You're using an outdated browser. Please upgrade to a modern browser for the best experience.
Please note this is a comparison between Version 1 by Shuen-Iu Hung and Version 2 by Dean Liu.
The clinical presentations of delayed drug hypersensitivity are various. They range from mild reactions (e.g., maculopapular exanthema (MPE) and fixed drug eruption (FDE)) to drug-induced liver injury (DILI) and severe cutaneous adverse reactions (SCARs) (e.g., Stevens–Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug reaction with eosinophilia and systemic symptoms (DRESS)). The common culprits of delayed drug hypersensitivity include anti-epileptics, antibiotics, anti-gout agents, anti-viral drugs, etc. Delayed drug hypersensitivity is proposed to be initiated by different models of molecular recognition, composed of drug/metabolite antigen and endogenous peptide, HLA presentation, and T cell receptor (TCR) interaction. Increasing genetic variants of HLA loci and drug metabolic enzymes have been identified as responsible for delayed drug hypersensitivity. The genetic susceptibility of delayed drug hypersensitivity showed drug-specific, phenotype-specific, and ethnic variation.
- Genetics
- drug hypersensitivity
- T lymphocytes
- HLA
- Genomics
- Pharmacogenomics
- Allergy
1. Genetic Susceptibility of Antiepileptics-Induced Hypersensitivity Reactions
Chung W. H. et al. first reported that HLA-B*15:02 is a genetic marker for carbamazepine (CBZ)-induced SJS/TEN in Han Chinese patient populations in Taiwan in 2004 [1][36] (Table 1). This association has been further validated in other Asian countries, including Hong Kong, Singapore, Vietnam, Thailand, Malaysia, and India [2][3][4][5][6][7][8][37,38,39,40,41,42,43] (Table 1). A prospective study showed that the genetic screening of HLA-B*15:02 before CBZ administration prevented the occurrence of CBZ-induced SJS/TEN [9][44]. None of the 4877 recruited patients who received CBZ treatment with preemptive pharmacogenomic testing developed SJS/TEN [9][44]. Additionally, HLA-B*57:01 was associated with CBZ-induced SJS/TEN in Europeans [10][45] (Table 1). By comparison, HLA-A*31:01 is associated with CBZ-induced MPE and DRESS, which was first reported in Han Chinese in Taiwan in 2006 [11][46]. This genetic susceptibility was then validated in different populations, including Europeans and Japanese [10][11][12][13][14][15][16][45,46,47,48,49,50,51] (Table 1).
Aside from being a risk allele for CBZ-induced SJS/TEN, HLA-B*15:02 has also been associated with SCARs induced by other antiepileptics that have a similar aromatic structure to CBZ, such as oxcarbazepine [17][18][52,53], phenytoin [5][18][19][20][40,53,54,55], and lamotrigine [21][56] (Table 1). In addition, HLA-A*32:01 was reported to be associated with oxcarbazepine-induced MPE in the Eastern Han Chinese population [22][57]. HLA-B*13:01 and B*51:01 are suggested to be related to phenytoin-induced SCARs in different studies in Asians, including Han Chinese, Japanese, and Malaysian patient populations [20][23][55,58] (Table 1). In addition to HLA alleles, the loss of function in the allele of cytochrome P450 2C9 (CYP2C9), CYP2C9*3, affecting drug metabolism, was responsible for phenytoin-induced SCARs in Taiwan [23][58]. The genetic association was validated in the patient populations from Thailand and Japan [23][24][25][58,59,60] (Table 1). For lamotrigine-induced SCARs, HLA-A*31:01 and HLA-B*38:01 were reported to be risk alleles in patients of Asian or European descent [14][26][49,61] (Table 1).
[49][84], and HLA-B*13:01, associated with dapsone-induced DRESS in north-eastern and south-eastern Asians [50][51][52][53][54][55][85,86,87,88,89,90] (Table 3).
Table 1. Genetic variants associated with antiepileptics-induced hypersensitivity reactions.
