Meet the finalists looking to take on cancer's toughest challenges

There is a huge amount of scientific information about cancer but it is almost impossible for a human researcher to read, retain and process that volume of information. This, however, is not the case for computer and artificial intelligence (AI). The Biologia Ex Machina team aims to harness the power of AI and develop AI-co-scientists – software systems that use AI that can work semi-independently to process huge amounts of information while collaborating with human researchers.

These AI-co-scientists will generate theories about what drives cancer development and growth. The co-scientists will also be able to trouble-shoot to refine these theories. And they won’t just generate and test one theory at a time; they will be capable of rapidly generating thousands of ideas and testing them at scale. Biologia Ex Machina could significantly accelerate cancer research productivity and transform how cancer research is conducted. 

Biologia Ex Machina’s AI-co-scientists could ultimately allow the identification of new treatment targets across many tumour types, which have thus far remained hidden using conventional methods of discovery. Critically, the team’s work aims to accelerate their translation into the clinic to benefit patients. 

The SENTINEL team aims to develop a human–AI co-laboratory, with the capability to create a virtual model of how colorectal cancer starts, to understand the earliest changes to normal cells that may cause cancer to develop. The human–AI co-laboratory will be able to predict which cells are likely to become cancerous, and, critically, explain why.  

The team will focus its work on two groups of individuals who are high-risk for developing colorectal cancer: those with an inherited condition and obese individuals. This will allow the team to investigate the contributions of both genetics and diet, or nature versus nurture, in early colorectal cancer progression. The team has the ambitious long-term aim of reducing the incidence of colorectal cancer, by intercepting the disease before it can take hold.

The team’s learnings could have an impact beyond colorectal cancer, with insights relevant to other cancer types and general prevention strategies. This team plans to provide a roadmap for building a mechanistic understanding of how cancer develops through true human-AI collaboration.

The Cancer Antibody Atlas team plans to address the challenge by exploring the role of immune-modulating autoantibodies in cancer resistance. Imagine your body is a factory that builds and protects itself. Antibodies are like security guards. Their job is to patrol the factory and attack intruders (like viruses and bacteria). An autoantibody is a confused security guard who starts attacking the workers or machinery inside the factory instead of invaders. The guard thinks a normal part of the factory is dangerous—even though it’s not. 

The team plans to build on pioneering work from its research team in identifying the presence of these autoantibodies in people with the most severe, often life-threatening, COVID-19 disease. These autoantibodies blocked important signals from the immune system preventing it from responding to the virus, i.e. the security guards attacked other members of the security team, allowing intruders to slip past and cause chaos. 

Autoantibodies aren’t always harmful. In some cases, these security guards can also attack cancer cells, if they can recognise these traitors among the normal cells and remove them before they can grow or spread. But these security guards aren’t always successful and this anti-cancer protection is not yet understood. 

Expanding on its ground-breaking work during the pandemic, the team hypothesise that these autoantibodies may exist in cancer and could help to explain why some high-risk individuals don’t get cancer versus those that do.

The dark proteome is a group of mysterious and largely unknown proteins that exist in our cells. Unlike most proteins, which have a single, stable 3D shape, proteins in the dark proteome are either unstructured and highly flexible, or they are unusual in a way that makes them difficult to study. This has made them invisible to many traditional research techniques, like the dark matter of the protein universe.

Team DARK-MATTERS aims to decode and harness the dark proteome in cancer. The team will uncover how common dark proteins are in cancer, where they originate from and what they do. To achieve this, DARK-MATTERS will create an atlas using state-of-the-art experimental tools to allow unprecedented identification of dark proteins that are found in cancers. In parallel, the team will determine how important the dark proteome is for tumour growth, fitness and therapy resistance. 

The team plans to go on to determine whether the cancer dark proteome produces dark ‘antigens’. An antigen is something the immune system sees as a threat and reacts to — like a name tag that helps it spot cancer cells and attack them. The immune system uses T cells as security guards to recognise these name tags, but they need specific receptors to see specific antigens, like the right pair of glasses to read the name tag. The team will use cutting edge approaches to match dark antigens with the T-cell receptors that recognise them, identifying which glasses can see which name tags. 

DARK-MATTERS aims to then develop strategies to target dark antigens, training the immune system so that it can recognise and eliminate cancer cells, providing it with the correct glasses for the job. This work could lead to next-generation immunotherapies for patients with limited treatment options.

The ILLUMINE team aims to unlock the treatment potential of the cancer dark proteome (a group of mysterious and largely unknown proteins that exist in our cells) by comprehensively mapping and characterising its function. The team plans to focus on hard-to-treat cancers, creating an unprecedented atlas of the cancer dark proteome. It plans to use cutting-edge experimental techniques to uncover where dark proteins come from, how they are made, and what they do.

The team will also ask whether the cancer dark proteome produces dark ‘antigens’.  An antigen is something the immune system sees as a threat and reacts to — like a name tag that helps it spot cancer cells and attack them. The immune system uses T cells as security guards to recognise these name tags, but they need specific receptors to see specific antigens, or the right pair of glasses, to read the name tag. The team will go on to identify the T-cell receptors in our bodies that can recognise dark antigens, identifying which glasses can see which name tags. 

