Cancer Care in the Era of AI: How Artificial Intelligence Can Improve Prognosis

In 2024, roughly 2 million Americans will be diagnosed with cancer which is also the second leading cause of death in the United States. These numbers show the absolute necessity of ensuring that early detection mechanisms and treatment options, such as cancer screening, are readily available. Yet the reality indicates that there is a disparity in cancer care outcomes across the United States; for example, the 46 million Americans living in rural communities are more likely to die from cancer than their urban counterparts. 

However, the rise of artificial intelligence (AI) in recent years has paved the way for ground-breaking developments in cancer treatments, and this would not only improve diagnosis but facilitate closing the gap in care for cancer patients and promoting greater health equity. More and more hospitals are adopting AI into patient care, but what role will AI play in treating cancer? Learning about the uses of AI in medical care has become essential knowledge for FNP online programs and various other medical courses, and if you have been diagnosed with cancer or have a family with cancer, keep reading on to learn about the exciting and boundless potential AI has in the future of cancer diagnosis. 

Cancer: What Is It?

Cancer is generally defined as a disease where the growth of certain cells within the body becomes uncontrollable and spreads to other parts of the body. 

Under normal circumstances, human cells grow and multiply under cell division to form newer cells in the body. The older cells degenerate and eventually die, with new cells replacing them. However, dysfunction in this process begins when damaged cells grow and multiply. They can turn into tumors which are a group of these abnormal cells. Tumors can either be cancerous (malignant) or non-cancerous (benign). Doctors determine whether a tumor is malignant or benign by seeing if it spreads or penetrates nearby tissue; cancerous tumors spread while benign ones do not. The most common types of cancers such as breast, colon, bladder, prostate, or lung cancer are classified as solid tumors, while blood cancers such as leukemias are called liquid tumors.

Generally, benign tumors do not grow back when removed, while malignant cancerous tumors sometimes do, and it can be difficult to predict whether they will or will not. It evidences the need for better and more precise treatment mechanisms and personalized plans to improve cancer care, a field in which AI can potentially shine the brightest. 

Disease Forecasting

One of the most transformative features of AI is its predictive models, and such technology can shepherd a new era of precision in dealing with cancer. It can be challenging for oncologists to give accurate predictions on how a particular cancer might develop and what treatment regimen is most effective for each individual patient. AI can however use data based on previous clinical assessments, investigations, scans, medical histories, and information to predict disease risk, prognosis, progression, and patient survival outcomes, potentially overcoming such difficulties.  

For example, a study conducted by Harvard Medical School and the University of Copenhagen (Denmark) in 2023 used AI to predict pancreatic cancer. Researchers used the technology to analyze data from 6 million individuals (24,000 with pancreatic cancer) in Denmark and 3 million individuals (3900 with pancreatic cancer) in the U.S. The results were that the AI was able to identify people most at risk (or have elevated risk) of pancreatic cancer at approximately 36 months (three years) before any actual diagnosis was made. 

Pancreatic cancer is renowned for being deadly, so a mechanism that can provide such early detection can go miles in ensuring adequate prevention, providing timely treatment, and preventing unnecessary deaths for a disease that has a reputation for late diagnosis. 

Personalized Treatments

AI technologies can play a critical role in enhancing the effectiveness of future treatment options like immunotherapy. Compared to traditional chemotherapy and radiation treatment methods, immunotherapy directs the body’s own immune system to fight the cancer. As the treatment method is still in its early stages, cure rates still sit at around 50% for patients, which indicates the need to move beyond generalized strategies to a more personalized application directly suited to each individual’s cancer diagnosis and stage – and this is where AI will play a decisive role. 

Previously mentioned data models can be used by AI algorithms to predict the effectiveness of immunotherapy for a patient. Oncologists can gather these conclusions to develop personalized treatment plans, which have proven to have far longer-lasting patient outcomes. AI technology for example can analyze a patient’s genetic data to identify mutations or any problems within the body that may make immunotherapy less effective. It can also predict side effects and adverse reactions, allowing oncologists to ensure patients are informed and, most importantly, avoid the avoidable. 

One study has already used AI algorithms to predict patient response to immunotherapy treatment methods for advanced melanoma cancer patients. Another collaborative study has experimented with an AI-derived biomarker, quantitative vessel tortuosity (QVT), which can pinpoint problematic tumor-associated blood vessels to predict responses to immunotherapy amongst lung cancer patients. Moreover, AI has also been used to decide which treatment options are necessary and unnecessary for women with breast cancer, as well as its potential future uses to determine an adequate duration and intensity of chemotherapy for patients. 

Reducing Costs and Greater Affordability 

Harnessing the power of AI can also ensure better affordability in treatment, which is a key step in closing the gap in cancer care. In 2019, the national patient economic burden for cancer treatments was estimated to be $21.09 billion in the U.S., with estimated out-of-pocket costs at approximately $16.22 billion and patient time costs at $4.87 billion. These high costs indicate an urgent need to lower the price of cancer treatments, and AI technologies with their cost-saving can make this a reality. 

Just this year, researchers at Shanghai Jiaotong University (China) have produced and tested a new tool that can diagnose cancer with a mere drop of blood, with a focus on pancreatic, gastric, and colorectal cancer. The system runs on an AI machine-learning model that analyzes the patient’s blood. It is quicker and more cost-effective than current testing methods, such as computed tomography (CT) scans used for pancreatic and colonoscopy cancer and gastroscopy used for gastric cancer. Researchers concluded that upon using the technology, undiagnosed cases fell dramatically, with colorectal cancer falling from 84.30% to 29.20%; gastric cancer from 77.57% to 57.22%; and pancreatic cancer from 34.56% to 9.30%, with an overall reduction being between 20.35-55.10%. 

But AI can not only be used in the field of testing or treatment, it has also been used in cancer drug discovery, which can lead to quicker development of safer, and more inexpensive and effective drugs for patients. For example, in 2023, the University of Toronto (Canada) used an AI-powered database called ‘AlphaFold’ to create a new liver cancer drug in just 30 days. Making drug discovery and production faster would inevitably reduce costs and be more affordable for patients. 

Looking Forward: Potential Challenges

Potential future challenges of AI in the medical field always revolve around ethical and privacy concerns. Some questions may include:

• Is the data accurately representing the population? There is always the potential for AI to make erroneous decisions, so this should be constantly monitored. 
• Is the AI using data neutrally concerning sex, race, and other factors? Any dataset can be biased towards certain demographics, so it must be neutral. 
• Is there a balance between privacy protection and technological development? AI models use huge datasets, so it is important to strike a proper balance between preserving confidentiality and continuing development. 

The future is bright for AI to be a key part of cancer care. But it is also important that basic ethical principles are upheld, to ensure that individual, private, and public interests are considered.