medical illness. up of subjects from the DCCT more than there is no consensus to guide the choice The American Diabetes Association/ 10 years after the active treatment com- of injection or pump therapy in a given JDRF Type 1 Diabetes Sourcebook notes ponent of the study demonstrated less patient, and research to guide this 0.5 units/kg/day as a typical starting dose macrovascular as well as less microvas- decision-making isneeded(13).Thearrival in patients with type 1 diabetes who are cular complications in the group that ofcontinuousglucosemonitors toclinical metabolically stable, with half adminis- received intensive treatment (2,4). practice has proven beneficial in specific tered as prandial insulin given to control Over the last 25 years, rapid-acting and circumstances. Reduction of nocturnal blood glucose after meals and the other long-acting insulin analogs have been hypoglycemia in people with type 1 di- half as basal insulin to control glycemia developed that have distinct pharma- abetes using insulin pumps with glucose in the periods between meal absorption cokinetics compared with recombinant sensors is improved by automatic sus- (20); this guideline provides detailed human insulins: basal insulin analogs have pension of insulin delivery at a preset information on intensification of ther- longer duration of action with flatter, more glucose level (13–15). When choosing apy to meet individualized needs. In constant plasma concentrations and activ- among insulin delivery systems, patient addition, the American Diabetes Asso- ity profiles than NPH insulin; rapid-acting preferences, cost, insulin type and dosing ciation position statement “Type 1 Di- analogs (RAA) have a quicker onset and regimen, and self-management capabil- abetes Management Through the Life peak and shorter duration of action than ities should be considered (See Section Span” provides a thorough overview of regular human insulin. In people with 7 “Diabetes Technology,” https://doi type 1 diabetes treatment (21). type 1 diabetes, treatment with analog .org/10.2337/dc21-S007). Typical multidose regimens for pa- insulins is associated with less hypogly- TheU.S. and Drug Administration tients with type 1 diabetes combine cemia and as well as lower (FDA) has now approved two hybrid premeal use of shorter-acting insulins A1C compared with human insulins (5–7). closed-loop pump systems. The safety with a longer-acting formulation, usually More recently, two new injectable insulin and efficacy of hybrid closed-loop sys- at night. The long-acting basal dose is formulations with enhanced rapid action temshas been supported in the literature titrated to regulate overnight, profiles have been introduced. Inhaled in adolescents and adults with type 1 glucose. Postprandial glucose excursions human insulin has a rapid peak and short- diabetes (16,17), and recent evidence are best controlled by a well-timed in- ened duration of action compared with suggests that a closed-loop system is jection of prandial insulin.Theoptimal time RAA and may less hypoglycemia and superior to sensor-augmented pump to administer prandial insulin varies, weight gain (8), and faster-acting insulin therapy for glycemic control and reduc- based on the of the aspart and -aabc may reduce tion of hypoglycemia over 3 months of formulation (regular, RAA, inhaled), the prandial excursions better than RAA comparison in children and adults with premeal blood glucose level, and carbo- (9,9a,9b); further investigation is needed type 1 diabetes (18). In the International hydrate consumption. Recommendations to establish a clear place for these agents Diabetes Closed Loop (iDCL) trial, a for prandial insulin dose administration in . In addition, new 6-month trial in patients with type 1 should therefore be individualized. Phys- longer-acting basal analogs (U-300 glargine diabetes at least 14 years of age, the iologic insulin secretion varies with glyce- or degludec) may confer a lower hypogly- use of a closed-loop system was associ- mia, meal size, and tissue demands for cemiariskcomparedwithU-100glargine ated with a greater percentage of time glucose. To approach this variability in in patients with type 1 diabetes (10,11). spent in the target glycemic range, re- people using insulin treatment, strategies Despite the advantages of insulin analogs duced mean glucose and A1C levels, and have evolved to adjust prandial doses in patients with type 1 diabetes, for some lower percentage of time spent in hypo- based on predicted needs. Thus, edu- patients the expense and/or intensity of glycemia compared with use of a sensor- cation of patients on how to adjust treatment required for their use is pro- augmented pump (19). prandial insulin to account for carbohy- hibitive. There are multiple approaches Intensive insulin management using a drate intake, premeal glucose levels, to insulin treatment, and the central version of CSII and continuous glucose and anticipated activity can be effective precept in the management of type 1 should be considered in most and should be offered to most patients diabetes is that some form of insulin be patients. Automated insulin delivery sys- (22,23). For individuals in whom carbo- giveninaplannedregimentailoredtothe tems may be considered in adults with hydrate counting is effective, estimates individual patient to keep them safe and type 1 diabetes who have the skills to use of the and content of meals out of and to avoid them in order to improve time in range can be incorporated into their prandial significant hypoglycemia, with every and reduce A1C and hypoglycemia (19). dosing for added benefit(24). care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S113

Insulin Injection Technique of this therapy and, as such, holds the PHARMACOLOGIC THERAPY FOR Ensuring that patients and/or caregivers potential for improved clinical outcomes. understand correct insulin injection tech- Recommendations Noninsulin Treatments for Type 1 nique is important to optimize glucose 9.4 is the preferred ini- Diabetes control and insulin use safety. Thus, it is tial pharmacologic agent for the Injectable and oral glucose-lowering important that insulin be delivered into treatment of type 2 diabetes. A drugs have been studied for their efficacy the proper tissue in the correct way. 9.5 Once initiated, metformin should as adjuncts to insulin treatment of type 1 Recommendations have been pub- be continued as long as it is tol- diabetes. is based on the lished elsewhere outlining best practi- erated and not contraindicated; naturally occurring b-cell ces for insulin injection (25). Proper other agents, including insulin, and is approved for use in adults with insulin injection technique includes in- should be added to metformin. A type 1 diabetes. Results from random- jecting into appropriate body areas, in- 9.6 Early combination therapy can ized controlled studies show variable jection site rotation, appropriate care of be considered in some patients reductions of A1C (0–0.3%) and body injection sites to avoid or other at treatment initiation to extend weight (1–2 kg) with addition of pramlin- complications, and avoidance of intra- the time to treatment failure. A tide to insulin (27,28). Similarly, results muscular (IM) insulin delivery. 9.7 The early introduction of insulin have been reported for several agents Exogenously delivered insulin should should be considered if there is currently approved only for the treat- be injected into subcutaneous tissue, not evidence of ongoing catabolism ment of type 2 diabetes. The addition of intramuscularly. Recommended sites for (), if symptoms of metformin in adults with type 1 diabetes insulin injection include the abdomen, hyperglycemia are present, or caused small reductions in body weight thigh, buttock, and upper arm. Because when A1C levels (.10% [86 and levels but did not improve A1C insulin absorption from IM sites differs mmol/mol]) or blood glucose (29,30). The addition of the -like according to the activity of the muscle, levels ($300 mg/dL [16.7 mmol/L]) peptide 1 (GLP-1) (RA) inadvertent IM injection can lead to un- are very high. E or to insulin therapy predictable insulin absorption and vari- 9.8 A patient-centered approach caused small (0.2%) reductions in A1C able effects on glucose, with IM injection should be used to guide the compared with insulin alone in people being associated with frequent and un- choice of pharmacologic agents. with type 1 diabetes and also reduced explained hypoglycemia in several re- Considerations include effect body weight by ;3 kg (31). Similarly, the ports. Risk