By Michael N. Liebman, Ph.D., Managing Director, IPQ Analytics, Adjunct Professor, Pharmacology and Physiology, Drexel College of Medicine
Biomedical research is evolving with an increasing emphasis on data science, e.g., data integration and storage, data privacy and security, data analytics and data representation, driven by the transformative technologies that have become the currency of genomics in precision medicine. In spite of numerous “beachhead” successes, however, the gap between data and clinical utility continues to grow. I would propose that at least some of this reflects an imbalance between highly accurate and increasingly precise technology and the current limitations that exist within real world clinical practice and in the complexity of real world patients.
In this blog, I’ll describe some of the key issues that contribute to this disparity. I’d like this to be the starting point of discussions aimed at identifying limitations and solutions to close these gaps, and to begin engagement among the clinical and research communities as follow-up to my recent NCI CBIIT Speaker Series presentation.
As a systems-oriented person, I view disease as a process. An individual progresses from pre-disease risk, under the influence of disease, to present signs and symptoms that lead to diagnosis and treatment. These last two steps involve a physician’s application of existing guidelines and personal experience which result in prescribing treatment and the expectation that patients will adhere to this recommendation. The final step reflects the outcome of the process, although it only positions the “patient” for “recycling” into the next pre-disease risk phase. Issues that I see include:
Risk: There is a need to evolve from correlative to causal analysis to optimize both the opportunity to detect risk and the potential to mitigate it. This includes the recognition that factors that affect risk, such as smoking, may vary across physiological development of an individual because of the changes in accompanying biological processes and may also vary for specific conditions, e.g., breast cancer and heart disease, at different stages of development.
Diagnosis: Diagnosis requires greater understanding of the complexity of disease processes and evaluating the suitability of current guidelines and diagnostics to accurately describe them, especially regarding syndromes and complex diseases. Criteria for disease staging and stratification are typically not developed with an understanding of the underlying process and its pathway of progression, and may not be aligned with the critical determinants necessary for optimal patient management and early detection and/or prevention.
Treatments: Syndromes and complex diseases represent the majority of conditions that challenge the identification of best treatment for a specific patient but are not the only confounders. All patients present with a complex clinical history impacted by environment, lifestyle, and genetics. In addition, almost all patients present with multiple co-morbidities whether previously managed, current, or current and undiagnosed. Guidelines rarely address this level of complexity. In cancer, tumor heterogeneity is the rule, not the exception, and this impacts molecular analysis and resulting effectiveness of treatment.
Patient Adherence/Medication Adherence: Patients who do not fill prescriptions or follow medication instructions present a growing problem for themselves and their families, as well as for the whole healthcare ecosystem. The cost of this problem extends from personal health to major economic issues. To adequately address this issue we need to go beyond technologies that administer pills, mobile apps, telephone reminders, or reducing transportation and economic barriers. It is critical to understand that the initial communication between physician and patient concerning the diagnosis, treatment, and/or risks needs to be personalized to optimize the understanding, appreciation, and potential response of the patient.
Outcome: In consideration of catastrophic disease, it is not sufficient to focus solely on improved survival as the metric indicating advanced understanding of disease and enhanced treatment. Although a convenient and readily quantifiable measure, we need to examine and evaluate the quality of life of patients who are now “being managed,” the cost of their health maintenance and also whether we are actually having an impact on disease prevalence. We also need to consider hospital readmissions and subsequent risk of secondary disease as a result of primary treatment.
The successes of current efforts in personalized medicine deserve to be recognized but it is also critical to acknowledge the gap between our successes and our true understanding of disease and disease processes. The complexity comprising real-world clinical practice and real-world patients needs to be integrated into our planning and efforts to drive personalized medicine towards general practice.