Non-invasive diagnostic imaging in Crohn’s disease
Multispectral optoacoustic tomography uses haemoglobin as a proxy for the degree of inflammation in the intestinal tract
Crohn’s disease is an inflammatory condition of the gastrointestinal tract for which no cure has been developed, and the available therapeutics mostly alleviate the symptoms of the condition. The disease manifests itself with a series of symptoms common to other inflammatory bowel diseases (IBDs), such as inflammation of the intestinal mucosae, fibrogenesis, ulceration, as well as the formation of strictures and fistulae, which for most patients increases the need to undergo surgery. Diagnosing Crohn’s disease is particularly difficult because inflammation can occur in only certain tracts of the bowel, which means that the entire ileum and colon — regions within the intestine that are common sites of inflammation — need to be examined. Endoscopic imaging, specifically a colonoscopy, is at present one of the most reliable diagnostic tools, as it not only enables the visualisation of the intestinal epithelium and mucosae, but can also direct core–needle biopsy sampling. Recently developed multimodal endoscopic techniques incorporate imaging probes, such as an ultrasound transducer in combination to optical fibres. Results from clinical, endoscopic and histological data are then used to determine the disease activity score.
Reporting in the March issue of the New England Journal of Medicine, Ferdinand Knieling and co-authors evaluated the use of multispectral optoacoustic tomography (MSOT), in a combined phase I and II clinical trial, as a non-invasive diagnostic method for detecting localized inflammation from the intestines of 108 patients with Crohn’s disease. The transabdominal MSOT measurements used blood as a source of contrast and the authors measured parameters relating to total haemoglobin, oxygenated haemoglobin, deoxygenated haemoglobin and oxygen saturation. The experimental workflow included clinical evaluation by ultrasound, histopathology and a blinded colonoscopy.
High-resolution ultrasound was used to measure wall thickness and to identify anatomical regions of the colon with signal intensities corresponding to areas of inflammation; those regions were then traced using the ultrasound B-mode signal obtained via the transducer mounted on the MSOT handheld probe, which could then be switched to optoacoustic acquisition. These steps allowed the ultrasound, which has excellent penetration depth and resolution, to guide the MSOT acquisition to a defined region of interest of 10 x 15 mm and whose reconstructed echo signals have a maximal field of view of 40 x 40 mm. Given the constraints of light penetration in tissue, the MSOT imaging was performed up to 3.5 cm of depth, which meant that some patients had to be excluded from the study if no suitable signal could be established within that depth or in those with signal artefacts caused by air in the intestine — the air–tissue interface causes an index mismatch for the ultrasonic wave propagation that results in high intensity peaks originating from the refracted waves. Complete datasets were obtained for 88 patients and the MSOT parameters (Fig. 1) were then statistically compared with the three clinical scoring systems used to assess Crohn’s disease, and to differentiate between cases of active disease from patients in remission. The MSOT parameters were tested using 2-sample t-tests against the Harvey-Bradshaw index (HBI), the endoscopic score (SES-CD), the histological modified Riley score, as well as other tests including white-blood-cell count, C-reactive protein and intestinal wall thickness (US-Limberg score). The ability of the MSOT data to discriminate between disease activity states was also evaluated using a Sidaks correction for ANOVA multiple testing.
Figure 1: The MSOT and ultrasound signal is collected via a handheld probe and is sensitive to the changes in microvasculature within the intestinal wall. Oxygenated blood (HbO2) signal is detected in patients with active Crohn's disease, and is absent in cases of disease remission (Figure courtesy of iThera Medical, GmbH).
The results show that the MSOT imaging data, which reflect properties of the tissue’s vascular system, correlate well with histological and endoscopic scoring systems, but are instead distinct from the HBI scores, which are symptom-based and known to be poor predictors of mucosal inflammation. The histological scoring for Crohn’s disease includes the assessment of lymphocyte aggregates, granulomas, eosinophils, infiltrating neutrophils and other inflammatory cells, mucin depletion, crypt abscesses and other architectural irregularities, as well as surface epithelial integrity. The correlation with haemoglobin levels, attributed to the increased vascular flow typical of inflamed tissue, can be evaluated using receiver operator characteristics (ROC) analyses.
In conclusion, the MSOT data from Crohn’s disease patients demonstrated that the technique is highly sensitive to tissue inflammation. This result is in accordance with other clinical data published recently showing ultra-broadband optoacoustic sensitivity to inflammation in the epidermis of psoriasis patients in which a scoring system was derived using the optoacoustic signal that correlated with the psoriasis area severity index. However, the difference in the Crohn’s disease study is the demonstration that MSOT diagnostics can be achieved also within the human gut wall, an area that is particularly challenging to image. The potential impact of this imaging technique in the clinic is high, given its speed (the MSOT acquisition takes about 10 minutes) and the non-invasive nature of the procedure.
Knieling et al., “Multispectral Optoacoustic Tomography for Assessment of Crohn’s Disease Activity”, N. Engl. J. Med., 376, 1292-1294 (2017).
Waldner et al., “Multispectral Optoacoustic Tomography in Crohn's Disease: Noninvasive Imaging of Disease Activity”, Gastroenterol., 151 (2), 238-240 (2016).
Aguirre et al., “Precision assessment of label-free psoriasis biomarkers with ultra-broadband optoacoustic mesoscopy”, Nat Biomed. Eng., 1, 0068 (2017).
Powell et al., “The mucosal immune system: master regulator of bidirectional gut–brain communications”, Nat. Rev. Gastroenterol. Hepatol., 14, 143–159 (2017).