The Pharmacogenomics Journal (2014) 14, 493–502 © 2014 Macmillan Publishers Limited All rights reserved 1470-269X/14 www.nature.com/tpj

REVIEW

Testing for thiopurine methyltransferase status for safe and effective thiopurine administration: a systematic review of clinical guidance documents HF Burnett1, R Tanoshima2, W Chandranipapongse2, P Madadi2, S Ito2,3 and WJ Ungar1,4 A common pharmacogenomic test is for thiopurine S-methyltransferase (TPMT) status prior to treatment with thiopurine drugs, used to treat auto-immune conditions and pediatric cancer. Guidelines assist practitioners with decisions regarding testing and treatment. The objectives were to conduct a systematic review and critical appraisal of guidance documents with statements regarding TPMT testing and thiopurine dosing. Guidelines, clinical protocols and care pathways from all disciplines were eligible. A quality appraisal was carried out by three appraisers using the Appraisal of Guidelines for Research and Evaluation II. Of the 20 documents found, 5 recommended genotyping while 4 recommended phenotyping. Thirteen documents provided dosing recommendations based on TPMT status. The quality appraisal revealed wide variation across documents. The National Institute for Health and Clinical Excellence and Cincinnati Children’s Hospital guidelines demonstrated the highest overall quality with scores of 79 and 76, respectively. Low-scoring documents failed to use systematic methods to develop recommendations or to provide evidence to support recommendations. Guidance documents that included dosing recommendations demonstrated higher quality. The Pharmacogenomics Journal (2014) 14, 493–502; doi:10.1038/tpj.2014.47; published online 26 August 2014

INTRODUCTION Although pharmacogenomic testing applications are increasing, the use of pretreatment testing by clinical practitioners to avert adverse drug events may be impeded by a lack of education and/or awareness, uncertainty surrounding which test to order and skepticism that test results will translate into improved outcomes.1,2 Clinical guidance can assist physicians in the appropriate use of testing to guide drug therapies.3,4 This requires the development of rigorous evidence-based guidelines, protocols or care maps.3 A common application is testing for deficiency in thiopurine Smethyltransferase (TPMT), the enzyme that metabolizes thiopurines.5 Thiopurines are a class of immunosuppressive and chemotherapeutic drugs that are used to treat inflammatory bowel disease (IBD), autoimmune hepatitis, idiopathic arthritis and dermatological conditions. Thiopurines are also used as maintenance therapy in acute lymphoblastic leukemia (ALL) and to prevent posttransplant organ transplant rejection.6,7 Commonly used thiopurines include azathioprine (AZA), 6-mercaptopurine (6MP) and 6-thioguanine (6-TG). Approximately 89% of Caucasians have normal (that is, fully functional) TPMT activity, 11% are genetically heterozygous and are intermediate metabolizers and 0.3% have genetic variants resulting in undetectable enzyme activity.8,9 Patients with reduced or undetectable TPMT activity treated with standard doses of thiopurines are at risk of serious life-threatening adverse events, including myelosuppression, anemia, bleeding, leukopenia and severe infection.10 These adverse events can result in lengthy

hospital admissions and substantial morbidity and reduced quality of life for patients already coping with a serious illness.11,12 It is therefore important to identify the presence of TPMT deficiencies in patients prescribed thiopurine drugs. In the absence of TPMT testing, patients begin treatment with standard doses of thiopurines and are monitored for neutropenia by means of white blood cell counts. In these patients, up- or down-titration is often required to achieve an optimal therapeutic dose but with delayed benefit for patients with fully functional TPMT activity and risk of toxicity for patients with reduced or deficient TPMT activity.13 When TPMT status is known, patients can achieve an optimal therapeutic dose faster and avoid the risk of toxicity.14 However, even when TPMT testing reveals a normal result, the presence of undetected deficiencies in other enzymes involved in the metabolic pathway, such as inosine triphosphate pyrophosphatase, suggests that the risk of adverse events persists and complete blood count monitoring should be maintained.15 There are two approaches to testing for TPMT status. A phenotype test measures the level of TPMT enzyme activity by first isolating the enzyme and then adding a thiopurine substrate and quantifying metabolite levels.16–19 Results of the enzymatic assay can be confounded by concomitant medications or recent blood transfusions.20 Washing of red blood cells prior to analysis is necessary to avoid interference by other drugs. Genotype tests that detect the presence of a number of known single-nucleotide polymorphisms that code for a deficient TPMT enzyme are more versatile, but most commercially available tests capture only a portion of known genetic variants.21,22 There is disagreement

1 Program of Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, ON, Canada; 2Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, ON, Canada; 3Departments of Medicine, Pharmacology and Pharmacy, and Paediatrics, Faculty of Medicine, University of Toronto, Toronto, ON, Canada and 4Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada. Correspondence: Dr WJ Ungar, Program of Child Health Evaluative Sciences, The Hospital for Sick Children Peter Gilgan Centre for Research and Learning, 11th Floor, 686 Bay Street, Toronto, ON M5G 0A4, Canada. E-mail: [email protected] Received 27 March 2014; revised 10 June 2014; accepted 14 July 2014; published online 26 August 2014

Systematic review of TPMT testing HF Burnett et al

494 regarding whether the enzyme activity assay (phenotype) or genotype test is the most appropriate strategy. The objective of this study was to conduct a systematic review and critical appraisal of clinical guidance documents that include a statement regarding TPMT testing prior to the administration of thiopurine drugs. MATERIALS AND METHODS Literature search As a test to be used to inform drug selection and dosing, it was recognized that statements regarding TPMT testing could emanate from a wide range of guidance documents. An inclusive search strategy was therefore devised to include clinical practice guidelines (CPGs) or treatment protocols or care pathways, regardless of scope or purpose, that included statements or recommendation(s) regarding genotype or phenotype TPMT testing for any clinical condition, specialty or discipline for thiopurine administration in any age group (adults, children, mixed populations). Multiple documents released from the same organization in different years were included. Guidance documents that discussed TPMT activity but did not make a recommendation for or against testing were excluded. Laboratory protocols and non-English documents were excluded. Medline, Embase, and the Cumulative Index to Nursing and Allied Health Literature were searched between 1980 and September 2012. Medical subject headings included ‘Practice Guideline’, ‘Guideline’, ‘Clinical Protocols’, ‘Critical Pathways’, ‘Decision Support Systems, Clinical’, ‘6-mercaptopurine’, ‘azathioprine’ and ‘thioguanine’. Keywords included but were not limited to ‘TPMT’, ‘thiopurine methyltransferase’, ‘recommendation’, ‘clinical consensus’ and ‘consensus statement’. Grey literature searches included the National Guideline Clearinghouse, Guidelines International Network and other international guideline databases, national guideline agency websites, government agency websites and medical organization websites. Reference lists of identified articles were hand-searched for eligible guidelines. Results from the literature search were exported into a single EndNote library and duplicate documents were removed. Titles and abstracts were screened by a single reviewer (HB) to determine eligibility. The same reviewer then examined the full text of all remaining studies, applying the inclusion and exclusion criteria. Data extraction A data extraction tool was devised to collect data, including publication year, authors, target audience, target population or condition, organization or group the document was produced for or endorsed by and whether or not systematic methods were used to create the document. Details of TPMT testing recommendations were extracted as were recommendation statements for test type and dosing, evidence grades or quality assigned to TPMT recommendations and the sections of the document where TPMT recommendations were found. TPMT recommendations were categorized as recommending genotype or phenotype testing. Recommendations that included vague statements to ‘test’, ‘measure’, ‘check’ or ‘assess’ TPMT or TPMT activity were categorized as not having specified a test type. Guidelines that recommended ‘genotype or phenotype’ testing were also categorized as such. Recommendations that explicitly referred to genotyping or gene polymorphisms, including those that referred to testing only a specific patient population (for example, patients who had undergone a recent blood transfusion) or genotyping prior to another adjunct test were categorized as genotyping recommendations. Recommendations that referred to TPMT levels or TPMT testing and thiopurine metabolites were categorized as phenotype testing recommendations. The Pharmacogenomics Journal (2014), 493 – 502

