Genetic and clinical markers for predicting treatment responsiveness in rheumatoid arthritis

Abstract

Although many drugs and therapeutic strategies have been developed for rheumatoid arthritis (RA) treatment, numerous patients with RA fail to respond to currently available agents. In this review, we provide an overview of the complexity of this autoimmune disease by showing the rapidly increasing number of genes associated with RA. We then systematically review various factors that have a predictive value (predictors) for the response to different drugs in RA treatment, especially recent advances. These predictors include but are certainly not limited to genetic variations, clinical factors, and demographic factors. However, no clinical application is currently available. This review also describes the challenges in treating patients with RA and the need for personalized medicine. At the end of this review, we discuss possible strategies to enhance the prediction of drug responsiveness in patients with RA.

Keywords

rheumatoid arthritis / gene / clinical markers / therapy

Introduction

Rheumatoid arthritis (RA) is a common immune-mediated disease characterized by chronic inflammatory arthritis and synovitis that destroy the bone and the cartilage. In the United States, the incidence of RA is approximately 1.3 million adults [¹,²]. Patients with RA may also suffer from joint damage, functional impairment, and decreased quality of life.

Insufficient evidence has been published to determine the overall RA pathogenesis. However, an increasing number of studies have demonstrated the involvement of genetics, environmental cues, and autoimmune system dysfunction in the mechanism of RA and provided emerging new targets for treatment. Biologics, such as antitumor necrosis factor (TNF) agents, the B-cell depleting agent rituximab (RTX), the selective co-stimulation modulator abatacept, and the anti-IL-6 receptor monoclonal antibody tocilizumab, have been effective treatments for RA [³–⁶]. These drugs can be used individually or in combination with disease-modifying anti-rheumatic drugs (DMARDs), such as methotrexate, hydroxychloroquine, leflunomide, and sulfasalazine [⁷]. However, some patients still fail to respond to treatment regardless of therapeutic options administered to patients with RA. As such, further studies should focus on how to identify these patients before therapy and how to predict treatment responses in patients with RA.

In this review, we provide an overview of the genes implicated in RA by mining MEDLINE/PubMed records. Despite many decades of research in the field of molecular biology and medicine, the number of genes implicated in RA is increasing rapidly, indicating the underlying complexity of this autoimmune disease. We then systematically review various factors, including genetics, clinical characteristics, and demographics, which have a predictive value for determining the potential response to different drugs in RA treatment. These results highlight the challenges in treating patients with RA and the need for personalized medicine. Lastly, we discuss possible strategies to predict drug responsiveness during RA treatment effectively.

Overview of genes implicated in the pathogenesis of RA

To provide an overview of previous RA research, we employed a text-mining technique to search MEDLINE/PubMed abstracts for genes associated with RA and used keywords, such as “associated” and “implicated,” to identify studies that examined genes that were differentially expressed in patients versus healthy controls or in patients who responded well to treatment versus those who did not. This approach allowed us to identify 3500 associations between 610 human genes and RA (Supplementary Table 1). Grouping the results by the year of publication, we found that the number of relevant publications per year has increased dramatically since the 1980s, and the number of implicated genes has accumulated (Fig. 1). These results, especially the increasing number of associated genes, suggest that RA pathogenesis has yet to be fully elucidated, and this disease may be even more complicated than it currently appears.

To provide an overview of the broad functional categories of the 610 RA-associated genes, we categorized the genes into gene families based on their annotated InterPro [⁸] functional domains (using the data from Ensembl BioMart [⁹]). As expected, immunoglobulin (Ig)-like genes are the most abundant gene family in our gene set; as shown below, 5 of the top 10 largest gene families contain genes that encode Ig-domain-containing proteins and cytokines. The 5 other families are all related to various kinases (Table 1). Kinases often participate in signal transduction cascades, regulate a large number of downstream proteins, and play important roles in RA pathogenesis [¹⁰,¹¹]. The large number of kinases associated with RA further emphasizes the complexity of this autoimmune disease.

Consistent with these observations, our results revealed that the top 10 most studied genes, ranked by the number of MEDLINE/PubMed records in which these genes were mentioned, include mostly Igs, cytokines, and kinases (Table 2).

Genetic variants as predictors of responsiveness to RA treatment

TNF has been extensively studied in RA [¹²] because it is a cytokine produced by the immune system and responsible for inflammation. TNF expression is inhibited in healthy individuals. Conversely, in patients with RA or similar autoimmune conditions, the TNF expression in the blood often increases, leading to more inflammation [¹³]. Thus, anti-TNF agents or TNF-blocking drugs have been widely used to treat inflammatory conditions, including RA [³,⁴]. The commonly used anti-TNF drugs are infliximab, etanercept, and adalimumab [⁵,⁶]. These drugs can be used individually or in combination with DMARDs, such as methotrexate, hydroxychloroquine, leflunomide, and sulfasalazine [⁷].

