Bidimensional measurement of nasopharyngeal carcinoma: a simple method to predict outcomes

To determine whether the standard techniques of measuring tumour size could be applied to the measurement of nasopharyngeal carcinoma.

A retrospective review of case notes from the Buddhist Tzu Chi Dalin General Hospital archives was performed.

The Buddhist Tzu Chi Dalin General Hospital is a teaching hospital in Taiwan.

All patients with nasopharyngeal carcinoma were included.

Ninety-eight patients with newly diagnosed nasopharyngeal carcinoma were treated with high-dose radiotherapy and chemotherapy were enrolled in this study. Computed tomography-derived primary tumour volume, bidimensional measurement and unidimensional measurement were recorded. Intrarater reliability was measured. To examine the validity of various measurements, we estimated the Spearman's correlation co-efficients between those measurements and the gold standard value (primary tumour volume measurement with summation of area techniques). Univariate and multivariate analyses were performed and Kaplan-Meier survival curves constructed.

There was a significant association between primary tumour volume and bidimensional measurement with respect to tumour size at diagnosis (Spearman's correlation co-efficient = 0.845, P < 0.001). Using age, gender, chemotherapy status and T-stage as covariate, bidimensional measurement remained an independent prognostic factor for any relapse [Hazard ratio = (HR) 1.066; P = 0.029], and overall survival (HR = 1.097; P = 0.007). Patients with small bidimensional measurement (<10 cm(2)) had better prognosis and fewer recurrences.

When using simple measurement to evaluate nasopharyngeal carcinoma, the bidimensional measurement may be used to measure size at diagnosis. Patients with small bidimensional measurement had better prognosis and fewer recurrences. Bidimensional measurement may be further considered to improve the current staging system.