Summary of study identification, inclusion, and exclusion. HuGENet indicates Human Genome Epidemiology Network.
Forest plot shows associations between HLA-B*1301 and dapsone-induced cADRs using dapsone-tolerant controls. Each data marker and its size indicate a mean value and a magnitude of the data point compared with the total. Error bars indicate 95% CIs and vertical dotted line, overall estimated odds ratio (OR). Data marker size reflects the weight of the study using random-effects meta-analysis. Data were evaluated using R, version 3.4.3 (The R Foundation).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Tangamornsuksan W, Lohitnavy M. Association Between HLA-B*1301 and Dapsone-Induced Cutaneous Adverse Drug Reactions: A Systematic Review and Meta-analysis. JAMA Dermatol. 2018;154(4):441–446. doi:10.1001/jamadermatol.2017.6484
Is HLA-B*1301 genotype associated with dapsone-induced cutaneous adverse reactions?
In this systematic review and meta-analysis of 3 studies that included 111 patients with dapsone-induced cutaneous adverse drug reactions, 1165 dapsone-tolerant patients, and 3026 healthy controls, a statistically significant association between HLA-B*1301 and dapsone-induced cutaneous adverse reactions was found.
Because HLA-B*1301 is associated with dapsone-induced cutaneous adverse reactions, genetic screening for HLA-B*1301 before initiating dapsone therapy is warranted.
Dapsone-induced hypersensitivity syndrome (DHS) is a life-threatening adverse drug reaction. Based on available epidemiologic studies, HLA genotypes may play an important role in DHS.
To assess the association between HLA-B*1301 and dapsone-induced cutaneous adverse drug reactions (cADRs).
Human studies investigating associations between HLA-B*1301 and dapsone-induced cADRs were systematically searched without language restriction from the inception of each database until September 12, 2017, in PubMed, the Human Genome Epidemiology Network), and the Cochrane Library. Combinations of HLA genotypes, dapsone, and synonymous terms were used; reference lists were searched in selected articles.
Two reviewers identified studies investigating the associations between HLA-B*1301 and dapsone-induced cADRs that reported sufficient data for calculating the frequency of HLA-B*1301 carriers among case and control patients, in which all patients received dapsone before HLA-B*1301 screening. An initial search of the databases identified 391 articles, of which 3 studies (2 in Chinese populations and 1 in a Thai population) met the inclusion criteria.
Data Extraction and Synthesis
Overall odds ratios (ORs) with 95% CIs were calculated using a random-effects model to determine the association between HLA-B*1301 and dapsone-induced cADRs. Subgroup analyses by type of cADR were also performed. PRISMA guidelines were used to abstract and assess data.
Main Outcomes and Measures
Primary outcomes were associations between HLA-B*1301 and dapsone-induced cADRs in dapsone-tolerant controls. The outcomes are reported as overall OR. Statistical heterogeneity was assessed using the Q statistic and I2 tests.
From the 3 included studies, there were 111 unique patients with dapsone-induced cADRs (subsequently used in the meta-analysis), 1165 dapsone-tolerant patients, and 3026 healthy controls. The cases included 64 men and 49 women (2 patients were missing from the meta-analysis; 1 each from 2 of the 3 studies); mean age was 39.7 years. An association between HLA-B*1301 and dapsone-induced cADRs was identified (summary OR, 43.0; 95% CI, 24.0-77.2). Subgroup analyses among types of cADRs produced similar findings in DHS (OR, 51.7; 95% CI, 16.9-158.5), dapsone-induced severe cADRs (Stevens-Johnson syndrome and toxic epidermal necrolysis [SJS/TEN] plus drug rash [adverse skin reaction to a drug] along with eosinophilia and systemic symptoms [DRESS]) (OR, 54.0; 95% Cl, 8.0-366.2), dapsone-induced SJS/TEN (OR, 40.5; 95% CI, 2.8-591.0), and dapsone-induced DRESS (OR, 60.8; 95% CI, 7.4-496.2). There was no heterogeneity (I2 = 0%, P = .38).