Quality appraisal The quality of guidance documents was assessed by three independent appraisers (HB, WC, RT) using the Appraisal of Guidelines for Research and Evaluation II (AGREE-II) Instrument.23 AGREE-II was used to assess each document across six independent domains consisting of scope and purpose (three items), stakeholder involvement (three items), rigor of development (eight items), clarity of presentation (three items), applicability (four items) and editorial independence (two items). Websites of guideline developers were examined for additional information when necessary. Independent scoring of each item was carried out using a 7-point scale (anchored at 1—strongly disagree and 7—strongly agree). Higher scores signified higher quality. Domains scores were calculated by summing the scores for each item within a domain and scaling the total as a percentage of the maximum possible score for that domain (assuming all items scored 7). This allowed standardization of domain scores from 0 (lowest score) to 100 (best score). Scores assigned by each appraiser for individual guidance documents were required to be within two points of agreement. When disagreement occurred, face-to-face discussion occurred to reach consensus within 2 points. Final domain scores were used to rank the quality of each guideline by domain. An overall score for each guideline was determined from the mean of the domain scores. RESULTS Scope of guidance documents A total of 370 documents were identified and reviewed for eligibility; 158 were excluded because they were not guidance documents, 104 because they did not include a TPMT recommendation statement and 88 because they were written in a language other than English. Twenty guidance documents were included spanning a wide range of patient populations: IBD (including Crohn’s disease and ulcerative colitis) (n = 8), inflammatory skin disorders (n = 3), autoimmune hepatitis (n = 3), rheumatic disease (n = 2), ALL (n = 2), and general pharmacogenetic testing (n = 2). Six of the guidance documents focused on the treatment of pediatric patients. Table 1 provides an overview of the document characteristics. Quality of guidance documents Wide variation in the quality of the guidance documents was observed across all AGREE domains (Table 2). The mean total score was 47.14 (s.d. = 18.94, range 10.42–78.59). The three highest quality documents were the IBD guideline produced by the National Institute for Health and Clinical Excellence (NICE),24 the pediatric IBD guideline produced by Cincinnati Children’s Hospital25 and the rheumatology guideline produced by the British Health Professionals in Rheumatology (BHPR).26 Overall, the guidance documents scored the highest in terms of objective and scope (domain 1) and the lowest in terms of applicability (domain 5). For objective and scope (domain 1), the highest quality guidance documents were the IBD guidelines produced by NICE24 and Cincinnati Children’s25 and the rheumatology guideline produced by the BHPR.26 The mean score for domain 1 was 57.8 (s.d. 22.2). For stakeholder involvement (domain 2), the highest quality guidance documents were the NICE IBD guidelines,24 Cincinnati Children’s25 and 2011 British Association of Dermatologists (BAD) guidelines.27 The mean score for domain 2 was 42.9 (s.d. 22.8). Low scores resulted from very few documents providing sufficient information regarding members of the guideline development group and the fact that the preferences of patients were not sought in the development process. For rigor of development (domain 3), the highest quality guidance documents were the NICE IBD guidelines,24 Cincinnati Children’s25 and © 2014 Macmillan Publishers Limited

© 2014 Macmillan Publishers Limited

Table 1.

Characteristics of guidelines that include recommendations for TPMT testing

Identifier, year

Organization

Focus

Target audience

Target condition/field

CPG CPG

Uni-disciplinary Uni-disciplinary

Gastroenterologists (pediatric) Gastroenterologists

Pediatric ulcerative colitis Crohn’s disease

Consensus statement CPG CPG CPG

Uni-disciplinary Uni-disciplinary Uni-disciplinary Uni-disciplinary

Gastroenterologists Gastroenterologists Gastroenterologists Gastroenterologists (pediatric)

CPG + algorithm

Uni-disciplinary

Gastroenterologists (pediatric)

Medical Position Statement

Uni-disciplinary

Gastroenterologists

Inflammatory bowel disease Inflammatory bowel disease Inflammatory bowel disease Pediatric inflammatory bowel disease Pediatric inflammatory bowel disease Inflammatory bowel disease

Inflammatory skin disorders BAD, 201127 British Association of Dermatologists BAD, 200441 British Association of Dermatologists 31 American Academy of Dermatology AAD, 2009

CPG CPG CPG

Uni-disciplinary Uni-disciplinary Uni-disciplinary

Dermatologists Dermatologists Dermatologists

Inflammatory dermatoses Inflammatory dermatoses Psoriasis

Autoimmune hepatitis BSG, 201138 AASLD, 201029

British Society of Gastroenterology American Association for the Study of Liver Diseases American Association for the Study of Liver Diseases

CPG CPG

Uni-disciplinary Uni-disciplinary

Gastroenterologists and hepatologists Gastroenterologists and hepatologists

Autoimmune hepatitis Autoimmune hepatitis

CPG

Uni-disciplinary

Gastroenterologists and hepatologists

Autoimmune hepatitis

The British Society for Paediatric and Adolescent Rheumatology British Health Professionals in Rheumatology

Medical Position Statement CPG

Uni-disciplinary

Rheumatologists (pediatric)

Pediatric rheumatology

Multi-disciplinary

Healthcare professionals, health service managers, patients, national societies

Rheumatic and dermatological conditions

CPG Clinical Protocol

Uni-disciplinary Uni-disciplinary

Oncologists (pediatric) Oncologists (pediatric)

Acute lymphoblastic leukemia Acute lymphoblastic leukemia

CPG

Multi-disciplinary

Clinicians

TPMT genotyping and dosing

CPG

Multi-disciplinary

Medical practitioners (physicians, nurses, pharmacists, clinical researchers)

Pharmacogenetic testing

Inflammatory bowel disease ECCO, 201245 European Crohn’s and Colitis Organization National Institute for Health and Clinical NICE, 201224 Excellence 54 Asian Pacific Association of Gastroenterology APAG, 2010 28 British Society of Gastroenterology BSG, 2010 36 WGO 2010 World Gastroenterology Organization 40 British Society of Paediatric Gastroenterology BSPGHN, 2008 Hepatology and Nutrition 25 CCHMC, 2007 Cincinnati Children's Hospital Medical Center AGA, 2006

39

AASLD, 200237 Rheumatic diseases BSPAR, 201146 BHPR, 2008

26

American Gastroenterological Association

The Pharmacogenomics Journal (2014), 493 – 502

Acute lymphoblastic leukemia NCCN, 201232 National Comprehensive Cancer Network Children's Oncology Group COG, 200820 General pharmacogenetic testing CPIC, 201130 Clinical Pharmacogenetics Implementation Consortium 33 NACB, 2010 The National Academy of Clinical Biochemistry

Systematic review of TPMT testing HF Burnett et al

Guidance type

Abbreviation: CPG, clinical practice guideline. Note: The ECCO guideline45 was also endorsed by the European Society for Paediatric Gastroenterology, Hepatology, and Nutrition. The British Health Professionals in Rheumatology guideline26 was also endorsed by the British Society for Rheumatology.

