Discussion
Our study showed for the first time the simulated benefits of utilizing pre-stored HLA information to prevent type B ADRs or SCARs without additional testing in a real-life setting. Consistent with the findings of previous studies, the risk of developing SCARs due to allopurinol and carbamazepine use was significantly higher in patients with the risk alleles. In addition, type B ADRs due to allopurinol, carbamazepine, and oxcarbazepine use were more common in patients with the risk alleles. Conversely, if the data on the HLA PGx alleles were available before prescribing the risk drugs to the patients, a significant number of SCARs and type B ADRs could have been prevented, resulting in improved patient safety and cost effectiveness.
Our findings support the claim that PGx information should be integrated into EMRs using a clinical decision support system.5,15 In case of organ transplantation patients who have already undergone tests for HLA genotyping, physicians could use this HLA PGx information to reduce ADR risks. Although our study did not find statistically significant differences in the rate of ADRs for some drugs, possibly owing to the small number of the study population who took them, the available HLA data could be useful in preventing ADRs.
In addition to the data on the HLA PGx alleles, other readily available HLA allele data can be used to diagnose and identify individuals at a risk of various autoimmune diseases.16,17 Associations between certain HLA alleles and autoimmune diseases, such as HLA-DRB1 for rheumatoid arthritis, HLA-B51 for Behcet’s disease, HLA-B27 for ankylosing spondylitis, HLA-DQ2/DQ8 for celiac disease, and HLA-DQB1*06:02 for narcolepsy, are widely known. A broadened screening of pre-stored HLA information to detect disease-associated variants of HLA alleles may enable early identification of susceptible patients, which may facilitate early diagnosis of diseases.
The exact incidence of SCARs is unknown; however, the incidence of SJS/TEN is estimated at 1–2 cases per 1,000,000 people per year.18 Although the incidence of SCARs is very low, it carries significant morbidity, and the mortality rates are 10% for SJS, 30% for SJS/TEN overlap, 50% for TEN, and 5% for DRESS.18 Furthermore, SCARs may severely damage the affected mucosa or skin and leave permanent sequelae. Therefore, although the absolute risk reduction is small owing to its rare occurrence, the potential benefits of preventing SCARs are substantial. Considering patient data for the HLA PGx alleles is already available, it is reasonable to integrate this data and ensure its availability for clinicians at the point-of-care to help prevent SCARs. To achieve this, a clinical decision support system that instantly informs the personalized estimated risk to physicians by automatically linking the existing genetic information with the prescription should be implemented.
Pharmacogenomic studies showed that certain HLA genotypes induce T cell activation to a specific drug, resulting in the development of a SCAR. In 2005, the HLA-B*58:01 allele was first reported to be strongly associated with allopurinol-induced SCAR in a case–control study of the Han Chinese population in Taiwan (OR = 580.3).19 In Korea, the HLA-B*58:01 allele was also strongly associated with allopurinol-induced SCAR (OR = 97.8).20 This study is consistent with these findings, confirming the usefulness of detecting the HLA-B*58:01 allele in actual clinical practice.
However, the association between a specific allele and a particular drug-induced hypersensitivity varies between ethnicities. For example, 100% of carbamazepine-induced SJS patients were positive for the HLA-B*15:02 allele, whereas only 3% of the tolerant patients were positive (OR = 2,504) in the Han Chinese population, in which the HLA-B*15:02 allele frequency is relatively high.21This strong association between the occurrence of carbamazepine-induced SJS/TEN and HLA-B*15:02 was not replicated in Koreans, in whom the HLA-B*15:02 allele frequency is low.22 Instead, HLA-B*15:11 has been proposed as an additional allele type associated with carbamazepine-induced SJS in Koreans.23 In patients with carbamazepine-induced DRESS, the HLA-A*31:01 allele was reported as a risk marker in Europeans, Japanese, and Koreans.23 Oxcarbazepine, a 10-keto analog of carbamazepine, has also been associated with the HLA-B*15:02 and HLA-A*31:01 alleles in the development of SJS/TEN and maculopapular rash, respectively.24
Methazolamide, a carbonic anhydrase inhibitor used as an intraocular pressure-lowering drug to treat glaucoma, was found to be associated with SJS/TEN in individuals with the HLA-B*59:01 allele in Northeast Asia, including Korea, Japan, and China.25 However, we did not find significant results regarding methazolamide because its use was minimal in our study. Dapsone, an antimicrobial agent used in the treatment of leprosy or Pneumocystis jirovecii pneumonia, can induce a hypersensitivity syndrome similar to DRESS syndrome,26 and it was significantly associated with the HLA-B*13:01 allele in Chinese, Thai, and Koreans.27 Recently, the HLA-A*32:01 allele was reported to be strongly associated with vancomycin-induced DRESS. Although vancomycin is one of the main culprit drugs of SCARs in Korea, the HLA-A*32:01 allele is unlikely to be used as a screening test considering its rare genotype frequency in Koreans, which is at 0.6%.27 Further studies to investigate the risk alleles of vancomycin-induced DRESS in the Korean population are warranted. Nonetheless, if pre-stored data on HLA-A*32:01 were readily available, it could still be used to minimize vancomycin prescription in susceptible patients with the HLA-A*32:01 allele.
