Now that the pandemic is a full year old, I’m starting to find that certain topics that I wrote about early in the pandemic, only to forget about them for nearly a year, are starting to pop up again, like bad science or pseudoscience zombies coming to life after having been killed, à la the walkers in The Walking Dead. So it was that recently a “reader” (if you can call him that) sent me an indignant email about a post that I had written eleven months ago about an improbable bit of technology being touted at the time called the Healight. Basically, Healight is a device that would shine UV-A radiation into the lungs and thereby kill SARS-CoV-2, the coronavirus that causes COVID-19. Now, nearly a year later, the company is touting a clinical study to show that its device works. it’s a study that is—shall we say?—not particularly compelling. However, before I get to the study, itself, let’s go on a trip down memory lane, back to the early days of the COVID-19 pandemic, a time when a device like Healight might have seemed not so wacky.
You might recall an incident in April last year when our former President Donald Trump “just asked questions” about using ultraviolet light or disinfectants internally—yes, internally—to treat COVID-19. So divorced from reality were Trump’s statements and wild speculations that they earned a bit of the ol’ not-so-Respectable Insolence and even led to Trump’s disappearing from the daily coronavirus briefings that had been so transfixing the nation (and not in a good way) up until then. In fact, I can’t help but repost this rather iconic reaction of Dr. Deborah Birx, as her President said these things:
What a difference a year makes! And, for once, actually in a good way!
Those wild and crazy early days (and not in a good way) of the pandemic aside, there actually was a device proposed whose inventors appeared to take advantage of the kerfuffle. Dubbed the “Healight,” the device was intended to treat COVID-19 by shining UV light internally into the lungs because, you know, UV light inactivates most viruses (and can kill many bacteria) and therefore you should be able to inactivate SARS-CoV-2, the coronavirus that causes COVID-19, by shining light into the lungs, right? At the time, I even noted that the company that had developed Healight, Aytu Bioscience, its CEO Josh Disbrow, and its scientists Drs. Mark Pimentel, Ruchi Mathur, Gil Melmed, and Ali Rezaie, who were working with Cedars Sinai Medical Center in Los Angeles had developed a catheter with an LED emitting UV-A light. They had even presented an abstract at a gastroenterology conference testing their device in the colon to kill pathogenic bacteria and in a press release were proudly touting their plan to apply the device to the treatment of COVID-19.
At the time, I explained why Healight was so implausible as a COVID-19 treatment, even assuming that it worked exactly as claimed by Aytu (in vitro, at least) and killed only cells infected with SARS-CoV-2 while sparing surrounding normal cells. (Never mind that the majority of the cells in the respiratory epithelial lining of the bronchial passages are likely to be infected when the disease is severe enough to require intubation, the method by which Healight is designed to be introduced into the lungs—wait, sorry, I was trying to give Healight the benefit of the doubt.) I even further assumed that Healight does the same thing that Aytu claimed it did in cell culture when the light is inserted into the trachea via endotracheal tube hooked up to the ventilator. First of all, shining a light inside the trachea could potentially kill all those infected cells within range of the light en masse, rather than in a random fashion. What does that mean? It means that, instead of cells lining the inside of the trachea and bronchial tubes being killed over time by the immune system and through dying as virus is released, there would likely be a wave of cell death over a much shorter period of time, thus greatly exacerbating the inflammatory response that has damaged the lungs to the point where the patient needs to be on a ventilator.
Of course, that was also being generous and assuming that the light’s selectivity for virus-infected cells would perfect and that it wouldn’t damage normal cells as well. In medicine, there is no such intervention, be it drug or light (as in radiation therapy) that is perfectly selective for the target cells. Radiation therapy is selective for cancer cells because they replicate faster than normal cells and their DNA repair mechanisms are impaired, but still can damage and destroy normal cells that are replicating. The same is true of chemotherapy. The same, I thought, would be likely true of Healight.
I also noted another big problem with Healight, and it’s an obvious one. In brief, the Healight catheter only extends to the upper reaches of the trachea. It doesn’t even go past the bifurcation of the trachea into the right and left mainstem bronchi. Of course, I’m sure it could be pushed further down into the mainstem bronchi and even maybe into some of the larger bronchi that branch off from the mainstem bronchi. However, even if it were possible to do that and that the light could kill all the coronavirus-containing infected cells lining the trachea and the larger bronchi, given how many branches there are, it would be a painfully tedious process to treat them all. Of course, it isn’t just the tracheal and bronchial epithelial cells that are the target of SARS-CoV-2. It’s the type 1 and 2 pneumocytes, in the alveoli of the lungs. (The alveoli are the air sacs where oxygen and CO2 exchange occurs.) I saw no way that a catheter could get that far down the respiratory passages and to kill infected cells in the alveoli, and even if it could it would likely result in more fluid, more inflammation, and worse gas exchange. Again, that was assuming that the Healight works exactly as Aytu Bioscience claimed that it could. Worse, even back then we knew that COVID-19 was a multi system disease, with the virus infecting more than just the lung. What about all the other organs SARS-CoV-2 could potentially infect?
All of this brings us to a press release from last week:
Aytu BioScience , Inc. (NASDAQ:AYTU), a specialty pharmaceutical company focused on commercializing novel products that address significant patient needs, announced today that data from a first in-human, open label, clinical trial in SARS-CoV-2 patients has been released.
The pre-print publication titled “Endotracheal application of ultraviolet A light in critically ill severe acute respiratory syndrome coronavirus-2 patients: A first-in-human study” concluded that endotracheal UVA light treatment was associated with a significant reduction of SARS-CoV-2 viral load and improvement in WHO clinical severity scores. Additionally, the endotracheal UVA light treatment did not result in any serious adverse device effects and was well tolerated.
