Scientists are constantly working to find a cure for cancer. Their research takes them in a lot of often unexpected directions — including steps that incorporate artificial intelligence (AI). How is artificial intelligence accelerating cancer research?
Understanding Drug-Resistant Cancers
Around 39.5% of men and women will be diagnosed with some form of cancer at some point in their lives. However, the earlier a patient receives a cancer diagnosis, the better their prognosis and long-term survivability become.
Anti-cancer drugs are as many and varied as the cancers they treat. One major challenge researchers face is the tendency of cancer to resist the drugs that doctors use to save lives.
Cancer drug resistance can come in a variety of forms, including:
- Drug inactivation: The cancer cells directly stop the treatment drugs from working.
- Multi-drug resistance: Cancer cells become resistant to multiple different types of drugs, making them harder to treat.
- Cell death inhibition (apoptosis suppression): Blocking the pathways the drugs follow to trigger cancer cell death.
- Changes to drug metabolism: Some chemotherapy treatments rely on enzymes. Drug resistance can occur when the cells change how the body processes these enzymes.
- Epigenetic changes: Mutations, either within the cancer cells or the therapeutic agents, can prevent the drugs from working as intended.
- Altering drug targets: If cancer treatments miss the target, they aren’t effective and can do damage to healthy tissues.
- Enhance DNA repair: Some therapeutic agents work by changing the DNA in the cancer cell, which triggers cell death. Resistance to this sort of genetic manipulation can render the treatment ineffective.
- Target gene amplification: This treatment works by increasing the number of targeted genes within a cancer cell. If it goes wrong, it could trigger drug resistance.
The biggest challenge here isn’t figuring out what might work. Rather, predicting what might cause these cancer cells to become resistant to treatment is critical.
Improving Cancer Treatment Efficacy
Chemotherapy is one of the primary tools doctors use to treat cancer and manage symptoms, but it is a delicate balance between a successful treatment and poisoning the patient. Two similar patients can be diagnosed with the same type of cancer, but have their chemo treatments react differently because of some unseen variable.
Artificial intelligence has the ability to account for all of those variables, predicting which type of chemotherapy regimen will have the best success rate based on information collected from previous patients.
The more information an AI or machine learning system is allowed to collect, the smarter it becomes. With enough data, these systems can even begin to make extremely accurate predictions about future treatment plans before they ever become necessary. In the case of cancer treatment, this enables researchers and oncologists to optimize and customize treatment plans for each individual patient, while taking steps to reduce the risk that treatment-resistant cancer cells could develop.
Creating More Accurate Radiation Treatment Plans
Radiation treatments are often the other facet of cancer management, but this can be a bit of a double-edged sword. It has the potential to kill cancer cells and aid in recovery, but exposure to specific types of radiation increases the risk of harmful side effects. Biological cells don’t respond well to radiation, taking damage or even mutating and causing additional problems.
Artificial intelligence helps doctors create better treatment plans that specifically target cancerous cells without causing damage to the rest of the patient’s body. This can be vital when the cancer cells are located in a delicate part of the body.
One study found that an AI program could delineate between healthy tissue and nasopharyngeal carcinoma with 79% accuracy. Creating a treatment plan can also take place much faster, which can help improve patient outcomes.
Managing Immunotherapy Treatments
Radiation and chemotherapy aren’t the only treatment options being bolstered by AI. Instead of using chemicals and radiation, immunotherapy works by boosting the patient’s immune system and using it to fight the cancer cells. It isn’t an option for everyone, as not everyone’s immune system is up to the task, so finding the patients that it will work for is often a case of trial and error.
AI algorithms can skip a lot of the trial and error by determining who will best respond to immunotherapy by studying biomarkers from those who have experienced successful immunotherapy treatments in the past.
As with many AI and machine learning applications, these diagnostics rely on past data to form their conclusions. But with sufficiently large datasets, their predictions will likely become more accurate than anything even experienced diagnosticians, researchers, and oncologists can manage unassisted.
When it comes to cancer diagnostics, overtreatment is just as dangerous as undertreatment, especially when considering how toxic some of the treatments can be. There are also some conditions, such as lesions of breast tissue, that have the potential to become malignant.
The temptation is to treat these lesions as cancerous, even if they don’t prove to be malignant. This sort of preemptive treatment can do more harm than good.
Artificial intelligence and machine learning systems can identify pre-cancerous lesions and accurately determine whether cancer treatments are even necessary. Preventing overtreatment keeps patients safe and healthy while ensuring those who need the treatments are able to access them.
Supporting Clinical Decision-Making
Clinical decision-making, both in laboratory and treatment settings, can be a time-consuming process prone to human error. Introducing artificial intelligence won’t replace trained oncologists, but it can make the process easier and more streamlined. This is especially valuable for patients who may be working with oncologists in different parts of the world.
In this situation, AI systems are valuable assistants for doctors and researchers, helping them to find the most suitable treatment plan for patients in a timely manner. The faster these treatment plans can be put into motion, the better the patient’s overall outcome tends to be.
Developing New Anti-Cancer Drugs
The discovery and testing of new drugs is a long, drawn-out process. Researchers have to:
- Discover each new molecule
- Determine whether it might have any clinical applications
- Move through testing, development, and trials.
This process can take years, and every day that passes during development could mean another life these drugs aren’t able to save. Using machine learning and AI takes a lot of the trial and error out of this development process.
AI systems can sort through molecules and test them in simulations. This narrows down the selection process so researchers can address and test the ones that show the most potential in a lab setting. These systems have the tools to sort through years of stored data in a fraction of the time it would take human researchers to do the same.
Artificial Intelligence Accelerating Cancer Research and More
AI might be the villain in popular science fiction stories. However, it’s taking its place in the research and development sector and becoming an invaluable tool. Adopting new technology, such as AI and machine learning systems, can help researchers make the most of decades of information while working toward the ultimate goals of curing cancer and saving lives. As more research emerges, let’s hope that artificial intelligence continues accelerating cancer research and brings us one step closer to a cure.
Follow Us On
Get the latest tech stories and news in seconds!
Sign up for our newsletter below to receive updates about technology trends