Personalised treatment in biologically naive patients with Crohn's disease: a systematic review of the last 10 years.

BACKGROUND

Biological treatment has revolutionised the management of Crohn's disease (CD), however, at baseline (pre-treatment), it is not possible to predict which biologic a patient will respond the best to.

AIM: This systematic review aimed to identify pre-treatment predictors of response/non-response to infliximab, adalimumab, vedolizumab and ustekinumab and to critically evaluate the available literature.

METHODS

We searched Embase, MEDLINE and Cochrane Registry of Controlled Trials (published from 2014 to 2024) for studies reporting on pre-treatment predictors of response/non-response to infliximab, adalimumab, vedolizumab, ustekinumab in adult (≥18 years) biologically naïve CD patients. Two reviewers independently completed title/abstract screening, full-text reading, data extraction, and quality assessment. The quality of the included studies was evaluated using 'Prediction model Risk Of Bias ASsessment Tool' (PROBAST).

RESULTS

A total of 39 studies were eligible for final inclusion. Vedolizumab and ustekinumab were investigated in one study each. For infliximab and adalimumab, we found several potential predictors such as age, body mass index (BMI), disease duration and behaviour, C-reactive protein (CRP), interleukins (ILs) and different imaging characteristics, that could be integrated into future prediction model studies. All studies were judged high risk of bias (RoB) - mainly due to problems regarding analyses and outcome definitions and determinations.

CONCLUSIONS

No pre-treatment predictors fulfilled the criteria of being an appropriate predictor of response/non-response. This was due to high RoB, conflicting results and a large heterogeneity between the studies. Future studies must be well-designed and follow strict protocols to produce robust results.