Prognostic biomarkers to identify patients likely to develop severe Crohn's disease: a systematic review

Notes

Permissions

Chapter 1 Background

The research detailed in this monograph is a response to the National Institute for Health Research (NIHR) Health Technology Assessment (HTA) programme call of 2014, HTA 14/210, ‘Prognostic markers in early Crohn’s disease’. Some of the text presented in this section has been published previously as part of the study protocol. 1

The commissioning brief stated that the NHS decision problem turned on the fact that:

Some patients with Crohn’s disease follow a relatively limited course with rapid response to relatively simple treatment and few subsequent acute flare-ups. Others progress rapidly to severe disease needing escalation of medical treatment or surgery: as many as one-third of patients require surgery within a year of beginning oral steroids. Rapid disease control may allow sufferers to return to work and other normal activities more rapidly and may reduce the need for surgery. NICE [National Institute for Health and Care Excellence] does not currently recommend early use of anti-TNF [tumour necrosis factor] drugs although there is some evidence of better early outcomes with their use. Rapid escalation of medical treatment is allowed for within current NICE guidelines. Identification of patients who might benefit from early intensive treatment is needed before the cost-effectiveness of treatment strategies can be determined.

Reproduced with permission, 2021, NIHR HTA programme2

Reproduced with permission, 2021, NIHR HTA programme

The HTA programme stipulated that the commissioned research should comprise a systematic review that evaluated ‘clinical criteria for treatment and/or use of biomarkers or biopsy informatics to develop a clinical tool to identify patients most likely to benefit from early intensive treatment’. Outcomes of the research would therefore be ‘[a]n overview of current evidence’ combined with ‘a signal as to which biomarkers may be most useful’. It was also noted that, contingent on the findings, any prediction tool that was developed during the research may be subsequently tested in a prospective study, in research commissioned by the HTA programme.

Chapter 2 Research questions

Our overarching research question was to identify biomarkers that are potentially able to predict which patients with Crohn’s disease are destined to subsequently develop severe disease. This question (and related others) was answered through the research objectives listed below.

Chapter 3 Review methods

At the design stage, we anticipated that there would be many potential biomarkers, which could make for an unwieldy review. We, therefore, set a priori quality/quantity thresholds for review inclusion to prevent extraction of data for biomarkers that have not been sufficiently studied for the findings to be deemed generalisable/reproducible and/or that have weak methodology. For example, at the time of writing the review protocol, more than 70 separate genes had been implicated in Crohn’s disease. 13 Given that genetic sequencing is currently very expensive and there are multiple potential individual genetic predictors, we anticipated that few genetic predictors will have been studied in sufficient depth. However, sequencing will probably become increasingly cost-effective in the near future. Our systematic review, therefore, considered only those genes for which sufficient primary studies exist to provide a useful signal of their prognostic potential.

Genetic makeup is also linked to response to biological therapy. Given that genetic makeup is fixed, these factors need to be measured only once (in contrast to other biomarkers that fluctuate with disease activity). There are also multiple antibody candidates, and prognostic strategies have focused on both titres of individual antibodies and the number of different antibodies. For example, patients with three or more positive antibodies are eight times more likely to need surgery than patients who are antibody negative. 14

Chapter 4 Studies included in the reviews

This review is reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines; the PRISMA flow chart is illustrated in Figure 1. In total, 29,947 abstracts were identified (PubMed, n = 11,292; EMBASE, n = 18,655). A total of 15,923 duplicates were removed and 14,024 abstracts were screened. After applying the eligibility criteria, 247 full-text articles were assessed, which included four of the biomarkers that were considered essential by the expert panel. Ultimately, 71 articles were included in the review, which corresponded to 56 non-overlapping patient cohorts. Reasons for exclusions are given in Figure 1.

FIGURE 1.

The PRISMA flow chart. Some articles were included in more than one review. a, Hand-searching of conference proceedings failed to yield any additional research.

Table 3 shows the predictors that were deemed important for meta-analysis by the expert panel. Ultimately, this involved the addition of only one predictor: CRP. Nine predictors that were deemed important were already included. Seven predictors that were deemed important yielded too few individual studies to be viable for meta-analysis. For example, there were no prognostic studies of faecal calprotectin. ‘Severe endoscopic lesions’ were deemed important but prognostic data were provided by only two studies.

Ultimately, the number of papers included in the three reviews were as follows:

• general overview – 58 papers (including 11 contained in the genetic review and six in the serological review)

• genetic review – 17 papers (including 11 contained in the general overview and two in the serological review)

• serological review – 13 papers (including six contained in the general overview and two in the genetic review).

Chapter 5 Results of the reviews

Most of the studies were European (37/71, 52%), with 14 (20%) from the USA/Canada. A total of 40 (56%) studies were multicentre (Table 4). In total, 36 (51%) studies were prospective, 33 (46%) were retrospective and two (3%) were unclear. Of 71 studies, 11 (16%) were paediatric only, 23 (32%) were adults only and 37 (52%) were mixed. Individual study characteristics are shown in Table 4.

The recruitment dates varied (see Table 4), with 22 (31%) studies not reporting recruitment dates. Follow-up also varied (median 8 years, IQR 5–10 years, range 0.8–18 years; Table 4). Nine studies allowed extraction of prediction of severe disease events at multiple time points during follow-up. 25–33 Studies frequently presented several definitions of severe disease (Table 5).

The forest plots for predictors indicate data characteristics, including event, paediatric versus adult versus mixed population, and whether or not patients with severe disease were included at baseline. For each predictor meta-analysis, we ensured that patients were included in analyses only once. Figure 2 presents the ROB summarised across all studies.

FIGURE 2.

Risk of bias summarised across all studies.

Most of the studies were rated as having high ROB in at least one domain. Accordingly, ‘Overall ROB’ was rated as ‘high’ in 65 (92%) studies. Only three studies were rated as having ‘low’59,89,91 ROB and three were rated as ‘unclear’. 28,58,79 Concern regarding ‘Overall applicability’ was rated as ‘low’ or ‘unclear’, except for five studies. 30,46,60,67,73 ROB and applicability for individual studies are presented in Box 1 and Figures 3–5.

FIGURE 3.

Risk-of-bias and applicability concerns summary: review of author judgements regarding each domain for each individual study in the clinical review. Orange circles indicate high ROB; blue circles indicate low ROB; purple circles indicate uncertain ROB.

FIGURE 4.

Risk-of-bias and applicability concerns summary: review of author judgements regarding each domain for each individual study in the serological review. Orange circles indicate high ROB; blue circles indicate low ROB; purple circles indicate uncertain ROB.

FIGURE 5.

Risk-of-bias and applicability concerns summary: review of author judgements regarding each domain for each individual study in the genetic review. Orange circles indicate high ROB; blue circles indicate low ROB; purple circles indicate uncertain ROB.

Overall, we identified 12 individual predictors eligible for inclusion: three serological, one genetic and eight clinical. There were no radiological, endoscopic or histological predictors that met our criteria for inclusion in the review.

Paediatric data are presented in the forest plots, but there were insufficient for subgroup analysis.

Table 6 shows the summary of findings across the meta-analyses performed.

Meta-analysis of clinical predictors

The following clinical predictors had prognostic data available from fewer than five studies and, therefore, did not meet the inclusion criteria for the review: submucosal plexitis, fever, weight loss, poor growth, medication, upper disease, Jewish ethnicity, joint problems, abdominal pain, oral contraception (women), ethnic origin, systematic manifestations, systematic steroid use, azathioprine/biologicals, granuloma, depression, cancer, diarrhoea, time to diagnosis from symptom onset, severe endoscopic lesions, visceral fat area, rectal bleeding, fatigue, use of total parenteral nutrition, bowel stenosis, internal fistula, alternative initial diagnosis, abscess, stricture, development of strictures during follow-up, development of fistulae during follow-up, initial diagnosis at surgery, steroids per year, flares per year, married/common law, low conscientiousness, high neuroticism, low adherence to medication, adverse childhood experience, childhood sexual abuse, childhood physical abuse, high childhood adversity score, asthma, eczema, glandular fever, kidney stones, liver disease, mental illness, bronchiectasis, tonsillectomy, chole, grommet, immunised against measles, immunised against mumps, immunised against tuberculosis, antibiotic consumption, medication consumption, breastfed, alcohol, vegetarian, takeaways, public swimming, sand pit, farm, shared bedroom, continuous course, frequent relapse, symptoms at diagnosis, symptoms at 1 year, persistent mucosal erosions, jejunal involvement, Crohn’s Disease Activity Index (CDAI), educational level, type of centre, nausea/vomiting, diabetes mellitus, coronary artery disease and hypertension.

We were able to meta-analyse seven clinical markers from 58 studies. These were Montreal disease behaviour, age, disease duration, disease location, smoking, sex and family history.

A meta-analysis of 12 studies (4376 participants, 1551 events) found that B2 stricturing, B3 fistulating and either severe or disabling disease predicted severe disease more powerfully than B1 inflammation alone (see Table 6 and Figure 6). B2 and B3 Montreal disease behaviours were the strongest predictors of subsequent severe disease in our review (B2: OR 4.09, 95% CI 2.59 to 6.48; B3: OR 6.25, 95% CI 3.68 to 10.63).

FIGURE 6.

Forest plot of unadjusted ORs investigating Montreal disease behaviour as predictive of severe Crohn’s disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

The participants’ age at diagnosis was examined in 43 studies (19,623 participants, 7010 events). Most of the studies categorised age using the Montreal/Vienna thresholds, so data were greatest for three groups: aged < 17 years, aged 17–40 years and aged > 40 years. Overall, diagnosis at < 17 years of age was associated with a lower risk of severe Crohn’s disease than other ages (see Table 6 and Figures 7–9). Disease duration was examined in 14 studies (8690 participants, 1714 events). A meta-analysis found that increased duration of disease was associated with significant risk of severe disease at all durations examined (see Table 6 and Figures 7–9).

FIGURE 7.

Forest plot of unadjusted ORs investigating age and disease duration as predictive markers for severe Crohn’s disease: age at diagnosis < 17 years compared with > 17 years, 17–40 years and > 40 years. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

FIGURE 8.

Forest plot of unadjusted ORs investigating age and disease duration as predictive markers for severe Crohn’s disease: age at diagnosis 17–40 years compared with > 40 years. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

FIGURE 9.

Forest plot of unadjusted ORs investigating age and disease duration as predictive markers for severe Crohn’s disease: duration of disease, prognostic studies only. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Disease location was examined in 32 studies (10,877 participants, 4193 events). Studies were diverse regarding the segments and/or segment combinations described. We could analyse ‘colonic disease alone’ and ‘any colonic disease’ versus other locations. Overall, colonic disease alone conferred significantly lower surgery risk than other locations (OR 0.42, 95% CI 0.31 to 0.58) (see Table 6 and Figure 10).

FIGURE 10.

Forest plot of unadjusted ORs investigating disease location as predictive of severe Crohn’s disease: colonic disease only. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Similarly, any colonic disease (23 studies, 7373 participants, 3086 events) predicted lower surgical or B2/B3 risk than no colonic disease (see Table 6 and Figure 11).

