J Antimicrob Chemother 2015; 70: 2648 – 2651 doi:10.1093/jac/dkv158 Advance Access publication 3 July 2015
Lack of cross-toxicity between isoniazid and ethionamide in severe cutaneous adverse drug reactions: a series of 25 consecutive confirmed cases Rannakoe J. Lehloenya1,2, Rudzani Muloiwa3, Sipho Dlamini4, Nomphelo Gantsho1, Gail Todd1 and Keertan Dheda2* 1
Division of Dermatology, Department of Medicine, University of Cape Town, Cape Town, South Africa; 2Lung Infection and Immunity Unit, Department of Medicine, University of Cape Town, Cape Town, South Africa; 3Department of Paediatrics and Child Health, University of Cape Town, Cape Town, South Africa; 4Division of Infectious Disease and HIV Medicine, Department of Medicine, University of Cape Town, Cape Town, South Africa *Corresponding author. Tel: +27-21-404-7654; Fax: +27-21-404-7651; E-mail:
[email protected]
Received 5 March 2015; returned 11 April 2015; revised 19 May 2015; accepted 25 May 2015 Background: Isoniazid and ethionamide are important first- and second-line anti-TB drugs (FLDs and SLDs), respectively. Ethionamide is a structural analogue of isoniazid and the two drugs share other similarities, including their metabolism, therapeutic targets, hepato-toxicity patterns and drug resistance. As a result, there has always been concern about possible cross-reactivity between them. Methods: Among 69 patients with drug rash with eosinophilia and systemic symptoms (DRESS), Stevens – Johnson syndrome (SJS) or toxic epidermal necrolysis (TEN) to FLDs, FLDs were stopped and SLDs added when the skin and laboratory parameters had settled. This was followed by sequential and additive rechallenge with FLDs. We report 25 consecutive cases that developed confirmed cutaneous adverse drug reactions (CADRs) to isoniazid or ethionamide used as FLD and SLD, respectively. Results: Sixty-nine participants who developed CADRs on FLDs were enrolled in the study. Twenty developed a rechallenge reaction to isoniazid and five reacted to ethionamide. Four of the 20 isoniazid cases were patch test positive, 3/20 were skin prick test positive and 13/20 reacted to oral rechallenge. All seven cases that were patch and skin prick test positive were associated with systemic reactions. Twenty of the 25 cases had DRESS and 5 had SJS/TEN. Twenty-three of the 25 cases with rechallenge reactions were HIV infected. Importantly, none of the cases that reacted to ethionamide during the rechallenge reacted to isoniazid and none who subsequently reacted to isoniazid reacted to ethionamide. Conclusions: Our findings strongly suggest that the risk of cross-reactivity of isoniazid and ethionamide in DRESS syndrome and SJS/TEN is low. These findings have implications for clinical management.
Introduction Isoniazid forms a cornerstone of the first-line treatment of TB and ethionamide is one of the most frequently used second-line drugs. There are several similarities between the two drugs. Ethionamide is a structural analogue of isoniazid and both target InhA, an enzyme involved in mycolic acid biosynthesis, a major component of the mycobacterial wall. Both are pro-drugs that require activation by different mycobacterial enzymes (catalase-peroxidase KatG and mono-oxygenase EthA, respectively) for their action.1 Productive activation of the two drugs results essentially in the same final metabolite aimed at the same cellular target, namely InhA.1,2 Reactive intermediates of drug metabolism by the host
are thought to form the basis of the hapten theory of drug hypersensitivity reactions by covalently binding with proteins to form antigens.3 – 5 Hepatotoxicity of ethionamide and isoniazid is well established and their hepatocellular pattern of enzyme elevation is similar, as is the overlapping drug resistance pattern.2 Based on these similarities, there has always been concern about possible cross-toxicities between the two drugs. This has obvious implications for clinical management. We report our findings in 25 consecutive patients, 20 of whom had initially developed cutaneous adverse drug reactions (CADRs) that were subsequently confirmed to be due isoniazid, and 5 of whom had reacted to ethionamide used as a second-line anti-TB drug (SLD).
