drug · intermediate · partly written
TGN1412 (anti-CD28 superagonist antibody)
TGN1412 is an anti-CD28 antibody that in a 2006 first-in-human trial caused a life-threatening cytokine storm in all six healthy volunteers who received it, though animal safety testing had not predicted it.
Also called: TGN1412
This describes published medicine for readers who want to understand how it works and what happened. It is not medical advice and says nothing about what any person should take.
The short version
TGN1412 was an antibody that switches T cells on by pressing CD28. In March 2006 six healthy volunteers received a single infusion. Within 90 minutes all six had a systemic inflammatory response, within 12 to 16 hours they were critically ill, and all survived. Animal safety testing had not predicted this. 12
Reported, not yet replicated
What happened more
The clinical report describes six healthy young male volunteers at a contract research organisation, enrolled in the first phase 1 trial of the drug. Within 90 minutes of a single intravenous dose all six had a systemic inflammatory response with rapid induction of proinflammatory cytokines, headache, muscle pain, nausea, diarrhoea, hypotension and other signs. Within 12 to 16 hours they were critically ill with lung injury, renal failure and disseminated intravascular coagulation, and severe depletion of lymphocytes and monocytes followed within 24 hours. They received intensive care including dialysis, high-dose methylprednisolone and an anti-interleukin-2 receptor antibody. Two developed prolonged shock and lung failure and needed intensive support for 8 and 16 days. All six survived. Only the abstract was opened. 1
Reported, not yet replicated
What the regulators found more
A BMJ news item on the final UK regulator (MHRA) report says the failure could not be blamed on how the drug was made or given, while the report criticised aspects of the trial administration. The Expert Scientific Group final report, as relayed by trade press, said preclinical work did not predict a safe human dose even though regulatory requirements were met, and recommended basing first doses on the minimum anticipated biological effect level (MABEL) rather than the no-observed-adverse-effect level from animals. Both accounts are second-hand; the reports themselves were not opened. 23
Reported, not yet replicated
Why animals missed it more
Eastwood and colleagues found that in humans, IL-2 and IFN-gamma release from CD4 effector memory T cells were key indicators of a TGN1412-type response, and that CD28 is not expressed on the CD4 effector memory T cells of all species used for preclinical safety testing, so TGN1412 cannot stimulate them there. The authors call this explanation likely and almost certainly responsible, which is strong but is evidence from cells, not a direct test in the volunteers. Other antibody drugs that cause reactions act mainly through natural killer cells. 4
Reported, not yet replicated
What stays uncertain deep
The doses, the intended indication and the comparison with the placebo recipients are not in the abstracts opened. The explanation rests on cell experiments; no sources opened show the cytokine storm reproduced in an animal model. 14
Our inference
Modality
Targets
Pathways
Indications
Identifiers
Trial
Phase 1, March 2006, Northwick Park Hospital area, London (as described in the BMJ item)
Mechanism
Delivery route
Must survive
Side effects
Systemic inflammatory response
All six volunteers had it within 90 minutes of a single dose: rapid proinflammatory cytokine induction, headache, myalgias, nausea, diarrhoea, erythema, vasodilatation and hypotension. 1
Reported, not yet replicated
Organ failure
All six were critically ill within 12 to 16 hours, with lung injury, renal failure and disseminated intravascular coagulation. 1
Reported, not yet replicated
Lymphocyte and monocyte depletion
Severe and unexpected depletion within 24 hours in all six. 1
Reported, not yet replicated
Verdict
Outcome
failed
Failure mode
All six recipients developed a systemic inflammatory response within 90 minutes and critical illness within 12 to 16 hours. It was missed because, in the species used for preclinical safety testing, CD28 is not expressed on the CD4 effector memory T cells that respond in humans (Eastwood 2010). The Expert Scientific Group said preclinical work did not predict a safe human dose although requirements were met. 1423
Reported, not yet replicated
Teach this
In one sentence
A 2006 first-in-human trial of an anti-CD28 antibody put all six healthy recipients into a cytokine storm because the animals used for safety testing lacked CD28 on the human cell type that matters. 1423
Reported, not yet replicated
The five-minute version
TGN1412 pressed CD28, a switch on T cells, in a way that activates them directly. Six healthy men received a single infusion. All six had a rapid inflammatory response and then organ failure; all survived with intensive care. Animals had been tested first and gave no warning. Later work showed that the key human T cell subset carries CD28 while the same subset in the animal species used does not, so the animals could not reveal the effect. Regulators later recommended basing first doses on the lowest dose expected to have any biological effect. 1423
Reported, not yet replicated
What a listener will ask
Did anyone die?
Was it a manufacturing error?
The regulator report, as relayed by the BMJ, said the cause was not how the drug was made or given.
Why did animals not show it?
The common wrong idea
If an antibody is safe at the animal no-effect dose it is safe in the first human, because antibodies are exquisitely specific.
Specificity is to the human protein and the human cell type; if the test animal lacks that cell-surface switch, the animal test is silent, as here.
The one citation to hand over: 4
Still open here
- The doses given to the volunteers and to animals, and any safety margin, were not in the sources opened.
- Full text of the clinical report, the MHRA investigation report and the Expert Scientific Group report were not opened.
- The intended indication is not stated in the sources opened.
Not yet written: Assays used, Off targets.
Linked from
Sources
- Suntharalingam G, et al.. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med 355:1018-1028 (2006). paper; opened 2026-11-17 all records citing this
- Day M. Agency criticises drug trial. BMJ 332:1290 (2006). news, not peer reviewed; opened 2026-11-17 all records citing this
- Final ESG report recommends new safeguards for first-in-man trials. PharmaTimes (2006). news, not peer reviewed; opened 2026-11-17 all records citing this
- Eastwood D, Findlay L, Poole S, et al.. Monoclonal antibody TGN1412 trial failure explained by species differences in CD28 expression on CD4+ effector memory T-cells. Br J Pharmacol 161:512-526 (2010). paper; opened 2026-11-17 all records citing this
History of this record (5)
This record has been changed 5 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.
- Year
- 2006
- Regulatory status
- failed-in-trials
- Status
- partly written
- Last updated
- 2026-11-19
- Record
- content/drugs/tgn1412.json