Clinical and technical evidence

A literature search was carried out for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.

Published evidence

Two published cohort studies and 1 prospective cohort study (currently under review) are summarised in this briefing. The studies provide biochemical 'proof of concept' evidence for the signature underlying PredictSure-IBD. They also provide evidence that the assay can predict severe relapsing Crohn's disease and severe ulcerative colitis in patients with inflammatory bowel disease (IBD). Both of the studies use the underlying signature for PredictSure-IBD but do not use the actual test. The prospective study looks at the use of PredictSure-IBD in clinical practice and follows up patients for around 5 years. These studies are summarised in table 1.

Overall assessment of the evidence

One cohort study used existing assays together to detect patterns of gene expression related to T-cell exhaustion. This showed some specific signatures in several infectious and autoimmune diseases, including IBD. A specific transcriptional signature can identify patients with active, untreated Crohn's disease who could have beneficial long-term outcomes if anti-tumour necrosis factor (TNF) therapy is started early. The company used this algorithm to do further applications of the assay in different cohorts. They present data for the test used in 350 patients with any type of autoimmune disease, but this is limited to 210 patients with Crohn's disease and ulcerative colitis.

Most of the evidence is retrospective analyses of cohorts of blood samples from patients recruited to the study when first diagnosed. They have been followed through to development of Crohn's disease from first presentation with IBD. There are few demographics or other details of the patients available.

The prospective cohort study recruited patients with IBD for testing with PredictSure-IBD. These patients were followed-up for around 5 years. PredictSure-IBD identified 2 patient subgroups with different severity of disease. Further research is being done by recruiting patients to have the test and then have treatment based on the results.

Table 1 Summary of selected studies

McKinney et al. (2015)

Study size, design and location

Observational cohort study of 58 people with Crohn's disease.

Location: UK.

Intervention and comparator(s)

Polymerase chain reaction analysis of blood samples to detect CD8 T-cell transcriptional signature.

No comparator.

Key outcomes

Results showed a transcriptional signature of the expression of surrogate markers of co-stimulation/exhaustion signatures in independent data sets. This confirmed an association with poor outcome in autoimmune and inflammatory disease. The results are based on a retrospective analysis of a relatively small group of 58 patients. The study also looked at other immune-related disorders (type 1 diabetes, adeno-associated virus, systemic lupus erythematosus, idiopathic pulmonary fibrosis and dengue haemorrhagic fever) as well as association of signatures with good clinical outcome or response to therapy in infections (hepatitis C virus) and vaccination (yellow fever, malaria, influenza).

Strengths and limitations

This is a high-quality biochemical study looking at markers in blood samples characteristic of immune response in terms of up and down gene regulation in response to various infectious and autoimmune diseases. It showed there was a clear signature in the blood of untreated people with Crohn's disease at the point they presented as patients with IBD, which was associated with poor outcome.

This is the same underlying signature as that in the PredictSure-IBD test but not identical to that used in the test.

Lee et al. (2011)

Study size, design and location

Observational prospective study of 35 people with Crohn's disease and 32 people with ulcerative colitis.

Location: UK.

Intervention and comparator(s)

Standard care step-up strategy given by doctors blinded to the PredictSure-IBD signature test results (note – this study did not use the actual test but did use the same signature that is used in PredictSure-IBD).

No comparator.

Key outcomes

The study used statistical techniques to identify 2 subgroups of patients, IBD1 and IBD2. This showed CD8+ T-cell transcriptional signatures that identified 2 subgroups that had very different disease courses. Patients in the subgroup with elevated expression of genes involved in antigen-dependent T-cell responses had substantially higher incidence of frequently relapsing disease. The authors comment that this suggests that the course of otherwise distinct autoimmune and inflammatory conditions may be influenced by common pathways and identifies the first biomarker that can predict prognosis in Crohn's disease at first diagnosis. This is a step toward personalised therapy.

Strengths and limitations

This study presents a data analysis of results from the PredictSure-IBD signature in a cohort of Crohn's disease patients, 66% of whom were previously undiagnosed. The patients were recruited specifically for this study and all treatment was blinded to the test results. The study is relatively small and gives basic demographic information for the patients. Very detailed descriptions of the treatment course and outcomes are presented in the supplementary information as are details of the complex statistical methods used in the analysis.

Biasci et al. (2019)

Study size, design and location

Prospective cohort study of 66 people with Crohn's disease and 57 people with ulcerative colitis.

Location: UK.

Intervention and comparator(s)

People with previously untreated Crohn's and ulcerative colitis were tested with PredictSure-IBD. Clinicians were blinded to the test result.

Key outcomes

There were 33 people with Crohn's disease and 24 people with ulcerative colitis classified as IBD1 phenotype. These people had more aggressive disease and needed earlier and more frequent escalations in treatment in comparison to people identified as having the IBD2 phenotype over a median follow up of 5.3 years.

Strengths and limitations

People included in the study were treated according to national guidelines at the discretion of their clinician rather than following a formal protocol. This may have led to some differences in the way individuals were treated but is representative of clinical practice in the NHS. The study results show that people with IBD1 phenotype have more severe disease but does not show how basing therapy choices on this might improve outcomes (this is being assessed in the PROFILE trial).

Abbreviations: CD8, cluster of differentiation 8; IBD, inflammatory bowel disease.

Recent and ongoing studies