Nature Genetics: doi: /ng Supplementary Figure 1. Study design.

Transcription

1 Supplementary Figure 1 Study design. Leukopenia was classified as early when it occurred within the first 8 weeks of thiopurine therapy and as late when it occurred more than 8 weeks after the start of therapy. The discovery and replication analyses were initially focused on early leukopenia. The SNPs identified from the initial analyses were confirmed in patients with late leukopenia. 1

2 Supplementary Figure 2 Quantile-quantile plot for the Immunochip association P values of SNPs in 33 cases and 307 controls. log 10 P values are plotted against the expected null distribution. Black data points and red data points represent the results of 95,405 genotyped SNPs and 531,416 genotyped + imputed SNPs in the Immunochip set, respectively. 2

3 Supplementary Figure 3 Manhattan plot for the discovery Immunochip association analysis in 33 cases and 307 controls. Data represent the trend test result of 531,416 genotyped + imputed SNPs. log 10 P values are shown according to chromosomal location. SNPs within the SUCLA2-NUDT15-MED4 locus at 13q14 achieve genome-wide significance. The red horizontal line indicates SNPs at a genome-wide significance level (here P = ). The blue horizontal line indicates SNPs with P <

4 Supplementary Figure 4 Regional association plots after conditional analysis of the SUCLA2-NUDT15-MED4 locus at 13q14. (a,b) Regional association plots after conditional analysis was performed to control for the association at rs (a) and rs (b). 4

5 Supplementary Figure 5 Receiver operating characteristic curve for an additive prediction model of early leukopenia using NUDT15 SNP rs The area under the curve with 95% CI = is Overall sensitivity is 89.4%, whereas specificity is 93.2%. 5

6 Supplementary Figure 6 Sensitivity of Jurkat cells to 7.5 µm 6-MP following transfection with wild-type or mutant NUDT15. (a) NUDT15 mrna levels in Jurkat cells transfected with wild-type or mutant NUDT15, as determined by RT-PCR analysis. (b) The number of viable mutant NUDT15-transfected cells decreases compared to wild type transfected control cells. (c) Cytotoxicity assay with annexin V and propidium iodide staining showing that cells are more sensitive to 6-MP when they have been transfected with mutant NUDT15, compared to cells transfected with wild-type NUDT15. Data are presented as mean ± s.d. of three independent experiments. 6

7 Supplementary Figure 7 Principal-component analysis (PCA). (a) PCA of 340 samples and 194 reference DNA samples from HapMap. (b) Plots of first eight principal components from the principalcomponent analysis using 340 samples (33 cases, 307 controls). 7

8 Supplementary Information A common missense variant in NUDT15 confers susceptibility to thiopurine-induced leukopenia Suk-Kyun Yang, Myunghee Hong, Jiwon Baek, Hyunchul Choi, Wanting Zhao, Yusun Jung, Talin Haritunians, Byong Duk Ye, Kyung-Jo Kim, Sang Hyoung Park, Soo-Kyung Park, Dong-Hoon Yang, Marla Dubinsky, Inchul Lee, Dermot P. B. McGovern, Jianjun Liu, and Kyuyoung Song * Supplementary Note and Supplementary Tables 1-7 1

9 Contents Supplementary Note Study population... 3 Use of thiopurines... 3 Assessment of leukopenia... 4 Supplementary references... 5 Supplementary Tables Supplementary Table 1. Baseline characteristics of case patients with leukopenia and control patients without leukopenia... 6 Supplementary Table 2. Association of 10 single nucleotide polymorphisms with thiopurine-induced early leukopenia in the discovery, replication, and combined samples... 7 Supplementary Table 3. Association of 2 nonsynonymous single nucleotide polymorphisms with thiopurineinduced early leukopenia in the discovery, replication, and combined samples... 8 Supplementary Table 4. Association of NUDT15 R139C with thiopurine-induced late leukopenia in the replication samples... 9 Supplementary Table 5. Association between NUDT15 R139C genotype and thiopurine-induced leukopenia Supplementary Table 6. Prevalence of early and late leukopenia for the different NUDT15 genotypes Supplementary Table 7. Association between TPMT Y240C genotype and thiopurine-induced leukopenia

