The addition of tumour-stroma ratio to the 8th AJCC staging system improves survival prediction in oral tongue squamous cell carcinoma.

2020 
BACKGROUND One of the objectives of current researches is to customise the treatment of cancer patients. This objective requires stratification of patients based on the most significant prognostic factors. The aim of this study was to evaluate the prognostic value of tumour-stroma ratio (TSR), defined as the proportion of tumour cells relative to surrounding stroma, in patients with Oral Tongue Squamous Cell Carcinoma (OTSCC) and to develop a prognostic nomogram based on the most significant clinico-pathological features. METHODS Clinico-pathological data of 211 patients treated at "Ospedali Riuniti" General Hospital (Ancona, Italy) for OTSCC were collected. 139 patients were re-staged according to the 8th AJCC edition. Evaluation of TSR was performed on haematoxylin and eosin-stained slides and correlation with survival outcomes evaluated. In addition, aiming to integrate the independent value of TSR with the 8th edition of AJCC, a prognostic nomogram for OTSCC has been developed. RESULTS OTSCC with low TSR (i.e. high proportion of stroma and low proportion of tumour cells) showed to have a negative prognostic value in terms of disease-specific survival with a hazard ratio (HR) of 1.883 and 95% confidence interval (CI) of 1.033-3.432 (P=0.039) and overall survival (HR=1.747, 95% CI 0.967-3.154; P=0.044) independently from other histological and clinical parameters. For the cohort of 139 patients re-staged according to the 8th AJCC edition, variables correlating with a poor prognosis were: TSR, perineural invasion, and gender. The nomogram built on these parameters showed a good predictive capacity, over performing the 8th AJCC staging system in stratifying disease-specific survival in OTSCC. CONCLUSIONS Including TSR in the predictive model could improve risk stratification of OTSCC patients and aid in treatment decision.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    49
    References
    17
    Citations
    NaN
    KQI
    []