Table 3. Genetic variants associated with antibiotics-induced hypersensitivity reactions.
| Causative Drugs | Reactions | Genetic Factors | Ethnicity | OR (95% CI) |
p-Value | Reference | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Amoxicillin | MPE | HLA-DRB3*02:02 | European | 8.88 (3.37–23.32) 1 | <0.0001 | [37][72] | |||||||||
| [ | 28 | ] | [63] | ||||||||||||
| Co-amoxiclav | DILI | HLA-DRB1*15:01 | European | 2.59 (1.44–4.68) | 0.002 | [38 | 7.27 (2.04–25.97) | 4.46 × 10−13 | [8][43] | ||||||
| ] | [ | 73 | ] | Japanese | 62.8 (21.2–185.8) | 5.388 × 10−12 | [32][67] | ||||||||
| Flucloxacillin | DILI | HLA-B*57:01 | European | 80.6 (22.8–284.9) | 8.97 × 10−19 | [39][74] | Indian | 71.40 (3.0–1698) | 0.0014 | ||||||
| Sulfamethoxazole (SMX) | SJS/TEN | HLA-B*38 | [7][42] | ||||||||||||
| Malaysian | |||||||||||||||
| Thai | 348.3 (19.2–6336.9) | 1.6 × 10−13 | [30][65] | European | 8.6 (3.5–21) | <0.003 | [28][63] | 16.15 (4.57–62.4) | 7.87 × 10−6 | [6][41] | |||||
| 579.0 (29.5–11,362.7) | <0.001 | [ | |||||||||||||
| Co-trimoxazole (SMX/TMP) | SJS/TEN | 31][66] | HLA-B*15:02 | Thai | 3.91 (1.42–10.92) | 0.0037 | [40][75] | Vietnamese | 33.78 (7.55–151.03) | <0.0001 | [4][39] | ||||
| DRESS | HLA-B*58:01 | Han Chinese | 47.7 (18.2–125.4) | 1.0 × 10−26 | [36][71] | ||||||||||
| HLA-C*06:02 | 11.84 (1.24–566.04) | 0.0131 | Singaporean | 27.20 (2.67–∞) | 0.004 | [3][38] | |||||||||
| Thai | 430.3 (22.6–8958.9) | <0.001 | [31][ | ||||||||||||
| HLA-C*08:01 | 66 | ] | 3.53 (1.21–10.40) | 0.0108 | HLA-B*57:01 | European | 9.0 (4.2–19.4) | 9.62 × 10−7 | [10] | ||||||
| SCARs | [ | 45 | ] | ||||||||||||
| SCARs | HLA-B*58:01 | Korean | 97.8 (18.3–521.5) | 2.45 × 10−11 | [ | HLA-A*11:01-B*15:02 haplotype,33 | Thai (HIV-infected patients only) | 4.36 (1.43–13.34) | 0.0108][68] | [42][77] | DRESS | HLA-A*31:01 | Han Chinese | 23.0 (4.2–125) | <0.001 |
| HLA-B*58:01 | Han Chinese | 44.0 (21.5–90.3) | 2.6 × 10 | ||||||||||||
| HLA-B*13:01-C*03:04 haplotype | 3.77 (1.27–11.19) | −41 | 0.0251 | [36][ | |||||||||||
| HLA-B*13:01 | Han Chinese, Thai, Malaysia | 11.7 (5.7–24) | 1.3 × 10−13 | [41][76] | |||||||||||
| Vancomycin | DRESS | HLA-A*32:01 | European | NA | 1 × 10−8 | [49][84] | |||||||||
| Dapsone | SCARs | HLA-B*13:01 | Thai | 39.00 (7.67–198.21) | 5.34 × 10−7 | [50][85] | |||||||||
| 54.00 (7.96–366.16) | 0.0001 | [53][88] | |||||||||||||
| DRESS | HLA-B*13:01 | Han Chinese | 20.53 (11.55–36.48) | 6.84 × 10−25 | [51][86] | ||||||||||
| Taiwanese, Malaysian | 49.64 (5.89–418.13) | 2.92 × 10−4 | [52][87] | ||||||||||||
| HLA-B*13:01 | Korean | 73.67 (2.56–2119.93) | 0.012 | [54][89] | |||||||||||
| HLA-B*13:01 | Papua | 233.46 (1.7–67.7) | 7.11 × 10−9 | [55][90] |
Abbreviations: SJS/TEN, Stevens–Johnson syndrome/toxic epidermal necrolysis; DRESS, drug reaction with eosinophilia and systemic symptoms; SCARs, severe cutaneous adverse reactions; NA, not available; OR, odds ratio. 1 The odds ratio represents the increased risk of delayed type compared to immediate type reaction.