FIGURE 11.

Forest plot of unadjusted ORs investigating disease location as predictive of severe Crohn’s disease: any colonic involvement. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Perianal disease (24 studies, 13,483 participants, 5510 events) was associated with a significantly increased risk of subsequent severe disease overall (1.84 OR, 95% CI 1.29 to 2.62) (see Table 6 and Figure 12).

FIGURE 12.

Forest plot of unadjusted ORs investigating disease location as predictive of severe Crohn’s disease: perianal disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

A total of 34 studies reported an association of smoking with severe disease (11,475 participants, 5097 events); a meta-analysis of 26 studies found that current smoking increased the risk of severe Crohn’s disease significantly (OR 1.53, 95% CI 1.30 to 1.79) (see Table 6 and Figure 13).

FIGURE 13.

Forest plot of unadjusted ORs investigating smoking as a predictive marker for severe Crohn’s disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Sex was examined in 35 studies (14,489 participants, 5350 events) and, overall, was not a significant predictor of severe Crohn’s disease (see Table 6 and Figure 14).

FIGURE 14.

Forest plot of unadjusted ORs investigating sex as a predictive marker for severe Crohn’s disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Family history was examined in 18 studies (5687 participants, 1413 events). A meta-analysis found no consistent direction and no significant association of family history and severe Crohn’s disease (OR 1.05, 95% CI 0.81 to 1.36) (see Table 6 and Figure 15).

FIGURE 15.

Forest plot of unadjusted ORs investigating family history as a predictive marker for severe Crohn’s disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Meta-analysis of serological predictors

The following serological predictors had prognostic data available from fewer than five studies and, therefore, were not subject to meta-analysis: thrombocytosis, faecal calprotectin, anti-Saccharomyces cerevisiae antibodies (ASCA)-immunoglobulin (Ig)G, anti-CBir1, white blood cell (WBC), albumin, erythrocyte sedimentation rate (ESR), platelets, anaemia, CRP, haematocrit, ASCA-IgA, ferritin, anti-Fla2, anti-FlaX, anti-L IgA, anti-C IgA, anti-Saccharomyces chitobioside antibody (ACCA)-IgA, anti-Saccharomyces laminaribioside antibody (ALCA)-IgG, anti-Saccharomyces mannobioside antibody (AMCA)-IgG and anti-I2.

Three serological markers (ASCA, anti-CBir1 and CRP) from 13 studies were meta-analysed. ASCA and anti-CBir1 showed potential prognostic association; ASCA-positive patients had significantly increased odds of developing severe disease (OR 2.29, 95% CI 1.31 to 3.99; six studies) (see Table 6 and Figure 16).

FIGURE 16.

Forest plot of unadjusted ORs investigating serological markers as predictors of severe Crohn’s disease: ASCA. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Anti-CBir1 data were identified from five studies; meta-analysis from ORs did not reach significance (see Table 6 and Figure 17), but additional evidence from hazard ratios indicated a significant association between anti-CBir1 and severe Crohn’s disease (Table 7).

FIGURE 17.

Forest plot of unadjusted ORs investigating serological markers as predictors of severe Crohn’s disease: anti-CBir1. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

TABLE 7 - Summary of the data not presented in forest plots

Adj, adjusted; HR, hazard ratio; MA, meta-analysis; NR, not reported; NS, not significant.

95% CI is as reported in the original publication.

Predictor	Study (first author and year)	Definition of serious disease	Number of participants	Number of events	Outcome	Significant	Point estimate (95% CI)
ASCA	Annese 200536	B2	316	NR	p-value	Yes	0.046
	Dubinsky 200845	Surgery	592	61	HR	Yes	3.2 (1.1 to 9.5)
	Loly 200820	Other	73	NR	p-value	No	NS
	Siegel 201685	B2/B3/surgery	243	142	HR	Yes	1.42 (1.24 to 1.63)
CBir-1	Dubinsky 200845	B2/B3	536	37	HR	Yes	2.5 (1.2 to 5.2)
	Siegel 201685	B2/B3/surgery	243	142	HR	Yes	1.47 (1.24 to 1.75)
CRP	All studies in MA
NOD2	Shaoul 200983	Surgery	119	38	p-value	No	NS
Disease behaviour (B3)	Vester-Andersen 201492	Surgery	213	49	p-value	No	NR
	Park 201370	Surgery	1403	471	Adj HR	Yes	5.67 (4.51 to 7.11)
Disease behaviour (NR)	Brant 200337	Surgery	257	183	p-value	NR	NR
Disease behaviour (B2/B3)	Nunes 201364	Surgery	3118	1269	Adj HR	Yes	4.42 (3.87 to 5)
Age at diagnosis	Annese 200536	Surgery	316	87	Adj OR	Yes	1.37 (1.05 to 1.79)
Duration of disease	Renda 200875	Surgery	182	110	HR	No	0.9 (0.9 to 1.02) at 1.5 years
	Goel 201346	B2/B3	223	93	p-value	No	> 0.3 at 6 years
Location: colonic only	Malmborg 201559	B2/B3/surgery	161	25	HR	No	0.72 (0.33 to 1.59)
	Aldhous 200734	Surgery	251	113	Adj HR	Yes	0.27 (0.16 to 0.45)
	Park 201370	Surgery	1403	471	Adj HR	Yes	0.36 (0.2 to 0.64)
Location: any colonic	Papi 200569	B3	139	47	Adj HR	No	0.7 (0.3 to 1.6)
	Annese 200536	Surgery	316	87	p-value	No	NS
Perianal	Annese 200536	Surgery	316	87	p-value	No	NS
	Brant 200337	Surgery	257	183	p-value	No	NS
	Chen 201541	Surgery	197	64	HR	No	0.68 (0.32 to 1.42)
	Law 201357	Other	79	34	p-value	No	NS
	Nasir 201362	Surgery	503	240	Adj OR	Yes	2.84 (1.83 to 4.38)
	Nunes 201364	Surgery	3201	1313	Adj HR	Yes	1.32 (1.17 to 1.5)
	Park 201370	Surgery	1403	471	Adj HR	No	1.09 (0.91 to 1.32)
	Ryan 201378	Surgery	182	77	p-value	No	0.06
	Siegel 201685	B2/B3/surgery	243	142	HR	No	0.86 (0.54 to 1.37)
	Tarrant 200888	B2/B3	715	189	HR	Yes	1.62 (1.28 to 2.05)
	Vernier-Massouille 200891	Surgery	394	176	HR	No	0.94 (0.53 to 1.65)
Smoking	Lovasz 201327	B2/B3	287	110	HR	No	1.48 (0.96 to 2.37)
	Law 201357	Other	79	34	Adj HR	Yes	4.68 (1.03 to 4.09)a
	Brant 200337	Surgery	257	183	p-value	No	NS
	Goel 201346	Surgery	223	73	p-value	No	> 0.3
	Ryan 201378	Surgery	182	77	p-value	No	0.05
	Vester-Andersen 201492	Surgery	213	49	p-value	No	NS
	Annese 200536	Surgery	316	87	Adj OR	Yes	1.42 (1.06 to 1.88)
	Papi 200569	B3	139	47	Adj OR	No	1.2 (0.5 to 2.6)
Sex	Nasir 201363	Surgery	503	240	Adj OR	No	1.16 (0.77 to 1.74)
	Papi 200569	B3	139	47	Adj OR	No	1.67 (0.33 to 2.17)
	Gupta 200647	Surgery	989	128	HR	No	1.42 (1.00 to 2.01)
	Malmborg 201559	B2/B3/surgery	161	25	HR	No	0.59 (0.31 to 1.15)
	Siegel 201184	B2/B3	579	67	HR	No	0.55 (0.3 to 1.01)
	Charpentier 201439	Surgery	367	103	HR	No	1.3 (0.8 to 1.9)
	Vernier-Massouille 200891	Surgery	394	176	HR	No	0.96 (0.71 to 1.3)
	Aldhous 200734	B2/B3	274	105	Adj HR	No	0.89 (0.6 to 1.3)
	Annese 200536	Surgery	316	87	p-value	No	NS
	Brant 200337	Surgery	257	183	p-value	NR	NR
	Law 201357	Other	79	34	p-value	No	NS
	Ryan 201378	Surgery	182	77	p-value	No	NS
Family history	Papi 200569	B3	139	47	Adj OR	No	2 (0.5 to 7.6)
	Malmborg 201559	B2/B3/surgery	161	17	HR	No	0.21 (0.03 to 1.56)
	Aldhous 200734	Surgery	251	113	Adj HR	No	1.06 (0.65 to 1.73)
	Brant 200337	Surgery	257	183	p-value	No	NS
	Ryan 201378	Surgery	182	77	p-value	No	NS
	Tarrant 200888	B2/B3	715	85	p-value	No	NS

For CRP, data were identified from only three studies (the expert panel requested that CRP was included) and the results found no significant predictive association of CRP and severe Crohn’s disease (see Table 6 and Figure 18).

FIGURE 18.

Forest plot of unadjusted ORs investigating serological markers as predictors of severe Crohn’s disease: CRP. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Meta-analysis of genetic predictors

The following genetic predictors had prognostic data available from fewer than five studies and, therefore, were not subject to meta-analysis: LOC441108, TNFSF15, 5p13.1, NCF4, CX3CR1, JAK2, SBNO2, ZPBP, PTGER4, PUS10, PRDM1, C13ORF31, SLC22A23, TAB2/MAP3K7IP2, PTPN22, ICOSLG, STAT3, PTPN2, NKX2-3, POU2E1, U10, U7, AK097548, CDKAL1, HERC2, ATG4A, NALP3, IL21, CARD8, nucleotide-binding oligomerisation domain-containing protein 2 (NOD2) – all 3 versus 0, TLR4 D299G, FOXO3a, IBD5, DLG5, PAI-1, SMAD3, MMP, TIMP, ATG2A, FNBP1L and ATG4D.

Of the genetic markers, results for NOD2 in at least one variant were reported in 17 studies, although one study reported only p-values and was excluded from meta-analysis. 83 Meta-analysis of 16 studies (5407 participants, 2645 events) suggested higher risk of severe Crohn’s disease with a NOD2 variant gene overall (OR 1.69, 95% CI 1.43 to 2.00); studies were a mix of prognosis and prognosis/diagnosis (see Table 6 and Figure 19).

FIGURE 19.

Forest plot of unadjusted ORs investigating NOD2 (any variant) as a predictor of severe Crohn’s disease. A, adult; AC, both adults and children; C, child; P, predictive effect only; P + D, predictive and diagnostic effect combined.

Chapter 6 Discussion

The research detailed in this monograph is a response to the NIHR HTA programme call of 2014, HTA 14/210, ‘Prognostic markers in early Crohn’s disease’. The outcomes stipulated by the HTA programme in their commissioning brief were ‘[a]n overview of current evidence’ and ‘[a] signal as to which biomarkers may be most useful’. Although a large amount of existing research has investigated the potential of a multitude of biomarkers to predict severe Crohn’s disease, individual studies are usually small, single centre and restricted to one or a handful of biomarkers. Furthermore, results from individual studies often diverge, providing no clarity. The consequence is that individual clinicians cannot synthesise these data in a way that is meaningful for individual patients. In response to the brief, we have carried out a comprehensive overview across all biomarker types, the findings of which are detailed in this report. Although we identified seven previous systematic reviews that investigated associations between severe Crohn’s disease and mostly serological and some genetic biomarkers, many confounded biomarker associations with both current disease (i.e. a diagnostic role) and future disease (i.e. a prognostic or predictive role), with no distinction between the two roles.

