Newly Diagnosed Crohn's/infliximab article, part 1

[David in Seattle]

Clinical Outcome of Newly Diagnosed Crohn's Disease: A Comparative, Retrospective Study before and after Infliximab Availability

Alimentary Pharmacology & Therapeutics. 2009;31(2):233-239.

Abstract

Background Infliximab (IFX) could change the course of Crohn's disease (CD) by reducing steroid use, surgery or prompting earlier introduction of immunomodulators (IMM).
Aim To evaluate the impact of IFX availability on the course of early CD.
Methods Two cohorts of newly diagnosed CD patients were identified: The first cohort included patients diagnosed from January 1994 to December 1997 and the second from January 2000 to December 2003. All patients were diagnosed, treated and followed up in the same centre until December 1999 (first cohort) or December 2005 (second cohort). Development of disease-related complications, steroid, IMM or IFX requirements and intestinal resections during follow-up were registered.
Results A total of 328 patients were included (146 first cohort, 182 second cohort). A similar proportion of patients in both cohorts received steroids, but steroid exposure resulted significantly more intense in the first cohort (P = 0.001). In the second cohort, 14% of patients received IFX. Thiopurines were used more (P = 0.001) and earlier (P = 0.012) in the second cohort. No differences in surgical requirements or the development of disease-related complications were found.
Conclusions Following a step-up therapeutic algorithm, IFX availability did not reduce surgical requirements or the development of disease-related complications.
Introduction

Crohn's disease (CD) is a chronic relapsing and remitting inflammatory condition of the intestine. Although most patients present with uncomplicated disease at the time of diagnosis, a great proportion of them will either present chronic inflammatory activity despite conventional therapy or develop penetrating complications, intestinal stenosis or perianal disease within the first five to 10 years after CD diagnosis,[1–3] leading to the deterioration of the patient's quality of life. Efforts have been made to identify early those patients who will have a more aggressive course of the disease. In this sense, Beaugerie and colleagues evaluated the development of what they arbitrarily called 'disabling disease' in a retrospective series of more than one thousand CD patients, and found an estimated prevalence of 59–92% within the first 5 years from diagnosis. In this study, the initial need for systemic steroids and the presence of perianal disease at CD onset or disease diagnosis under the age of 40 years were independent predictive factors of disabling disease.[3]

Thiopurines have proven to be effective in reducing steroid exposure, maintaining disease in remission and inducing mucosal healing.[4, 5] Under this perspective, it was thought that they might also change the natural course of CD; in a classical retrospective study, Cosnes et al. showed that the increasing use of thiopurines in the last decades was not associated with a diminished rate of intestinal resection.[6] To prevent or delay the occurrence of disease complications, a trend towards an earlier and/or more intensive therapeutic approach has emerged in the last years. Markowitz et al. demonstrated in a RCT with new onset paediatric CD patients that early introduction of mercaptopurine significantly reduced steroid requirements and increased the likelihood to remain in clinical remission as compared to placebo.[7] More recently, D'Haens et al. showed that in adults with early CD, a more intensive therapeutic regimen using an induction regimen with infliximab (IFX) followed by maintenance therapy with thiopurines achieved disease control significantly earlier and led to mucosal healing in a greater proportion of patients as compared with conventional therapy.[8]

The aim of the present study was to assess if the availability of IFX for the treatment of CD has had any impact on the CD outcome within the first years from diagnosis. In the 1990s, the use of immunomodulators (IMM) for CD management was widely established in our centres and open-label IFX use for induction and maintenance of disease remission was approved in Europe in 2000. This was the reason why we chose to compare two historical periods to elucidate, in a clinical practice setting, whether the availability of IFX for the treatment of CD has had any impact on CD outcome within the first years from diagnosis. For this, we evaluated the clinical outcome and therapeutic requirements within the first 2–5 years of disease in two hospital-based inception cohorts of CD patients, basically differing by the availability of IFX treatment.

Two cohorts of CD patients were identified: the IMM cohort included patients diagnosed with CD between January 1994 and December 1997 and the IFX cohort included patients diagnosed between January 2000 and December 2003. All patients were identified from the IBD databases of five referral Spanish centres. The study was approved by the Institutional Review Board of the steering centre (Hospital Universitari Germans Trias i Pujol, Badalona). Diagnosis of CD was based on the classic Lennard-Jones criteria.[9] Patients were diagnosed, treated and followed up in a single centre until death or the end of follow-up that was decided to be December 1999 (first cohort) and December 2005 (second cohort). Patients lost to follow-up for any cause (except death) for more than 6 months within the study period, were excluded. As one of the main objectives of the study was to assess the development of disease complications within the first years from CD diagnosis, those patients in whom an intestinal resection was performed within the first 3 months were excluded. Demographic and epidemiological baseline characteristics, as well as the initial Montreal classification (that includes location and behaviour of the disease, and the presence of perianal involvement)[10] were recorded. To evaluate disease outcomes, the use and timing of introduction of systemic steroids, IMM or IFX, together with surgery requirements during follow-up were carefully addressed. In addition, any change in the initial Montreal classification (development of intestinal stenosis, intra-abdominal abscess or fistulae, perianal disease, changes in disease location) was also assessed and recorded. In summary, we evaluated retrospectively the clinical outcome and therapeutic requirements within the first 2–5 years of disease in two hospital-based inception cohorts of CD patients, basically differing by the availability of IFX treatment.
Statistical Analysis

