T Cell Involvement in the Pathogenesis of Drug‐induced Enterocolitis Syndrome

Recently, Mori et al. published three new cases of druginduced enterocolitis syndrome (DIES). A 6yearold boy (Case 1) was investigated for a suspected IgEmediated hypersensitivity reaction (angioedema of hands and feet) to amoxicillin/clavulanic acid (AMX/CL). He was orally provoked with AMX/CL with a graded challenge of 1/10 – 2/10 – 7/10 of the therapeutic dose every 30 minutes. After 2 hours of observation, the patient was discharged from the hospital in good conditions. After two hours and thirty minutes from the last drug intake, the child began vomiting, and his general condition deteriorated progressively. He was pale and lethargic, although vital signs were in range. For the severity of symptoms, he was admitted to our emergency care unit and needed intravenous saline solution infusion. During the acute phase, leukocytosis (20,480/μL; normal range 5000– 15,000/μL) with neutrophilia 13,170/μL (not eosinophilia) and an increased methemoglobin level (1.1%; normal value 0.2– 0.6%) were observed. In the following hours, he completely recovered and was discharged with the diagnosis of DIES to AMX/CL.1 It has been postulated that food protein– induced enterocolitis syndrome (FPIES) and DIES may share common pathogenetic aspects as they are both nonimmediate hypersensitivity reactions involving adaptive immunity in addition to other still unknown mechanisms.2 Here, we aimed to investigate the role of T cells in DIES by performing lymphocyte transformation test (LTT) toward AMX/CL and detecting cytokine production after hapten stimulation in haptenspecific Tcell lines. Skin prick tests (SPTs) and intradermal tests (IDTs) with AMX/ CL were completed. The mother solution and serial dilutions of the culprit drug were freshly prepared at the time of skin testing. IDTs were read after 20 minutes (immediate IDTs) and after 24– 48 hours (delayed IDTs) according to the European Network of Drug Allergy Guidelines. Skin tests were considered positive when the diameter was greater than 3 mm from the initial wheal or increased from the initial wheal in association with a flare and negative saline control. The patient underwent the drug provocation test (DPT) with the culprit drug the day after the completion of the skin tests. The DPT followed the following protocol: AMX/CL was orally administered at refracted doses (1/102/107/10 of the pediatric therapeutic dose per day [50/mg/kg/day]) every 30 minutes in an inpatient setting. The patient was observed for 2 hours after the last drug intake in case of a negative outcome. DPT was considered positive if, during the challenge or within 48 hours after the end of the challenge, any objective skin, respiratory and/or cardiovascular, neurologic, or gastrointestinal clinical manifestation was observed and documented with a picture by a physician or parents. In case of a reaction, the child was reevaluated in the allergy unit. The culture medium was VLERPMI 1640 (Biochrom GmbH) supplemented with 2 mm endotoxinfree Lglutamine, 1% nonessential amino acids, 1% sodium pyruvate (SigmaAldrich), and 2x10−5 M 2ME (Merck; complete medium). Penicillin (BP), ampicillin (AM), AMX (all from Sigma), and AMX/CL (GSK) were used; rIL2 (Proleukin) was from Novartis; the polyclonal activator phytohemagglutinin (PHA) was from Biochrom AG; and the recall antigen streptokinase (SK) was from CSL Behring. The LTT toward AMX/CL was performed 8 months afterward without the interference of corticosteroids. Peripheral blood mononuclear cells (PBMCs) were isolated by Lymphoprep (Sentinel Diagnostic). 2 × 105 cells were cultured in triplicate in roundbottomed 96well plates in a complete medium with 5% autologous serum in a final volume of 0.2 mL for 5 days in the presence of 4 increasing doses of individual βLs (BP and AM 2.5– 0.5– 0.1– 0.02 mg/ mL, AMX 1– 0.5– 0.1– 0.02 mg/mL, AMX/CL 0.5– 0.1– 0.02– 0.004 mg/ mL) or in a medium alone as a negative control at 37°C in 5% CO2humidified atmosphere. As positive controls, 2500 UI/mL SK and 1% vol/vol PHA were used. After 16hour pulsing with 0.5 μCi/well 3HTdR (PerkinElmer), the cultures were harvested, and radionuclide uptake was measured by scintillation counting. A stimulation index (SI) (ratio between the radioactivity from stimulated and unstimulated cultures) ≥3 was considered positive.3 Additionally, 1 × 106 PBMCs were cultured in the presence of a single βL (BP, AM, or AMX 0.5 mg/mL, AMX/CL 0.1 mg/mL) for 6 days. Activated T cells were then expanded for a further 8 days by adding rIL2 (25 U/mL). For the hapten specificity of Tcell lines, Tcell blasts were finally recovered, washed, counted, adjusted to 1 × 106/mL, and assessed for specificity by thymidine incorporation, cultured for 3 days in the presence of autologous irradiated PBMC (1:1 ratio) and the βL used for the induction (BP, AM, and AMX 0.5 mg/mL, AMX/CL 0.1 mg/mL) or crossreactive. SI was considered positive when ≥3. In assessing cytokine synthesis by the haptenspecific Tcell lines at the singlecell level, 1 × 106 Tcell blasts were polyclonally