Translational medicine: how a lab discovery reaches the clinic

A mouse with diabetes gets an injection of cells, and its blood sugar comes down. The paper is published, the press release goes out, and patients start asking when they can have it. The honest answer is usually "not for a long time, and possibly never." The space between that experiment and a treatment a doctor can prescribe is what translational medicine is about.

The term covers the work of moving a discovery from the bench to the bedside, and also the reverse: bringing what doctors see in patients back into the lab to ask better questions. It is less glamorous than discovery and much more expensive.

The valley of death

Researchers use the phrase "valley of death" for the gap between an academic finding and a product a company will pay to develop. On one side, government grants fund discovery and reward publication. On the other, investors fund products that already have human data or a clear path to it. In between sits work that neither side is set up to pay for: repeating the key experiments to make sure they hold, making the product consistently, running formal toxicology and writing the regulatory file.

NIH launched the National Center for Advancing Translational Sciences (NCATS) in fiscal year 2012 to work on exactly this gap. At the time, NIH noted that it could take almost 15 years for a discovery to become an approved drug, with a failure rate above 95 percent. NCATS funds the Clinical and Translational Science Awards (CTSA) Program, which supports more than 60 medical research institutions. These hubs offer shared services such as regulatory help, research coordinators, biostatistics and pilot grants, and they train the people who do translational work.

Stage by stage

Preclinical work

Preclinical research happens before any person receives the product. The FDA describes two broad kinds. In vitro work uses cells, tissues or organoids in the lab to show the product does what it is supposed to do and to find a potential dose. In vivo work uses animals to study how the product behaves in a whole body.

Animal models are informative and also misleading in predictable ways. Many mouse models of type 1 diabetes respond to treatments that later fail in people, a pattern described in our stem cells and diabetes explainer. Large animals such as pigs or sheep are often used for cardiac and orthopedic work because their anatomy is closer to ours.

Studies meant to support human testing, particularly toxicology studies, follow good laboratory practice (GLP) rules in 21 CFR Part 58. GLP covers personnel, facilities, equipment, written protocols and quality assurance. The point is that someone else can audit the study and trust the result. For cell therapies, preclinical packages also look at where the cells travel in the body, whether they form tumors and how potency will be measured from batch to batch.

Talking to the FDA early

Developers of cell and gene therapies can request an early, informal INTERACT meeting with the FDA's Office of Therapeutic Products before definitive toxicology studies are designed, and a formal pre-IND meeting later. These meetings exist because expensive preclinical programs sometimes answer the wrong questions.

The IND

Before a new drug or biologic can be tested in people in the US, the sponsor submits an investigational new drug application (IND). It contains the preclinical data, a detailed description of how the product is made and tested (the chemistry, manufacturing and controls section), the clinical protocol and information on the investigators. Under 21 CFR 312.40, the IND takes effect 30 days after the FDA receives it unless the agency places it on clinical hold. A university can be the sponsor; many first-in-human cell therapy trials are run by academic investigators holding their own IND.

Making the product

A treatment cannot be tested unless it can be made the same way every time. Small molecules go to chemistry contract manufacturers. Cell therapies need clean rooms, trained operators and release testing, which is why many academic medical centers built their own GMP cell processing facilities. Hospitals that run stem cell transplant programs already process cells under strict standards, and FACT (the Foundation for the Accreditation of Cellular Therapy) accredits many of them. When a product moves toward larger trials, the work often shifts to a contract development and manufacturing organization (CDMO). Stem cell technology covers isolation, expansion and freezing in more detail.

Clinical trials

Phase 1 trials test safety and dosing in a small group, phase 2 looks for signs of effect and refines the dose, and phase 3 tests the treatment in a larger group against a control. Each stage is registered publicly; our guide on how to read a ClinicalTrials.gov listing shows how to tell what a trial is actually testing.

Hospitals do most of the hands-on trial work: enrolling patients, giving the treatment, collecting samples and reporting side effects. Sponsors, whether a company or a university, design the trial, pay for it and own the data.

Cell therapy after a heart attack shows why phases matter. Early small studies of bone marrow cells reported encouraging changes in heart function, and larger randomized trials mostly found small or no benefit. The details are in cell therapy after heart attack.

Review and approval

When trials are done, the sponsor files a marketing application (an NDA for drugs, a BLA for biologics, including cell therapies). Under the FDA's user fee goals, a new product gets a decision within about 10 months of the 60-day filing date for a standard review, or 6 months for a priority review.

How long each step takes

These are rough ranges, and individual programs vary widely.

The phase sizes and durations come from the FDA's patient page on clinical research. The overall figure comes from the 2021 "Clinical Development Success Rates" report by BIO, Informa Pharma Intelligence and QLS Advisors, which also found that roughly 8 percent of programs entering phase 1 reached approval.

International networks, and clinics that move to avoid oversight

Much translational work is international by design. Large trials enroll patients in many countries under shared Good Clinical Practice standards from the ICH, so that one data set can support approvals in several places. Regulators differ in the routes they offer. Japan introduced a conditional, time-limited approval route for regenerative medicine products in 2014, and the European Union allows a limited "hospital exemption" for some advanced therapies made and used within a single hospital. Both still involve regulators.

Some clinics move for the opposite reason: to sell unproven cell treatments where oversight is weaker, or to market them to travelers from countries where they would not be allowed. The FDA has issued repeated consumer alerts about unapproved regenerative medicine products, including stem cell and exosome products, and the ISSCR warns against paying for stem cell interventions outside regulated trials. The CDC's chapter on medical tourism notes that infections are the most common complications among people who travel for care, and follow-up at home can be hard when records are incomplete.

Common questions

Why do so many promising lab results never reach patients?

Some do not replicate. Some work in animals and not in people. Some cannot be manufactured consistently or affordably. And some work, but not better than existing treatment. Each of these can end a program at a different stage.

Can a hospital make a cell therapy itself?

Hospitals process cells for transplant routinely, under FDA rules and accreditation standards. Making a new, more than minimally manipulated cell product for patients requires an IND for trials and a license for routine use, whoever makes it.

Is a clinic abroad that cites published research offering the same thing?

Not necessarily. A citation shows that someone studied a related idea, not that the clinic's product is the same, made to the same standard or shown to work. The useful questions are whether the treatment is part of a registered trial, what exactly the product is and who inspects the facility. Outside a registered trial, a patient is paying to step outside the system designed to catch problems.