The combination of methotrexate and TNF-blocking agents is commonly prescribed for RA treatment [¹⁴] and has demonstrated positive results in most clinical trials. However, up to 45% of patients are slightly responsive or nonresponsive to methotrexate [¹⁵], suggesting that treatment responses vary among individuals. However, other studies have shown that this particular combination may increase RA remission rates [¹⁶].

Genetic polymorphisms, such as single nucleotide polymorphisms (SNPs), copy number variations, and insertion/deletions (indels), have been implicated in RA treatment responsiveness. Table 3 provides a shortlist of genes, their associated genetic variations, and effects on RA treatment.

TNF polymorphisms and their effects on responsiveness to DMARDs and TNF-blocking agents have been extensively studied because of the central role of TNF in RA treatment [³,⁴,¹³]. Many studies have focused on SNPs in the promoter region at positions −238 and −308 of the TNF gene. The −238 promoter polymorphism has been associated with RA severity, unresponsiveness to methotrexate [³³], and treatment response to infliximab [³⁴]. Conversely, the promoter SNP at −308 is not associated with treatment responses to infliximab [³⁴] or etanercept [³²]. However, SNP −308 has been linked to treatment responses to adalimumab [³¹]. These promoter SNPs contribute to treatment responses by presumably altering TNF gene expression [³¹]. An SNP at position 489 of the coding region has also been studied, but no remarkable contribution to methotrexate responsiveness has been revealed [³³].

TNFRSF1B, a member of the TNF family, has been associated with the treatment responses to infliximab [³⁵]. A previous study in a French population showed that the M196R SNP results in decreased responses to infliximab therapy, although this SNP and another one investigated in the same study are not related to RA susceptibility [³⁵].

Polymorphisms in other cytokines have been associated with responses to DMARDs and TNF-blocking agents. For example, promoter microsatellite polymorphisms in the IL10 gene can be used to predict clinical responses to etanercept [²⁵].

Many Ig genes are implicated in RA. A major histocompatibility complex, class I, G (HLA-G) gene in the Ig family can be induced by methotrexate activation [²¹]; thus, its polymorphisms may influence the efficacy of methotrexate treatment. Indeed, a 14 bp indel polymorphism in exon 8 at the 3′ UTR in HLA-G is clinically advantageous in methotrexate responsiveness [²²]. However, controversial results have been observed in a follow-up study [²³]. In addition, polymorphisms in Ig receptor genes may affect their interactions with Ig genes, which may influence the efficacy of Ig-based therapies [¹⁹]. For example, Cañete et al. reported the association of several functional SNPs in two Fc γ receptor (FcγR) genes (FCGR2A and FCGR3A) and responses of patients with RA to infliximab [¹⁹]. Interestingly, treatment responses are remarkably associated with polymorphisms in the two genes at different timepoints, highlighting the dynamic nature of the interactions between FcγR and Ig [¹⁹]. Another study has confirmed the association between FCGR3A and the response to infliximab [²⁰].

The effectiveness of methotrexate is largely due to its dihydrofolate reductase inhibitor activity. Its metabolites also inhibit other folate enzymes, including 5,10-methylenetetrahydrofolate reductase (MTHFR) [³⁶]. Therefore, variations in MTHFR may influence the effectiveness of methotrexate treatment. Several studies have confirmed this association and shown that the C667T SNP in the coding region of MTHFR is related to an increased methotrexate toxicity in Spanish [²⁸], Korean [²⁷], African-American, and Caucasian individuals [²⁶]. Ethnic differences influence the response to methotrexate treatment [²⁶]. Conversely, in a study involving Japanese subjects, C667T SNP does not remarkably contribute to methotrexate responses [²⁹]. The A1298C SNP has also been studied [²⁶,²⁹] and found to be associated with methotrexate treatment responses in African-American and Caucasian individuals [²⁶] but not in Korean individuals [²⁷]. Polymorphisms in the 3′ UTR of thymidylate synthetase, another member of the folate pathway, are also associated with responses to methotrexate [²⁷].

Genes with protein products that can directly interact with administered drugs, such as those involved in drug resistance or metabolism, can influence treatment efficacy. For example, ATP-binding cassette, subfamily B (MDR/TAP), member 1 (ABCB1) is a multi-drug resistance gene that encodes an ATP-dependent drug efflux pump. A C3435T polymorphism in ABCB1 affects methotrexate sensitivity in patients with RA, and patients with the 3435TT allele are significantly more nonresponsive than those with the 3435CC allele [³⁰]. Leflunomide is also a commonly used DMARD, and its metabolism involves genes in cytochrome P450, mainly CYP1A2 and CYP2C19 [¹⁸,³⁷]. Polymorphisms in both genes are associated with the response to leflunomide [¹⁷,¹⁸,³⁷]; in particular, certain CYP2C19 genotypes can hasten leflunomide degradation [¹⁸].

