Knowledge and opinions among Canadian academic physicians regarding genetic screening to prevent severe cutaneous adverse drug reactions

To the Editor: Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are rare drug reactions with high mortality and long-term physical and psychological complications.1Olteanu C. Shear N.H. Chew H.F. et al.Severe physical complications among survivors of Stevens–Johnson syndrome and toxic epidermal necrolysis.Drug Saf. 2018; 41: 277-284Google Scholar, 2Dodiuk-Gad R.P. Olteanu C. Feinstein A. et al.Major psychological complications and decreased health-related quality of life among survivors of Stevens-Johnson syndrome and toxic epidermal necrolysis.Br J Dermatol. 2016; 175: 422-424Crossref PubMed Scopus (51) Google Scholar Carbamazepine and allopurinol are frequent inducers of SJS and TEN. Certain populations are genetically predisposed to SJS and TEN caused by carbamazepine and allopurinol: human leukocyte antigen B (HLA-B)*15:02-positive Asians taking carbamazepine and HLA-B*58:01-positive Asians and Europeans taking allopurinol. HLA-B*15:02 is not associated with SJS and TEN in the Japanese, Koreans, or Europeans, and the association between HLA-B*58:01 and allopurinol-induced SJS and TEN has not been studied in populations other than Asians and Europeans.3Pan R.Y. Dao R.L. Hung S.I. Chung W.H. Pharmacogenomic advances in the prediction and prevention of cutaneous idiosyncratic drug reactions.Clin Pharmacol Ther. 2017; 102: 86-97Crossref PubMed Scopus (26) Google Scholar In a study conducted in Taiwan, screening for HLA-B*15:02 and avoidance of carbamazepine produced no cases of SJS and TEN compared with a historical incidence of 0.23%.4Chen P. Lin J.J. Lu C.S. et al.Carbamazepine-induced toxic effects and HLA-B*1502 screening in Taiwan.N Engl J Med. 2011; 364: 1126-1133Crossref PubMed Scopus (543) Google Scholar Similar findings were demonstrated for allopurinol-induced SJS and TEN.5Ko T.M. Tsai C.Y. Chen S.Y. et al.Use of HLA-B*58:01 genotyping to prevent allopurinol induced severe cutaneous adverse reactions in Taiwan: national prospective cohort study.BMJ. 2015; 351: h4848Crossref PubMed Scopus (155) Google Scholar International regulatory agencies and medical associations have recommended routine genetic screening before using these medications in at-risk populations.3Pan R.Y. Dao R.L. Hung S.I. Chung W.H. Pharmacogenomic advances in the prediction and prevention of cutaneous idiosyncratic drug reactions.Clin Pharmacol Ther. 2017; 102: 86-97Crossref PubMed Scopus (26) Google Scholar We sought to assess the knowledge and clinical practices on this topic among Canadian academic physicians. An anonymous online SurveyMonkey-based questionnaire was distributed to physicians affiliated with the Departments of Medicine, Psychiatry, and Family Medicine at the University of Toronto. Questions were on knowledge and medical practice regarding carbamazepine- and allopurinol-induced SJS and TEN (Tables I and II). Responses were collected during April and May 2018. The study was approved by the Sunnybrook Health Sciences Centre Research Ethics Board. Chi-squared and Fisher's exact tests were used to compare the responses across 5 groups of physician specialties (R version 3.5.1).Table IParticipant responses to knowledge-related questions on genetic screening to prevent carbamazepine- and allopurinol-induced SJS and TENDrug, question [correct response]No. participants answering correctly/total (%)∗Percentages might not sum to 100% due to uncategorized responses. Percentages calculated relative to number of respondents for each question.Overall, n = 294Dermatology, n = 28Hematology,Nephrology,Oncology, and Rheumatology,†Specialties more likely to routinely prescribe allopurinol. n = 37Other Internal Medicine, n = 96Neurology and Psychiatry,‡Specialties more likely to routinely prescribe carbamazepine. n = 78Other, n = 55P value§P values were calculated by using Chi-squared or Fisher's exact test comparing question responses across various physician specialties.Carbamazepine Most common inducer of SJS and TEN listed [carbamazepine]211/294 (72)27/28 (96)21/37 (57)74/96 (77)58/78 (74)31/55 (56)<.001 Ethnicities at highest risk of carbamazepine-induced SJS and TEN [Asian]82/290 (28)21/28 (75)4/37 (11)22/96 (23)30/78 (39)5/51 (10)<.001 Genetic background strongly associated with carbamazepine-induced SJS and TEN [HLA-B*15:02]46/287 (16)16/28 (57)4/37 (11)8/96 (8)14/78 (18)4/48 (8)<.001 Aware of FDA recommendations¶US FDA guidelines (2007) recommend genetic screening of patients of Asian ancestry for HLA-B*15:02 before starting carbamazepine. HLA-B*15:02-positive patients should not be treated with carbamazepine. This recommendation is based on data from Hung SI, Chung WH, Jee SH, et al. Genetic susceptibility to carbamazepine-induced cutaneous adverse drug reactions. Pharmacogenet Genomics. 