Clinical and genetic factors associated with clinical relapse during anti-tumor necrosis factor therapy in Japanese patients with Crohn’s disease

DOI: https://doi.org/10.21203/rs.3.rs-2180536/v1

Abstract

Background: Little is known about clinical and genetic factors that predict the long-term response of anti-TNF therapy are limited in Japanese patients with Crohn’s disease (CD).

Methods: Association between clinical factors and cumulative clinical relapse-free rates were investigated in 464 patients with CD (373 anti-TNF naïve and 91 anti-TNF switch patients). A genome-wide association study (GWAS) was performed using Cox proportional hazards model. Genotype data of 5,657,947 SNPs from 275 anti-TNF naïve patients were used for GWAS.

Results: Lower serum albumin level, perianal disease, and younger age at disease onset were identified as risk factors for earlier clinical relapse in the anti-TNF naïve group (hazard ratio: HR = 1.76, 1.43, and 1.36; P = 0.00029, 0.044, and 0.045, respectively). Previous intestinal resection was associated with clinical relapse in the anti-TNF switch group (HR = 0.42; P = 0.0075). In the GWAS, rs12613485, which is located between RFX8 and MAP4K4, showed the strongest association with relapse (HR = 2.44; P = 3.42E-7). Pathway analysis indicated the association of the TGF-β signaling pathway (P = 3.06E-4).

Conclusions: We identified several reasonable clinical factors and candidate genetic factors associated with early relapse during anti-TNF treatments in Japanese CD patients.

Introduction

Crohn’s disease (CD) is a chronic inflammatory bowel disease. Patients with long-standing CD may experience relapse and develop other GI complications, such as stenosis, fistula, or perforation, which ultimately requires surgical resection 1. Repeated surgeries or hospitalizations lower a patient’s quality of life. Currently, as there is no curative treatment for CD, the key to improving long-term prognosis is to administer effective maintenance treatments after remission induction.

Anti-tumor necrosis factor (anti-TNF) therapies, such as infliximab (IFX) and adalimumab (ADA), have drastically changed the therapeutic strategy for CD due to their high efficacy for induction and maintenance treatments. Therefore, they have become key drugs in the treatment of CD.

The reported relapse rates in patients with CD who have been treated with anti-TNF have been increasing annually, indicating that both IFX and ADA have had a loss of response (LOR) 2 3 4 5. The estimated LOR rates of IFX and ADA are 13 and 20.3% per patient-year, respectively 6 7. One systematic review reported that the clinical remission rate at 12 months after switching anti-TNF agents ranged from 19 to 68% 8. LOR induces complications, such as stenosis and intestinal fistula, leading to surgeries and an increase in medical expenses. In cases of partial LOR, IFX and ADA could be optimized by increasing the dose, shortening intervals before infusions, and switching to other biologics. Thus, elucidating the predictors of early clinical relapse, including LOR during maintenance anti-TNF therapy is critical.

Although several studies have investigated clinical and genetic factors associated with clinical relapse during anti-TNF therapy in Western populations 9 10 11 12 13, there are few such studies in the Japanese population. Furthermore, clinical factors that affect clinical relapse after switching anti-TNF agents are unclear. The factors for the clinical relapse during anti-TNF therapy in Asian patients with CD need to be investigated as their clinical and genetic backgrounds differ from those of Western patients 14 15.

This study primarily aimed to identify the clinical factors that affect cumulative relapse-free survival rates in Japanese patients with CD who were treated with either IFX or ADA in the anti-TNF naïve and switch groups. Additionally, we aimed to identify the genetic factors associated with cumulative relapse-free survival rates, by performing an unbiased genome-wide association analysis (GWAS); a first in this field.

Results

Clinical factors

1. Patients’ characteristics and baseline data

The study design is summarized in Figure 1.

In the anti-TNF naïve group, the cumulative clinical relapse rates were analyzed in 373 patients; 75.9% (283 patients) in IFX-treated and 24.1% (90 patients) in ADA-treated. The baseline characteristics of patients are summarized in Table 1. The median age at initial CD diagnosis and disease duration at the start of the anti-TNF therapy were 21.0 and 6.2 years, respectively. The identified disease locations were as follows: 51 ileal, 54 colonic, and 268 ileocolonic. As for disease behavior, 152, 144, and 77 patients had inflammation, stenosis, and fistula types, respectively. Two hundred and fifty-four patients (68.1%) had perianal lesions, and 203 patients (54.4%) had a history of intestinal resection. Fifty-six patients (15.0%) were treated with concomitant thiopurine. 

In the anti-TNF switch group, the cumulative clinical relapse rates were analyzed in 91 cases (IFX to ADA, N = 80; 87.9% and ADA to IFX, N = 11; 12.1%). The baseline characteristics of patients are summarized in Table 2. The median age at CD diagnosis and disease duration at the start of the anti-TNF were 21.0 and 12.4 years, respectively. The identified disease locations were as follows: 5 ileal, 16 colonic, and 70 ileocolonic. As for disease behavior, 32, 32, and 27 patients had inflammation, stenosis, and fistula types. Seventy-four patients (81.3%) had anal lesions, and 63 patients (69.2%) had a history of intestinal resection. Twenty-two patients (24.2%) were treated with concomitant thiopurine. The reason for ADA to IFX switch was LOR, whereas the reasons for IFX to ADA switch were LOR in 45 patients (56.2 %), adverse events in 31 patients (38.8 %), and other reasons in four patients (5.0 %).

