Bridging Efficacy and Adherence in Glucagon-Like Peptide-1 Receptor Agonist Therapy: The Emerging Role of Oral Agents

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have significantly advanced the management of type 2 diabetes mellitus (T2DM) and obesity, with much of the literature focusing on the synthesis of associated risks [1,2]—including gastrointestinal events, psychiatric symptoms, and ocular complications—that warrant careful consideration in clinical practice. However, an equally important dimension of GLP-1 RA therapy is patient adherence, which directly influences both safety and therapeutic effectiveness. Several pharmaceutical companies are advancing oral GLP-1 RAs to improve adherence and patient acceptability in the treatment of T2DM and obesity. Semaglutide (Rybelsus, Novo Nordisk, Bagsvaerd, Denmark) remains the only U.S. Food and Drug Administration-approved oral GLP-1 RA to date [3]. Investigational oral GLP-1 RAs under development include orforglipron (Eli Lilly, Indianapolis, IN, USA; phase 3 [4]), GSBR-1290 (Structure Therapeutics, South San Francisco, CA, USA; phase 2), and HS‑10535 (Hansoh Pharma [Lianyungang, China], licensed to Merck [Rahway, NJ, USA] in 2023 [5]), while earlier-stage candidates such as the dual GLP-1/glucose-dependent insulinotropic polypeptide receptor agonist HS-20094—licensed by Regeneron from Hansoh Pharma in June 2025 and currently in phase 3 trials—underscore the growing industry-wide pursuit of orally delivered incretin-based therapies.

Research comparing real-world adherence to oral GLP-1 RAs versus injectable GLP-1 RAs remains scarce. Our recent analysis of over 78,000 commercially insured adults in the United States from 2010 to 2022 offers additional context to the discussion. We conducted a retrospective analysis of IQVIA PharMetrics Plus for Academics (IQVIA, Durham, NC, USA), a 25% random sample of a national longitudinal health plan database that includes fully adjudicated medical and pharmacy claims for commercially insured individuals in the United States. We identified adults aged 18 to 65 who initiated a GLP-1 RA between 2010 and 2022 and had more than 6 months of follow-up data. Given that newer agents like oral semaglutide were introduced later in the study period, we restricted our analysis to each agent’s index prescription. The index prescription was defined as the first observed GLP-1 RA prescription following a minimum 90-day period without any prior GLP-1 RA use [6]. Baseline comorbidities, including T2DM and obesity, were identified via International Classification of Diseases, 10th Revision (ICD-10) codes from the index year and preceding years. Adherence was assessed using the proportion of days covered (PDC), with PDC ≥80% [7] indicating high adherence, calculated over both 6-month [6] and 12-month [8] periods. We monitored prescription fills beyond the 6- or 12-month adherence window, classifying gaps >90 days as discontinuation and any reinitiation after such a gap as a new treatment episode with a new index date. This approach enabled a more balanced comparison of adherence across agents within a contemporaneous clinical context. All analyses were conducted using Python version 3.11.8 (Python Software Foundation, Wilmington, DE, USA). This study was deemed exempt from Institutional Review Board approval due to unidentifiable data.

In our analytic sample of 78,297 patients prescribed GLP-1 RAs with >6 months follow-up, the median age was 53 years, and 55.3% were female. More than half of the patients (54.7%) had both T2DM and obesity concurrently, while 40.4% had T2DM only and 4.9% had obesity only. Most patients (98.9%) were prescribed injectable GLP-1 RAs; liraglutide accounted for the largest share of prescriptions (60.3%), followed by dulaglutide (25.0%) and injectable semaglutide (8.6%). At 6 months, mean PDC ranged between 58% to 69.5%, reflecting moderate overall adherence (Table 1). Less than half (range 28% to 49.7%) of the patients achieved high adherence (PDC ≥80%). At 12-month follow-up, mean PDC improved across all agents (69.8% to 82.4%). The proportion of patients achieving high adherence (PDC ≥80%) was highest for oral semaglutide (65.1%), followed by dulaglutide (59.1%), and lowest for liraglutide/insulin degludec (34.9%) and injectable semaglutide (38.8%).

Table 1.