In addition to anti-TNF drugs, many other drugs targeting the interleukin family, a group of cytokines that play central roles in the regulation of immune and inflammatory responses, and genes important in immune cell development have been used to treat RA. For example, RTX is a chimeric monoclonal antibody against CD20, which is a B-cell surface protein, and has been used to treat RA [³⁸]. A promoter polymorphism at −174 in an IL6 gene is associated with the response to RTX treatment, but patients with the IL6 −174CC genotype often show no response to this drug [²⁴]. Nevertheless, associations between genes and relatively new non-TNF blocking drugs are seldom reported. The latter have been proven to be useful in treating RA, and many researchers proposed cocktail therapies for RA by using combinations of anti-TNF drugs and other cytokine antagonists [³⁹].

In addition to genetic factors, epigenetic factors, such as DNA methylation and histone modifications, can be used to predict responses to drugs in RA treatment. Plant et al. summarized this topic in a recent review [¹⁴]. Considering the technical complexity in identifying epigenetic modifications and the recently revealed links between epigenetic factors and drug responses during RA treatment, we will explore this topic in our future studies and request our readers to consult the review by Plant et al. and the references therein.

Differential gene expression as predictors of responsiveness in RA treatment

Many genetic variations affect gene expression. Thus, determining differentially expressed genes by comparing between patients with RA and healthy controls or among patients with different responses to RA treatment may provide useful information for identifying the predictors of drug responsiveness. Gene expression is often measured at a transcriptome level by using microarray or quantitative real-time polymerase chain reaction (qPCR) or at a proteome level from blood samples by conducting enzyme-linked immunosorbent assay (ELISA) and comparative serum proteome analyses.

The TNF −308 G/G promoter polymorphism contributes to adalimumab responsiveness by altering serum TNF levels; the median serum TNFα level of the G/G group was remarkably higher than that of the G/A group [³¹]. In addition to TNF expression levels, the expression levels of other cytokines, such as interferon (IFN) and IFN receptor genes, have been extensively studied and linked to RA treatment responsiveness. For example, using the data from whole-blood qPCR analysis, van Baarsen et al. [⁴⁰] found that a high IFN signature, that is, an abundant expression of genes activated by type I IFNs, is associated with poor responses in infliximab-treated patients with RA. de Jong et al. [⁴¹] reported that the use of prednisone, a synthetic immunosuppressant drug, suppresses the IFN signature and increases the responsiveness to RTX in patients with RA. The use of the IFN signature as a predictor of drug response has also been reviewed [⁴²].

The use of drugs in RA treatment can lead to immunogenicity, which can lead to the development of antidrug antibodies (ADAbs) and low effective drug levels. Jani et al. recently reported that patients who have developed ADAbs and low drug levels after 3 months of adalimumab treatment respond poorly to continued treatment [⁴³]. Thus, early responses to a particular drug can predict the long-term outcomes of patients administered with the same drug or possibly other drugs. Similar results have been observed in Japanese patients with RA [⁴⁴].

By applying comparative serum proteome analyses in TNF-antagonist responders and nonresponders, Blaschke et al. [⁴⁵] identified four genes, namely, haptoglobin-a1 (Hp-a1), -a2 (Hp-a2), vitamin D-binding protein (VDBP) and apolipoprotein C-III (ApoC-III), as predictors of etanercept responses in RA treatment. After 6 months of etanercept treatment, the first three genes (Hp-a1, Hp-a2, and VDBP) are remarkably more upregulated in the serum samples of the responders than that of the nonresponders. By contrast, ApoC-III is considerably downregulated. Therefore, high-throughput techniques can accelerate the discovery of genes essential for the clinical outcomes in RA treatment.

Clinical and demographic factors as predictors of responsiveness in RA treatment

Clinical factors, such as disease histories, previous use of certain drugs, and even RA severity, can serve as predictors of drug responsiveness in RA treatment. For example, Atzeni et al. [⁴⁶]studied 1300 patients with various levels of RA disease severity. The authors evaluated the associations between several clinical factors and the response to anti-TNF agents and found that the enhanced responses of patients with severe RA are associated with no previous use of corticosteroids (an immunosuppressant), the use of adalimumab versus infliximab (both anti-TNF drugs), and the absence of comorbidities [⁴⁶]. However, these associations are unremarkable in patients with moderate RA disease severity [⁴⁶], indicating that disease severity itself can be a predictor of drug responsiveness in RA treatment. Gender and age at the start of treatment are also associated with responsiveness to anti-TNF therapy; in particular, males have been higher RA remission than females [⁴⁷]. However, these results partly contradicted those of a study of Swedish patients with RA in which gender does not influence treatment responses to TNF-blocking therapy [⁴⁸]. These discrepancies can be attributed to the different populations investigated.

The presence of rheumatoid factor (RF) and anticyclic citrullinated peptide (anti-CCP) is a characteristic of RA [⁴⁹,⁵⁰]. Thus, the serum levels of these molecules in patients have been extensively studied and linked to drug responsiveness. Narváez et al. [⁵¹] investigated 108 Spanish patients with RA and found that patients with high levels of serum anti-CCP and/or RF achieve enhanced responses to RTX.

