Merck’s eye drug for diabetes met its goal while restoring less sight than the standard drug

In Short. In a trial against the standard injection for retinal swelling in diabetes, Merck’s remigromig restored less sight and was linked to more growth of harmful new vessels. It passed because it only had to prove it was not much worse.

DIABETES CAN damage the small blood vessels of the retina, the layer of light-sensing cells at the back of the eye, so that they leak fluid into its center and blur vision. This condition, diabetic macular edema, affects nearly 1.6m people in America, according to Merck. For more than a decade it has been treated with injections into the eye of drugs such as ranibizumab (sold as Lucentis), which block VEGF, a signal that makes vessels leak and grow. Merck’s remigromig, an antibody injected in the same way, works differently. It switches on the Wnt pathway, a set of signals that, according to the presenting investigator quoted by Retinal Physician, helps the cells lining retinal vessels seal the gaps between them. Merck says up to 40% of patients still have swelling after six months of the older injections, which is the case for trying something new.

Merck tested remigromig in a trial called Brunello, whose full one-year results were presented on October 10th at the American Academy of Ophthalmology meeting in New Orleans. The trial did not compare the drug with a dummy injection. Instead it set remigromig against ranibizumab itself, an established drug used as the comparison, which trialists call an active control. According to the trial’s registry record, 984 adults were assigned at random in equal numbers to one of two doses of remigromig or to monthly ranibizumab, and neither patients, doctors nor the people measuring outcomes knew which.

The measure chosen in advance to judge success, the primary endpoint, was the change after 52 weeks in how many letters each patient could read on a standard eye chart. Each line of that chart holds five letters, and a gain of 15 letters, or three lines, counts as a moderate change and is commonly used as an outcome in trials, according to a review published by America’s National Library of Medicine. Patients on the two doses of remigromig gained an average of 9.1 and 8.7 letters. Those on ranibizumab gained 11.8.

That result counts as a pass because Brunello was designed to show only that remigromig was not much worse than ranibizumab. In such a trial the sponsor fixes beforehand the largest shortfall it will accept, called the non-inferiority margin. America’s Food and Drug Administration, in its 2016 guidance on such trials, describes this as “the largest clinically acceptable difference (degree of inferiority) of the test drug compared to the active control.” Merck says both doses fell within its prespecified margin and that the differences were “not considered clinically meaningful”. Merck has not published the margin. Nor has it published the ranges of uncertainty around the gap between each dose and ranibizumab, though the ranges it gave for each group on its own do not overlap. Merck’s figures imply that the gap was roughly between one and a half and four and a half letters, and that the margin must therefore have been wider than that. Neither dose beat ranibizumab on any of the trial’s secondary measures, which included retinal thickness.

The safety results raise a separate question. Proliferative diabetic retinopathy is the growth of fragile new vessels on the surface of the retina, which can bleed and scar. Side-effects indicating it occurred in 6.7% and 6.1% of patients on the two doses of remigromig, against 0.9% of those on ranibizumab. Treatment was stopped because of side-effects by 4.9% and 4.5% of patients on remigromig and by 0.9% on ranibizumab, while inflammation inside the eye was similar in all three groups. Merck says the extra vessel growth may reflect the natural course of the disease, because remigromig repairs the vessel lining rather than suppressing new growth. That explanation fits older evidence that drugs which block VEGF hold back this stage of the disease. In the earlier RISE and RIDE trials, a review of the evidence reports, it developed in 1.6-5.6% of eyes given ranibizumab against 11.5-15% of eyes given dummy injections.

The trial therefore establishes that remigromig restores sight to within an amount of ranibizumab that Merck judged acceptable, while the long-term meaning of the extra vessel growth remains unknown. Merck says further analyses of those cases are under way. BAROLO, a second trial with the same design that enrolled 1,054 patients and is due to finish its main analysis in March 2027, will show whether the smaller vision gain and the extra vessel growth hold up in a new group of patients.