The Antibiotic That Sat in a Drawer for Six Years

Eli Lilly stopped developing daptomycin because the clinical programme of the late 1980s and early 1990s had not been convincing, and because skeletal muscle was the tissue the molecule most readily affected. In 1997 a young Massachusetts company, Cubist Pharmaceuticals, licensed the worldwide rights, at a time when few firms were working on antibiotics at all, and made a discovery that turned the problem round: the muscle effects depended heavily on how the doses were spaced 12.

The title of this piece counts six years, from the early 1990s to the licence. Sources do not give the same date for the moment Lilly stopped, and none of those consulted for this article gives it as a single day, so the figure should be read as an approximation. The shape of the story does not depend on it. A drug was not disproved; it was left, and then someone reopened the file.

A soil sample and a code number

Like several antibiotics of its era, daptomycin began in soil. The producing organism was an actinomycete, Streptomyces roseosporus, and the early work at Lilly identified a family of related compounds from its culture, named the A21978C complex, whose chemistry was reported in the Journal of Antibiotics in 1987 3. One review of the drug's later career is titled 'from the mountain to the clinic', a nod to the place from which the original sample is said to have come 2. This piece takes the organism and the company from the primary literature and leaves the geography of the sample to those who have documented it.

The molecule picked from the complex was coded LY 146032. It is a cyclic peptide with a fatty-acid tail, which is what the word lipopeptide describes. Both parts matter to what follows: the peptide ring gives the molecule its shape, and the lipid tail allows it to insert into the membrane of a bacterium 4.

LY 146032 and the trials of the 1980s

A retrospective review published in 2000 sets out what Lilly did. Over the 1980s and into the early 1990s the company ran nineteen phase 1 studies and two phase 2 studies, in more than 370 people in all. In the trial settings of skin and soft-tissue infection and of bacteria in the bloodstream, the results were described as encouraging, and the endocarditis data hinted at activity at higher doses 1.

Encouraging is not the same as sufficient. The review records that Lilly was not satisfied with the overall clinical results of the regimen it had used, which was a twice-daily schedule 1. Evidence of this kind is from human studies of mixed design, many of them small, and the company's decision was a judgement about whether the programme justified further investment, not a verdict on the molecule.

The muscle signal, and a decision to stop

The other half of the reason was safety. In animal studies, skeletal muscle was the most sensitive tissue. The damage was mild and reversible, but it was consistent, and at higher doses, around four times higher, degeneration of axons in nerves appeared as well 1. In human volunteers, muscle events were seen only at the highest multiple-dose level that was tested, which was given twice daily 1.

To a company weighing the cost of a large phase 3 trial against a lukewarm efficacy signal, a muscle finding in animals and at the top dose in people is a reason to stop. The detail that matters is in the primary sources: the schedule that produced the effects was twice daily, and nobody had yet tried a different one.

An industry leaving antibiotics behind

The decision was also made against a backdrop. Through the 1990s, large pharmaceutical firms moved away from antibiotics. A new antibiotic is taken for days, not years, and must be held in reserve if it is to remain useful, which is a poor commercial profile next to a drug for a chronic condition. Many companies closed or shrank their programmes. Bacterial resistance, in the meantime, was spreading in hospitals, and the shortage of new agents was beginning to be noticed 1.

The consequence for a molecule like daptomycin was a kind of administrative limbo. It had not failed in the sense of being shown ineffective, and it had not succeeded in the sense of earning a market. The rights, the data and the production know-how sat inside a company whose priorities lay elsewhere, and a file in that condition is easy to forget. This is the situation the title has in mind: the drawer was a corporate one, and nothing about the molecule had changed when it was opened again.

It is a pattern that recurs in the history of drugs of every class, and the sister article on the decades when peptides fell out of fashion describes the wider climate in which such decisions were made. A cyclic lipopeptide that needed an injection, with an unfamiliar safety signal and a use limited to hospitals, was the kind of project a cautious portfolio committee would defer.

Cubist, 1997: the licence nobody else wanted

In 1997 Cubist licensed worldwide rights to daptomycin from Lilly for the development of a treatment for hospitalised patients with serious Gram-positive infections 1. Cubist was a young company with no marketed product, and the molecule it acquired came with a reputation that had two halves: it worked in the laboratory against resistant bacteria, and it hurt muscle.

The 2010 review whose title names Francis Tally credits him with essential help in carrying the drug from the shelf to the clinic 2. Tally was an infectious-disease physician with long experience of antibiotic development, and the story told about the rescue is partly a story of one person's persistence. It is worth saying that the papers consulted for this piece establish the contribution, not its size.

A licence of this kind is a bet by both sides. For Lilly, it was a way of recovering something from a programme it no longer wished to fund. For Cubist, it was the bulk of what the company had, and the success or failure of the molecule would decide whether the company survived. The 2000 review, written while the work was still in progress, is notable for its tone: it sets out the old data and the new plan side by side, and does not yet know the ending 1.

The dosing-interval idea

The detail that changed the project was already in the old data. The animal studies showed that the muscle effects were worse when a given total daily dose was divided into several portions than when it was given once. A lipopeptide that disappears from the blood slowly and accumulates in tissue over the day, then, might be better tolerated as a single daily exposure 1.

From March 1999 Cubist ran a phase 2 study in bacteraemia to test once-daily regimens against the older twice-daily one, and its phase 3 studies in skin and soft-tissue infection used a once-daily schedule chosen on the strength of the preclinical data 1. This is the sense in which the molecule was rescued by a schedule: the chemistry that Lilly shelved was unchanged.

2003 and after: approval, resistance, acquisition

The US regulator approved daptomycin in September 2003, for adults, and Cubist marketed it as Cubicin. The drug was by then one of very few new antibiotics aimed at resistant Gram-positive organisms, and a 2005 review of its properties describes it as a lipopeptide for serious infections with such bacteria 4.

As with other antibiotics, reduced susceptibility has been described in the years since approval. In December 2014 the pharmaceutical group Merck agreed to acquire Cubist for about $9.5 billion, a price that suggests how far the small company with a borrowed molecule had come from the licence of 1997.

What a lipopeptide is, and where the pharmacology articles start

A lipopeptide is a peptide joined to a fatty acid. In daptomycin the fatty-acid tail anchors the molecule in a bacterial membrane, and the action reported for it depends on calcium, which is why the drug is described as acting on the cell membrane rather than on the cell wall 4. The structure of the A21978C family was first worked out from mass-spectral data and reported in 1987 3.

The mechanism and the pharmacology are matters for the technical articles of this network. The point that belongs here is the business and human point: a company can be right to stop a programme on the evidence it has, and the evidence can still be incomplete.

In one sentence

Daptomycin was abandoned over a muscle signal and a twice-daily schedule, and rescued when a small company asked whether the schedule, not the molecule, was the problem 12.