Who paid for this trial
Eli Lilly and Company funded TRANSCEND-T2D-1, and the paper states plainly that the sponsor was involved in study design, data collection, data review, analyses, and drafting of the report.
That is not an accusation. It is standard practice for industry-funded phase 3 trials, disclosed exactly as it should be, and this is a peer-reviewed publication in The Lancet, one of the most rigorously reviewed venues in medicine. But naming the funder before discussing the results is the same discipline applied to the paper covered in the first series in this publication, where the same standard was applied to a narrative review with far weaker methodology behind it. Consistency matters more than any single verdict.
The limitation the trial cannot get around
Every result in this series came from a specific kind of person: newly diagnosed, drug-naive, no other diabetes medication in the picture. That population choice is what made Parts 1 through 3 possible. It is also, by the authors' own account, the trial's central limitation.
The paper states it directly: the largely medication-naive participant population reduces the generalisability of the study findings to the broader population of people with type 2 diabetes who might already be on prescription antihyperglycaemic medications or an insulin regimen.
Most people living with type 2 diabetes today are not newly diagnosed. They are years or decades in, often on metformin, sometimes on insulin, sometimes on two or three drugs stacked together. Nothing in this trial tells you what happens when retatrutide is added on top of that existing regimen, how it interacts with those drugs, or whether the exceptional weight loss numbers from Part 2 would hold in a system that has been managing diabetes pharmacologically for years already. That is a real, unanswered question, not a footnote to skim past.
The three-country problem
Recruitment ran across the United States, Mexico and India. Three countries, 48 sites. The paper flags this too: restricting the trial to these countries limits generalisability to broader populations. Genetics, diet, healthcare access, baseline metabolic patterns and dozens of other variables differ meaningfully across the world's population of people with type 2 diabetes, and three countries, however thoughtfully chosen, are not a substitute for a broader geographic footprint.
The trial ended while the drug was still working
This is the limitation most likely to get lost in a topline summary, and it might be the most important one.
Part 3 noted that the weight loss curve showed no evidence of a plateau at week 40. That phrase is doing real work. It means the trial's own measurement window closed before the drug's effect on body weight had finished playing out. The 15.3 percent weight loss reported for the 12 mg group at week 40 is not necessarily the ceiling. It might be a point partway up a curve that was still descending when the researchers stopped measuring.
The paper is explicit that the 40-week treatment duration is itself a limitation, and that a longer trial would be needed to assess the full extent of efficacy, including whether outcomes like HbA1c and weight reduction eventually plateau, and to characterise long-term safety over a duration that actually resembles how long someone would take this drug in practice.
What was compared, and what wasn't
Part 3's comparison against semaglutide and tirzepatide monotherapy data was informative and it was not a randomised trial. Different studies, different eras, different populations beyond the matched baseline HbA1c, different statistical designs. The paper says so in its own words: cross-trial comparisons are restricted by differences in study design, and differences in diabetes duration, baseline bodyweight, and study design preclude definitive conclusions.
That caveat earns real weight here, because the comparison in Part 3 was one of the more striking findings in this entire series. Striking findings from indirect comparisons are exactly the kind that deserve the most scrutiny before anyone treats them as settled, and the field itself has treated it that way: two head-to-head trials are already registered specifically to resolve this, NCT06260722 against semaglutide and NCT06662383 against tirzepatide.
What's actually still coming
The researchers did not leave the open questions to sit. Two further trials in the same TRANSCEND-T2D programme are already running, and they target the exact two gaps this trial could not close on its own.
TRANSCEND-T2D-2 will compare retatrutide against semaglutide, head-to-head, over 80 weeks, specifically in participants who are already using metformin, with or without an SGLT2 inhibitor. That is the trial that answers the medication-naive limitation directly, by testing the drug in the population TRANSCEND-T2D-1 could not include: people already on background therapy.
TRANSCEND-T2D-3 will compare retatrutide against placebo over 52 weeks in participants with type 2 diabetes and renal impairment, a population usually excluded from early trials in this class and one where drug clearance and dosing safety need dedicated study before broader use can be recommended with confidence.
Between them, those two trials are built to close the two gaps named above: the drug-naive-only population, and the short observation window. That is worth knowing, because it tells you this is not a finished story being oversold as one. It is a programme that knows exactly what it has not yet proven, and has already funded the trials to find out.
What can honestly be said right now
Strip away everything that needs more data and what is left is still substantial. A double-blind, randomised, placebo-controlled phase 3 trial, in a genuinely under-studied population of newly diagnosed people with type 2 diabetes, produced HbA1c reductions competitive with the strongest data available in this drug class, weight loss that broke the pattern seen in every prior trial of a diabetic population, zero severe hypoglycaemia, and a safety profile that grew more tolerable, not less, between phase 2 and phase 3.
What cannot yet be said: how it performs on top of existing medication, how it performs beyond 40 weeks, how it stacks up in a randomised comparison against its two closest competitors, and what a rare, longer-term signal might look like in a drug most patients would take for years.
That is not a discouraging place to land. It is where every genuinely promising phase 3 result sits the week it is published; strong enough to matter, incomplete enough to require the next trial before anyone should treat it as the final word.
The trials to close those gaps are already enrolling. That is where this one leaves off.
This series covers a peer-reviewed phase 3 clinical trial published in The Lancet. Retatrutide has not been approved by the FDA for type 2 diabetes or any other indication. Nothing here is medical advice. Speak to a qualified clinician about your own circumstances.