for IM insulin delivery is on cardiovascular and renal co- addition of a sodium–glucose cotrans- increased in younger, leaner patients morbidities, efficacy, hypogly- porter 2 (SGLT2) inhibitor to insulin therapy when injecting into the limbs rather than cemia risk, impact on weight, has been associated with improvements truncalsites(abdomenandbuttocks)and cost, risk for side effects, and in A1C and body weight when compared when using longer needles. Recent ev- patient preferences (Table 9.1 with insulin alone (32,33); however, SGLT2 idence supports the use of short needles and Fig. 9.1). E inhibitor use in type 1 diabetes is asso- (e.g., 4-mm pen needles) as effective and 9.9 Among patients with type 2 di- ciated with a two- to fourfold increase in well tolerated when compared with lon- abetes who have established ketoacidosis. The risks and benefits of ger needles, including a study performed atherosclerotic cardiovascular adjunctive agents continue to be evalu- in adults with (26). or indicators of high risk, ated, but onlypramlintide isapproved for Injection site rotation is additionally established disease, or treatment of type 1 diabetes. necessary to avoid lipohypertrophy, an failure, a sodium–glucose accumulation of subcutaneous fat in re- 2 inhibitor or SURGICAL TREATMENT FOR TYPE sponse to the adipogenic actions of in- glucagon-like peptide 1 receptor 1 DIABETES sulin at a site of multiple injections. agonist with demonstrated car- Lipohypertrophy appears as soft, smooth and Islet Transplantation diovascular disease benefit raised areas several centimeters in breadth Successful pancreas and islet transplan- (Table 9.1, Table 10.3B, Table and can contribute to erratic insulin tation can normalize glucose levels and 10.3C) is recommended as part absorption, increased glycemic variabil- mitigate microvascular complications of of the glucose-lowering regi- ity, and unexplained hypoglycemic epi- type 1 diabetes. However, patients re- men independent of A1C and sodes. Patients and/or caregivers should ceiving these treatments require lifelong in consideration of patient-spe- receive education about proper injection immunosuppression to prevent graft re- cificfactors(Fig. 9.1 and Section site rotation and to recognize and avoid jection and/or recurrence of autoimmune 10). A areas of lipohypertrophy. As noted in islet destruction. Given the potential 9.10 In patients with type 2 diabetes, Table 4.1, examination of insulin injec- adverse effects of immunosuppressive a glucagon-like peptide 1 recep- tion sites for the presence of lipohyper- therapy, should tor agonist is preferred to in- trophy, as well as assessment of injection be reserved for patients with type 1 dia- sulin when possible. A device use and injection technique, are betes undergoing simultaneous renal trans- 9.11 Recommendation for treatment key components of a comprehensive plantation, following renal transplantation, intensification for patients not diabetes medical evaluation and treat- or for those with recurrent ketoacidosis meeting treatment goals should ment plan. Proper insulin injection tech- or severe hypoglycemia despite intensive not be delayed. A nique may lead to more effective use glycemic management (34). S114 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

Care” (https://doi.org/10.2337/dc21-S011) consider a drug from another class de- 9.12 The regimen and have recommendations for the use of picted in Fig. 9.1. When A1C is $1.5% medication-taking behavior glucose-lowering drugs in the manage- (12.5 mmol/mol) above the glycemic should be reevaluated at regular ment of cardiovascular and renal disease, target (see Section 6 “Glycemic Targets,” intervals (every 3–6months)and respectively. https://doi.org/10.2337/dc21-S006, adjusted as needed to incorpo- for appropriate targets), many patients rate specific factors that impact will require dual combination therapy choice of treatment (Fig. 4.1 and Initial Therapy to achieve their target A1C level (42). Table 9.1). E Metformin should be started at the time Insulin has the advantage of being effec- 9.13 Clinicians should be aware of type 2 diabetes is diagnosed unless there tive where other agents are not and the potential for overbasaliza- are contraindications; for many patients should be considered as part of any tion with insulin therapy. Clinical this will be monotherapy in combination combination regimen when hyperglyce- signals that may prompt evalu- with lifestyle modifications. Additional mia is severe, especially if catabolic fea- ation of overbasalization include and/or alternative agents may be con- tures (weight loss, hypertriglyceridemia, basal dose more than ;0.5 IU/ sidered in special circumstances, such ketosis) arepresent.Itis commonpractice kg, high bedtime-morning or as in individuals with established or in- to initiate insulin therapy for patients who post-preprandial glucose differ- creased risk of cardiovascular or renal present with blood glucose levels $300 ential, hypoglycemia (aware or complications (see Section 10 “Cardio- mg/dL (16.7 mmol/L) or A1C .10% (86 unaware), and high variability. vascular Disease and Risk Management,” mmol/mol)orifthepatienthassymptoms Indication of overbasalization https://doi.org/10.2337/dc21-S010, and of hyperglycemia (i.e., or poly- should prompt reevaluation to Fig. 9.1). Metformin is effective and safe, dipsia) or evidence of catabolism (weight further individualize therapy. E is inexpensive, and may reduce risk of cardiovascular events and (37). loss) (Fig.9.2).Asglucosetoxicityresolves, simplifying the regimen and/or changing The American Diabetes Association/ Metformin is available in an immediate- to oral agents is often possible. However, European Association for the Study of release form for twice-daily dosing or as there is evidence that patients with un- Diabetes consensus report “Management an extended-release form that can be controlled hyperglycemia associated with of Hyperglycemia in Type 2 Diabetes, given once daily. Compared with sulfo- type 2 diabetes can also be effectively 2018” and the 2019 update (35,36) nylureas, metformin as first-line therapy treated with a (43). recommend a patient-centered approach has beneficial effects on A1C, weight, and to choosing appropriate pharmacologic cardiovascular mortality (38); there is Combination Therapy treatment of blood glucose. This includes little systematic data available for other Because type 2 diabetes is a progressive consideration of efficacy and key patient oral agents as initial therapy of type 2 disease in many patients, maintenance of factors: 1) important such diabetes. glycemic targets with monotherapy is as atherosclerotic The principal side effects of metformin often possible for only a few years, after (ASCVD) and indicators of high ASCVD are gastrointestinal intolerance due to which combination therapy is necessary. risk, chronic (CKD), and bloating, abdominal discomfort, and di- Current recommendations have been to (see Section 10 “Cardiovas- arrhea; these can be mitigated by gradual use stepwise addition of to cular Disease and Risk Management,” dose titration. The drug is cleared by metformintomaintain A1Cat target. This https://doi.org/10.2337/dc21-S010, and renal filtration, and very high circulating allows a clearer assessment of the pos- Section 11 “Microvascular Complications levels (e.g., as a result of overdose or itive and negative effects of new drugs and Foot Care,” https://doi.org/10.2337/ acute renal failure) have been associated and reduces patient risk and expense dc21-S011), 2) hypoglycemia risk, 3)ef- with lactic . However, the occur- (44); based on these factors, sequential fects on body weight, 4)sideeffects,5) renceofthiscomplicationisnowknownto addition of oral agents to metformin has cost, and 6) patient preferences. Lifestyle bevery rare, and metformin may besafely been the standard of care. However, modifications that improve health (see used in patients with reduced estimated there are data to support initial combi- Section 5 “Facilitating Behavior Change glomerular filtration rates (eGFR); the nation therapy for more rapid attain- and Well-being to Improve Health FDA has revised the label for metformin ment of glycemic goals (45,46) and later Outcomes,” https://doi.org/10.2337/ to reflect its safety in patients with combination therapy for longer durabil- dc21-S005) should be emphasized along eGFR $30 mL/min/1.73 m2 (39). A ran- ity of glycemic effect (47). The VERIFY with any pharmacologic therapy. Section 12 domized trial confirmed previous obser- ( Efficacy in combination with “Older