Abstract
The pharmacodynamics of taxol in human head and neck squamous cell carcinoma were studied using histocultures of surgical specimens from patients (n = 22). Tumors were treated with taxol for 24 h. The inhibition of DNA synthesis was determined by 48 h cumulative bromodeoxyuridine (BrdUrd) incorporation. The induction of apoptosis was measured by morphological changes, in situ DNA end labeling, post-exonuclease III-BrdUrd labeling, and DNA fragmentation. Inhibition of the BrdUrd labeling index (LI) by taxol was incomplete, with 11 tumors showing a maximal inhibition (Emax) of 30–50% and the remaining 11 tumors showing an Emax of 50–80%. For both groups, the inhibition approached maximum values at 1 µm taxol concentration; an additional 10-fold increase in drug concentrations did not significantly enhance the inhibition. The taxol concentrations required for a 30% inhibition (IC30) were 4.2 and 0.3 µm for the first and second groups, respectively. The IC30 correlated with the Emax (r2 = 0.39; P < 0.001). Taxol induced apoptosis in all tumors; 11 tumors showed a maximal fraction of apoptotic tumor cells between 3 and 10% and 11 tumors between 13 and 28%, whereas untreated controls showed a maximal apoptotic index of <1%. For individual tumors, the maximal apoptotic index occurred between 0.1 and 3 µm, and correlated with the BrdUrd LI for the untreated control (r2 = 0.37; P < 0.01). It is interesting that > 95% of apoptotic cells were BrdUrd labeled, whereas not all BrdUrd-labeled cells were apoptotic. To investigate the basis of the variable tumor response to taxol, we determined the expression of multidrug resistance P-glycoprotein (Pgp), p53, and bcl-2 proteins, using immunohistochemical staining and Western blot analysis. Eleven (50%), 10 (45%), and 7 (32%) tumors expressed Pgp, p53, and bcl-2, respectively. Patients with Pgp-positive tumors showed a higher number of affected lymph nodes than those with Pgp-negative tumors (P < 0.05). Compared with moderately and well differentiated tumors, the poorly differentiated tumors expressed p53 and Pgp more frequently and showed a lower maximum inhibition of DNA synthesis and a higher apoptotic fraction after taxol treatment (P < 0.05 in both cases). Pgp expression correlated differently with taxol-induced inhibition of DNA synthesis than with apoptosis; Pgp-positive tumors showed a significantly higher Emax (63%) and IC30 (4.2 µm) but also a higher apoptotic index (17%) than Pgp-negative tumors (Emax, 36%; IC30, 0.3 µm; and apoptotic index, 6%; P < 0.05 for all cases). p53 and bcl-2 expression did not correlate with taxol-induced inhibition of DNA synthesis or apoptosis. The data indicate that taxol acts through apoptosis and inhibition of proliferation in human head and neck cancer. Pgp overexpression appears to protect cells from the antiproliferative effect of taxol but correlates with a higher apoptosis.
This study was supported in part by Research Grant R37CA49816 and Research Career Development Award K04CA01497 from the National Cancer Institute, NIH, Department of Health and Human Services, and a grant from Bristol-Myers Squibb. The Ohio State University Comprehensive Cancer Center Tissue Procurement Service was supported in part by Grant P30CA16058 from the National Cancer Institute, NIH, Department of Health and Human Services.