6533b870fe1ef96bd12cfaef

RESEARCH PRODUCT

The role of sarcopenia in patients with intrahepatic cholangiocarcinoma: Prognostic marker or hyped parameter?

Aline Mähringer-kunzSebastian SchottenFabian BartschPeter R. GalleArndt WeinmannRoman KloecknerLukas MüllerChristoph DüberHauke LangFabian StöhrFelix Hahn

subject

MaleSarcopeniamedicine.medical_specialtyGastroenterologyResectionCholangiocarcinoma03 medical and health sciences0302 clinical medicineRisk FactorsGermanyInternal medicinemedicineHumansIn patientIntrahepatic CholangiocarcinomaAgedPsoas MusclesRetrospective StudiesUnivariate analysisHepatologybusiness.industryCancerMiddle AgedPrognosismedicine.diseaseMagnetic Resonance ImagingSurvival AnalysisPredictive valueBile Duct Neoplasms030220 oncology & carcinogenesisSarcopeniaMultivariate AnalysisCohortFemale030211 gastroenterology & hepatologyTomography X-Ray Computedbusiness

description

BACKGROUND & AIMS Sarcopenia has emerged as a prognostic parameter in numerous cancer entities. Current research favours its role as a determining factor for overall survival (OS) in patients with intrahepatic cholangiocarcinoma (ICC); however, it is unclear whether sarcopenia is a truly independent survival predictor if combined with established prognostic factors. METHODS Between 1997-2018, 417 patients with histopathologically confirmed ICC were referred to our centre, of whom 293 were included in this study. Cross-sectional imaging, laboratory examinations and histopathological reports were retrospectively analysed. Psoas muscle index (PMI) as easy-to-measure marker of sarcopenia was calculated. Using optimal stratification, sex-specific PMI cut-offs were calculated and tested in hazard regression models against previously published risk factors-for the entire cohort, and within resected and non-resected subgroups. RESULTS Median OS for patients with low respectively high PMI was 23.5 and 34.5 months in the resected subgroup (P = 0.008) and 5.1 and 7.8 months (P = 0.01) in the non-resected subgroup. In multivariate hazard regression models for the entire cohort, low PMI exhibited independent predictive value (P = 0.01) as did translobar tumour spread (P = 0.005), extrahepatic extension (P = 0.03), tumour boundary type (P < 0.001), carbohydrate antigen 19-9 (CA 19-9) levels (P = 0.001), alkaline phosphatase levels (P = 0.001) and distant metastasis (P < 0.001). In subgroup analyses, low PMI remained predictive among non-resected patients (P = 0.03), but lost its predictive value among resected patients (P = 0.15). CONCLUSIONS Psoas muscle index strongly predicted OS in univariate analysis. However, addition of established risk factors eliminated its predictive value among resected patients. Thus, when resection is deemed oncologically reasonable, patients should not be excluded from surgery because of sarcopenia alone.

https://doi.org/10.1111/liv.14132