Conclusions and Relevance
Associations between HLA-B*1301 and dapsone-induced cADRs were found in dapsone-tolerant and healthy control groups. For patient safety, genetic screening for HLA-B*1301 in Asian populations before dapsone therapy is warranted.
Dapsone is an antimicrobial anti-inflammatory drug widely used in the treatment of some diseases.1 As an antimicrobial agent, dapsone has been used to treat leprosy and actinomycetoma and for prophylaxis and treatment of Pneumocystis carinii pneumonia and malaria. It has also been used as an anti-inflammatory agent to treat dermatologic conditions, including dermatitis herpetiformis, linear IgA dermatosis, and pustular psoriasis.2,3 Dapsone has pharmacologic (eg, hemolytic anemia, methemoglobinemia, and agranulocytosis) and idiosyncratic (dapsone-induced hypersensitivity syndrome [DHS]) adverse effects.1Quiz Ref ID Idiosyncratic adverse effects of dapsone range from mild cutaneous reactions to severe life-threatening reactions, such as exfoliated dermatitis, liver failure, agranulocytosis, Stevens-Johnson syndrome and toxic epidermal necrolysis (SJS/TEN), nephritis, and renal failure.4,5 Dapsone-induced hypersensitivity syndrome is a life-threatening drug reaction with a mortality rate of 9.9% to 12.5%.6-8 It is characterized by fever, adverse skin reaction to a drug, and systematic involvement and usually manifests in the first 6 weeks after dapsone treatment initiation. However, Quiz Ref IDHLA genotypes are a family of genes involved in immune reactions. HLA genotypes are the most polymorphic gene cluster in the human genome.9,10 Specific polymorphisms in HLA genotypes are associated with various adverse drug reactions, for example, HLA-B*5801 and allopurinol-induced SJS/TEN,11HLA-B*5701 and abacavir-induced hypersensitivity,12HLA-B*1502 and carbamazepine-induced SJS/TEN,13 and HLA-B*1502 and oxcarbazepine-induced SJS/TEN.14
Associations between HLA genotypes and dapsone-induced cutaneous adverse drug reactions (cADRs) have been investigated in epidemiologic studies.15-17 We aimed to summarize and elucidate the associations between HLA-B*1301 and dapsone-induced cADRs using systematic review and meta-analysis techniques.
A systematic literature search in PubMed, the Human Genome Epidemiology Network (HuGENet), and the Cochrane Library was performed from database inception until September 2017. Our search terms were combinations of keywords and synonyms for HLA genotypes and dapsone without language or study-design restrictions. Only human studies were included. Additional studies were retrieved from references listed in the included articles. We independently screened titles, abstracts, or both for relevance; studies deemed relevant underwent full-text article assessments for inclusion. Studies were included for case and control groups if (1) HLA-B*1301 and dapsone-induced cADR associations were investigated; (2) all patients received dapsone before HLA-B*1301 screening; and (3) sufficient data to calculate frequency of HLA-B*1301 carriers were reported. If 2 or more studies shared the same population, the study with more complete data or larger sample size was included. For studies that met inclusion criteria but did not provide sufficient data for meta-analysis, we contacted corresponding authors for additional information. This additional information was not included in the present analysis.
The PRISMA guidelines were used to abstract and assess data. We extracted data by study design; eligibility criteria, definition, and diagnostic criteria for cases and controls; patient demographics; dose and duration of dapsone exposure; HLA genotyping technique; and Hardy-Weinberg equilibrium information. The frequency of genotypes was examined by the Hardy-Weinberg equilibrium to determine whether patients from the selected studies were representative of the population.18,19 The Newcastle-Ottawa scale was used to determine the quality of the selected studies.20 Throughout the study, disagreements between reviewers were resolved through discussion until consensus was reached.