Adults” (https://doi.org/10.2337/ vations that metformin use is associated metfoRmIn For earlY treatment of type 2 dc21-S012) and Section 13 “Children and with deficiency and wors- diabetes) trial demonstrated that initial Adolescents” (https://doi.org/10.2337/ ening of symptoms of neuropathy (40). combination therapy is superior to se- dc21-S013) have recommendations spe- This is compatible with a report from the quential addition of medications for ex- cific for older adults and for children Diabetes Prevention Program Outcomes tending primary and secondary failure and adolescents with type 2 diabetes, Study (DPPOS) suggesting periodic testing (48). In the VERIFY trial, participants respectively. Section 10 “Cardiovascular of vitamin B12 (41). receiving the initial combination of met- Disease and Risk Management” (https:// In patients with contraindications or formin and the dipeptidyl peptidase doi.org/10.2337/dc21-S010) and Section intolerance to metformin, initial therapy 4 (DPP-4) inhibitor vildagliptin had 11 “Microvascular Complications and Foot should be based on patient factors; a slower decline of glycemic control aedaeejunl.r hraooi prahst lcmcTetetS115 Treatment Glycemic to Approaches Pharmacologic care.diabetesjournals.org Table 9.1—Drug-specific and patient factors to consider when selecting antihyperglycemic treatment in adults with type 2 diabetes

ASCVD, atherosclerotic cardiovascular disease; CV, cardiovascular; CVOT, cardiovascular outcomes trial; DPP-4, dipeptidyl peptidase 4; DKA, diabetic ketoacidosis; DKD, diabetic kidney disease; eGFR, estimated glomerular filtration rate; GI, gastrointestinal; GLP-1 RAs, glucagon-like peptide 1 receptor ; HF, heart failure; NASH, nonalcoholic ; SGLT2, sodium–glucose cotransporter 2; SQ, subcutaneous; T2D, type 2 diabetes. *For agent-specific dosing recommendations, please refer to the manufacturers’ prescribing information. †FDA-approved for cardiovascular disease benefit. ‡FDA-approved for heart failure indication. §FDA-approved for indication. 16PamclgcApoce oGyei Treatment Glycemic to Approaches Pharmacologic S116 ibtsCare Diabetes oue4,Splmn ,Jnay2021 January 1, Supplement 44, Volume

Figure 9.1—Glucose-lowering medication in type 2 diabetes: 2021 ADA Professional Practice Committee (PPC) adaptation of Davies et al. (35) and Buse et al. (36). For appropriate context, see Fig. 4.1. The 2021 ADA PPC adaptation of the Fig. 9.1 “Indicators of high-risk or established ASCVD, CKD, or HF” pathway has been adapted based on trial populations studied. ASCVD, atherosclerotic cardiovascular disease; CKD, chronic kidney disease; CVD, cardiovascular disease; CVOTs, cardiovascular outcomes trials; DPP-4i, dipeptidyl peptidase 4 inhibitor; eGFR, estimated glomerular filtration rate; GLP-1 RA, glucagon-like peptide 1 receptor agonist; HF,heartfailure; HFrEF,heartfailurewithreducedejection fraction; LVEF,leftventricular ejection fraction; LVH, left ventricularhypertrophy; SGLT2i, sodium–glucose cotransporter2 inhibitor; SU, sulfonylurea;T2D, type 2 diabetes; TZD, . care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S117

Figure9.2—Intensifyingtoinjectabletherapies.DSMES,diabetesself-managementeducationandsupport;FPG,fastingplasmaglucose;FRC,fixed-ratio combination; GLP-1 RA, glucagon-like peptide 1 receptor agonist; max, maximum; PPG, postprandial glucose. Adapted from Davies et al. (35). S118 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

compared with metformin alone and consideration of patient-specific factors Cost for diabetes has in- with vildagliptin added sequentially to (Fig. 9.1). For patients without estab- creased dramatically over the past two metformin. These results have not lished ASCVD, indicators of high ASCVD decades, and an increasing proportion been generalized to oral agents other risk, heart failure, or CKD, the choice of is now passed on to patients and their than vildagliptin, but they suggest that a second agent to add to metformin is families (61). Table 9.2 provides cost more intensive early treatment has some not yet guided by empiric evidence com- information for currently approved benefits and should be considered through paring across multiple classes. Rather, drug noninsulin therapies. Of note, prices a shared decision-making process with choice is based on efficacy, avoidance listed are average wholesale prices patients, as appropriate. Moreover, since of side effects (particularly hypoglycemia (AWP) (62) and National Average Drug the absolute effectiveness of most oral and weight gain), cost, and patient pref- Acquisition Costs (NADAC) (63), separate medications rarely exceeds 1%, initial com- erences (51). Similar considerations are measures to allow for a comparison of bination therapy should be considered in applied in patients who require a third drug prices, but do not account for patients presenting with A1C levels 1.5– agent to achieve glycemic goals. A recent discounts, rebates, or other price adjust- 2.0% above target. systematic review and network meta- ments often involved in prescription Recommendations for treatment in- analysis suggests greatest reductions in sales that affect the actual cost incurred tensification for patients not meeting A1C level with insulin regimens and specific by the patient. Medication costs can be a treatment goals should not be delayed. GLP-1 RAs added to metformin-based back- major source of for patients with Shared decision-making is important in ground therapy (52). In all cases, treatment diabetes and contribute to worse adher- discussions regarding treatment intensi- regimens need to be continuously reviewed ence to medications (64); cost-reducing fication. The choice of medication added for efficacy, side effects, and patient burden strategies may improve in to metformin is based on the clinical (Table 9.1). In some instances, patients will some cases (65). characteristics of the patient and their require medication reduction or discontin- preferences. Important clinical charac- uation. Common reasons for this include Cardiovascular Outcomes Trials teristics include the presence of estab- ineffectiveness, intolerablesideeffects,ex- There are now multiple large randomized lished ASCVD or indicators of high ASCVD pense, or a change in glycemic goals (e.g., in controlled trials reporting statistically risk, heart failure, CKD, other comorbid- response to development of comorbidities significant reductions in cardiovascular ities, and risk for specific adverse drug or changes in treatment goals). Section 12 events in patients with type 2 diabetes effects, as well as safety, tolerability, and “Older Adults” (https://doi.org/10.2337/ treated with an SGLT2 inhibitor (- cost. Although there are numerous trials dc21-S012) has a full discussion of treat- gliflozin, canagliflozin, dapagliflozin) or comparing dual therapy with metformin ment considerations in older adults, in GLP-1 RA (liraglutide, , dula- alone, there is little evidence to support whom changes of glycemic goals and glutide); see Section 10 “Cardiovascular one combination over another. A com- de-escalation of therapy are common. Disease and Risk Management” (https:// parative effectiveness meta-analysis The need for the greater potency of doi.org/10.2337/dc21-S010) for details. suggests that each new class of non- injectable medications is common, par- The subjects enrolled in the cardiovas- insulin agents added to initial therapy ticularly in people with a longer duration cular outcomes trials using empagliflozin, with metformin generally lowers A1C of diabetes. The addition of basal insulin, canagliflozin, dapagliflozin, liraglutide, and approximately 0.7–1.0% (49,50). If the either human NPH or one of the long- semaglutide had A1C $6.5%, and more A1C target is not achieved after approx- acting insulin analogs, to oral agent regi- than 70% were taking metformin at base- imately 3 months, metformin can be mens is a well-established approach that line. Thus, a practical extension of these combined with any one of the preferred is effective for many patients. In addition, results to clinical practice is to use these six treatment options: sulfonylurea, thia- recent evidence supports the utility of drugs preferentially in patients with zolidinedione, DPP-4 inhibitor, SGLT2 in- GLP-1 RAs in patients not at glycemic goal. type 2 diabetes and established ASCVD hibitor, GLP-1 RA, or basal insulin; the While most GLP-1 RAs are injectable, an or indicators of high ASCVD risk. For these choice of which agent to add is based on oral formulation of semaglutide is now com- patients, incorporating one of the