IFN-γ ELISpot-enabled machine learning for culprit drug identification in nonimmediate drug hypersensitivity.

BACKGROUND: Diagnosing drug-induced allergy, especially nonimmediate phenotypes, is challenging. Incorrect classifications have unwanted consequences. OBJECTIVE: We sought to evaluate the diagnostic utility of IFN-γ ELISpot and clinical parameters in predicting drug-induced nonimmediate hypersensitivity using machine learning. METHODS: The study recruited 393 patients. A positive patch test or drug provocation test (DPT) was used to define positive drug hypersensitivity. Various clinical factors were considered in developing random forest (RF) and logistic regression (LR) models. Performances were compared against the IFN-γ ELISpot-only model. RESULTS: Among the 102 patients who had 164 DPTs, most patients had severe cutaneous adverse reactions (35/102, 34.3%) and maculopapular exanthems (33/102, 32.4%). Common suspected drugs were antituberculosis drugs (46/164, 28.1%) and ß-lactams (42/164, 25.6%). Mean (SD) age of patients with DPT was 52.7 (20.8) years. IFN-γ ELISpot, fixed drug eruption, Naranjo categories, and nonsteroidal anti-inflammatory drugs were the most important features in all developed models. The RF and LR models had higher discriminating abilities. An IFN-γ ELISpot cutoff value of 16.0 spot-forming cells/106 PBMCs achieved 94.8% specificity and 57.1% sensitivity. Depending on clinical needs, optimal cutoff values for RF and LR models can be chosen to achieve either high specificity (0.41 for 96.1% specificity and 0.52 for 97.4% specificity, respectively) or high sensitivity (0.26 for 78.6% sensitivity and 0.37 for 71.4% sensitivity, respectively). CONCLUSIONS: IFN-γ ELISpot assay was valuable in identifying culprit drugs, whether used individually or incorporated in a prediction model. Performances of RF and LR models were comparable. Additional test datasets with DPT would be helpful to validate the model further.

A multidisciplinary approach to verify and de-label of drug allergic histories in a university hospital in Thailand: a retrospective descriptive study.

BACKGROUND: Mislabeling of drug allergic histories causes avoidable negative impacts on patients and healthcare system. Although multidisciplinary adverse drug reaction (ADR) services to verify and de-label drug allergic histories have been operated in particular hospitals in Thailand, their performances have not been reported. This research aimed to examine the effectiveness of verification of drug allergic history and de-labeling (VD) services of the physician-led multidisciplinary ADR clinic. METHODS: This research was a retrospective descriptive study. Medical charts of patients with at least one drug allergic history who received VD services at the multidisciplinary clinic between January 2017 to December 2018, were reviewed. Data on the history of drug allergy, VD services, and results were analyzed and presented using descriptive statistics. RESULTS: Seventy patients' charts were reviewed, and 171 unconfirmed drug allergic histories were identified. 79.53% of the reported reactions involved skin and soft tissues. The most found adverse skin reactions were maculopapular rash, pruritic and erythematous rash, and angioedema. The remaining 20.47% were systemic reactions which included drug reaction with eosinophilia and systemic symptoms (DRESS), anaphylaxis, and nausea/vomiting was the most prevalent. Antituberculosis, beta-lactam antibiotics, and nonsteroidal anti-inflammatory drugs (NSAIDs) were the most reported suspected drugs. Drug allergic history reviewing by physicians or pharmacists could confirm and de-label for 3 and 20 reactions, respectively. Seven and one reactions were confirmed by enzyme-linked immunospot assay and patch test, respectively. The provocation tests with the suspected or alternative drug were conducted in 64 reactions. Twelve reactions were confirmed, and 45 reactions were de-labeled. Totally, 65/171 (38.01%) allergic histories were successfully de-labeled, 23/171 (13.45%) were confirmed, and 83/171 (48.53%) were inconclusive. CONCLUSIONS: More than half of drug allergic histories were successfully confirmed or de-labeled by the multidisciplinary ADR team. The collaborative activities of various healthcare professionals, consisting of physicians, nurse, and pharmacists as presented in the study were effective in VD services and should be implemented in other healthcare settings.

Beta-lactam hypersensitivity diagnosis in ambulatory and hospitalized settings require different approaches.