2006;16(4):297-306. for genetic screening for HLA-B*15:02 in Asian patients before carbamazepine use [yes]52/286 (18)16/28 (57)3/37 (8)9/96 (9)21/78 (27)3/47 (6)<.001 Aware of clinical availability of genetic testing for HLA-B*15:02 in Canada [yes]75/285 (26)16/28 (57)6/37 (16)22/96 (23)22/78 (28)9/46 (20).006Allopurinol Most common inducer of SJS and TEN listed [allopurinol]153/270 (57)27/28 (96)29/37 (78)71/96 (74)18/78 (23)8/31 (26)<.001 Ethnicities at highest risk of allopurinol-induced SJS and TEN [Asian and European]21/267 (8)8/28 (29)2/37 (5)6/96 (6)4/78 (5)1/28 (4)<.001 Genetic background strongly associated with allopurinol-induced SJS and TEN [HLA-B*58:01]28/265 (11)8/28 (29)10/37 (27)8/96 (8)1/78 (1)1/26 (4)<.001 Aware of 2012 ACR guideline recommendations‖The ACR guidelines (2012) recommend the consideration of genetic screening of subpopulations at higher risk for severe allopurinol hypersensitivity syndrome (ie, Koreans with stage 3 or worse chronic kidney disease and all Han Chinese and Thai patients) for HLA-B*58:01 before prescribing allopurinol. HLA-B*58:01-positive patients should not be treated with allopurinol; febuxostat should be instead considered. This recommendation is based on data from Hung SI, Chung WH, Liou LB, et al. HLA-B*5801 allele as a genetic marker for severe cutaneous adverse reactions caused by allopurinol. Proc Natl Acad Sci U S A. 2005;102(11):4134-4139. for genetic screening for HLA-B*58:01 in high-risk populations before allopurinol use [yes]37/264 (14)9/28 (32)8/37 (22)17/96 (18)3/78 (4)0/25 (0)<.001 Aware of clinical availability of genetic testing for HLA-B*58:01 in Canada [yes]120/264 (45)13/28 (46)16/37 (43)54/96 (56)29/78 (37)8/25 (32).18ACR, American College of Rheumatology; FDA, Food and Drug Administration; HLA-B, human leukocyte antigen B; SJS, Stevens-Johnson syndrome; TEN, toxic epidermal necrolysis.∗ Percentages might not sum to 100% due to uncategorized responses. Percentages calculated relative to number of respondents for each question.† Specialties more likely to routinely prescribe allopurinol.‡ Specialties more likely to routinely prescribe carbamazepine.§ P values were calculated by using Chi-squared or Fisher's exact test comparing question responses across various physician specialties.¶ US FDA guidelines (2007) recommend genetic screening of patients of Asian ancestry for HLA-B*15:02 before starting carbamazepine. HLA-B*15:02-positive patients should not be treated with carbamazepine. This recommendation is based on data from Hung SI, Chung WH, Jee SH, et al. Genetic susceptibility to carbamazepine-induced cutaneous adverse drug reactions. Pharmacogenet Genomics. 2006;16(4):297-306.‖ The ACR guidelines (2012) recommend the consideration of genetic screening of subpopulations at higher risk for severe allopurinol hypersensitivity syndrome (ie, Koreans with stage 3 or worse chronic kidney disease and all Han Chinese and Thai patients) for HLA-B*58:01 before prescribing allopurinol. HLA-B*58:01-positive patients should not be treated with allopurinol; febuxostat should be instead considered. This recommendation is based on data from Hung SI, Chung WH, Liou LB, et al. HLA-B*5801 allele as a genetic marker for severe cutaneous adverse reactions caused by allopurinol. Proc Natl Acad Sci U S A. 2005;102(11):4134-4139. Open table in a new tab Table IIParticipant responses to practice and opinion-related questions on genetic screening to prevent carbamazepine- and allopurinol-induced Stevens-Johnson syndrome and toxic epidermal necrolysisDrug, question, responseNo. participants responding/total (%)∗Percentages might not sum to 100% due to uncategorized responses. Percentages calculated relative to number of respondents for each question.Overall, n = 294Dermatology, n = 28Hematology, NEphrology, ONcology, and RHeumatology,†Specialties more likely to routinely prescribe allopurinol. n = 37Other Internal Medicine, n = 96Neurology and psychiatry,‡Specialties more likely to routinely prescribe carbamazepine. n = 78Other, n = 55P value§P values were calculated using Chi-squared or Fisher's exact tests comparing question responses across various physician specialties.Carbamazepine Does or has ever prescribed, yes152/291 (52)2/28 (7)6/37 (16)50/93 (54)67/78 (86)27/55 (49)<.001 Prescribing pattern<.001Regularly19/151 (13)0/2 (0)0/6 (0)0/50 (0)18/67 (27)1/26 (4)Rarely130/151 (86)2/2 (100)6/6 (100)50/50 (100)47/67 (70)25/26 (96) Refers patients for genetic screening before use.43Always or almost always7/147 (5)0/2 (0)0/6 (0)2/50 (4)5/67 (7)0/22 (0)Sometimes8/147 (5)1/2 (50)0/6 (0)2/50 (4)5/67 (7)1/22 (5)Never100/147 (68)1/2 (50)2/6 (33)37/50 (74)45/67 (67)15/22 (68)Other32/147 (22)0/2 (0)4/6 (67)9/50 (18)12/67 (18)6/22 (27) Finds it worthwhile to refer at-risk patients for genetic screening before use.48Strongly agree89/145 (61)1/2 (50)5/6 (83)29/50 (58)43/67 (64)11/20 (55)Agree45/145 (31)1/2 (50)1/6 (17)17/50 (34)21/67 (31)5/20 (25)Neutral10/145 (7)0/2 (0)0/6 (0)4/50 (8)2/67 (3)4/20 (20)Strongly disagree1/145 (1)0/2 (0)0/6 (0)0/50 (0)1/67 (1)0/20 (0)Allopurinol Does or has ever prescribed, yes125/269 (46)6/28 (21)34/36 (94)72/96 (75)1/77 (1)12/31 (39)<.001 Prescribing pattern<.001Regularly46/125 (37)0/6 (0)25/35 (74)18/72 (25)1/1 (100)2/12 (17)Rarely76/125 (61)5/6 (83)9/35 (26)52/72 (72)0/1 (0)10/12 (83) Refers patients for genetic screening before use.10Always or almost always5/124 (4)2/6 (33)1/34 (3)2/72 (3)0/1 (0)0/11 (0)Sometimes5/124 (4)1/6 (17)1/34 (3)4/72 (6)0/1 (0)0/11 (0)Never102/124 (82)2/6 (33)29/34 (85)61/72 (85)1/1 (100)9/11 (82)Other12/124 (10)1/6 (17)3/34 (9)5/72 (7)0/1 (0)2/11 (18) Finds it worthwhile to refer at-risk patients for genetic screening before use.02Strongly agree74/123 (60)5/6 (83)22/34 (65)41/72 (57)0/1 (0)6/10 (60)Agree38/123 (31)1/6 (17)8/34 (24)28/72 (39)0/1 (0)1/10 (10)Neutral10/123 (8)0/6 (0)4/34 (12)3/72 (4)1/1 (100)2/10 (20)Strongly disagree1/123 (1)0/6 (0)0/34 (0)0/72 (0)0/1 (0)1/10 (10)∗ Percentages might not sum to 100% due to uncategorized responses. Percentages calculated relative to number of respondents for each question.† Specialties more likely to routinely prescribe allopurinol.‡ Specialties more likely to routinely prescribe carbamazepine.§ P values were calculated using Chi-squared or Fisher's exact tests comparing question responses across various physician specialties. Open table in a new tab ACR, American College of Rheumatology; FDA, Food and Drug Administration; HLA-B, human leukocyte antigen B; SJS, Stevens-Johnson syndrome; TEN, toxic epidermal necrolysis. From the 2188 invited physicians, 261 complete and 33 partial responses were returned. Participant responses to knowledge-related questions and practice and opinion-related questions are summarized in Tables I and II, respectively. Physicians were cognizant of carbamazepine and allopurinol as common inducers of SJS and TEN, but knowledge of the at-risk ethnicities and genetic associations was limited. Awareness of clinical availability of pertinent genetic screening and usage guidelines was generally low. The educational component of our study successfully highlighted this topic's clinical importance. Almost all participants (∼90%) agreed that genetic screening before prescribing these medications would be worthwhile. Furthermore, we received many responses regarding practical implementation of the knowledge provided. An internal medicine program director stated, “It's one of the few studies I've learned from! What's the route for testing? We have a large vulnerable Toronto population.” An academic dermatologist replied, “This is my reality. We have a huge Asian population. I need to know more … and persuade rheumatologists and nephrologists to do the test.” In Canada and the United States, testing for HLA-B*15:02 and HLA-B*58:01 can be ordered from academic centers and private laboratories. Turnaround times vary from 3 to 10 days. Costs are $150-300 USD/test and expected to decline. Study limitations include a reliance on self-reported information, small cohort, and restricted population. Our study found a major knowledge gap between the validated literature on genetic screening to prevent SJS and TEN relative to the (lack of) clinical insight into this topic among Canadian academic physicians. Our results identify a need for health authorities to educate practitioners regarding pharmacogenetic screening to prevent SJS and TEN and ultimately, as done in some countries, to create a national screening program of personalized safe prescribing. We thank Leah Szadkowski and George Tomlinson (Biostatistics Research Unit, University Health Network, Toronto, Canada) for their assistance with data analysis. They were compensated for their contributions.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.