Jun 28, 2022

Case Study: Target discovery via a reverse immunology approach

What if we could use the immune system itself to help us identify new therapeutic targets? At Absci we are merging #AI and #syntheticbiology to do just that. Learn more with VP of Immunology Daniele Biasci in this short video.


Daniele Biasci:

Hi everyone, I’m Daniele Biasci and I am VP of Immunology at Absci. Today I want to give you a quick presentation and give you a flavor of the antibody and target discovery program we are currently working on.  

Daniele Biasci:

So, the problem we are trying to solve is the following: if you look at all the therapeutic antibodies approved by the FDA at the moment, about 40% of them, so a very large fraction, focus on a relatively small number of drug targets. Now, this phenomenon can have multiple explanations and for sure, there are historical reasons for that. But one thing is clear, finding new therapeutic targets in a systematic manner is still an unsolved problem.  

And while there have been a lot of promises around genomics, and they use a large dataset to find new therapeutic targets. Most of the promises of those approaches still remain up to feel. We’ve tried to develop a different approach to target discovery which I’m going to highlight briefly in the next few slides. 

Daniele Biasci:

We essentially attempt to ask the immune system to help us identify new therapeutic targets. In order to do that, we start from the observation that inside tumors, inside human tumors, you have the formation of ectopic germinal centers. Now, those are essentially biological machines which are evolved to select and identify high affinity antibodies, presumably against the local antigen in the tumor microenvironment. The presence of this structure has been known for quite a few years but the exact nature of the antibody producing the ectopic germinal centers inside the tumor is still not completely known. Now, germinal centers are exceptional biological immunological structures. The reaction that happens inside the germinal center, the so-called germinal center reaction, can be considered the apex of adaptive immunity. 

And it’s a remarkable biological phenomenon where essentially B cells that have encountered the antigen can migrate in the dark zone of the germinal center. Here they’re going to mutate their B cell receptor and then by a mechanism that is called somatic epimutation. When this has happened enough, they migrate to the light zone of the germinal center when essentially, they attempt to use their B cell receptor to steal pockets of antigen from the follicular dendritic cell, which present it on their arms, after which they internalize the antigen, process it, present it and then compete for very limited T follicular helpers provided by TFH. 

The successful clones of B cells compete successfully for T follicular help. They are licensed to become intra memory B cells or plasma blast and then eventually plasma cells, which are one of the terminal stages of differentiation of B cells and is essentially a stage where the B cell becomes a cell able to produce a very large amounts of antibodies, is essentially an antibody factory. Now, the B cell clones cannot compete successfully for the antigen binding, can go back to the dark zone where they mutate some more and then enter back to the light zone in the competition and in this way when this cycle has happened a few times, you have selected for B cells that may produce eventually antibodies with very high affinity for the target antigen.  

Now, because this reaction happened inside the tumor, it’s very likely that the germinal center inside the tumor selects high affinity antibodies against cancer antigen and helps the intratumoral immune response. Here on the left you see the example of germinal center highlighted by expression C 21, which is a germinal center marker and although without and other types of B cells the image you see here, it’s a relatively new technique which is essentially allowed to see the expression of multiple genes on in the context of the sample slide. In this case, a breast cancer tumor is a slide of breast cancer tissue and it’s called spatial transcriptomics. So, we can see a gene expression overlaid on the special context of the original tissue now because you have intratumoral germinal centers the core idea of the antibody program and target discovery program we are performing at Absci is why we can’t just use the antibodies produced by those ectopic germinal centers to identify the antigens which the immune system actually think are important to target in order to trigger as a clinically successful response against the tumor. 

Now, before we can do that, we need to substantiate the evidence that makes us think that presence of intratumoral germinal centers and the antibodies produced by the structure are actually important in the immune response against the tumor. And there are a few lines of evidence which I’m going to highlight briefly.  

Daniele Biasci:

So the first line of evidence is that at least since 2015, and probably a bit before, it’s known that in presence of intratumoral plasma cells those cells that produce very large amounts of antibodies are associated with much longer survival, more so in some cancer types such as breast cancer or lung carcinoma then the cells that usually we think is associated with better survival in cancer which are severely T cells which have a killing effect on cancer cells. So, the presence of intratumoral antibody producing cells seems to be associated with longer survival in those patients, quite to a large degree.  