SGLT2 drug-specific effects and patient factors mercially available (53). In trials comparing inhibitors or GLP-1 RAs that have been (Fig. 9.1 and Table 9.1). the addition of an injectable GLP-1 RA or demonstrated to have cardiovascular dis- For patients with established ASCVD insulin in patients needing further glu- ease benefit is recommended (Table 9.1). or indicators of high ASCVD risk (such as cose lowering, glycemic efficacy of inject- In cardiovascular outcomes trials, em- patients $55 years of age with coronary, able GLP-1 RA was similar or greater than pagliflozin, canagliflozin, dapagliflozin, carotid, or lower-extremity steno- that of basal insulin (54–60). GLP-1 RAs in liraglutide, semaglutide, and sis .50% or left ventricular hypertro- these trials had a lower risk of hypogly- all had beneficial effects on indices of phy), heart failure, or CKD, an SGLT2 cemia and beneficial effects on body CKD, while dedicated renal outcomes inhibitor or GLP-1 RA with demonstrated weight compared with insulin, albeit studies have demonstrated benefitof CVD benefit(Table 9.1, Table 10.3B, with greater gastrointestinal side effects. specific SGLT2 inhibitors. See Section Table 10.3C,andSection10“Cardiovas- Thus, trial results support GLP-1 RAs as the 11 “Microvascular Complications and cular Disease and Risk Management,” preferred option for patients requiring Foot Care” (https://doi.org/10.2337/ https://doi.org/10.2337/dc21-S010) is the potency of an injectable therapy for dc21-S011) for discussion of how CKD recommended as part of the glucose- glucose control (Fig. 9.2). However, high may impact treatment choices. Addi- lowering regimen independent of A1C, costs and tolerability issues are impor- tional large randomized trials of other independent of metformin use, and in tant barriers to GLP-1 RA use. agents in these classes are ongoing. care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S119

Table 9.2—Median monthly (30-day) AWP and NADAC of maximum approved daily dose of noninsulin glucose-lowering agents in the U.S. Dosage strength/product Median AWP Median NADAC Maximum approved Class Compound(s) (if applicable) (min, max)† (min, max)† daily dose* c Metformin 850 mg (IR) $108 ($6, $109) $3 2,550 mg 1,000 mg (IR) $87 ($4, $88) $2 2,000 mg 1,000 mg (ER) $242 ($242, $188 ($188, $572) 2,000 mg $7,214) (2nd c 4 mg $74 ($71, $198) $4 8 mg generation) c 10 mg (IR) $75 ($67, $97) $5 40 mg (IR) 10 mg (XL) $48 $11 20 mg (XL) c Glyburide 6 mg (micronized) $52 ($48, $71) $10 12 mg (micronized) 5 mg $93 ($63, $103) $11 20 mg c 45 mg $348 ($283, $349) $5 45 mg c 4 mg $407 $330 8 mg a-Glucosidase inhibitors c 100 mg $106 ($104, $106) $28 300 mg c 100 mg $241 $311 300 mg (glinides) c 120 mg $155 $31 360 mg c 2 mg $878 ($162, $897) $38 16 mg DPP-4 inhibitors c 25 mg $234 $175 25 mg c 5 mg $530 $424 5 mg c 5 mg $555 $444 5 mg c 100 mg $568 $456 100 mg SGLT2 inhibitors c Ertugliflozin 15 mg $354 $284 15 mg c Dapagliflozin 10 mg $621 $496 10 mg c Empagliflozin 25 mg $627 $501 25 mg c Canagliflozin 300 mg $622 $499 300 mg GLP-1 RAs c Exenatide (extended 2 mg powder for $882 $706 2 mg** release) suspension or pen c Exenatide 10 mg pen $752 $720 20 mg c Dulaglutide 4.5/0.5 mL pen $957 $766 4.5 mg** c Semaglutide 1 mg pen $973 $779 1 mg** 14 mg () $927 $738 14 mg c Liraglutide 18 mg/3 mL pen $1,161 $930 1.8 mg c 300 mg/3 mL pen $774 N/A 20 mg Bile acid sequestrant c 625 mg tabs $710 ($674, $712) $105 3.75 g 3.75 g suspension $804 $318 3.75 g Dopamine-2 agonist c 0.8 mg $960 $772 4.8 mg Amylin mimetic c Pramlintide 120 mg pen $2702 $2,097 120 mg/injection†† AWP, average wholesale price; DPP-4, dipeptidyl peptidase 4; ER and XL, extended release; GLP-1 RA, glucagon-like peptide 1 receptor agonist; IR, immediate release; max, maximum; min, minimum; N/A, data not available; NADAC, National Average Drug Acquisition Cost; SGLT2, sodium–glucose cotransporter 2. †Calculated for 30-day supply (AWP [62] or NADAC [63] unit price 3 number of doses required to provide maximum approved daily dose 3 30 days); median AWP or NADAC listed alone when only one product and/or price. *Utilized to calculate median AWP and NADAC (min, max); generic prices used, if available commercially. **Administered once weekly. ††AWP and NADAC calculated based on 120 mg three times daily.

Insulin Therapy management is beneficial. For example, individualized titration over days to weeks Many patients with type 2 diabetes instructionofpatientsinself-titrationof as needed. The principal action of basal eventually require and benefit from in- insulin doses based on glucose monitoring insulinistorestrainhepaticglucosepro- sulin therapy (Fig. 9.2). See the section improves glycemic control in patients with duction and limit hyperglycemia overnight INSULIN INJECTION TECHNIQUE, above, for guid- type 2 diabetes initiating insulin (66). Com- and between meals (67,68). Control of ance on how to administer insulin safely prehensive education regarding self-mon- fasting glucose can be achieved with hu- and effectively. The progressive nature of itoring of blood glucose, , and the man NPH insulin or a long-acting insulin type 2 diabetes should be regularly and avoidance and appropriate treatment of analog. In clinical trials, long-acting basal objectively explained to patients, and hypoglycemia are critically important in analogs (U-100 glargine or detemir) have providers should avoid using insulin any patient using insulin. been demonstrated to reduce the risk of as a threat or describing it as a sign of Basal Insulin symptomatic and nocturnal hypoglycemia personal failure or punishment. Rather, Basal insulinalone isthe most convenient compared with NPH insulin (69–74), al- the utility and importance of insulin to initial insulin regimen and can be added though these advantages are modest and maintain glycemic control once progres- to metformin and other oral agents. may not persist (75). Longer-acting basal sion of the disease overcomes the effect Starting doses can be estimated based analogs (U-300 glargine or degludec) may of other agents should be emphasized. on body weight (0.1–0.2 units/kg/day) convey a lower hypoglycemia risk com- Educating and involving patients in insulin and the degree of hyperglycemia, with pared with U-100 glargine when used in S120 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

Table 9.3—Median cost of insulin products in the U.S. calculated as AWP (62) and NADAC (63) per 1,000 units of specified dosage form/product Median AWP Median Insulins Compounds Dosage form/product (min, max)* NADAC* Rapid-acting c Lispro follow-on product U-100 vial $157 $125 U-100 prefilled pen $202 $161 c Lispro U-100 vial $165† $132† U-100 cartridges $408 $326 U-100 prefilled pen $212† $170† U-200 prefilled pen $424 $339 c Lispro-aabc U-100 vial $330 N/A U-100 prefilled pen $424 N/A U-200 prefilled pen $424 N/A c Glulisine U-100 vial $341 $272 U-100 prefilled pen $439 $350 c Aspart U-100 vial $174† $139† U-100 cartridges $215 $344 U-100 prefilled pen $223† $179† c Aspart (“faster acting U-100 vial $347 $278 product”) U-100 cartridge $430 N/A U-100 prefilled pen $447 $356 c Inhaled insulin Inhalation cartridges $924 $606 Short-acting c human regular U-100 vial $165†† $133†† Intermediate-acting c human NPH U-100 vial $165†† $133†† U-100 prefilled pen $208 $167 Concentrated human regular c U-500 human regular U-500 vial $178 $143 insulin insulin U-500 prefilled pen $229 $183 Long-acting c Glargine follow-on product U-100 prefilled pen $190 (118, 261) $210 U-100 vial $190 (118, 261) N/A c Glargine U-100 vial; U-100 prefilled pen $340 $272 U-300 prefilled pen $340 $272 c Detemir U-100 vial; U-100 prefilled pen $370 $296 c Degludec U-100 vial; U-100 prefilled pen; U-200 $407 $325 prefilled pen Premixed insulin products c NPH/regular 70/30 U-100 vial $165†† $133†† U-100 prefilled pen $208 $167 c Lispro 50/50 U-100 vial $342 $273 U-100 prefilled pen $424 $338 c Lispro 75/25 U-100 vial $342 $274 U-100 prefilled pen $212 $340 c Aspart 70/30 U-100 vial $180 $144 U-100 prefilled pen $224 $179 Premixed insulin/GLP-1 RA c Glargine/Lixisenatide 100/33 prefilled pen $589 $471 products c Degludec/Liraglutide 100/3.6 prefilled pen $874 $701 AWP, average wholesale price; GLP-1 RA, glucagon-like peptide 1 receptor agonist; N/A, not available; NADAC, National Average Drug Acquisition Cost. *AWP or NADAC calculated as in Table 9.2. †Generic prices used when available. ††AWP and NADAC data presented do not include vials of regular human insulin and NPH available at Walmart for approximately $25/vial; median listed alone when only one product and/or price.

