Dr. Rhonda Patrick: Cancer is an awful disease. Anyone listening who has had cancer, or knows someone who has, understands that.

Many people think of cancer as one disease. Non-scientists and non-physicians often think of it that way, but you and I both know that it is not one disease. It is hundreds of diseases.

We still do not have a cure for cancer. We have made a lot of progress, and some cancers can be treated better than others. Why has it been so hard to find treatments for cancer?

Dr. Derya Unutmaz: The important thing to clarify is that cancer is not one disease. It is probably 100 different diseases with hundreds of subtypes.

Some cancers are nearly 100 percent curable. Some childhood leukemias that were fatal a few decades ago now have cure rates near 90 percent or higher. If certain cancers are caught early enough, cure rates can also approach 100 percent.

These diseases are very different. Pancreatic cancer is very different from lung cancer or breast cancer. Some cancers grow so slowly that when they are found in an 80-year-old, doctors may monitor rather than treat them because they are unlikely to shorten that person's life. Certain prostate cancers are an example.

Cancer is complex, but one central challenge is that cancer cells are part of us. When a bacterium or virus infects us, the immune system can recognize it as foreign and fight back. We can also target a bacterium specifically with an antibiotic without directly targeting normal human cells.

Cancer is different. A treatment that stops rapidly dividing cancer cells can also stop rapidly dividing normal cells. That is why people can lose their hair and why the immune system can become weakened during chemotherapy. Hair cells and immune cells also divide. Treatment side effects can sometimes be severe.

The recent revolution in cancer treatment is immunotherapy. The question was whether we could make the immune system recognize cancer as a foreign threat. Cancer cells can resemble normal cells closely enough that the immune system does not know what to do.

If we can teach the immune system, or remove some of its regulatory brakes, it can recognize and attack cancer cells. That hypothesis worked. Cancer immunotherapy has become a powerful treatment, although chemotherapy and radiotherapy still have important roles.

Another goal is to make treatments very specific. Chemotherapy is not very specific. If we identify a mutation in a target such as the EGFR receptor, we may be able to develop a small molecule that acts only when that mutation is present. These targeted drugs can be very effective in the patients whose tumors carry the matching mutation.

We can also engineer immune cells. In CAR T-cell therapy, cells are removed from a patient and genetically engineered to recognize a marker on cancer cells. The engineered cells seek out cells carrying that marker and kill them. This approach has produced remarkable results, although cancers can still escape by suppressing immunity or losing the targeted marker.

I think personalized mRNA cancer vaccines will also be revolutionary. A patient's breast cancer may have mutations that another patient's breast cancer does not have. We can identify those tumor mutations, synthesize RNA encoding them, and give it as a vaccine. The vaccine trains that patient's immune system to recognize those mutations as a threat.

If the cancer later acquires different mutations, another mRNA vaccine could be created to train the immune system against those changes. This directs the immune response toward an internal threat unique to that patient. These approaches are difficult, but I think we can see light at the end of the tunnel.

Dr. Rhonda Patrick: Will cancer be 100 percent curable?

Dr. Derya Unutmaz: Probably within less than a decade.

Dr. Rhonda Patrick: How is AI going to make that happen?

Dr. Derya Unutmaz: AI is already helping. You may have heard a story from Australia about a computer scientist who used ChatGPT and other AI models to help develop an mRNA vaccine for his dog. The dog had what I think was melanoma.

The tumor was sequenced, and the sequence was given to an AI model. The model designed a precise mRNA molecule intended to train the dog's immune system. The vaccine was synthesized and applied within about three months. The tumor began to regress, and the dog remained alive beyond the expected prognosis.

That is a simple example. There are hundreds of cancer types, and we may eventually have hundreds of treatments for a single type of lung cancer. Some will be mRNA vaccines. Others will be small molecules or other targeted treatments.

To develop these on demand or very rapidly, we will need AI. AI can model possible mutations and screen millions of compounds. We may reach a point where hundreds of new drugs are developed each month. A large set of drugs could be available for breast cancer, with treatment combinations selected according to a tumor's mutations, stage, and other features.