More recently in 2020 the presence of tertiary lymphoid structures which are the immunological structures that contain ectopic germinal centers inside tumors have been associated quite compellingly with response to immune point inhibitors and with the subsequent survival that was a paper published in Nature in 2020. And was accompanied by two more papers that essentially show a similar thing across different indications. And so, it seems that the presence of antibody-secreting cells inside the tumor is very important both for survival and for response to immune stimulating therapy.  

That’s of course, just an association, but let’s line these up with additional evidence which is available to us. 

Daniele Biasci:

More compelling evidence came recently in 2022 in March when essentially people started realizing what is behind the association between antibody secreting cells and better outcomes and start disentangling the mechanism of it. So, this paper shows many different observations but the one core observation that is done in this small cohort of patient with ovarian carcinoma is that not only the presence of antibody secreting cells inside the tumor is associated both with disease free survival and with ovarian survival but the actual number of cancer cells to which endogenous antibodies are bound.  

So, the presence of the ability of the endogenous antibodies to bind cancer cells on their surface is also associated with the disease-free survival and overall survival in this small but well characterized cohort of patients. 

This provides an additional level of mechanistic evidence that antibodies, that the presence of antibodies, that combined onto the cancer cells seemed to be associated with longer survival. And I want in this presentation to list all the author’s lines of evidence, but I will leave some papers in the comment section. So, the folks that are interested can dig a bit deeper into the biological rationale, behind the antibody and target discovery program at Absci. 

So, let’s assume that we are correct and that intratumoral B cells, plasma cells, secret antibodies which are important to fight cancer proliferation, and you know, trigger a productive immune response against the tumor.  

Daniele Biasci:

Then what we do at Absci, it’s quite peculiar. For instance, tumor samples from people that have exceptional immune responses, such as people that respond to mutual point inhibitors, we get to select all the antibody encoding RNA sequencing reads from those patients from those tumors. We then assemble them computationally in the case of bulk RNA sequencing. In other circumstances we could be using a single cell depending on what’s available. Then we computationally construct the full sequence of the original antibody. And we tend to focus on the ones that are mostly clonal, because if you remember what I said about the germinal center reaction before, only the very high affinity antibodies, the most effective one to compete for antigen binding, they usually get expressed to a very high level because they are a product of plasma cells which express a very large amount of those antibodies and so they are transcribing very often.   

Once we have the sequence of those exceptional antibodies then we produce them in the lab as monoclonal antibodies. After which we have essentially optimized a pipeline to discover what they bind to. Because at this point, we only have the antibody. We screen them against very large collections of human proteins usually covering the entire protein or the entire set of proteins which are expressed on the cell surface.  

There are usually six thousand potential proteins. Once we get a hit, we then confirm it by plasma surface plasmon resonance or the bilayer interferometry, so we get the full kinetic of the binding and I am happy to report that we have been doing this for quite a few years, actually before, somewhere so much evidence was available regarding the importance of intratumoral antibodies. We have now constructed a pretty large library of antibodies and antigen derived from cancer patients. And actually, I’m happy to report that when they do bind, they bind with very high affinities with KDs in the learning model range which is what you would predict from what we know about the mechanics of germinal center reaction. Once we have those antibodies, those are fully human derived different patients. And we use them as the basis for the development of therapeutic antibodies in different formats. And the original point to this program is that it’s one of the very first approaches to discover at the same time, the target antigen and a fully human ready-to-go antibody against it. So, we essentially use a reverse translational immunology approach to discover new therapeutic antibodies and targets at the same time. 

Daniele Biasci:

I am also happy to report that we have published our first findings from the first four years, essentially, for this very long running project, which is in bioRxiv currently under review. And if you screen the QR code in this slide you’ll be able to access the paper and see some of the exciting results that this program has already generated. Thank you very much for your attention and I’ll leave some more references in the comment section. Please feel free to contact Absci or me anytime if you are interested in this program or if you want to discuss this further. 


• Rakocevic, G., et al. “The landscape of high-affinity human antibodies against intratumoral antigens.” bioRxiv, 2021, (under review).

• Gentles, Andrew J., et al. “The prognostic landscape of genes and infiltrating immune cells across human cancers.” Nature Medicine, (21)8, 2015, 938-945.

• Helmink, Beth A., et al. “B cells and tertiary lymphoid structures promote immunotherapy response.” Nature (577)7791, 2020, 549-555.

• Mazor, Roei D., et al. “Tumor-reactive antibodies evolve from non-binding and autoreactive precursors.” Cell (185)7, 2022, 1208-1222.​​