Prognostic or Predictive Factors in Biliary Tract Cancer: A Systematic Literature Review
Author(s)
John Bridgewater, MRCO PhD FRCP1, Juan Valle, MB ChB MSc FRCP2, Wayne Su, MSc3, Javier Sabater, B Pharm4, Sam Mettam, PhD4, Lin Fan, PhD3, Sally Bowditch, BSc, MPH4, Amit Ahuja, M.Pharm LLB5, Vijay Sharma, M.Pharm LLB5, Suzy Paisley, PhD6, Ankit Rohilla, M.Pharm5, Aikyadeep Mandal, M.Pharm6, Andrea Garcia, MSc7, Farshid Dayyani, MD8, Richard Kim, MD9;
1UCL Cancer Institute, London, United Kingdom, 2Cholangiocarcinoma Foundation, Herriman, UT, USA, 3Jazz Pharmaceuticals, Philadelphia, PA, USA, 4Jazz Pharmaceuticals, Oxford, United Kingdom, 5Lumanity, Gurugram, India, 6Lumanity, London, United Kingdom, 7Lumanity, Utrecht, Netherlands, 8UCI Health Chao Family Comprehensive Cancer Center, Orange, CA, USA, 9Moffitt Cancer Center, Tampa, FL, USA
1UCL Cancer Institute, London, United Kingdom, 2Cholangiocarcinoma Foundation, Herriman, UT, USA, 3Jazz Pharmaceuticals, Philadelphia, PA, USA, 4Jazz Pharmaceuticals, Oxford, United Kingdom, 5Lumanity, Gurugram, India, 6Lumanity, London, United Kingdom, 7Lumanity, Utrecht, Netherlands, 8UCI Health Chao Family Comprehensive Cancer Center, Orange, CA, USA, 9Moffitt Cancer Center, Tampa, FL, USA
OBJECTIVES: Biliary tract cancer (BTC) is a rare and aggressive heterogenous group of tumors with poor outcomes. Identifying prognostic factors can help optimize treatment for specific patient groups, thereby improving resource allocation. This review aims to identify variables assessed as potential prognostic/predictive factors for progression-free survival (PFS) and overall survival (OS) in patients with unresectable, locally advanced/metastatic BTC.
METHODS: Embase®, MEDLINE® (using Embase.com), and Cochrane Library were searched from their inception until 24 September 2023. Studies assessing the association between variables and OS or PFS in the target population were included. Variables with statistically significant associations with OS and PFS, along with study counts for each, were summarized in an evidence map. Variables reported in >3 studies were reviewed for their prognostic/predictive association.
RESULTS: From 5,634 citations, 77 studies were included: 75 observational, one randomized controlled and one single-arm trial. Most studies (48 of 77) involved patients receiving first-line chemotherapy. Sample sizes ranged from 18-1,333 (median=106) patients, with median age ranging from 53-75 years.
Multiple variables were identified and categorized into demographics and health status, serological markers, inflammatory markers, disease stage, tumor location, treatment factors and genetic alterations. ECOG status, carcinoembryonic antigen (CEA), cancer antigen 19-9 (CA19-9), albumin and neutrophil-to-lymphocyte (NLR) ratio were repeatedly identified as independent prognostic/predictive factors for PFS and OS. Tumor location, treatment factors, including treatment regimen, prior resection, number of treatment lines, were statistically associated with PFS and OS. The significance of these associations varied between studies due to heterogeneity. The prognostic value of genetic alterations was rarely assessed and found as inconclusive.
CONCLUSIONS: This review identified key variables such as ECOG status, CEA, CA19-9, albumin, and NLR ratio as prognostic/predictive factors of PFS and OS in multiple studies. Statistical associations of other factors varied, and the prognostic value of genetic alterations was inconclusive.
METHODS: Embase®, MEDLINE® (using Embase.com), and Cochrane Library were searched from their inception until 24 September 2023. Studies assessing the association between variables and OS or PFS in the target population were included. Variables with statistically significant associations with OS and PFS, along with study counts for each, were summarized in an evidence map. Variables reported in >3 studies were reviewed for their prognostic/predictive association.
RESULTS: From 5,634 citations, 77 studies were included: 75 observational, one randomized controlled and one single-arm trial. Most studies (48 of 77) involved patients receiving first-line chemotherapy. Sample sizes ranged from 18-1,333 (median=106) patients, with median age ranging from 53-75 years.
Multiple variables were identified and categorized into demographics and health status, serological markers, inflammatory markers, disease stage, tumor location, treatment factors and genetic alterations. ECOG status, carcinoembryonic antigen (CEA), cancer antigen 19-9 (CA19-9), albumin and neutrophil-to-lymphocyte (NLR) ratio were repeatedly identified as independent prognostic/predictive factors for PFS and OS. Tumor location, treatment factors, including treatment regimen, prior resection, number of treatment lines, were statistically associated with PFS and OS. The significance of these associations varied between studies due to heterogeneity. The prognostic value of genetic alterations was rarely assessed and found as inconclusive.
CONCLUSIONS: This review identified key variables such as ECOG status, CEA, CA19-9, albumin, and NLR ratio as prognostic/predictive factors of PFS and OS in multiple studies. Statistical associations of other factors varied, and the prognostic value of genetic alterations was inconclusive.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
CO49
Topic
Clinical Outcomes
Topic Subcategory
Relating Intermediate to Long-term Outcomes
Disease
SDC: Gastrointestinal Disorders, SDC: Oncology