To generate meaningful results, we stipulated an a priori criterion that excluded biomarkers with insufficient evidence (defined by us as fewer than five component primary studies), and considered meta-analysis only where data for an individual biomarker could be extracted from three or more studies. Given the potential literature – we identified nearly 30,000 abstracts – we encountered relatively few genuinely prognostic studies, that is studies that investigated the development of future severe disease. We were surprised that common biomarkers, such as CRP, white blood cell count and haemoglobin, were rarely investigated despite being in widespread clinical use. We found no endoscopic, radiological or histopathological biomarkers that were sufficiently examined for meaningful meta-analysis. Although we anticipated an inevitable delay between validation of new biomarkers and subsequent prognostic studies, we were surprised that well-established biomarkers were poorly researched. This dearth of research may reflect researcher preoccupation with novel biomarkers, publication bias towards novel biomarkers or bias against studies finding no benefit.

Ultimately, across all biomarker types we identified only 71 individual studies with prognostic data, representing just 56 non-overlapping patient cohorts. From these, this report presents a meta-analysis of 11 potential predictors, of which eight displayed some evidence of predictive utility following meta-analysis. We present results for NOD2 (any variant), the genetic predictor quoted in clinical guidelines as being associated with an increased risk of severe disease. We present a meta-analysis of three serological biomarkers, finding both ASCA-positive and anti-CBir1-positive patients to be at a higher risk of severe disease. It may be that biomarker levels vary too widely to be useful; acute phase reactant levels probably represent the patient status at diagnosis, for example during a flare. The levels of antimicrobial antigens at subsequent times remain uncertain. Although studies with paired sera at 5-year intervals imply stability, others suggest that levels of antimicrobial antigens change with disease progression and treatment. We excluded research without baseline serum draw. To have clinical utility, biomarkers must be prognostic in early disease.

We meta-analysed seven clinical predictors: Montreal behaviour, age, disease duration, disease location, smoking, sex and family history. Five of these showed prognostic potential (sex and family history did not). The strongest association with subsequent severe disease across the entire review was found for Montreal B2 and B3 categories, with ORs of 4.09 and 6.25, respectively. Meta-analysis also suggested that increased risk of developing severe disease was associated with perianal disease, longer disease duration and smoking. We found that the risk of subsequent severe disease was potentially decreased in patients who were diagnosed at a younger age and in those with any colonic disease or colonic disease alone.

Our review does have limitations, the majority of which are contingent on the quality of the primary component studies and difficulties with definitions of severe disease. As noted already, we found that true prognostic research was relatively scarce and was often reported confounded with diagnosis. Moreover, methods were heterogeneous and we were obliged to meta-analyse across different methods of biomarker measurement, different positivity thresholds and different follow-up durations. Such variability probably underlies the disparity between study effect estimates displayed, as seen, for example, in the ASCA forest plot.

Because data were heterogeneous, we stress that interpretation of our findings should focus on which biomarkers were found to have predictive potential rather than the precise strength of that prediction. Readers should not draw conclusions regarding the effectiveness of different predictors given that our estimates are based on results from different studies. Particularly for serological markers, the number and quality of studies are presently such that findings from future studies will be important; the majority of genetic and serological biomarkers that were identified were examined by insufficient studies for meaningful meta-analysis.

Furthermore, results from individual studies probably differ because of varying patient populations, study designs or outcome definitions. For example, we were obliged to use a range of definitions for ‘severe’ disease because there is no single universally accepted/reported definition. Where the measure is relatively inclusive, for example Beaugerie et al. ,18 around 60% of patients will have ‘disabling disease’ within 5 years. Conversely, although the Liege criteria define severity more precisely,20 no article in our review stated the extent of intestinal surgery with detail sufficient to allow post hoc classification; single, early ileocaecal resection may be curative. Subsequently, only Loly et al. 20 (who proposed the Liege criteria) published data using this end point.

It is also possible that patients undergoing complex, expensive serological testing at diagnosis may be spectrum-biased towards severe disease. It is self-evident that biomarkers predictive of future severe disease have no clinical utility when such disease is established at diagnosis. For example, we were careful not to use Montreal B3 at baseline to predict development of B3 disease. B1, B2 and B3 are used mostly as predictors of surgery. We found that young age was not predictive of severe disease. However, children often display a panenteric phenotype that precludes surgical therapy, even when disease is severe.

We found considerable ROB, mostly because of suspected reporting bias as studies reported fewer than three standard predictors with non-significant results and results were not reported for standard follow-up times. Accordingly, no results were available for follow-up extending 5 to 10 years. Our analyses included some retrospective studies recruiting before 2004. Current treatment paradigms are evolving rapidly, and it is unclear whether or not biomarkers will behave similarly pre- and post-biologic introduction; we present individual study results with information sufficient for readers to explore this. Likewise, it was beyond our remit to consider anti-TNF response/failure to respond/loss of response. Recent studies have found that this in itself may identify patients destined to develop severe disease. 182,183 Predictive studies of multigene assays are currently under way, but we identified insufficient existing research for their inclusion in this review.

Owing to the large number of citations identified, we were unable to perform initial independent abstract screening by two reviewers; Darren Boone screened all abstracts alone. However, any uncertainty regarding eligibility, no matter how minor, was raised with Lucinda Archer (a statistician) initially and the with other members of the team if any uncertainty persisted. However, it is possible that some relevant data were discarded on the initial sift by using a single researcher. All full-text articles retrieved were screened independently by Sue Mallett (the senior statistician) to verify that data extraction for meta-analysis was correct, in particular to make the distinction between diagnostic and prognostic material.

This work has taken a considerable amount of time and resource, which was largely because of the nature of the prognostic reviews in diagnosis. Diagnostic studies remain poorly and imprecisely indexed in comparison with studies of therapeutic interventions. This problem is confounded further when the aim is to separate diagnostic from prognostic information, which is often confounded within the same article (as explained elsewhere). The result is that search strings must be inclusive to avoid missing important data, but this inflates the amount of primary research that must be examined. We carried out an updated search in August 2020 to attempt to quantify the number of potentially relevant data subsequent to our original search. We identified an additional 4005 indexed articles in PubMed and 6878 in EMBASE, yielding 80 potentially relevant full-text papers. We estimate that around 23 of these would ultimately yield useful prognostic data, the large majority dealing with clinical biomarkers. Accordingly, we do not believe that our research has missed important serological or genetic biomarker research and we believe that our findings are likely to stand overall.

In summary, we carried out a systematic review of biomarkers potentially predictive of subsequent severe Crohn’ disease, across all biomarker types, and focused on those contained in clinical guidelines. We found that prognostic research was heterogeneous and at a high ROB, and relatively few biomarkers relevant to clinical guidelines were investigated sufficiently for meta-analysis. Of those that were, we found evidence of prognostic potential for two serological (ASCA and anti-CBir1), one genetic (NOD2) and five clinical biomarkers (Montreal behaviour, age, disease duration, disease location and smoking). Considerably more prognostic research in this area is required, with methodological rigour, informed by consensus agreement around definitions of severe disease and following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guidelines,184 so that study results can be used by other researchers.

Chapter 7 Conclusions and recommendations for future research

We can make two broad conclusions and recommendations that arise from the research presented in this monograph. The first is that the quality and extent of prognostic research to predict future disease severity is somewhat lacking. We recommend that funders consider commissioning studies that are clearly prognostic and methodologically sound, and that adhere to established reporting guidelines for prognostic research. Because of the evidence gaps we identified, any such research should investigate all potentially important predictors, not just those that are perceived as ‘cutting-edge’ and/or ‘novel’.

The second recommendation stems from the predictors identified by our review and meta-analysis; in the face of the limitations that we describe, we have identified a limited number of relatively easily accessible predictors that could be combined and tested in a multivariable prognostic model. This model would aim to accurately predict those patients with a new diagnosis of Crohn’s disease who are destined to develop severe disease in the future, thereby facilitating individualised, early, targeted biological therapy. At the time of writing, we are currently developing and validating such a model using historic data obtained from the Royal Devon and Exeter NHS Trust; this work has been interrupted by the 2020 covid-19 pandemic that required clinical academics to return to full-time clinical work. It is our intention to complete this work when possible. If that validation proves sufficiently promising, we will ask that funders consider commissioning an external validation and/or clinical trial. Although the cleanest methodology would be to randomise application of the predictive model, this may prove ethically difficult given the widespread use of biological therapies currently. Due consideration should, therefore, be given to use of historic databases.

As noted elsewhere, because the existing literature is deficient it is likely that we have ‘missed’ some potentially important predictors simply because they have not been assessed sufficiently at the time of writing. As a result, development of a multivariable model should be flexible and not be restricted to the predictors identified by us. Development and validation could also incorporate multigene arrays, which are garnering considerable current interest: their incremental benefit (or otherwise) can then be quantified by their effect on prediction when added as one of the variables within an existing prognostic model based on more easily accessible data.

The work described in this monograph has not had substantial PPI input because it is a review of existing research performed by others. By contrast, we will seek PPI input from our expert patient collaborator when we have completed our planned work on the predictive model. Specifically, to inform clinical implementation of any model, we need to understand, from a patient perspective, how the presumed benefits of early biological therapy might be weighed against potential harmful side effects. These considerations must be made within the context of predictive accuracy, namely the degree to which the model may deny biological therapy to some deserving patients while simultaneously over-treating others.

Acknowledgements

The authors are grateful to Professor Brian Saunders, Professor Simon Travis, Dr John Hamlin and Dr Andrew Millar for their help with the expert panel.

The authors are grateful to the UK NIHR HTA programme for funding this research (project number: HTA 14/210/07).

