Number of negative points: a novel method for predicting radiosensitivity in head and neck tumor cell lines

The present study was aimed at establishing a method that combines multiple factors of protein and genetic changes that enables prediction of radiosensitivity in the head and neck squamous cell carcinoma (HNSCC) cell lines. In nine HNSCC cell lines, the quantity of protein expression and the type of genetic alterations were translated into a point system, called the Number of Negative Points. The expression of 14 proteins involved in growth control and/or apoptosis was quantified using a densitometric assessment of Western blots. The blots were adjusted to actin and standardised to normal oral keratinocytes classifying them into four groups depending on the amount of protein expressed (0-3 points). Mutations of the p53 gene were classified into three groups and each mutation was given one point. Since the cell lines each had a known intrinsic radiosensitivity, a multivariate statistical calculation could then be performed to select for the combination of factors having the strongest correlation to radiosensitivity. The strongest correlation of the investigated factors was the combination of epidermal growth factor receptor, survivin and splice site/missense p53 mutations (R=0.990 and P<0.0001). No single factor had a significant correlation to the intrinsic radiosensitivity. Since a significant correlation to the intrinsic radiosensitivity was achieved only when two or more factors were combined, we conclude that a method such as the Number of Negative Points is necessary for prediction of treatment response. We present a novel method to combine factors which enables the prediction of radiosensitivity of HNSCC cell lines.