The association between clinical variables and response to several drugs was analyzed; only a significant or near-significant associations are shown in Table 3.
We observed that older patients and those diagnosed after the age of 40 responded better to 5-ASA compounds. In contrast, duration of disease for more than 5 years was a negative predictor of response both for 5-ASA and for azathioprine, although the latter did not reach statistical significance (P=0.008 and 0.07, respectively). Disease extent negatively influenced response to 5-ASA and azathioprine, with patients with pancolitis showing poorer responses (Table 3). With respect to EIMs, we observed that it was a negative predictor of response to 5-ASA, corticosteroids, and azathioprine (P=0.001, 0.06, and 0.02, respectively) but it seemed to positively influence response to biologics, although not significantly (OR 8.0, 95% CI 0.72–88.22; P=0.09).
Interestingly, previous therapies also seemed to influence response to several drugs. Thus, previous users of corticosteroids, azathioprine, and/or biologics responded better to 5-ASA. The same was observed for corticosteroids and azathioprine, suggesting that the use of IS or biologics may allow recapture of patients to respond to therapies to which they were previously nonresponsive.
Table 4 shows the associations between IL23R SNPs and clinical characteristics including response to therapy. We observed that carriers of IL23R C2370A and IL23R_G4305A alleles were at an increased risk of showing EIMs (Table 4). No significant association was observed with disease extension or other phenotypic characteristics. With respect to response to therapy, we observed that IL23R_C2370A negatively influenced response to 5-ASA and corticosteroids, whereas homozygous carriers for IL23R_G9T were more likely to respond to azathioprine.
We also examined whether there was any interaction between clinical and genetic predictors of response. We observed that IL23R_C2370A CA carriers with EIMs had a significantly lower chance of responding to 5-ASA compounds (OR 0.12, 95% CI 0.018–0.86; P=0.03).
Compared with CD, the individual natural history of patients with UC is harder to predict than in CD 21,22 and the relevance of clinical predictors has not been validated in independent cohorts 13. Recent papers strongly suggest that UC, as CD, should be considered as progressive diseases resulting in cumulative bowel damage, thereby affecting bowel function and the patient’s quality of life 10–12.
As most patients with UC respond to 5-ASA-based therapies, clinicians should focus on early identification of nonresponders to these therapies as these should be started on more aggressive and effective therapies as early as possible. Most series published so far identified young age, female sex, and extensive colitis as poor prognostic markers, theoretically but not necessarily, with less probability of responding to therapy with 5-ASA only 2–4. Serologic and genetic predictors have been far less explored in UC than in CD and only a few clinical settings have been investigated 13. In a recent study carried out in 94 children with IBD, the authors identified pANCA positivity and a diagnosis of UC as predictors for a primary nonresponse to infliximab 7. Genetic markers are more attractive as they are already present at the onset of the disease and remain stable over time, which is not the case for clinical and serologic parameters. In the present study, we focused on IL23R SNPs because there is accumulating evidence that IL23 cytokine is essential to drive the chronic intestinal inflammation observed in IBD, namely UC. IL23 is produced by activated dendritic cells and macrophages present in the gut mucosa in response to microbial stimulation 17. Although initially found to support the expansion and maintenance of CD4+ T helper 17 cells, IL23 is now recognized as having multiple effects on the immune response, including restraining Foxp3+ regulatory T-cell activity and inducing the expression of Th17-type cytokines from non-T-cell sources 17. Certainly related to this diversity and multiplicity of upstream and downstream effects, certain of these SNPs confer a protective effect against developing IBD 19, whereas others increase the susceptibility to develop IBD 7,15. Furthermore, there are significant ethnic differences, with some variants increasing the risk of developing IBD in Whites but not in Asian individuals 18. These apparent discrepancies may be explained by the fact that IL23R contributes to colitis pathogenesis through various pathways that may be tackled in a number of different ways, thereby resulting in various types of responses. The exact mechanism by which the studied polymorphisms exert their effect on disease pathogenesis and on metabolism and mechanism of action of drugs is still not known. When we consider SNPs affecting coding regions, such as the IL23R G1142A or G9T polymorphisms, we can hypothesize changes in protein structure, namely an effect on the helical structure of the protein product 23 and consequently modifications in receptor affinity. In what concerns the SNP IL23R C2370A, located in the 3′ untranslated region, the latter may be associated with RNA stability or even translation efficiency, which in turn lead to changes in protein expression. Finally, the IL23R G43045A with intronic localization could be associated with RNA splicing or control of transcription, and indirectly affect RNA stability or gene expression. However, we cannot exclude that these polymorphisms could be in linkage with other variants in genes located nearby, responsible for UC pathogenesis and/or disease phenotype. Finally, the reason why certain of these variants are associated with different responses to therapy is probably through the modulation of the phenotype that they produce as opposed to a direct interference with drug metabolism.