495

Systematic review of TPMT testing HF Burnett et al

496 Table 2.

Results of AGREE-II quality appraisal D1—Scope and purpose

D2—Stakeholder involvement

D3—Rigor of development

D4—Clarity of presentation

D5—Applicability

D6—Editorial independence

Score

Rank

Score

Rank

Score

Rank

Score

Rank

Score

Rank

Score

Rank

Inflammatory bowel disease ECCO, 2012 75.9 NICE, 2012 94.4 APAG, 2010 53.7 BSG, 2010 68.5 WGO, 2010 35.2 BSPGHN, 2008 68.5 CCHMC, 2007 87.0 AGA, 2006 42.6

5 1 10 7 17 7 3 16

35.2 79.6 55.6 68.5 11.1 46.3 81.5 59.3

11 2 7 3 18 8 1 6

50.7 86.8 43.8 46.5 5.6 42.4 79.2 31.9

9 1 11 10 20 12 3 15

63.0 64.8 53.7 37.0 48.1 66.7 87.0 55.6

8 7 11 16 14 6 1 10

16.7 70.8 11.1 20.8 18.1 25.0 50.0 20.8

12 2 14 7 11 6 3 7

56.0 75.0 55.6 38.9 0.0 75.0 69.4 33.3

Inflammatory skin disorders BAD, 2011 53.7 BAD, 2004 31.5 AAD, 2009 31.5

10 18 18

68.5 24.1 20.4

3 15 16

83.3 61.8 56.9

2 4 6

85.2 59.3 53.7

2 9 11

34.7 30.6 9.7

4 5 16

Autoimmune hepatitis BSG, 2011 51.9 AASLD, 2010 48.1 AASLD, 2003 48.1

12 13 13

27.8 38.9 35.2

14 9 11

34.0 32.6 30.6

13 14 16

72.2 50.0 33.3

4 13 17

11.1 8.3 8.3

Rheumatic diseases BSPAR, 2011 BHPR, 2008

20 2

9.3 68.5

19 3

9.7 56.9

19 6

31.5 42.6

18 15

13 4

29.6 NA

13 NA

19.4 52.8

18 8

16.7 68.5

20 5

Identifier, year

9.3 88.9

Acute lymphoblastic leukemia NCCN, 2012 48.1 COG, 2008 81.5

General pharmacogenetic testing CPIC, 2011 72.2 6 NACB, 2010 64.8 9 Mean (s.d.) 57.8 (22.2)

37.0 10 18.5 17 42.9 (22.9)

59.0 5 27.1 17 45.6 (22.6)

79.6 3 31.5 18 55.0 (19.1)

Overall Score

Overall Rank

9 6 10 13 19 6 8 15

49.6 78.6 45.6 46.7 19.7 54.0 75.7 40.6

8 1 11 10 19 7 2 13

80.6 88.9 97.2

5 4 1

67.7 49.3 44.9

4 9 12

14 17 17

13.9 38.9 2.8

16 13 17

35.1 36.1 26.4

15 14 18

0.0 77.8

19 1

2.8 91.7

17 3

10.4 71.1

20 3

19.4 NA

10 NA

47.2 NA

11 NA

30.1 67.6

17 5

20.8 7 12.5 13 24.6 (20.8)

97.2 1 41.7 12 52.9 (32.5)

61.0 6 32.7 16 47.1 (18.9)

Abbreviations: AAD, American Academy of Dermatology; AASLD, American Association for the Study of Liver Diseases; AGA, American Gastroenterological Association; AGREE-II, Appraisal of Guidelines for Research and Evaluation II; APAG, Asian Pacific Association of Gastroenterology; BAD, British Association of Dermatologists; BHPR, British Health Professionals in Rheumatology; BSPAR, The British Society for Paediatric and Adolescent Rheumatology; BSG, British Society of Gastroenterology; BSPGHN, British Society of Paediatric Gastroenterology Hepatology and Nutrition; CCHMC, Cincinnati Children's Hospital Medical Center; COG, Children’s Oncology Group; CPIC, Clinical Pharmacogenetics Implementation Consortium; ECCO, European Crohn's and Colitis Organization; NACB, National Academy for Clinical Biochemistry; NCCN, National Comprehensive Cancer Network; NICE, National Institute for Excellence; WGO, World Gastroenterology Organization.

the 2011 BAD guidelines.27 The mean score for domain 3 was 45.6 (s.d. 22.6). This domain focused on the quality of TPMT recommendations. In general, low-scoring guidance documents failed to use systematic methods to develop their recommendation statements. In cases where systematic reviews were carried out, very few documents provided sufficient evidence to support the recommendations or failed to link the recommendations to supporting evidence. In some cases, evidence used to support recommendations contradicted recommendation statements. For example, the 2010 British Society of Gastroenterology (BSG) guideline28 referred to several studies illustrating that TPMT status is a poor predictor of myelosuppression and other adverse events in patients with IBD, and the evidence to support TPMT testing prior to treatment with thiopurines was deemed ‘controversial’. However the authors nevertheless recommended ‘all patients be tested for TPMT levels before starting thiopurines, to avoid administration in patients with no functional TPMT in whom thiopurine administration may be fatal.’ Similarly, the 2010 American Association for the Study of Liver Disease guidance document29 stated that thiopurine toxicity was not well predicted by ‘genotyping or phenotyping for TPMT activity’ yet recommended TPMT testing as a ‘reasonable precaution’ that should be considered in all patients, especially those with pretreatment cytopenia, those with cytopenia developing during therapy or those patients who require higher than conventional doses of AZA. For clarity of presentation (domain 4), the highest quality The Pharmacogenomics Journal (2014), 493 – 502

guidance documents were produced by Cincinnati Children’s,25 the BAD27 and the Clinical Pharmacogenetics Implementation Consortium (CPIC).30 The mean score for domain 4 was 55.0 (s.d. 19.1). For applicability (domain 5), the highest quality guidance documents were the NICE IBD guidelines,24 Cincinnati Children’s25 and the joint rheumatology guideline produced by the BHPR.26 The mean score for domain 4 was 24.6 (s.d. 20.8), representing the lowest scoring domain. Low scores resulted from very few documents describing how guidelines can be implemented and monitored. For editorial independence (domain 6), the highest quality guidance documents were produced by the American Association of Dermatologists31, the BHPR26 and the CPIC.30 The mean score for domain 6 was 53.0 (s.d. 32.5), and this domain had the widest score variation across guidance documents. Genotype versus phenotype testing Five documents made explicit recommendations for genotype testing prior to initiation of thiopurine therapy.20,27,30,32,33 The CPIC30 and National Academy for Clinical Biochemistry (NACB)33 documents focused on the use of genotyping technologies (Table 3), but none of the documents recommended a specific type of genetic assay. The CPIC guideline recommends phenotype testing in conjunction with genotype testing,30 the 2011 BAD guideline recommends genotyping for patients with intermediate phenotypes,27 and the Children’s Oncology Group (COG) protocol © 2014 Macmillan Publishers Limited

