On Movember, a month dedicated to raising awareness of men’s health issues, it’s crucial to highlight the significant strides being made toward improving outcomes for those affected. Movember’s mission to “change the face of men’s health” aims to create awareness and transform how we understand and treat conditions such as prostate cancer.
Globally, prostate cancer remains one of the most significant health challenges faced by men. Fortunately, advancements in data science are offering promising solutions to improve outcomes. While traditional methods of tracking prostate cancer progression through structured data in electronic health records (EHRs) have provided valuable insights, they often miss critical details needed for personalized treatment. Much of the vital information, such as disease-specific characteristics and nuanced patient responses, remains buried in unstructured clinical notes, which can create gaps in our ability to effectively stratify patients by risk, tumor stage and grade, and optimize care.
Artificial intelligence (AI) and real-world data (RWD) are key to closing these gaps and revolutionizing prostate cancer treatment. By unlocking the potential of unstructured data, AI is enabling life sciences companies to identify patients more efficiently while accelerating the development of therapies tailored to specific patient profiles.
The data problem in prostate cancer
Despite advancements in health care, traditional methods of capturing prostate cancer progression through structured data in EHRs have significant limitations. While structured data (e.g., age, race, and comorbidities) can provide useful information, it often lacks the nuanced detail required to accurately stratify patients by risk and tumor stage and grade to tailor treatments. Important details about disease characteristics and how patients respond to treatments are often written in free-text or unstructured clinical notes instead of in specific data boxes.
AI and RWD help fill the gaps by leveraging advanced techniques (e.g., machine learning and natural language processing) to analyze unstructured clinical notes. AI-powered large language models can identify keywords and phrases indicative of metastatic status in patients with prostate cancer, leading to improved accuracy in stratifying patients for downstream analyses or clinical trial identification. For example, when Verana Health examined urologists’ EHR clinical notes for signs of cancer spreading, the team discovered five times more cases by expanding the search beyond the term “M1,” a standardized term that refers to a stage of cancer where it has metastasized, to include other keywords for prostate cancer. This capability can improve how treatments are developed and tested, potentially leading to more effective care.
Other key variables involved in the diagnosis and staging of prostate cancer are Gleason scores, which are based on biopsy samples and describe how aggressive cancer cells are, and prostate-specific antigen (PSA) levels, determined by a lab test used to track disease progression. These measures are not captured in standard medical claims databases, and since they are not in EHR-structured fields, they are not recorded the same way by every clinician. Attempts to extract these meaningful insights from unstructured datasets have historically required labor-intensive, manual searches that are neither efficient nor scalable. AI-driven models are required to help capture Gleason scores. By analyzing patterns of diagnosis and patient PSA levels over time, it’s possible to identify patients with localized cancer and evaluate treatment patterns and outcomes.
How AI improves trial design and treatment development
AI and RWD are revolutionizing the design of genitourinary oncology clinical trials and treatment development, especially in prostate cancer. One of the primary challenges with traditional clinical trial methods is the difficulty in identifying and recruiting the right participants. It’s estimated that “85 percent of clinical trials fail to recruit enough patients and 80 percent of trials are delayed due to recruitment problems.” Patients often have diverse disease profiles and treatment histories, which can make it hard to find suitable candidates to meet the requirements of many trials. This not only delays the study timeline but can also skew results, as the data may not accurately reflect the broader patient population.
Verana Health leverages AI and RWD to help life sciences companies pinpoint urology practices that treat patients with specific prostate cancer profiles. This targeted approach ensures that clinical trials are better aligned with the needs of various patient subgroups. The end result is more efficient study enrollment, as potential participants can be identified and approached more effectively.
Leveraging AI and RWD can also enhance the development of more personalized therapies. By tailoring treatments to specific patient profiles—such as genetic markers, disease stage, or treatment history—life sciences companies can improve the effectiveness of their drugs. This targeted approach accelerates the development process and increases the likelihood that the resulting therapies will benefit the intended patient population.
Tracking the patient journey for better outcomes
The ability to track the patient journey provides a comprehensive view of disease progression over time. Understanding the natural history of the disease, treatment patterns, and outcomes helps life sciences companies identify critical health care turning points, allowing for improved therapies at varying stages.
AI and RWD also enhance the understanding of treatment effectiveness, ensuring therapies are tailored to the clinical realities of patients as they navigate their treatment paths. By capturing the timing and sequence of events in prostate cancer progression, AI and RWD provide deeper insights into patient health trajectories.
Utilizing AI and RWD to track the clinical characteristics and demographics of patients who are receiving different treatments can help life sciences companies allocate marketing and distribution resources more effectively. This approach ensures that new therapies reach appropriate patient populations faster, optimizing resources to quickly bring treatments to market.
Transforming prostate cancer care
The integration of AI and RWD is reshaping how we understand, treat, and manage prostate cancer. Integrating both into the clinical trial design process can help address the limitations of traditional methods by facilitating a more precise, efficient, and patient-centered approach to drug development. This evolution is critical for advancing treatments in genitourinary oncology and ensuring that patients receive therapies that truly meet their needs.
As we recognize Movember, let’s commit to leveraging these advancements for a brighter future in men’s health.
Sujay Jadhav is a health care executive.