The serum levels of proteins/antibodies and the number of circulating cells can be used as predictors of drug responsiveness. Chara et al. [⁵²] measured the number of circulating monocytes and CD14(+high)CD16(−), CD14(+high)CD16(+), and CD14(+low)CD16(+) subsets in treatment-naive patients with RA through flow cytometry. They found that patients who are nonresponsive to methotrexate treatment have a higher number of circulating monocytes and the subsets than those of responsive patients, who showed normal number of these cells [⁵²]. Their study suggests a new strategy for predicting the clinical response to methotrexate in treatment-naive patients.

Lifestyle activities, such as smoking and drinking, have been associated with several autoimmune diseases, including RA [⁵³–⁵⁵]. Smoking has also been linked to drug responsiveness in RA treatment. For example, a study on British patients has revealed that nonsmokers tend to have higher responses to TNF therapy than smokers do [⁵⁶].

Concluding remarks and future directions

Despite efforts devoted to identifying the genetic risk factors of RA, the number of associated genes has been rapidly increasing (Fig. 1), highlighting the complex pathogenesis of this autoimmune disease. In this review, we focused primarily on the genetic and clinical factors that have predictive power to determine drug responses during RA treatment, especially recent developments. On the basis of the collected research articles, we revealed that multiple factors, including genetic variation, gene expression abundance, comorbidities, the use of other drugs, treatment history, and even patient lifestyles, can influence treatment outcomes in RA. To date, most studies have focused on only a limited number of these factors, but combinations of these factors are clearly the most effective predictors of drug responsiveness.

How can we use the available information to predict drug responsiveness? A recent study utilizing machine-learning methods may provide insights into this topic. Lin et al. recently reported their efforts to diagnose RA by mining patient clinical data stored in Electronic Medical Records [⁵⁷]. The authors extracted clinically relevant information from 2792 clinical notes with human diagnoses and used these data as a training set for machine-learning algorithms (MLAs). Then, they applied MLA-generated statistic models to a test set of 2093 clinical notes and obtained a machine-derived diagnosis of disease activity. Consequently, the MLAs achieved a performance similar to human diagnosis. Although drug responsiveness was not explicitly tested in the study, the methods can be applied to study responses to therapies. Interestingly, a rigorous community-based assessment of using SNP data for predicting anti-TNF treatment efficacy in patients with RA has shown no substantial improvement over clinical information [⁵⁸]. In other studies, haplotypes of HLA-DRB1 gene are correlated with disease susceptibility, severity, and mortality but are inversely correlated with TNF inhibitor treatment response [⁵⁹]. Nevertheless, these results may indicate that information other than SNP data, such as the expression abundance of target genes and the extent of DNA methylation, should be considered [⁶⁰]. Dose and intervals of biologics during treatment should be taken into account [⁶⁰].

In addition to pre-treatment screening, short-term drug response is a reliable predictor of long-term clinical outcomes. Hence, therapeutic strategies should be flexible during treatment and revised whenever necessary.

In summary, despite the availability of many drugs and therapeutic strategies, a considerable proportion of patients with RA fail to respond to currently available agents during treatment. As such, challenges in treating patients with RA should be addressed, and additional disease-associated genes and mutations should be identified. More omics data, including gene expression, DNA methylation, and post-translational modifications, should be obtained, and intelligent methods, such as MLAs, should be established to build predictive RA models with an enhanced performance.

References

[1] Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H, Deyo RA, Gabriel S, Hirsch R, Hochberg MC, Hunder GG, Jordan JM, Katz JN, Kremers HM, Wolfe F; National Arthritis Data Workgroup. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum 2008; 58(1): 26–35

[2] Helmick CG, Felson DT, Lawrence RC, Gabriel S, Hirsch R, Kwoh CK, Liang MH, Kremers HM, Mayes MD, Merkel PA, Pillemer SR, Reveille JD, Stone JH; National Arthritis Data Workgroup. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part I. Arthritis Rheum 2008; 58(1): 15–25

[3] Taylor PC, Feldmann M. Anti-TNF biologic agents: still the therapy of choice for rheumatoid arthritis. Nat Rev Rheumatol 2009; 5(10): 578–582

[4] Seymour HE, Worsley A, Smith JM, Thomas SH. Anti-TNF agents for rheumatoid arthritis. Br J Clin Pharmacol 2001; 51(3): 201–208

[5] Costa JO, Lemos LL, Machado MA, Almeida AM, Kakehasi AM, Araújo VE, Cherchiglia ML, Andrade EI, Acurcio FA. Infliximab, methotrexate and their combination for the treatment of rheumatoid arthritis: a systematic review and meta-analysis. Rev Bras Reumatol 2015; 55(2): 146–158