And here’s the Healight study, published as a preprint on medRxiv. It is every bit as…impressive…as the press release says. (Yes, that’s sarcasm.) I couldn’t resist heading over to ClincalTrials.gov to look for the study, UVA Light Device to Treat COVID-19, noting that the study didn’t open until October 1, 2020 and that it took two months to recruit five patients. That made me ask: Why did it take two months to recruit five patients during the worst period of the pandemic, when Southern California hospitals were buckling under the load of COVID-19 cases, which by December health officials were characterizing as a “surge on top of a surge“? It should have been easy to line up five patients in a single week, much less over two months.
Far be it from me to be too critical about not being able to accrue enough patients for a clinical trial, having had my share of failures on that score myself over the course of my career, though; so I won’t dwell on this issue. (I know. It’s hard.) So let’s move on to the paper itself. The inclusion criteria included: age over 18 years, positive PCR test result for SARS-CoV-2 on nasal swab, and mechanical ventilation with an endotracheal tube inner diameter of ≥7.5 mm. Pregnant women were excluded. Subjects received all standard supportive care; concomitant use of any other COVID-19 treatments was permitted. In other words, patients received whatever care patients at the time were receiving.
Here’s a description of the intervention:
Within 24 hours of enrollment, subjects underwent 20 minutes of endotracheal UVA therapy, which was repeated once daily for a total of 5 consecutive days. All subjects received 100% FiO2 for 30 minutes prior to the procedure (see Supplemental Materials and Methods for protocol). The UVA catheter was inserted to the distal end of the ETT, with concomitant ventilator adjustments to flow rate and tidal volume to maintain optimal oxygenation. A plastic clamp fixed the catheter base to the access port to assure stability and consistent depth of catheter insertion throughout the 20-minute treatment session. The procedural instructional video can be accessed at: https://cedars.box.com/s/0lqm2slw1vlyt4j5em70xregfro4592t. UVA dosing was chosen based on the optimal response of coronavirus 229E-infected human primary tracheal cells to UVA exposure observed in in vitro experiments. Controlled narrow-band UVA emission (peak wavelength 340-345nm) of maximum 2 milliwatt/cm2 was delivered at the level of tracheal mucosa. Predetermined criteria for treatment cessation and withdrawal of the UVA catheter included O2 saturation drop below 88% or hemodynamic instability.
The first thing to note is that, again, the catheter was only inserted to the distal end of the endotracheal tube. Endotracheal tubes are inserted so that their distal end is above the bifurcation of the trachea to the two mainstem bronchi, each of which heads to one lung. As a result, the light could not possibly have penetrated very deep into the lungs, down to the alveoli, consistent with my previous assessment.
The second thing to note is that this is a very small case series, with no control group. There’s nothing wrong with that as a proof of principle, but did this even prove the principle behind the Healight? Let’s look at the reported endpoints, first the primary endpoints:
Subjects had elevated viral loads at baseline (range 3.4×104 -1.64×107 copies/ml) except for study subject 2 who had an undetectable viral load at all time points, demonstrating that virus had cleared since the last nasal swab (Fig. S2). There was no significant correlation between symptom onset date and either baseline (Spearman rho=-0.70, p=0.23) or day 6 viral loads (Spearman rho=-0.21, p=0.83).
There was a significant reduction of SARS-CoV-2 levels in endotracheal aspirates during UVA treatment. The average log10 changes in endotracheal viral load from baseline to day 5 and day 6 were -2.41 (range -1.16 to -4.54; Friedman p=0.002) and -3.2 (range -1.2 to -6.77; Friedman p<0.001), respectively (Fig. 2, Fig. S2).
I note that subject 2 was the only one of the five patients who died. This patient apparently developed bleeding in the brain because of anticoagulation due to having been placed on extracorporeal membrane oxygenation (ECMO) and was placed on comfort care.
Secondary endpoints were:
Among the secondary outcome measures, quantification of absolute endotracheal bacterial load at baseline ranged from 1×103 -1.7x 106 CFU/ml and remained statistically unchanged during the UVA treatment sessions (Fig. S3).
In other words, Healight had no detectable effect on bacterial load in the lungs. The authors also report a correlation with the observed decrease in coronavirus in the lungs and WHO severity scores by day 30 and try their best to make it sound as though this correlation equals causation. Unfortunately, given such a small number of patients, it’s impossible to know if this correlation is real or spurious, and, more importantly, without a control group it’s impossible to know if viral load wouldn’t have decreased anyway as the patient started to recover from severe COVID-19. Presumably, it would have in patients who got better, or at least have declined faster in such patients than in patients who did not.
So what do I say about Healight now? Certainly, I’m nowhere near as impressed as the “reader” (in reality someone who was likely Googling “Healight” and came across my old post on the subject) was. The best that can be said of this study is that it does indeed show that, at least, Healight is not dangerous and can be used on intubated COVID-19 patients. A sober assessment of its results is that it demonstrates nothing particularly compelling. Adding to that the implausibility of the principles behind the device, I can only conclude that this study is what I like to refer to as underwhelming. It is not compelling evidence, over a year into the pandemic, that a breakthrough has been made in treating critically ill COVID-19 patients on a ventilator. As time goes on and more and more people are vaccinated, thus decreasing (it is fervently hoped) the number of COVID-19 patients requiring mechanical ventilation to very low levels, one can’t help but wonder what the rationale is for such a device any more.