Imaging Fibrosis in Crohn's Disease

A survey of promising methods for imaging to detect and measure intestinal fibrosis in Crohn's disease.

Transcript

1. Imaging for the Detection and Measurement of Intestinal Fibrosis Peter D.R. Higgins University of Michigan @ibddoctor
2. Agenda • What are we Measuring? • Why Measure Fibrosis? • Molecular Imaging • Magnetization Transfer-MRI • Photoacoustic Imaging • Ultrasound • Strain imaging • Shear Wave imaging • Vascular Imaging by MRI
3. What Are We Measuring? • Pathology • Increased deposition of extracellular matrix • Collagen, fibronectin, etc. • Increased activation of intestinal myofibroblasts • Synthesis of ECM and contractile proteins (aSMA, MYLK) • Increased bowel wall muscle – reaction to obstruction • Increased thickness, interspersed with ECM – both hypertrophy and hyperplasia • Loss of organization: layers of circular, longitudinal muscle less clear • Evidence of inflammation – often refractory to medications • Why – altered blood flow? Hypoxia → increased vascular branching. • Does ECM tension affect immune activation?
4. Are Strictures Inflamed or Fibrotic? 4 Normal ileal margin Surgically resected stricture Arrows: Smooth muscle hyperplasia interspersed with blue-stained fibrosis Chen, W., et al. JCC 2017; 92-104.
5. Strictures Are Inflamed AND Fibrotic • Histopathologic studies show: • A mixture of inflammation (chronic>active) and fibrosis • Damaged,highly branched blood vessels and slowed blood flow • Dramatic hypertrophy of smooth muscle • Loss of organization of muscle layers • Infiltrated by fibrosis • Creeping fat • Neuronal hypertrophy 5 AND Hypermuscular AND Fat-Wrapped AND Hypersensitive
6. The Problem with Intestinal Fibrosis • Inflammation initiates intestinal fibrosis • A gene expression signature of fibrogenesis at diagnosis predicts future surgery in children with Crohn’s • Over time, fibrosis becomes independent of inflammation • Not detected by conventional CT, MR Preclinical inflammatory (minimal symptoms) Symptomatic inflammatory (not yet diagnosed) IBD diagnosis made (start treatment) Bowel Damage Accumulating Inflammation Improves Fibrosis Keeps Going Hyams, J., et al. Lancet 2017; 389: 1710-1718.
7. Why Measure Fibrosis? • To identify bowel damage. • To identify progression of bowel damage over time. • To determine which patients are unlikely to respond to medical anti-inflammatory therapy and need mechanical therapy. • Endoscopic balloon dilation • Surgery • To determine if medical anti-fibrotic therapies are effective. • Presumes that we can identify and develop effective therapies.
8. An Ideal Fibrosis Measurement tool • Reproducible, certifiable quality at each site • Widely available, low operator-dependence • Can detect/differentiate several levels of progression of fibrosis • None, mild, moderate, severe • Easily integrated into clinical workflow • Low cost • Low invasiveness • Low radiation/toxicity to patient • Accepted by FDA/EMA as a surrogate endpoint
9. Molecular Imaging MT-MRI and Photoacoustic Imaging
10. Magnetization Transfer MRI • Protons of free water interact with protons in large macromolecules, such as collagen and myosin (quantified using MT ratio [MTR]) • Signal from both fibrosis and accumulated muscle in the bowel wall. Dillman, JR, et al. JMRI 2015;42:801-10.
11. Inflammation/Fibrosis Ratio • Inflammation – T2 bright • Fibrosis – MTR high • Ratio: T2WSI/MTR AuROC = 0.98 Acute TNBS Chronic TNBS Dillman, JR, et al. JMRI 2015;42:801-10.
12. MT-MRI before Surgery • 31 consecutive patients scanned before surgery Li, et al. Radiology. 2018 May;287(2):494-503.
13. Magnetization Transfer MRI Strengths • Objective measures of collagen and myosin • Promising sensitivity and specificity. • Can separate multiple levels • Can evaluate entire bowel in one pass. Limitations • MRI very expensive (in US) • Needs external validation at multiple centers.
14. Photoacoustic Imaging Laser light at 532 nm causes Hgb to be energized then relax, releasing ultrasound waves detected by the probe CD in remission CD inflamed Waldner, et al. Gastro 2016;151:238–240 HgbO2 signal
15. PA Imaging: Rabbit Intestinal Fibrosis 532 nm (red) = Hgb 1310 nm (green) = Collagen
16. Longitudinal PA Study in Rabbits 532 nm (red) = Hgb 1310 nm (green) = Collagen
17. PA Imaging – Transcutaneous Patient Scans
18. Endoscopic PA imaging
19. Photoacoustic Imaging Strengths • Objective measures of hemoglobin and collagen • Detects longitudinal change in rabbit model • Transcutaneous PA imaging is feasible in humans Limitations • Transcutaneous PA imaging limited to BMI < 35 • Endoscopic probe requires invasive scope • Needs external validation • For focal lesions, impractical for entire bowel
