Dexamethasone, a corticosteroid on the World Health Organization’s essential medicines list since the 1970s, reduced 28-day mortality in hospitalized COVID-19 patients receiving respiratory support from 25.7% to 22.9% in the RECOVERY trial, a randomized study of 6,425 patients published by the RECOVERY Collaborative Group in the New England Journal of Medicine in 2021. It is one of the most cited recent examples of drug repurposing, a strategy reviewed in Nature Reviews Drug Discovery in 2019 by Pushpakom and colleagues, in which existing approved drugs are tested for new indications rather than developed from scratch.
The economics favor repurposing when they can
Why repurposing matters
Developing a new drug from initial target identification through regulatory approval takes about 10 to 15 years and, by the industry standard estimate, costs on the order of $1 to $2 billion when the cost of failures is included. The attrition is severe: about 90% of candidates entering phase 1 do not reach approval. Repurposing shortens the timeline because early-phase safety, pharmacokinetics, and toxicology data are already available, and often skips or shortens dose-finding trials. Pushpakom and colleagues, reviewing the field in Nature Reviews Drug Discovery, note that repositioned drugs can reach approval at rates several times higher than de novo candidates because the compounds are de-risked before the new-indication trial begins.
The classic examples show both breadth and origin
Sildenafil, developed in the early 1990s as a candidate for angina, produced an unexpected side effect in a UK clinical trial that led to its approval for erectile dysfunction under the brand name Viagra. Thalidomide, withdrawn in the 1960s after causing severe birth defects, has been reintroduced with strict pregnancy prevention programs for multiple myeloma and complications of leprosy, based on its immunomodulatory and anti-angiogenic effects. Minoxidil began as an antihypertensive before its hair-growth effect was identified. Aspirin, developed as an analgesic, is now used at low dose for cardiovascular prevention. Most of these discoveries were serendipitous. The current wave of repurposing tries to systematize what used to be an accident.
Dexamethasone in COVID-19 is the field’s fastest success
When SARS-CoV-2 emerged in early 2020, the pathophysiology of severe disease was recognized as involving damaging immune hyperactivation in the lungs. Corticosteroids, cheap and widely available, were an obvious candidate. The RECOVERY trial, an adaptive platform study run across UK hospitals, randomized 6,425 patients to dexamethasone 6 mg daily for up to 10 days or usual care. The primary result, published in the New England Journal of Medicine in 2021, showed 28-day mortality of 22.9% on dexamethasone against 25.7% for usual care (rate ratio 0.83, 95% confidence interval 0.75 to 0.93). The benefit concentrated among patients on invasive mechanical ventilation, where mortality fell from 41% to 29%. The authors concluded that “the RECOVERY trial provides evidence that treatment with dexamethasone at a dose of 6 mg once daily for up to 10 days reduces 28-day mortality in patients with Covid-19 who are receiving respiratory support.” NHS England estimated the intervention saved approximately one million lives worldwide within a year of the RECOVERY result.
Computational screening changes what gets tested
From serendipity to computation
Traditional repurposing was serendipitous. The current approach applies computational tools to large biomedical datasets: gene expression signatures, protein-protein interaction networks, electronic health records, and mechanistic models of disease. Machine learning systems compare the molecular signature of a drug against the signature of a disease to predict which drugs might reverse the disease state, or which drugs are already reversing symptoms in patients who take them for unrelated reasons. Platforms such as Open Targets and the NIH’s National Center for Advancing Translational Sciences maintain public repositories of candidate matches. The predictions still need to be tested in clinical trials, but the hit rate on the trials that get run is higher than a random guess.
The main barrier is economic, not scientific
The economics problem
Repurposing works best when the drug is off-patent and cheap, and that is also when the pharmaceutical industry has the weakest financial incentive to fund the definitive phase 3 trial needed for a new label. A generic drug with strong signal in a small early trial often stalls because no company will run the pivotal study. The gap is being filled by academic consortia, government funding through agencies including the NIH and its National Center for Advancing Translational Sciences, disease foundations, and patient organizations. Some jurisdictions allow off-label prescribing based on published evidence, but that is not a substitute for regulatory approval when the goal is standard-of-care recognition and consistent access.
What the evidence cannot yet answer
The high-profile successes are visible; the failure rate of computational repurposing screens is not as widely published. Not every predicted match works in patients, and the translation from in silico signal to trial-ready hypothesis has not been rigorously benchmarked. Regulatory pathways for off-patent repositioning remain patchwork across jurisdictions. And whether the strategy can meaningfully reduce the cost of drug development at a system level, or only at the level of individual drugs, is an open question.
References
- Pushpakom S, Iorio F, Eyers PA, et al. Drug repurposing: progress, challenges and recommendations. Nature Reviews Drug Discovery, 2019; 18: 41-58. DOI: 10.1038/nrd.2018.168
- RECOVERY Collaborative Group; Horby P, Lim WS, Emberson JR, et al. Dexamethasone in Hospitalized Patients with Covid-19. New England Journal of Medicine, 2021; 384: 693-704. DOI: 10.1056/NEJMoa2021436
- National Center for Advancing Translational Sciences (NCATS). Therapeutic repurposing program. Available at: ncats.nih.gov
- Open Targets Platform. Drug target identification and repositioning. Available at: opentargets.org