At the European Association for the Study of Diabetes meeting in Milan, Novo Nordisk announced real-world data on a question many people on GLP-1 injections quietly ask: what happens if you move to a daily pill instead? The headline was that people who switched lost an average of 4.1% of their body weight over the next three months.
That is an interesting data point. It is also a small, uncontrolled one, and the gap between “interesting” and “informative about you” is the point of this post.
What the study did
The analysis is called OCTANE, presented as a late-breaking abstract at EASD on September 30. It is a retrospective cohort study using de-identified records from Ro, a telehealth platform, combining health records, patient-reported data, and pharmacy records.
- 194 patients were in the final analysis, sampled from adults with overweight or obesity and at least one related condition such as high blood pressure, abnormal cholesterol, sleep apnea, or cardiovascular disease.
- Everyone had been on injectable semaglutide or tirzepatide for weight management and then started the Wegovy pill (oral semaglutide), on average about 37 days after their last injectable prescription.
- Average weight at the switch was about 100.5 kg (roughly 222 lb), with an average BMI of 34.5.
- Follow-up was three months on the pill.
What it reported
- Average weight change of 4.1% (about 4.0 kg, or 8.8 lb) after three months.
- 40.7% of participants had lost at least 5% of their body weight.
- The share of participants with a BMI of 30 or higher moved from 87.1% to 65.5%.
- Most participants with available data reported at least one self-described improvement, such as better-fitting clothes.
What it does not establish
There is no comparison group. Nobody in the study stayed on injections, and nobody was assigned to anything. Without a comparator, the data cannot say whether switching beat continuing, or whether similar changes would have happened regardless. A 4.1% figure with nothing next to it is a description, not a result.
Only people who stuck with the pill were counted. The analysis required continuous persistence on oral semaglutide with no gap longer than 30 days, plus self-reported weights within set time windows. The release does not say how many patients were screened out or stopped. That is a classic source of selection bias: people who tolerate and keep taking a medication tend to look better than everyone who started it.
Many weights were self-reported. Routine telehealth data are uneven in timing and completeness, and self-reporting introduces its own error. The study’s own authors list these limits, along with the reminder that real-world analyses “evaluate associations and cannot determine causality.”
It does not tell you the cost of switching. The study counted the people who lost weight. It does not answer how the pill compares with an injection at the same point, what happens at one year, or whether some people regain weight after moving. The Clinical Trials Arena write-up noted that longer-term outcomes remain to be established.
Who paid for it matters. The announcement came from Novo Nordisk, which makes the product studied, and describes the work as part of a collaboration with Ro. That does not make the numbers wrong. It means you should weigh the framing accordingly and wait for peer-reviewed publication.
There is also a small inconsistency worth flagging: the inclusion criteria as described include a lower BMI range than the group’s reported average, and the release does not reconcile the two. Details like this are why a conference abstract is a first look, not a final word.
Why a trial average still isn’t your number
Even a well-controlled study reports an average across many people. Some people in OCTANE presumably lost far more than 4.1% and some lost less or none. An average cannot tell you which group you would fall into, which is why the choice of medication, form, and dose is a conversation between you and a licensed provider who knows your history, not a conclusion to draw from a headline.
The part that applies to compounded medications
OCTANE studied one specific product: Novo Nordisk’s manufactured, branded Wegovy pill. The same is true of the trials that established oral semaglutide’s profile in the first place. That evidence covers that product and does not transfer to a compounded preparation, which is made by a state-licensed pharmacy for an individual patient, is not FDA-approved, and has not been through an equivalent trial. We spell out that distinction in our GLP-1 education guide, and it applies here as much as anywhere. If a company implies this study is evidence for a different product it sells, it is stretching the data past what it covers.
For what NoBsRx does offer, how the process works and what each program costs, see how it works and pricing.
This post is general information, not medical advice. It does not describe a benefit of any NoBsRx program, and it is not a recommendation to switch, start, stop, or change any medication. Compounded medications are not FDA-approved. Treatment is never guaranteed, and individual results vary. Eligibility and every treatment decision are made by independent licensed providers based on an individual clinical assessment, and availability varies by state. Talk to a licensed clinician before changing any treatment.