Serum Biomarkers as Prognostic Factors for Metastatic Sarcoma

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review


Aims: To investigate the prognostic value of combining inflammatory biomarkers in a prognostic index (Aarhus composite biomarker score: ACBS), adjusted for known confounders, including comorbidity, in patients with metastatic sarcoma. Materials and methods: All patients diagnosed with metastatic sarcoma from 1993 until 2008 were extracted from the Aarhus sarcoma database. The levels of serum albumin, C-reactive protein, serum sodium, haemoglobin, neutrophils and lymphocytes were collected. ACBS as well as the neutrophil to lymphocyte ratio (NLR), Glasgow prognostic score (GPS) and a combined score of GPS and NLR known as CNG were calculated. The prognostic importance of the biomarkers on disease-specific mortality was analysed. Adjustments were made for age, comorbidity, histological type and site of metastasis using the Cox proportional hazard model. Harrell's concordance index (C-index) was used to evaluate whether the ACBS adds prognostic information to already known prognostic factors. The data were validated using the bootstrapping method. Results: In total, 265 patients with metastatic sarcoma were included. The 2-year disease-specific mortality was 74% (95% confidence interval 68–80) and 79% (95% confidence interval 68–88) for soft-tissue sarcoma and bone sarcoma, respectively. Comorbidity was present in 21% of soft-tissue sarcoma patients and 13% of the bone sarcoma patients. All six biomarkers were independent prognostic factors. The various scoring systems (NLR, GPS, CNG and ACBS) combining more than one biomarker were also prognostic for disease-specific mortality. Conclusion: The biomarker scoring systems are independent prognostic factors for adult patients with metastatic sarcoma. However, a modified ACBS was superior to all the other scoring systems in predicting outcome.

TidsskriftClinical Oncology
Sider (fra-til)242-249
Antal sider8
StatusUdgivet - 2019

Se relationer på Aarhus Universitet Citationsformater

ID: 148718245