2. Overall relapse-free survival

The cumulative relapse-free survival rates of the anti-TNF naïve group were 73.5%, 52.6%, and 35.5% at 1, 3, and 5 years, respectively (Supplementary Figure 1). Among patients in the anti-TNF switch group, 75 had a clinical relapse during their disease course. The cumulative relapse-free survival rates of the anti-TNF switch group were 59.0%, 32.8%, and 19.3% at 1, 3, and 5 years, respectively. The cumulative relapse-free survival rate was significantly higher in the anti-TNF naïve group than in the anti-TNF switch group (P = 3.32E-5).

3. Factors Associated with relapse-free survival

3-1. Univariate Analysis

The association between clinical factors and relapse-free survival rate of the anti-TNF naïve group is summarized in Table 1. The cumulative relapse-free survival rates were significantly higher when the age at diagnosis was ≥ 21 years (P = 0.012), in the group without anal disease (P = 0.026), and in the group with baseline serum albumin levels ≥ 3.7 g/dL (P = 0.00022). Kaplan–Meier curves for each clinical factor are shown in Figures 2a, 2b, and 2c. 

The association between clinical factors and relapse-free survival rate of the anti-TNF switch group is summarized in Table 2. The cumulative relapse-free survival rates were significantly lower in the group treated with IFX after ADA (P = 2.95E-7), the group with inflammatory disease behavior (P = 0.045), the group without previous intestinal resection (P = 8.42E-5), and those with baseline serum albumin levels < 3.6 g/dL (P = 0.014) (Supplementary Figures 2 a, 2b, 2c and 2d).  

3-2. Multivariate Analysis

The results of the multivariate analysis of the anti-TNF naïve group are summarized in Table 1. Age at diagnosis < 21 years, anal disease, and baseline serum albumin levels < 3.7 g/dL were identified as independent risk factors for the early clinical relapse (hazard ratio: HR = 1.76, 1.43 and 1.36; P = 0.00029, 0.044 and 0.045, respectively). 

The results of the multivariate analysis of the anti-TNF switch group are summarized in Table 2. Previous intestinal resection was significantly associated with the cumulative relapse-free survival rates (HR = 0.42; P = 0.0075). 

Genetic factors

1. GWAS of clinical relapse-free rates

A total of 275 patients with CD were analyzed in the study. The independent risk factors that were identified by the univariate and multivariate analysis were used as covariates in GWAS (age at diagnosis, anal disease, and baseline serum albumin levels). 

The Manhattan plot of the GWAS is shown in Figure 3. The genomic inflation factor (lambda GC) was 1.03. Although no SNPs reached genome-wide significance, 80 SNPs were identified as candidate SNPs for relapse-free survival. These candidate SNPs were summarized into 15 loci (Table 3). The SNP with the lowest P-value was rs12613485 (HR = 2.44; P = 3.42E-7). The regional plot around rs12613485 is shown in Figure 4a. This SNP was located near Mitogen-activated protein kinase kinase kinase kinase 4 (MAP4K4) and Regulatory Factor X 8 (RFX8). Additionally, rs1029898 (HR = 1.76; P = 2.71E-7) and rs338887 (HR = 2.78; P = 6.15E-7) showed relatively strong associations to relapse (P < 1E-6). Rs1029898 is located near DnaJ Heat Shock Protein Family (Hsp40) Member B6 (DNAJB6) and rs338887 is an intronic variant of PRELI Domain-Containing Protein 2 (PRELID2). Rs338887 is associated with the expression of PRELID2 in multiple tissues with the strongest association in the colon (P = 3.0E-52). The regional plots around rs1029898 and rs338887 are shown in Figures 4b and 4c. Kaplan–Meier curves for the each SNP are shown in Supplementary Figures 3a, 3b and 3c.

Furthermore, we conducted a replication analysis of the SNPs which were reportedly associated with the therapeutic outcomes of anti-TNFα for inflammatory bowel disease (IBD) 21 22 23 24 25 (Supplementary Table 1). Only one out of the 13 candidate SNPs (rs1801274) was associated with clinical relapse in our study (P =1.96E-2).

2. Pathway analysis

Pathway analysis found nominally significant associations with several pathways (Supplementary Table 2). The top two pathways were REACTOME Signaling by transforming growth factor (TGF) β family members (P = 3.06E-4) and REACTOME Signaling by TGF β receptor complex (P = 7.80E-4). 

Discussion

This study demonstrated that serum albumin level, perianal disease, and age at disease onset are associated with clinical relapse in the anti-TNF naïve group, whereas previous intestinal resection is associated with relapse in the anti-TNF switch group. Furthermore, we identified candidate genetic factors which affect clinical relapse during anti-TNF therapy by performing GWAS; the first report in Japanese patients with CD.

Low serum albumin levels reportedly increase the clearance of IFX 26 27 28, and thus, could decrease the concentration of the anti-TNF. Furthermore, lower drug concentrations at the induction phase could lead to the generation of large amounts of anti-drug antibodies by inhibiting the immunological tolerance 29. These could explain why lower albumin levels at baseline could cause earlier clinical relapse. Several studies have reported that the incidence of LOR to anti-TNF therapy was increased in patients with CD who had perianal lesions and were first diagnosed at a young age 30 31 32.

Patients with CD in whom these independent risk factors exist require strict follow-up. Furthermore, treating modifiable factors, such as improvement of nutritional status and seton drainage for perianal abscess, prior to induction of anti-TNF could possibly improve the disease course.

