Goldsmith, Manning M., Cresson, David H., Arnold, Larry A., Postma, Duncan S., Askin, Frederic B., and Pillsbury, Harold C.
The prognostic significance of deoxyribonucleic acid (DNA) flow cytometry has been investigated for many solid tumors, but few data have been accumulated for squamous cell carcinomas of the head and neck. To our knowledge, we report the largest number of patients (69) with head and neck primary carcinomas to be studied by DNA flow cytometry. In the first part of this study, we reviewed 109 consecutive patients with laryngeal or hypopharyngeal primary carcinomas which were treated at North Carolina Memorial Hospital during the period of 1981 to 1984. The final analysis comprised 139 DNA histograms (mean coefficient of variation: 8.02) on paraffin-embedded specimens from 48 patients. Of the 48 patients with primary carcinomas, 24 had glottic, 18 had supraglottic, and 6 had carcinomas from the piriform sinus. Patients had follow-up for a minimum of 12 months, with a mean follow-up period of 23 months. Twenty-three of the 48 primary carcinomas (48%) were clearly aneuploid, and the remaining 52% were tetraploid (22%) or diploid (30%). We have concluded that patients with clearly aneuploid primary carcinomas had significantly better prognoses than those with diploid tumors (p= 0.008). High DNA amounts (greater than 40% of cells beyond the diploid peak, DNA G1GO) also correlated with a favorable prognosis when compared with low DNA amounts (p<0.01), and this remained significant when the clinical outcome was adjusted for staging of the primary site (T), nodal status, and stage of disease. Ploidy was the most significant prognostic variable for the laryngeal group of patients. In the second part of the study, twenty-one patients with oral cavity squamous cell carcinomas were studied in a similar fashion as the group with laryngeal carcinomas. In this group, a low DNA amount, with 40% as the cutoff point, was associated with a favorable prognosis (p= 0.024), and this remained significant while controlling for I, nodal status, and stage of disease. Numbers were too small to permit evaluation of the impact of ploidy in this group, but there was a slight trend toward aneuploidy and tetraploidy, correlating with a poor treatment outcome (p= 0.228). DNA amount was the most significant prognostic variable for the group of patients with oral cavity carcinomas. We conclude that DNA flow cytometry may be a powerful prognostic indicator in malignant conditions of the head and neck. The implications of these data for the management of head and neck cancer are discussed.