20. Mechanical Properties and US Stiffness of Bowel Revealed with Sound
21. Mechanical Stiffness with Ultrasound • Strain during bowel compression = strain elastography • Stiffer tissues compress less • Softer tissues compress more • Shear Wave Velocity • Sound waves travel faster through stiff tissue
22. Detection of Intestinal Fibrosis • Measure bowel wall mechanical properties • Stiff walls • Ultrasound Elasticity Imaging (UEI) -During bowel wall compression, ultrasound can track and map compressibility Normal soft bowel wall Fibrotic firm bowel wall
23. Native B-scan Ultrasound Strain Map Overlay Strictured Ileum (Transverse View) Normal bowel (Transverse View) UEI in Crohn’s Disease
24. Unit (%) (a) (b) Unit (dB) 2.0 cm 2.0 cm 2.0 cm - 60 - 40 - 20 - 8 - 6 - 4 - 2 - 20 - 16 - 12 - 8 - 4 - 60 - 40 - 20 (a) (b) Unit (%) Unit (dB) 3.4 cm 3.4 cm 2.4 cm - 20 - 16 - 12 - 8 - 4 - 60 - 40 - 20 (a) (b) Unit (%) Unit (dB) 3.0 cm 3.0 cm 2.0 cm Subject A Subject D Subject C - 8 - 6 - 4 - 2 Unit (%) - 60 - 40 - 20 Unit (dB) (a) (b) 2.0 cm 2.0 cm 2.0 cm Subject B UEI Strain Estimates Mean Strain % (Std Dev) Fibrotic bowel -2.6 (0.67) Adjacent bowel -6.3 (1.9) Fibrotic vs. Normal UEI strain significantly differ p = 0.03 UEI Results from 16 CD Patients Strain % = % change in volume during compression Stidham, RW, et al., Gastroenterology 2011,141: 819-826.
25. Shear Wave Velocity Detects Bowel Wall Fibrosis in mens A. Johnson2, Jonathan M. Rubin1, Peter D. R. Higgins2 Internal Medicine, 3Pathology, Ann Arbor, Michigan, USA FIGURE 3 Mean SWS measurements. Box plots comparing mean SWS measurements for low (0-2) vs. high (3) fi brosis score bowel segments obtained using VTQ shear wave elastography method. Whiskers represent minimum and maximum SWS measurements. *** p <0.001. S (non-fi brotic). Q shear wave elastography of a 20 cting CD. (A) VTQ shear wave elastography 0.91 [95%, 0.67-0.99; p<0.0001 method (area under the curve = 0 FIGURE 5 Relationship between VTQ and VT-IQ measurements. Scatter plot showing relationship Ultrasound Pulse The pulse initiates a shear wave In the tissue The ultrasound probe Can detect the shear wave And measure its speed. The shear wave speed (SWS) is Proportional to the stiffness of the tissue
26. Rapidly Growing Literature • Interobserver agreement in Ultrasound in Crohn’s disease • Inflamm Bowel Dis. 2018 Apr 28 • Good agreement for lesion location, fistula, phlegmon, abscess • Poor for mesenteric adipose tissue alteration, lesion extent, stenosis, and prestenotic dilation. • Real-time Elastography correlates with fibrosis in CD • Radiology. 2015 Jun;275(3):889-99. • Shear wave elastography differentiates inflammation from fibrosis in CD • Inflamm Bowel Dis. 2018 Sep 15;24(10):2183-2190.
27. Ultrasound Strengths • Measures tissue stiffness • Inexpensive • Noninvasive • Relatively easy to do longitudinally Limitations • Challenging at higher BMI • Concerns about operator dependence, applied pressure • Variation in stricture • How many ROIs needed to measure reliably in one slice? • How many points along the length of a stricture? • For focal lesions, impractical for entire bowel
28. MRE and Vascular Properties Is there abnormal blood flow/contrast enhancement in intestinal fibrosis?
29. Delayed Enhancement Rimola, J. Panes, J, et al. AJG 110:432, 2015 Enhancement by gadolinium 70 seconds 7 minutes • Evaluated 44 segments in 41 patients undergoing surgical resection • The degree of histopathologic fibrosis correlated with • The percentage of enhancement gain (P<0.01) • The pattern of enhancement at 7 min (P<0.01) • Using percentage of enhancement gain, MRI discriminated between mild–moderate and severe fibrosis • Sensitivity of 0.94 and a specificity of 0.89.
30. Persistence of Bowel Findings on MRE • Evaluated 28 patients in endoscopic remission in 73 segments • Despite endoscopic remission, persistent findings • Wall thickening in 23/72 segments with BWT at baseline • Strictures in 6/8 • Creeping fat in 7/10 • Intramural fat deposits in 20/24 • Despite resolution of inflammation, many lesions consistent with bowel damage persist on MRE Rimola, J. Panes, J, et al. APT 48:1232, 2018
31. Imaging for Intestinal Fibrosis • Conventional imaging (x-ray, CT, MR) not very helpful • Do not directly measure fibrosis • Can detect residual damage (fat, wall thickness, strictures) • Molecular imaging – MT-MRI, PA imaging promising • Mechanical properties – ultrasound promising for focal lesions, had good features for longitudinal monitoring • MR washout – can detect severe fibrosis • Could be useful for surgical decisions
32. Thank You