Artificial antibody guides cancer to immune execution cells

We have developed many different ways of fighting cancer, from chemotherapy to radiotherapy; more recently we’ve begun to harness our own immune systems to attack tumours too. And because the immune system can access all areas and is pre-made to work with our tissues, it’s a logical way forward. To make the process as effective as possible, the same team of scientists from the company BioNtech, which developed one of the Covid vaccines, have now come up with a way to use their mRNA vaccine technology to introduce the genetic instructions for making special modified antibodies - called bispecific antibodies - that can lock onto two things at once: they bind simultaneously onto tumour cells AND grab hold of specific immune cells capable of killing them; by bringing the two together like this, the immune system is stimulated to attack the tumour much more powerfully. The lead author on the study is BioNTech’s Christiane Stadler…

Christiane - Our aim was to get the patient's immune system directly to the site of the tumour by using the technology of bispecific antibodies. Bispecific antibodies are capable of redirecting immune cells to tumour cells.

Chris - Can you just tell us a bit about what a bispecific antibody is? Just paint a picture for us of what one is, what one does, how I would recognise one.

Christiane - Sure. So our bispecific antibodies we are talking about here have a kind of 'y' shape, but that's a little smaller than natural antibodies. And it has two arms binding to a tumour antigen and has one arm that binds to a molecule on T cells. It's called CD3. And this antibody can function like a bridge between tumour cells and T cells.

Chris - It's almost like introducing the victim to its executioner.

Christiane - Exactly. You got it. So, we have everything in our body that we need to kill cancer cells. However, cancer cells are pretty smart and know how to evade the immune system. They're hiding, they're not showing their mutations, so the immune system is kind of blind. Therefore bispecific antibodies are one tool in helping to make these mutated cells visible again and get the cytotoxic T cells directly to the tumour.

Chris - Presumably you've got some kind of molecule or marker on the surface of the cancer cell that the antibody can recognise, which is unique to the cancer.

Christiane - That's correct. So, in our case we use Claudin 6, and Claudin 6 is the surface molecule or antigen of cancer cells. The great thing about Claudin 6 is that it is exclusively expressed on cancer cells, but not on any adult normal tissue apart from the placenta. In humans, Claudin 6 is expressed during embryogenesis, which explains also the expression in the placenta, but then it's completely silenced and absent from adult tissue, which makes it very safe as a target.

Chris - So the rationale here is, we come up with these antibodies, they go around the body, they would lock onto a T cell capable of killing a cancer cell, and they would lock onto a cancer cell, introduce the two together, and this would help the immune system to obliterate the cancer in the body.

Christiane - Exactly. That's the way it works. You bridge the two and T cells get activated and they release a lot of molecules to really kill the cancer cells. And then the T cells proliferate, they become more and more and it's a very highly active system.

Chris - So the key question is, how do you make these antibodies in the first place?

Christiane - This is very different actually in our company. So, 12 years ago we started to implement the RNA technology for antibodies. That's actually how the whole RNA business started at BioNTech. We thought that we could directly inject RNA that carries the genetic information for the therapeutic antibodies into patients so the liver produces the protein. The liver cells would take up the RNA and produce the antibody, release it to the bloodstream, then we could save a lot of time in manufacturing, have a quicker development for patients, and also we could have a longer half life of the bispecific antibodies in the blood because these are small proteins that get eliminated quickly. But since this action of translation in the liver cells goes on for several days, we get a longer sustainable expression of our antibody.

Chris - So you basically have engineered the genetic message for what you want the antibody to recognise on each side of itself - the T-cell and the cancer cell - you put the genetic message in to the bloodstream and liver cells helpfully make those antibodies for you, chuck them into the bloodstream and then they go around the body for a while doing this job.

Christiane - That's the way it works. Yes.

Chris - And does it work?

Christiane - It works. So first we started of course in individual systems, cell cultures, and we saw that cells are able to assemble beautiful antibodies. Then, we went to the mouse for preclinical studies. You always have to use an animal model. And here we intravenously inject the RNA, holding the information for the antibody, and this RNA is packed into lipids to protect it. After intravenous injection, it's delivered to the liver. In the blood we can detect the antibody, but we can also see when we inject human tumours to the mice that these tumours get eliminated. So it's working.

Chris - Are there any risks from doing this? Is it possible you could end up with too much of these antibodies and they could start to cause a runaway immune response or they could prime the immune system to attack something it shouldn't?

Christiane - There's no risk to prime the immune system to attack anything else. However, as always with antibody therapies, it can be too much. So this is carefully investigated by different dose levels: how high can we go? But since we have this tumour exclusive target, Claudin 6, the risk of any adverse events is really, really low. But bispecific antibodies and activating T cells, you always have to be a little careful to not overshoot the cytokine release. This has been carefully monitored but, in our case, the risk is lower than usual.