AI & Cancer: Life-saving Innovation or A Gimmick
AI in Oncology: Life-saving innovation or just hot air?
Back in the 1930s, when a patient showing symptoms of cancer entered the doctor’s office, filled with an air of uncertainty, it offered very little comfort as it often ended with a death sentence due to a lack of early detection and treatment plans.
Fast forward to 2024, the narrative has transformed drastically with cancer patients walking into modern clinics, buzzing with innovation. A bright screen pops up, showing their entire medical report and exclusive treatment plans. Not only addressing the disease but also their hopes and fears. This is the kind of revolutionary healthcare AI can bring to the table.
Is AI making cancer detection easier?
Ryan Schoenfeld, CEO of Mark Foundation for Cancer Research, highlights that advancements in diagnosis, especially in radiography, have resulted in easier image analysis and detection. He said, “AI can now analyze scans faster and more accurately, helping doctors catch cancer earlier.”
Philip Lieberman, founder and president of Analog Informatics, contradicts Schoenfeld’s statement. He believes that AI might not be as accurate as a trained technician; however, AI can detect difficult clues that might not be recognizable by the human eye.
In rural areas or underdeveloped countries that do not have access to a trained technician, AI imaging and analysis could be a game changer. It is not only readily utilizable but also faster and cheaper, making it a smart move.
Jeffery Sorenson co-founder and CEO at Yono, emphasizes that AI could reconstruct patient data and organize and group them accordingly. Streamlining patient data makes it easier for AI predictive diagnosis based on individual patient data and predicts patient outcomes.
Mayo Clinic’s AI system enhances early diagnosis and predictive analysis, which are very important in oncology, as each individual could have varied outcomes.
Ai powered liquid- biopsy method (MRD-EDGE) analyzes tumor DNA and catches cancer recurrence months or years earlier than usual methods. It has been proven successful in cases of lung, colorectal, and breast cancer.
Early cancer diagnosis is very crucial for patient care, but AI also has a huge role to play in drug discovery.
AI in Cancer drug development and discovery
Tools like DeepMind’s AlphaFold have transformed how we study disease-related proteins, speeding up the identification of new therapeutic targets,” said Schoenfeld.
Alphafold, an AI tool developed by DeepMind.It targets predicting the 3D structure of proteins, which enables identifying potential drug targets. In 2022 it has already predicted 200 million protein structures, generating umpteen amounts of datasets, which can be used for studying more potential drug targets in cancer research.
Collaboration between Lawrence Livermore National Laboratory (LLNL) and BridgeBio has given rise to a magical drug-BBO-8520, which targets KRAS mutations. Utilizing AI-powered simulations, the drug has reached phase 1 of clinical trials. Exscientia, a drug discovery company, has also used similar technologies to develop novel cancer treatments.
Not just drug development and discovery, AI also has a hand in repurposing drugs. For example, GL1 agonists, a drug developed for diabetes, have drawn approval for obesity and cancer as well.
AI model designed by Harvard Medical School -TxGNN connects well-understood disease information with ill-understood diseases. In terms of cancer, it could help repurposing drugs of well-understood diseases.
Dreaming of a Six Figure Salary Job in a Bio-IT Company? Here’s Your BIO-IT Career Plan!
Remember, the first step always seems the hardest. But fear not! Contact us! https://bit.ly/aiml-whatsapp
Dive in with BioTecNika scientists, and watch yourself transform into a Success Magnet! – https://btnk.org/ai-ml-drug-discovery
Join our AI ML in Drug Discovery, BioPharma & Chemistry Training Program starting from 11th Nov 2024!
AI and Precision Oncology
Just as some people have allergies to peanuts while others react to seafood, each person responds differently to medical treatments. This difference in outcomes highlights the vitality of personalized and precision medicine analysis databases containing information regarding patient genetics, clinical history, and tumor profiles to simulate treatment outcomes.
Sorenson says cancer isn’t the only challenging thing, it’s instead hundreds of different variations – variations in the cancers itself but also patient’s genetic profiles. AI is enabled to calculate different possibilities, it matches the patient’s genetics with the drug molecules and customizes therapies. Tailored therapies target specific cancer cell types, reducing the risk of toxicity and cancer relapse.
Researchers created adaptive therapy schedules for prostate cancer patients, using Deep reinforcement learning (DRL). Mathematical models simulate how individual patients react to different treatments. The results were assuring as it doubled the time to relapse, compared to usual methods. This technology could prove its importance in metastatic cancers where drug resistance is a huge problem.
EVX-01, an AI-designed personalized cancer vaccine was presented by Evaxion Biotech. The trial combines EVX-01 with Keytruda for patients with advanced melanoma. Key outcomes presented a 69% overall response rate, with 15 out of 16 patients experiencing a reduction in tumor size validating the accurate prediction.
Challenges
The regulations for AI in healthcare are still being figured out in Europe by the European Commission on guidelines to make sure AI is safer to use and the FDA is working on regulating AI-dependent medical devices.
The regulatory bodies must be up to date with all the advances, as it is important to make the frame and explicit rules that encourage innovation with patient safety.
If the AI model is being trained mainly based on one particular group of people for example western population, AI might not achieve the same expected results in people belonging to ethnic backgrounds or other groups of people.
These issues need to be addressed to get the best of AI in cancer care. Regulations are an important framework for patient safety.
What’s Next? AI Breakthroughs You Can’t Ignore
We are already witnessing a great transformation in cancer treatment, in the next 10 years, AI is positioned to revolutionize the field of oncology. Here Are three areas where AI is expected to have a greater hand in making a difference.
First, AI predicts which unique T cell receptor will bind to specific antigens at a structural level, revealing methods for cancer immunotherapies, cell-based therapies, and other treatments.
Second, AI models enabling protein designing mutations in key proteins like cytokines, repurposing them as therapeutic agents. The collaboration of immunotherapy and designer proteins has a huge scope for personalized medicine.
Third, using LLMs(Large language models) for predictive analysis. IBM’s Watson failed to meet its high expectations in clinical practices, but LLMs seem to have the potential to enhance clinical record analysis.
Snyder believes that combining AI with biological data will allow real-time analysis of treatment plans based on patient conditions. However, Sorensen highlights healthcare organizations might face financial difficulties in procuring these AI models. Pharma and drug-developing companies need to work closely with AI companies to overcome all the challenges.
AI holds a promising revolution in cancer care, but it needs to approach this transformation in an optimistic way. It might not overcome every challenge overnight, but its role in revolutionizing cancer care is undeniable. As we continue to innovate we move closer to a future where cancer is not just managed but truly transformed. It’s just the beginning, possibilities are vast.
AI & Cancer: Life-saving Innovation or A Gimmick
