Cancer care is a multifaceted and ever-evolving space. And, while the ongoing evolution and new technologies in cancer care are undoubtedly valuable, it simultaneously introduces an additional layer of complexity, offering an expanded array of treatment options. It’s also essential to bear in mind that there are currently more than 200 distinct types of cancer that we know of today.
There are a number of factors that go into selecting a treatment option, and in the case of prostate cancer, for example, the choice of treatment is (and should be) highly individualized. This means the clinician must wade through a vast amount of data to determine the course of treatment. These data include the stage and grade of cancer, the patient’s age, general health, any pre-existing medical conditions, risk category, and side effects, among many other elements.
Integrating artificial intelligence (AI) into the process of evaluating data and selecting a treatment based on the findings, cannot only reduce the time it takes clinicians to comb through the patient information, but also serves as a tool to hone in on the best treatment choices to narrow down the options. Therefore, AI when used in the appropriate capacity can help increase confidence in the treatment option selected, for the patient and clinician alike. Overall, AI should not be in the driver’s seat but it can serve as a “GPS” guiding the patient and clinician to the destination they are both looking to go.
Predictive and Prognostic Analytics
AI can also help clinicians more accurately prognosticate outcomes when compared to utilizing current clinical guidelines alone. For example, when a physician is prognosticating prostate cancer, they typically gather as much information as possible and use guidelines to determine a treatment plan for their patient. While guideline recommendations have been a great tool thus far in helping to personalize medicine, there is still room for greater precision. AI, however, can use an individual patient’s clinical information and pathology slide images to provide greater insights into how the disease will progress and whether a patient may benefit from specific treatment options. It has been found that the AI model can outperform the current guideline recommendations available for treating men with localized prostate cancer.
Personal Approach to Care
AI-derived biomarkers have demonstrated the ability to assist in selecting the most appropriate treatment approach based on a patient’s individual characteristics. For example, for localized prostate cancer, an AI-derived biomarker test can identify patients who are likely to benefit from androgen deprivation therapy (ADT) in combination with radiation therapy and those who instead may achieve similar outcomes with radiation therapy alone. This targeted approach ensures that patients receive treatments that are most likely to work for them.
Enabling a patient to make an informed decision with their clinician surrounding their treatment selection is powerful. For prostate cancer specifically, adding ADT to radiation therapy is a common practice. While ADT has a place in prostate cancer treatment, it is also known to have negative side effects that can impact a patient’s quality of life and overall health. Recent studies suggest many men may not need ADT as part of their treatment plan, and that radiation therapy alone is effective. If given the opportunity to leverage the biomarker, clinicians and patients can make a more informed decision when creating a treatment plan, and possibly avoid the added risks of ADT if it would not benefit them.
AI can also directly benefit patients as it can be utilized to generate easy-to-understand reports around a patient’s diagnosis, treatment options, and potential outcomes. By providing patients with clear and comprehensive information, AI can help them become active participants in their treatment decisions, promoting a sense of confidence and increased treatment adherence.
Bridging Racial Disparities
When AI is trained and clinically validated via diverse patient populations, the tool can also help create greater confidence among minority populations. For decades these populations have been underrepresented in clinical trials, which has resulted in inadequate care and mistrust. This is especially concerning as African Americans are at greater risk of developing and dying from several different types of cancer.
Recent data shows that for oncology clinical trials, less than 5% of participants were Black men. When creating AI tools, researchers need to be aware of this historical underrepresentation and ensure that they are using diverse data that adequately represent the population. Doing so helps train the AI algorithm in a more inclusive manner, allowing for personalized care for all patients.
Enhanced Patient-Provider Communication
As mentioned above, the information provided by AI tools can greatly benefit the patient and clinician relationship. With more information available from AI, clinician and patient communication is enhanced to improve shared-decision making for a more personalized treatment plan. AI tools can also provide easy-to-understand reports, which can make patients feel even more confident in their treatment decisions. This technology helps strengthen communication and ensures that patients are well-informed and reassured throughout their cancer journey.
Tim Showalter, MD, MPH
Tim Showalter, MD, MPH, is a radiation oncologist and cancer researcher who serves as Vice President of Clinical Development at ArteraAI. In this role, he provides scientific and clinical leadership for product development and evidence generation needed to bring ArteraAI biomarkers into clinical use. Tim maintains a part-time clinical practice at the University of Virginia as Professor of Radiation Oncology. During his 13 years in academia, Tim led clinical programs in prostate, breast and gynecologic cancers, published over 175 peer-reviewed manuscripts, and oversaw the design and conduct of clinical trials and research programs funded by the National Cancer Institute. He is experienced in outcomes research and cost-effectiveness analyses and was the founder of a medical device company funded by federal grants.