combination with oral agents (76–82). Indication of overbasalization should prompt (e.g., individuals with relaxed A1C goals, Despite evidence for reduced hypogly- reevaluation to further individualize ther- low rates of hypoglycemia, and prom- cemia with newer, longer-acting basal apy (84). inent , as well as those insulin analogs in settings, in The cost of insulin has been rising with cost concerns), human insulin (NPH practice these effects may be modest steadily over the past two decades, at a and regular) may be the appropriate compared with NPH insulin (83). Clini- pace several fold that of other medical choice of therapy, and clinicians should cians should be aware of the potential for expenditures (85). This expense contrib- be familiar with its use (83). Human regular overbasalization with insulin therapy. utes significant burden to patients as insulin, NPH, and 70/30 NPH/regular prod- Clinical signals that may prompt evalu- insulin has become a growing “out-of- ucts can be purchased for considerably ation of overbasalization include basal pocket” cost for people with diabetes, less than the AWP and NADAC prices dosegreaterthan;0.5IU/kg,highbedtime- and direct patient costs contribute to listed in Table 9.3 at select pharmacies. morning or post-preprandial glucose differ- treatment nonadherence (85). Therefore, ential (e.g. bedtime-morning glucose dif- consideration of cost is an important Prandial Insulin ferential $50 mg/dL), hypoglycemia component of effective management. Many individuals with type 2 diabetes (aware or unaware), and high variability. For many patients with type 2 diabetes require doses of insulin before meals, in care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S121

addition to basal insulin, to reach glyce- evidence that compared with injectable 30) formulations, are less costly alterna- mic targets. A dose of 4 units or 10% of rapid-acting insulin, supplemental doses tives to insulin analogs. Figure 9.2 outlines the amount of basal insulin at the largest of inhaled insulin taken based on post- these options as well as recommendations meal or the meal with the greatest post- prandial glucose levels may improve for further intensification, if needed, to prandial excursion is a safe estimate for blood glucose management without achieve glycemic goals. When initiating initiating therapy. The prandial insulin additional hypoglycemia or weight gain combination injectable therapy, metfor- regimen can then be intensified based on (92), although results from a larger study min therapy should be maintained, while patient needs (see Fig. 9.2). People with are needed for confirmation. Inhaled sulfonylureas and DPP-4 inhibitors are type 2 diabetes are generally more in- insulin is contraindicated in patients with typically weaned or discontinued. In pa- sulin resistant than those with type 1 chronic lung disease, such as asthma and tients with suboptimal blood glucose diabetes, require higher daily doses (;1 chronic obstructive pulmonary disease, control, especially those requiring large unit/kg), and have lower rates of hypo- andisnotrecommendedinpatientswho insulin doses, adjunctive use of a thia- glycemia (86). Titration can be based on smoke or who recently stopped . zolidinedione or an SGLT2 inhibitor home glucose monitoring or A1C. With All patients require spirometry (forced may help to improve control and reduce significant additions to the prandial in- expiratory volume in 1 s [FEV1]) testing to the amount of insulin needed, though sulin dose, particularly with the evening identify potential lung disease prior to potential side effects should be consid- meal, consideration should be given to and after starting inhaled insulin therapy. ered.Once a basal/ insulin regimen decreasing basal insulin. Meta-analyses is initiated, dose titration is important, of trials comparing rapid-acting insulin Combination Injectable Therapy with adjustments made in both mealtime analogs with human in If basal insulin has been titrated to an and basal insulins based on the blood patients with type 2 diabetes have not acceptable fasting blood glucose level (or glucose levels and an understanding of reported important differences in A1C or if the dose is .0.5 units/kg/day with the pharmacodynamic profile of each for- hypoglycemia (87,88). indications of need for other therapy) mulation (pattern control). As people with Concentrated Insulins and A1C remains above target, consider type 2 diabetes get older, it may become Several concentrated insulin prepara- advancing to combination injectable necessary to simplify complex insulin regi- tions are currently available. U-500 reg- therapy (Fig. 9.2). This approach can use a mens because of a decline in self-manage- “ ular insulin is, by definition, five times GLP-1 RA added to basal insulin or multi- ment ability (see Section 12 Older ” more concentrated than U-100 regular ple doses of insulin. The combination of Adults, https://doi.org/10.2337/dc21- insulin. Regular U-500 has distinct phar- basal insulin and GLP-1 RA has potent S012). macokinetics with delayed onset and glucose-lowering actions and less weight longer duration of action, has character- gain and hypoglycemia compared with References fi – istics more like an intermediate-acting intensi edinsulinregimens (93 95),with 1. Cleary PA, Orchard TJ, Genuth S, et al.; DCCT/ one study suggesting greater durability EDIC Research Group. The effect of intensive (NPH) insulin, and can be used as two or fi of glycemic effect compared with addi- glycemic treatment on coronary artery calci ca- three daily injections (89). U-300 glargine tion in type 1 diabetic participants of the and U-200 degludec are three and two tion of basal insulin alone (47). Two Diabetes Control and Complications Trial/ fi timesasconcentratedastheirU-100 different once-daily, xed dual-combination Interventions and Complications (DCCT/EDIC) study. Diabetes 2006; formulations and allow higher doses of products containing basal insulin plus a GLP-1 RA are available: 55:3556–3565 basal insulin administration per volume 2. Nathan DM, Cleary PA, Backlund J-YC, et al.; used. U-300 glargine has a longer dura- plus lixisenatide and Diabetes Control and Complications Trial/ tion of action than U-100 glargine but plus liraglutide. Epidemiology of Diabetes Interventions and fi modestly lower efficacy per unit admin- Intensi cation ofinsulintreatmentcan Complications (DCCT/EDIC) Study Research Group. Intensive diabetes treatment and car- istered (90,91). The FDA has also approved be done by adding doses of prandial to basal insulin. Starting with a single pran- diovascular disease in patients with type 1 a concentrated formulation of rapid-acting diabetes. N Engl J Med 2005;353:2643–2653 insulin lispro, U-200 (200 units/mL) and dial dose with the largest meal of the day 3. Diabetes Control and Complications Trial insulin lispro-aabc (U-200). These concen- is simple and effective, and it can be (DCCT)/Epidemiology of Diabetes Interventions trated preparations may be more conve- advanced to a regimen with multiple and Complications (EDIC) Study Research Group. Mortality in type 1 diabetes in the DCCT/EDIC nient and comfortable for patients to inject prandial doses if necessary (96). Alter- natively, in a patient on basal insulin in versus the general population. Diabetes Care and may improve adherence in those 2016;39:1378–1383 with insulin resistance who require large whom additional prandial coverage is 4. Writing Team for the Diabetes Control and doses of insulin. While U-500 regular in- desired, the