References

1. Halligan S, Boone D, Bhatnagar G, Ahmad T, Bloom S, Rodriguez-Justo M, et al. Prognostic biomarkers to identify patients destined to develop severe Crohn’s disease who may benefit from early biological therapy: protocol for a systematic review, meta-analysis and external validation. Syst Rev 2016;5. https://doi.org/10.1186/s13643-016-0383-5.
2. National Institute for Health Research (NIHR) Health Technology Assessment (HTA) programme . HTA No. 14 210: Prognostic Markers in Early Crohn’s Disease 2014. https://njl-admin.nihr.ac.uk/document/download/2023257 (accessed March 2021).
3. Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002;359:1541-9. https://doi.org/10.1016/S0140-6736(02)08512-4.
4. Colombel JF, Sandborn WJ, Reinisch W, Mantzaris GJ, Kornbluth A, Rachmilewitz D, et al. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med 2010;362:1383-95. https://doi.org/10.1056/NEJMoa0904492.
5. D’Haens G, Baert F, van Assche G, Caenepeel P, Vergauwe P, Tuynman H, et al. Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn’s disease: an open randomised trial. Lancet 2008;371:660-7. https://doi.org/10.1016/S0140-6736(08)60304-9.
6. Baert F, Moortgat L, Van Assche G, Caenepeel P, Vergauwe P, De Vos M, et al. Mucosal healing predicts sustained clinical remission in patients with early-stage Crohn’s disease. Gastroenterology 2010;138:463-8. https://doi.org/10.1053/j.gastro.2009.09.056.
7. Feagan BG, Panaccione R, Sandborn WJ, D’Haens GR, Schreiber S, Rutgeerts PJ, et al. Effects of adalimumab therapy on incidence of hospitalization and surgery in Crohn’s disease: results from the CHARM study. Gastroenterology 2008;135:1493-9. https://doi.org/10.1053/j.gastro.2008.07.069.
8. Khanna R, Bressler B, Levesque BG, Zou G, Stitt LW, Greenberg GR, et al. Early combined immunosuppression for the management of Crohn’s disease (REACT): a cluster randomised controlled trial. Lancet 2015;386:1825-34. https://doi.org/10.1016/S0140-6736(15)00068-9.
9. Peyrin-Biroulet L, Fiorino G, Buisson A, Danese S. First-line therapy in adult Crohn’s disease: who should receive anti-TNF agents?. Nat Rev Gastroenterol Hepatol 2013;10:345-51. https://doi.org/10.1038/nrgastro.2013.31.
10. Biomarkers Definitions Working Group . Biomarkers and surrogate endpoints: preferred definitions and conceptual frameworks. Clin Pharmacol Ther 2001;69:89-95. https://doi.org/10.1067/mcp.2001.113989.
11. Blonski W, Buchner AM, Lichtenstein GR. Clinical predictors of aggressive/disabling disease: ulcerative colitis and crohn disease. Gastroenterol Clin North Am 2012;41:443-62. https://doi.org/10.1016/j.gtc.2012.01.008.
12. Waugh N, Cummins E, Royle P, Kandala NB, Shyangdan D, Arasaradnam R, et al. Faecal calprotectin testing for differentiating amongst inflammatory and non-inflammatory bowel diseases: systematic review and economic evaluation. Health Technol Assess 2013;17. https://doi.org/10.3310/hta17550.
13. Franke A, McGovern DP, Barrett JC, Wang K, Radford-Smith GL, Ahmad T, et al. Genome-wide meta-analysis increases to 71 the number of confirmed Crohn’s disease susceptibility loci. Nat Genet 2010;42:1118-25. https://doi.org/10.1038/ng.717.
14. Mow WS, Vasiliauskas EA, Lin YC, Fleshner PR, Papadakis KA, Taylor KD, et al. Association of antibody responses to microbial antigens and complications of small bowel Crohn’s disease. Gastroenterology 2004;126:414-24. https://doi.org/10.1053/j.gastro.2003.11.015.
15. Moons KG, de Groot JA, Bouwmeester W, Vergouwe Y, Mallett S, Altman DG, et al. Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLOS Med 2014;11. https://doi.org/10.1371/journal.pmed.1001744.
16. Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005;19:5A-36A. https://doi.org/10.1155/2005/269076.
17. Levine A, Griffiths A, Markowitz J, Wilson DC, Turner D, Russell RK, et al. Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification. Inflamm Bowel Dis 2011;17:1314-21. https://doi.org/10.1002/ibd.21493.
18. Beaugerie L, Seksik P, Nion-Larmurier I, Gendre JP, Cosnes J. Predictors of Crohn’s disease. Gastroenterology 2006;130:650-6. https://doi.org/10.1053/j.gastro.2005.12.019.
19. Satsangi J, Silverberg MS, Vermeire S, Colombel JF. The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut 2006;55:749-53. https://doi.org/10.1136/gut.2005.082909.
20. Loly C, Belaiche J, Louis E. Predictors of severe Crohn’s disease. Scand J Gastroenterol 2008;43:948-54. https://doi.org/10.1080/00365520801957149.
21. National Institute for Health and Care Excellence (NICE) . Infliximab and Adalimumab for the Treatment of Crohn’s Disease (TA187) 2010.
22. Teml A, Kratzer V, Schneider B, Lochs H, Norman GL, Gangl A, et al. Anti-Saccharomyces cerevisiae antibodies: a stable marker for Crohn’s disease during steroid and 5-aminosalicylic acid treatment. Am J Gastroenterol 2003;98:2226-31. https://doi.org/10.1111/j.1572-0241.2003.07673.x.
23. Wolff RF, Moons KGM, Riley RD, Whiting PF, Westwood M, Collins GS, et al. PROBAST: a tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med 2019;170:51-8. https://doi.org/10.7326/M18-1376.
24. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177-88. https://doi.org/10.1016/0197-2456(86)90046-2.
25. Israeli E, Ryan JD, Shafer LA, Bernstein CN. Younger age at diagnosis is associated with panenteric, but not more aggressive, Crohn’s disease. Clin Gastroenterol Hepatol 2014;12:72-9.e1. https://doi.org/10.1016/j.cgh.2013.06.027.
26. Louis E, Michel V, Hugot JP, Reenaers C, Fontaine F, Delforge M, et al. Early development of stricturing or penetrating pattern in Crohn’s disease is influenced by disease location, number of flares, and smoking but not by NOD2/CARD15 genotype. Gut 2003;52:552-7. https://doi.org/10.1136/gut.52.4.552.
27. Lovasz BD, Lakatos L, Horvath A, Szita I, Pandur T, Mandel M, et al. Evolution of disease phenotype in adult and pediatric onset Crohn’s disease in a population-based cohort. World J Gastroenterol 2013;19:2217-26. https://doi.org/10.3748/wjg.v19.i14.2217.
28. Pittet V, Rogler G, Michetti P, Fournier N, Vader JP, Schoepfer A, et al. Penetrating or stricturing diseases are the major determinants of time to first and repeat resection surgery in Crohn’s disease. Digestion 2013;87:212-21. https://doi.org/10.1159/000350954.
29. Smith BR, Arnott ID, Drummond HE, Nimmo ER, Satsangi J. Disease location, anti-Saccharomyces cerevisiae antibody, and NOD2/CARD15 genotype influence the progression of disease behavior in Crohn’s disease. Inflamm Bowel Dis 2004;10:521-8. https://doi.org/10.1097/00054725-200409000-00005.
30. Solberg IC, Cvancarova M, Vatn MH, Moum B. Risk matrix for prediction of advanced disease in a population-based study of patients with Crohn’s Disease (the IBSEN study). Inflamm Bowel Dis 2014;20:60-8. https://doi.org/10.1097/01.mib.0000436956.78220.67.
31. Van Limbergen J, Russell RK, Drummond HE, Aldhous MC, Round NK, Nimmo ER, et al. Definition of phenotypic characteristics of childhood-onset inflammatory bowel disease. Gastroenterology 2008;135:1114-22. https://doi.org/10.1053/j.gastro.2008.06.081.
32. Yang CH, Ding J, Gao Y, Chen X, Yang ZB, Xiao SD. Risk factors that predict the requirement of aggressive therapy among Chinese patients with Crohn’s disease. J Dig Dis 2011;12:99-104. https://doi.org/10.1111/j.1751-2980.2011.00484.x.
33. Zabana Y, Garcia-Planella E, Van Domselaar M, Mañosa M, Gordillo J, López San Román A, et al. Does active smoking really influence the course of Crohn’s disease? A retrospective observational study. J Crohns Colitis 2013;7:280-5. https://doi.org/10.1016/j.crohns.2012.03.020.
34. Aldhous MC, Drummond HE, Anderson N, Smith LA, Arnott ID, Satsangi J. Does cigarette smoking influence the phenotype of Crohn’s disease? Analysis using the Montreal classification. Am J Gastroenterol 2007;102:577-88. https://doi.org/10.1111/j.1572-0241.2007.01064.x.
35. Alvarez-Lobos M, Arostegui JI, Sans M, Tassies D, Plaza S, Delgado S, et al. Crohn’s disease patients carrying Nod2/CARD15 gene variants have an increased and early need for first surgery due to stricturing disease and higher rate of surgical recurrence. Ann Surg 2005;242:693-700. https://doi.org/10.1097/01.sla.0000186173.14696.ea.
36. Annese V, Lombardi G, Perri F, D’Inca R, Ardizzone S, Riegler G, et al. Variants of CARD15 are associated with an aggressive clinical course of Crohn’s disease – an IG–IBD study. Am J Gastroenterol 2005;100:84-92. https://doi.org/10.1111/j.1572-0241.2005.40705.x.
37. Brant SR, Picco MF, Achkar JP, Bayless TM, Kane SV, Brzezinski A, et al. Defining complex contributions of NOD2/CARD15 gene mutations, age at onset, and tobacco use on Crohn’s disease phenotypes. Inflamm Bowel Dis 2003;9:281-9. https://doi.org/10.1097/00054725-200309000-00001.
38. Büning C, Genschel J, Bühner S, Krüger S, Kling K, Dignass A, et al. Mutations in the NOD2/CARD15 gene in Crohn’s disease are associated with ileocecal resection and are a risk factor for reoperation. Aliment Pharmacol Ther 2004;19:1073-8. https://doi.org/10.1111/j.1365-2036.2004.01967.x.
39. Charpentier C, Salleron J, Savoye G, Fumery M, Merle V, Laberenne JE, et al. Natural history of elderly-onset inflammatory bowel disease: a population-based cohort study. Gut 2014;63:423-32. https://doi.org/10.1136/gutjnl-2012-303864.
40. Chatzicostas C, Roussomoustakaki M, Potamianos S, Paspatis G, Mouzas I, Romanos J, et al. Factors associated with disease evolution in Greek patients with inflammatory bowel disease. BMC Gastroenterol 2006;6. https://doi.org/10.1186/1471-230X-6-21.
41. Chen M, Yi F, Zhou F, Huang M, Li J, Yan W, et al. Risk factors for initial surgery in patients with Crohn’s disease in Central China. Surg Today 2015;45:1002-8. https://doi.org/10.1007/s00595-014-1093-z.
