Goodchild, George; (2025) The diagnosis and management of biliary strictures. Doctoral thesis (M.D(Res)), UCL (University College London).

Preview

Text Goodchild_Thesis.pdf - Submitted Version Download (5MB) | Preview

Abstract

Biliary strictures are considered indeterminate when the aetiology remains elusive despite standard diagnostic work-up. Misdiagnosis can have a dramatic impact on patient outcomes. Cholangiocarcinoma and IgG4 related disease (IgG4-RD) are two diseases which often mimic each other. Distinction can be challenging but is vital as treatment approaches and prognosis are very different. We describe the experience of the first inter-regional specialist IgG4-RD multidisciplinary team meeting (MDM) and demonstrate that the MDM improved awareness of a rare disease and enabled important diagnostic and management decisions as well as serving as a platform for patients to access third line treatment with Rituximab. As part of the evaluation of indeterminate biliary strictures, digital single-operator cholangioscopy (d-SOC) with cholangioscopic biopsies is often used to provide both a visual assessment of the stricture and to obtain targeted biopsies. Some studies suggest a higher sensitivity for visual impression compared to histology, although in these studies endoscopists were unblinded to previous investigations. We investigated the diagnostic accuracy and inter-observer agreement (IOA) of d-SOC in the visual assessment of biliary strictures when blinded to background information. An international, multicentre, cohort study was performed. Cholangioscopic videos in patients with a known diagnosis were systematically scored. Anonymised videos were reviewed by 19 experts in 2 steps: 1: blinded for patients’ history and investigations and 2: unblinded. Forty-four videos of 19 benign and 25 malignant cases were reviewed. The sensitivity and specificity for the diagnosis of malignancy was 74.2% and 46.9% (blinded) and 72.7% and 62.5% (unblinded). Cholangioscopic certainty of a malignant diagnosis led to over-diagnosis (sensitivity 90.6%, specificity 33%), especially if no background clinical information was provided. The IOA for the presence of malignancy was fair for both assessments (Fleiss’ kappa (κ) 0.245 (blinded) and κ 0.321 (unblinded)). For individual visual features, the IOA ranged from slight to moderate for both assessments (κ 0.059 – 0.400 versus κ 0.031 – 0.452). Our study shows low sensitivity and specificity for blinded and unblinded d-SOC video appraisal of indeterminate biliary strictures, with significant inter-observer variation. Whilst reaching consensus on the optical features of biliary strictures remains important, optimising visually-directed biopsies may be the most important role of cholangioscopy in biliary stricture assessment. Whilst cholangioscopy and visually targeted biopsies remain an important tool in the assessment of patients with possible cholangiocarcinoma, this invasive test has considerable associated costs and carries some risks to the patient. Several non-invasive tests are currently available for clinical use, such as serum CA19-9 and CEA However, these lack sufficient sensitivity and specificity for standalone use. There is an urgent unmet need for further reliable non-invasive markers for the diagnosis of cholangiocarcinoma (CCA). Cytokeratin fragment 21-1 (CYFRA) and Mucin 5AC (MUC5AC) have reported utility for differentiating CCA from benign disease. Herein, serum levels of these markers were tested in 81 malignant cases verses 89 healthy controls regardless of time to diagnosis. Neither marker was significantly elevated in malignant cases versus benign disease (median 2.25ng/ml versus 1.95ng/ml p=0.89). Next, to assess the potential utility of both markers for early diagnosis of biliary tract cancer, levels were performed on a set of 89 pre-diagnosis serum samples taken from 55 cases of cancer and 91 matched benign controls obtained from the UKCTOCS biobank. The median time from sample collection to diagnosis was 31.5 months. Serum levels of CYFRA21-1 and MUC5AC in CCA samples, as measured by ELISA, were comparable to cancer free controls at 0-1 years, 1-2 years, 2-3 years and >3 years prior to diagnosis. When stratifying by time to diagnosis, both MUC5AC and CYFRA remained unchanged across all time groups. When combined into panels with previously described biomarkers, a panel combining CA19-9, ALP and CYFRA21-1 reached an AUC (cross validated) of 0.826 (95% CI 0.641-0.962) and sensitivity of 0.778 (95% CI 0.444-1) for the diagnosis of CCA.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item