TGN1412 – a clinical trials disaster

In 2006, six volunteers signed up to a clinical trial at Northwick Park Hospital, testing the new drug TGN1412.
Normally an early-stage clinical trial is straightforward, checking the safety of a new treatment in a group of healthy individuals before it’s used in longer, larger studies. But this time, the trial didn’t go to plan, and within hours of receiving the drug all the volunteers were in intensive care.

As a centre of the Medicines and Healthcare products Regulatory Agency (MHRA), we played a key role in uncovering what went wrong. We review what happened and how our findings helped to transform clinical trials, making them safer for volunteers taking part every year. 

Calm before the storm

TGN1412 was developed as a potential therapy for leukaemia and autoimmune conditions such as rheumatoid arthritis. It works by binding to a structure known as the CD28 receptor, that sits on the surface of white blood cells called regulatory T-cells. When the drug binds to the receptor on the surface of these cells, it activates the cell helping them to carry out their job working to controlling immune responses.

The drug had been proven safe in pre-clinical studies, however when healthy volunteers received the drug, they experienced a severe reaction. Many suffered from multiple organ failure, with two participants even being diagnosed as being in a critical condition. Doctors identified that the participants had experienced a ‘cytokine storm’- where the immune system goes into overdrive producing high levels of molecules known as cytokines.

This reaction had never been seen in a clinical trial and no-one understood exactly what had happened. 

Unearthing the answers 

Once doctors were aware of the drug’s effects, the MHRA immediately became involved to check the quality of the drug and confirm it was the same as the one used in earlier studies. They immediately removed the drug from the trial and sent it on to researchers at NIBSC who quickly confirmed that this batch of TGN1412 was no different from that used in pre-clinical studies.

NIBSC scientists soon realised that if the problem didn’t lie with the quality of the drug, the early testing could be to blame. They found that changing the way that TGN1412 was added to human blood cells in the lab altered their response. If they added the drug directly to cells, nothing happened, but if they put the drug onto the dish the cells grow on instead- they observed a dramatic response. So, the way that these early cell studies were carried out meant researchers were unable to predict the devastating side effects observed in humans.

But new drugs are also tested in animals before human clinical trials- so why weren’t the dangerous side effects identified at this stage? Looking closely at the responses of cells, the team found that monkey cells just don’t respond to TGN1412 in the same way as humans.

"We know that the monkey and human T-cell receptors are virtually identical. But despite the fact the monkey T-cell binds TGN1412 and has a very similar receptor to humans- it does not respond to the TGN1412 stimulus in the way the human T-cell does."

Dr Adrian Bristow, Former head Division of Biotherapeutics, NIBSC

This meant that for TGN1412 the monkey studies just couldn’t predict the severe reactions that were observed in humans or evaluate safe doses of the drug for use in the trial.

Making clinical trials safer

The findings from the team at NIBSC were vital for developing new European legislation to regulate how a new drug is first given to volunteers as part of a clinical trial.

A new wave of laboratory tests have also been introduced to help researchers predict the likelihood of new medicines causing a similar “cytokine storm” reaction, improving safety of clinical trials. And NIBSC scientists are again at the forefront of this work,  preparing key control materials that help ensure these tests work with the same sensitivity and accuracy wherever they are carried out.