The included studies demonstrating associations between HLA-B*1301 and dapsone-induced cADRs were characterized and summarized based on the most recent data. Overall odds ratios (ORs) with 95% CIs were calculated to determine associations between HLA-B*1301 and dapsone-induced cADRs. All analyses were performed using the DerSimonian and Laird method under a random-effects model.21 Analyses were also performed separately on studies using different types of control patients (eg, dapsone-tolerant or healthy controls) and different types of cADRs. Statistical heterogeneity was assessed using the Q statistic and I2 tests.22 Two-sided P ≤ .10 indicated a statistically significant heterogeneity between studies. Heterogeneity I2 values of 25%, 50%, and 75% denote low, moderate, and high heterogeneity across studies, respectively.23 All statistical analyses were performed using R, version 3.4.0 (R Foundation for Statistical Computing).
Our literature search strategy and study selection process are summarized in Figure 1. The initial search of the databases identified 391 articles. After duplicate records were removed, 365 articles were screened on the basis of title, abstract, or both to determine eligibility. Three articles15-17 met inclusion criteria. No additional articles were identified in the bibliographies of the included studies.
Characteristics of the included studies are summarized in Table 1. All of the included studies15-17 were case-control studies. One hundred eleven unique case patients with dapsone-induced cADRs,15-17 1165 dapsone-tolerant controls,15-17 and 3026 healthy controls15-17 were included in our systematic review and meta-analysis. Mean age of included patients was 39.7 years in case patients,15-17 32.2 years in dapsone-tolerant control patients,15,16 and 39.5 years in healthy controls.15 Because some of the selected studies in this meta-analysis did not provide individual data points, SDs could not be calculated from the available information. Male patients made up 64 of 113 cases (56.6%)15-17 (2 patients were missing from the meta-analysis; 1 each from the study by Wang et al15 and Zhang et al16), 817 of 1144 dapsone-tolerant controls (71.4%),15,16 and 53 of 100 healthy controls (53.0%).15 The mean dose of dapsone was 94.5 mg/d (range, 2-100 mg/d).17 Mean duration of dapsone-induced cADRs was 30.4 days (range, 1-53 days).15,16 All included studies15-17 were conducted in Asian populations (2 in Chinese populations15,16 and 1 in a Thai population17). The studies by Wang et al15 and Zhang et al16 investigated the association between HLA-B*1301 and DHS, whereas Tempark et al17 investigated the association between HLA-B*1301 and dapsone-induced severe cADRs. Diagnostic criteria for DHS, SJS/TEN, and DRESS for each study are summarized in Table 1. The included studies identified HLA genotypes by using polymerase chain reaction sequence-specific primers,15 polymerase chain reaction next-generation sequencing,16 and polymerase chain reaction sequence-specific oligonucleotide primers.17 In the study by Zhang et al,16 reported controls deviated significantly from the Hardy-Weinberg equilibrium (P < 10−8). The mean score of quality assessment using the Newcastle-Ottawa scale for case-control studies was 6.3 (range, 5-7) (Table 1).
The associations between HLA-B*1301 and dapsone-induced cADRs are summarized in Table 2. Quiz Ref IDOne hundred eleven patients had dapsone-induced cADRs and 1165 were dapsone tolerant. In these studies, 95 cases and 157 controls carried at least 1 HLA-B*1301 allele. We found an association between HLA-B*1301 and dapsone-induced cADRs (summary OR, 43.0; 95% CI, 24.0-77.2). There was no heterogeneity (I2 = 0%, P = .38) (Table 2 and Figure 2).
In a subgroup analysis by type of cADR, ORs were 51.7 (95% CI, 16.9-158.5; I2 = 46%; P = .17) in the DHS group; 54.0 (95% CI, 8.0-366.2) in the dapsone-induced severe cADRs (SJS/TEN plus DRESS) group; 40.5 (95% CI, 2.8-591.0) in the dapsone-induced SJS/TEN group; and 60.8 (95% CI, 7.4-496.0) in the dapsone-induced DRESS group (Table 2).