# The Author 2015. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail:
[email protected]
2648
No cross-reactivity between isoniazid and ethionamide in severe drug eruptions
JAC
Patients and methods
Results
Study setting
Demographic and clinical characteristics
The patients initially presented with TB-associated CADRs to a dermatology ward at Groote Schuur Hospital, a tertiary referral centre in Cape Town, South Africa, between January 2010 and June 2014. This series was part of an ongoing randomized controlled trial comparing sequential, additive and incremental dosing versus sequential, additive and full therapeutic dose re-introduction of first-line anti-TB drugs (FLDs) following severe TB-associated CADRs. The study was approved by our Human Research Ethics Committee (REF: 246/2009). All the participants provided informed consent for study inclusion.
Sixty-nine patients (62 of whom were HIV infected) who had developed CADRs while on at least two FLDs were enrolled in the study. Twenty participants developed rechallenge reactions to isoniazid and five reacted to ethionamide. Their median age was 38 (31 – 57) and 36 (17 – 51) years, respectively. There were 10 males and 10 females in the isoniazid group compared with 3 males and 2 females in the ethionamide group. Four cases in the ethionamide group had pulmonary TB and one case had disseminated disease, and in the isoniazid group there were 13 and 7 cases, respectively. Only 18/20 isoniazid cases were HIV infected compared with all 5 ethionamide cases. The median CD4 count of the two groups was 91 and 126 cells/mm3, respectively. At presentation, 3/20 isoniazid cases were on ART compared with 1/5 of those who reacted to ethionamide. None of the 25 participants reported a previous drug reaction. At the time of developing the initial CADR, 24 cases were on Rifafourw (a combination pill made up of rifampicin, isoniazid, pyrazinamide and ethambutol). One of the cases with rifampicin mono-resistance was on the other three FLDs. Nine of the 20 isoniazid cases and 4 of the 5 ethionamide cases were on co-trimoxazole for Pneumocystis jirovecii pneumonia prophylaxis when they developed the initial CADR. There were 16/20 cases of DRESS and 4/20 of SJS/TEN in the isoniazid group compared with 4/5 cases of DRESS and 1/5 of SJS/TEN in the ethionamide group.
Definitions CADRs refer to uncommon skin disorders that are often characterized by severity (associated with significant morbidity and mortality and usually leading to hospitalization), are non-predictable (idiosyncratic and probably with an immunological mechanism) and are induced by drugs.6 Amongst others, CADRs include drug rash with eosinophilia and systemic symptoms (DRESS), Stevens –Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). DRESS was defined as a drug eruption associated with a long latency period (.3 weeks) and was characterized by fever, widespread maculopapular rash, white cell abnormalities and multi-organ involvement.7 SJS and TEN were considered as a spectrum of the same disease, characterized by epidermal detachment and mucous membrane erosions. In SJS, ,10% of the body surface area is affected, while in TEN .30% of the body surface area is affected. The SJS/TEN overlap lies between these two extremes.8 A rechallenge reaction was defined as a recurrence of some or all of the signs or symptoms that previously resulted in discontinuation of drug(s) on re-exposure.9
Management of medications The rechallenge process was performed as we previously described.10 In brief, we stopped all anti-TB drugs when the initial CADR was suspected. Once the CADR had resolved and laboratory parameters returned to baseline, we initiated three SLDs to which the patient had not previously been exposed as bridging therapy. This was to minimize the risk of developing mycobacterial resistance during the prolonged rechallenge process with the individual drugs. We used a combination of aminoglycosides (streptomycin, amikacin or kanamycin), fluoroquinolones (ofloxacin or moxifloxacin), terizidone, ethionamide, linezolid, clofazimine or para-aminosalicylic acid depending on our hospital’s treatment guidelines, drug availability and toxicity profile in relation to the participant’s co-morbidities. The rechallenge with FLDs was started 2 weeks after initiating SLD bridging therapy to avoid confusion in cases of multiple drug hypersensitivity, as we previously reported with both FLDs and SLDs in this population.10 The drugs used depended on the participant’s drug sensitivities, completion of the intensive phase of treatment and their co-morbidities. The sequence of rechallenge was sequential and additive in the order isoniazid, rifampicin, pyrazinamide and