Investigating the Role of Skin-Resident Memory T Cells in Cutaneous Delayed-Type Drug Hypersensitivity Reactions

Abstract

Cutaneous delayed-type drug hypersensitivity reactions (dtDHRs) span the range of clinical severity from mild rash that is self-limited to severe sloughing of skin and mucosal surfaces with high mortality and no proven treatment. T cells are generally thought to be the principal actors driving these reactions. However, basic mechanisms underlying them, such as the phenotype and function of the T cells involved, remain poorly understood. Research into this topic has long been limited by lack of access to specimens sufficient for laboratory analysis. This project investigates whether skin-resident memory T cells mediate dtDHRs, using innovative techniques to overcome previous barriers to research. To begin addressing this hypothesis, we first developed a robust database of clinically- and pathologically-confirmed cases of three types of skin dtDHR: Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis (SJS/TEN), drug reactions with eosinophilia and systemic symptoms (DRESS) syndrome, and morbilliform drug eruption (MDE). The database included extensive clinical data, which allowed us to conduct a nested clinical study on clinicians’ ability to identify culprit drugs correctly and the consequences of incorrect culprit drug identification. It revealed that culprit identification is complicated by patients taking multiple drugs (69% on 4+) before disease onset, clinicians used no testing or validated approaches to determine culprits, 40% of patients had concurrent infections that may have confounded a drug-induced etiology, and patients’ allergy lists subsequently contained possible inaccuracies that could have adverse consequences for the care provided to them and even for public health in general. We performed transcript analysis of 187 target genes using Nanostring on formalin-fixed, paraffin-embedded (FFPE) skin samples from adult and pediatric patients with MDE (n=6), DRESS (n=6), and SJS/TEN (n=13). Only SJS/TEN > 10% TBSA blistered skin were included to reduce diagnostic error. Healthy skin served as controls (n=10). Preliminary results revealed significantly increased CD3 and CD8, non-specific memory marker CD45R0, and central memory T cell marker CD62L in SJS/TEN and DRESS compared to healthy controls, but not in MDE samples. Skin-resident memory T cell markers CD69 and CD103 were not elevated in any DHR group. The dtDHRs largely demonstrated a Th1/Tc1 skewing. Microscopy confirmed that the majority of T cells (CD3+) were CD45RO+ in all three dtDHR types yet a minority of T cells were CD103+. T cells consisted of both CD4+ and CD8+ subsets and were largely CLA+. A group of patients with MDE were identified that were profoundly lymphopenic, indicating that they were nearly depleted of circulating T cells compared to healthy controls. These lymphopenic skin samples contained CD4+ and CD8+ T cell subsets that were predominantly CD45RO+ and CLA+, and were of equivalent numbers as healthy controls. These data suggest that skin-resident memory T cells can mediate MDE, but that central memory T cells are recruited to skin in SJS/TEN and DRESS. These findings may explain why MDE is largely skin-limited while SJS/TEN and DRESS involve multiple tissues/organs. This work provides a strong platform on which future basic and clinical investigations into dtDHRs can be conducted, and it offers an appealing framework for other skin researchers interested in extracting valuable information from rare cutaneous diseases.