Systematic review of TPMT testing HF Burnett et al

497 Table 3.

Guidelines recommending genotype testing in order to determine TPMT status

Identifier, year

Target condition/field

BAD, 2011

‘TPMT genotyping is only required for patients with indeterminate phenotype (i.e. borderline values) or those who have had a recent blood transfusion’ Acute lymphoblastic ‘TPMT testing should be performed if myelosupression leukaemia leads to delays in therapy. Genotyping may be preferable to phenotype testing in cases where a history of red cell transfusions would potentially confound assessments of TPMT activity.’ Acute lymphoblastic ‘For patients receiving 6-MP, consider testing for TPMT leukaemia gene polymorphisms, particularly in patients that develop severe neutropenia after starting 6-MP’ General pharmacogenetic ‘Genotype tests have a high likelihood of being testing informative. Complementary phenotype tests can be helpful adjuncts to genotyping tests’ General pharmacogenetic ‘TPMT genotyping is recommended as a useful adjunct testing to a regimen for prescribing azathioprine’

COG, 2008

NCCN, 2012 CPIC, 2011 NACB, 2010

Recommendation for TPMT testing

Reported strength of recommendation

Inflammatory skin disorders

Rigor of development score (rank)

Grade D, Level 4

83.3 (2)

Not reported

52.8 (8)

Not reported

19.4 (18)

Not reported

59.0 (5)

Grade A, Level I

27.1 (17)

Abbreviations: BAD, British Association of Dermatologists; COG, Children’s Oncology Group; CPIC, Clinical Pharmacogenetics Implementation Consortium; NACB, National Academy for Clinical Biochemistry; NCCN, National Comprehensive Cancer Network; TPMT, thiopurine S-methyltransferase.

Table 4.

Guidelines recommending phenotype testing in order to determine TPMT status

Identifier, year

Target condition/field

Recommendation for TPMT testing

Reported strength of recommendation

AAD, 2009

Inflammatory skin disorders Inflammatory bowel disease

‘TPMT levels are generally used to guide dosing’

Not reported

56.9 (7)

‘All patients should have measurement of TPMT levels before starting thiopurines, mainly to avoid administration to a patient with no functional TPMT’ ‘Where available, TPMT and thiopurine metabolite testing for 6-thioguanine and 6-methylmercaptopurine may assist dose optimization of AZA/6-MP’ ‘TPMT genotyping will be informative in all patients, if at least one mutant allele is identified. If not, and myelosuppression continues, send samples for TPMT activity and/or metabolites since TPMT genotyping will miss 5-10% of mutants.’

Grade B, Level 4

34.0 (12)

Not reported

43.8 (10)

Not reported

52.8 (8)

BSG, 2010 APAG, 2010

Inflammatory bowel disease

COG, 2008

Acute lymphoblastic leukaemia

Rigor of development score (rank)

Abbreviations: AAD, American Academy of Dermatology; APAG, Asian Pacific Association of Gastroenterology; BSG, British Society of Gastroenterology; COG, Children’s Oncology Group; TPMT, thiopurine S-methyltransferase.

recommends genotyping in patients with a history of blood transfusions.20 The National Comprehensive Cancer Network32 and NACB33 documents scored low for rigor of development, despite the NACB reporting Grade A (‘good evidence that it improves health outcomes and the benefits substantially outweigh harms’), Level I (‘consistent results from well-designed, wellconducted studies in representative populations’) evidence.33 The NACB recommendation was based on two review articles6,34 and results from a single retrospective cohort study of 171 kidney transplant patients.35 The cohort study included 12 patients heterozygous for TPMT status, 58% of whom required AZA dose reductions as a result of leukopenia (compared with 30% of wildtype patients).35 A statement in the introduction of the NACB guideline claims that, in rapidly evolving fields such as pharmacogenetics where evidence is uncertain, there is a need for robust recommendations regardless of whether or not rigorous evidencebased approaches can be applied.33 Three guidance documents recommend phenotype testing (Table 4). The COG protocol recommends phenotype testing for © 2014 Macmillan Publishers Limited

patients in whom genotyping was not informative.20 All phenotyping recommendations were moderate in terms of their score for rigor of development. Although several guidelines recommended either genotype or phenotype testing, many failed to specify the type of test. Thirteen documents made general statements about the need for TPMT testing, with several recommending either genotyping or phenotyping and others disregarding the test method, making vague statements to ‘test’, ‘measure’, ‘check’ or ‘assess’ TPMT status (Table 5). Thiopurine dose adjustments A total of 13 documents included thiopurine dosing recommendations based on TPMT status. The majority of dosing recommendations were statements to avoid AZA or 6-MP in patients who are homozygous mutant or who have extremely low or absent TPMT activity,24,26,27,29,36–40 and to reduce thiopurine doses in patients who are heterozygous or who have intermediate TPMT activity.24,26,40 Five documents included guidance on how to adjust thiopurine doses based on TPMT status, including a specific The Pharmacogenomics Journal (2014), 493 – 502

498

Guidelines recommending TPMT testing without specification of test type

Identifier, year

Target condition

Recommendation for TPMT testing

Reported strength of recommendation

BAD, 2011

Inflammatory skin disorders

‘TPMT activity should be checked in all patients prior to receiving azathioprine’

Grade A, Level 1+

ECCO, 2012

Inflammatory bowel disease NICE, 2012 Inflammatory bowel disease WGO, 2010 Inflammatory bowel disease BSPGHN, 2008 Inflammatory bowel disease CCHMC, 2007 Inflammatory bowel disease AGA, 2006 Inflammatory bowel disease BAD, 2004 Inflammatory skin disorders BSG, 2011 Autoimmune hepatitis AASLD, 2010

Autoimmune hepatitis

AASLD, 2003

Autoimmune hepatitis

BHPR, 2008 BSPAR, 2011

Rheumatic disease Pediatric rheumatic disease

‘TPMT testing only identifies a proportion of individuals at increased risk of Grade B, Level 2++ haematological toxicity, hence the continued need for regular monitoring of blood counts irrespective of TPMT status’ ‘The determination of TPMT genotype or phenotype, if available, is encouraged Not reported to identify patients at greater risk for early profound myelosuppression’ ‘Assess TPMT activity before offering AZA or 6-MP’ Not reported ‘Before starting AZA or 6MP measuring TPMT by phenotype (enzyme levels) or genotype will help direct dosing’ ‘TPMT should be checked prior to initiating treatment and is probably best done at diagnosis’ ‘It is recommended that TPMT genotype or phenotype be determined prior to initiation of 6-MP or AZA’ ‘Individuals should have TPMT genotype or phenotype assessed before initiation of therapy with AZA or 6-MP’ ‘Pre-treatment TPMT measurement should be performed in all patients prescribed AZA’ ‘TPMT measurement should be considered to exclude homozygous TPMT deficiency and is recommended in patients with pre-existing leucopenia’ ‘Azathioprine therapy should not be started in patients with known complete deficiency of TPMT activity’ ‘Pre-treatment testing for TPMT is a reasonable precaution, and it should be considered in all patients, especially those with pretreatment cytopenia’ ‘Perform TPMT assay prior to treatment with AZA’ ‘Pre-treatment testing: TPMT activity’