ethambutol. Rechallenge started with a patch test with the drug, which, if negative, was followed by a skin prick test with the same drug. If both the patch and prick tests were negative, oral drug rechallenge followed.10 The severity of the rechallenge reactions and the reactions to the SLD bridging therapy were graded according to the National Cancer Institute (NCI) common terminology criteria for adverse events (CTCAE), version 4.03 (NCI, Bethesda, MD, USA).11 We used the validated Naranjo score to assess the probability of isoniazid and ethionamide being the offending drugs in each case. The score is based on the pattern of response, temporal relationship with the drug, withdrawing the drug, rechallenge with the drug, alternative causes, placebo response, drug levels in body fluids or tissues, dose– response relationship, previous patient experience with the drug, and confirmation by objective evidence.12
Frequency and severity of the drug reactions These are summarized in Table 1. Four of the 20 isoniazidassociated rechallenge reactions were patch test positive, 3/20 were skin prick test positive and 13/20 reacted to oral rechallenge. Surprisingly, all seven cases that were patch and skin prick test positive were associated with systemic features rather than reactions localized to the test area. These are summarized in Table 2. All five participants who reacted to ethionamide did not develop a rechallenge reaction to isoniazid. Based on the CTCAE, the isoniazid rechallenge reactions were graded as follows: 8/20 were grade 1 (mild), 7/20 grade 2 (moderate), 3/20 grade 3 (severe) and 2/20 grade 4 (life threatening). The life threatening cases were two anaphylactic reactions. Amongst the ethionamide cases, 3/5 were grade 1 and there was a case each of grade 2 and grade 4. The grade 4 reaction was a severe case of hepatitis in a participant with low hepatic baseline from severe hepatitis since the initial CADR. Based on the Naranjo score, 19/20 cases that reacted to isoniazid were classified as definite cases due to isoniazid toxicity, and 1 of the 20 patients was a probable case. Similarly, 3/5 cases that reacted to ethionamide were definite cases and 2/5 were probable cases.
Discussion We report a lack of cross-reactivity between isoniazid and ethionamide in a series of predominantly HIV-infected patients who had initially developed severe TB-associated CADRs, including SJS/TEN and DRESS. The patients subsequently developed rechallenge reactions to isoniazid and CADRs secondary to ethionamide, the latter used as bridging therapy during the rechallenge process.
2649
Lehloenya et al.
Table 1. Drugs used and characteristics of 25 cases of CADRs that reacted to isoniazid and ethionamide
Case
Initial CADR
SLDs used
Offending drug
Method of rechallenge
Interval before reaction
CTCAE grade of reactiona
Final regimen
1 2 3 4 5 6 7 8 9 10 11 12
SJS DRESS DRESS SJS/TEN DRESS DRESS DRESS DRESS SJS/TEN DRESS DRESS DRESS
TRD; OFX; ETH STR; TRD; OFX; KAN; ETH STR; OFX; KAN; ETH STR; TRD; OFX; ETH STR; OFX; ETH STR; TRD; OFX; ETH STR; OFX; ETH STR; OFX; AMK; ETH STR; OFX; ETH STR; OFX STR; MOX; ETH STR; MOX; ETH
INH INH INH INH INH ETH ETH INH INH INH INH INH
patch prick prick patch patch oral oral oral oral prick oral oral
24 h 24 h 24 h 48 h 24 h 96 h 2h 48 h 24 h 24 h 24 h 24 h
1 1 2 3 2 1 4 2 3 2 1 1
lost to follow-up RIF; PZA; STR; TRD EMB; ETH; OFX; AMK RIF; ETH; TRD RIF; ETH; OFX; STR INH; PZA; EMB; OFX; STR RIF; INH RIF; PZA; EMB; ETH; STR RIF; PZA; EMB RIF; PZA; EMB RIF; PZA; EMB; ETH deceased due to unrelated cause RIF; PZA; TRD; MOX RIF; PZA; STR; TRD; ETH EMB; ETH; MOX RIF; INH RIF; INH; PZA
13 14 15 16 17
SJS DRESS DRESS DRESS SJS
INH INH INH ETH ETH
oral patch oral oral oral
24 h 24 h 30 min 72 h 96 h
1 2 4 2 1
18 19 20 21 22
DRESS DRESS DRESS DRESS DRESS
INH INH INH INH ETH
oral oral oral oral oral
48 h 24 h 24 h 24 h 48 h
2 3 1 1 1
23 24
DRESS DRESS
INH INH
oral oral
24 h 72 h
2 4
RIF; PZA; EMB; MOX TRD; MOX; RFB PZA; EMB; ETH; RFB lost to follow-up PZA; TRD; OFX; AMK; PAS; KNM PZA; EMB; RFB PZA; EMB; MOX; RFB
25
DRESS
STR; TRD; OFX; MOX; ETH STR; TRD; MOX; ETH STR; TRD; MOX; ETH STR; TRD; MOX; ETH STR; TRD; MOX; PAS; LZD; AMK; ETH STR; MOX; ETH STR; MOX; TRD STR; MOX; KAN; ETH STR; TRD; OFX; ETH TRD; OFX; KAN; PAS; AMK; CLF; ETH STR; MOX; KAN; ETH TRD; MOX; KAN; AMK; ETH MOX; KAN; ETH
INH
oral
24 h
1
RIF; PZA
STR, streptomycin; OFX, ofloxacin; MOX, moxifloxacin; ETH, ethionamide; TRD, terizidone; PAS, para-aminosalicylic acid; LZD, linezolid; AMK, amikacin; KAN, kanamycin; CLF, clofazimine; INH, isoniazid; RIF, rifampicin; PZA, pyrazinamide; EMB, ethambutol; RFB, rifabutin. a 1, mild; 2, moderate; 3, severe; 4, life threatening; 5, death.