The cumulative relapse-free survival rates decreased annually in the anti-TNF naïve and switch groups. The relapse-free survival rate in the switch group was significantly lower than that in the naïve group, which was consistent with previous reports 4 8 33. Several studies have shown that patients who have antibodies against IFX develop antibodies to ADA more frequently than patients who do not have antibodies against IFX 34 35. This is considered to result in poorer outcomes in the anti-TNF switch group.

Kiss et al. reported that previous surgical resection was associated with clinical remission in patients who were treated with anti-TNF 36. Intestinal resection prior to the anti-TNF switch is thought to reduce the risk of clinical relapse by resetting bowel complications, such as strictures, fistulas, and abscesses, thus decreasing the anti-TNF therapy effect. Therefore, when considering a switch in the anti-TNF therapy agent, bowel complications should be investigated and treated first. The reason for the anti-TNF switch is a predictor of response to the switched therapy; i.e., if the reason for switching anti-TNF therapy is intolerance, such as infusion reaction, the response rate is better than that in cases where the reason is LOR 33. In our study, the reason for ADA to IFX switch was LOR, whereas the reasons for the IFX to ADA switch were intolerance or other adverse events. Furthermore, the switch from ADA to IFX showed a tendency for worse clinical relapse rate in our analysis. We suspect that these differences in the switch reasons contributed to the results in our analysis.

The clinical factors associated with clinical relapse during anti-TNF therapy that were identified in our study are reasonable and consistent with previous reports.

In the GWAS study, we identified 80 candidate SNPs in 15 loci. Among these, rs12613485 showed the strongest association with relapse and is located near RFX8 and MAP4K4. Although very little of RFX8 has been known, MAP4K4 was reported to activate the expression of TNF-α and IL-1β and silencing Map4k4 in macrophages prevented lipopolysaccharide-induced lethality by inhibiting TNF-α and IL-1β production37. Furthermore, rs12613485 is reported to be associated with IL1RL1 measurement38. IL1RL1, also known as IL33R, is a member of the interleukin-1 receptor family and receptor of IL-33; IL-1 plays an important role in many inflammatory conditions including IBD. Pastorelli et al. showed that IL-33/ST2 system plays an important role in IBD and is modulated by anti-TNF therapy39. These findings suggest, that rs12613485 could be a reasonable candidate SNP to predict clinical relapse in patients with CD undergoing anti-TNF therapy.

The candidate SNP, rs338887, showed the third strongest association to relapse. This SNP is strongly associated with the expression level of PRELID2 in multiple tissues, especially in the colon. Although the effect of PRELID2 on the response to anti-TNF therapy in CD is unknown, the functional study of PRELID2 in patients with CD might be interesting. Additionally, to integrate the effects of genetic factors on relapse, we performed a pathway analysis, which indicated an association between TGF β signaling pathway and the response to anti-TNF therapy. Serum TGF β level was reported to be significantly higher in patients who do not respond to anti-TNF therapy than in those who do respond40. It is reasonable to suggest that genetic factors involved in the TGF β pathway can affect the response to anti-TNF therapy. However, considering no SNP reached a genome-wide significant level in our study and only one SNP (rs1801274) could replicate previous results, we concluded that the effect of genetic factors on the response to anti-TNF alpha therapy is limited; the effect is polygenic and not monogenic, consisting of many genetic loci which harbor weak effects.

There are several limitations in this study. First, the sample size was relatively small. Second, this was a retrospective study. Finally, we did not perform any replications to validate our GWAS results because recruiting a sufficient number of samples was difficult in a single-center study design. A replication study using an independent cohort should be performed to externally validate our results. Unconfirmed factors, such as intestinal flora and cytokine balance in the inflammatory regions, may affect the response to anti-TNF therapy; future studies are required to investigate this.

Despite these limitations, this is the first study to investigate the SNPs influencing the response to anti-TNF using GWAS in Japanese patients with CD. Furthermore, this is the first GWAS survival analysis in this field. It is well known that the genetic backgrounds of patients with CD differ among races, especially between the Asian and the Western populations. Therefore, this study is the first of its kind in a Japanese population.

In conclusion, our results identified several independent predictive factors and suggested a polygenic genetic effect on the clinical relapse of anti-TNF therapy.

Materials And Methods

1. Study design

This was a retrospective, observational cohort study conducted at a single center. The study protocol was reviewed and approved by the Tohoku University Hospital’s Ethics Committee (2018-1-138). All patients provided written informed consent. All methods in this study were performed in accordance with ethical guidelines for medical and health research involving human subjects established by the Ministry of Health, Labour and Welfare in Japan.

2. Subjects

We enrolled consecutive, self-reported, Japanese patients with CD, who had a history of treatment with either IFX (Mitsubishi-Tanabe Pharma, Tokyo, Japan) or ADA (EA pharma, Tokyo, Japan) at Tohoku University Hospital, from August 2002 to December 2020. We excluded patients who did not receive the scheduled maintenance treatment within 8 weeks due to either primary non-response or intolerance to the agents. Primary non-response was defined as a case where anti-TNF was stopped in the induction phase (first three administrations) due to no or reduced response to the agents. Intolerance was defined as a case where anti-TNF was stopped due to an adverse event.

CD was diagnosed based on endoscopic, radiological, and/or histological findings, in patients who presented with specific features as proposed by the Japanese Ministry of Health, Labor, and Welfare, such as longitudinal ulcer, a cobblestone appearance, and noncaseous epithelioid cell granuloma.

Disease locations were categorized into ileal, ileocolonic, and colonic, as determined by endoscopic or radiological findings. Patient characteristics were obtained from their respective medical records.