BACKGROUND: Data on beta-lactam hypersensitivity (BLH) are mainly focused on immediate or mild nonimmediate reactions in the ambulatory setting, but limited in patients with concurrent illness and moderate-to-severe nonimmediate reactions in the hospitalized setting. OBJECTIVE: To investigate the entire spectrum of BLH in Thai tertiary hospital. METHODS: Clinical characteristics of 357 patients with suspected BLH were evaluated in a 7-year period. Culprit drug identification was performed in 335 patients by combined skin testing, in vitro testing, or drug provocation tests. RESULTS: The predominant BLH presentations were non-immunoglobulin (Ig)E-mediated reactions with severe cutaneous adverse reactions of 18.9%, and BLH status was definitively confirmed in 18.1%. The most common verified culprits were cephalosporins (34.8%), particularly in hypersensitivity type IV reactions. Natural penicillins were the main implicated drugs in 48.5% of ambulatory patients. In contrast, cephalosporins and carbapenems were the main implicated drugs in hospitalized patients. Non-IgE-mediated anaphylaxis and serum sickness-like reaction remained diagnostically challenged. New generations of beta-lactams, hospitalized patients, recent allergic history, and underlying malignancies or autoimmune diseases were associated with increased BLH risk. CONCLUSION: At present, cephalosporins are the leading causes of BLH, particularly in non-IgE-mediated reactions. More research on the verification of non-IgE hypersensitivity reactions from new generations of beta-lactams should be better emphasized. CLINICAL TRIAL REGISTRATION: The registry was approved by the Ethics and Research Committee of the Faculty of Medicine, Chulalongkorn University, and listed on ClinicalTrials.gov (Identifier: NCT01667055; https://www. CLINICALTRIALS: gov/ct2/show/NCT01667055).

Clinical Characteristics, Urinary Leukotriene E4 Levels, and Aspirin Desensitization Results in Patients With NSAID-Induced Blended Reactions.

PURPOSE: Data on non-steroidal anti-inflammatory drug (NSAID) hypersensitivity in Southeast Asia are scarce. Increased urinary leukotriene E4 (uLTE4) levels have been suggested as a biomarker of NSAID-exacerbated respiratory disease (NERD). This study investigated clinical patterns of NSAID sensitivity in Thailand and the diagnostic roles of uLTE4 measurement in various phenotypes. METHODS: The clinical phenotypes in 92 Thai adults with cross-reactive NSAID hypersensitivity were characterized based on the clinical history and drug provocation. The uLTE4 levels were measured at baseline, after aspirin provocation and after desensitization. RESULTS: More than half of the patients (56.5%) presented with cutaneous symptoms (NSAID-exacerbated cutaneous disease), while one-third (33.7%) developed symptoms in at least 2 systems (NSAID-induced blended reactions; NIBR). Fifty-two patients underwent drug provocation and 59.6% of them yielded positive results. After drug provocation, a significant number of patients with confirmed NSAID cross-reactivity experienced clinical symptoms in more than one organ system. The uLTE4 levels at baseline were comparable between the NSAID-tolerant and NSAID-sensitive groups, but were substantially increased after aspirin provocation predominantly in NERD (983.4 pg/mg creatinine) and NIBR (501.0 pg/mg creatinine) compared to NSAID-tolerant subjects (122.1 pg/mg creatinine, P < 0.01 and 0.05, respectively). The uLTE4 levels were elevated after aspirin desensitization, although nasal polyposis and asthma were under control in 3 NERD and 3 NIBR subjects. CONCLUSIONS: NIBR is not uncommon among NSAID-sensitive patients in Thailand. The diagnostic value of basal uLTE4 levels was limited, but increased uLTE4 levels upon aspirin provocation suggest NSAID cross-reactivity with respiratory components. This study indicates that aspirin desensitization, if necessary, might be effective in both NERD and NIBR. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03849625.

Reliability and validity of the Thai Drug Hypersensitivity Quality of Life Questionnaire: a multi-center study.

OBJECTIVE: To adapted the Drug Hypersensitivity Quality of Life (DrHy-Q) Questionnaire from Italian into Thai and assessed its validity and reliability. DESIGN: Prospectively recruited during January 2012-May 2017. SETTING: Multicenter; six Thai tertiary university hospitals. STUDY PARTICIPANTS: Total of 306 patients with physician-diagnosed drug hypersensitivity. INTERVENTIONS: Internal consistency and test-retest reliability were evaluated among 68 participants using Cronbach's ɑ and intra-class correlation coefficient (ICC). The validity of Thai DrHy-Q was assessed among 306 participants who completed World Health Organization Quality of Life-BREF (WHOQOL-BREF-THAI). Construct and divergent validities were assessed for Thai DrHy-Q. Known-groups validity assessing discriminating ability was conducted in Thai DrHy-Q and WHOQOL-BREF-THAI. MAIN OUTCOME MEASURES: Validity; reliability; single vs. multiple drug allergy; non-severe cutaneous adverse reactions (SCAR) vs. SCAR. RESULTS: Thai DrHy-Q showed good reliability (Cronbach's ɑ = 0.94 and ICC = 0.8). Unidimensional factor structure was established by confirmatory factor analysis (CFI&TLI = 0.999, RMSEA = 0.02). Divergent validity was confirmed by weak correlation between Thai DrHy-Q and WHOQOL-BREF-THAI domains (Pearson's r = -0.41 to -0.19). Known-groups validity of Thai DrHy-Q was confirmed with significant difference between patients with and without life-threatening SCAR (P = 0.02) and patients with multiple implicated drug classes vs. those with one class (P < 0.01); while WHOQOL-BREF-THAI could differentiate presence of life-threatening SCAR (P < 0.01) but not multiple-drug allergy. CONCLUSIONS: Thai DrHy-Q was reliable and valid in evaluating quality of life among patients with drug hypersensitivity. Thai DrHy-Q was able to discriminate serious drug allergy phenotypes from non-serious manifestations in clinical practice and capture more specific drug-hypersensitivity aspects than WHOQOL-BREF-THAI.