The reactions to isoniazid patch and skin prick tests were all systemic rather than confined to the test area. All the reactions were definite or probable cases caused by each drug based on the Naranjo score. In this series we characterized 20 participants who reacted to isoniazid and 5 who reacted to ethionamide. The most significant finding is that 22/25 (88%) of the participants had been exposed to isoniazid and ethionamide and none reacted to both. There has been a strong suggestion that there is no cross-reactivity between the two drugs in hepatic injury and there are reports of patients with ethionamide-associated liver injury who later tolerated isoniazid.13 There has been limited direct evidence on the crossreactivity between isoniazid and ethambutol in CADRs. In this study we provide substantial evidence supporting the lack of crossreactivity between the two drugs in predominantly HIV-infected persons with SJS/TEN and DRESS syndrome, both considered severe CADRs. There is a limited number of effective anti-TB drugs, and this study affirms that replacement of isoniazid with ethionamide is safe in complicated cases of severe drug reactions. It is therefore relevant and helpful to clinical management.
2650
However, larger studies in different populations and clinical settings are needed to confirm our findings. We used the validated Naranjo score to confirm the likelihood of the two drugs being the offender in each case. The rechallenge process facilitated the majority of isoniazid cases being assessed as definite cases as there was a recurrence of the reaction on re-exposure to the drug. Three of the five ethionamide cases recurred on re-exposure to the drug and hence their Naranjo score was ‘definite’ in nature. In the other two cases the clinical features of CADRs improved on withdrawal of ethionamide while continuing with the other SLDs, hence they could only be classified as probable cases. Surprisingly, all the patch and skin prick tests in this series resulted in systemic reactions rather than being localized to the test area. This is a subject of ongoing investigation and analysis in the parent study. The CTCAE were used to determine the severity of the original CADRs, rechallenge reactions to isoniazid and reactions to ethionamide. The rechallenge reactions to isoniazid ranged from mild to life threatening. The rechallenge was performed under close supervision, in a clinical setting of limited alternative treatment
No cross-reactivity between isoniazid and ethionamide in severe drug eruptions
Table 2. Characteristics of the rechallenge reactions in seven cases who reacted to patch and skin prick tests with isoniazid
JAC
Transparency declarations None to declare.
Case
Mode of diagnosis
1
patch
2 3
prick prick
4
patch
5
patch
10
prick
14
patch
a
Clinical features
CTCAE grade of reactiona
transaminitis, eosinophilia, fever, morbilliform rash morbilliform rash, nausea transaminitis, eosinophilia, fever, morbilliform rash, oedema transaminitis, eosinophilia, fever, epidermal necrosis ,30%, mucositis transaminitis, eosinophilia, fever, morbilliform rash, nausea, vomiting transaminitis, eosinophilia, morbilliform rash, oedema transaminitis, morbilliform rash
1
References
1 2
1 Morlock GP, Metchock B, Sikes D et al. ethA, inhA, and katG loci of ethionamide-resistant clinical Mycobacterium tuberculosis isolates. Antimicrob Agents Chemother 2003; 47: 3799– 805.
3
2 DeBarber AE, Mdluli K, Bosman M et al. Ethionamide activation and sensitivity in multidrug-resistant Mycobacterium tuberculosis. Proc Natl Acad Sci USA 2000; 97: 9677– 82.