Abacavir, a nucleoside analog used to treat HIV infections, can cause severe delayed systemic hypersensitivity reactions in association with the HLA-B*57:01 allele. However, the use of the HLA-B*57:01 allele as a screening test has no clinical relevance owing to its very low allelic frequency (0.2%) in Koreans.28 Our results are consistent with this observation.
Of the 4,092 individuals with 8 HLA PGx alleles identified in our study, 1,597 (39%), or 13% of the total number of study subjects, were prescribed one or more of the seven drugs associated with these alleles. If all these patients had their HLA test results evaluated at the time of drug prescription, doctors could have made better therapeutic decisions, including whether to prescribe alternative drugs, change the dose, or carefully monitor the patients according to their individual PGx profile. In addition, if a prescription was changed by a physician due to an automatic PGx warning at the time of prescription, not only SCARs but also a considerable number of type B ADRs, specifically 15.3% (53/347) for allopurinol, 33.3% (3/9) for carbamazepine, 23.5% (4/17) for oxcarbazepine, 66.7% (2/3) for dapsone, and 2.6% (9/341) for vancomycin, could have been prevented.
Our study is a retrospective study and the preventative effects proposed need to be assessed further in prospective studies and real clinical practice. Nonetheless, the proposed approach has clear benefits because it utilizes pre-tested HLA data to prevent ADR occurrences without additional costs or patient discomfort. Given the rapid increase in the genomic data collection for research and clinical purposes, timely use of these valuable genomic data in clinical practice should be prioritized.29,30 For the secondary use of the point-of-care of genomic tests, it is necessary to evaluate and confirm the clinical validity of the tests for the specific uses other than the primary purpose. Additionally, it is necessary to devise an effective and safe approach to deliver genomic information at the right time and in the right manner. The most appropriate method so far is to design and use a clinical decision support system that integrates the personal genomic information of the patient at the time of drug prescription and provides an appropriate PGx warning. Considering the diversity of genetic tests, various drugs, and individual genetic variations, it is necessary to design an optimal system to ensure efficiency and safety in clinical practice.
Although this study provided important results, there are several limitations. First, the HLA genotypes selected in this study were obtained from results accumulated over a long period using various HLA testing methods. Therefore, the results of the lower resolution HLA tests were converted to the matched 4-digit results based on the previously reported population-wise allele frequencies. This means that some results may not be accurate in some patients. However, this is unlikely to result in significant error based on the allele frequency data.28 For example, the HLA-B58 status was also shown to be strongly associated with allopurinol hypersensitivity and HLA-B*58:01 is the only HLA-B58 allele in the Korean population.31
Second, the causal relationships in this study were not assessed by confirmative tests, and therefore, misdiagnosis and overestimation are possible. In our analysis, we assumed that the reported SCARs or type B ADRs were related to the prescribed drug. Therefore, SCARs reported in an individual could have been caused by another drug other than the risk drug associated with the HLA PGx alleles. The study relied on ADR reporting, and thus, under-reporting could have been a problem. Nonetheless, our study could replicate the significant associations between the HLA-B*58:01 allele and allopurinol and the HLA PGx alleles for carbamazepine, suggesting that our overall findings are valid.
In summary, the results of this study highlight the importance of utilizing pre-stored HLA data to predict potential ADRs in a clinical setting. Because HLA PGx alleles are already available, no additional cost is required; moreover, if these HLA PGx alleles were readily available at the point-of-care, some SCARs and type B ADRs could have been prevented. Therefore, further studies are needed to integrate HLA data and existing EMRs to achieve a personalized medicine approach.