Evaluation of Selected Laboratory Parameters as Predictive Biomarkers of Clinical Response to Dupilumab Therapy in Patients with Atopic Dermatitis

Background: Clinical response to dupilumab in patients with atopic dermatitis is variable, and no established laboratory biomarkers currently enable reliable identification of patients likely to achieve a predictable response prior to treatment initiation or early during therapy. Methods: This prospective, real-world study conducted in a Polish population evaluated the discriminative capacity of selected laboratory biomarkers for predicting short- and long-term clinical response to dupilumab in patients with atopic dermatitis. Results: Lactate dehydrogenase (LDH) was the most consistently informative continuous baseline biomarker (area under the receiver operating characteristic curve [AUC] for at least 75% improvement in the Eczema Area and Severity Index [EASI-75] at week 16: 0.60; EASI-90 at week 16: 0.63; EASI-75 at week 40: 0.66; EASI < 7 at week 40: 0.64). The systemic inflammatory response index (SIRI) was the second most stable baseline predictor (AUC = 0.63 and 0.64 for EASI-75 and EASI < 7 at week 40, respectively) and demonstrated the highest overall discriminative performance among week 16 biomarkers. Conclusions: LDH and SIRI appear to be the most promising candidates for further evaluation in larger cohorts owing to their cross-outcome and cross-time consistency rather than a single dominant AUC value. The predictive potential of individual biomarkers, while detectable, remained below the threshold of clinical utility for every biomarker–outcome combination examined. Composite multi-biomarker models may be required to achieve clinically meaningful predictive accuracy.