Biologic Treatment Responses in Psoriasis

Juul van den Reek, MD, PhD, is a physician-researcher at Radboud University Medical Center, Nijmegen, the Netherlands. Dr van den Reek has published multiple studies on psoriasis, particularly on analyses for psoriasis responses to biologics. She has also studied many other topics, including atopic dermatitis, nonmelanoma skin cancer, and localized scleroderma. In this insightful interview, Dr van den Reek discusses her commentary titled, “Data-driven Prediction of Biologic Treatment Responses in Psoriasis: Steps Towards Precision Medicine,”¹ which discussed the findings presented in a recent study conducted by Geifman et al.²

How does predicting treatment success through patient-specific trajectories decrease the number of nonresponders per biologic despite the increase in treatment options for psoriasis?
The newest drugs are, indeed, very promising. Especially in the trials you see very high efficacy rates, but in practice, the effectiveness is always somewhat lower. We still have many patients that we cannot perfectly help. It is therefore still important to predict success of biologics despite the new promising drugs.

What we also know—and we don't have a very good explanation for—is that even when patients respond well initially, every year a part of patients will stop responding. We have to think of another drug when this happens, and predictors would be helpful to make the next choice.

How exactly can trajectory modelling techniques help outline individual-level medication-utilization trajectories or time-varying exposures?
This is a very new approach that has not been done often in psoriasis literature, but we know that the response to biologics, or response to any drug, is very heterogeneous among patients with psoriasis.

Normally, we look at the question "Can gene X predict if patients respond to drug Y?" It is a binary outcome, and we often say, "If they have a PASI90 [Psoriasis Area and Severity Index score decrease of 90% compared to baseline], they are a responder at some point in time." However, the response in time changes and this approach is too simple and static. With trajectory modeling, they can look at the response over time. It is more a dynamic measure of treatment response, and it provides a lot more information to assess different responder groups.

Based on the study of Geifman et al,² how does the sensitivity analysis on adalimumab suggest that future studies on biologic-specific trajectories might be needed to determine different treatment responses over time?
It would be very informative if different drugs have different trajectories. We can also use it for prediction of response, but in the article by Geifman et al,² they did not see a big difference between the drugs. They all had similar trajectories.

If you have large groups per biologic, it’s interesting to study if there are differences in the trajectories between biologics, but we also need much more data on the newest biologics. That will take a lot of time.

Also based on Geifman et al,² how can future analyses on additional molecular and pharmacologic data define subgroups and lead to more accurate predictions in treatment responses?
Since psoriasis is a multifactorial disease and the pharmacological metabolism differs between patients, it's sensible to also incorporate variables like molecular and pharmacologic parameters. In the end, that would be the way to predict treatment response in which you look at many aspects of the disease.

I know that many investigators already look at molecular predictors and pharmacologic data to predict treatment response, but as stated, we now often look at a binary response outcome at one time point. If in the future we can add many other molecular data and pharmacologic data to the analysis that Geifman² proposed, that would be more informative in the end.

Are there any tips or insights you’d like to share with your dermatology colleagues regarding biologic treatment responses and psoriasis?
Sometimes, people say for psoriasis that “we're already there” because we have so many treatment options, but that is not the case. Everyone who sees a lot of patients with psoriasis and treats them with biologics knows that we still have patients that do not or insufficiently respond.

Another thing that is important, and may seem a bit contradictory, is if you have so many options, especially for people who do not prescribe biologics that much, it's difficult to choose a biologic. Research on treatment predictors provides guidance to dermatologists, especially in a field in which we have so many treatment options, to still choose the best option for the individual patients.   

Reference

1. van der Schoot LS, van den Reek J. Data-driven prediction of biologic treatment responses in psoriasis: steps towards precision medicine. Br J Dermatol. Published online August 1, 2021. doi:10.1111/bjd.20625

2. Geifman N, Azadbakht N, Zeng J, et al. Defining trajectories of response in patients with psoriasis treated with biologic therapies. Br J Dermatol. Published online April 7, 2021. doi:10.1111/bjd.20140