By Ian Chuang, MD, MS, CCFP, Chief Medical Officer, EMEALAAP Health at Elsevier
The status quo
Controlled Clinical Trial. Just by the name, we can see its juxtaposition from medical treatment in the real-world. Patients in controlled trials are carefully selected based on matching a set of criteria. Not knowing whether the treatment is likely to work, we are comfortable blinding the patient and potentially withholding treatment or using a placebo. At the end of the clinical trial, we are able to conclude that for a population of patients within a set of criteria, the treatment did or didn’t work more effectively or more often than the alternative.
When it comes to patient care, from both the physician’s and patient’s perspectives, the question is more specifically whether the recommended treatment is going to work for this specific individual. Using our existing methods, we are taking insights at a population level and having to make decisions for the individual. This model and approach is far from precise.
When physicians take population level research and apply it to the individual patient, that patient will match some high-level clinical criteria centred on the diagnosis, severity, and a few patient attributes. Beyond that, the treatment is really a therapeutic trial.
Traditionally, we thought of that as building experience of the treatment. The more we treat, the more experience we have with all the different patients that exist to uncover who it works best for, and how. Population-wide, we are now able to leverage the large real-world treatment data along with better computing power and approaches to uncover the nuanced patient attributes and treatment efficacy. The premise is not new, for decades we have looked to mine pools of data to understand how to more effectively treat our patients. The pharmaceutical industry has been a great proponent of this through Phase IV studies. However, all too often the data lived in silos and the process takes years of data mining to generate.
Being equipped to face the challenges of medicine
Remaining at the status quo will not address the significant challenges facing healthcare practitioners, hospital managers, and health systems globally today. We must find a way to interlink real-world data and practice faster than previously. Health tech is what sits at the frontier between evidence-based data and real-world data. Health tech combines the evidence from clinical research data and real-world data obtained from practical care, to identify optimum treatment in every variation that can be present in a patient across situations and scenarios the physician will never have had personal experience with.
Let’s look at oncology as an example. Life-threatening cancer often has the least luxury of time for trial and error with treatment. We suspected for a long time that cancers of a particular type are not all the same, however we didn’t know where and how they were different, and more importantly, we didn’t know how to treat them differently. Now, gene sequencing is revealing that certain genetic findings actually determine the body’s or tumour’s response to treatment. In this example, practice-based knowledge is about the real-world data from patients who responded favourably as well as those who didn’t. Precision medicine is then about physicians learning from and leveraging this knowledge to save cost by reducing the trial and error routine, provide a more precise care plan for their specific patient, and quite possibly save the life of that patient with a more timely and effective action.
This type of practice-based learning is leveraging the data from patients who responded favourably as well as those who didn’t across larger populations than a single physician, hospital, or system can possibly encounter. Unlike Controlled Clinical Trials, the more variables available to analyse and the larger the pool of patients to analyse, the better. Only through technology is it practical to churn through the big data. As such, technology is the enabler. It allows physician to make use of all the data available to them, evidence-based or real-world. With it, we can leverage the large real-world approaches to uncover the nuanced patient attributes. Insight about a broad group based on diagnosis can be refined to smaller cohorts based on other factors that can impact response.
Generating value through real-world data and technology
From a healthcare financing and cost perspective, technology has strong cost implications as well. The waste of both treatments and the associated cost from therapeutic trial and error with treatments is neither cost effective nor efficient. Unused medications are wasted, and the net cost is the sum of the failed treatment plus the cost of the eventual treatment that actually works. Precision in diagnostic and treatment now reduces the rate of non-response and can improve the cost benefit analysis of new treatments.
The breadth of treatment options only expands with each new discovery. The financial cost to achieve the desired treatment outcomes cannot follow the prices of new therapies. Without precision knowledge, the clinical risk, the associated cost, and the emotional toll on patients with such broad based therapeutic treatment trial in real world practice is not sustainable. The opportunity is to better slot treatment options to align to what known genotypic and phenotypic traits are better predictors of treatment response and achieve the desired outcomes.
Precision medicine insights and capabilities shouldn’t deter or discourage pharma from continuing to research and seek novel treatment options. However, these expensive treatments should be judiciously used where we know they will more likely work. The healthcare systems cannot sustain sub-optimal or ineffective uses of high-cost therapies. As such, pharma’s financial success in the future will be more dependent on their efforts towards uncovering the precision use of treatment of highest known efficacy.
Health technology and real-world data are intricately interlinked with the future of medicine, and it is a future to be welcomed. An ongoing challenge will be to ensure that we combine data so that sufficient pools of information can be created and high-quality evidence can be generated. Through this collaboration we will begin the process of democratising data, meaning that we will ensure that knowledge based on evidence from the real world is available to each and every physician, enabling physicians to make equally informed care decisions. That will require more than simply pooling: it will require vastly complex data management that only technology can provide, and due to the importance of what is at stake, trust in the provider who is holding and managing the technology will be crucial for success.