Researchers at Oxford techbio Etcembly hope to find new targets for cancer therapies by analysing the immune cells of cancer survivors, as principal scientist Mathew Jones explains.
Despite all the advances in medicine over the past few decades, curing cancer is still beyond our grasp. Only half of all people diagnosed survive for ten years or more, and survival rates vary widely by cancer type. Furthermore, many of the new treatments that are being fast-tracked through regulatory approval are failing to deliver significant clinical benefit.
Immunotherapies – drugs that harness the power of a patient’s own immune system to kill cancer cells – have huge promise, but they don’t work for everyone and can have significant side effects.
As a result, the search for more effective immunotherapeutics aimed at directing the power of the immune system against cancer cells has intensified. However, traditional methods for discovering suitable targets within tumours have yielded limited success.
Etcembly’s new ETCh study flips the script on target discovery. Rather than going from tumour to target, we’re working backwards from people who have already beaten their disease to identify new targets for next-generation immunotherapies.
The opportunities for TCR-based therapies
T cells and T cell receptor (TCR) therapeutics are a rapidly growing area of oncology, harnessing the power of the immune system to fight cancer. Once activated, these cells rapidly multiply and attack pathogens or cancer cells, as well as trigger other immune cells.
One major breakthrough T cell therapy is CAR-T cell therapy, which uses engineered antibodies to direct T cells to specific targets on cancer cells. But researchers are exploring even more promising approaches with biological drugs based on TCRs themselves.
TCRs recognise protein fragments (peptides) displayed on a cancer cell’s surface by special molecules called HLA complexes, triggering T cell activation. Compared with antibodies, TCRs can recognize a wider range of targets on cancer cells, including those hiding inside the cell.
But to develop new T cell and TCR-based immunotherapies, we need to discover new cancer targets.
Where are the TCR targets?
In the early days of TCR therapeutic research, advances in technology – primarily mass spectrometry – brought the promise of unearthing new targets for biotherapeutics and cell therapies.
TCR companies invested heavily, and amassed mountains of data on potentially promising new targets, but converting these into a reliable source of blockbuster therapeutics has been a challenge.
“The drugs being developed today are still mostly targeting the same old suspects,” Nick Pumphrey, Chief Scientific Officer at Etcembly explains.
“Targets like MAGEA4, MAGEA1, NYESO1 and gp100 have been known about for years, before the advent of mass spectrometry validation.”
While a few new leads have emerged in recent years, these targets remain unvalidated in the clinic. Etcembly’s first program ETC-101 – the world’s first generative AI-designed bispecific T cell engager with picomolar affinity – targets PRAME, one of the very few of those older targets that is highly expressed in a number of cancer types. Beyond this, we have a more ambitious vision.
“Our goal is to develop a pipeline of TCR therapeutics that take aim at new targets within cancer cells,” says Nick. “We don’t want to go about finding these new targets in the same ways that have had limited success before. Instead, we’re trying a different approach.”
Working backwards from TCR to target
Over the last few years, a few approaches have emerged to investigate the T cell repertoires of long-term cancer survivors who are still alive five years after diagnosis.
One of the first studies to do this was led by Professor Andy Sewell from Cardiff University, in collaboration with Professor Inge Marie Svane at the Centre for Cancer Immune Therapy (CCIT) in Denmark.
The researchers extracted T cells from the tumours of melanoma patients, grew them up in the lab and reintroduced them to the patients.
This technique, now known as tumour infiltrating lymphocyte (TIL) therapy, has become a well-established method for treating melanoma and other cancers.
The team then delved deeper, conducting a further study within patients who had received TILs to identify the specific T cells that recognised and attacked the tumours, investigating whether any could potentially target cancers in other patients. They saw reactivities to some traditional melanoma and transplant targets, as well as unexpected novel ones.
This way of working backwards from a T cell to its target is known as ‘de-orphanisation’ – an approach that Etcembly are applying in their new study.
Nick explains, “By coming at the problem from the opposite direction, we can identify TCRs and targets from cancer survivors that have proven their ability to beat cancer. This will allow us to develop therapies that are more likely to work for others.”
The search begins
To do this, we have launched a groundbreaking research study called ETCh, gathering participants who are living with or have survived cancer as well as healthy individuals. Our research team will conduct an extensive analysis of the immune repertoire at an unprecedented scale by sequencing millions of antibodies and TCRs from each individual.
Etcembly’s advanced AI platform, EMLyTM, will then perform an in-depth analysis to identify which of these are likely to play a role in recognising and destroying cancer cells, and determine the aberrant molecules in tumours that they target.
These molecules could become new targets for next-generation immunotherapies that harness the power of a patient’s own immune system to combat cancer.
Our approach involves:
Single-cell sequencing: Etcembly’s in-house proprietary microfluidics technology allows for efficient and cost-effective DNA sequencing of TCR and antibody repertoires, including those that respond to multiple types of cancer cells in lab tests.
Paired TCR analysis: Most databases only contain single TCR chains (alpha or beta), but TCRs function as pairs. Etcembly’s technology isolates and sequences both chains together, providing a complete picture of each TCR.
Computational modelling: EMLy™, Etcembly’s generative AI platform, analyses the immune repertoires of long-term cancer survivors to predict which TCRs hold the greatest potential for binding new targets within tumours.
Laboratory validation: Promising TCRs are made and tested in the lab to see how well they work. The data from these tests is fed back into EMLy to further optimise specificity and affinity.
The future: New targets, new possibilities
This approach has the potential to identify new targets more precisely than conventional target identification methods. And as more patients are sequenced, common themes across different cancers may emerge, which could lead to a wider range of effective therapies for many patients.
“Because we’re focusing on T cells that respond to antigens that are common across multiple cancer types, we might be able to find broadly applicable targets,” says Nick.
Etcembly’s long-term vision is to develop in silico TCR
de-orphanisation. This means using AI to predict a TCR’s target antigen solely based on its underlying DNA sequence. This would revolutionise our understanding of the immune system and rapidly accelerate the development of targeted therapies.
“At Etcembly, we’re starting with baby steps, first, we are focusing on TCR specificity & engineering.” explains Nick.
“We’re modelling interactions and predicting specificities computationally, but there’s some way to go before we can accurately predict a TCR’s specificity from a sequence if we’ve never seen anything like it before.
“This is an ongoing challenge, but we’re making progress. By combining our approach with established methods like peptide-HLA display libraries, we’re gathering valuable data to validate and train our models, steadily paving the way for full in silico TCR de-orphanisation.”
