In the early 2000s, a firm known as Alteon Inc. believed it had addressed a dilemma that had perplexed biochemists for twenty years: how to reverse a specific type of chemical damage that accumulates in aged proteins. Its compound, alagebrium (ALT-711), demonstrated promising results in a small open-label pilot — patients with stiffened hearts experienced measurable enhancements in how effectively the heart’s principal chamber filled with blood between beats. However, a subsequent larger placebo-controlled trial discovered no significant improvements in exercise tolerance or cardiac function, obscuring that initial positive indication. A follow-up study was halted before completion due to the company’s financial downfall, and the drug never attained approval — raising questions about how much of alagebrium’s narrative was due to inadequate chemistry and how much stemmed from financial struggles.
A recent article in Nature Communications, released on 14 July 2026, represents the most significant effort to tackle this issue since the fall of alagebrium. It is essential to consider this new finding in light of that history, as the two failure modes encountered in this field — the biology failing and the business failing to withstand clinical challenges — are distinctly different issues, with only one of them being addressed here.
The damage nobody could reverse
The focus in both scenarios is a class of molecules referred to as advanced glycation end-products, or AGEs — created when sugar interacts with long-lasting proteins in a gradual process that chemists often liken to bread toasting, except it occurs inside the body over years instead of minutes in an oven. Once these cross-links develop, there is no known biological mechanism that eliminates them. This assumption — permanent, unidirectional, irreparable — was what alagebrium attempted to counter chemically, and what the new study is endeavoring to challenge using an engineered enzyme.
The new research, led by Aaron Cravens at Revel Pharmaceuticals alongside collaborators from Calico Life Sciences and the University of Colorado Anschutz Medical Campus, developed an enzyme named CMLase, targeting a specific AGE known as Nε-carboxymethyl-lysine, or CML. Starting with a bacterial glycine oxidase, the team screened over 500 million variants through directed evolution before arriving at a version capable of recognizing and dismantling CML specifically, tested on tissues donated by deceased individuals: skin, an artery, and lens material. In donated skin, the intervention reduced CML levels by more than fifty-five percent — sufficient to lower the tissue below levels typically observed in a 31-year-old. In the abdominal aorta of a 75-year-old donor, overnight treatment decreased CML by over seventy percent. In lens tissue from a 64-year-old donor, reductions ranged from 45 to 78 percent depending on the measurement technique.
What’s genuinely different this time
The authors of the paper explicitly reference alagebrium’s past, which is significant — they could have easily chosen to overlook it. Implicitly, they argue that CMLase fails differently than alagebrium did, should it fail at all.
Alagebrium was a small molecule that chemically disrupted cross-links indiscriminately wherever it identified the right bond — a blunt instrument that worked, to the extent it worked, in a somewhat random manner. CMLase, in contrast, is an engineered enzyme with a singular, specific target: it does not affect glucosepane, another AGE potentially more significant as the primary cross-linker in aged collagen, at all. This specificity marks a true advancement in precision. However, it does not inherently safeguard against the failure mode that ultimately led to alagebrium’s demise — financial issues rather than chemical ones. Revel is a private entity that has previously announced a $3.8 million NIH grant and a $12 million seed round; the paper does not outline a clinical timeline, and the gap between “this works in a dish” and “this endures the years and funding a clinical program demands” is the very gap alagebrium never bridged.
Alagebrium did not have to confront a second, emerging failure mode, as it was a small molecule rather than a protein: CMLase is derived from bacteria, and the paper hints — without providing a resolution — at the potential for a person’s immune system to react to it during repeated dosing. This presents a biological risk, as opposed to a financial one — a challenge that alagebrium’s chemistry never had to contend with.
The background this is set against
None of this implies that CMLase is doomed to a similar fate. It indicates that the pertinent comparison isn’t “will the chemistry work” — initial indications on that front are more targeted and mechanistically clearer than those presented by alagebrium. The relevant comparison is: this represents a niche field with essentially one previous serious attempt, and that attempt is held in memory more for the fact that it never received the backing to determine if the science was sound than for whether the science itself was upheld. One pharmacologist, responding to the new paper