In an era where cancer rates continue to rise, the emotional toll a cancer diagnosis and the subsequent treatment journey has on a patient is becoming a critical concern. A 2021 study found that, globally, 27% of cancer patients have depression. The same study notes that this rate is rising by 0.6% every year.
Most patients and loved ones confronting cancer can be faced with a flood of emotions, grappling with feelings of fear, confusion, and uncertainty. Not to mention, both patients and clinicians are faced with the daunting task of navigating complex treatment decisions, as well as the ins and outs of a complicated healthcare system, amidst this emotional turbulence. Today more treatment options are available than ever before, and most patients will be on a combination of treatments. While the increased options offer more hope, they also offer more roads to take to arrive at the optimal treatment path. Determining the course for each patient based on their unique clinical characteristics and needs can be an uphill battle without the right tools to guide the way.
[time-brightcove not-tgx=”true”]
In the cancer journey, clinicians are forced to strike a balance in matching the aggressiveness of treatment to the aggressiveness of the tumor. Without the foresight of how cancer will progress, and how quickly, both clinicians and patients may find themselves navigating the path forward without clarity or confidence.
But what if technology could help us simplify the problem? With the rapid advancements in AI, we are witnessing a shift in how we deliver healthcare. AI-enabled cancer tests, in particular, have the potential to provide personalized insights, empowering both patients and clinicians to make more informed treatment decisions and ultimately improving patient outcomes, both mentally and physically.
Read More: The Unique Hell of Getting Cancer as a Young Adult
One of the highest contributing factors to the mental anguish of cancer patients is the overwhelming uncertainty associated with treatment decisions. Patients are afraid of what they are signing up for, often asking “Will this therapy make a difference for me” or “Is it worth the adverse effects?” These concerns are especially valid as cancer treatments are known to have potential life-altering side effects such as loss of sexual function, bone density loss, decreased muscle mass, and even increased risk of death. When these questions go unanswered, they can hugely impact a patient’s psychosocial adjustment, health behaviors, and quality of life.
A patient’s mental health can also prevent them from seeking out treatment or consistently following their treatment plan. A 2024 study in Cureus found that patients who do not follow their treatment plan are more likely to have worse outcomes including increased physician visits, higher hospitalization rates, longer stays, disease progression, and elevated mortality rates.
AI can serve as an aid. Today there are AI-enabled tests that can swiftly analyze real-world data and translate it into digestible and personalized insights, allowing for a more personalized approach to cancer therapy. For example, there are tests for localized prostate cancer that use enormous amounts of data from digitized pathology slides to extract hidden patterns and corresponding knowledge about an individual patient’s disease. These are data that clinicians are not capable of interpreting as the human eye cannot uncover the patterns within these large datasets without technological support. The current standard of care is to rely on a pathologist, who will review pathology slides, determine if cancer is present, and grade the cancer risk, e.g. Gleason grading for prostate cancer. While these general categories are effective, the reality is that digitizing the pathology slide can identify between 10-40k image patterns of data from just one patient alone. With the power of AI, the test can analyze the digitized slide to determine with far greater detail the specific risks of the individual patient, and provide detailed insights into prognosis and treatment benefit. This benefits the clinician by not only saving time and alleviating the cognitive burden but also by creating confidence when it comes to determining whether or not to advise a patient to undergo a specific course of treatment.
Integrating AI into the decision-making process not only places the patient at the center but also provides more clarity throughout the cancer treatment journey. How does it achieve this? By equipping patients with precise information derived from their unique data and potential outcomes. For instance, in prostate cancer, AI tests now have the capability to analyze digital biopsy images and patient clinical data to identify those who would benefit from hormone therapy. A 2023 study validating this test showed that only a third of the patient population receiving the therapy demonstrated benefits, meaning two-thirds of patients were overtreated and could’ve safely avoided the therapy.
Read More: AI-Driven Behavior Change Could Transform Health Care
Aside from the toll cancer therapy takes on the body, there’s also the physical strain and apprehension regarding the potential necessity for multiple biopsies to procure tissue samples.
For example, genomic tests can offer the potential for personalized care, however, they often require the consumption of tissue samples. As new tests emerge, a portion of the tissue sample is needed for re-analysis. Tissue is finite, which poses a significant challenge as each genomic test diminishes the available tissue for future testing or procedures. However, the rise of digital pathology allows the opportunity to forgo additional tissue consumption, and instead, AI tests can tap into the same digitized samples as many times as needed, yielding fresh insights without the need for further invasive procedures. This approach not only preserves patient tissue but also streamlines prognostic procedures, potentially obviating the necessity for subsequent biopsies.
It is also important to recognize the challenges of implementing AI into clinical care. One of the reasons racial bias exists in AI, for instance, is because there is a lack of racially diverse clinical data, leaving algorithms to be trained from a patient population that is primarily white. This is why researchers must properly train algorithms on racially diverse data that represent the patient population. The technology is only as great as the data that it receives.
There is no argument that AI will have a significant impact on the healthcare industry. While the technology can aid in alleviating administrative burdens and streamline workflows, its more powerful potential can be found in its ability to simplify complex problems and signal solutions. Harnessing this capability will be what paves the way for patients and clinicians to find hope, confidence, and comfort amid a cancer journey.