4. Genetic Susceptibility of Antiviral Agents-Induced Hypersensiticity
Abacavir is a nucleoside reverse transcriptase inhibitor, usually used in combined therapy for treating patients with HIV infection. Approximately 5–8% of European patients treated with abacavir developed immune-related adverse drug reaction (ADR) during the first six weeks of treatment [56][91]. HLA-B*57:01 was identified as a genetic predisposition for abacavir-related hypersensitivity in Caucasians in 2002 [57][58][92,93] (Table 4). The subsequent randomized clinical trials, recruiting 1956 patients from 19 countries, demonstrated that the carriage of HLA-B*57:01 could be a genetic predictor, in order to prevent abacavir hypersensitivity [59][94].
Nevirapine, a non-nucleoside reverse transcriptase inhibitor to treat HIV infection, has been reported to be associated with a hypersensitivity reaction; it has the clinical presentations of fever, a skin rash, or hepatitis. HLA-B*35:05 and HLA-Cw*04:01 were found to be related to a nevirapine-induced skin rash in Thailand [60][61][95,96] and nevirapine-induced SJS/TEN among Africans, respectively [62][97] (Table 4). Of note, HLA-Cw*04 was also found to be associated with cutaneous adverse reactions in multiple ethnicities [61][96] (Table 4). HLA-DRB1*01:01 was proposed to contribute to nevirapine-induced DRESS in patients in West Australia [63][98] (Table 4). Studies of multiple ethnicities show that the HLA-DRB1*01:01 allele is associated with DILI in white people [61][96] (Table 4). The other studies showed HLA-Cw*08 was associated with Nevirapine-induced hepatitis in Sardinian and Japanese populations [64][65][99,100]. The association between nevirapine-induced hepatotoxicity and HLA-Cw*04 was reported in Han Chinese people; this needs further validation [66][101]. In addition, an SNP (rs3099844) of the HCP gene was proposed to be associated with nevirapine-SJS/TEN in Africans [67][102] (Table 4).
Raltegravir, an HIV integrase inhibitor introduced in 2007, was associated with DRESS in Africans. HLA-B*53:01 was implicated as a risk allele of raltegravir-DRESS in the African population [68][103] (Table 4).
Table 4. Genetic variants associated with antiviral agents-induced hypersensitivity reactions.
| Causative Drugs | Reactions | Genetic Factors | Ethnicity | OR (95% CI) |
p-Value | Reference |
|---|---|---|---|---|---|---|
| Abacavir | DiHS | HLA-B*57:01 | Caucasians | 23.6 (8.0–70.0) | <0.0001 | [57][92] |
| 117 (29–481) | <0.0001 | [58][93] | ||||
| NA | NA | [59][94] | ||||
| Nevirapine | SJS/TEN | rs3099844 (HCP5) | Mozambique | 2.03 (na) | 0.039 | [67][102] |
| HLA-C*04:01 | African | 4.84 (2.71–8.61) | 8.47 × 10−8 | [62][97] | ||
| DiHS | HLA-Cw*04 | Thai | 2.43(1.22–4.84) | 0.17 | [61][96] | |
| NA | 0.0088 | [69][104] | ||||
| Asians, Blacks, Whites | 2.51 (1.73–3.62) | 8.7 × 10−6 | [61][96] | |||
| Han Chinese | 3.611 (1.135–11.489) | 0.030 | [66][101] | 71] | [12][47] | |
| HLA-B*35:05 | Thai | 18.96 (4.87–73.44) | 4.6 × 10−6 | [60][95] | ||
| HLA-B*35 | Asians | 3.47 (1.58–7.61) | 0.053 | [61][96] | ||
| HLA-DRB1*01 | Whites | 3.02 (1.66–5.49) | 0.0074 | [61][96] | ||
| 4.8 (na) | 0.14 | [63][98] | ||||
| HLA-Cw*08 | Sardinian | NA | 0.05 | [64][99] | ||
| Japanese | NA | 0.03 | [65][100] | |||
| Raltegravir | DRESS | HLA-B*53:01 | African | NA | NA | [68][103] |
Abbreviations: SJS/TEN, Stevens–Johnson syndrome/toxic epidermal necrolysis; DiHS, drug-induced hypersensitivity syndrome; DRESS, drug reaction with eosinophilia and systemic symptoms; NA, not available; OR, odds ratio.