[6] Scott LJ. Etanercept: a review of its use in autoimmune inflammatory diseases. Drugs 2014; 74(12): 1379–1410

[7] Joshi P, Dhaneshwar SS. An update on disease modifying antirheumatic drugs. Inflamm Allergy Drug Targets 2014; 13(4): 249–261

[8] Mitchell A, Chang HY, Daugherty L, Fraser M, Hunter S, Lopez R, McAnulla C, McMenamin C, Nuka G, Pesseat S, Sangrador-Vegas A, Scheremetjew M, Rato C, Yong SY, Bateman A, Punta M, Attwood TK, Sigrist CJ, Redaschi N, Rivoire C, Xenarios I, Kahn D, Guyot D, Bork P, Letunic I, Gough J, Oates M, Haft D, Huang H, Natale DA, Wu CH, Orengo C, Sillitoe I, Mi H, Thomas PD, Finn RD. The InterPro protein families database: the classification resource after 15 years. Nucleic Acids Res 2015; 43(Database issue): D213–D221

[9] Kinsella RJ, Kahari A, Haider S, Zamora J, Proctor G, Spudich G, Almeida-King J, Staines D, Derwent P, Kerhornou A, Kersey P, Flicek P. Ensembl BioMarts: a hub for data retrieval across taxonomic space. Database (Oxford) 2011; 2011: bar030

[10] Okamoto H, Kobayashi A. Tyrosine kinases in rheumatoid arthritis. J Inflamm (Lond) 2011; 8(1): 21

[11] Opar A. Kinase inhibitors attract attention as oral rheumatoid arthritis drugs. Nat Rev Drug Discov 2010; 9(4): 257–258

[12] Vasanthi P, Nalini G, Rajasekhar G. Role of tumor necrosis factor-alpha in rheumatoid arthritis: a review. APLAR J Rheumatol 2007; 10(4): 270–274

[13] Moelants EA, Mortier A, Van Damme J, Proost P. Regulation of TNF-a with a focus on rheumatoid arthritis. Immunol Cell Biol 2013; 91(6): 393–401

[14] Plant D, Wilson AG, Barton A. Genetic and epigenetic predictors of responsiveness to treatment in RA. Nat Rev Rheumatol 2014; 10(6): 329–337

[15] Barrera P, van der Maas A, van Ede AE, Kiemeney BA, Laan RF, van de Putte LB, van Riel PL. Drug survival, efficacy and toxicity of monotherapy with a fully human anti-tumour necrosis factor-a antibody compared with methotrexate in long-standing rheumatoid arthritis. Rheumatology (Oxford) 2002; 41(4): 430–439

[16] Saleem B, Keen H, Goeb V, Parmar R, Nizam S, Hensor EM, Churchman SM, Quinn M, Wakefield R, Conaghan PG, Ponchel F, Emery P. Patients with RA in remission on TNF blockers: when and in whom can TNF blocker therapy be stopped? Ann Rheum Dis 2010; 69(9): 1636–1642

[17] Bohanec Grabar P, Rozman B, Tomsic M, Suput D, Logar D, Dolzan V. Genetic polymorphism of CYP1A2 and the toxicity of leflunomide treatment in rheumatoid arthritis patients. Eur J Clin Pharmacol 2008; 64(9): 871–876

[18] Bohanec Grabar P, Grabnar I, Rozman B, Logar D, Tomsic M, Suput D, Trdan T, Peterlin Masic L, Mrhar A, Dolzan V. Investigation of the influence of CYP1A2 and CYP2C19 genetic polymorphism on 2-Cyano-3-hydroxy-N-[4-(trifluoromethyl)phenyl]-2-butenamide (A77 1726) pharmacokinetics in leflunomide-treated patients with rheumatoid arthritis. Drug Metab Dispos 2009; 37(10): 2061–2068

[19] Cañete JD, Suárez B, Hernández MV, Sanmartí R, Rego I, Celis R, Moll C, Pinto JA, Blanco FJ, Lozano F. Influence of variants of Fcγ receptors IIA and IIIA on the American College of Rheumatology and European League Against Rheumatism responses to anti-tumour necrosis factor α therapy in rheumatoid arthritis. Ann Rheum Dis 2009; 68(10): 1547–1552

[20] Morales-Lara MJ, Conesa-Zamora P, García-Simón MS, Pedrero F, Santaclara V, Perez-Guillermo M, Soriano-Navarro E. Association between the FCGR3A V158F polymorphism and the clinical response to infliximab in rheumatoid arthritis and spondyloarthritis patients. Scand J Rheumatol 2010; 39(6): 518–520

[21] Rizzo R, Rubini M, Govoni M, Padovan M, Melchiorri L, Stignani M, Carturan S, Ferretti S, Trotta F, Baricordi OR. HLA-G 14-bp polymorphism regulates the methotrexate response in rheumatoid arthritis. Pharmacogenet Genomics 2006; 16(9): 615–623