Capitalising on its unique expertise the team will leverage and integrate cutting-edge experimental approaches to uncover hidden layers of cancer cell biology. ILLUMINE aims to ultimately identify new tumour antigens that are potentially found across cancers and develop innovative immunotherapies for hard-to-treat cancers, transforming patient outcomes.  

DNA is like an instruction manual for our cells, telling them how to grow and function. But over time, DNA can be damaged by harmful substances in the environment (like pollution, tobacco smoke, or UV rays from the sun) as well as by natural mistakes that happen inside our bodies. This damage, called mutations, can sometimes lead to cancer. Each mutation leaves behind a unique pattern of DNA damage, known as a mutational signature.

These signatures act like forensic clues, helping scientists figure out what caused the DNA damage in the first place. Imagine an autograph on a piece of paper, if you don’t recognise the name, it’s just a scribble. But if you can identify the person who signed it, it tells you something important. In the same way, mutational signatures can reveal what has caused  damage to our DNA. So far, scientists have identified different mutational signatures, but for many of them, we still don’t know what caused the damage. Without this knowledge, we can’t fully understand what triggers cancer, or how to prevent it.  

The CAUSE team aims to fill in the biological gaps between the autograph (the mutational signature) and its author (the source), thereby identifying the DNA adducts, or the pen or pencil in our analogy. The team will systematically identify and characterise DNA adducts (or DNA pieces) to understand the process by which insults – external exposures or internal processes - leave a mutational signature on DNA. Identifying all the possible pens that can sign autographs, and how they do it,  will help reveal the author. 

This would provide actionable insights for real-world impact, potentially transforming cancer prevention and our understanding of how cancer starts.

Growing evidence suggests that communication between the nervous system and cancer plays an important role in how the disease develops and spreads. But we still don’t fully understand how nerves and cancer cells communicate or how this interaction affects cancer progression. The CoNNECTED team’s theory is that nerves across the body may influence how cancers initiate, grow, spread and respond to treatment.

To fully unleash the therapeutic potential of cancer neuroscience, the CoNNECTED team aims to go beyond the tumour microenvironment, the complex supportive network in which the tumour resides, to map whole-body neural circuits involved in cancer. To do this the team will leverage the powerful tools of modern neuroscience and sophisticated imaging to understand how the brain talks to the tumour and the tumour talks to the brain.

The team plans to also study how nerves connect to the immune system and how these nerve-immune interactions regulate cancer progression and immunotherapy. 

The team also aims to understand the high incidence of depression, anxiety and cognitive symptoms in cancer patients and uncover how cancer-induced brain changes may in turn fuel tumour progression. 

The CoNNECTED team aims to push the boundaries of the cancer neuroscience field, laying the foundations for new therapies that target both tumour progression and brain dysfunction, with the goal of improving patient survival as well as quality of life. 

This team aims to approach the challenge by investigating the role of interoception – the ability of the brain to monitor organ states and exert control over the body via the nervous system. The team thinks that interoception could allow the brain to monitor tumour formation and influence cancer progression, via the brain–nervous system–tumour axis.

InteroCANCEption plans to figure out how this axis sends signals back and forth, how the brain talks to the tumour, and the tumour talks to the brain and make detailed maps of these connections.

Looking to the future, InteroCANCEption would also explore whether state-of-the-art devices called neuroprosthetics, a type of brain or nerve implant already used in other conditions, could be adapted to help treat cancer. By changing brain activity in precise ways, these tools could one day slow cancer growth, boost the immune system, or reduce symptoms.  

Team NEUROIMPACT propose that the nervous system plays an active and stage-specific role in shaping cancer. The team thinks that the nervous system may be able to spot early tumours and help the body fight them off. But that’s not the whole story. The team thinks the nervous system may also be hijacked by cancer cells to help them grow, spread to other parts of the body, and avoid being killed by treatments.

The team aims to uncover how the nervous system detects early-stage tumours and initiates anti-tumour responses. Then once a tumour takes hold, the team plans to determine how these same nervous system-tumour interactions change to support tumour progression and spread via brain-mediated mechanisms. 

The team aims to study how the nervous and immune systems act as a surveillance system and how tumour cells sneak past the body’s natural defences. NEUROIMPACT will also study the impact of tumours on circadian rhythms, our internal clock that controls daily patterns of sleep and activity, as well as hormonal systems and stress.

This diverse team aims to uncover how the brain and body together shape cancer, and how those insights could be turned into better outcomes for patients. 

REWIRE-CAN plan to tackle the challenge by turning cancer’s own survival mechanisms against itself, transforming what was an advantage into a disadvantage.

Cancer cells finely tune their signalling levels, selecting for an optimal level of signalling to drive their growth, rather than an extremely high level, which may trigger stress responses. It’s like how one cup of coffee can wake you up and allow you to focus on your work, whereas multiple cups might leave you feeling anxious and unable to concentrate. The team aims to take advantage of this, developing precise signalling modulators to disrupt this finely tuned ‘Goldilocks’ state. The team also plans to rewire treatment-resistant cancer cells to become sensitive ones.

The team will focus on colorectal cancer, the second most common cause of cancer death, with an alarming rise in incidence in young adults. REWIRE-CAN team’s work has the potential to transform treatment options and patient outcomes in colorectal cancer and provide a roadmap for rewiring of cancer cells across tumour types.