18
kilkennyobserver.ie
The Kilkenny Observer Friday 10 June 2022
Science & Wellbeing Activating the immune system at the site of a tumour can recruit and stimulate immune cells to destroy tumour cells. A new therapeutic approach prevents the growth of metastatic tumours in mice by forcing cancer cells into a dormant state in which they are unable to proliferate. The study, published in the Journal of Experimental Medicine (JEM), could lead to new treatments that prevent the recurrence or spread of various cancer types, including breast cancer and head and neck squamous cell carcinoma (HNSCC). Many cancer patients relapse, often years or decades after their initial treatment, and develop new tumours that regrow in the same location or metastasise (spread) to other parts of the body. These secondary tumours are often resistant to treatment and are produced by individual tumour cells that may remain dormant for long periods before being reactivated to start proliferating again. Patient relapse might therefore be prevented if researchers could find a way to keep remaining cancer cells in a dormant state. In a previous study, Maria Soledad Sosa from the Icahn School of Medicine at Mount Sinai and Julio A. Aguirre-Ghiso, now at Albert Einstein College of Medicine, discovered that the ability of cancer cells to remain dormant is controlled by a protein called NR2F1. This receptor protein can enter the cell nucleus and turn numerous genes on or off to activate a programme that prevents the cancer cells from proliferating. NR2F1 levels are usually low in primary tumours but are elevated in dormant disseminated cancer cells.
Cancer drug stops tumour growth by putting cancer cells to sleep Levels of the NR2F1 protein then decline once more when cancer cells start proliferating again and form recurrent or metastatic tumours. “We therefore thought that activating NR2F1 using a small molecule could be an attractive clinical strategy to induce cancer cell dormancy and prevent recurrence and metastasis,” AguirreGhiso explains. In the new JEM study, Sosa and Aguirre-Ghiso’s teams used a computer-based screening approach to identify a drug, named C26, that activates NR2F1. The researchers found that treat-
ing patient-derived HNSCC cells with C26 boosted the levels of NR2F1 and arrested cell proliferation. The researchers then tested whether C26 would prevent metastasis in mice. Animals injected with patient-derived HNSCC cells typically form large primary tumours that spread to the lungs after the original tumour is surgically removed. Treatment with C26 reduced the size of primary tumours, and, after surgery, further doses of C26 completely blocked the growth of metastatic tumours. Instead, the rodent’s lungs contained just a few dor-
mant disseminated cancer cells unable to proliferate even after cessation of the treatment. Sosa and Aguirre-Ghiso’s teams determined that, by activating NR2F1, C26 forces cancer cells into a longlived state of dormancy characterised by a unique pattern of gene activity. Cancer patients whose tumours display a similar pattern of gene activity tend to go longer without relapsing, suggesting that inducing this dormancy program with C26-type drugs could be effective in humans. “Drugs that activate NR2F1 might be particularly
useful in breast cancer,” says Sosa. “NR2F1 is highly enriched in ER-positive tumours when compared to ER-negative tumours, and activating NR2F1 might be able to suppress reawakening of dormant cancer cells kept in that state by antiestrogen therapies.” However, because C26 treatment elevates the levels of NR2F1, the approach may also be useful for other cancers with inherently low levels of the receptor protein. “Overall, our study reveals a mechanism-based and rationally designed strategy to exploit NR2F1-activated dormancy as a therapeutic
option to prevent metastatic relapse,” Aguirre-Ghiso says. The researchers also showed that the treatment can also target multiple tumours in an animal. “After more than 10 years of work on PIP, it is rewarding to experience this convergence of expertise from laboratories around Stanford, which allowed us to develop a highly promising new cancer treatment strategy,” Cochran says. The researchers are now studying the treatment in other types of cancer, and in combination with other immunotherapies.
The long road to General Artificial Intelligence A versatile new AI is fuelling speculation that machines will soon think like humans. Last month, DeepMind, a subsidiary of technology giant Alphabet, set Silicon Valley abuzz when it announced Gato, perhaps the most versatile artificial intelligence model in existence. Billed as a ‘generalist agent’, Gato can perform more than 600 different tasks. It can drive a robot, caption images, identify objects in pictures, and more. It is probably the most advanced AI system on the planet that isn’t dedicated to a singular function. And, to some computing experts, it is evidence that the industry is on the verge of reaching a long-awaited, much- hyped milestone: Ar-
tificial General Intelligence. Unlike ordinary AI, Artificial General Intelligence wouldn’t require giant troves of data to learn a task. Whereas ordinary artificial intelligence has to be pretrained or programmed to solve a specific set of problems, a general intelligence can learn through intuition and experience. An AGI would in theory be capable of learning anything that a human can, if given the same access to information. Basically, if you put an AGI on a chip and then put that chip into a robot, the robot could learn to play tennis the same way you or I do: by swinging a racket around and getting a feel for the game. That doesn’t
necessarily mean the robot would be sentient or capable of cognition. It wouldn’t have thoughts or emotions, it’d just be really good at learning to do new tasks without human aid. This would be huge for humanity. Think about everything you could accomplish if you had a machine with the intellectual capacity of a human and the loyalty of a trusted canine companion — a machine that could be physically adapted to suit any purpose. That’s the promise of AGI. It’s C-3PO without the emotions, Lt. Commander Data without the curiosity, and Rosey the Robot without the personality. In the hands of the right developers, it could
Robot epitomise the idea of human-centred AI. But how close, really, is the dream of AGI? And does Gato actually move us closer to it? For a certain group of scientists and developers Gato and similar systems based on transformer models of deep learning have already given us the blueprint for building AGI. Essentially, these transformers use humongous databases and billions or trillions of adjustable parameters to predict what will happen next in a sequence. Deep-learning AIs are librarians and, as such, they are only as good as their training libraries. All that remains is to make them bigger and faster.