[22] Baricordi OR, Govoni M, Rizzo R, Trotta F. In rheumatoid arthritis, a polymorphism in the HLA-G gene concurs in the clinical response to methotrexate treatment. Ann Rheum Dis 2007; 66(8): 1125–1126

[23] Stamp LK, O’Donnell JL, Chapman PT, Barclay ML, Kennedy MA, Frampton CM, Roberts RL. Lack of association between HLA-G 14 bp insertion/deletion polymorphism and response to long-term therapy with methotrexate response in rheumatoid arthritis. Ann Rheum Dis 2009; 68(1): 154–155

[24] Fabris M, Quartuccio L, Lombardi S, Saracco M, Atzeni F, Carletto A, Cimmino M, Fabro C, Pontarini E, Pellerito R, Bambara LM, Sarzi-Puttini P, Cutolo M, Manfredi M, Benucci M, Morassi P, Fischetti F, Padovan M, Govoni M, Curcio F, Tonutti E, De Vita S. The CC homozygosis of the −174G>C IL-6 polymorphism predicts a lower efficacy of rituximab therapy in rheumatoid arthritis. Autoimmun Rev 2012; 11(5): 315–320

[25] Schotte H, Schlüter B, Drynda S, Willeke P, Tidow N, Assmann G, Domschke W, Kekow J, Gaubitz M. Interleukin 10 promoter microsatellite polymorphisms are associated with response to long term treatment with etanercept in patients with rheumatoid arthritis. Ann Rheum Dis 2005; 64(4): 575–581

[26] Hughes LB, Beasley TM, Patel H, Tiwari HK, Morgan SL, Baggott JE, Saag KG, McNicholl J, Moreland LW, Alarcón GS, Bridges SL Jr. Racial or ethnic differences in allele frequencies of single-nucleotide polymorphisms in the methylenetetrahydrofolate reductase gene and their influence on response to methotrexate in rheumatoid arthritis. Ann Rheum Dis 2006; 65(9): 1213–1218

[27] Kim SK, Jun JB, El-Sohemy A, Bae SC. Cost-effectiveness analysis of MTHFR polymorphism screening by polymerase chain reaction in Korean patients with rheumatoid arthritis receiving methotrexate. J Rheumatol 2006; 33(7): 1266–1274

[28] Cáliz R, del Amo J, Balsa A, Blanco F, Silva L, Sanmarti R, Martínez FG, Collado MD, Ramirez MC, Tejedor D, Artieda M, Pascual-Salcedo D, Oreiro N, Andreu JL, Graell E, Simon L, Martínez A, Mulero J. The C677T polymorphism in the MTHFR gene is associated with the toxicity of methotrexate in a Spanish rheumatoid arthritis population. Scand J Rheumatol 2012; 41(1): 10–14

[29] Inoue S, Hashiguchi M, Takagi K, Kawai S, Mochizuki M. Preliminary study to identify the predictive factors for the response to methotrexate therapy in patients with rheumatoid arthritis. Yakugaku Zasshi 2009; 129(7): 843–849

[30] Takatori R, Takahashi KA, Tokunaga D, Hojo T, Fujioka M, Asano T, Hirata T, Kawahito Y, Satomi Y, Nishino H, Tanaka T, Hirota Y, Kubo T. ABCB1 C3435T polymorphism influences methotrexate sensitivity in rheumatoid arthritis patients. Clin Exp Rheumatol 2006; 24(5): 546–554

[31] Cuchacovich M, Soto L, Edwardes M, Gutierrez M, Llanos C, Pacheco D, Sabugo F, Alamo M, Fuentealba C, Villanueva L, Gatica H, Schiattino I, Salazaro L, Catalan D, Valenzuela O, Salazar-Onfray F, Aguillón JC. Tumour necrosis factor (TNF)α –308 G/G promoter polymorphism and TNFα levels correlate with a better response to adalimumab in patients with rheumatoid arthritis. Scand J Rheumatol 2006; 35(6): 435–440

[32] Stojanović S, Jevtović-Stoimenov T, Stanković A, Pavlović D, Neković J, Stamenković B, Dimić A, Marinković M. Association of TNF-α polymorphism (−308 A/G) with high activity of rheumatoid arthritis and therapy response to Etanercept. Srp Arh Celok Lek 2011; 139(11-12): 784–789 (in Serbian)

[33] Fabris M, Di Poi E, Sacco S, Damante G, Sinigaglia L, Ferraccioli G. TNF-α gene polymorphisms in rheumatoid arthritis patients treated with anti-TNF-α agents: preliminary results. Reumatismo 2002; 54(1): 19–26 (in Italian)

[34] Lee YH, Ji JD, Bae SC, Song GG. Associations between tumor necrosis factor-α (TNF-α) –308 and –238 G/A polymorphisms and shared epitope status and responsiveness to TNF-α blockers in rheumatoid arthritis: a metaanalysis update. J Rheumatol 2010; 37(4): 740–746