2
3 Park BK, Naisbitt DJ, Gordon SF et al. Metabolic activation in drug allergies. Toxicology 2001; 158: 11– 23.
2 2
1, mild; 2, moderate; 3, severe; 4, life threatening; 5, death.
and having assessed the possible outcomes of sub-optimal TB treatment to be worse than those of a possible rechallenge reaction.14,15 Thus, until further data are available we do not recommend routine rechallenge with FLDs following severe CADRs unless serious consideration is given to associated risks compared with benefits. In South Africa in 2013, 90% of all TB cases were of known HIV status and 62% of these were HIV infected in 2013. The incidence of CADRs is also much higher in HIV-infected persons.16,17 Consequently, the high proportion of our cases that were HIV infected is not an unexpected finding. It is not clear whether our findings are generalizable to HIV-uninfected populations. Further studies are needed to answer this question. The major limitation of the study is the relatively small number of participants who reacted to ethionamide in comparison with isoniazid. There were also substantially fewer cases of SJS/TEN, making it difficult to generalize the findings to include both SJS/ TEN and DRESS syndrome. However, it is the largest reported series and its importance lies in the implications for clinical management in complex cases. Our findings add to the literature assuaging clinical anxiety about the use of ethionamide in patients with TB who develop CADRs, including DRESS syndrome and SJS/TEN, to FLDs, including isoniazid. However, larger prospective studies are needed in different populations and phenotypes of CADRs to confirm these findings.
Funding This work was supported by grants from the Discovery Foundation (to R. J. L.), the Dermatological Society of South Africa (to R. J. L.) and the Carnegie Corporation (to R. J. L.).
4 Schnyder B, Mauri-Hellweg D, Zanni M et al. Direct, MHC-dependent presentation of the drug sulfamethoxazole to human alphabeta T cell clones. J Clin Invest 1997; 100: 136– 41. 5 Zanni MP, von Greyerz S, Schnyder B et al. HLA-restricted, processingand metabolism-independent pathway of drug recognition by human alpha beta T lymphocytes. J Clin Invest 1998; 102: 1591 –8. 6 Roujeau JC, Allanore L, Liss Y et al. Severe cutaneous adverse reactions to drugs (SCAR): definitions, diagnostic criteria, genetic predisposition. Dermatol Sinica 2009; 27: 203–9. 7 Rive CM, Bourke J, Phillips EJ. Testing for drug hypersensitivity syndromes. Clin Biochem Rev 2013; 34: 15–38. 8 Bastuji-Garin S, Rzany B, Stern RS et al. Clinical classification of cases of toxic epidermal necrolysis, Stevens-Johnson syndrome, and erythema multiforme. Arch Dermatol 1993; 129: 92– 6. 9 Girard M. Conclusiveness of rechallenge in the interpretation of adverse drug reactions. Br J Clin Pharmacol 1987; 23: 73–9. 10 Lehloenya RJ, Wallace J, Todd G et al. Multiple drug hypersensitivity reactions to anti-tuberculosis drugs: five cases in HIV-infected patients. Int J Tuberc Lung Dis 2012; 16: 1260– 4. 11 US Department of Health and Human Services. Common Terminology Criteria for Adverse Events (CTCAE), Version 4.03. Bethesda, MD, USA: National Institutes of Health, National Cancer Institute, 2010. 12 Naranjo CA, Busto U, Sellers EM et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981; 30: 239–45. 13 Gex-Collet C, Helbling A, Pichler WJ. Multiple drug hypersensitivity— proof of multiple drug hypersensitivity by patch and lymphocyte transformation tests. J Investig Allergol Clin Immunol 2005; 15: 293– 6. 14 Lehloenya RJ, Dheda K. Cutaneous adverse drug reactions to antituberculosis drugs: state of the art and into the future. Expert Rev Anti Infect Ther 2012; 10: 475– 86. 15 Lehloenya RJ, Todd G, Badri M et al. Outcomes of reintroducing antituberculosis drugs following cutaneous adverse drug reactions. Int J Tuberc Lung Dis 2011; 15: 1649 –57. 16 WHO. Global Tuberculosis Report 2014. Geneva: WHO, 2014. 17 Coopman SA, Johnson RA, Platt R et al. Cutaneous disease and drug reactions in HIV infection. N Engl J Med 1993; 328: 1670 –4.
2651
