By Dr. Paul Billings

Innovative technologies for early disease detection and ongoing surveillance of high-risk individuals are vital components of prevention and treatment strategies. To bolster engagement with risk identification programs, it is essential to identify people with the potential to develop disease before the onset of clinical symptoms, which are often indicative of advanced disease. Artificial intelligence (AI) is now being used alongside traditional health screening methods to identify at-risk individuals who are candidates for surveillance programs. Once identified, advanced diagnostic assays can be used for monitoring, leading to timely interventions and more targeted care. Combining AI and new surveillance technologies can significantly impact the care and management of high-risk patients.

The state of early detection for cancer
Populations are comprised of unique individuals, including many with higher or lower than average risk of any disease or treatment response. Certain cancers provide a good example of high-risk disease that is often caught too late for treatment efficacy. Current health guidelines recommend screenings for a limited number of cancers and only for those with risk factors such as age and family history of cancer. Historically, high-risk individuals (representing fewer than 1% of the population) have been challenging to identify because of incomplete or inaccurate medical records and unknown or incomplete family medical histories. However, a better understanding of the genetics that leads to cancer, along with emerging technologies, is helping cast a wider net and increase the eligibility for cancer surveillance programs. Through surveillance, research has shown an improvement in early-stage cancer detection, leading to lower mortality rates for more challenging cancers, such as pancreatic cancer.

Pancreatic cancer represents 8.3% of all cancer-related deaths and is one of several cancer types commonly diagnosed after the disease has metastasized. Because most pancreatic cancer cases are diagnosed at later stages, the 5-year survival rate is just 12.5%, much lower than the average survival rate of 68.4% for all cancers. Recent research from the U.S. and the Netherlands shows that enrolling people at risk for pancreatic cancer in surveillance programs can lead to a stage shift where most cancers are detected when they are still localized.

Pancreatic cancer risk is currently assessed based on factors including age, family history, and behavioral and clinical indicators, with the more recent addition of genetic status. However, a lack of consistent medical records, heterogeneous clinical recommendations, and uneven application of clinical practices has led to only a fraction of eligible people entering surveillance programs.

Expanding surveillance eligibility through EMR and AI
Electronic health records (EHR) are increasingly used as real-world data to inform predictive AI models, which are now being created to identify people likely to develop challenging cancers. These models analyze extensive clinical records of larger populations, weighing an array of clinical measurements to generate criteria for high-risk surveillance cohorts beyond current monitoring programs, and continue to evolve as new information emerges for analysis.

The recent proliferation of advanced genetic tests is creating another data set that can be evaluated by AI to identify disease markers. Comprehensive EMR databases have enhanced the accuracy of genetic testing, and these tests are now commonly used for newborns, rare diseases, and cancer risk identification in families. AI can create more accurate risk profiles by combining common gene alterations associated with people who have developed certain cancer types with family history and clinical data.

A powerful duo – AI and surveillance
Predictive diagnostic tests are influenced by a person's prior probability of developing the disease, making the positive predictive value of these tests closely correlated with disease prevalence in a group. AI systems can detect those at risk for cancer using probabilities and predictions even when the available information is incomplete or uncertain, enriching disease prevalence in a surveillance cohort. Once individuals with a high risk of disease are identified, clinicians can use appropriate surveillance measures to detect early disease onset.

When individuals are initially surveyed, oncologists often find that they do not fit the criteria to enter active surveillance for all cancers. For instance, if someone is identified as a BRCA carrier in their 20s or 30s, it is unlikely they will be diagnosed with pancreatic cancer at that time. However, since breast and ovarian cancers are more likely to develop during this time, conducting regular, cancer-specific screenings is essential. As a person enters their 50s and 60s, pancreatic cancer in BRCA carriers is more likely to occur, and therefore, it may be appropriate to initiate surveillance around 35 to 45 years of age. The primary objective of predictive cancer diagnostics is to enable early-stage detection without burdening people with unnecessary testing.

In a recent study published in Nature Medicine, an AI model was developed and trained with the goal of predicting pancreatic cancer risk three years before diagnosis based on a sequence of specific conditions subjects experienced over time. Implementing AI on actual clinical records could aid the development of a more adaptable workflow that would enable larger populations to be identified for surveillance.

Beyond risk-stratifying people, it is vital to have tools that accurately detect cancer. Historically, the lack of a clinically useful biomarker assay has challenged reliable detection. Recent studies conducted by Biological Dynamics show the effectiveness of next-generation biomarkers and advanced technology for early disease detection. Promising blood-based assays are now enabling higher sensitivity to early-stage disease using exosome-based biomarkers circulating in the blood. Combining AI identification and surveillance of high-risk individuals with these early detection tests has the potential to change our standard of care for those who need it most.

The future of surveillance and early detection
Traditional public health models offer general wellness guidelines to large, undifferentiated populations based on exposure and health history. Looking toward the future, health directives will become more tailored to individuals based on factors like family history, geographic location, diet, daily activities, and environmental exposures. The next wave of personalized medicine will also likely include genetic assessments as a routine part of medical and wellness care.

The use of AI to identify high-risk subjects, in conjunction with advanced diagnostic assays for ongoing surveillance, is bringing a promising new era to the fight against challenging diseases—each technology complementing the other and strengthening the chances of people receiving successful outcomes and a brighter future.

About the author: Dr. Paul R. Billings, MD, Ph.D., FACP, FACMGG, is the CEO and director of Biological Dynamics. He is devoted to studying and teaching medicine and genetics while accelerating the use of a broad range of novel genomic technologies in clinical settings. Over his decades in healthcare, he has established key business partnerships, driving the adoption of innovative discoveries and commercial success.