Historically, test validation focused on the performance of the technology itself.
Jonathan E. Schmitz, Vanderbilt University Medical Center
Optical technologies have revolutionized many aspects of modern health care. From automated microscopy and digital pathology to diverse bioanalytic platforms, optics are central to laboratory diagnostics.
Molecular (i.e., DNA/RNA-based) diagnostics epitomize the value of photonic-driven technology. Nucleic acid amplification testing and next-generation sequencing rely largely upon fluorescence — such as real-time polymerase chain reaction and DNA sequencing-by-synthesis. Infectious diseases are a prominent application of these techniques, especially in light of the COVID-19 pandemic.
Yet the ability of clinical laboratories to generate vast quantities of data can outpace the ability of physicians to clearly interpret or act on it. For instance, it may be analytically straightforward to generate a next-generation sequencing profile of the commensal bacteria of a patient’s microbiome. But this information may not translate into guidance on treatment or lifestyle modifications.
Even when a diagnostic result carries a straightforward interpretation, it may not alter the course of a patient’s care. Highly multiplexed polymerase chain reaction panels can detect any of the numerous viruses that elicit respiratory infections. However, a majority of these infections typically represent different forms of, colloquially speaking, the common cold, without virus-specific treatment strategies. So, these panels do not often provide significant advantages compared with targeted polymerase chain reaction.
Historically, test validation focused on the performance of the technology itself: How does sensitivity or specificity compare to existing gold standards? This information is an important parameter that regulatory agencies such as the FDA consider when approving new tests. But the question of whether a new test correlates with improved outcomes does not directly enter this calculus.
This process contrasts starkly with the development and approval of new drugs, for which clinical outcomes, generally assessed through randomized trials, are paramount. In one sense, incorporating outcomes-based research into the diagnostic sphere may align it more closely with therapeutic development, but it can also entail additional significant complexities.
Who would be responsible for supporting outcomes-based research for next-generation diagnostics? Some manufacturers have devoted increased attention to the need for this data. But detailed outcomes studies are often the exception, and such studies remain outside the purview of many traditional funding mechanisms (i.e., government and foundation support).
A modest starting point is to recognize the complexity of the challenge. This includes the incorporation of stakeholders even at the early stages of test development, such as experts in the underlying optical technology. Advancing novel diagnostics from concept to care has always been a painstaking endeavor. And these technologies must be harnessed for both their analytical capabilities and their ability to drive clinical outcomes.
Meet the author
Jonathan E. Schmitz, M.D., Ph.D., D(ABMM), serves as both a clinical microbiologist and investigator of microbial pathogenesis. Schmitz provides service at Vanderbilt University Medical Center as an attending physician on the Clinical Microbiology Service, and he previously served for six years as the medical director of the Molecular Infectious Diseases Laboratory; email: [email protected].
The views expressed in ‘BioOpinion’ are solely those of the author and do not necessarily represent those of Photonics
Media. To submit a BioOpinion, send a few sentences outlining the proposed topic to [email protected].
Accepted submissions will be reviewed and edited for clarity, accuracy, length, and conformity to Photonics Media style.