Data are expressed as mean ± standard deviation or frequencies. Baseline characteristics between both cohorts were compared with chi-square analysis for qualitative variables and Student t-test for quantitative variables. Risk factors for IMM introduction and intestinal resection were univariately analysed by the log-rank test. Multivariate analysis of those variables reaching statistical significance was then performed with a Cox regression. Cumulative probabilities of IMM introduction, IFX introduction and intestinal resection for the whole series as well as for each study cohort were calculated by the Kaplan–Meier method. All statistical analyses were performed using SPSS 12.0 package for Windows (SPSS Inc., Chicago, IL, USA).

 

[David in Seattle]

Newly Diagnosed Crohn's/infliximab article, part 2

Results

A total of 328 CD patients were included, 146 of them in the first cohort and 182 in the second cohort. Baseline characteristics were similar in both cohorts, although patients in the first cohort were significantly younger at diagnosis (Table 1). None of the Montreal's classification parameters differed at the time of CD diagnosis between both cohorts. A majority of patients were diagnosed between 17 and 40 years of age, most of them had ileal disease (either alone or together with colonic involvement) and 78% had an initial inflammatory behaviour of disease. Perianal disease was present at onset in 12% of patients. Interestingly, almost half of the patients were current smokers at diagnosis.

Follow-up features of patients in both cohorts are summarized in Table 2. No death was recorded during the study period, after a mean follow-up of 45 months. At least one-third of patients continued smoking during follow-up with a significant greater proportion of current smokers in the first cohort. Smoking was not associated with a worse outcome as judged by the development of stenosing or penetrating complications, perianal disease, need for steroids, IMM, or intestinal resections.

Twenty-three percent of patients modified their initial Montreal classification after a mean of 40 ± 16 months from disease diagnosis, mainly because of the change in the behavioural pattern of CD (15%) or the development of perianal disease (8%). No differences between the two cohorts were found either in the proportion of patients developing stenosing or intra-abdominal penetrating disease complications or in the time for their occurrence (Table 2). The development of de novo perianal disease was similar in both cohorts, but this occurred earlier in the first cohort than in the second cohort (P = 0.007).

Seventy percent of patients required at least one course of steroids during the follow-up. Although the proportion of patients requiring a course of steroids was not different between the two cohorts, the total time on steroids during follow-up was significantly longer in the first cohort (P = 0.001). Thiopurines were introduced in 43% of patients during follow-up. Main indications for IMM introduction were steroid-dependence (67%), perianal disease (20%), and prevention of post-operative recurrence (15%). According to clinical outcomes, indications were similar in both cohorts, except for steroid-dependence that accounted for a greater proportion of IMM-treated patients of the first cohort (76% vs. 62%). The cumulative probability of IMM introduction in the whole series was 19%, 33% and 41% after 1, 2 and 3 years from diagnosis respectively. Of interest, IMM were started significantly earlier (P = 0.003) and in a higher proportion of patients (P = 0.049) in the second cohort, leading to a cumulative probability of IMM introduction in the first and the second cohorts of 11% and 26% at 1 year, 22% and 40% at 2 years and 34% and 48% at 3 years respectively (P = 0.0089) (Figure 1). The multivariate analysis (Cox model) showed that a change in the initial Montreal classification (P = 0.001), male gender (P = 0.01) and belonging to the first cohort (P = 0.002) were the only independent predictors of IMM introduction.

Click to zoom
(Enlarge Image)
Figure 1.

Cumulative probability to introduce immunomodulators during follow-up.
[ CLOSE WINDOW ]

Figure 1.
Cumulative probability to introduce immunomodulators during follow-up.

Infliximab was introduced in 14% of patients of the second cohort and only in 1% in the first cohort, as expected. All patients treated with IFX also received thiopurines: 12 for at least 6 months, nine for less than 6 months and six started thiopurines concomitantly to IFX. When those patients in whom IFX was introduced were excluded, no differences between the two cohorts were found in the timing for IMM introduction.

One quarter of patients required intestinal resection during follow-up. Indication for surgery was intestinal stenosis in 57%, intra-abdominal penetrating complications in 41% and refractory luminal disease in 2%. The cumulative probability of intestinal resection was 11%, 15% and 21% at 1, 2 and 3 years after diagnosis respectively. According to the data related to the changes in Montreal classification, no differences between the two cohorts were found regarding the timing, indication and the proportion of patients requiring intestinal resection (Table 3, Figure 2). The multivariate analysis demonstrated that only an aggressive disease behaviour pattern (fistulizing or stenosing) either at CD diagnosis (P < 0.0001) or during follow-up (P < 0.0001) was an independent predictor of resectional surgery.