regimen can be converted to Complications Trial/Epidemiology of Diabetes sulin is available in both prefilled pens and two doses of a premixed insulin. Each Interventions and Complications Research Group. Effect of intensive therapy on the microvascular vials, other concentrated insulins are avail- approach has advantages and disadvan- tages. For example, basal/prandial regi- complications of type 1 diabetes mellitus. JAMA able only in prefilled pens to minimize the 2002;287:2563–2569 fl risk of dosing errors. mens offer greater exibility for patients 5. Tricco AC, Ashoor HM, Antony J, et al. Safety, who eat on irregular schedules. On the effectiveness, and cost effectiveness of long Inhaled Insulin other hand, two doses of premixed in- acting versus intermediate acting insulin for Inhaled insulin is available as a rapid- sulin is a simple, convenient means of patients with type 1 diabetes: systematic re- view and network meta-analysis. BMJ 2014; acting insulin; studies in people with spreading insulin across the day. More- 349:g5459 type 1 diabetes suggest rapid pharma- over, human insulins, separately, self- 6. Bartley PC, Bogoev M, Larsen J, Philotheou A. cokinetics (8). A pilot study found mixed, or as premixed NPH/regular (70/ Long-term efficacy and safety of S122 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

compared to Neutral Protamine Hagedorn in- a multicentre, 12-week randomised trial. Lancet controlled type 1 diabetes (DEPICT-1): 24 week sulin in patients with type 1 diabetes using a 2018;392:1321–1329 results from a multicentre, double-blind, phase treat-to-target basal-bolus regimen with insulin 19. Brown SA, Kovatchev BP, Raghinaru D, et al.; 3, randomised controlled trial. Lancet Diabetes aspart at meals: a 2-year, randomized, controlled iDCL Trial Research Group. Six-month random- Endocrinol 2017;5:864–876 trial. Diabet Med 2008;25:442–449 ized, multicenter trial of closed-loop control in 33. Rosenstock J, Marquard J, Laffel LM, et al. 7. DeWitt DE, Hirsch IB. Outpatient insulin ther- type 1 diabetes. N Engl J Med 2019;381:1707– Empagliflozin as adjunctive to insulin therapy in apy in type 1 and type 2 diabetes mellitus: 1717 type 1 diabetes: the EASE trials. Diabetes Care scientific review. JAMA 2003;289:2254–2264 20. Peters AL, Laffel L (Eds.). American Diabetes 2018;41:2560–2569 8. BodeBW,McGillJB,LorberDL,GrossJL,Chang Association/JDRF Type 1 Diabetes Sourcebook. 34. Dean PG, Kukla A, Stegall MD, Kudva YC. PC, Bregman DB; Affinity 1 Study Group. Inhaled Alexandria, VA, American Diabetes Association, Pancreas transplantation. BMJ. 2017;357:j1321 technosphere insulin compared with injected 2013 35. Davies MJ, D’Alessio DA, Fradkin J, et al. prandial insulin in type 1 diabetes: a randomized 21. Chiang JL,KirkmanMS,LaffelLMB,Peters AL; Management of hyperglycemia in type 2 diabe- 24-week trial. Diabetes Care 2015;38:2266–2273 Type 1 Diabetes Sourcebook Authors. Type 1 di- tes, 2018. A consensus report by the American 9. Russell-Jones D, Bode BW, De Block C, et al. abetes through the life span: a position state- Diabetes Association (ADA) and the European Fast-acting improves glycemic ment of the American Diabetes Association. Association for the Study of Diabetes (EASD). control in basal-bolus treatment for type 1 Diabetes Care 2014;37:2034–2054 Diabetes Care 2018;41:2669–2701 diabetes: results of a 26-week multicenter, 22. Bell KJ, Barclay AW, Petocz P, Colagiuri S, 36. Buse JB, Wexler DJ, Tsapas A, et al. 2019 active-controlled, treat-to-target, randomized, Brand-Miller JC. Efficacy of carbohydrate count- update to: management of hyperglycemia in parallel-group trial (onset 1). Diabetes Care ing in type 1 diabetes: a systematic review and type2diabetes,2018.Aconsensus reportbythe 2017;40:943–950 meta-analysis. Lancet Diabetes Endocrinol 2014; American Diabetes Association (ADA) and the 9a. Klaff L, Cao D, Dellva MA, et al. Ultra rapid 2:133–140 European Association for the Study of Diabetes lispro improves postprandial glucose control 23. Vaz EC, Porf´ırio GJM, Nunes HRC, Nunes- (EASD). Diabetes Care 2020;43:487–493 compared with lispro in patients with type 1 Nogueira VDS. Effectiveness and safety of car- 37. Holman RR, Paul SK, Bethel MA, Matthews diabetes: results from the 26-week PRONTO-T1D bohydrate counting in the management of adult DR, Neil HAW. 10-year follow-up of intensive study. Diabetes Obes Metab 2020;22:1799–1807 patients with type 1 diabetes mellitus: a system- glucose control in type 2 diabetes. N Engl J Med 9b. Blevins T, Zhang Q, Frias JP, Jinnouchi H, atic review and meta-analysis. Arch Endocrinol 2008;359:1577–1589 Chang AM; PRONTO-T2D Investigators. Random- Metab 2018;62:337–345 38. Maruthur NM, Tseng E, Hutfless S, et al. Di- ized double-blind clinical trial comparing ultra 24. Bell KJ, Smart CE, Steil GM, Brand-Miller JC, abetes medications as monotherapy or metformin- rapid lispro with lispro in a basal-bolus regimen King B, Wolpert HA. Impact of fat, protein, and based combination therapy for type 2 diabetes: in patients with type 2 diabetes: PRONTO-T2D. on postprandial glucose control in a systematic review and meta-analysis. Ann In- Diabetes Care 2020;43:2991–2998 type 1 diabetes: implications for intensive diabetes tern Med 2016;164:740–751 10. Lane W, Bailey TS, Gerety G, et al.; Group management in the continuous glucose monitor- 39. U.S. Food and Drug Administration. FDA Information; SWITCH1. Effect of insulin degludec ing era. Diabetes Care 2015;38:1008–1015 Drug Safety Communication: FDA revises warn- vs insulin glargine U100 on hypoglycemia in patients 25. FridAH,KreugelG,GrassiG,etal.Newinsulin ings regarding use of the diabetes medicine with type 1 diabetes: the SWITCH 1 randomized deliveryrecommendations.MayoClinProc 2016; metformin in certain patientswithreduced kidney clinical trial. JAMA 2017;318:33–44 91:1231–1255 function. Accessed 6 March 2020. Available from 11. Home PD, Bergenstal RM, BolliGB, et al. New 26. Bergenstal RM, Strock ES, Peremislov D, https://www.fda.gov/drugs/drug-safety-and- insulin glargine 300 units/mL versus glargine Gibney MA, Parvu V, Hirsch LJ. Safety and efficacy availability/fda-drug-safety-communication-fda- 100 units/mL in people with type 1 diabetes: ofinsulintherapydeliveredviaa4mmpenneedle revises-warnings-regarding-use-diabetes-medicine- a randomized, phase 3a, open-label clinical trial in obese patients with diabetes. Mayo Clin Proc metformin-certain (EDITION 4). Diabetes Care 2015;38:2217–2225 2015;90:329–338 40. Out M, Kooy A, Lehert P, Schalkwijk CA, 12. Yeh H-C, Brown TT, Maruthur N, et al. 27. RatnerRE,DickeyR, FinemanM,et al.Amylin Stehouwer CDA. Long-term treatment with Comparative effectiveness and safety of meth- replacement with pramlintide as an adjunct to metformin in type 2 diabetes and methylmalonic ods of insulin delivery and glucose monitoring for insulin therapy improves long-term glycaemic acid: post hoc analysis of a randomized con- diabetes mellitus: a systematic review and meta- and weight control in type 1 diabetes mellitus: trolled 4.3 year trial. J Diabetes Complications analysis. Ann Intern Med 2012;157:336–347 a1-year,randomizedcontrolledtrial.DiabetMed 2018;32:171–178 13. Pickup JC. The evidence base for diabetes 2004;21:1204–1212 41. Aroda VR, Edelstein SL, Goldberg RB, et al.; technology: appropriate and inappropriate meta- 28. Edelman S, Garg S, Frias J, et al. A double- Diabetes Prevention Program Research Group. analysis. J Diabetes Sci Technol 2013;7:1567–1574 blind, placebo-controlled trial assessing pramlin- Long-term metformin