Rigor of developmentscore (rank) 83.3 (2)

30.6 (16) 86.8 (1)

Not reported

5.6 (20)

Not reported

42.4 (11)

1 large prospective study, 1 retrospective study, expert opinion and consensus Grade B

79.2 (3) 31.9 (14)

Not reported

61.8 (4)

Grade B2, Level II-iii

34.0 (12)

Class 3, Level C

32.6 (13)

Not reported

30.6 (15)

Not reported Not reported

56.9 (6) 9.7 (19)

© 2014 Macmillan Publishers Limited

Abbreviations: AASLD, American Association for the Study of Liver Diseases; AGA, American Gastroenterological Association; AZA, azathioprine; BAD, British Association of Dermatologists; BHPR, British Health Professionals in Rheumatology; BSPAR, The British Society for Paediatric and Adolescent Rheumatology; BSG, British Society of Gastroenterology; BSPGHN, British Society of Paediatric Gastroenterology Hepatology and Nutrition; CCHMC, Cincinnati Children's Hospital Medical Center; ECCO, European Crohn's and Colitis Organization; NICE, National Institute for Excellence; WGO, World Gastroenterology Organization; TPMT, thiopurine S-methyltransferase.

Systematic review of TPMT testing HF Burnett et al

The Pharmacogenomics Journal (2014), 493 – 502

Table 5.

Systematic review of TPMT testing HF Burnett et al

© 2014 Macmillan Publishers Limited

Abbreviations: AAD, American Academy of Dermatology; AZA, azathioprine; BAD, British Association of Dermatologists; CCHMC, Cincinnati Children's Hospital Medical Center; CPIC, Clinical Pharmacogenetics Implementation Consortium; NR, not reported; TPMT, thiopurine S-methyltransferase; U, units. aStrength of evidence for heterozygous dosing is Grade C, Level 2+. bStrength of evidence for heterozygous dosing is Grade B, Level III.

59.0 (5) 61.8 (4) 83.3 (2) 79.2 (3)

56.9 (6)

Grade A, Level II-ii NR Grade A, Level 2+ NR

Reported strength of recommendation Rigor of development score (rank)

Do not prescribe use of AZA

TPMT 5–13.7U: 0.5 mg kg − 1 Do not prescribe or, if used, (max) TPMT 13.7–19U: dose of 0.5–1 mg kg − 1 daily −1 1.5 mg kg (max) with more frequent monitoringb TPMT o5U: Do not Alternative therapies use AZA recommended Lowered dose, 1–1.5 mg kg − 1 dailya

1.5 mg kg − 1 daily and if labs are ok, advance over 4 weeks to 2.5 mg kg − 1 daily Do not use AZA Low or absent activity (homozygous mutant)

Normal activity (wild type)

Intermediate activity (heterozygous)

1–3 mg kg − 1 daily TPMTo19U: 2.5 mg kg − 1

Normal dose (2–3 mg kg − 1 daily), adjust based on disease-specific guidelines, allow 2 weeks to reach steady state 30–70% of target dose and titrate based on tolerance, allow 2–4 weeks to reach steady state Consider an alternative therapy, or, if using, reduce dose by 10-fold and titrate based on tolerance and disease-specific guidelines, allow 4–6 weeks to reach steady state Strong

Inflammatory skin disorder Inflammatory skin disorder

Inflammatory skin disorder Conventional dose (2–3 mg kg − 1 daily) Inflammatory bowel disease (pediatric) 2.5 mg kg − 1 daily Target condition

CPIC, 2011 BAD, 2004 AAD, 2009 BAD, 2011 CCHMC, 2007 Identifier, year

Dosing recommendations for azathioprine based on TPMT status

Implications for special populations Physiological factors, including age, sex and disease states, are known to contribute to individual variation in the pharmacokinetic and pharmacodynamic properties of administered drugs.43 A 2011 systematic review by the Agency for Healthcare Research and Quality44 concluded that there is currently insufficient evidence to support the clinical validity and utility of TPMT testing across the board in patients with any auto-inflammatory disease. Aside from the six documents focused on pediatric populations,20,25,32,40,45,46 none of the other documents considered or discussed age in the context of TPMT testing. The CPIC addressed the issue of age in a 2013 update, which included five new studies and concluded that ‘the original dosing recommendations can be used in both the adult and pediatric populations’.47 The authors justified this statement because a large proportion of the evidence supporting the original recommendations focused

Table 6.

DISCUSSION The present review reveals disagreement and a lack of methodological rigor in guidance documents for TPMT testing. The documents varied not only in scope but also in terms of the type of TPMT test recommended. Only a few of the documents scored high across more than three AGREE-II domains. Documents that paired recommendations with dose adjustments tended to provide more details on the methods used to generate recommendations, with most describing a systematic literature review. However, the mention of a systematic literature review in guideline development did not always result in a recommendation based on high-quality evidence. For example, the 2006 American Gastroenterological Association guideline39 assigned a high level of evidence to recommendations for TPMT testing (grade B), but the reference associated with the statement was not a clinical study but the Food and Drug Administration drug label warning, which recommends ‘TPMT genotyping or phenotyping (red blood cell TPMT activity) can identify patients who are homozygous deficient or have low or intermediate TPMT activity’.42 The Food and Drug Administration warns about the use of phenotype tests in patients who have received recent blood transfusions and the need for regular complete blood cell count monitoring.

None, general TPMT testing

499 adjusted dose or a percentage of the normal dose for each of the TPMT genotypes or phenotypes.20,24,27,30,41 The dosing recommendations are summarized for AZA, 6-MP and 6-TG in Tables 6–8, respectively. Overall, consistency was observed in recommended AZA dosing for patients with full, intermediate or low TPMT activity. The only exception was the CPIC guideline which recommends a 10-fold dose reduction with titration based on tolerance in patients with low or absent TPMT activity. All other documents recommend avoiding AZA in homozygous mutant patients. For 6-MP, slight variation was observed in recommended dose adjustments, with COG20 recommending 30–50% of a normal dose and CPIC30 recommending 30–70% in patients with intermediate activity. For patients with low or absent TPMT activity, COG20 recommends 10–20 mg m − 2 daily while CPIC30 recommends a 10-fold reduction in non-cancer patients. The Cincinnati Children’s25 document recommends avoiding 6-MP in patients with pediatric IBD. Only CPIC provides dosing recommendations for 6-TG based on TPMT status and advises a 10-fold dose reduction with dose adjustment based on tolerance and disease-specific guidelines.30 The CPIC30 and 2011 BAD27 guidelines acknowledged that alternative treatments should be administered in non-malignant patients with low TPMT activity, with the BAD guideline providing a list of alternative treatments to AZA. The CPIC guideline recommends dose reductions and does not recommend alternatives to 6-MP and 6-TG for patients with malignancy.