42. Choung RS, Princen F, Stockfisch TP, Torres J, Maue AC, Porter CK, et al. Serologic microbial associated markers can predict Crohn’s disease behaviour years before disease diagnosis. Aliment Pharmacol Ther 2016;43:1300-10. https://doi.org/10.1111/apt.13641.
43. Cleynen I, González JR, Figueroa C, Franke A, McGovern D, Bortlík M, et al. Genetic factors conferring an increased susceptibility to develop Crohn’s disease also influence disease phenotype: results from the IBDchip European Project. Gut 2013;62:1556-65. https://doi.org/10.1136/gutjnl-2011-300777.
44. Degenhardt F, Dirmeier A, Lopez R, Lang S, Kunst C, Roggenbuck D, et al. Serologic anti-GP2 antibodies are associated with genetic polymorphisms, fibrostenosis, and need for surgical resection in Crohn’s disease. Inflamm Bowel Dis 2016;22:2648-57. https://doi.org/10.1097/mib.0000000000000936.
45. Dubinsky MC, Kugathasan S, Mei L, Picornell Y, Nebel J, Wrobel I, et al. Increased immune reactivity predicts aggressive complicating Crohn’s disease in children. Clin Gastroenterol Hepatol 2008;6:1105-11. https://doi.org/10.1016/j.cgh.2008.04.032.
46. Goel A, Dutta AK, Pulimood AB, Eapen A, Chacko A. Clinical profile and predictors of disease behavior and surgery in Indian patients with Crohn’s disease. Indian J Gastroenterol 2013;32:184-9. https://doi.org/10.1007/s12664-012-0293-y.
47. Gupta N, Cohen SA, Bostrom AG, Kirschner BS, Baldassano RN, Winter HS, et al. Risk factors for initial surgery in pediatric patients with Crohn’s disease. Gastroenterology 2006;130:1069-77. https://doi.org/10.1053/j.gastro.2006.02.003.
48. Henckaerts L, Van Steen K, Verstreken I, Cleynen I, Franke A, Schreiber S, et al. Genetic risk profiling and prediction of disease course in Crohn’s disease patients. Clin Gastroenterol Hepatol 2009;7:972-80. https://doi.org/10.1016/j.cgh.2009.05.001.
49. Jauregui-Amezaga A, Rimola J, Ordás I, Rodríguez S, Ramírez-Morros A, Gallego M, et al. Value of endoscopy and MRI for predicting intestinal surgery in patients with Crohn’s disease in the era of biologics. Gut 2015;64:1397-402. https://doi.org/10.1136/gutjnl-2014-308101.
50. Kugathasan S, Denson LA, Walters TD, Kim MO, Marigorta UM, Schirmer M, et al. Prediction of complicated disease course for children newly diagnosed with Crohn’s disease: a multicentre inception cohort study. Lancet 2017;389:1710-18. https://doi.org/10.1016/S0140-6736(17)30317-3.
51. Kugathasan S, Collins N, Maresso K, Hoffmann RG, Stephens M, Werlin SL, et al. CARD15 gene mutations and risk for early surgery in pediatric-onset Crohn’s disease. Clin Gastroenterol Hepatol 2004;2:1003-9. https://doi.org/10.1016/S1542-3565(04)00452-5.
52. Kwon JH, Im JP, Ye BD, Cheon JH, Jang HJ, Lee KM, et al. Disease phenotype, activity and clinical course prediction based on C-reactive protein levels at diagnosis in patients with Crohn’s disease: results from the CONNECT study. Gut Liver 2016;10:595-603. https://doi.org/10.5009/gnl15411.
53. Lacher M, Helmbrecht J, Schroepf S, Koletzko S, Ballauff A, Classen M, et al. NOD2 mutations predict the risk for surgery in pediatric-onset Crohn’s disease. J Pediatr Surg 2010;45:1591-7. https://doi.org/10.1016/j.jpedsurg.2009.10.046.
54. Laghi L, Costa S, Saibeni S, Bianchi P, Omodei P, Carrara A, et al. Carriage of CARD15 variants and smoking as risk factors for resective surgery in patients with Crohn’s ileal disease. Aliment Pharmacol Ther 2005;22:557-64. https://doi.org/10.1111/j.1365-2036.2005.02629.x.
55. Lakatos PL, Lakatos L, Szalay F, Willheim-Polli C, Osterreicher C, Tulassay Z, et al. Toll-like receptor 4 and NOD2/CARD15 mutations in Hungarian patients with Crohn’s disease: phenotype-genotype correlations. World J Gastroenterol 2005;11:1489-95. https://doi.org/10.3748/wjg.v11.i10.1489.
56. Lakatos PL, Czegledi Z, Szamosi T, Banai J, David G, Zsigmond F, et al. Perianal disease, small bowel disease, smoking, prior steroid or early azathioprine/biological therapy are predictors of disease behavior change in patients with Crohn’s disease. World J Gastroenterol 2009;15:3504-10. https://doi.org/10.3748/wjg.15.3504.
57. Law ST, Li KK. Age-related differences in the clinical course of Crohns disease in an Asian population: a retrospective cohort review. Indian Pediatr 2013;50:1148-52. https://doi.org/10.1007/s13312-013-0301-z.
58. Lunney PC, Kariyawasam VC, Wang RR, Middleton KL, Huang T, Selinger CP, et al. Smoking prevalence and its influence on disease course and surgery in Crohn’s disease and ulcerative colitis. Aliment Pharmacol Ther 2015;42:61-70. https://doi.org/10.1111/apt.13239.
59. Malmborg P, Grahnquist L, Ideström M, Lindholm J, Befrits R, Björk J, et al. Presentation and progression of childhood-onset inflammatory bowel disease in Northern Stockholm County. Inflamm Bowel Dis 2015;21:1098-108. https://doi.org/10.1097/MIB.0000000000000356.
60. Mazor Y, Maza I, Kaufman E, Ben-Horin S, Karban A, Chowers Y, et al. Prediction of disease complication occurrence in Crohn’s disease using phenotype and genotype parameters at diagnosis. J Crohns Colitis 2011;5:592-7. https://doi.org/10.1016/j.crohns.2011.06.002.
61. Moon CM, Park DI, Kim ER, Kim YH, Lee CK, Lee SH, et al. Clinical features and predictors of clinical outcomes in Korean patients with Crohn’s disease: a Korean association for the study of intestinal diseases multicenter study. J Gastroenterol Hepatol 2014;29:74-82. https://doi.org/10.1111/jgh.12369.
62. Nasir BF, Griffiths L, Nasir A, Roberts R, Barclay M, Gearry R, et al. Perianal disease combined with NOD2 genotype predicts need for IBD-related surgery in Crohn’s disease patients from a population-based cohort. J Clin Gastroenterol 2013;47:242-5. https://doi.org/10.1097/MCG.0b013e318258314d.
63. Nasir BF, Griffiths LR, Nasir A, Roberts R, Barclay M, Gearry RB, et al. An envirogenomic signature is associated with risk of IBD-related surgery in a population-based Crohn’s disease cohort. J Gastrointest Surg 2013;17:1643-50. https://doi.org/10.1007/s11605-013-2250-1.
64. Nunes T, Etchevers MJ, Domènech E, García-Sánchez V, Ber Y, Peñalva M, et al. Smoking does influence disease behaviour and impacts the need for therapy in Crohn’s disease in the biologic era. Aliment Pharmacol Ther 2013;38:752-60. https://doi.org/10.1111/apt.12440.
65. Odes HS, Fich A, Reif S, Halak A, Lavy A, Keter D, et al. Effects of current cigarette smoking on clinical course of Crohn’s disease and ulcerative colitis. Dig Dis Sci 2001;46:1717-21. https://doi.org/10.1023/a:1010609722315.
66. Oh K, Oh EH, Baek S, Song EM, Kim GU, Seo M, et al. Elevated C-reactive protein level during clinical remission can predict poor outcomes in patients with Crohn’s disease. PLOS ONE 2017;12. https://doi.org/10.1371/journal.pone.0179266.
67. Oriuchi T, Hiwatashi N, Kinouchi Y, Takahashi S, Takagi S, Negoro K, et al. Clinical course and longterm prognosis of Japanese patients with Crohn’s disease: predictive factors, rates of operation, and mortality. J Gastroenterol 2003;38:942-53. https://doi.org/10.1007/s00535-003-1177-9.
68. Pandey A, Salazar E, Kong CSC, Lim WC, Ong J, Ong DEH, et al. Risk of major abdominal surgery in an Asian population-based Crohn’s disease cohort. Inflamm Bowel Dis 2015;21:2625-33. https://doi.org/10.1097/MIB.0000000000000525.
69. Papi C, Festa V, Fagnani C, Stazi A, Antonelli G, Moretti A, et al. Evolution of clinical behaviour in Crohn’s disease: predictive factors of penetrating complications. Dig Liver Dis 2005;37:247-53. https://doi.org/10.1016/j.dld.2004.10.012.
70. Park SK, Yang SK, Park SH, Park SH, Kim JW, Yang DH, et al. Long-term prognosis of the jejunal involvement of Crohn’s disease. J Clin Gastroenterol 2013;47:400-8. https://doi.org/10.1097/MCG.0b013e3182705f9e.
71. Pigneur B, Seksik P, Viola S, Viala J, Beaugerie L, Girardet JP, et al. Natural history of Crohn’s disease: comparison between childhood- and adult-onset disease. Inflamm Bowel Dis 2010;16:953-61. https://doi.org/10.1002/ibd.21152.
72. Polito JM, Childs B, Mellits ED, Tokayer AZ, Harris ML, Bayless TM. Crohn’s disease: influence of age at diagnosis on site and clinical type of disease. Gastroenterology 1996;111:580-6. https://doi.org/10.1053/gast.1996.v111.pm8780560.
73. Posovszky C, Pfalzer V, Lahr G, Niess JH, Klaus J, Mayer B, et al. Age-of-onset-dependent influence of NOD2 gene variants on disease behaviour and treatment in Crohn’s disease. BMC Gastroenterol 2013;13. https://doi.org/10.1186/1471-230x-13-77.
74. Protic MB, Pavlovic ST, Bojic DZ, Krstic MN, Radojicic ZA, Tarabar DK, et al. CARD15 gene polymorphisms in Serbian patients with Crohn’s disease: genotype-phenotype analysis. Eur J Gastroenterol Hepatol 2008;20:978-84. https://doi.org/10.1097/MEG.0b013e328302f45e.
75. Renda MC, Orlando A, Civitavecchia G, Criscuoli V, Maggio A, Mocciaro F, et al. The role of CARD15 mutations and smoking in the course of Crohn’s disease in a Mediterranean area. Am J Gastroenterol 2008;103:649-55. https://doi.org/10.1111/j.1572-0241.2007.01589.x.
76. Rieder F, Schleder S, Wolf A, Dirmeier A, Strauch U, Obermeier F, et al. Serum anti-glycan antibodies predict complicated Crohn’s disease behavior: a cohort study. Inflamm Bowel Dis 2010;16:1367-75. https://doi.org/10.1002/ibd.21179.
77. 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 Gastroenterol 2009;104:371-83. https://doi.org/10.1038/ajg.2008.38.
78. Ryan JD, Silverberg MS, Xu W, Graff LA, Targownik LE, Walker JR, et al. Predicting complicated Crohn’s disease and surgery: phenotypes, genetics, serology and psychological characteristics of a population-based cohort. Aliment Pharmacol Ther 2013;38:274-83. https://doi.org/10.1111/apt.12368.