use and vitamin B12 de- 14. Bergenstal RM, Klonoff DC, Garg SK, et al.; tide treatment in the setting of intensive insulin ficiency in the diabetes prevention program out- ASPIRE In-Home Study Group. Threshold-based therapy in type 1 diabetes. Diabetes Care 2006; comes study. J Clin Endocrinol Metab 2016;101: insulin-pump interruption for reduction of hy- 29:2189–2195 1754–1761 poglycemia. N Engl J Med 2013;369:224–232 29. Meng H,ZhangA,Liang Y,Hao J,Zhang X,Lu J. 42. Henry RR, Murray AV, Marmolejo MH, 15. Buckingham BA, Raghinaru D, Cameron F, Effect of metformin on glycaemic control in Hennicken D, Ptaszynska A, List JF. Dapagliflozin, et al.; In Home Closed Loop Study Group. Pre- patients with type 1 diabetes: a meta-analysis of metformin XR, or both: initial pharmacotherapy dictive low-glucose insulin suspension reduces randomized controlled trials. Diabetes Metab fortype 2diabetes, arandomisedcontrolledtrial. duration of nocturnal hypoglycemia in children Res Rev 2018;34:e2983 Int J Clin Pract 2012;66:446–456 without increasing ketosis. Diabetes Care 2015; 30. Petrie JR, Chaturvedi N, Ford I, et al.; RE- 43. Babu A, Mehta A, Guerrero P, et al. Safe and 38:1197–1204 MOVAL Study Group. Cardiovascular and meta- simple emergency department discharge therapy 16. Bergenstal RM, Garg S, Weinzimer SA, et al. bolic effects of metformin in patients with type 1 for patients with type 2 diabetes mellitus and severe Safety of a hybrid closed-loop insulin delivery diabetes (REMOVAL): a double-blind, rando- hyperglycemia. Endocr Pract 2009;15:696–704 system in patients with type 1 diabetes. JAMA mised, placebo-controlled trial. Lancet Diabe- 44. Cahn A, Cefalu WT. Clinical considerations 2016;316:1407–1408 tes Endocrinol 2017;5:597–609 for use of initial combination therapy in type 2 17. Garg SK, Weinzimer SA, Tamborlane WV, 31. Wang W, Liu H, Xiao S, Liu S, Li X, Yu P. Effects diabetes. Diabetes Care 2016;39(Suppl. 2):S137– et al. Glucose outcomes with the in-home use of insulin plus glucagon-like peptide-1 receptor S145 of a hybrid closed-loop insulin delivery system in agonists (GLP-1RAs) in treating type 1 diabetes 45. Abdul-Ghani MA, Puckett C, Triplitt C, et al. adolescents and adults with type 1 diabetes. mellitus: a systematic review and meta-analysis. Initial combination therapy with metformin, Diabetes Technol Ther 2017;19:155–163 Diabetes Ther 2017;8:727–738 pioglitazone and exenatide is more effective 18. Tauschmann M, Thabit H, Bally L, et al.; 32. Dandona P, Mathieu C, Phillip M, et al.; than sequential add-on therapy in subjects with APCam11 Consortium. Closed-loop insulin de- DEPICT-1 Investigators. Efficacy and safety of new-onset diabetes. Results from the Efficacy livery in suboptimally controlled type 1 diabetes: dapagliflozin in patients with inadequately and Durability of Initial Combination Therapy care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S123

for Type 2 Diabetes (EDICT): a randomized trial. a randomised, open-label, parallel-group, multi- initiating insulin glargine 100U/mL or neutral Diabetes Obes Metab 2015;17:268–275 centre, multinational, phase 3a trial. Lancet Di- protamine Hagedorn insulin analysed according 46. Phung OJ, Sobieraj DM, Engel SS, Rajpathak abetes Endocrinol 2017;5:355–366 to concomitant oral antidiabetes therapy. Di- SN. Early combination therapy for the treatment 59. Davies M, Heller S, Sreenan S, et al. Once- abetes Res Clin Pract 2017;124(Suppl. C):57–65 of type 2 diabetes mellitus: systematic review weekly exenatide versus once- or twice-daily 73. Riddle MC, Rosenstock J, Gerich J; Insulin and meta-analysis. Diabetes Obes Metab 2014; insulin detemir: randomized, open-label, clinical Glargine 4002 Study Investigators. The treat-to- 16:410–417 trial of efficacy and safety in patients with type 2 target trial: randomized addition of glargine or 47. Aroda VR, Gonzalez-Galvez´ G, Grøn R, et al. diabetes treated with metformin alone or in human NPH insulin to oral therapy of type 2 Durability of insulin degludec plus liraglutide combination with sulfonylureas. Diabetes Care diabetic patients. Diabetes Care 2003;26:3080– versus insulin glargine U100 as initial injectable 2013;36:1368–1376 3086 therapy in type 2 diabetes (DUAL VIII): a multi- 60. Diamant M, Van Gaal L, Stranks S, et al. Once 74. HermansenK,DaviesM,DerezinskiT,Martinez centre, open-label, phase 3b, randomised con- weekly exenatide compared with insulin glargine Ravn G, Clauson P, Home P. A 26-week, random- trolled trial. Lancet Diabetes Endocrinol 2019;7: titrated to target in patients with type 2 diabetes ized, parallel, treat-to-target trial comparing in- 596–605 (DURATION-3): an open-label randomised trial. sulin detemir with NPH insulin as add-on therapy 48. Matthews DR, Paldanius´ PM, Proot P, Chiang Lancet 2010;375:2234–2243 to oral glucose-lowering drugs in insulin-naive Y, Stumvoll M, Prato SD. Glycaemic durability of 61. Riddle MC, Herman WH. The cost of diabetes people with type 2 diabetes. Diabetes Care 2006; an early combination therapy with vildagliptin caredan elephant in the room. Diabetes Care 29:1269–1274 and metformin versus sequential metformin 2018;41:929–932 75. Yki-Jarvinen¨ H, Kauppinen-Makelin¨ R, Tiikkainen monotherapy in newly diagnosed type 2 di- 62. Truven Health Analytics. Micromedex 2.0: M, et al. Insulin glargine or NPH combined with abetes (VERIFY): a 5-year, multicentre, rand- Introduction to RED BOOK Online. Accessed 10 metformin in type 2 diabetes: the LANMET study. omised, double-blind trial. Lancet 2019;394: September 2020. Available from https://www. Diabetologia 2006;49:442–451 1519–1529 micromedexsolutions.com/micromedex2/4.34.0/ 76. Bolli GB, Riddle MC, Bergenstal RM, et al.; on 49. Bennett WL, Maruthur NM, Singh S, et al. WebHelp/RED_BOOK/Introduction_to_REDB_BOOK_ behalf of the EDITION 3 study investigators. New Comparative effectiveness and safety of medi- Online.htm insulinglargine 300 U/mlcomparedwith glargine cations for type 2 diabetes: an update including 63. Centers for Medicare & Services. 100 U/ml in insulin-na¨ıve people with type 2 new drugs and 2-drug combinations. Ann Intern NADAC (National Average Drug Acquisition Cost) diabetes on oral glucose-lowering drugs: a ran- Med 2011;154:602–613 drug pricing and payment. Accessed 2 October domized controlled trial (EDITION 3). Diabetes 50. MaloneyA, RosenstockJ,Fonseca V. A model- 2020. Available from https://data.medicaid.gov/ Obes Metab 2015;17:386–394 based meta-analysis of 24 antihyperglyce- Drug-Pricing-and-Payment/NADAC-National- 77. Terauchi Y, Koyama M, Cheng X, et al. New mic drugs for type 2 diabetes: comparison of Average-Drug-Acquisition-Cost-/a4y5-998d insulin glargine 300 U/ml versus glargine 100 U/ treatment effects at therapeutic doses. Clin Phar- 64. Kang H, Lobo JM, Kim S, Sohn M-W. Cost- ml in Japanese people with type 2 diabetes using macol Ther 2019;105:1213–1223 related medication non-adherence among U.S. basal insulin and oral antihyperglycaemic drugs: 51. Vijan S, Sussman JB, Yudkin JS, Hayward RA. adults with diabetes. Diabetes Res Clin Pract glucose control and hypoglycaemia in a random- Effectof patients’ risks andpreferenceson health 2018;143:24–33 ized controlled trial (EDITION JP 2). Diabetes gainswith plasma glucose level lowering in type 2 65. Patel MR, Piette JD, Resnicow K, Kowalski- Obes Metab 2016;18:366–374 diabetes mellitus. JAMA Intern Med 2014;174: Dobson T, Heisler M. Social determinants of 78. Yki-Jarvinen¨ H, Bergenstal RM, Bolli GB, et al. 1227–1234 health, cost-related nonadherence, and cost- Glycaemic