Study shows oral semaglutide matches DPP-4i in adherence and persistence for type 2 diabetes

By Hugo Francisco de SouzaReviewed by Susha Cheriyedath, M.Sc.Jul 16 2024

In a recent large-scale retrospective cohort study published in the journal Primary Care Diabetes, researchers used a comprehensive (n = 10,465) database-derived dataset to evaluate the adherence and persistence of newly initiating oral semaglutide type 2 diabetes (T2D) patients compared to dipeptidyl peptidase-4 inhibitor (DPP-4i) users. While the latter has been established as a well-tolerated daily dose medication, oral semaglutide is relatively novel, with limited previous literature on its market acceptance or patient adherence.

Statistical analyses from the study reveal that despite being more expensive than DPP-4i, patient adherence over 12 months was similar between both medications. Additionally, semaglutide initially presented a steep drop in persistence but then stabilized to DPP-4i levels over the subsequent nine months. Notably, semaglutide-consuming patients were observed to require fewer concurrent anti-diabetic medications (ADMs) than DPP-4i, highlighting the potentially higher enduring efficacy of the former.

Study: Adherence and persistence among people with type 2 diabetes newly initiating oral semaglutide versus DPP-4is in a US real-world setting. Image Credit: Ti_A / Shutterstock

Background

Type 2 diabetes (T2D) is a chronic condition characterized by abnormally high blood sugar levels as a result of the body's inability to produce or utilize insulin properly. The International Diabetes Federation (IDF) estimates that 11.3% of all adults have diabetes (422 million individuals), with more than 90% of these patients having T2D. In the United States (US) alone, more than 95% of its 37 million diabetic patients report T2D, with projections expecting this prevalence to only increase in the coming decades.

Alongside health behavior counseling and diet-based glycemic control, pharmacotherapy is the most common mode of T2D treatment. The US Food and Drug Administration (FDA) has approved numerous anti-diabetic medications (ADMs), of which dipeptidyl peptidase-4 inhibitors (DPP-4is) and glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are the best-studied and most often prescribed ADM classes. Unfortunately, the World Health Organization (WHO) has highlighted that many T2D patients present low adherence to these medications, likely due to socioeconomic-, health system-, condition-, therapy-, or patient-related factors.

Conventional ADMs, including GLP-1 RAs, were often injection-based, possibly contributing to patients' discontinuation of medication. Advances in ADM research have resulted in the development of novel oral consumption-based medications. DPP-4is have been orally administered for years and are now established as safe and well-tolerated with minimal side effects. Semaglutide is a relatively novel once-daily oral GLP-1 RA. Despite clinical trials validating its safety and efficacy, resulting in the FDA approving its use in 2019, real-world patient adherence and persistence to the drug have never been formally investigated.

About the study

The present study evaluated US-based T2D patient adherence and persistence to semaglutide by comparing their patterns to those of DPP-4i. The study was a non-interventional, retrospective, cohort-based study using data from two administrative healthcare databases – the Merative MarketScan Commercial and the Medicare databases. These databases contain data across age groups and from all 50 US states.

For this study, 'adherence' was defined as the proportion of days covered (PDC) by the index drug (semaglutide or DPP-4i) using database-sourced insurance claims as proxies. 'Persistence' was defined as the number of days between the initial index drug prescription and its subsequent discontinuation using a cutoff of >45 days of no available medication supply. The study further investigated evidence of concurrent or alternate ADM utilization by patients prescribed either index drug, especially during the pre-index (6 months before index drug prescription) or post-index (12 months following index drug termination). The post-index period outcomes assessed the long-term efficacy of semaglutide or DPP-4i.

Statistical analyses included propensity score (ps) inverse probability weighting (IPW) of both semaglutide and DPP-4i cohorts to validate their statistical comparisons. Models used for these analyses included patients' anthropometric and demographic variables (age, sex, region, socioeconomic status) as covariates, adjusted for clinical parameters (index year, diabetes comorbidities, BMI, and concurrent medication). Differences in outcomes between included cohorts were evaluated using t-tests for continuous variables and Chi-squared tests for categorical variables. Finally, sensitivity analyses were carried out for covariates revealed to be statistically different between cohorts to ensure that these variables did not alter study results.

Study findings

After excluding participants with missing data, the final dataset comprised 5,485 and 4,980 T2D patients prescribed semaglutide and DPP-4i, respectively. Ps-IPW analyses revealed that demographic characteristics between both cohorts were statistically comparable, with 'out-of-pocket costs' and obesity being the only exception. Semaglutide costs amounted to an average of $103 compared to the DPP-4i average of $67 per 30 days. Notably, sensitivity analyses confirmed that these exceptions did not alter study outcomes.

Adherence evaluations revealed that despite mean PDC being slightly lower in the semaglutide cohort, the percentage of adherent patients was statistically similar between semaglutide and DPP-4i-consuming patients. In contrast, persistence patterns displayed temporal differences between cohorts – for the first six months following index drug prescription, semaglutide displayed substantially lower persistence than DPP-4i. However, as the study progressed, persistence in the semaglutide cohort stabilized at DPP-4i levels. This highlights that patients prescribed semaglutide may require additional education or support during the initial months of their medication course.

Encouragingly, concurrent ADM utilization was observed to be significantly lower in the semaglutide cohort compared to their DPP-4i-consuming counterparts. This trend continued even following the completion of patients' respective index drugs, with a larger proportion of DPP-4i cohort members requiring ADM post-index compared to semaglutide.

Conclusions

The present study found similar levels of adherence and persistence between patients prescribed daily oral DPP-4i and those initiating daily oral semaglutide. Notably, semaglutide patients displayed a marked reduction in persistence during the first six months following index drug prescription but then stabilized by the end of the 12-month-long study. This may be due to the substantially higher out-of-pocket cost of semaglutide compared to DPP-4i, though the retrospective nature of this study makes confirming that hypothesis presently impossible.

Semaglutide-consuming patients were observed to require fewer concurrent ADMs both during and, more importantly, after their medication course, highlighting this drug's improved and longer-lasting efficacy compared to DPP-4i. In summary, while costing more and potentially requiring additional support during the first few months of its use, semaglutide is at least as good if not better than DPP-4i as an anti-diabetic oral medication.