79. Sabate JM, Ameziane N, Lamoril J, Jouet P, Farmachidi JP, Soule JC, et al. The V249I polymorphism of the CX3CR1 gene is associated with fibrostenotic disease behavior in patients with Crohn’s disease. Eur J Gastroenterol Hepatol 2008;20:748-55. https://doi.org/10.1097/MEG.0b013e3282f824c9.
80. Sands BE, Arsenault JE, Rosen MJ, Alsahli M, Bailen L, Banks P, et al. Risk of early surgery for Crohn’s disease: implications for early treatment strategies. Am J Gastroenterol 2003;98:2712-18. https://doi.org/10.1111/j.1572-0241.2003.08674.x.
81. Savoye G, Salleron J, Gower-Rousseau C, Dupas JL, Vernier-Massouille G, Fumery M, et al. Clinical predictors at diagnosis of disabling pediatric Crohn’s disease. Inflamm Bowel Dis 2012;18:2072-8. https://doi.org/10.1002/ibd.22898.
82. Schaefer ME, Machan JT, Kawatu D, Langton CR, Markowitz J, Crandall W, et al. Factors that determine risk for surgery in pediatric patients with Crohn’s disease. Clin Gastroenterol Hepatol 2010;8:789-94. https://doi.org/10.1016/j.cgh.2010.05.021.
83. Shaoul R, Karban A, Reif S, Weiss B, Shamir R, Tamir A, et al. Disease behavior in children with Crohn’s disease: the effect of disease duration, ethnicity, genotype, and phenotype. Dig Dis Sci 2009;54:142-50. https://doi.org/10.1007/s10620-008-0326-7.
84. Siegel CA, Siegel LS, Hyams JS, Kugathasan S, Markowitz J, Rosh JR, et al. Real-time tool to display the predicted disease course and treatment response for children with Crohn’s disease. Inflamm Bowel Dis 2011;17:30-8. https://doi.org/10.1002/ibd.21386.
85. Siegel CA, Horton H, Siegel LS, Thompson KD, Mackenzie T, Stewart SK, et al. A validated web-based tool to display individualised Crohn’s disease predicted outcomes based on clinical, serologic and genetic variables. Aliment Pharmacol Ther 2016;43:262-71. https://doi.org/10.1111/apt.13460.
86. Solberg IC, Vatn MH, Høie O, Stray N, Sauar J, Jahnsen J, et al. Clinical course in Crohn’s disease: results of a Norwegian population-based ten-year follow-up study. Clin Gastroenterol Hepatol 2007;5:1430-8. https://doi.org/10.1016/j.cgh.2007.09.002.
87. Song XM, Gao X, Li MZ, Chen ZH, Chen SC, Hu PJ, et al. Clinical features and risk factors for primary surgery in 205 patients with Crohn’s disease: analysis of a South China cohort. Dis Colon Rectum 2011;54:1147-54. https://doi.org/10.1097/DCR.0b013e318222ddc3.
88. Tarrant KM, Barclay ML, Frampton CM, Gearry RB. Perianal disease predicts changes in Crohn’s disease phenotype-results of a population-based study of inflammatory bowel disease phenotype. Am J Gastroenterol 2008;103:3082-93. https://doi.org/10.1111/j.1572-0241.2008.02212.x.
89. 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. https://doi.org/10.1053/j.gastro.2010.06.070.
90. van der Heide F, Dijkstra A, Weersma RK, Albersnagel FA, van der Logt EM, Faber KN, et al. Effects of active and passive smoking on disease course of Crohn’s disease and ulcerative colitis. Inflamm Bowel Dis 2009;15:1199-207. https://doi.org/10.1002/ibd.20884.
91. Vernier-Massouille G, Balde M, Salleron J, Turck D, Dupas JL, Mouterde O, et al. Natural history of pediatric Crohn’s disease: a population-based cohort study. Gastroenterology 2008;135:1106-13. https://doi.org/10.1053/j.gastro.2008.06.079.
92. Vester-Andersen MK, Vind I, Prosberg MV, Bengtsson BG, Blixt T, Munkholm P, et al. Hospitalisation, surgical and medical recurrence rates in inflammatory bowel disease 2003–2011 – a Danish population-based cohort study. J Crohns Colitis 2014;8:1675-83. https://doi.org/10.1016/j.crohns.2014.07.010.
93. Yaari S, Benson A, Aviran E, Lev Cohain N, Oren R, Sosna J, et al. Factors associated with surgery in patients with intra-abdominal fistulizing Crohn’s disease. World J Gastroenterol 2016;22:10380-7. https://doi.org/10.3748/wjg.v22.i47.10380.
94. Adler J, Rangwalla SC, Dwamena BA, Higgins PD. The prognostic power of the NOD2 genotype for complicated Crohn’s disease: a meta-analysis. Am J Gastroenterol 2011;106:699-712. https://doi.org/10.1038/ajg.2011.19.
95. Bonneau J, Dumestre-Perard C, Rinaudo-Gaujous M, Genin C, Sparrow M, Roblin X, et al. Systematic review: new serological markers (anti-glycan, anti-GP2, anti-GM-CSF Ab) in the prediction of IBD patient outcomes. Autoimmun Rev 2015;14:231-45. https://doi.org/10.1016/j.autrev.2014.11.004.
96. Kaul A, Hutfless S, Liu L, Bayless TM, Marohn MR, Li X. Serum anti-glycan antibody biomarkers for inflammatory bowel disease diagnosis and progression: a systematic review and meta-analysis. Inflamm Bowel Dis 2012;18:1872-84. https://doi.org/10.1002/ibd.22862.
97. Mao R, Xiao YL, Gao X, Chen BL, He Y, Yang L, et al. Fecal calprotectin in predicting relapse of inflammatory bowel diseases: a meta-analysis of prospective studies. Inflamm Bowel Dis 2012;18:1894-9. https://doi.org/10.1002/ibd.22861.
98. Prideaux L, De Cruz P, Ng SC, Kamm MA. Serological antibodies in inflammatory bowel disease: a systematic review. Inflamm Bowel Dis 2012;18:1340-55. https://doi.org/10.1002/ibd.21903.
99. Xiong Y, Wang GZ, Zhou JQ, Xia BQ, Wang XY, Jiang B. Serum antibodies to microbial antigens for Crohn’s disease progression: a meta-analysis. Eur J Gastroenterol Hepatol 2014;26:733-42. https://doi.org/10.1097/MEG.0000000000000102.
100. Zhang Z, Li C, Zhao X, Lv C, He Q, Lei S, et al. Anti-Saccharomyces cerevisiae antibodies associate with phenotypes and higher risk for surgery in Crohn’s disease: a meta-analysis. Dig Dis Sci 2012;57:2944-54. https://doi.org/10.1007/s10620-012-2244-y.
101. Dubinsky MC, Lin YC, Dutridge D, Picornell Y, Landers CJ, Farrior S, et al. Serum immune responses predict rapid disease progression among children with Crohn’s disease: immune responses predict disease progression. Am J Gastroenterol 2006;101:360-7. https://doi.org/10.1111/j.1572-0241.2006.00456.x.
102. Aaltonen G, Carpelan-Holmström M, Keränen I, . Risk factors for proctectomy in consecutive Crohn’s colitis surgical patients in a reference colorectal centre. Int J Colorectal Dis 2019;34:1401-6. https://doi.org/10.1007/s00384-019-03337-8.
103. Aggarwal V, Day AS, Connor S, Leach ST, Brown G, Singh R, et al. Role of capsule endoscopy and fecal biomarkers in small-bowel Crohn’s disease to assess remission and predict relapse. Gastrointest Endosc 2017;86:1070-8. https://doi.org/10.1016/j.gie.2017.09.011.
104. Alexakis C, Saxena S, Chhaya V, Cecil E, Majeed A, Pollok R. Smoking status at diagnosis and subsequent smoking cessation: associations with corticosteroid use and intestinal resection in Crohn’s disease. Am J Gastroenterol 2018;113:1689-700. https://doi.org/10.1038/s41395-018-0273-7.
105. Arieira C, Cúrdia Gonçaves T, Dias de Castro F, João Moreira M, Cotter J. Clinical course in Crohn’s disease: factors associated with behaviour change and surgery. Scand J Gastroenterol 2018;53:1222-7. https://doi.org/10.1080/00365521.2018.1503709.
106. Arora U, Ananthakrishnan AN, Kedia S, Bopanna S, Mouli PM, Yadav DP, et al. Effect of oral tobacco use and smoking on outcomes of Crohn’s disease in India. J Gastro Hepatol 2018;33:134-40. https://doi.org/10.1111/jgh.13815.
107. Arora U, Kedia S, Garg P, Bopanna S, Jain S, Yadav DP, et al. Colonic Crohn’s disease is associated with less aggressive disease course than ileal or ileocolonic disease. Dig Dis Sci 2018;63:1592-9. https://doi.org/10.1007/s10620-018-5041-4.
108. Assa A, Amitai M, Greer ML, Castro DA, Kuint RC, Martínez-León M, et al. Perianal pediatric Crohn disease is associated with a distinct phenotype and greater inflammatory burden. J Pediatr Gastroenterol Nutr 2017;65:293-8. https://doi.org/10.1097/MPG.0000000000001484.
109. Assa A, Rinawi F, Shamir R. The long-term predictive properties of the paris classification in paediatric inflammatory bowel disease patients. J Crohns Colitis 2018;12:39-47. https://doi.org/10.1093/ecco-jcc/jjx125.
110. Birimberg-Schwartz L, Wilson DC, Kolho KL, Karolewska-Bochenek K, Afzal NA, Spray C, et al. pANCA and ASCA in children with IBD-unclassified, Crohn’s colitis, and ulcerative colitis-a longitudinal report from the IBD Porto Group of ESPGHAN. Inflamm Bowel Dis 2016;22:1908-14. https://doi.org/10.1097/MIB.0000000000000784.
111. Bossuyt P, Debeuckelaere C, Ferrante M, de Buck van Overstraeten A, Vanbeckevoort D, Billiet T, et al. Risk stratification for surgery in stricturing ileal Crohn’s disease: the BACARDI risk model. J Crohn’s Colitis 2018;12:32-8. https://doi.org/10.1093/ecco-jcc/jjx110.
112. Brückner A, Werkstetter KJ, de Laffolie J, Wendt C, Prell C, Weidenhausen T, et al. Incidence and risk factors for perianal disease in pediatric crohn disease patients followed in CEDATA-GPGE registry. J Pediatr Gastroenterol Nutr 2018;66:73-8. https://doi.org/10.1097/MPG.0000000000001649.
113. Buisson A, Mak WY, Andersen MJ, Lei D, Kahn SA, Pekow J, et al. Faecal calprotectin is a very reliable tool to predict and monitor the risk of relapse after therapeutic de-escalation in patients with inflammatory bowel diseases. J Crohns Colitis 2019;13:1012-24. https://doi.org/10.1093/ecco-jcc/jjz023.
114. Burisch J, Kiudelis G, Kupcinskas L, Kievit HAL, Andersen KW, Andersen V, et al. Natural disease course of Crohn’s disease during the first 5 years after diagnosis in a European population-based inception cohort: an Epi-IBD study. Gut 2019;68:423-3. https://doi.org/10.1136/gutjnl-2017-315568.
115. Chaparro M, Garre A, Ricart E, Iglesias-Flores E, Taxonera C, Domènech E, et al. Differences between childhood- and adulthood-onset inflammatory bowel disease: the CAROUSEL study from GETECCU. Aliment Pharmacol Ther 2019;49:419-28. https://doi.org/10.1111/apt.15114.