5. Genetic Susceptibility of Hypersensitivity Reactions to Anti-Thyroid Drugs and Methazolamide
Anti-thyroid drugs (ATD), including carbimazole and methimazole, have been reported to induce agranulocytosis, and their association with HLA genotypes has been found in different ethnicities. Table 5 lists the genetic variants associated with drug-induced agranulocytosis (DIA). Methimazole-induced agranulocytosis was associated with HLA-DRB1*08:03:02 in Japanese people [70][105] (Table 5). HLA-B*27:05, HLA-B*38:02, and HLA-DRB1*08:03 alleles were found to be related to ATD-induced agranulocytosis in Taiwan [71][106] (Table 5). HLA-B*38:02 and HLA-DRB1*08:03 alleles were reported to be associated with ATD-induced agranulocytosis in Han Chinese people [72][73][107,108] (Table 5). HLA-B*27:05 was reported in a European population [74][109], as well as in Han Chinese people from northern China [75][110] (Table 5).
Methazolamide, an intraocular pressure-lowering drug, may cause SJS/TEN in Asians. HLA-B*59:01 has been proposed to be associated with methazolamide-induced SJS/TEN in Korean, Japanese, and Han Chinese patients [76][77][78][79][111,112,113,114] (Table 5).
Table 5. Genetic variants associated with hypersensitivity reactions to anti-thyroid drugs and methazolamide.
| Causative Drugs | Reactions | Genetic Factors | Ethnicity | OR//break//(95% CI) | p-Value | Reference | |||
|---|---|---|---|---|---|---|---|---|---|
| Methimazole | DIA | HLA-DRB1*08:03:02 | Japanese | 5.42 (na) | 0.002 | [70][105] | |||
| HLA-B*38:02, DRB1*08:03 haplotype | Han Chinese | 48.41 (21.66–108.22) | 3.32 × 10−21 | [72][107] | |||||
| HLA-B*38:02 | 21.48 (11.13–41.48) | 6.75 × 10−32 | |||||||
| HLA-DRB1*08:03 | 6.13 (3.28–11.46) | 1.83 × 10−9 | |||||||
| Carbimazole/Methimazole | DIA | HLA-B*38:02:01 | Han Chinese | 265.5 (27.87–2528.0) | 2.5 × 10−14 | [73][108] | |||
| Carbimazole, Methimazole, Propylthiouracil | DIA | HLA-B*27:05 | Caucasian | 7.30 (3.81–13.96) | 1.91 × 10−9 | [74][109] | |||
| Methimazole, Propylthiouracil | DIA | HLA-B*27:05 | Han Chinese | 60.11 (3.27–1104.4) | 1.1 × 10−4 | [75][110] | |||
| HLA-B*38:02 | 6.55 (2.11–20.36) | 2.41 × 10−4 | |||||||
| HLA-DRB1*08:03 | 3.95 (1.60–9.79) | 6.86 (2.4–19.9) | 2.7 × 10−3 | [15][50 | |||||
| 1.57 × 10 | −3 | European] | |||||||
| 39.11 (4.49–340.51) | 5.9 × 10 | −4 | [29][64] | ||||||
| Methazolamide | SJS/TEN | HLA-B*59:01 | Han Chinese | 305.0 (11.3–8259.9) | 6.3 × 10−7 | [76][111] | European | 49.9 (12.9–193.6) | 4.0 × 10 |
