Novartis shares drop 10% as its second late-stage drug trial fails

Novartis said Tuesday that del-desiran, its experimental treatment for a rare inherited muscle disorder, failed a late-stage Phase 3 trial, the company's second major late-stage collapse in a matter of days. Shares fell as much as 10% in early Swiss trading.

The trial, called HARBOR, tested del-desiran in myotonic dystrophy type 1, or DM1, a rare inherited condition that causes progressive muscle stiffness and weakness and, for many patients, a steady loss of the ability to relax muscles after using them. The drug missed its primary endpoint, a measure of how quickly patients' hands release after squeezing, known as video hand opening time, or vHOT. The endpoint is unusual outside the DM1 field but central to the disease, because the inability to relax the hand is one of the condition's most disabling daily effects.

The miss is the second within days. Earlier this week, Novartis disclosed a separate late-stage trial failure that, like the DM1 result, raised fresh questions about how reliably the company can refill the revenue it will lose as key patents expire. Together the two flops form the kind of news cycle that turns a single disappointing readout into a larger verdict on a company's pipeline. STAT+ reported both setbacks as evidence of mounting pressure on the company's pipeline and acquisition strategy at a moment when it is navigating a major patent cliff, the wave of patent expirations expected to erode revenue from blockbuster drugs this decade.

Each miss now lands harder on Novartis because of a structural mismatch the company cannot fully control. Its bestselling drugs are losing patent protection on a calendar set by regulators, but the replacements arrive on biology's clock, dictated by whether experimental compounds actually work in late-stage trials. The two are not in sync, and a single late-stage miss now has to carry the weight of multiple missing replacements at once. The del-desiran trial is the latest datapoint in that pattern: the readout the company was leaning on to demonstrate that biology is keeping pace with the calendar came back negative.

Novartis does have a recent win to point to. Last week the company reported positive results for a multiple sclerosis drug, a different therapeutic area but a reminder that the pipeline is not uniformly broken. The contrast is part of what makes the del-desiran failure sting: it is the second miss, not the first, and the MS-drug success sits close enough in memory to underline that this one did not have to fail. For a company whose pitch to investors has been that an acquisition-driven pipeline can keep up with the patent cliff, two misses in a week is harder to absorb than one.

The human cost of the DM1 failure is concrete. There is no approved disease-modifying therapy for myotonic dystrophy type 1, so del-desiran was being tested in a population whose only option today is symptom management. Patients and families in the DM1 community were waiting on a readout that would at least have shown whether the mechanism Novartis was pursuing could work. The trial said it did not.

Del-desiran is an antisense oligonucleotide, a class of RNA-targeted drug designed to interfere with the toxic RNA that drives DM1. The Phase 3 miss is a setback for that mechanism, not necessarily a final verdict on antisense drugs in DM1, but it removes one of the few late-stage candidates in a very small field. Future development will now depend on whether the full HARBOR dataset, including secondary endpoints, points to a subgroup or signal worth pursuing.

For Novartis, the next test of the calendar-vs-biology tension will come from the other late-stage readouts the company has scheduled. Each one now carries more weight, because the market is repricing not just a single drug but the company's ability to keep replacing what it loses. Watch the next two scheduled readouts: if those trials also miss, the "pipeline problem" framing will harden; if one or both succeed, the del-desiran miss will look more like a one-drug problem inside a still-functional portfolio.