In the present study, besides clinical and genetic predictors of response to infliximab, we also focused on predictors of nonresponse to other therapies including 5-ASA and azathioprine, which is certainly useful in clinical practice, as only a minority of patients will need escalation to biologics. We also decided to use long-term, more than 1-year, response, because we believe that this concept is clinically more relevant. Thus, a patient who responded to 5-ASA, but before 1 year relapsed with the need for corticosteroids or escalation of therapy, was considered a nonresponder to 5-ASA. Early institution of IS and/or biologic therapy in these patients could be a way of maintaining the patient in remission and avoiding the complications of undertreated disease.
In the present study, we examined whether clinical and/or IL23R SNPs, alone or in combination, could be used as predictors of response to several drugs used in the treatment of UC. Similar to previous series, over 60% of patients responded to 5-ASA compounds. We had estimated that at least 40% of patients would be refractory to 5-ASA and would require azathioprine and/or biologics. As a step-up approach was followed in all instances of patients not responding to 5-ASA, most of these patients were started on azathioprine, 58% of whom responded to this therapy. Thus, only 25 of 174 patients ultimately required biologics, which may preclude some statistical significance.
Consistent with previous studies 2–5,24,25, we observed that older patients responded better to 5-ASA, whereas no influence was observed in terms of probability of response to azathioprine or biologics. Duration of disease for more than 5 years was associated with a lower probability of response both for 5-ASA and for azathioprine. In agreement with previous studies, disease extent 2–5,24,25 was also a negative predictive factor for responding to 5-ASA and azathioprine but not to steroids or biologics. EIMs were strongly associated with poor response to 5-ASA, steroids, and azathioprine, whereas a nonsignificant but positive association was observed in terms of biologics. To our knowledge, this has not been clearly reported before. EIMs are among the clinical risk factors that may be associated with disease extent and severity 26, and a recent review on this topic 27 suggests that early aggressive therapy may be required for treating several EIMs to prevent chronic damage. According to our results, together with young age of disease onset and greater disease extent, the presence of EIMs may predict nonresponse to 5-ASA and need for aggressive therapy.
A very interesting association was also observed with previous therapies. Thus, previous users of steroids, azathioprine, or biologics were more likely to respond to 5-ASA. This recapture phenomenon is frequently observed in clinical practice. Similarly, previous users of azathioprine and biologics had a greater chance of responding to corticosteroids and that of biologics with a greater likelihood of responding to azathioprine. These observations suggest that patients who did not respond to 5-ASA and required escalation of therapy, after remission was achieved, may be maintained on 5-ASA.