The Pharmacogenomics Journal (2014), 493 – 502

Systematic review of TPMT testing HF Burnett et al

500 Table 7.

Dosing recommendations for 6-mercaptopurine based on TPMT status

Identifier, year

CCHMC, 2007

COG, 2008

CPIC, 2011

IBD (pediatric) 1.5 mg kg − 1 daily

ALL (pediatric) Normal dose

0.75–1 mg kg − 1 daily and if labs are ok, advance over 4 weeks to 1.5 mg kg − 1kg daily Do not use 6-MP

30–50% of normal dose

Target condition Normal activity (wildtype) Intermediate activity (heterozygous) Low or absent activity (homozygous mutant)

Reported strength of recommendation Rigor of development score (rank)

NR

NR

None, general TPMT testing Normal dose (1.5 mg kg − 1 daily) allow 2 weeks to reach steady state 30–70% of target dose and titrate based on tolerance and disease-specific guidelines, allow 2–4 weeks to reach steady state Non-malignant condition: consider alternative therapy; Malignancy: reduce daily dose 10-fold and frequency to weekly instead of daily, allow 4–6 weeks to reach steady state Strong

79.2 (3)

52.8 (8)

59.0 (5)

o 10% of normal dose—reduce normal dose by 10–20 mg m − 2 daily

Abbreviations: ALL, acute lymphoblastic leukemia; CCHMC, Cincinnati Children's Hospital Medical Center; COG, Children’s Oncology Group; CPIC, Clinical Pharmacogenetics Implementation Consortium; IBD, inflammatory bowel disease; NR, not reported; TPMT, thiopurine S-methyltransferase.

Table 8.

Dosing recommendations for 6-thioguanine based on TPMT status

Identifier, year

CPIC, 2011

Target condition Normal activity (wild type) Intermediate activity (heterozygous)

None, general TPMT testing Normal dose, adjust along with other Myelosuppressive agents as needed 30–50% of target dose, adjust based on tolerance and disease-specific guidelines, allow 2–4 weeks to reach steady state Non-malignant conditions: consider alternative therapy; Malignancy: reduce daily dose 10-fold and thrice weekly, adjust dose based on tolerance and disease-specific guidelines, allow 4–6 weeks to reach steady state Strong 59.0 (5)

Low or absent activity (homozygous mutant) Reported strength of recommendation Rigor of development score (rank)

Abbreviations: CPIC, Clinical Pharmacogenetics Implementation Consortium; TPMT, thiopurine S-methyltransferase.

on studies of children and because dosing recommendations were presented in units of mg m − 2 and mg kg − 1.47 It is also important to consider that genotyping in children is associated with ethical concerns related to obtaining informed consent and also to testing of other family members.48

Validity of recommendation statements Variation in recommendations for test type and dose adjustment was observed within each disease group. For example, COG recommends reducing doses by 30–50% in patients taking 6-MP (a maintenance therapy for ALL)20 while the CPIC recommends 30–70% reduction.30 The applicability and relevance of wide dose reduction recommendations are questionable. The observed differences in recommendations across common disease categories could be a result of changes in clinical practice over time, which may or may not be driven by evidence. These differences may also reflect the perspectives of the individuals/organizations producing or endorsing the guidelines. It is known that different medical specialties have different risk–benefit perspectives.49,50 Alternatively, the differences observed could reflect variation in the evidence used to support recommendation statements or variation in the methods of guideline development. For example, when comparing 200441 and 2011 BAD27 recommendations for the treatment of dermatological conditions, the most recent guideline scored much higher in terms of rigor of development. The Pharmacogenomics Journal (2014), 493 – 502

Quality of guidance documents Although AGREE-II allowed for assessment of the methodological rigor and transparency of each document, it did not allow for appraisal of the quality of the evidence cited to support the testing recommendations. Not all of the guidance documents that scored high in terms of rigor of development were based on highquality evidence. Some of the guidance documents provided references for review articles or case-studies to support TPMT testing recommendations (for example, BSPR46) while others presented evidence that appeared to contradict recommendation statements. For example, the 2010 BSG guideline states that TPMT-deficient IBD patients may not be at the same risk as ALL patients with regard to myelotoxicity but nevertheless recommends testing for all patients.28 The 2010 BSG guideline also acknowledges that evidence to support testing is limited, and the decision to test for TPMT status is controversial. A critical appraisal of evidence linked to recommendation statements was beyond the scope of this review. It is also important to note that recommendations for TPMT testing were not the primary objective of the guidance documents included (with the exception of CPIC). However, the rigor of development domain of AGREE was applied only to TPMT recommendations. Authority of sources for recommendations Differences in recommendations between pharmacogenetic organizations such as CPIC and clinical bodies such as the BAD and COG raise the question of who should be responsible for © 2014 Macmillan Publishers Limited

Systematic review of TPMT testing HF Burnett et al

501 guiding the use of pharmacogenetic testing and whether a single authoritative source is appropriate. The tolerance for risk of adverse events may differ between clinical sub-specialties, such as oncology and dermatology, and discipline-specific guidance may be needed. However, all clinical specialties benefit from the development of high-quality practice guidelines in pharmacogenetic testing. The quality of these guidelines can be enhanced through interdisciplinary collaboration between experts in genetics, pharmacology and the clinical disciplines treating patients with thiopurines. The findings of this systematic review could catalyze collaboration between clinical and academic societies, which would promote the sharing of evidence and joint guideline development and endorsement. Such an effort would recognize differences between clinical disciplines and make explicit the reasons for divergence in guidelines, for example, based on differences in age of the target population or the acceptable level of risk. A consensus approach may also be favorable in cases where evidence is lacking. Systematic reviews of available evidence can be used to identify gaps in the literature that can help inform judgments about the value of a test in particular clinical treatment paradigms, as well as identify areas for future research. Given the growth in pharmacogenomics testing, such a model of consensus-based guideline development may be instructive for other test-and-treat applications. Uptake of TPMT testing The availability of high-quality evidence-based guidelines is a necessary first step for uptake of TPMT testing. Evidence to support the cost-effectiveness of tests for health-care systems with constrained budgets is also essential. Economic evaluations require high-quality evidence of costs and outcomes and must use data that are relevant for the jurisdiction and target population. For example, the findings of an economic evaluation of TPMT testing for children with ALL51 cannot be applied to TPMT testing in IBD. Uptake of testing is also hampered by a lack of electronic networks with which data can be stored, interpreted and shared with clinicians and patients. Guidance documents are most useful when they are accessible through point-of-care devices and when test results are readily available through centralized electronic health records. The lack of strong consensus in recommendations suggests that it may be premature to issue universal recommendations for TPMT testing across patient target populations, and testing practice may evolve more rapidly in some clinical domains compared with others. Regardless, there is a need to establish authoritative trusted sources for guidance of a technology that has the potential to span different patient populations and clinical applications. CPIC52 and others3,53 have asked that pharmacogenetic CPGs go beyond making recommendations regarding clinical utility and address optimal treatment dosing for specific genotypes. As new research deepens understanding of the genetic basis of response to therapy, increasingly detailed guidelines will be needed to clarify which genetic variants that relate to a patient should be considered with respect to a given treatment and which of them do not add critical information. There are a number of limitations to this systematic review. The AGREE-II tool is intended to be applied to CPGs and not to consensus statements, medical position statements and clinical protocols. Not all of the AGREE-II domains could be applied to the COG protocol, limiting the ability to compare the quality of this document in terms of stakeholder involvement, applicability and editorial independence. Similarly, medical position statements are often deliberately brief compared with CPGs and lack details on the process of development. However, as practitioners rely on a wide range of documents to guide practice, it is important that the quality of these documents be appraised. Another limitation is © 2014 Macmillan Publishers Limited