 

[David in Seattle]

Newly Diagnosed Crohn's/infliximab article, part 3

Discussion

Several studies have shown that CD management by means of a conventional therapeutic algorithm (introducing IMM only in case of chronically active or refractory disease) is associated with a high risk of early development of disease-related complications.[1, 2] Most of these studies were performed before biological agents became available in clinical practice. Recently, the concept that a more intensive treatment strategy could change the natural history of the disease has emerged. This intensive treatment strategy, although yet to be clearly defined, implies the use of powerful drugs (i.e. IMM or biologicals) from the time of disease diagnosis. Early introduction of azathioprine after inducing clinical remission with prednisone[7] or IFX[8] has been shown to reduce the likelihood of clinical relapse and steroid requirements and also to increase the rate of mid-term mucosal healing. However, there are many drawbacks in this intensive therapeutic approach: first, the follow-up periods of these RCTs were not long enough to assess if these strategies also reduce the development of stenosing and penetrating complications or even surgery requirements. Second, data comparing early introduction of IMM or/and early use of biologicals as maintenance treatments are still lacking. Finally, early introduction of these drugs would result in the overtreatment of 30–50% of patients who will not develop a disabling disease within the first 5 years from diagnosis, putting them at risk for drug-related adverse effects. Our study aimed to assess if the availability of IFX changed the initial course of CD in patients managed in a conventional manner. The obtained results suggest that although the availability of IFX prompted an earlier and wider use of IMM, it did not lead to an improvement in CD natural history as measured by the development of disease-related complications. Nevertheless, we also found that IFX availability was associated with a significant reduction in steroid exposure. There are several explanations for these findings. First, patients in the first cohort could have a more aggressive course of the disease as long as they were younger at disease diagnosis and had a greater proportion of active smokers during follow-up; however, this was not reflected in a higher prevalence of disease complications such as intestinal stenosis or intra-abdominal abscesses. Second, a change in general treatment strategies of CD among the treating physicians could occur in the last decade. Finally, this reduction in steroid use might be related to the use of IFX given that the earlier and wider use of IMM was clearly because of the availability of the former. Although the incidence of infectious complications was not addressed in this study, the reduction in steroid exposure might be relevant as it has been recently shown that steroid use is one of the major risk factors to develop severe infections among patients with inflammatory bowel disease.[11, 12] Some clinical observations suggest that the earlier inflammatory activity is controlled, the higher is the likelihood to remain in clinical remission[7, 13] or to achieve mucosal healing.[8] In addition, the effectiveness of a given drug may depend on how early it is introduced. For instance, an induction scheme with 3 infusions of IFX appears to be more effective for inducing remission in new onset CD[8] than in late CD.[14] Opposite to the results obtained by Cosnes et al.,[6] it has been recently shown that thiopurines are able to reduce the risk of intestinal resection when introduced early in the course of the disease.[15] This improved efficacy and better outcomes of a given drug in early CD as compared to late CD seems to be related to different cytokine profiles and immune responses at different time settings of the disease.[16]

In the present study, only 14% of patients in the second cohort were treated with IFX and when used, IFX was not introduced early after diagnosis in most cases. Our aim was to assess the impact of the availability (not the use) of IFX on the natural course of CD. Consequently, as it occurred with IMM,[6] we can only conclude that IFX do not alter CD natural history when used in the setting of a step-up therapeutic algorithm. Although those patients who underwent an initial intestinal resection were excluded from the study, we still found a high incidence of disease-related complications early in the course of the disease. About one-fourth of patients underwent an intestinal resection (most of them because of stenosis or intra-abdominal penetrating complications) after a mean time of 2 years from diagnosis, and almost 10% of patients developed perianal disease within the first 3 years. It could be argued that IMM were indicated in only 43% of patients, a figure that is clearly lower than that reported in the step-up/top-down study where almost 70% of patients in the step-up arm required IMM within the first year.[8] However, when evaluating the second cohort alone, IMM were introduced in almost 50% of patients after a mean of 14 months from diagnosis (closer to the figures reported by D'Haens), but with a similar incidence of disease-related complications as in the first cohort. It could also be argued that the proportion of patients who used IFX in the second cohort was unusually low (14%) and that this precludes a proper comparison between the two study cohorts. However, this percentage of patients treated with IFX is far from unusual: an identical proportion of patients needing IFX were reported after 2 years among patients in the 'step-up' group in the D'Haens study, showing that our data reflect a conventional therapeutic algorithm.[8] From this point of view, it is uncertain if a cohort with patients whose diagnosis was made within 2003–2005 would have resulted in an earlier and/or more widespread prescription of IFX. These data reinforce the idea that IMM and/or IFX should be introduced at the time of CD diagnosis if a change in natural history of CD is warranted.

In conclusion, our results seem to indicate that the availability of IFX has little impact in the initial course of new onset CD if a conventional step-up therapeutic algorithm is used. The development of stenosing and penetrating complications, perianal disease and the need for intestinal resections remain unchanged. However, the opportunity to use biologicals seems to accelerate the introduction of IMM and to spare steroid exposure. A change in the natural history of CD is not likely to occur with the currently available drugs when used in the setting of a conventional step-up algorithm.