116. Chaudhry NA, Riverso M, Grajo JR, Moser PP, Zou F, Homsi M, et al. A fixed stricture on routine cross-sectional imaging predicts disease-related complications and adverse outcomes in patients with Crohn’s disease. Inflamm Bowel Dis 2017;23:641-9. https://doi.org/10.1097/MIB.0000000000001054.
117. Chen GB, Lee SH, Montgomery GW, Wray NR, Visscher PM, Gearry RB, et al. Performance of risk prediction for inflammatory bowel disease based on genotyping platform and genomic risk score method. BMC Med Genet 2017;18. https://doi.org/10.1186/s12881-017-0451-2.
118. Chhaya V, Saxena S, Cecil E, Subramanian V, Curcin V, Majeed A, et al. Emerging trends and risk factors for perianal surgery in Crohn’s disease: a 20-year national population-based cohort study. Eur J Gastroenterol Hepatol 2016;28:890-5. https://doi.org/10.1097/MEG.0000000000000651.
119. Chun J, Im JP, Kim JW, Lee KL, Choi CH, Kim H, et al. Association of perianal fistulas with clinical features and prognosis of Crohn’s disease in Korea: results from the CONNECT study. Gut Liver 2018;12:544-54. https://doi.org/10.5009/gnl18157.
120. de Barros KSC, Flores C, Harlacher L, Francesconi CFM. Evolution of clinical behavior in Crohn’s disease: factors associated with complicated disease and surgery. Dig Dis Sci 2017;62:2481-8. https://doi.org/10.1007/s10620-017-4685-9.
121. Dias CC, Pereira Rodrigues P, Fernandes S, Portela F, Ministro P, Martins D, et al. The risk of disabling, surgery and reoperation in Crohn’s disease - A decision tree-based approach to prognosis. PLOS ONE 2017;12. https://doi.org/10.1371/journal.pone.0172165.
122. Dias CC, Rodrigues PP, Coelho R, Santos PM, Fernandes S, Lago P, et al. Development and validation of risk matrices for Crohn’s disease outcomes in patients who underwent early therapeutic interventions. J Crohns Colitis 2017;11:445-53.
123. Diederen K, Hoekman DR, Leek A, Wolters VM, Hummel TZ, de Meij TG, et al. Raised faecal calprotectin is associated with subsequent symptomatic relapse, in children and adolescents with inflammatory bowel disease in clinical remission. Aliment Pharmacol Ther 2017;45:951-60. https://doi.org/10.1111/apt.13950.
124. Dong Y, Xu L, Fan Y, Xiang P, Gao X, Chen Y, et al. A novel surgical predictive model for Chinese Crohn’s disease patients. Medicine (Baltimore) 2019;98. https://doi.org/10.1097/MD.0000000000017510.
125. Foster AJ, Smyth M, Lakhani A, Jung B, Brant RF, Jacobson K. Consecutive fecal calprotectin measurements for predicting relapse in pediatric Crohn’s disease patients. World J Gastroenterol 2019;25:1266-77. https://doi.org/10.3748/wjg.v25.i10.1266.
126. Fumery M, Pariente B, Sarter H, Charpentier C, Armengol Debeir L, Dupas JL, et al. Natural history of Crohn’s disease in elderly patients diagnosed over the age of 70 years: a population-based study. Inflamm Bowel Dis 2016;22:1698-707. https://doi.org/10.1097/MIB.0000000000000821.
127. Fumery M, Pariente B, Sarter H, Savoye G, Spyckerelle C, Djeddi D, et al. Long-term outcome of pediatric-onset Crohn’s disease: A population-based cohort study. Dig Liver Dis 2019;51:496-502. https://doi.org/10.1016/j.dld.2018.11.033.
128. Gasparetto M, Guariso G, Pozza LV, Ross A, Heuschkel R, Zilbauer M. Clinical course and outcomes of diagnosing Inflammatory Bowel Disease in children 10 years and under: retrospective cohort study from two tertiary centres in the United Kingdom and in Italy. BMC Gastroenterol 2016;16. https://doi.org/10.1186/s12876-016-0455-y.
129. Guizzetti L, Zou G, Khanna R, Dulai PS, Sandborn WJ, Jairath V, et al. Development of clinical prediction models for surgery and complications in Crohn’s disease. J Crohns Colitis 2018;12:167-77. https://doi.org/10.1093/ecco-jcc/jjx130.
130. Herman Y, Rinawi F, Rothschild B, Nir O, Shamir R, Assa A. The characteristics and long-term outcomes of pediatric Crohn’s disease patients with perianal disease. Inflamm Bowel Dis 2017;23:1659-65. https://doi.org/10.1097/MIB.0000000000001171.
131. Herzog D, Fournier N, Buehr P, Rueger V, Koller R, Heyland K, et al. Age at disease onset of inflammatory bowel disease is associated with later extraintestinal manifestations and complications. Eur J Gastroenterol Hepatol 2018;30:598-607. https://doi.org/10.1097/MEG.0000000000001072.
132. Hou JK, Feagins LA, Waljee AK. Characteristics and behavior of elderly-onset inflammatory bowel disease: a multi-center US study. Inflammatory Bowel Diseases 2016;22:2200-5. https://doi.org/10.1097/MIB.0000000000000849.
133. Huguet JM, Iborra M, Bosca-Watts MM, Maroto N, Gil R, Cortes X, et al. Inflammatory bowel disease in patients over the age of 70 y. Does the disease duration influence its behavior?. Scand J Gastroenterol 2018;53:1079-84. https://doi.org/10.1080/00365521.2018.1501603.
134. Hwang SW, Kim JH, Im JP, Ye BD, Koo HS, Huh KC, et al. Influence of age at diagnosis on the clinical characteristics of Crohn’s disease in Korea: results from the CONNECT study. J Gastroenterol Hepatol 2017;32:1716-22. https://doi.org/10.1111/jgh.13775.
135. Jeuring SF, van den Heuvel TR, Liu LY, Zeegers MP, Hameeteman WH, Romberg-Camps MJ, et al. Improvements in the long-term outcome of Crohn’s disease over the past two decades and the relation to changes in medical management: results from the population-based IBDSL cohort. Am J Gastroenterol 2017;112:325-36. https://doi.org/10.1038/ajg.2016.524.
136. Jeuring SF, van den Heuvel TR, Zeegers MP, Hameeteman WH, Romberg-Camps MJ, Oostenbrug LE, et al. Epidemiology and long-term outcome of inflammatory bowel disease diagnosed at elderly age-an increasing distinct entity?. Inflamm Bowel Dis 2016;22:1425-34. https://doi.org/10.1097/MIB.0000000000000738.
137. Jones GR, Fascì-Spurio F, Kennedy NA, Plevris N, Jenkinson P, Lyons M, et al. Faecal calprotectin and magnetic resonance enterography in ileal Crohn’s disease: correlations between disease activity and long-term follow-up. J Crohns Colitis 2019;13:442-50. https://doi.org/10.1093/ecco-jcc/jjy187.
138. Kaur M, Panikkath D, Yan X, Liu Z, Berel D, Li D, et al. Perianal Crohn’s disease is associated with distal colonic disease, stricturing disease behavior, IBD-associated serologies and genetic variation in the JAK-STAT pathway. Inflamm Bowel Dis 2016;22:862-9. https://doi.org/10.1097/MIB.0000000000000705.
139. Kayar Y, Baran B, Ormeci AC, Akyuz F, Demir K, Besisik F, et al. Risk factors associated with progression to intestinal complications of Crohn disease. Chin Med J (Engl 2019;132:2423-9. https://doi.org/10.1097/CM9.0000000000000489.
140. Kim HJ, Oh SH, Kim DY, Lee HS, Park SH, Yang SK, et al. Clinical characteristics and long-term outcomes of paediatric Crohn’s disease: a single-centre experience. J Crohns Colitis 2017;11:157-64. https://doi.org/10.1093/ecco-jcc/jjw146.
141. Kim JW, Lee HS, Ye BD, Yang SK, Hwang SW, Park SH, et al. Incidence of and risk factors for free bowel perforation in patients with Crohn’s disease. Dig Dis Sci 2017;62:1607-14. https://doi.org/10.1007/s10620-017-4539-5.
142. Kim OZ, Han DS, Park CH, Eun CS, Kim YS, Kim YH, et al. The clinical characteristics and prognosis of Crohn’s disease in Korean patients showing proximal small bowel involvement: results from the CONNECT study. Gut Liver 2018;12:67-72. https://doi.org/10.5009/gnl16500.
143. Kostas A, Siakavellas SI, Kosmidis C, Takou A, Nikou J, Maropoulos G, et al. Fecal calprotectin measurement is a marker of short-term clinical outcome and presence of mucosal healing in patients with inflammatory bowel disease. World J Gastroenterol 2017;23:7387-96. https://doi.org/10.3748/wjg.v23.i41.7387.
144. Kühn F, Nixdorf M, Schwandner F, Klar E. Risikofaktoren für einen frühen OP-Zeitpunkt und chirurgische Komplikationen bei Morbus Crohn [Risk factors for early surgery and surgical complications in Crohn’s disease]. Zentralbl Chir 2018;143:596-602. https://doi.org/10.1055/a-0645-1489.
145. Kunovsky L, Kala Z, Marek F, Dolina J, Poredska K, Kucerova L, et al. The role of the NOD2/CARD15 gene in surgical treatment prediction in patients with Crohn’s disease. Int J Colorectal Dis 2019;34:347-51. https://doi.org/10.1007/s00384-018-3122-7.
146. Kupka T, Simova J, Dvorackova J, Martinek L, Motyka O, Uvirova M, et al. Crohn’s disease – genetic factors and progress of the disease. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2018;162:139-43. https://doi.org/10.5507/bp.2017.058.
147. Kwapisz L, Gregor J, Chande N, Yan B, Ponich T, Mosli M. The utility of fecal calprotectin in predicting the need for escalation of therapy in inflammatory bowel disease. Scand J Gastroenterol 2017;52:846-50. https://doi.org/10.1080/00365521.2017.1315740.
148. Levine A, Chanchlani N, Hussey S, Ziv-Baran T, Escher JC, Amil Dias J, et al. Complicated disease and response to initial therapy predicts early surgery in paediatric Crohn’s disease: results from the Porto Group GROWTH study. J Crohns Colitis 2020;14:71-8. https://doi.org/10.1093/ecco-jcc/jjz111.
149. Liu W, Zhou W, Xiang J, Cao Q, Zhu J, Qi W, et al. Lémann Index at diagnosis predicts the risk of early surgery in Crohn’s disease. Dis Colon Rectum 2018;61:207-13. https://doi.org/10.1097/DCR.0000000000000930.
150. Mańkowska-Wierzbicka D, Karczewski J, Poniedziałek B, Grzymisławska M, Staszewski R, Królczyk A, et al. C-reactive protein as a diagnostic and prognostic factor in inflammatory bowel diseases. Postepy Hig Med Dosw 2016;70:1124-30. https://doi.org/10.5604/17322693.1223798.