There were 100 patients with dapsone-induced cADRs and 3026 healthy controls. In these studies, 95 cases and 39 controls carried at least 1 HLA-B*1301 allele. Statistically significant associations between HLA-B*1301 and dapsone-induced cADRs were identified (Table 2).
To our knowledge, this is the first systematic review and meta-analysis to identify the associations between HLA-B*1301 and dapsone-induced cADRs. An association was found between the HLA-B*1301 genotype and dapsone-induced cADRs (summary OR, 43.0; 95% CI, 24.0-77.2) (Figure 2). This statistically significant association was similar to the results of the subgroup analyses in the DHS, dapsone-induced severe cADRs (SJS/TEN plus DRESS), dapsone-induced SJS/TEN, and dapsone-induced DRESS groups (Table 2). In addition, the association between HLA-B*1301 and dapsone-induced cADRs was found in healthy controls. In the healthy control group, there were duplicate cases in Zhang et al.16 Therefore, summary ORs could not be calculated and this study was not included in the meta-analysis. However, the observed ORs as shown in Table 2 are still relatively high.
Quiz Ref IDAllele frequency of HLA-B*1301 (ƒ) was 0 to 0.019 in European patients, 0.021 in Korean patients, and 0 to 0.015 in Japanese patients. This allele is more prevalent among Chinese (ƒ = 0.009-0.211) and Southeast Asian (ƒ = 0.091-0.021) populations.26,27 Prevalence of HLA-B*1301 is higher in Papua New Guinean (ƒ = 0.283-0.132) and Australian Aboriginal (ƒ = 0.132-0.270) populations.26,27 Nonetheless, outcomes of the dapsone-induced cADRs among persons of other races/ethnicities carrying HLA-B*1301 might be different from those in Chinese and Southeast Asian populations.
Our study demonstrated associations between HLA-B*1301 and dapsone-induced DHS in 2 Chinese populations15,16 and between HLA-B*1301 and dapsone-induced severe cADRs (SJS/TEN plus DRESS) in 1 Thai population.17 To identify any association between HLA-B*1301 and dapsone-induced cADRs in other populations from other geographic areas, future studies are needed.
Quiz Ref IDRecently, a mechanism by which HLA-B*1301 causes DHS was proposed by Watanabe et al.28 Dapsone specifically binds to HLA-B*1301 at the residue Ile95, a proposed binding site called “pocket F.” This specific binding between dapsone and HLA-B*1301 alters the structure of the antigen-recognition site of HLA-B*1301. As a result, the structurally altered protein can recognize its ligands and this leads to the dapsone-induced cADRs.
In this report, there were 3 studies in Asian populations (2 studies in Chinese populations and 1 study in a Thai population). Thus, more studies in more diverse populations are needed.
Associations between HLA-B*1301 and dapsone-induced cADRs were identified in both dapsone-tolerant and healthy control groups. Screening for HLA-B*1301 before initiation of dapsone therapy is warranted in Asian populations.
Accepted for Publication: December 24, 2017.
Corresponding Author: Manupat Lohitnavy, PhD, Center of Excellence for Environmental Health and Toxicology, Faculty of Pharmaceutical Sciences, Naresuan University, Phitsanulok 65000, Thailand (firstname.lastname@example.org).
Published Online: March 14, 2018. doi:10.1001/jamadermatol.2017.6484
Author Contributions: Both authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Both authors.
Acquisition, analysis, or interpretation of data: Both authors.
Drafting of the manuscript: Both authors.
Critical revision of the manuscript for important intellectual content: Both authors.
Statistical analysis: Both authors.
Administrative, technical, or material support: Lohitnavy.
Study supervision: Lohitnavy.
Conflict of Interest Disclosures: None reported.