By Yehudis Kundin, Staff Writer
Medicine has progressed rapidly in the last 200 years, increasing the average life expectancy and equipping us with treatments for many of the diseases prevalent today. However, cancer has remained a medical mystery. Much research has been dedicated to uncovering its causes and characteristics, and many resources have been devoted to developing effective treatments. This has significantly reduced the number of fatal cases, with current statistics showing a 34% decrease in the cancer mortality rate in the United States from 1991 to 2022. However, we still haven’t completely cured it, and millions of people are still diagnosed with the deadly disease every year.
There is a reason that cancer is so difficult to treat. Cancer occurs when cells in the body begin to multiply uncontrollably. While systematic cell division is necessary for normal growth and development, cancer emerges when damaged or abnormal cells continue multiplying when they shouldn’t. This process occurs due to genetic mutations that interfere with normal cellular regulation. The heterogeneity of cancer, caused by diverse initial mutations and ongoing genetic changes throughout its progression, renders it extremely challenging to manage therapeutically.
Traditional approaches to treating cancer include radiotherapy, chemotherapy and surgery. Recently, however, immunotherapy has emerged as another viable option. Immunotherapy is unique in that it harnesses the body’s natural immune system to fight cancer and control its progression throughout the body. The two main types of immunotherapy are CAR (chimeric antigen receptor) T-cell therapy and TCR (T-cell receptor) therapy.
T-cells are white blood cells that destroy harmful pathogens that enter the body. In CAR T-cell immunotherapy, T-cells are extracted from the body and genetically engineered to produce chimeric antigen receptors, or CAR, that are displayed on the surface of the cell. These synthetic proteins help navigate the T-cells to cancer cells by targeting a specific protein that appears on the surface of cancer cells. Once they bind to the cancer cells, the T-cells take action and destroy them. CAR T-cell therapy is limited in that it can only target cancer cells that display their cancer-causing proteins on their external surfaces. This accessibility constraint is why the therapy has only been approved for blood cancers, which are generally easier to target. The cancer cells of solid tumors, on the other hand, are often buried deep within the tissue, hiding their internal proteins and effectively blocking the CAR T-cells from detection and penetration. In 2017, the FDA approved 2 versions of the CAR T-cell therapy.
TCR therapy is a more traditional immunotherapy. Like CAR T-cell therapy, it also genetically engineers T-cells to be able to easily identify specific cancer cells. However, rather than inducing T-cells to produce receptors that will bind to the proteins on the surfaces of cancer cells, this method harnesses a natural T-cell mechanism used in identifying foreign intruder cells in the body. The method relies on human leukocyte antigen (HLA). HLA is a molecule that acts like a basket, containing bits of degraded proteins from inside a cell. It is found on the surface of most cells, and upon binding to this antigen, T-cells can then identify whether the proteins the cell is producing are normal or abnormal.
However, not all T-cells are able to recognize cancerous proteins on the cell. TCR therapy involves first identifying the few T-cell receptors in the person’s body that do recognize the neoantigens, or the proteins, of the cancerous cell. Next, other T-cells from the patient are genetically engineered to produce these exact receptors, so that the patient has more T-cells with which to recognize and then destroy the cancerous cells. The benefit of TCR therapy is that it is able to detect those proteins that are embedded within cancer cells, and it is therefore more beneficial for solid tumors than CAR T-cell therapy. While the FDA did approve TCR therapy for various solid tumors, one major limitation is its effective specificity. It can only recognize the proteins presented by a specific HLA, and HLAs vary greatly between people.
Recently, however, the research team of Timothy Jenkins, a medical biotechnologist at the Technical University of Denmark in Lyngby, discovered a more effective and efficient form of immunotherapy. The researchers used AI to design custom proteins that T-cells were then genetically engineered to carry. Just like CAR T-cell therapy and TCR therapy, these proteins would guide the T-cells directly to the cancer cells for destruction. The difference, however, is that with the help of AI, the scope of cancer immunotherapy has now been expanded.
To design these proteins, scientists input the structure of the target cancer into a generative AI model called RFdiffusion, which is trained on known protein structures and their amino acid sequences. The AI then proposed different possible protein structures that would fit perfectly with the cancer shape. Then, another AI model proposed possible amino acid sequences that would create that protein structure. A third AI model narrowed down the options. When testing these AI models, researchers found one protein design that, when expressed by a human T-cell, could effectively kill melanoma (skin cancer) cells and stop their growth. These AI models can produce custom protein designs within days, and after just a few weeks of lab testing, they are ready to be used. On the other hand, methods like TCR therapy, which involve searching for T-cell receptors that already exist in the body to use in fighting the cancer cells, can take months and may not even produce a viable therapeutic candidate.
This is an exciting discovery that can potentially contribute to an even larger decrease in cancer mortality rates. But there is still a long way to go. Christopher Klebanoff, a medical oncologist and researcher at Memorial Sloan Kettering Cancer Center in New York City, warns that more years of laboratory and animal testing are necessary before this therapy can be advanced to human clinical trials. It’s important first to determine how these AI-designed proteins will behave inside a human body. Nevertheless, this groundbreaking, novel approach stands as a pivotal next step in cancer therapy development, harnessing the power of AI to forge new frontiers in medical treatment.
Photo Credit: Unsplash