In total, 464 patients with CD were enrolled in this study. Patients were divided into two groups according to their anti-TNF use, the anti-TNF naïve group (N = 373) and the anti-TNF switch group (N = 91). Analysis for predictive clinical factors was performed in both groups separately. A GWAS was performed in the anti-TNF naïve group. Out of the 373 patients, genotyping data of only 344 patients were available. We further excluded 69 patients whose medical records lacked any characteristics data that were used as covariates. Finally, the GWAS included 275 cases.

3. Protocol of anti-TNFα antibody administration

Anti-TNF therapy was administered to patients with moderate to severe CD, who had an active luminal or perianal disease. There was no indication for anti-TNF therapy in patients with CD who had severe stenosis or internal fistulas; surgical treatments were performed for such complications first.

During the induction phase of IFX treatment, 5 mg/kg of the drug was administered at weeks 0, 2, and 6. Subsequently, 5 or 10 mg/kg of IFX was administered every 8 weeks as maintenance therapy. For the induction phase of ADA treatment, 160 mg of the drug was injected subcutaneously at week 0, and 80 mg at week 2. After week 4, ADA at a dose of 40 or 80 mg was administered every 2 weeks subcutaneously. Concomitant thiopurine use and elemental diet were initiated at the start of the anti-TNF therapy and were continued until an adverse event occurred.

4. Definition of clinical relapse

Clinical data of all enrolled patients were obtained from medical records. Clinical relapse was defined as the necessity of additional internal treatments (e.g., steroids, thiopurine, dose escalation, or switching the anti-TNFαagent) due to LOR, hospitalization, or surgery due to the worsening of CD. Surgery was defined as any intestinal resection attributable to CD.

5. Clinical factors investigated in this analysis

The clinical factors that were evaluated were: sex, age at diagnosis (< 21 or ≥ 21 years), duration of the disease at the start of anti-TNF therapy (< 6 or ≥ 6 years in the anti-TNFαnaïve group, < 12 or ≥ 12 years in the anti-TNFα switch group), BMI at the start of anti-TNF therapy (< 19 or ≥ 19), disease location (ileal, ileocolonic, or colonic), disease behavior (inflammation, stenosis, or fistula), presence of perianal disease (perianal fistulas and abscess, anal ulcers and stenosis), history of intestinal resection, smoking at the start of anti-TNF therapy, concomitant elemental diet (< 900 or ≥ 900 kcal/day), concomitant thiopurine use, serum albumin levels at the start of anti-TNF therapy (< 3.7 or ≥ 3.7 g/dl in the anti-TNF naïve group, < 3.6 or ≥ 3.6 g/dl in the anti-TNF switch group), and CRP levels at the start of anti-TNF therapy (< 0.6 or ≥ 0.6 mg/dl). The median values in each cohort were adopted as the cutoff values of disease duration, BMI, and serum albumin and CRP levels at the start of the anti-TNF therapy.

6. Genotyping and quality control

Genomic DNA was extracted from peripheral blood leukocytes by standard phenol-chloroform extraction precipitation using the NA1000 Automated Nucleic Acid Extraction Machine (Kurabo, Osaka, Japan) or the PAX gene DNA Kit (BD Bioscience, Franklin Lakes, NJ, USA). The genotyping for GWAS was performed using the Japonica Array V1 (Thermo Fisher, Tokyo, Japan), a single nucleotide polymorphism (SNP) array designed specifically for Japanese individuals 16. The genotype calling was conducted using the Affymetrix Power Tools (version 2.10.2.2; Thermo Fisher Scientific, Waltham, MA, USA). The quality control criteria, as recommended by Affymetrix, were a dish quality control of > 0.82 and a sample call rate of > 97%. The SNPs were categorized by cluster separation using the SNPolisher package (version 1.5.2; Thermo Fisher Scientific, Waltham, MA, USA). A subsequent analysis was conducted on 643,411 SNPs categorized as “recommended.” Identity by descent probabilities (PI_HAT) was estimated using plink software (version 1.90 b) 17, and cryptic relatives were detected by the maximum unrelated set identification (IMUS) method implemented in PRIMUS (version 1.8.0) using a minimum PI_HAT value of 0.1.As part of quality control, samples of cryptic relatives (PI_HAT > 0.5) and those with genotyping rates < 97% or call rates < 0.97 were excluded from further analysis.

The subsequent SNP and the sample quality-controlled genotype data of 643,496 SNPs from 275 cases were used for further analysis. The genomic coordinates of this data were converted from hg19 to GRCh38 using the CrossMap program 18 to match the genomic coordinates of the imputation panel.

7. Imputation

For quality control before imputation, SNPs with Hardy-Weinberg equilibrium P-value < 1E-4, call rate < 0.99, or minor allele frequency (MAF) < 0.005 were excluded, and 613,834 SNPs on autosomal chromosomes were included for further analysis. The imputation panel was an in-house constructed panel comprising the haplotypes of 5,765 individuals from diverse populations: 2,493 from the International 1000 Genomes, 820 from the Human Genome Diversity Project, 278 from the Simons Genome Diversity Project, 90 from the Korean Personal Genome Diversity Project, and 1,634 Japanese individuals. The Japanese data included the genomic data from 608 volunteers and 1,026 participants of the BioBank, Japan (National Bioscience Database Center [NBDC] human database [Accession ID: JGAS000114]). Variants that did not match alleles in the reference panel were removed using the comfort-gt program distributed with BEAGLE. Subsequently, we ran an imputation in BEAGLE 5.2 with default parameters. SNPs with call rate < 0.97, MAF < 0.05, Hardy-Weinberg equilibrium P < 1E-6, or information metric (INFO score) < 0.5 were excluded. After exclusion, the genotyped or imputed data of 5,657,947 SNPs from 275 patients with were used for the GWAS.