| Korean | −8 | 249.8 (13.4–4813.5) | [10][45] | ||||||
| <0.001 | [ | 77 | ][112] | 13.2 (8.4–20.8) | <0.001 | [12][47] | |||
| Japanese | NA | NA | [79][114] | 12.41 (1.27–121.03) | 3.5 × 10 | ||||
| HLA-B*59:01 | −8 | Han Chinese | [13][48] | ||||||
| 146.00 (16.12–1321.98) | 6.19 × 10 | −10 | [ | 78][113] | 22.00 (1.03–1190.36) | 0.047 | [14][49] | ||
| Japanese | 10.8 (5.9–19.6) | 3.64 × 10−15 | |||||||
| 20.86 (9.03–48.20) | |||||||||
| 1.22 × 10 | |||||||||
| −13 | |||||||||
| DILI | HLA-B*58:01, HLA-B*53:01 cluster | African-American, Caucasian, Hispanic | NA | 0.0007 | [34][69] | ||||
| MPE | HLA-B*58:01 | Thai | 144.0 (13.9–1497.0) | <0.001 | [31][66] | [ | |||
| Han Chinese | 8.5 (4.2–17.5) | 2.3 × 10−9 | [16][51] | ||||||
| HLA-B*55:02 | MPE/DRESS | HLA-A*31:01 | Han Chinese | 17.5 (4.6–66.5) | 0.0022 | [11][46] | |||
| [ | MPE | HLA-B*15:02 | Thai | 7.27 (2.04–25.97) | 0.0022 | [8][43] | |||
| 36 | ] | [ | Oxcarbazepine (OXC) | SJS/TEN | HLA-B*15:02 | Han Chinese | 27.90 (7.84–99.23) | 1.12 × 10−9 | [17][52] |
| 71.00 (7.84–643.10) | 1.43 × 10 | −4 | 80.7 (3.8–1714.4) | 8.4 × 10−4 | [18][53] | ||||
| MPE | HLA-A*32:01 | Han Chinese | 15.877 (1.817–138.720) | 0.004 | [22][57] | ||||
| Phenytoin (PHT) | SCARs | CYP2C9*3 | Taiwanese | 14.00 (6.75–29.02) | 0.00001 | [23][58] | |||
| Japanese | 8.88 (2.20–35.83) | ||||||||
| Malaysian | 5.60 (0.56–56.20) | ||||||||
| Thai | 4.30 (1.41–13.09) | <0.05 | [24][59] | ||||||
| 25 | ] | [ | 60] | ||||||
| 71 | ] | Taiwanese, Japanese, Thai | HLA-B*15:02 | Asian (Han Chinese, Japanese, Malaysian) | 5.0 (2.0–13) | 0.025 | [23][58] | ||
| SJS/TEN | HLA-B*15:02 | Han Chinese | 5.1 (1.8–15.1) | 0.0041 | [18][53] | ||||
| 3.50 (1.10–11.18) | 0.045 | [20][55] | |||||||
| Thai | 18.5 (1.82–188.40) | 0.005 | [5][40] | ||||||
| Malaysian | 5.71 (1.41–23.10) | 0.016 | [19][54] | ||||||
| Lamotrigine (LTG) | SCARs | HLA-A*31:01 | Korean | 11.43 (1.95–59.77) | 0.0037 | [26][61] | |||
| HLA-B*38:01 | European | 147.00 (1.88–483) | 0.001 | [14][49] | |||||
| SJS/TEN | HLA-B*15:02 | Han Chinese | 4.98 (1.43–17.28) | 0.01 | [21][56] | ||||
| AEDs (CBZ, LTG, PHT, etc.) | SCARs | HLA-B*15:02 | Han Chinese | 17.6 (2.9–105.2) | 0.001 | [2][37] |
Abbreviations: SJS/TEN, Stevens–Johnson syndrome/toxic epidermal necrolysis; DRESS, drug reaction with eosinophilia and systemic symptoms; MPE, maculopapular exanthema; OR, odds ratio; SCARs, severe cutaneous adverse reactions.