Six-month and 1-year PDC for glucagon-like peptide-1 receptor agonists

Drug name (available years)	Analysis for 6-month PDC			Analysis for 1-year PDC
	Number	Mean value, % (95% CI)	Percentage of PDC ≥80% (95% CI)	Number	Mean value, % (95% CI)	Percentage of PDC ≥80% (95% CI)
Oral agents
Semaglutide (2019–2021)	871	61.6 (59.4–63.8)	42.7 (39.4–46.0)	404	82.4 (80.6–84.2)	65.1 (60.4–69.8)
Injectable agents
Semaglutide (2017–2021)	6,714	60.6 (59.9–61.3)	35.0 (33.9–36.1)	3,007	72.1 (71.4–72.8)	38.8 (37.0–40.5)
Liraglutide (2011–2021)	47,237	61.5 (61.2–61.7)	37.1 (36.7–37.5)	24,730	72.8 (72.6–73.1)	44.5 (43.9–45.1)
Exenatide (2014–2021)	1,865	58.0 (56.6–59.4)	33.8 (31.7–36.0)	836	71.8 (70.3–73.2)	44.4 (41.0–47.7)
Exenatide extended release (2017–2021)	1,279	62.2 (60.4–63.9)	39.3 (36.7–42.0)	602	74.4 (72.8–76.1)	45.3 (41.4–49.3)
Dulaglutide (2013–2021)	19,563	69.5 (69.1–69.9)	49.7 (49.0–50.4)	11,230	80.2 (79.9–80.6)	59.1 (58.1–60.0)
Insulin degludec and liraglutide (2016–2021)	243	59.2 (55.7–62.7)	28.0 (22.6–33.7)	126	69.8 (66.1–73.4)	34.9 (27.0–43.7)
Insulin glargine and lixisenatide (2016–2021)	525	62.1 (59.6–64.7)	37.9 (33.7–42.1)	272	73.0 (70.4–75.5)	42.6 (36.8–48.5)

PDC, proportion of days covered; CI, confidence interval.

In our nationally representative cohort of commercially insured adults prescribed GLP-1 RAs between 2010 and 2022, adherence was moderate and varied across agents over a 6-month follow-up. Liraglutide was the most commonly prescribed agent, but less than half of patients (44.5%) reached high adherence (>80% PDC) by 12-month follow-up. Oral semaglutide showed the highest adherence (65.1%) but was much less commonly prescribed. Oral semaglutide demonstrated higher adherence than injectable semaglutide at both 6 and 12 months. While this may suggest greater convenience [9] of oral agents, the adherence advantage could also reflect unmeasured confounding by indication, health literacy, socioeconomic status, insurance coverage, and adverse event profiles. Pharmaceutical companies are increasingly developing oral GLP-1 RA formulations to align with patient preferences and enhance treatment adherence. Given that higher adherence to GLP-1 RAs is generally associated with improved glycemic control and weight reduction [10,11], our finding of enhanced adherence with the oral formulation may have important implications for industry stakeholders.

Several limitations warrant consideration. First, administrative claims data lack clinical details (e.g., glycosylated hemoglobin, blood pressure, renal function) and reasons for discontinuation. Second, oral semaglutide users represented only 1.1% of the cohort (871 patients), which limited the statistical power to detect adherence differences compared to injectable formulations. Third, the study was restricted to commercially insured adults aged 18 to 65 in the United States, limiting generalizability to commercially insured and older adults. Interestingly, adherence rates at 12 months exceeded those at 6 months across several agents. It may reflect that only persistent users remain in the 12-month analytic sample. Individuals who continue therapy beyond 6 months likely represent a more adherent subpopulation. This phenomenon, commonly referred to as the ‘healthy adherer effect [12],’ was a common confounder in observational studies. Last, fixed-dose combinations of GLP-1 RAs with basal insulin (e.g., degludec and glargine) may have more flexible, patient-adjusted dosing schedules, which can reduce fill frequency despite good adherence. As such, our PDC estimates may underestimate adherence for these agents.

Our study demonstrates that real-world adherence to GLP-1 RAs varies substantially across agents, with adherence to oral semaglutide exceeding that observed for most injectable formulations. Adherence patterns in real-world settings are shaped by a complex interplay of measurable factors [13]—such as patient demographics, comorbidities, medication burden, and documented adverse events—and unmeasured determinants, including health literacy, treatment beliefs, and patient preferences. These findings suggest that formulation characteristics—particularly the route of administration—may meaningfully influence real-world persistence and adherence, which in turn could affect downstream clinical outcomes.