Measure The Right Thing
That nagging question of why promising therapeutics failed FDA approval got me to thinking. So, I called up Wiley Chambers (Supervisory Medical officer at the FDA) out of the blue in 2015 and he was kind enough to chat with me. The FDA is tasked with an important job to review the submitted data to determine two things: 1) drug safety and 2) drug efficacy. Fair enough. Mighty glad for that, too.
Those metrics are chosen by the clinical trial designers, not the FDA. We’ve learned a lot on how to design great clinical trials and today’s Dry Eye Disease trials look diﬀerent than 18 years ago. I feel a bit for the folks that write FDA trial protocols. They are like the field goal kicker in the final seconds of the 4th quarter of a 2-point spread football game…it is a high stakes game of hero (FDA approval for your company) or zero (miss the goal and lose all that investment in product development).
Part of the challenge with Dry Eye Disease is the metrics we follow. Dry Eye is a slippery fish in murky waters, so why do we try to describe how far frogs fly? Yes, I know frogs don’t fly but I feel like that’s what we’re looking at in some of our traditional non-sensitive, non-specific study metrics. There goes that pesky sign-symptom disconnect again (mentioned on the TFOS DEWS II Report, page 639). If you can’t measure efficacy differences, it could mean that your therapeutic doesn’t work, or it could mean you need to switch to a different measuring stick.
New Generation of Trials
The metrics we currently use could sure be better as far as sensitivity and specificity are concerned. When it comes to Dry Eye, a measuring stick like Schirmer’s strips is much more descriptive when the river is nearly dry. Schirmer’s is most helpful when aqueous production is dramatically reduced as it is in later stages of disease as a consequence of loss of homeostasis, hyperosmolarity, inflammation and neurosensory abnormalities. Schirmer’s can also be used to measure response to neural stimulation devices (eg. Allergan’s new TrueTear®). How about staining? Is it specific enough to rely upon as a metric? Or is it just describing the erosion of the river banks?
Ideally, we would be testing the quality of the river water well before the volume visibly starts to change. Even better, what if we could also measure the health and diversity of the fish population in the river (growth factors, lactoferrin, etc)? Advanced point of care testing such as osmolarity (hyperosmolarity) and cytokine panel analysis (inflammation) could potentially lead to a clear snapshot of specific disease activity and also measure the response to therapy. (Dr. Chambers expressed interest to me in seeing biometric data as well as patient reported symptom data). Even ultrastructural changes with disease and response to therapy using technology like confocal microscopy are worth considering as clinical trial metrics for looking at therapeutics’ impact on homeostasis and neurosensory abnormalities.
We will still see the familiar and traditional Schirmer’s scores included in clinical trials in part because the FDA is used to seeing Schirmer’s data. However, Schirmer’s is an uncommonly performed test in clinical practice. Faster, commonly available point of care tests with much higher sensitivity and specificity and improved reflection on real time disease activity will help bridge the worlds of clinical trials and clinical practice. We need better measuring sticks. We need continued dedication to Good Clinical Practice Guidelines. We need a deeper understanding of Dry Eye Disease, it’s subtypes and how to measure the impacts of our therapeutics. Our patients need more tools. Let’s keep going, together.