[35] Rooryck C, Barnetche T, Richez C, Laleye A, Arveiler B, Schaeverbeke T. Influence of FCGR3A-V212F and TNFRSF1B-M196R genotypes in patients with rheumatoid arthritis treated with infliximab therapy. Clin Exp Rheumatol 2008; 26(2): 340–342

[36] Ulrich CM, Yasui Y, Storb R, Schubert MM, Wagner JL, Bigler J, Ariail KS, Keener CL, Li S, Liu H, Farin FM, Potter JD. Pharmacogenetics of methotrexate: toxicity among marrow transplantation patients varies with the methylenetetrahydrofolate reductase C677T polymorphism. Blood 2001; 98(1): 231–234

[37] Grabar PB, Rozman B, Logar D, Praprotnik S, Dolzan V. Dihydroorotate dehydrogenase polymorphism influences the toxicity of leflunomide treatment in patients with rheumatoid arthritis. Ann Rheum Dis 2009; 68(8): 1367–1368

[38] Rath E, Zwerina J, Oppl B, Nell-Duxneuner V. Efficacy and safety of rituximab in rheumatic diseases. Wien Med Wochenschr 2015; 165(1-2): 28–35

[39] Taylor PC, Williams RO. Combination cytokine blockade: the way forward in therapy for rheumatoid arthritis? Arthritis Rheumatol 2015; 67(1): 14–16

[40] van Baarsen LG, Wijbrandts CA, Rustenburg F, Cantaert T, van der Pouw Kraan TC, Baeten DL, Dijkmans BA, Tak PP, Verweij CL. Regulation of IFN response gene activity during infliximab treatment in rheumatoid arthritis is associated with clinical response to treatment. Arthritis Res Ther 2010; 12(1): R11

[41] de Jong TD, Vosslamber S, Blits M, Wolbink G, Nurmohamed MT, van der Laken CJ, Jansen G, Voskuyl AE, Verweij CL. Effect of prednisone on type I interferon signature in rheumatoid arthritis: consequences for response prediction to rituximab. Arthritis Res Ther 2015; 17(1): 78

[42] Rodríguez-Carrio J, López P, Suárez A. Type I IFNs as biomarkers in rheumatoid arthritis: towards disease profiling and personalized medicine. Clin Sci (Lond) 2015; 128(8): 449–464

[43] Jani M, Chinoy H, Warren RB. Em Griffiths C, Plant D, Fu B, Morgan AW, Wilson AG, Isaacs JD, Hyrich KL, Barton A; Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate Collaborators. Clinical utility of random anti-TNF drug level testing and measurement of anti-drug antibodies on long-term treatment response in rheumatoid arthritis. Arthritis Rheumatol 2015; 385(8): S48

[44] Takeuchi T, Yamamoto K, Yamanaka H, Ishiguro N, Tanaka Y, Eguchi K, Watanabe A, Origasa H, Shoji T, Miyasaka N, Koike T. Early response to certolizumab pegol predicts long-term outcomes in patients with active rheumatoid arthritis: results from the Japanese studies. Mod Rheumatol 2015; 25(1): 11–20

[45] Blaschke S, Rinke K, Maring M, Flad T, Patschan S, Jahn O, Mueller CA, Mueller GA, Dihazi H. Haptoglobin-a1, -a2, vitamin D-binding protein and apolipoprotein C-III as predictors of etanercept drug response in rheumatoid arthritis. Arthritis Res Ther 2015; 17(1): 45

[46] Atzeni F, Bongiovanni S, Marchesoni A, Filippini M, Caporali R, Gorla R, Cavagna L, Favalli EG, Saccardo F, Sarzi-Puttini P. Predictors of response to anti-TNF therapy in RA patients with moderate or high DAS28 scores. Joint Bone Spine 2014; 81(1): 37–40

[47] Barnabe C, Homik J, Barr SG, Martin L, Maksymowych WP. The effect of different remission definitions on identification of predictors of both point and sustained remission in rheumatoid arthritis treated with anti-TNF therapy. J Rheumatol 2014; 41(8): 1607–1613

[48] Kristensen LE, Kapetanovic MC, Gülfe A, Söderlin M, Saxne T, Geborek P. Predictors of response to anti-TNF therapy according to ACR and EULAR criteria in patients with established RA: results from the South Swedish Arthritis Treatment Group Register. Rheumatology (Oxford) 2008; 47(4): 495–499

[49] Luban S, Li ZG. Citrullinated peptide and its relevance to rheumatoid arthritis: an update. Int J Rheum Dis 2010; 13(4): 284–287

[50] Pietrapertosa D, Tolusso B, Gremese E, Papalia MC, Bosello SL, Peluso G, Petricca L, Michelutti A, Faustini F, Fedele AL, Ferraccioli G. Diagnostic performance of anti-citrullinated peptide antibodies for the diagnosis of rheumatoid arthritis: the relevance of likelihood ratios. Clin Chem Lab Med 2010; 48(6): 829–834