8. Statistical analysis

Survival analysis of the clinical relapse and associated factors were performed using the Kaplan–Meier method. The probability of clinical relapse to the clinical factors was compared between the groups using the log-rank test. Multivariate analysis of clinical relapse was performed using a Cox proportional hazards model. P-value < 0.05 was considered statistically significant. Clinical factors that were identified as significant in the log-rank test were included in the multivariate analysis. These analyses were performed using R software (version 4.1.3, http://www.r-project.org/).

GWAS was performed using a Cox proportional hazards model with the significant clinical factors of the multivariate analysis and the first 2 principal components as covariates. The R package, gwasurvivr, (version 1.12.0, https://github.com/suchestoncampbelllab/gwasurvivr) was used. SNPs with P-values < 5E-8 were considered as genome-wide significant and SNPs with P-values < 5E-6 were considered as candidates. Among these candidate SNPs, we used the “clump” procedure in PLINK (v.1.90b) 17 to summarize candidate variants into independent candidate loci considering the LD information. We used R2 > 0.1 (--clump-r2 0.1) and SNPs within 250 kb from the lead SNP (--clump-kb 250) as the LD parameters. Manhattan plots and regional association plots were generated using the Locus Zoom application 19. For the lead SNPs of each locus which showed a P-value < 1E-6, we investigated expression quantitative trait loci information using the Genotype-Tissue Expression (GTEx) project (version 8.0).

9. Pathway Analysis

The GWAS results were subsequently used for pathway analysis with MAGMA 20. MAGMA first computed the gene-level P-values using the weighted sum of the associated statistics for SNP sites in the region (25 kbp upstream and downstream), considering the local LD structures. Thereafter, a pathway-level statistical inference was performed based on a multiple linear–principal components regression model using biologically functional databases, such as Reactome, BioCarta, Wikipathways, and KEGG. Pathways with P-values < 1.81E-5 (0.05/2,756) were considered as significant and pathways with P-values < 0.05 were considered as candidates.

Declarations

Acknowledgements:

This work was supported by JSPS KAKENHI Grant Numbers JP15H04805. This work was supported (in part) by the Tohoku Medical Megabank Project (Special Account for reconstruction from the Great East Japan Earthquake). Part of computational resources were provided by the ToMMo supercomputer system. We would like to thank past and present members of the IBD group for fruitful discussions and scientific contributions. 

Author contributions:

Y. Kakuta, F.S., T.N., M. Nagasaki and Y. Kinouchi designed the study. F.S., T.N., Y. Kawai and Y. Kakuta acquired data. F.S., T.N., Y.S., R.M., H.S. and Y. Kakuta recruited patients. F.S., T.N., Y.Kakuta, Y.Kawai and M.N analyzed data. F.S., T.N., Y. Kakuta., Y.Kawai and A.M. drafted the manuscript. All authors reviewed the manuscript. 

Competing interests:

Y. Kakuta received research grants from AbbVie Inc., Mitsubishi Tanabe Pharma Corporation, EA Pharma Co. Ltd., JIMRO Co., Mochida Pharmaceutical Co., Ltd., Nippon Kayaku Co. Ltd., Daiichi Sankyo Co. Ltd, Kyowa Kirin Co. Ltd., and Janssen Pharmaceutical K.K, patent royalties from Medical & Biological Laboratories Co., Ltd., lecture fees from AbbVie Inc., Takeda Pharmaceutical Co. Ltd., Pfizer Inc., Mitsubishi Tanabe Pharma Corporation, EA Pharma Co. Ltd., JIMRO Co., Mochida Pharmaceutical Co., Ltd., Nippon Kayaku Co. Ltd., Kyorin Pharmaceutical Co. Ltd., Zeria Pharmaceutical Co. Ltd., Sandoz K.K., Life Technologies Japan Ltd., and Towa Pharmaceutical Co., Ltd., and Janssen Pharmaceutical K.K. H. Shiga received lecture fees from AbbVie Inc., Takeda Pharmaceutical Co. Ltd., Pfizer Inc., Mitsubishi Tanabe Pharma Corp., EA Pharma Co. Ltd., and Janssen Pharmaceutical K.K. M. Nagasaki received research grants from Toshiba Corporation. A. Masamune received research grants from Takeda Pharmaceutical Co. Ltd., AbbVie Inc., Mitsubishi Tanabe Pharma Corporation, EA pharma Co. Ltd., JIMRO Co. Ltd., Mochida Pharmaceutical Co. Ltd., and Zeria Pharmaceutical Co. Ltd., lecture fees from AbbVie Inc., Takeda Pharmaceutical Co. Ltd., and EA Pharma Co. Ltd. 

Data availability statement:

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. 

Ethics declarations:

Patient consent and IRB approval: The study protocol was reviewed and approved by the Tohoku University Hospital’s Ethics Committee (2018-1-138). All patients provided written informed consent.

