Engineering Viruses to Save a Teenager

Isabelle Carnell-Holdaway was fifteen years old and running out of options. Born with cystic fibrosis, she had lived for eight years with a chronic infection of Mycobacterium abscessus, a notoriously drug-resistant relative of the tuberculosis bacterium. In 2017 she underwent a bilateral lung transplant at Great Ormond Street Hospital in London. The new lungs gave her a chance — but the immunosuppression required to prevent rejection allowed the mycobacterium to roar back and spread. The infection disseminated: her surgical wound would not heal, infected nodules erupted across her skin, and lesions appeared on her liver. Conventional antibiotics had failed. Her clinicians estimated her chance of survival at roughly one percent and moved toward palliative care.

Her mother pressed for any remaining possibility, and a doctor reached out to Graham Hatfull at the University of Pittsburgh — a microbiologist who, with an army of student contributors, had spent decades assembling one of the world's largest collections of bacteriophages, numbering over 10,000 by then. The challenge was formidable. Mycobacteria are slow-growing and hard to lyse, and the phages in Hatfull's freezers had been collected and characterized largely to study basic viral genetics, not to cure patients. Most had never been tested against M. abscessus, let alone Isabelle's specific clinical isolate.

Hatfull's lab screened their collection against the patient's strain. They found only a handful of promising candidates — and most were imperfect. One phage, named Muddy, killed the bacterium efficiently in its natural form. But two others, Mycobacteriophage BPs and ZoeJ, were 'temperate': rather than reliably bursting their host, they could integrate into the bacterial genome and lie dormant, useless for therapy. So the team did something unprecedented in a clinical setting. Using genetic engineering, they deleted the repressor gene responsible for the dormant lifestyle, converting the temperate phages into obligately lytic killers. The result was a three-phage cocktail — one natural, two engineered — tailored to a single patient.

After purification and safety testing, intravenous treatment with the cocktail began in June 2018, twice daily, alongside topical application to skin lesions. The team watched anxiously for an immune reaction or signs of failure. Instead, the response was steady and encouraging. The therapy was well tolerated. Within weeks her surgical wound began to close. Her liver function improved. The disseminated skin nodules — there had been more than twenty — gradually resolved. Over months of continued treatment, Isabelle returned to a near-normal life.

The case was published in Nature Medicine in May 2019 by Rebekah Dedrick, Hatfull, and an international team including her London clinicians. It marked the first reported therapeutic use of genetically engineered bacteriophages in a human being. The achievement was significant on multiple fronts. It showed that phages could be engineered to order to overcome a specific pathogen's defenses. It extended phage therapy to mycobacteria, among the hardest bacterial targets. And it underscored the deeply personalized nature of the approach: this exact cocktail was built for one patient's one strain.

The authors were careful about what a single case can and cannot establish — recovery in one immunosuppressed transplant patient receiving concurrent antibiotics is not proof of efficacy. But the result was striking enough to galvanize the field and to seed larger efforts. Hatfull's group went on to treat additional mycobacterial patients on a compassionate-use basis, accumulating a case series and informing the design of formal clinical trials. For Isabelle Carnell-Holdaway, the engineered viruses meant survival. For the field, they meant a new frontier: phages were no longer only found in nature — they could be designed.