10 Supplementary Note Study population A total of 1934 patients with CD were seen at the Asan Medical Center, a tertiary university hospital in Seoul, Korea, between June 1989 and July During this period, thiopurine drugs AZA and 6-MP were started in 1476 patients, of whom 1191 provided DNA samples. After review of their medical records, however, 213 patients were excluded as they were considered to have received insufficient thiopurine therapy. Patients who did not experience leukopenia but had not received 1 mg/kg/day of AZA for 8 weeks were considered to have received insufficient therapy. Therefore, a total of 978 patients were finally analyzed in the present study (Supplementary Fig. 1). All study participants in the discovery and validation cohorts were of Korean descent. Patient demographic and clinical data including gender, age at diagnosis of CD, age at the start of thiopurines, duration of CD before thiopurine therapy, thiopurine dose, and duration of thiopurine therapy were obtained from the Asan IBD registry and information about comedication at the start of thiopurines was obtained by review of medical records (Supplementary Table 1). A United States cohort were recruited from the IBD Center at Cedars-Sinai Medical Center and were restricted to patients taking adequate doses of thiopurine (as defined above) with a normal TPMT evaluation. All IBD diagnoses were based on conventional clinical, radiologic, endoscopic, and histopathologic criteria 1. This study was approved by the Institutional Review Board of Asan Medical Center, and written informed consent was obtained from all participants. Use of thiopurines In brief, we usually started AZA at a dose of mg daily and increased the medication at 3

11 a rate of 25 mg every 2 to 4 weeks or slower to the goal dose of mg/kg/day for AZA and mg/kg/day for 6-MP as long as there were no leukopenia or other adverse events to prevent further increase of the medication. The dose of 6-MP was adjusted to AZA equivalents by multiplying the 6-MP dose by The median doses (interquartile range) of AZA in the Korean and US cohorts were 1.70 mg/kg/day ( mg/kg/day) and 2.41 mg/kg/day ( mg/kg/day), respectively. Currently we monitor blood counts to assess for any evidence of bone-marrow toxicity. Monitoring of complete blood counts (CBC) was usually performed biweekly for the first 8 weeks, monthly for the following 1 to 2 months, and then every 2 to 3 months. The interval for CBC monitoring was shortened if the dose was escalated or if there was a trend towards leukopenia. Assessment of leukopenia The medical records of each patient were retrospectively reviewed and information regarding thiopurine therapy including dose, duration of treatment, and timing and severity of leukopenia were independently assessed by two of the three physicians who were blinded to the results of genotyping. We defined leukopenia as a WBC count of less than 3000/mm 3, which corresponds with grade 2 or higher according to the US National Cancer Institute s Common Terminology Criteria for Adverse Events (NCI-CTCAE) version : grade 2 between /mm 3, grade 3 between /mm 3, and grade 4 less than 1000/mm 3. A WBC count between /mm 3 (NCI-CTCAE grade 1) was not regarded as leukopenia, but was considered borderline leukopenia. In patients with several episodes of leukopenia, the grade of leukopenia was determined by the lowest observed count of the first leukopenia episode. 4

12 Supplementary References 1. Lennard-Jones, J.E. Classification of inflammatory bowel disease. Scand. J. Gastroenterol. Suppl. 170, 2-6, discussion (1989). 2. Sandborn, W.J. A review of immune modifier therapy for inflammatory bowel disease: azathioprine, 6-mercaptopurine, cyclosporine, and methotrexate. Am. J. Gastroenterol. 91, (1996). 3. National Cancer Institute. Common Terminology Criteria for Adverse Events v , Accessed November 30, 2013, at 5

13 Supplementary Table 1. Baseline characteristics of case patients with leukopenia and control patients without leukopenia. Discovery samples Replication samples Characteristics Case patients with early leukopenia Controls without leukopenia Case patients with early leukopenia Case patients with late leukopenia Controls without leukopenia No. of subjects Male sex - no. (%) 20 (60.6) 228 (74.3) 19 (57.6) 195 (69.6) 242 (74.5) Age at diagnosis of CD - yr a 23 (13-39) 21 (12-64) 24 (12-44) 22 (10-73) 21 (9-57) Age at the start of thiopurines - yr a 25 (13-43) 24 (13-64) 28 (12-46) 25 (11-74) 24 (13-58) Duration of CD before thiopurine therapy - mo a 15 (0-176) 14 (0-192) 12 (0-173) 14 (0-192) 8 (0-162) Thiopurine dose - mg/kg/day of azathioprine a 0.96 ( ) 1.92 ( ) 0.98 ( ) 1.44 ( ) 1.95 ( ) Duration of thiopurine therapy - day a 28 (12-56) 1455 ( ) 28 (9-56) 420 ( ) 1050 ( ) Comedication at the start of thiopurines - no. (%) Mesalamine 31 (93.9) 283 (92.2) 31 (93.9) 249 (88.9) 283 (87.1) Anti-TNF agents 1 (3.0) 3 (1.0) 0 (0.0) 0 (0.0) 1 (0.3) Corticosteroids 10 (30.3) 100 (32.6) 13 (39.4) 100 (35.7) 111 (34.2) a Values are medians (range). CD, Crohn's disease. 6