that AGREE-II is intended to appraise CPG documents as a whole not just a section of interest (for example, drug administration and safety sections that include recommendations related to TPMT). Many of the included documents focused on both diagnosis and treatment of the conditions of interest, but only sections that referred to TPMT testing were appraised. Guidance documents that scored high in rigor of development provided evidence to support recommendations for drug administration and/or safety monitoring following TPMT testing. It is important that quality appraisal tools retain flexibility for application to a wide range of guidance documents, including CPGs, care maps, treatment algorithms and, increasingly, electronic disease management tools. In conclusion, clinical guidance on the use of TPMT testing is required to assist health-care professionals with decisions regarding which test to order and how test results can be used to improve patient care. The present review revealed variation in recommendations for TPMT testing, reflecting a lack of clear evidence to support the clinical validity and utility of test options as well as a lack of rigor in the methods used to develop recommendation statements. Interdisciplinary collaboration between experts in the fields of genetics, pharmacology and the clinical disciplines responsible for administering the test–treatment combinations is necessary for consensus-based guideline development. Such an initiative would make differences between clinical disciplines explicit and evidence based and would ensure that rigorous evidence underlies test and treat recommendations. Systematic reviews of available evidence can be used to identify gaps in the literature that in turn can help inform judgments about the value of a test, as well as set research agendas. CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGMENTS Funding for this research was provided by a program grant from the Ontario Ministry of Health and Long-Term Care Drug Innovation Fund. The views expressed are those of the authors and not of the Ontario Ministry of Health and Long-Term Care.

REFERENCES 1 Stanek EJ, Sanders CL, Taber KA, Khalid M, Patel A, Verbrugge RR et al. Adoption of pharmacogenomic testing by US physicians: results of a nationwide survey. Clin Pharmacol Ther 2012; 91: 450–458. 2 Shah RR, Shah DR.. Personalized medicine: is it a pharmacogenetic mirage? Br J Clin Pharmacol 2012; 74: 698–721. 3 Amstutz U, Carleton BC. Pharmacogenetic testing: time for clinical practice guidelines. Clin Pharmacol Ther 2011; 89: 924–927. 4 Zielinski SL. As genetic tests move into the mainstream, challenges await for doctors and patients. J Natl Cancer Inst 2005; 97: 334–336. 5 MacDermott RP. 6-Mercaptopurine (6-MP) metabolite monitoring and TPMT testing in the treatment of inflammatory bowel disease with 6-MP or azathioprine. UpToDate. Wolters Kluwer Health: Waltham, MA, USA., 2013. 6 Evans WE. Pharmacogenetics of thiopurine S-methyltransferase and thiopurine therapy. Ther Drug Monit 2004; 26: 186–191. 7 Sahasranaman S, Howard D, Roy S. Clinical pharmacology and pharmacogenetics of thiopurines. Eur J Clin Pharmacol 2008; 64: 753–767. 8 Weinshilboum RM, Sladek SL. Mercaptopurine pharmacogenetics: monogenic inheritance of erythrocyte thiopurine methyltransferase activity. Amer J Hum Gen 1980; 32: 651–662. 9 Yates CR, Krynetski EY, Loennechen T, Fessing MY, Tai HL, Pui CH et al. Molecular diagnosis of thiopurine S-methyltransferase deficiency: genetic basis for azathioprine and mercaptopurine intolerance. Ann Intern Med 1997; 126: 608–614. 10 Baker GR, Norton PG, Flintoft V, Blais R, Brown A, Cox J et al. The Canadian Adverse Events Study: the incidence of adverse events among hospital patients in Canada. Can Med Assoc J 2004; 170: 1678–1686. 11 Lathia N, Mittmann N, DeAngelis C, Knowles S, Cheung M, Piliotis E et al. Evaluation of direct medical costs of hospitalization for febrile neutropenia. Cancer 2010; 116: 3742–748.

The Pharmacogenomics Journal (2014), 493 – 502

Systematic review of TPMT testing HF Burnett et al

502 12 Weycker D, Malin J, Edelsberg J, Glass A, Gokhale M, Oster G. Cost of neutropenic complications of chemotherapy. Ann Oncol 2008; 19: 454–460. 13 McLeod HL, Isaacs KL. Preemptive pharmacogenetic testing: insufficient data equal unsatisfactory guidance. Ann Intern Med 2011; 154: 842–U108. 14 Hanauer SB. Advances in IBD: current developments in the treatment of inflammatory bowel diseases. Gastroenterol Hepatol 2009; 5: 407–409. 15 Stocco G, Cheok MH, Crews KR, Dervieux T, French D, Pei D et al. Genetic polymorphism of inosine triphosphate pyrophosphatase is a determinant of mercaptopurine metabolism and toxicity during treatment for acute lymphoblastic leukemia. Clin Pharmacol Ther 2009; 85: 164–172. 16 Alves S, Amorim A, Ferreira F, Prata MJ. Influence of the variable number of tandem repeats located in the promoter region of the thiopurine methyltransferase gene on enzymatic activity. Clin Pharmacol Ther 2001; 70: 165–174. 17 Anglicheau D, Sanquer S, Loriot MA, Beaune P, Thervet E. Thiopurine methyltransferase activity: new conditions for reversed-phase high-performance liquid chromatographic assay without extraction and genotypic-phenotypic correlation. J Chromatogr B Analyt Technol Biomed Life Sci 2002; 773: 119–127. 18 Indjova D, Atanasova S, Shipkova M, Armstrong VW, Oellerich M, Svinarov D. Phenotypic and genotypic analysis of thiopurine s-methyltransferase polymorphism in the bulgarian population. Ther Drug Monit 2003; 25: 631–636. 19 Winter JW, Gaffney D, Shapiro D, Spooner RJ, Marinaki AM, Sanderson JD et al. Assessment of thiopurine methyltransferase enzyme activity is superior to genotype in predicting myelosuppression following azathioprine therapy in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2007; 25: 1069–1077. 20 Children’s Oncology Group (COG). AALL0232: High Risk B-precursor Acute Lymphoblastic Leukemia (ALL). A Phase III Group-Wide Study. Children’s Oncology Group (COG): Arcadia, CA, USA, 2008. 21 Stanulla M, Schaeffeler E, Flohr T, Cario G, Schrauder A, Zimmermann M et al. Thiopurine methyltransferase (TPMT) genotype and early treatment response to mercaptopurine in childhood acute lymphoblastic leukemia. JAMA 2005; 293: 1485–1489. 22 Ujiie S, Sasaki T, Mizugaki M, Ishikawa M, Hiratsuka M. Functional characterization of 23 allelic variants of thiopurine S-methyltransferase gene (TPMT*2 - *24). Pharmacogenet Genomics 2008; 18: 887–893. 23 Appraisal of Guidelines for Research and Evaluation (AGREE) Instrument. AGREE Next Steps Consortium (2009). The AGREE II Instrument. Retreived 27 March 2014, from http://www.agreetrust.org. 24 National Institute for Health and Clinical Excellence (NICE). National Institute for Health and Clinical Excellence (NICE): Crohn's Disease: Management in Adults and Children. National Health Service: London, UK, 2012. 25 Cincinnati Children's Hospital Medical Center. Evidence-Based Care Guideline for Management of Pediatric Moderate/Severe Inflammatory Bowel Disease. Cincinnati, OH, USA, 2007. 26 Chakravarty K, McDonald H, Pullar T, Taggart A, Chalmers R, Oliver S et al. BSR/ BHPR guideline for disease-modifying anti-rheumatic drug (DMARD) therapy in consultation with the British Association of Dermatologists. Rheumatology 2008; 47: 924–925. 27 Meggitt SJ, Anstey AV, Mohd Mustapa MF, Reynolds NJ, Wakelin S. British Association of Dermatologists' guidelines for the safe and effective prescribing of azathioprine 2011. Br J Dermatol 2011; 165: 711–734. 28 Mowat C, Cole A, Windsor A, Ahmad T, Arnott I, Driscoll R et al. Guidelines for the management of inflammatory bowel disease in adults. Gut 2011; 60: 571–607. 29 Manns MP, Czaja AJ, Gorham JD, Krawitt EL, Mieli-Vergani G, Vergani D et al. Diagnosis and management of autoimmune hepatitis. Hepatology 2010; 51: 2193–2213. 30 Relling MV, Gardner EE, Sandborn WJ, Schmiegelow K, Pui CH, Yee SW et al. Clinical Pharmacogenetics Implementation Consortium guidelines for thiopurine methyltransferase genotype and thiopurine dosing. Clin Pharmacol Ther 2011; 89: 387–391. 31 Menter A, Korman NJ, Elmets CA, Feldman SR, Gelfand JM, Gordon KB et al. Guidelines of care for the management of psoriasis and psoriatic arthritis: Section 4. Guidelines of care for the management and treatment of psoriasis with traditional systemic agents. J Am Acad Dermatol 2009; 61: 451–485. 32 National Comprehensive Cancer Network. NCNN Clinical Practice Guidelines in Oncology: Acute Lymphoblastic Leukemia.Version 1.2012. National Comprehensive Cancer Network: Fort Washington, PA, USA, 2012.