References

  1. Baumgart, DC., Sandborn, WJ. Crohn's disease. Lancet 2012;380:1590-605.
  2. Sprakes, MB., Ford, AC., Warren, L., Greer, D., Hamlin, J. Efficacy, tolerability, and predictors of response to infliximab therapy for Crohn's disease: a large single centre experience. J Crohns Colitis 2012;6:143-53.
  3. Orlando, A. et al. Six year adalimumab efficacy in steroid-dependent Crohn's disease patients: A prospective single-center real life study. Dig Liver Dis 2016;48:1314-7.
  4. Moroi, R. et al. Long-term prognosis of Japanese patients with biologic-naïve Crohn's disease treated with anti-tumor necrosis factor-α antibodies. Intest Res 2019;17:94-106.
  5. Moroi, R. et al. Long-Term Prognosis of Japanese Patients with Crohn's Disease Treated by Switching Anti-Tumor Necrosis Factor-α Antibodies. Inflamm Intest Dis 2020;5:11-9.
  6. Gisbert, JP., Panés, J. Loss of response and requirement of infliximab dose intensification in Crohn's disease: a review. Am J Gastroenterol 2009;104:760-7.
  7. Billioud, V., Sandborn, W.J., Peyrin-Biroulet, L. Loss of response and need for adalimumab dose intensification in Crohn's disease: a systematic review. Am J Gastroenterol 2011;106:674-84.
  8. Ma, C. et al. Systematic review: the short-term and long-term efficacy of adalimumab following discontinuation of infliximab. Aliment Pharmacol Ther 2009;30:977-86.
  9. Visuri, I. et al. Predictors of drug survival: A cohort study comparing anti-tumour necrosis factor agents using the Swedish inflammatory bowel disease quality register. Aliment Pharmacol Ther 2021;54:931-43.
  10. Blesl, A. et al. Limited long-term treatment persistence of first anti-TNF therapy in 538 patients with inflammatory bowel diseases: a 20-year real-world study. Aliment Pharmacol Ther 2021;54:667-77.
  11. Mascheretti, S. et al. Response to infliximab treatment in Crohn's disease is not associated with mutations in the CARD15 (NOD2) gene: an analysis in 534 patients from two multicenter, prospective GCP-level trials. Pharmacogenetics 2002;12:509-15.
  12. Koder, S. et al. Genetic polymorphism in ATG16L1 gene influences the response to adalimumab in Crohn's disease patients. Pharmacogenomics 2015;16:191-204.
  13. Prieto-Pérez, R., Almoguera, B., Cabaleiro, T., Hakonarson, H., Abad-Santos, F. Association between Genetic Polymorphisms and Response to Anti-TNFs in Patients with Inflammatory Bowel Disease. Int J Mol Sci 2016;17:225.
  14. Fuyuno, Y. et al. Genetic characteristics of inflammatory bowel disease in a Japanese population. J Gastroenterol 2016;51:672-81.
  15. Yang SK. Personalizing IBD Therapy: The Asian Perspective. Dig Dis 2016;34:165-74.
  16. Kawai, Y. et al. Japonica array: improved genotype imputation by designing a population-specific SNP array with 1070 Japanese individuals. J Hum Genet 2015;60:581-7.
  17. Chang, CC. et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 2015;4:7.
  18. Zhao, H. et al. CrossMap: a versatile tool for coordinate conversion between genome assemblies. Bioinformatics 2014;30:1006-7.
  19. Pruim, RJ. et al. LocusZoom: regional visualization of genome-wide association scan results. Bioinformatics 2010;26:2336-7.
  20. de Leeuw, C.A., Mooij, J.M., Heskes, T., Posthuma, D. MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput Biol 2015;11:e1004219.
  21. Matsukura, H., Ikeda, S., Yoshimura, N., Takazoe, M., Muramatsu, M. Genetic polymorphisms of tumour necrosis factor receptor superfamily 1A and 1B affect responses to infliximab in Japanese patients with Crohn's disease. Aliment Pharmacol Ther 2008;27:765-70.
  22. Urabe, S. et al. Genetic Polymorphisms of IL-17F and TRAF3IP2 Could Be Predictive Factors of the Long-Term Effect of Infliximab against Crohn's Disease. Biomed Res Int 2015;2015:416838.
  23. Kimura, K., Takayanagi, R., Yokoyama, H., Yamada, Y. Effects of tumor necrosis factor α-857C/T polymorphism on the expression of tumor necrosis factor α. Apmis 2016;124:669-74.
  24. López-Hernández, R. et al. Genetic polymorphisms of tumour necrosis factor alpha (TNF-α) promoter gene and response to TNF-α inhibitors in Spanish patients with inflammatory bowel disease. Int J Immunogenet 2014;41:63-8.
  25. Steenholdt, C. et al. Genetic polymorphisms of tumour necrosis factor receptor superfamily 1b and fas ligand are associated with clinical efficacy and/or acute severe infusion reactions to infliximab in Crohn's disease. Aliment Pharmacol Ther 2012;36:650-9.
  26. Ordás, I., Mould, D.R., Feagan, B.G., Sandborn, W.J. Anti-TNF monoclonal antibodies in inflammatory bowel disease: pharmacokinetics-based dosing paradigms. Clin Pharmacol Ther 2012;91:635-46.
  27. Fasanmade, A.A., Adedokun, O.J., Blank, M., Zhou, H., Davis, H.M. Pharmacokinetic properties of infliximab in children and adults with Crohn's disease: a retrospective analysis of data from 2 phase III clinical trials. Clin Ther 2011;33:946-64.
  28. Xiong, Y. et al. Real-World Infliximab Pharmacokinetic Study Informs an Electronic Health Record-Embedded Dashboard to Guide Precision Dosing in Children with Crohn's Disease. Clin Pharmacol Ther 2021;109:1639-47.
  29. van Schouwenburg, P.A., Rispens, T., Wolbink, GJ. Immunogenicity of anti-TNF biologic therapies for rheumatoid arthritis. Nat Rev Rheumatol 2013;9:164-72.
  30. Zhang, Q.W., Shen, J., Zheng, Q., Ran, Z.H. Loss of response to scheduled infliximab therapy for Crohn's disease in adults: A systematic review and meta-analysis. J Dig Dis 2019;20:65-72.
  31. Hlavaty, T. et al. Polymorphisms in apoptosis genes predict response to infliximab therapy in luminal and fistulizing Crohn's disease. Aliment Pharmacol Ther 2005;22:613-26.
  32. Papamichael, K. et al. Long-term outcome of patients with Crohn's disease who discontinued infliximab therapy upon clinical remission. Clin Gastroenterol Hepatol 2015;13:1103-10.
  33. Gisbert, JP., Marín, A.C., McNicholl, A.G., Chaparro, M. Systematic review with meta-analysis: the efficacy of a second anti-TNF in patients with inflammatory bowel disease whose previous anti-TNF treatment has failed. Aliment Pharmacol Ther 2015;41:613-23.
  34. Frederiksen, M.T. et al. Antibodies against infliximab are associated with de novo development of antibodies to adalimumab and therapeutic failure in infliximab-to-adalimumab switchers with IBD. Inflamm Bowel Dis 2014;20:1714-21.
  35. Bartelds, G.M. et al. Anti-infliximab and anti-adalimumab antibodies in relation to response to adalimumab in infliximab switchers and anti-tumour necrosis factor naive patients: a cohort study. Ann Rheum Dis 2010;69:817-21.
  36. Kiss, L.S. et al. Early clinical remission and normalisation of CRP are the strongest predictors of efficacy, mucosal healing and dose escalation during the first year of adalimumab therapy in Crohn's disease. Aliment Pharmacol Ther 2011;34:911-22.
  37. Aouadi, M. et al. Orally delivered siRNA targeting macrophage Map4k4 suppresses systemic inflammation. Nature 2009;458:1180-4.
  38. Pietzner, M. et al. Genetic architecture of host proteins involved in SARS-CoV-2 infection. Nat Commun 2020;11:6397.
  39. Pastorelli, L. et al. Epithelial-derived IL-33 and its receptor ST2 are dysregulated in ulcerative colitis and in experimental Th1/Th2 driven enteritis. Proc Natl Acad Sci U S A 2010;107:8017-22.
  40. Di Sabatino, A. et al. Peripheral regulatory T cells and serum transforming growth factor-β: relationship with clinical response to infliximab in Crohn's disease. Inflamm Bowel Dis 2010;16:1891-7.

