Steven R. Feldman, MD, PhD
Silverberg and colleagues described a very well-designed, vehicle-controlled, randomized 8-week study of a topical formulation of a purified strain of Nitrosomonas eutropha, an ammonia-oxidizing bacterium. In theory, this bacterium may reduce Staphylococcus aureus. The study compared two concentrations of the bacterium vs vehicle delivered as a spray twice per day. Study participants were adults with AD affecting 10%-40% of body surface area.
The study found "meaningful" improvements in itch and objective signs of disease, with clear separation between both doses of the bacterial spray compared with vehicle. At week 4, about 23% of participants treated with the bacterium were clear or almost clear (with a 2-point improvement) compared with 12% in the vehicle group (for comparison, in a phase 2 study comparing topical ruxolitinib with 0.1% triamcinolone cream, there was a 25% clear or almost clear rate [with 2-point improvement] in the triamcinolone-treated individuals).
Though an "all-natural" bacterial approach to managing AD may be appealing to some, it sounded like magic to me. But this well-done study makes it seem like the bacterial approach could be more promising than I had thought. This study also reported about twice as many adverse events (including gastrointestinal issues) with the bacterium-treated participants compared with those who received vehicle, adding to my belief that the bacterial product has efficacy. Whether any other topical will be more effective and safer than is topical triamcinolone remains to be seen. I'm still pessimistic about topicals because of patients' poor adherence to topical treatment, but perhaps an easy-to-use spray that isn't associated with patients' fear of "steroids" will be helpful.
Chen and colleagues. They analyzed data on hundreds of thousands of patients with and without AD. Adults with AD had a "significantly increased risk" of developing venous thromboembolism compared with adults without AD. The huge sample size of their study seems compelling. That huge sample size allows detection of effects so small that they may be clinically insignificant.
They report that patients with AD had a venous thromboembolism at a rate of 1.05/1000 patients-years; the rate was 0.82 for patients without AD. From that, we can calculate that there would be an additional 23 patients with venous thromboembolism for every 100,000 patient-years or about one more venous thromboembolism in the AD group in every 4000 patient-years. Though the finding was statistically significant, I don't think it is clinically meaningful.
The authors correctly conclude that "vascular examination and consultation with the emergency department, cardiologists, or pulmonologists are indicated for patients with AD who present with relevant symptoms (eg, unexplained dyspnea, chest tightness, and limb swelling)." But it is probably also true that vascular examination and consultation with the emergency department, cardiologists, or pulmonologists are indicated for patients without AD who present with those symptoms. I think the authors might have been on solid ground if they had concluded that there was a statistically significant but clinically insignificant increased risk for venous thromboembolism in patients with AD.
Eichenfeld and colleagues examined the use of topical crisaborole once per day as a maintenance treatment for patients with mild to moderate AD. The study compared patients given topical crisaborole with those randomly assigned to vehicle. The active treatment was effective because topical crisaborole treated patients had longer times to the first flare following treatment and fewer flares over the 1 year of treatment. The differences were not huge, but I think they were clinically meaningful. I'm guessing that the topical crisaborole maintenance treatment would have been even more effective had it been used regularly. The study did not, as far as I could tell, assess how well the treatment was used.
An interesting aspect of this study is that it began with nearly 500 participants who started on twice daily topical crisaborole. The 270 patients who responded to the treatment (achieving clear or almost clear with at least a 2-point improvement) were enrolled in the 1-year maintenance phase. Thus, the participants in the maintenance phase were preselected for patients who respond to topical crisaborole. We don't know why they were responders (I, of course, expect it is because they selected for patients who are better than others are at using a topical treatment), but it may be best not to try to generalize these results and assume this form of maintenance treatment would work equally well in a population who achieve initial success with an oral therapy regimen (for example, a quick course of oral prednisone).
Dupilumab was a revolutionary treatment for AD. I didn't think that I'd ever see a more effective treatment. It's so safe too! It has been a first-line treatment for AD since its introduction. Now, we also have oral Janus kinase inhibitor options. Blauvelt and colleagues examined what happens when patients who have been on dupilumab are switched to a high dose (30 mg/d) of upadacitinib (the standard starting dose of upadacitinib is 15 mg/d). Though dupilumab is very effective, upadacitinib is more so. After 4 weeks of switching to upadacitinib, nearly half the patients were completely clear of AD compared with only 16.0% after 24 weeks of dupilumab! The authors point out, optimistically, that "No new safety risks were observed." Though there were no cancers, gastrointestinal perforations, major adverse cardiovascular events, or venous thromboembolic events, there were cases of eczema herpeticum and zoster in patients treated with upadacitinib. Having upadacitinib available for patients who fail dupilumab is a clear benefit; the role of upadacitinib before dupilumab seems less clear.
Patients doing great on dupilumab for AD may be wondering: Do I still need to take it every 2 weeks? Spekhorst and colleagues may have the answer. They describe the response to tapering dupilumab in patients who had been on the drug for at least 1 year with well-controlled disease for at least 6 months. Patients in the study then continued dupilumab with the longest possible dosing interval while maintaining control of their AD.
Generally, patients maintained good control of their AD, with only a small increase in mean disease severity and in concomitant use of topical steroids. For the patients who attempted prolongation, 83% successfully continued dupilumab treatment with a prolonged interval. Not at all surprisingly, the authors calculated that prolonging the interval between dosing led to large savings in cost.
One of the nice features of dupilumab treatment is that loss of response over time seems unusual. Perhaps there is a low propensity for forming antidrug antibodies when dupilumab is used in the standard every 2-week dosing regimen. I don't know whether antidrug antibodies would be more likely with the intermittent dosing regimen. But now that we have other good systemic treatment options for AD, losing dupilumab efficacy would not be as critical a problem as it used to be. I also want to point out that patients' adherence to injection treatment, though better than adherence to topicals, is far from perfect. It's likely that many patients have already been prolonging the interval between taking their treatments. If you want to know, just ask them. The way I like to phrase the question is: "Are you keeping the extra injectors you've accumulated refrigerated like you are supposed to?"