control and hypoglycaemia with new 52. Tsapas A, Avgerinos I, Karagiannis T, et al. reducing behaviorsamong adultswithdiabetes: insulin glargine 300 U/ml versus insulin glargine Comparative effectiveness of glucose-lowering findings from the National Health Interview 100 U/ml in people with type 2 diabetes using drugs for type 2 diabetes: a systematic review Survey. Med Care 2016;54:796–803 basal insulin and oral antihyperglycaemic drugs: andnetworkmeta-analysis.AnnInternMed 66. Blonde L, Merilainen M, Karwe V, Raskin P; the EDITION 2 randomized 12-month trial in- 2020;173:278–286 TITRATE Study Group. Patient-directed titration cluding 6-month extension. Diabetes Obes Metab 53. Pratley R, Amod A, Hoff ST, et al.; PIONEER for achieving glycaemic goals using a once-daily 2015;17:1142–1149 4 investigators. Oral semaglutide versus subcutaneous basal insulin analogue: an assessment of two 79. Marso SP, McGuire DK, Zinman B, et al.; liraglutide and placebo in type 2 diabetes (PIO- different fasting plasma glucose targets - the DEVOTE Study Group. Efficacy and safety of NEER 4): a randomised, double-blind, phase 3a TITRATE study. Diabetes Obes Metab 2009;11: degludec versus glargine in type 2 diabetes. N trial. Lancet 2019;394:39–50 623–631 Engl J Med 2017;377:723–732 54. Singh S, Wright EE Jr, Kwan AYM, et al. 67. Porcellati F, Lucidi P, Cioli P, et al. Pharma- 80. Rodbard HW, Cariou B, Zinman B, et al.; Glucagon-like peptide-1 receptor agonists compared cokinetics and of insulin BEGIN Once Long trial investigators. Comparison with basal insulins for the treatment of type 2 glargine given in the evening as compared with in of insulin degludec with insulin glargine in insulin- diabetes mellitus: a systematic review and meta- the morning in type 2 diabetes. Diabetes Care naive subjects with type 2 diabetes: a 2-year analysis. Diabetes Obes Metab 2017;19:228–238 2015;38:503–512 randomized, treat-to-target trial. Diabet Med 2013; 55. Levin PA, Nguyen H, Wittbrodt ET, Kim SC. 68. Wang Z, Hedrington MS, Gogitidze Joy N, 30:1298–1304 Glucagon-likepeptide-1receptoragonists:asystem- et al. Dose-response effects of insulin glargine in 81. Wysham C, Bhargava A, Chaykin L, et al. atic review of comparative effectiveness research. type 2 diabetes. Diabetes Care 2010;33:1555– Effect of insulin degludec vs insulin glargine u100 Diabetes Metab Syndr Obes 2017;10:123–139 1560 onhypoglycemiainpatientswithtype 2diabetes: 56. Abd El Aziz MS, Kahle M, Meier JJ, Nauck MA. 69. Singh SR, Ahmad F, Lal A, Yu C, Bai Z, Bennett the SWITCH 2 randomized clinical trial. JAMA A meta-analysis comparing clinical effects of H. Efficacy and safety of insulin analogues for the 2017;318:45–56 short- or long-acting GLP-1 receptor agonists management of diabetes mellitus: a meta-analysis. 82. Zinman B, Philis-Tsimikas A, Cariou B, et al.; versus insulin treatmentfromhead-to-head CMAJ 2009;180:385–397 NN1250-3579 (BEGIN Once Long) Trial Investi- studies in type 2 diabetic patients. Diabetes Obes 70. Horvath K, Jeitler K, Berghold A, et al. Long- gators. Insulin degludec versus insulin glargine Metab 2017;19:216–227 acting insulin analogues versus NPH insulin in insulin-naive patients with type 2 diabetes: 57. Giorgino F, Benroubi M, Sun J-H, Zimmermann (human isophane insulin) for type 2 diabetes a 1-year, randomized, treat-to-target trial (BE- AG, Pechtner V. Efficacy and safety of once-weekly mellitus. Database Syst Rev 2007;2: GIN Once Long). Diabetes Care 2012;35:2464– dulaglutide versus insulin glargine in patients with CD005613 2471 type 2 diabetes on metformin and glimepiride 71. Monami M, Marchionni N, Mannucci E. 83. Lipska KJ, Parker MM, Moffet HH, Huang ES, (AWARD-2). Diabetes Care 2015;38:2241–2249 Long-acting insulin analogues versus NPH hu- Karter AJ. Association of initiation of basal insulin 58. Aroda VR, Bain SC, Cariou B, et al. Efficacy man insulin in type 2 diabetes: a meta-analysis. analogs vs neutral protamine hagedorn insulin and safety of once-weekly semaglutide versus Diabetes Res Clin Pract 2008;81:184–189 with hypoglycemia-related emergency department once-daily insulin glargine as add-on to metfor- 72. Owens DR, Traylor L, Mullins P, Landgraf W. visits or hospital admissions and with glycemic min (with or without sulfonylureas) in insulin- Patient-level meta-analysis of efficacy and hy- control in patients with type 2 diabetes. JAMA naive patients with type 2 diabetes (SUSTAIN 4): poglycaemia in people with type 2 diabetes 2018;320:53–62 S124 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 44, Supplement 1, January 2021

84. Cowart K. Overbasalization: addressing hes- severely insulin-resistant patients with type 2 93. Diamant M, Nauck MA, Shaginian R, et al.; 4B itancy in treatment intensification beyond basal diabetes. Endocr Pract 2016;22:653–665 Study Group. Glucagon-like peptide 1 receptor insulin. Clin Diabetes 2020;38:304–310 90. Riddle MC, Yki-Jarvinen¨ H, Bolli GB, et al. agonist or bolus insulin with optimized basal 85. Cefalu WT, Dawes DE, Gavlak G, et al.; Insulin One-year sustained glycaemic control and less insulin in type 2 diabetes. Diabetes Care 2014;37: Access and Affordability Working Group. Insulin hypoglycaemia with new insulin glargine 300 U/ 2763–2773 Access and Affordability Working Group: con- ml comparedwith 100 U/ml in people with type 2 94. Eng C, Kramer CK, Zinman B, Retnakaran R. clusions and recommendations. Diabetes Care diabetes using basal plus meal-time insulin: the Glucagon-like peptide-1 receptor agonist and 2018;41:1299–1311 EDITION 1 12-month randomized trial, including basal insulin combination treatment for the 86. McCall AL. Insulin therapy and hypoglyce- 6-month extension. Diabetes Obes Metab 2015; management of type 2 diabetes: a systematic mia. Endocrinol Metab Clin North Am 2012;41: 17:835–842 review and meta-analysis. Lancet 2014;384: 57–87 91. Yki-Jarvinen¨ H, Bergenstal R, Ziemen M, 2228–2234 87.MannucciE,MonamiM,MarchionniN. et al.; EDITION 2 Study Investigators. New insulin 95. Maiorino MI, Chiodini P, Bellastella G, Short-acting insulin analogues vs. regular human glargine 300 units/mL versus glargine 100 units/ Capuano A, Esposito K, Giugliano D. Insulin and insulin in type 2 diabetes: a meta-analysis. Di- mL in people with type 2 diabetes using oral glucagon-like peptide 1 receptor agonist com- abetes Obes Metab 2009;11:53–59 agents and basal insulin: glucose control and bination therapy in type 2 diabetes: a systematic 88. Heller S, Bode B, Kozlovski P, Svendsen AL. hypoglycemia in a 6-month randomized con- review and meta-analysis of randomized con- Meta-analysis of insulin aspart versus regular trolled trial (EDITION 2). Diabetes Care 2014; trolled trials. Diabetes Care 2017;40:614–624 human insulin used in a basal-bolus regimen for 37:3235–3243 96. Rodbard HW, Visco VE, Andersen H, Hiort LC, the treatment of diabetes mellitus. J Diabetes 92. Akturk HK, Snell-Bergeon JK, Rewers A, et al. Shu DHW. Treatment intensification with step- 2013;5:482–491 Improved postprandial glucose with inhaled wise addition of prandial insulin aspart boluses 89. Wysham C, Hood RC, Warren ML, Wang T, technosphere insulin compared with insulin as- compared with full basal-bolus therapy (Full- Morwick TM, Jackson JA. Effect of total daily dose part in patients with type 1 diabetes on multiple STEP Study): a randomised, treat-to-target clin- on efficacy, dosing, and safety of 2 dose titration dailyinjections:the STATstudy.DiabetesTechnol ical trial. Lancet Diabetes Endocrinol 2014;2: regimens of human regular U500 insulin in Ther 2018;20:639–647 30–37