Tables

Table 1. Baseline characteristics and associations of cumulative relapse-free survival rates of Japanese CD patients in anti-TNF therapy naïve group.

Characteristic

N

P-value (univariate)

P-value (multivariate)

HR (95％ CI)

Anti-TNF therapy               Infliximab (%)

283 (75.9)

0.135 

　

　

Adalimumab (%)

90 (24.1)

　

　

　

Sex (Female, %)

98 (26.3)

0.290 

　

　

Age at diagnosis (years; Median (interquartile range))

21 (18-27)

　

　

　

 ＜21 years (%)

166 (44.5)

0.012 

0.045 

1.36 (1.01-1.85)

BMI at the start of the biologics (Median (interquartile range))

19.3 (17.7-21.0)

　

　

　

 <19 (%)

140 (37.5)

0.755 

　

　

Disease duration at the start of Anti-TNF therapy (years; Median (interquartile range))

6.2 (1.4-13.1)

　

　

　

＜6 years (%)

184 (49.3)

0.771 

　

　

Disease location                     ileal (%)

51 (13.7)

0.478 

　

　

colonic (%)

54 (14.5)

　

　

　

ileocolonic (%)

268 (71.8)

　

　

　

Disease behavior             inflammation (%)

152 (40.8)

0.107 

　

　

stenosis (%)

144 (38.6)

　

　

　

fistula (%)

77 (20.6)

　

　

　

Perianal disease (Yes, %)

254 (68.1)

0.026 

0.044 

1.43 (1.01-2.01)

Previous intestinal resection (Yes, %)

203 (54.4)

0.242 

　

　

Smoking at the start of Anti-TNF therapy (Yes, %)

78 (20.9)

0.312 

　

　

Concomitant elemental diet (Yes, %)

47 (12.6)

0.542 

　

　

Concomitant thiopurine (Yes, %)

56 (15.0)

0.575 

　

　

Serum albumin levels at the baseline (g/dl; Median (interquartile range))

3.7 (3.2-4.0)

　

　

　

 ＜3.7 g/dl (%)

157 (42.1)

0.00022 

0.00029 

1.76 (1.30-2.39)

CRP levels at the baseline (mg/dl; Median (interquartile range))

0.6 (0.2-2.0)

　

　

　

＜0.6 mg/dl (%)

148 (39.7)

0.067 

　

　

Abbreviations: BMI, body mass index; Anti-TNF, anti-tumor necrosis factor; CRP, C-reactive protein; CD, Crohn’s disease; HR, hazard ratio; CI, confidence interval

Table 2. Baseline characteristics and associations of cumulative relapse-free survival rates of Japanese CD patients in the anti-TNF therapy switch group.

