Send to printer »

GEN News Highlights : Sep 27, 2011

Investigators Identify CD8 T Cell Transcription Signature that Predicts Prognosis in Crohn's and Ulcerative Colitis

Researchers claim patients can be stratified to accurately predict course of disease at time of diagnosis.

Scientists have identified analogous CD8+ T-cell transcription signatures in both Crohn disease (CD) and ulcerative colitis (UC) patients that they claim can predict at the time of diagnosis how likely patients are to develop relapsing disease and require escalation of treatment. The based team says the signatures can delineate between otherwise indistinguishable subsets of patients and represent the first biomarker for predicting prognosis in both UC and CD from diagnosis.

Reported in The Journal of Clinical Investigation, the findings also suggest that the two otherwise distinct autoimmune and inflammatory conditions may be influenced by common pathways, according to authors James C. Lee, M.D., Kenneth G. C. Smith, M.D., and colleagues at the Cambridge Institute for Medical Research and  the University of Cambridge School of Clinical Medicine. The results are described in a paper titled “Gene expression profiling of CD8+ T cells predicts prognosis in patients with Crohn disease and ulcerative colitis.”

The Cambridge team had previously identified a common CD8 T-cell transcriptional signature in the two unrelated autoimmune diseases SLE and ANCA-associated vasculitis (AAV), which predicted disease prognosis in both disorders. Building on this work, the researchers hypothesized that because UC and CD share a relapsing-remitting course driven by immunological responses to antigens, they may also display a CD8 T-cell signature prognostic of disease progression.

They carried out their search in a cohort of 35 patients with active CD and 32 patients with active UC, who hadn’t yet started therapy. CD4 and CD8 T cells were positively selected from peripheral blood mononuclear cells (PBMCs) for whole-genome transcriptional analyses, and the patients were then started on treatment regimens based on a conventional step-up strategy.

Consensus clustering analyses demonstrated that two distinct patient subgroups were present within each disease cohort. Further genetic analysis of the two patient subgroups in each disease highlighted 3,403 genes in the CD cohort and 4,186 genes in the UC cohort that were significantly differentially expressed between the patient subgroups.

Notably, there was a significant overlap in the gene signatures between diseases: “Indeed, this overlap was so considerable that both signatures could be used interchangeably, each being able to exactly reproduce—in the other disease cohort—the same subgroups that had been detected in the unsupervised analysis,” the authors write.

“Moreover, we were also able to generate smaller lists of genes, selected on their ability to distinguish the subgroups in half of the entire patient cohort, which could accurately designate patients with either CD or UC in the other independent half into their correct subgroups.”

In fact, the classifier gene sets performed similarly in terms of positive and negative predictive value irrespective of the methods used to generate them or whether they contained four or 100 genes, the team remarks.

“We therefore termed the corresponding patient subgroups defined by these analogous signatures as 'IBD1' (the subgroup characterized by elevated expression of the majority of differentially expressed genes) and 'IBD2' (the subgroup characterized by lower expression of these genes).” Interestingly, the team found that the IBD gene signatures also overlapped with the AAV/SLE prognostic signature previously identified.

The next step was to see whether the IBD1 and IBD2 gene signatures were associated with different disease courses in UC and CD patients. Analyses indicated that in the CD cohort, patients in the IBD1 subgroup experienced relapsing or chronically active disease and required treatment escalation more frequently than those in the IBD2 subgroup.

In addition, of all the patients needing treatment escalation at any stage, those in the IBD1 subgroup were more likely to continue experiencing active disease and require further treatment escalation. “Of note, all additional treatment escalations were due to ongoing disease activity rather than intolerance to the initial treatment,” the authors stress.

Similar results were obtained for UC patients, in that individuals in the IBD1 subgroup experienced a more aggressive disease course and had a higher incidence of recurrently active disease than those in subgroup IBD2.

To provide a direct estimate of the prognostic utility of the signature the team calculated the signature’s  specificity (CD, 89%; UC, 84%) and sensitivity (CD, 59%; UC, 77%) in terms of its ability to predict the future requirement for any treatment escalation. Based on these figures it was calculated that when the gene signatures were taken into account the probability that treatment would need to be stepped up was 83% for the IBD1 subgroup and 30% for the IBD2 subgroup.

Prestratification the probability of treatment escalation for the CD cohort overall was 48.6%. With respect to the UC cohort, the probability of treatment escalation was 40.6% prestratification. Post-test probabilities, however, were 77% for the IBD1 cohort and 16% for the IBD2 group. Significantly, the patient subgroups couldn’t be identified using any other means including clinical, serological, or laboratory parameters at enrolment or prior duration of disease.

Gene set enrichment analysis (GSEA) suggested that a number of pathways involved in T-cell activation and the development of antigen-specific T-cell memory were upregulated in IBD1 patients. These included TCR signaling, CD28 co-stimulation, IL-6 signaling, and IL-2 signaling. Because the findings implied that IBD1 and IBD2 subgroups may display differences in CD8 T-cell activation, the team then investigated whether a previously derived generic signature of CD8 T-cell activation was also enriched within the IBD1 transcriptional signature.

This confirmed that T-cell activation genes were expressed at a higher level in IBD1 patients, even though there was no detectable difference in the relative size or surface phenotype of the CD8 T-cell memory subsets.

Importantly, when the researchers used the same methods to analyze expression data from CD4 T-cell genes, the results were far more ambiguous, and any evidence of subgroups didn’t correlate with subsequent disease course or any other clinical parameter.

“The lack of clinically relevant transcriptional variation in CD4 T cells is noteworthy in the context of the CD8 T-cell prognostic signature, particularly as CD4 T cells are traditionally thought to be more important in IBD pathogenesis,” they write. Nevertheless, they admit, it is possible that the CD8 T-cell transcriptional signature is a secondary phenomenon resulting from other immunological events that drive disease pathogenesis including CD4 T-cell responses.