Bobby Zhao Sheng Lo
© Bobby Zhao Sheng Lo
Prediction of complicated disease course for children newly diagnosed with Crohn’s disease: A multicentre inception cohort study
Kugathasan S, Denson LA, Walters TD, Kim MO, Marigorta UM, Schirmer M, et al.
Early-onset Crohn’s Disease (CD) displays different characteristics compared with CD in adults .Up to one-third of children with CD show progression in terms of disease location and/or presence of complicated disease within less than 10 years . Management of the disease is important as it covers a period including important developmental milestones, such as growth and puberty, and affects the quality of life . Various clinical, genetic, microbial and serological markers have been separately associated with complicated disease in paediatric CD but they have yet to be combined in a risk model [4,5]. The introduction of biological treatment offers hope that we will be able to change the natural history of paediatric CD through stratification of patients into low- or high-risk groups with respect to disease complications, a measure also necessary given the very high cost of treatment with biologics, which are therefore not used as first-line treatment [6,7].
The aim of this study was to derive a risk model based on clinical, host biological and microbial factors defined at diagnosis, and treatment including anti-tumour necrosis factor alpha (TNF-α) therapy.
A multicentre inception cohort from 28 sites in the United States and Canada included 913 paediatric CD patients with non-stricturing/non-penetrating disease. Within the first 90 days, 191 patients received anti-TNF-α therapy. Progression to stricturing disease was similar in patients with and patients without early use of anti-TNF-α treatment. However, progression to penetrating behaviour was reduced threefold in those who received anti-TNF-α early compared to those who did not.
CBir1 (anti-flagellin) was the only seropositive factor associated with stricturing complications, while older age, African-American race, ASCA-IgA and CBir1 seropositivity were associated with penetrating disease. This model showed a sensitivity of 66%, a specificity of 63%, a positive predictive value of 14% and a negative predictive value of 95%.
The authors also accessed the microbiota from CD patients and non-IBD controls from ileum, rectum and stool, managing to find genera associated with complicated disease. Unfortunately this sub-cohort differed from the whole cohort and therefore was not incorporated in the risk prediction model.
Gene expression was accessed from ileal biopsies, but considerable heterogeneity was found in the gene expression. The extracellular matrix gene signature was associated with stricturing behaviour. Extracellular matrix gene signature was noted in those who were at low risk of a complicated disease course but who progressed anyway. When gene signature was incorporated into the risk model, the sensitivity increased to 69%, the specificity to 71% and the positive predictive value to 24%; the negative predictive value was 94%. Interestingly, increased expression of genes involved in mitochondrial function (mitochondrial gene signature) was identified in patients at risk of complications who nevertheless remained complication free.
This study managed to construct a risk model for complicated disease course among children with CD. The authors demonstrated that early use of anti-TNF-α protects against a penetrating disease course but not against stricturing disease. Older age, African-American race, ileal disease location, ASCA, CBir1 and extracellular matrix gene signature were risk factors for complicated disease.
Even though this study investigated an inception cohort from 28 sites, it was not population based, meaning that it does not reflect the population incidence, a possible source of bias. Further, as the authors stated, imaging was not done according to a protocol and this may have led to variation in phenotyping of the disease, with underestimation of a complicated disease course.
It is noteworthy that the study was able to collect microbiota samples from the ileum, rectum and stool of both CD and non-IBD patients but unfortunately this was not done continuously and therefore could not be included in the risk model.
The authors failed to address the heterogeneity in the gene expression, which can have a detrimental effect if relevant variables are not incorporated into the model . Therefore the genetic results must be interpreted with caution.
The risk model will now be applied in clinical practice, with investigation of the efficiency of its predictive value. As the authors stated, the fact that the positive predictive value of the model is low but the negative predictive value is high could permit risk stratification at diagnosis, with identification of patients who do not need early intensive treatment and hopefully a consequent reduction in the treatment-related healthcare cost.
Furthermore, bacterial genera and other gene expressions, such as the mitochondrial gene signature, are yet to be incorporated into the model, and a future risk model could access the full spectrum of variables.
The study group is to be congratulated as this type of study is needed to identify future biomarkers and derive necessary risk models for complicated disease in both paediatric and adult patients with Inflammatory Bowel Diseases.
To summarise and conclude, this study managed to develop a risk model with high negative predictive value. The risk model has yet to be used in clinical practice to fully investigate its potential. The role of microbiota and other gene expressions is yet to be incorporated.
1. Ruel J, Ruane D, Mehandru S, Gower-Rousseau C, Colombel J-F. IBD across the age spectrum — is it the same disease? Nat Rev Gastroenterol Hepatol. 2013;11:88–98.
2. Duricova D, Fumery M, Annese V, Lakatos PL, Peyrin-Biroulet L, Gower-Rousseau C. The natural history of Crohn’s disease in children. Eur J Gastroenterol Hepatol. 2017;29:125–34.
3. Kunz JH, Hommel KA, Greenley RN. Health-related quality of life of youth with inflammatory bowel disease: A comparison with published data using the PedsQL 4.0 Generic Core Scales. Inflamm Bowel Dis. 2010;16:939–46.
4. Kovács M, Müller KE, Papp M, Lakatos PL, Csöndes M, Veres G. New serological markers in pediatric patients with inflammatory bowel disease. World J Gastroenterol. 2014;20:4873.
5. Thia KT, Sandborn WJ, Harmsen WS, Zinsmeister AR, Loftus EV. Risk factors associated with progression to intestinal complications of Crohn’s disease in a population-based cohort. Gastroenterology. 2010;139:1147–55.
6. Sprakes MB, Ford AC, Suares NC, et al. Costs of care for Crohn’s disease following the introduction of infliximab: a single-centre UK experience. Aliment Pharmacol Ther. 2010;32:1357–63.
7. van der Valk ME, Mangen MJ, Leenders M, et al. Healthcare costs of inflammatory bowel disease have shifted from hospitalisation and surgery towards anti-TNFα therapy: results from the COIN study. Gut. 2014;63:72–9.
8. Leek JT, Storey JD. Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 2007;3:1724–35.
|Bobby Zhao Sheng Lo is a research year student at The Gastrounit, Medical Division, Hvidovre Hospital, University of Copenhagen, Denmark. He is currently in the first year of his Master’s degree in Medical Science at the University of Copenhagen. His main interests are Inflammatory Bowel Disease and Irritable Bowel Syndrome with a focus on epidemiology and treatment and optimisation of the disease course with the aim of changing the natural history of Crohn's Disease and Ulcerative Colitis patients.||
Bobby Zhao Sheng Lo
© Bobby Zhao Sheng Lo