The Pharmacogenomics Journal (2014), 493 – 502

33 Valdes R Jr., Payne DA, Linder MW (eds) Clinical Practice Considerations. In: Laboratory Medicine Practice Guidelines: Laboratory Analysis and Application of Pharmacogenetics to Clinical Practice. The National Academy of Clinical Biochemistry: Rockville, MD, USA, 2010. 34 Sanderson J, Ansari A, Marinaki T, Duley J. Thiopurine methyltransferase: should it be measured before commencing thiopurine drug therapy? Ann Clin Biochem 2004; 41: 294–302. 35 Fabre MA, Jones DC, Bunce M, Morris PJ, Friend PJ, Welsh KI et al. The impact of thiopurine S-methyltransferase polymorphisms on azathioprine dose 1 year after renal transplantation. Transpl Int 2004; 17: 531–539. 36 Bernstein CN, Fried M, Krabshuis JH, Cohen H, Eliakim R, Fedail S et al. World Gastroenterology Organization Practice Guidelines for the Diagnosis and Management of IBD in 2010. Inflamm Bowel Dis 2010; 16: 112–124. 37 Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002; 36: 479–497. 38 Gleeson D, Heneghan MABritish Society of Gastroenterology. British Society of Gastroenterology (BSG) guidelines for management of autoimmune hepatitis. Gut 2011; 60: 1611–1629. 39 Lichtenstein GR, Abreu MT, Cohen R, Tremaine W. American Gastroenterological Association Institute medical position statement on corticosteroids, immunomodulators, and infliximab in inflammatory bowel disease. Gastroenterology 2006; 130: 935–939. 40 Sandhu BK, Fell JME, Beattie RM, Mitton SG. British Society of Paediatric Gastroenterology Hepatology and Nutrition (BSPGHAN). Guidelines for the Management of Inflammatory Bowel Disease (IBD) in Children in the United Kingdom. UK IBD Working Group, 2008. 41 Anstey AV, Wakelin S, Reynolds NJ. Guidelines for prescribing azathioprine in dermatology. Br J Dermatol 2004; 151: 1123–1132. 42 PURINETHOL® (mercaptopurine) Product Monograph. (2013). US Food and Drug Agency. Accessed 27 March 2014 http://www.accessdata.fda.gov/drugsatfda_ docs/label/2011/009053s032lbl.pdf. 43 Ma Q, Lu AYH. Pharmacogenetics, pharmacogenomics, and individualized medicine. Pharmacological Rev 2011; 63: 437–459. 44 Booth RA, Ansari MT, Loit E, Tricco AC, Weeks L, Doucette S et al. Assessment of thiopurine S-methyltransferase activity in patients prescribed thiopurines: a systematic review. Ann Intern Med 2011; 154: 814–823, W-295-818. 45 Turner D, Levine A, Escher JC, Griffiths AM, Russell RK, Dignass A et al. Management of pediatric ulcerative colitis: joint ECCO and ESPGHAN evidence-based consensus guidelines. J Pediatr Gastroenterol Nutr 2012; 55: 340–361. 46 The British Society for Paediatric and Adolescent Rheumatology. The British Society for Paediatric and Adolescent Rheumatology. Azathioprine use in paediatric rheumatology. https://www.bspar.org.uk/DocStore/FileLibrary/PDFs/BSPAR %20guidance%20for%20Azathioprine%202011.pdf. (accessed 13 August 2014). 47 Relling MV, Gardner EE, Sandborn WJ, Schmiegelow K, Pui CH, Yee SW et al. Clinical Pharmacogenetics Implementation Consortium Guidelines for Thiopurine Methyltransferase Genotype and Thiopurine Dosing: 2013 update. Clin Pharmacol Ther 2013; 93: 324–325. 48 Bunnik EM, Schermer MHN, Janssens ACJW. Personal genome testing: test characteristics to clarify the discourse on ethical, legal and societal issues. BMC Med Ethics 2011; 12: 11. 49 McDonnell J, Meijler A, Kahan JP, Bernstein SJ, Rigter H. Panellist consistency in the assessment of medical appropriateness. Health Pol 1996; 37: 139–152. 50 Kahan JP, Park RE, Leape LL, Bernstein SJ, Hilborne LH, Parker L et al. Variations by specialty in physician ratings of the appropriateness and necessity of indications for procedures. Med Care 1996; 34: 512–523. 51 Donnan JR, Ungar WJ, Mathews M, Hancock-Howard RL, Rahman P. A cost effectiveness analysis of thiopurine methyltransferase testing for guiding 6mercaptopurine dosing in children with acute lymphoblastic leukemia. Pediatr Blood Cancer 2011; 57: 231–239. 52 Relling MV, Klein TE. CPIC: Clinical Pharmacogenetics Implementation Consortium of the Pharmacogenomics Research Network. Clin Pharmacol Ther 2011; 89: 464–467. 53 Relling MV, Guchelaar HJ, Roden DM, Klein TE. Pharmacogenetics: call to action. Clin Pharmacol Ther 2011; 90: 507; author reply 507–508. 54 Ooi CJ, Fock KM, Makharia GK, Goh KL, Ling KL, Hilmi I et al. The Asia-Pacific consensus on ulcerative colitis. J Gastroenterol Hepatol 2010; 25: 453–468.

© 2014 Macmillan Publishers Limited