By Gerald L. Messerschmidt, M.D., FACP, Chief Medical Officer at Precision for Medicine, Oncology and Rare Disease
The futuristic end-concept of expansive virtual clinical trials is an exciting one for modeling study conduct, product development, and eventual approval. Steps toward this end are beginning to appear in several predictable situations. Clinical study procedures and data collection can be done remotely, and eligibility criteria, adverse events, concomitant medications, in-home blood sampling for basic safety labs and pharmacokinetics, patient diaries, and patient-reported outcomes (PROs) are a few of the data points being collected and electronically entered outside of the clinical setting.
The technology of wearable devices is becoming more and more sophisticated to allow for collection of information of a number of human functions. These and future wearable and home devices require cutting-edge electronic infrastructure with accompanying telecommunication services for appropriate and regulated collection into current and future electronic databases.
However, virtual trials are still in their infancy. The word virtual connotes to “not physically exist as such but made by software to appear to do so.” This is a futuristic goal of modeling the normal or ill human biophysical state and calculating changes or “outcomes” with various medical interventions, eg, drugs or biologics. Our knowledge of the normal and biophysically abnormal person is growing rapidly, yet we have years, perhaps decades, of deeper understanding to reach this goal across medicine.
Step-by-step, however, we are making exciting progress. We are measuring blood glucose levels every few minutes in patients, and these are being electronically collected. It is only a matter of time before chemical, hematologic, and specialty blood measurements are collected via sophisticated wearable systems for analysis by a technician or nurse and ready for physician review and decision-making.
Exploratory electronic algorithms have been developed that can analyze and implement rapid patient interventions within limited parameters. The National Comprehensive Cancer Network has created “clinical practice guidelines appropriate for use by patients, clinicians, and other healthcare decision makers around the world.” The NCCN, an alliance of 28 leading cancer centers, supplies key opinion leaders and experts who discuss many cancers, current treatments, and new therapy data to develop comprehensive decision-making flowcharts and guidelines. Initial and periodic patient history, physical examinations, and laboratory, imaging, and other testing are included—and all being collected electronically.
These systems, though a fantastic step forward, often cannot communicate across patients or platforms and much of the data is not even electronically searchable. Human interface is required in nearly all healthcare electronic data input and review. This recognized hurdle is being worked on and hopefully will be rectified over the next decade. So many systems and so many users and so many patients are impacting this progress across the healthcare sector in the United States alone. As can be deduced from the NCCN’s approach, as well as today’s relatively simple electronic data capture, even complicated treatment decisions are possible in the next few decades by electronic data analysis and decision-making programs, building on those currently in use.
Clinical trials will be able to collect electronic data progressively over the next 5 to 10 years from the source patient electronic medical record, or EMR. For trials to move to “virtual status” more and more data needs to be collected from patients in a real-world environment and transmitted, along with intervention information, to an electronic collection environment. Virtual clinical trials may be possible in well-known disease states, such as specific infections in normal patients, in the next decade or so.
However, in situations where the human biophysiology is not completely understood, such data will be required, eg, immune system normal functioning. Or where the biophysiology of the abnormal human system is not known, such as cancer clones or cytokine expression and functioning, deep knowledge will be needed to program or “teach” the electronic systems. Thus, while complicated biophysiologic virtual clinical trials are still decades away, with continued basic science progress, coupled with electronic development toward artificial intelligence improvements that can read and combine widely diverse literature, virtual trials may become a reality among the next generation of scientists, physicians, nurses, and other healthcare specialists.