With respect to IL23R SNPs, these have been described as susceptibility genes but to our knowledge there are no previous reports on whether these SNPs influence phenotypic manifestations in UC patients. A recent study by a German group suggests that homozygous carriers of IL23R variants that increase the risk of developing IBD are more likely to respond to infliximab 7. In the present study, we found that IL23R SNPs were associated with a significantly increased risk of developing EIMs, with a lower chance of responding to 5-ASA and corticosteroids but with a higher chance of responding to azathioprine. No association was found between these SNPs and response to infliximab, which may be related to the low number of patients treated with this drug. We also found a significant interaction between the IL23R_C2370A allele and the presence of EIM in predicting nonresponse to 5-ASA. This is consistent with these two factors, alone or in combination, being early indicators of need for azathioprine and/or infliximab therapy in agreement with a previous study published by Jurgens et al. 7. According to our data, older patients, with limited disease, not showing EIMs and not carrying IL23R variants are more prone to respond to 5-ASA compounds. The effect of IL23R variant alleles may be independent or because of the fact that they are more often associated with EIM, which is also a negative predictor for responding to 5-ASA compounds.
This study had some limitations that need to be addressed. A major drawback is the fact that response to drugs was evaluated retrospectively from data recorded in the charts. The decision to change a drug was made according to the treating physician, which might not be uniform across the several centers involved in the study. Nonetheless, the numbers obtained for response to several drugs are close to the ones reported in the literature, thereby suggesting similar policies of treatment. In our opinion, evaluating the response to a specific drug after the long-term evolution and whether the patient required therapy escalation after a transitory response that lasted less than 1 year is clinically more relevant than evaluating short-term responses. Another drawback of this study was that disease activity was not recorded on admission and it is plausible that this influenced therapeutic decisions. However, our main purpose was to identify early predictors of response and not prognosis and a conventional step-up approach was followed in the large majority of patients. Finally, it would be interesting to have a better characterization of the type of EIMs to determine whether they all had the same prognostic value.
In the present study, we could identify a number of clinical variables that could predict nonresponse to 5-ASA compounds, namely, the existence of EIMs. The latter were significantly more frequent in IL23R SNPs carriers, which was also associated with poor response to 5-ASA but to a higher probability of responding to azathioprine. These findings might be valuable in clinical practice, although further confirmation is still needed.
This work was funded by a grant from GEDII 2009-2012.
Conflicts of interest
There are no conflicts of interest.
1. Dignass A, Lindsay JO, Sturn A, Sturm A, Windsor A, Colombel JF, et al..Second European evidence-based consensus of the diagnosis and management of ulcerative colitis
: current management.J Crohns Colitis2012;6:991–1030.
2. Langholz E, Munkholm P, Davidsen M, Binder V.Course of ulcerative colitis
: analysis of changes in disease activity over years.Gastroenterology1994;107:3–11.
3. Hoie O, Wolters F, Riis L, Solberg C, Bernklev T, Odes S, et al..Ulcerative colitis
: patients characteristics may predict 10-year disease recurrence in a European-wide population based cohort.Am J Gastroenterol2007;102:1692–1701.
4. Solberg IC, Lygren I, Janhsen J, Aadland E, Høie O, Cvancarova M, et al..Clinical course during the first 10 years of ulcerative colitis
: results from a population-based inception cohort (IBSEN Study).Scand J Gastroenterol2009;44:431–440.
5. Portela F, Magro F, Lago P, Cotter J, Cremers I, de Deus J, et al..Ulcerative colitis
in a Southern European country: a national perspective.Inflamm Bowel Dis2010;16:822–829.
6. Bessissow T, Lemmens B, Ferrante M, Bisschops R, Van Steen K, Geboes K, et al..Prognostic value of serologic and histologic markers on clinical relapse in ulcerative colitis
patients with mucosal healing.Am J Gastroenterol2012;107:1684–1692.
7. Jurgens M, Laubender RP, Hartl F, Weidinger M, Seiderer J, Wagner J, et al..Disease activity, ANCA, and IL-23R genotype status determine early response to infliximab in patients with ulcerative colitis
.Am J Gastroenterol2010;105:1811–1819.
8. Dubinsky MC, Mei L, Friedman M, Dhere T, Haritunians T, Hakonarson H, et al..Genome wide association predictors of anti-TNFα therapeutic responsiveness in pediatric inflammatory bowel disease (IBD).Inflamm Bowel Dis2010;16:1357–1366.