Characteristic

N

P-value (univariate)

P-value (multivariate)

HR (95％ CI)

Anti-TNF therapy    adalimumab→infliximab (%)

80 (87.9)

0.00000030 

0.078 

0.42 (0.16-1.10)

infliximab→adalimumab (%)

11 (12.1)

　

　

　

Sex (female; %)

34 (37.4)

0.108 

　

　

Age at diagnosis (years; Median (interquartile range))

21 (17-24)

　

　

　

 ＜21 years (%)

41 (45.1)

0.583 

　

　

BMI at the switch of the biologics (Median (interquartile range))

19.5 (18.3-21.1)

　

　

　

 <19 (%)

43 (47.3)

0.776 

　

　

Disease duration at the switch of the biologics (years; Median (interquartile range))

12.4 (5.5-19)

　

　

　

＜12 years (%)

44 (48.4)

0.838 

　

　

Disease location                          ileal (%)

5 (5.5)

0.337 

　

　

colonic (%)

16 (17.6)

　

　

　

ileocolonic (%)

70 (76.9)

　

　

　

Disease behavior                inflammation (%)

32 (35.2)

0.013 

　

　

stenosis (%)

32 (35.2)

　

0.650 

0.84 (0.39-1.80)

fistula (%)

27 (29.7)

　

0.718 

0.87 (0.41-1.84)

Perianal disease (Yes, %)

74 (81.3)

0.547 

　

　

Previous intestinal resection (Yes; %)

63 (69.2)

0.000084 

0.0075 

0.42 (0.22-0.79)

Smoking at the switch of Anti-TNF therapy (Yes, %)

20 (22)

0.242 

　

　

Concomitant elemental diet (Yes, %)

14 (15.4)

0.964 

　

　

Concomitant thiopurine (Yes; %)

22 (24.2)

0.564 

　

　

Serum albumin levels at the baseline (g/dl; Median (interquartile range))

3.6 (3.2-3.9)

　

　

　

 ＜3.6 g/dl (%)

35 (38.5)

0.014 

0.070 

1.70 (0.96-3.02)

CRP levels at the baseline (mg/dl; Median (interquartile range))

0.6 (0.2-1.9)

　

　

　

＜0.6 mg/dl (%)

34 (37.4)

0.993 

　

　

Abbreviations: BMI, body mass index; Anti-TNF, anti-tumor necrosis factor; CRP, C-reactive protein; CD, Crohn’s disease; HR, hazard ratio; CI, confidence interval

Table 3. Candidate SNPs and genes associated with cumulative relapse-free rates in Japanese CD anti-TNF therapy naïve group using GWAS (P ≤ 5E-6)

rsID

CHR

Position

A1

A2

AF_A2

P-value

HR (95％ CI)

Location

Near Genes

Distance from Genes

rs12613485

2

101538302

G

A

0.1382

1.39E-07

2.44 (1.75-3.40)

intergenic

RFX8; MAP4K4

dist=63190; dist=159401

rs1029898

7

157420907

C

T

0.3400

2.71E-07

1.77 (1.42-2.19)

intergenic

DNAJB6; LOC101927914

dist=3468; dist=45324

rs338887

5

145764856

T

G

0.0636

6.15E-07

2.77 (1.86-4.14)

intronic

PRELID2

.

rs12050455

14

19961274

C

T

0.1145

1.15E-06

2.32 (1.65-3.26)

intergenic

OR4K1; OR4K15

dist=24591; dist=14245

rs78078728

1

103115062

C

T

0.0909

1.32E-06

2.44 (1.70-3.51)

intergenic

COL11A1; LOC101928436

dist=6540; dist=301918

rs72321632

12

24237521

TTAATTTG

T

0.0527

1.69E-06

2.66 (1.78-3.97)

ncRNA_intronic

SOX5-AS1

.

rs562708696

2

64593617

TA

T

0.1927

1.85E-06

1.98 (1.49-2.62)

downstream

AFTPH

dist=613

rs7607317

2

45817500

T

C

0.1236

2.90E-06

2.19 (1.58-3.04)

intronic

PRKCE

.

rs11305762

12

102424239

CT

C

0.1436

3.15E-06

2.15 (1.56-2.97)

intronic

IGF1

.

rs5819105

17

6304482

T

TCA

0.3636

3.18E-06

1.75 (1.38-2.21)

intergenic

WSCD1; AIPL1

dist=180055; dist=119256

rs10938700

4

8247216

C

T

0.0800

3.22E-06

2.64 (1.75-3.97)

intergenic

SH3TC1; HTRA3

dist=6113; dist=22538

rs12354999

10

70263564

C

T

0.2800

3.23E-06

0.53 (0.40-0.69)

intronic

NPFFR1

.

rs17542397

4

23502505

T

G

0.1036

3.65E-06

2.44 (1.67-3.55)

intergenic

GBA3; PPARGC1A

dist=682933; dist=289516

rs55646402

10

48661955

T

C

0.0545

4.76E-06

2.80 (1.80-4.36)

intergenic

ARHGAP22; WDFY4

dist=5690; dist=22918

rs16962552

16

60767998

A

G

0.2655

4.85E-06

1.87 (1.43-2.45)

intergenic

LINC02141; MIR4426

dist=714027; dist=287709

 

Abbreviations: SNP, single-nucleotide polymorphism; CD, Crohn’s disease; GWAS, genome-wide association study; CHR, chromosome number; AF_A2, minor allele frequency of A2 allele in our cohort; HR, hazard ratio; CI, confidence interval

Positions are based on the Genome Reference Consortium human build 38 (GRCh38).