King's College London

Research portal

A clinical prognostic model compared to the newly adopted UICC staging in an independent validation cohort of P16 negative/positive head and neck cancer patients

Research output: Contribution to journalArticle

Jacob H Rasmussen, Katrin Håkansson, Gregers B Rasmussen, Ivan R Vogelius, Jeppe Friborg, Barbara M Fischer, Søren M Bentzen, Lena Specht

Original languageEnglish
Pages (from-to)52-60
Number of pages9
Early online date17 Apr 2018
Publication statusPublished - 1 Jun 2018

Bibliographical note

Copyright © 2018 Elsevier Ltd. All rights reserved.

King's Authors


OBJECTIVES: A previously published prognostic model in patients with head and neck squamous cell carcinoma (HNSCC) was validated in both a p16-negative and a p16-positive independent patient cohort and the performance was compared with the newly adopted 8th edition of the UICC staging system.MATERIALS AND METHODS: Consecutive patients with HNSCC treated at a single institution from 2005 to 2012 were included. The cohort was divided in three. 1.) Training cohort, patients treated from 2005 to 2009 excluding patients with p16-positive oropharyngeal squamous cell carcinomas (OPSCC); 2.) A p16-negative validation cohort and 3.) A p16-positive validation cohort. A previously published prognostic model (clinical model) with the significant covariates (smoking status, FDG uptake, and tumor volume) was refitted in the training cohort and validated in the two validation cohorts. The clinical model was used to generate four risk groups based on the predicted risk of disease recurrence after 2 years and the performance was compared with UICC staging 8th edition using concordance index.RESULTS: Overall 568 patients were included. Compared to UICC the clinical model had a significantly better concordance index in the p16-negative validation cohort (AUC = 0.63 for UICC and AUC = 0.73 for the clinical model; p = 0.003) and a borderline significantly better concordance index in the p16-positive cohort (AUC = 0.63 for UICC and 0.72 for the clinical model; p = 0.088).CONCLUSION: The validated clinical model provided a better prognostication of risk of disease recurrence than UICC stage in the p16-negative validation cohort, and similar prognostication as the newly adopted 8th edition of the UICC staging in the p16-positive patient cohort.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454