14 Supplementary Table 2. Association of 10 single-nucleotide polymorphisms (SNPs) with thiopurine-induced early leukopenia in the discovery, replication, and combined samples. SNP Chr Position Alleles F_A F_U OR P value b F_A F_U OR P value b P value c OR 95% CI P BD d rs a A/C ( ) CACNA1S rs A/G ( ) FBLN2 rs G/A ( ) TPMT rs C/T ( ) (CMAHP) rs a G/A ( ) LOC rs T/C ( ) (ST3GAL1) rs a T/C ( ) NUDT15 rs a A/G ( ) MED4 rs G/A ( ) NRXN3 rs a C/T NA NA NA NA NA NA NA NA GALC Chr, chromosome; F_A, minor allele frequency in the case patients; F_U, minor allele frequency in the controls; OR, odds ratio; CI, confidence interval. a These SNPs were selected based on imputation data in the discovery set P <10-5, the numbers shown were genotype confirmed data. b P values were calculated using allelic association test. c Combined P values were calculated using the Cochran-Mantel-Haenszel test. d P BD: asymptotic P value of the Breslow-Day test for heterogeneity of the odds ratio. Discovery Replication Combined Gene 7

15 Supplementary Table 3. Association of 2 nonsynonymous single-nucleotide polymorphisms (SNPs) with thiopurine-induced early leukopenia in the discovery, replication, and combined samples. SNP Chr Position Alleles Discovery Replication Combined F_A F_U OR P value a F_A F_U OR P value a P value b OR 95% CI P BD c Gene rs A/T ( ) SUCLA2 (S199T) rs A/G ( ) NUDT15 (V18I) Chr, chromosome; F_A, minor allele frequency in the case patients; F_U, minor allele frequency in the controls; OR, odds ratio; CI, confidence interval. a P values were calculated using allelic association test. b Combined P values were calculated using the Cochran-Mantel-Haenszel test. c P BD: asymptotic P value of the Breslow-Day test for heterogeneity of the odds ratio. 8

16 Supplementary Table 4. Association of NUDT15 R139C with thiopurineinduced late leukopenia in the replication samples. Study group Replication analysis (late leukopenia) No. of samples RAF OR (95% CI) P value a Case patients ( ) Controls RAF, risk allele frequency; OR, odds ratio; CI, confidence interval. a P values were calculated using allelic association test. 9

17 Supplementary Table 5. Association between NUDT15 R139C genotype and thiopurine-induced leukopenia. Study group (no. of samples) Homozygous (TT) Genotype frequency number (percent) Heterozygous (CT) Noncarrier (CC) Homozygous odds ratio (TT vs CC) OR (95% CI) P value a Heterozygous odds ratio (CT vs CC) OR (95% CI) P value a OR (95% CI) Allelic odds ratio (T vs C) P value b Case patients (346) 14 (4.1) 133 (38.4) 199 (57.5) NA ( ) Early leukopenia (66) 14 (21.2) 45 (68.2) 7 (10.6) NA ( ) ( ) ( ) Late leukopenia (280) 0 (0) 88 (31.4) 192 (68.6) NA NA 6.28 ( ) ( ) Controls (632) 0 (0) 43 (6.8) 589 (93.2) OR, odds ratio; CI, confidence interval; NA, not applicable. a P values were calculated using the Fisher s exact test. b P values were calculated using allelic association test. 10

18 Supplementary Table 6. Prevalence of early and late leukopenia for the different NUDT15 genotypes. Genotype Number of patients a Prevalence of early leukopenia Prevalence of late leukopenia Prevalence of leukopenia TT 14 (1.4%) 100% (14/14) 0% (0/14) 100% (14/14) CT 176 (18.0%) 25.6% (45/176) 50.0% (88/176) 75.6% (133/176) CC 788 (80.6%) 0.9% (7/788) 24.4% (192/788) 25.3% (199/788) Total 978 (100%) a The number of patients carrying a particular genotype (percentage) 11

19 Supplementary Table 7. Association between TPMT Y240C genotype and thiopurine-induced leukopenia. Study group (no. of samples) Homozygous (GG) Genotype frequency number (percent) Heterozygous (GA) Noncarrier (AA) Heterozygous odds ratio (GA vs AA) OR (95% CI) P value a OR (95% CI) Allelic odds ratio (G vs A) P value b Case patients (346) 0 (0) 13 (4) 333 (96) 1.61 ( ) ( ) Early leukopenia (66) 0 (0) 8 (12) 58 (88) 5.67 ( ) ( ) Late leukopenia (280) 0 (0) 5 (2) 275 (98) 0.75 ( ) ( ) Controls (632) 0 (0) 15 (2) 617 (98) OR, odds ratio; CI, confidence interval. Homozygous odds ratio was not calculable. a P values were calculated using the Fisher s exact test. b P values were calculated using allelic association test. 12