NMAs of Systemic Therapies for Advanced Biliary Tract Cancer: A Systematic Review and Critical Appraisal Using RoB NMA
Author(s)
Charlotte Ahmadu, MS1, Tiyi Morris, PhD2, Oliver T. Darlington3.
1Initiate Consultancy Ltd, Aberdeen, United Kingdom, 2Initiate Consultancy, London, United Kingdom, 3Head of HEOR, Initiate Consultancy, NA, United Kingdom.
1Initiate Consultancy Ltd, Aberdeen, United Kingdom, 2Initiate Consultancy, London, United Kingdom, 3Head of HEOR, Initiate Consultancy, NA, United Kingdom.
OBJECTIVES: Biliary tract cancers (BTC) are rare, aggressive malignancies and comprise of approximately 15% of all primary liver cancers. Due to their detection at unresectable or metastatic disease, they are highly fatal and associated with poor prognosis. First-line (1L) systemic treatment for untreated advanced BTCs include chemotherapies and targeted therapies, as single agents or in combination regimens. Network meta-analyses (NMAs) investigating the most effective 1L strategy have been published. This study aimed to conduct a qualitative appraisal of these NMAs using the Risk of Bias (RoB) in NMA tool.
METHODS: A systematic search of published literature via Ovid was undertaken on 23rd May 2025 in Embase, MEDLINE, and the Cochrane Database of Systematic Reviews. Searches of Google Scholar, Epistemonikos, and citation tracking supplemented. NMAs investigating the effectiveness of 1L strategies for advanced BTC were included. Two reviewers independently selected records and assessed methodological quality and robustness using ROBIS and RoB NMA (PROSPERO protocol registration: CRD420251059270).
RESULTS: Of the 15 records screened, 4 NMAs fulfilled the inclusion criteria and were qualitatively appraised according to review methodology, interventions and network geometry, effect modifiers, and statistical synthesis. While systematic methods were cited, according to ROBIS, review processes for 75% of NMAs were of unclear/ high risk of bias. Although Gem+Cis and GEMOX were common comparators, ROB NMA identified limitations in reporting potential treatment effect modifiers, publication bias and differences in primary studies (low scores for 75% of NMAs).
CONCLUSIONS: To our knowledge, this is the first published study utilising the ROB NMA to assess robustness. Patients, clinicians and policy makers should be aware of the substantial differences in the reporting quality of published 1L advanced BTC NMAs prior to applying relative efficacy assessments and conclusions to inform clinical decision-making. The confidence in systematic review plus NMA results can be further evaluated using CINeMA or GRADE NMA.
METHODS: A systematic search of published literature via Ovid was undertaken on 23rd May 2025 in Embase, MEDLINE, and the Cochrane Database of Systematic Reviews. Searches of Google Scholar, Epistemonikos, and citation tracking supplemented. NMAs investigating the effectiveness of 1L strategies for advanced BTC were included. Two reviewers independently selected records and assessed methodological quality and robustness using ROBIS and RoB NMA (PROSPERO protocol registration: CRD420251059270).
RESULTS: Of the 15 records screened, 4 NMAs fulfilled the inclusion criteria and were qualitatively appraised according to review methodology, interventions and network geometry, effect modifiers, and statistical synthesis. While systematic methods were cited, according to ROBIS, review processes for 75% of NMAs were of unclear/ high risk of bias. Although Gem+Cis and GEMOX were common comparators, ROB NMA identified limitations in reporting potential treatment effect modifiers, publication bias and differences in primary studies (low scores for 75% of NMAs).
CONCLUSIONS: To our knowledge, this is the first published study utilising the ROB NMA to assess robustness. Patients, clinicians and policy makers should be aware of the substantial differences in the reporting quality of published 1L advanced BTC NMAs prior to applying relative efficacy assessments and conclusions to inform clinical decision-making. The confidence in systematic review plus NMA results can be further evaluated using CINeMA or GRADE NMA.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
CO171
Topic
Clinical Outcomes
Disease
Oncology