9. Ferrante M, Vermeire S, Katsanos K, Noman M, Van Assche G, Schnitzler F, et al..Predictors of early response to infliximab in patients with ulcerative colitis
.Inflamm Bowel Dis2007;13:123–128.
10. Torres J, Billioud V, Sachar DB, Peyrin-Biroulet L, Colombel JF.Ulcerative colitis
as a progressive disease: the forgotten evidence.Inflamm Bowel Dis2012;18:1356–1363.
11. Ochsenkuhn T, D’Haens G.Current misunderstandings in the management of ulcerative colitis
12. Torres J, Billioud V, Peyrin-Biroulet L, Colombel JF.Ulcerative colitis
as a sole mucosal disease: another misunderstanding (Letter).Gut2012;61:633.
13. Beaugerie L.Clinical, serological and genetic predictors of inflammatory bowel disease course.World J Gastroenterol2012;18:3806–3813.
14. Vermeire S, Van Assche G, Rutgeerts P.Role of genetics in prediction of disease course and response to therapy.World J Gastroenterol2010;16:2609–2615.
15. Duerr RH, Taylor KD, Brant SR, Rioux JD, Silverberg MS, Daly MJ, et al..A genome-wide association study identifies IL23R as an inflammatory bowel disease gene.Science2006;314:1461–1463.
16. Sarra M, Pallone F, MacDonald T, Monteleone G.IL-23/IL-17 axis in IBD.Inflamm Bowel Dis2010;16:1808–1813.
17. Morrison PJ, Ballantyne SJ, Kullberg MC.Interleukin-23 and T-helper 17-type responses in intestinal inflammation: from cytokines to T-cell plasticity.Immunology2011;133:397–408.
18. Kim SW, Kim ES, Moon CM, Park JJ, Kim TI, Kim WH, Cheon JH.Genetic polymorphisms of IL23R and IL-17A and novel insights into their associations with inflammatory bowel disease.Gut2011;60:1527–1536.
19. Dubinski MC, Wang D, Picomell Y, Wrobel I, Katzir L, Quiros A, et al..IL-23 receptor (IL-23R) gene protects against pediatric Crohn’s disease.Inflamm Bowel Dis2007;13:511–515.
20. Guo SW, Thompson EA.Performing the exact test of Hardy–Weinberg proportion for multiple alleles.Biometrics1992;48:361–372.
21. Dignass A, Eliakim R, Magro F, Maaser C, Chowers Y, Geboes K, et al..Second European evidence-based consensus on the diagnosis and management of ulcerative colitis
: definitions and diagnosis.J Crohns Colitis2012;6:965–990.
22. Cosnes J, Gower-Rousseau C, Seksik P, Cortor A.Epidemiology and natural history of inflammatory bowel diseases.Gastroenterology2011;140:1785–1794.
23. Safrany E, Szabo M, Szell M, Kemeny L, Sumegi K, Melegh BI, et al..Difference of interleukin-23 receptor gene haplotype variants in ulcerative colitis
compared to Crohn’s disease and psoriasis.Inflamm Res2013;62:195–200.
24. Bitton T, Peppercorn MA, Antonioli DA, Niles JL, Shah S, Bousvaros A, et al..Clinical, biological, and histologic parameters as predictors of relapse in ulcerative colitis
25. Romberg-Camps MJ, Dagnelie PC, Kester AD, Hesselink-van de Kruijs MA, Cilissen M, Engels LG, et al..Influence of phenotype at diagnosis and of other potential prognostic factors on the course of inflammatory bowel disease.Am J Gastroenterol2009;104:371–383.
26. Farmer RG, Easley KA, Rankin GB.Clinical patterns, natural history, and progression of ulcerative colitis
. A long term follow-up of 1116 patients.Dig Dis Sci1993;38:1137–1146.
27. Veloso FT.Extraintestinal manifestations
of inflammatory bowel disease: do they influence treatment and outcome.World J Gastroenterol2011;17:2702–2707.
Keywords:Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
extraintestinal manifestations; IL23R single nucleotide polymorphisms; predictors of response to therapy; ulcerative colitis