[51] Narváez J, Díaz-Torné C, Ruiz JM, Hernandez MV, Torrente-Segarra V, Ros S, Rodriguez de la Serna A, Díaz-López C, Sanmartí R, Nolla JM. Predictors of response to rituximab in patients with active rheumatoid arthritis and inadequate response to anti-TNF agents or traditional DMARDs. Clin Exp Rheumatol 2011; 29(6): 991–997

[52] Chara L, Sánchez-Atrio A, Pérez A, Cuende E, Albarrán F, Turrión A, Chevarria J, del Barco AA, Sánchez MA, Monserrat J, Prieto A, de la Hera A, Sanz I, Diaz D, Alvarez-Mon M. The number of circulating monocytes as biomarkers of the clinical response to methotrexate in untreated patients with rheumatoid arthritis. J Transl Med 2015; 13(1): 2

[53] Jutley G, Raza K, Buckley CD. New pathogenic insights into rheumatoid arthritis. Curr Opin Rheumatol 2015; 27(3): 249–255

[54] Johannsen A, Susin C, Gustafsson A. Smoking and inflammation: evidence for a synergistic role in chronic disease. Periodontol 2000 2014; 64(1): 111–126

[55] Scott IC, Tan R, Stahl D, Steer S, Lewis CM, Cope AP. The protective effect of alcohol on developing rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology (Oxford) 2013; 52(5): 856–867

[56] Hyrich KL, Watson KD, Silman AJ, Symmons DP; British Society for Rheumatology Biologics Register. Predictors of response to anti-TNFα therapy among patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Rheumatology (Oxford) 2006; 45(12): 1558–1565

[57] Lin C, Karlson EW, Canhao H, Miller TA, Dligach D, Chen PJ, Perez RN, Shen Y, Weinblatt ME, Shadick NA, Plenge RM, Savova GK. Automatic prediction of rheumatoid arthritis disease activity from the electronic medical records. PLoS One 2013; 8(8): e69932

[58] Sieberts SK, Zhu F, García-García J, Stahl E, Pratap A, Pandey G, Pappas D, Aguilar D, Anton B, Bonet J, Eksi R, Fornés O, Guney E, Li H, Marín MA, Panwar B, Planas-Iglesias J, Poglayen D, Cui J, Falcao AO, Suver C, Hoff B, Balagurusamy VS, Dillenberger D, Neto EC, Norman T, Aittokallio T, Ammad-Ud-Din M, Azencott CA, Bellón V, Boeva V, Bunte K, Chheda H, Cheng L, Corander J, Dumontier M, Goldenberg A, Gopalacharyulu P, Hajiloo M, Hidru D, Jaiswal A, Kaski S, Khalfaoui B, Khan SA, Kramer ER, Marttinen P, Mezlini AM, Molparia B, Pirinen M, Saarela J, Samwald M, Stoven V, Tang H, Tang J, Torkamani A, Vert JP, Wang B, Wang T, Wennerberg K, Wineinger NE, Xiao G, Xie Y, Yeung R, Zhan X, Zhao C; Members of the Rheumatoid Arthritis Challenge Consortium, Greenberg J, Greenberg J, Kremer J, Michaud K, Barton A, Coenen M, Mariette X, Miceli C, Shadick N, Weinblatt M, de Vries N, Tak PP, Gerlag D, Huizinga TW, Kurreeman F, Allaart CF, Louis Bridges S Jr, Criswell L, Moreland L, Klareskog L, Saevarsdottir S, Padyukov L, Gregersen PK, Friend S, Plenge R, Stolovitzky G, Oliva B, Guan Y, Mangravite LM, Bridges SL, Criswell L, Moreland L, Klareskog L, Saevarsdottir S, Padyukov L, Gregersen PK, Friend S, Plenge R, Stolovitzky G, Oliva B, Guan Y, Mangravite LM. Crowdsourced assessment of common genetic contribution to predicting anti-TNF treatment response in rheumatoid arthritis. Nat Commun 2016; 7: 12460

[59] Viatte S, Plant D, Han B, Fu B, Yarwood A, Thomson W, Symmons DP, Worthington J, Young A, Hyrich KL, Morgan AW, Wilson AG, Isaacs JD, Raychaudhuri S, Barton A. Association of HLA-DRB1 haplotypes with rheumatoid arthritis severity, mortality, and treatment response. JAMA 2015; 313(16): 1645–1656

[60] Takeuchi T, Miyasaka N, Tatsuki Y, Yano T, Yoshinari T, Abe T, Koike T. Inhibition of plasma IL-6 in addition to maintenance of an efficacious trough level of infliximab associated with clinical remission in patients with rheumatoid arthritis: analysis of the RISING Study. Ann Rheum Dis 2012; 71(9): 1583–1585

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH, Germany, part of Springer Nature