2. Genetic Susceptibility of Allopurinol Hypersensitivity
Allopurinol, a xanthine oxidase inhibitor, is the first-line drug to treat hyperuricemia and gout. Allopurinol is also one of the common culprit drugs to induce drug hypersensitivity. Hung S.I. et al. first found HLA-B*58:01 to be associated with allopurinol-induced SCARs in Han Chinese people in Taiwan in 2005 [27][62] (Table 2). This association was further replicated and validated among various ethnicities, including European [28][29][63,64], Thai [30][31][65,66], Japanese [32][67], Korean [33][68], and African American [34][69] (Table 2). HLA-B*58:01 genetic screening has been shown to be a promising strategy for preventing allopurinol SCARs [35][70]. The renal dysfunction and increased plasma levels of the metabolite of allopurinol, i.e., oxypurinol, deteriorate the severity of allopurinol hypersensitivity [36][71]. This might explain the higher mortality rate of allopurinol-induced SCARs in patients with chronic kidney disease, because of the delayed clearance of oxypurinol. Furthermore, allopurinol-induced liver injury (DILI) was found to be associated with HLA-A*34:02, HLA-B*53:01, and HLA-B*58:01 [31][34][66,69] (Table 2). Some genomics studies validated the attribution of HLA-B*58:01 and proposed that other genetic variants, out of the HLA region, might also contribute to the development of allopurinol hypersensitivity [32][67] (Table 2).
Table 2. Genetic variants associated with allopurinol-induced hypersensitivity reactions.
| Reactions | Genetic Factors | Ethnicity | OR (95% CI) |
p-Value | Reference |
|---|---|---|---|---|---|
| SJS/TEN | HLA-B*58:01 | Han Chinese | 580.3 (34.4–9780.9) | 4.7 × 10−24 | [27][62 |
Abbreviations: SJS/TEN, Stevens–Johnson syndrome/toxic epidermal necrolysis; DRESS, drug reaction with eosinophilia and systemic symptoms; DILI, drug-induced liver injury; SCARs, severe cutaneous adverse reactions; MPE, maculopapular exanthema; NA, not available; OR, odds ratio.
3. Genetic Susceptibility of Antibiotics-Induced Hypersensitivity Reactions
Antibiotics can cause either immediate-type or delayed-type drug hypersensitivity. Antibiotic hypersensitivity has shown an immune-related genetic predisposition. The HLA-DRB3*02:02 allele, absent in Europeans, accounts for 83% of amoxicillin-induced MPE cases in Italy [37][72] (Table 3). A high level of HLA-DRB1*15:01 was observed in Europeans with amoxicillin-clavulanate-induced liver injury [38][73], and a high level of HLA-B*57:01 was observed in flucloxacillin-induced liver injury [39][74].
Co-trimoxazole, a combination of sulfamethoxazole (SMX) and trimethoprim (TMP), is associated with delayed drug hypersensitivity. HLA-B*38 was reported to be related to sulfamethoxazole-induced SJS/TEN in Europeans [28][63]. Kongpan T. et al. reported that carriers with HLA-B*15:02, HLA-C*06:02, or HLA-C*08:01 had an increased risk of co-trimoxazole-induced SJS/TEN (odds ratio: 11) [40][75]. By whole genome sequencing (WGS), the recent multi-country case-control study showed that HLA-B*13:01 was strongly associated with co-trimoxazole-induced SCARs in patients from Taiwan, Thailand, and Malaysia [41][76]. Notably, HLA-B*13:01 contributed to 85.4% of patients with co-trimoxazole-induced DRESS [41][76]. A multicentric study of the Thai population showed that HLA-B*15:02 and HLA-C*08:01 are associated with cotrimoxazole-SJS/TEN and HLA-B*13:01 in DRESS. Additionally, the haplotypes of HLA-A*11:01-B*15:02 and HLA-B*13:01-C*03:04 are associated with co-trimoxazole-induced SJS/TEN and DRESS, respectively [42][77]. Earlier studies indicate that gene variants involved in drug metabolisms, such as NAT2 [43][44][45][46][47][78,79,80,81,82] and GSTM1 [48][83] null genotypes, were associated with sulfonamide-induced hypersensitivity reactions. However, the associations were weak and lacked validation.
Other significant discoveries, regarding pharmacogenomic associations with antibiotics-induced SCARs, include HLA-A*32:01, is associated with vancomycin-induced DRESS in Caucasians
Abbreviations: DIA, drug-induced agranulocytosis; SJS/TEN, Stevens–Johnson syndrome/toxic epidermal necrolysis; NA, not available; OR, odds ratio.