Slow desensitization of imatinib-induced nonimmediate reactions and dynamic changes of drug-specific CD4+CD25+CD134+ lymphocytes.

BACKGROUND: Imatinib is a tyrosine kinase inhibitor indicated for the treatment of gastrointestinal stromal tumors (GISTs) and certain neoplastic diseases; however, nonimmediate adverse reactions are common. OBJECTIVE: To describe the process of imatinib slow desensitization in patients who experienced nonimmediate reactions to imatinib and the dynamic change in drug-specific CD4+CD25+CD134+ T-lymphocyte percentages. METHODS: Five patients diagnosed as having GISTs and with a recent history of imatinib-induced nonimmediate reactions (maculopapular exanthema with eosinophilia, exfoliative dermatitis, palmar-plantar erythrodysesthesia, and drug rash with eosinophilia and systemic symptoms) were desensitized using a slow desensitization protocol. The reintroduced imatinib dosage was stepped up every week starting from 10 mg/d and increasing to 25, 50, 75, 100, 150, 200, and 300 mg/d until the target dose of 400 mg/d was achieved. Prednisolone of up to 30 mg/d was allowed if allergic reactions recurred. The percentages of CD4+CD25+CD134+ T cells present after incubating peripheral blood mononuclear cells with imatinib, at baseline and after successful desensitization, were analyzed using flow cytometric analysis. RESULTS: By using a slow desensitization technique, all patients were able to receive 400 mg/d of imatinib, and prednisolone was gradually tapered off. The percentages of imatinib-induced CD4+CD25+CD134+ T cells decreased from a mean (SD) of 11.3% (6.5%) and 13.4% (7.3%) at baseline to 3.2% (0.7%) and 3.0% (1.1%) after successful desensitization, when stimulating peripheral blood mononuclear cells with 1 and 2 µM of imatinib, respectively. CONCLUSION: Slow desensitization is a helpful procedure in treating patients with imatinib-induced nonimmediate reactions other than simple maculopapular exanthema. The reduced percentages of imatinib-induced CD4+CD25+CD134+ T cells in these patients may be associated with immune tolerance.

Extranasal symptoms of allergic rhinitis are difficult to treat and affect quality of life.

BACKGROUND: Total nasal symptom score is widely used to evaluate the severity of allergic rhinitis, but the clinical significance of extranasal symptoms is largely unknown. We wished to analyze the presenting symptoms in allergic rhinitis, as well as their relationship with quality of life (QoL) and therapeutic outcomes. METHODS: Presenting symptoms and QoL were assessed using the Optum™ SF-12v2(®) Health Survey in 260 patients with allergic rhinitis. Clinical response at 3 months after treatment was measured. RESULTS: Ten most common symptoms presenting with at least moderate severity in allergic rhinitis were: blocked nose (82.7%), rhinorrhea (75.0%), sneeze (70.9%), itchy nose (68.5%), fatigue (63.6%), mouth breathing (61.1%), daytime somnolence (52.7%), postnasal drip (49.1%), itchy eyes (47.3%), and dry mouth (46.3%). Severity of sneeze was correlated with physical component summary (PCS) whereas postnasal drip and daytime somnolence were correlated with mental component summary (MCS). Severity of dry mouth was correlated with PCS and MCS. The symptoms with the highest severity scores after treatment were blocked nose, postnasal drip, fatigue, and dry mouth, respectively. CONCLUSIONS: Extranasal symptom scores correlated well with physical health and mental health in allergic rhinitis patients. Assessment of extranasal symptoms should be included to evaluate disease severity and assess therapeutic outcomes. Clinical trial NCT02000648, http://www.clinicaltrials.gov.