151. Mañosa M, Calafat M, de Francisco R, García C, Casanova MJ, Huelín P, et al. Phenotype and natural history of elderly onset inflammatory bowel disease: a multicentre, case-control study. Aliment Pharmacol Ther 2018;47:605-14. https://doi.org/10.1111/apt.14494.
152. Mosli M, Sabbahi H, Alyousef H, Abdulhaq M, Hadadi A, Aljahdali E, et al. Risk stratification of patients with Crohn’s disease: a retrospective analysis of clinical decision making and its impact on long-term outcome. Dig Dis 2018;36:49-55. https://doi.org/10.1159/000477613.
153. Müller KE, Lakatos PL, Kovacs JB, Arato A, Varkonyi A, Nemes E, et al. Baseline characteristics and disease phenotype in inflammatory bowel disease. J Pediatr Gastroenterol Nutr 2016;62:50-5. https://doi.org/10.1097/MPG.0000000000000885.
154. Naganuma M, Hisamatsu T, Matsuoka K, Kiyohara H, Arai M, Sugimoto S, et al. Endoscopic severity predicts long-term prognosis in Crohn’s disease patients with clinical remission. Digestion 2016;93:66-71. https://doi.org/10.1159/000441767.
155. Nahon S, Lahmek P, Paupard T, Lesgourgues B, Chaussade S, Peyrin-Biroulet L, et al. Diagnostic delay is associated with a greater risk of early surgery in a french cohort of Crohn’s disease patients. Dig Dis Sci 2016;61:3278-84. https://doi.org/10.1007/s10620-016-4189-z.
156. Ng SC, Zeng Z, Niewiadomski O, Tang W, Bell S, Kamm MA, et al. Early course of inflammatory bowel disease in a population-based inception cohort study from 8 countries in Asia and Australia. Gastroenterology 2016;150:86-95.e3. https://doi.org/10.1053/j.gastro.2015.11.019.
157. Nguyen GC, Bernstein CN, Benchimol EI. Risk of surgery and mortality in elderly-onset inflammatory bowel disease: a population-based cohort study. Inflamm Bowel Dis 2017;23:218-23. https://doi.org/10.1097/MIB.0000000000000993.
158. Ouaz A, Fekih M, Labidi A, Ben Mustapha N, Serghini M, Zouiten L, et al. Changes of Crohn’s disease phenotype over time. Tunis Med 2016;94:167-70.
159. Pallotta N, Vincoli G, Pezzotti P, Giovannone M, Gigliozzi A, Badiali D, et al. A risk score system to timely manage treatment in Crohn’s disease: a cohort study. BMC Gastroenterol 2018;18. https://doi.org/10.1186/s12876-018-0889-5.
160. Park SH, Aniwan S, Scott Harmsen W, Tremaine WJ, Lightner AL, Faubion WA, et al. Update on the natural course of fistulizing perianal Crohn’s disease in a population-based cohort. Inflamm Bowel Dis 2019;25:1054-60. https://doi.org/10.1093/ibd/izy329.
161. Park Y, Cheon JH, Park YL, Ye BD, Kim YS, Han DS, et al. Development of a novel predictive model for the clinical course of Crohn’s disease: results from the CONNECT study. Inflamm Bowel Dis 2017;23:1071-9. https://doi.org/10.1097/MIB.0000000000001106.
162. Parker D, Karmazyn B, Steiner SJ. Radiologic predictors of surgery in newly diagnosed pediatric Crohn disease patients. J Pediatr Gastroenterol Nutr 2016;63:e182-5. https://doi.org/10.1097/MPG.0000000000001217.
163. Pernat Drobež C, Repnik K, Gorenjak M, Ferkolj I, Weersma RK, Potočnik U. DNA polymorphisms predict time to progression from uncomplicated to complicated Crohn’s disease. Eur J Gastroenterol Hepatol 2018;30:447-55. https://doi.org/10.1097/MEG.0000000000001055.
164. Rinawi F, Assa A, Hartman C, Mozer Glassberg Y, Nachmias Friedler V, Rosenbach Y, et al. Evolution of disease phenotype in pediatric-onset Crohn’s disease after more than 10 years follow up – Cohort study. Dig Liver Dis 2016;48:1444-50. https://doi.org/10.1016/j.dld.2016.08.118.
165. Rispo A, Imperatore N, Testa A, Bucci L, Luglio G, De Palma GD, et al. Combined endoscopic/sonographic-based risk matrix model for predicting one-year risk of surgery: a prospective observational study of a tertiary centre severe/refractory Crohn’s disease cohort. J Crohns Colitis 2018;12:784-93. https://doi.org/10.1093/ecco-jcc/jjy032.
166. Rönnblom A, Holmström T, Karlbom U, Tanghöj H, Thörn M, Sjöberg D. Clinical course of Crohn’s disease during the first 5 years. Results from a population-based cohort in Sweden (ICURE) diagnosed 2005-2009. Scand J Gastroenterol 2017;52:81-6. https://doi.org/10.1080/00365521.2016.1230777.
167. Saad AM, Czul F, Sakuraba A, Rubin DT, Cohen RD. Age of diagnosis is associated with disease presentation and therapeutic complications in patients with Crohn’s disease. Inflamm Bowel Dis 2016;22:1027-31. https://doi.org/10.1097/MIB.0000000000000732.
168. Sharma Y, Bousvaros A, Liu E, Bender Stern J. Natural history of children with mild Crohn’s disease. World J Gastroenterol 2019;25:4235-45. https://doi.org/10.3748/wjg.v25.i30.4235.
169. Smids C, Horjus Talabur Horje CS, Nierkens S, Drylewicz J, Groenen MJM, Wahab PJ, et al. Candidate serum markers in early Crohn’s disease: predictors of disease course. J Crohns Colitis 2017;11:1090-100. https://doi.org/10.1093/ecco-jcc/jjx049.
170. Sollelis E, Quinard RM, Bouguen G, Goutte M, Goutorbe F, Bouvier D, et al. Combined evaluation of biomarkers as predictor of maintained remission in Crohn’s disease. World J Gastroenterol 2019;25:2354-64. https://doi.org/10.3748/wjg.v25.i19.2354.
171. Song EM, Kim N, Lee SH, Chang K, Hwang SW, Park SH, et al. Clinical characteristics and long-term prognosis of elderly-onset Crohn’s disease. Scand J Gastroenterol 2018;53:417-25. https://doi.org/10.1080/00365521.2018.1437927.
172. Stallmach A, Bokemeyer B, Helwig U, Lügering A, Teich N, Fischer I, et al. Predictive parameters for the clinical course of Crohn’s disease: development of a simple and reliable risk model. Int J Colorectal Dis 2019;34:1653-60. https://doi.org/10.1007/s00384-019-03369-0.
173. Sun XW, Wei J, Yang Z, Jin XX, Wan HJ, Yuan BS, et al. Clinical features and prognosis of Crohn’s disease with upper gastrointestinal tract phenotype in chinese patients. Dig Dis Sci 2019;64:3291-9. https://doi.org/10.1007/s10620-019-05651-1.
174. Szántó K, Nyári T, Bálint A, Bor R, Milassin Á, Rutka M, et al. Biological therapy and surgery rates in inflammatory bowel diseases - Data analysis of almost 1000 patients from a Hungarian tertiary IBD center. PLOS ONE 2018;13. https://doi.org/10.1371/journal.pone.0200824.
175. Torres J, Caprioli F, Katsanos KH, Lobatón T, Micic D, Zerôncio M, et al. Predicting Outcomes to optimize disease management in inflammatory bowel diseases. J Crohns Colitis 2016;10:1385-94. https://doi.org/10.1093/ecco-jcc/jjw116.
176. Wang XQ, Xiao Y, Xu X, Yu Y, Shan CY, Guo Y, et al. Study of disease phenotype and its association with prognosis of paediatric inflammatory bowel disease in China. BMC Pediatr 2018;18. https://doi.org/10.1186/s12887-018-1212-x.
177. Wu J, Lubman DM, Kugathasan S, Denson LA, Hyams JS, Dubinsky MC, et al. Serum protein biomarkers of fibrosis aid in risk stratification of future stricturing complications in pediatric Crohn’s disease. Am J Gastroenterol 2019;114:777-85. https://doi.org/10.14309/ajg.0000000000000237.
178. Ye L, Chen BQ, Wang SD, Shi H, Yang Z, Wang FY. Fecal calprotectin is a strong predictive marker of relapse in Chinese patients with Crohn’s disease: a two-year prospective study. Scand J Gastroenterol 2017;52:1113-19. https://doi.org/10.1080/00365521.2017.1346704.
179. Zhao M, Lo BZS, Vester-Andersen MK, Vind I, Bendtsen F, Burisch J. A 10-year follow-up study of the natural history of perianal Crohn’s disease in a Danish population-based inception cohort. Inflamm Bowel Dis 2019;25:1227-36. https://doi.org/10.1093/ibd/izy374.
180. Zhulina Y, Cao Y, Amcoff K, Carlson M, Tysk C, Halfvarson J. The prognostic significance of faecal calprotectin in patients with inactive inflammatory bowel disease. Aliment Pharmacol Ther 2016;44:495-504. https://doi.org/10.1111/apt.13731.
181. Ziv-Baran T, Hussey S, Sladek M, Amil Dias J, Martin de Carpi J, Miele E, et al. Response to treatment is more important than disease severity at diagnosis for prediction of early relapse in new-onset paediatric Crohn’s disease. Aliment Pharmacol Ther 2018;48:1242-50. https://doi.org/10.1111/apt.15016.
182. Mazor Y, Almog R, Kopylov U, Ben Hur D, Blatt A, Dahan A, et al. Adalimumab drug and antibody levels as predictors of clinical and laboratory response in patients with Crohn’s disease. Aliment Pharmacol Ther 2014;40:620-8. https://doi.org/10.1111/apt.12869.
183. Zitomersky NL, Atkinson BJ, Fournier K, Mitchell PD, Stern JB, Butler MC, et al. Antibodies to infliximab are associated with lower infliximab levels and increased likelihood of surgery in pediatric IBD. Inflamm Bowel Dis 2015;21:307-14. https://doi.org/10.1097/MIB.0000000000000284.
184. Collins GS, Reitsma JB, Altman DG, Moons KG. Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD): the TRIPOD statement. Ann Intern Med 2015;162:55-63. https://doi.org/10.7326/M14-0697.

List of abbreviations

anti-CBir1

anti-flagellin antibody

anti-TNF

anti-tumour necrosis factor

ASCA

anti-Saccharomyces cerevisiae antibodies

CDAI

Crohn’s Disease Activity Index

CHARMS

CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies

CI

confidence interval

CRP

C-reactive protein

ESR

erythrocyte sedimentation rate

HTA

Health Technology Assessment

IBD

inflammatory bowel disease

Ig

immunoglobulin

IQR

interquartile range

JAK

Janus kinase

MeSH

medical subject heading

NICE

National Institute for Health and Care Excellence

NIHR

National Institute for Health Research

NOD2

nucleotide-binding oligomerisation domain-containing protein 2

OR

odds ratio

PPI

patient and public involvement

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROBAST

Prediction model Risk Of Bias Assessment Tool

ROB

risk of bias

STAT

signal transducer and activator of transcription

TA

Technology Appraisal

TNF

tumour necrosis factor

TRIPOD

Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis

WBC

white blood cell