Ground-breaking developments in immunotherapy benefit greatly from the introduction of AI while diagnosis is more precise and cost-effective.
As the father of a cancer-surviving son, these days I often find myself investigating any new forms of treatment that may be of value for patients like him. That’s just one driving force behind my enthusiasm for the growing use of AI in the development and efficacy of immunotherapy.
One significant plus to AI across the spectrum of cancer treatments is a simple matter of speed. AI-enabled devices, including advanced computed tomography (CT) scans, magnetic resonance imaging (MRIs), and ultrasounds, can perform repetitive tasks more accurately and faster than human personnel.
Not only does that reduce the possibility of misdiagnosis, but it also shaves costs by recommending care before patients’ conditions become more serious – and expensive.
One such example is an Israeli start-up that has created AI algorithms for diagnosing conditions including osteoporosis, brain hemorrhage, malignant tissue in breast mammography, and coronary aneurysms. The results come faster than those where humans are involved.
Even better, the accuracy of diagnosis improves. According to a recent report, AI has demonstrated 99% accuracy in evaluating and analyzing mammograms. This has made it possible to diagnose breast cancer more quickly, lowering the cost of diagnosis.
AI’s cost-saving potential is taking hold in other areas related to cancer care. At the Center for Cancer and Blood Disorders in Texas, oncologists employ clinical AI to help patients avoid unnecessary emergency room trips. AI enables care teams to predict which patients are likely to go into the ER in the next 30 days and recommends proactive steps oncologists and case managers can take to keep patients stable and out of emergency care altogether. Estimated savings: $3m.
AI and immunotherapy
AI’s role in enhancing the effectiveness of immunotherapy, a game-changing cancer treatment, is particularly exciting and provocative. Unlike traditional cancer treatments such as chemotherapy and radiation, immunotherapy leverages the body’s immune system – a promising approach that has shown significant success in treating various types of cancer.
Currently, overall cure rates for advanced lung, breast, and colon cancer are typically less than 50%. This indicates a need for more focused treatment programs, ones that are suited to an individual’s cancer diagnosis and stage. This is where AI steps up to the plate.
Powered by machine learning, AI algorithms can analyze vast amounts of data, including genetic information, medical histories, and imaging scans. These algorithms can pinpoint patterns and correlations, particularly those that might be missed by human observers. This allows AI to predict the likely effectiveness of immunotherapy for individual patients. In turn, that can be used to develop personalized treatment plans and, further, more effective and longer-lasting patient outcomes.
For instance, AI can analyze a patient’s genetic data to identify mutations that make the cancer more vulnerable to immunotherapy. This helps oncologists select the most effective treatment strategy for each patient. Moreover, AI can predict potential side effects and adverse reactions, enabling doctors to take preventive measures and manage treatment more effectively without encountering unwanted results.
“OSAIRIS” can save many hours of physicians’ time in preparing scans while also reducing patient wait time between referral for radiotherapy and treatment. The system works by significantly cutting the amount of time a doctor needs to spend drawing around healthy organs on scans before radiotherapy. Specialists can plan radiotherapy treatments approximately two-and-a-half times faster than if they were working alone, ensuring more patients can get treatment sooner and improving outcomes. (Source: Cambridge University Hospitals, June, 2023.)
Artificial intelligence has developed a treatment for an aggressive form of cancer in just 30 days. University of Toronto researchers worked with Insilico Medicine to develop potential treatment for hepatocellular carcinoma (HCC) using an AI drug discovery platform called Pharma. AI discovered a previously unknown treatment pathway. In a separate development, researchers from the University of British Columbia and BC Cancer have developed an artificial intelligence (AI) model that predicts cancer patient survival more accurately and with more readily available data than previous tools. The model is reportedly 80% accurate. (Source: Stacy Liberatore, Daily Mail, March, 2023.)
Artificial intelligence (AI) and machine learning platforms offer physicians the opportunity to optimize efficiency in revenue cycle management. By reviewing large amounts of health care financial data and leveraging that information to guide future claims to a more successful outcome, machine learning platforms have the potential to reduce the need for repeated insurance submissions and appeals, significantly improving the practice’s clean claim rate. More efficient revenue cycle results can help reduce the potential financial burden on patients. (Source: Dan Lodder, OncLive, May 15, 2023.)
In one study, a team of researchers from several healthcare systems and universities discovered a new AI-derived biomarker, quantitative vessel tortuosity (QVT). Tumor-associated blood vessels are chaotically arranged and twisted, blocking the normal flow of blood and redirecting it to tumors, allowing them to grow even faster and spread throughout the body. The good news is that the QVT biomarker can help pinpoint these problematic tumors. Using routine imaging scans, physicians can now predict with far greater certainty which patients with lung cancer will respond to immunotherapy.First and foremost, the findings, published recently in Science Advances, offer guidance for patients and their physicians making critical treatment decisions. In effect, AI can help doctors determine what sort of patient will respond to particular forms of immunotherapy – and those who will not.
Trim the cost of the fight
AI protocols can also blunt the financial burden associated with immunotherapy. Given the staggering costs – roughly $200,000 a year per patient – the need to non-invasively identify potentially positive responses before beginning therapy becomes financially essential. AI can help show doctors which patients will get a bang for their buck through immunotherapy as well as those for whom the cost will essentially prove a waste of money.
But AI’s use in immunotherapy isn’t limited to just highly individualized patient care. AI can also analyze global cancer research data. By doing so, it can identify new potential targets for immunotherapy, accelerating the development of new forms of treatment. AI’s ability to sift through vast amounts of research data and identify relevant findings is an invaluable tool in the overall fight against cancer.
Not surprisingly, the use of AI in immunotherapy is not without its challenges. For one thing, as sensitive patient information is used to develop AI algorithms, data privacy and security are significant concerns, including ownership of a patient’s healthcare records and history as well as whose data will be shared and under what circumstances. Patient consent also comes up. Additionally, transparency is essential for how AI draws its conclusions and treatment recommendations. This is particularly important to avoid systemic and repeated AI-generated mistakes and missteps due to inherent bias. Ethics are, understandably, also important. But for cancer patients, the use of AI in treatment such as immunotherapy is not only a matter of effective treatment and quality of life – but a growing source of hope.
This article is based on REINVENT: Navigating Business Transformation in a Hyperdigital Era by Faisal Hoque (Fast Company Press, 2023), in association with IMD. All rights reserved. All proceeds from REINVENT book sales are pledged to multiple myeloma cancer research.
Originally published on IbyIMD.
Related: 5 ways to kick bad habits by Faisal Hoque
Copyright (c) 2023 by Faisal Hoque. All rights reserved.
