Hepatic arterial infusion of verapamil and doxorubicin with complete hepatic venous isolation and extracorporeal chemofiltration: pharmacological evaluation of reduction in systemic drug exposure and assessment of hepatic toxicity.

Tumour resistance to chemotherapeutic drugs through expression of the multidrug resistance phenotype is a major impediment in the treatment of hepatic malignancies. We performed hepatic arterial infusion of verapamil (at a dose known to block P-glycoprotein activity) and doxorubicin combined with complete hepatic venous isolation and extracorporeal chemofiltration in non-tumour-bearing pigs with normal livers to evaluate the pharmacology and toxicology of this drug combination. The complete hepatic venous isolation-chemofiltration system significantly reduced system exposure to both verapamil and doxorubicin (P < 0.01). Hepatic arterial infusion of verapamil (2 mg/kg) alone did not result in hepatocellular toxicity. However, the combination of verapamil and doxorubicin (3 mg/kg or 5 mg/kg) produced significant elevations in liver enzymes (P < 0.01), and gross histological evidence of liver damage in 90% of the treated animals. The results of this study indicate that hepatic arterial infusion of verapamil and doxorubicin, in an attempt to improve treatment response in unresectable liver tumours expressing the multidrug resistance phenotype, may not be tolerated by patients with limited hepatic reserve.

Assessment of Ki-67-derived tumor proliferative activity in colorectal adenocarcinomas.

Analysis of tumor growth fraction (TGF) has become essential as cell cycle-directed modalities have been increasingly used for the treatment of solid neoplasms. We studied TGF in fresh tissue samples from 74 surgically resected colorectal carcinomas (50 colon and 24 rectum; 44 men and 30 women) by flow cytometry (FCM) and by immunostaining using Ki-67 monoclonal antibody. In 12 cases, samples of uninvolved colonic mucosa adjacent to tumor (transitional mucosa) and samples of normal mucosa (at least 5 cm away from tumor) were available for growth fraction analysis. The mean Ki-67 and S-phase values were 17.1% (range, 0-60%) and 17.5% (range, 3-39%), respectively. The mean percentage of Ki-67 positivity in tumor samples from women was significantly lower than that in samples from men (P = 0.001). Ki-67-derived TGF values, however, did not correlate with patient age, lymph node status, or tumor size, site, stage, degree of differentiation, or DNA ploidy. The correlation between Ki-67-derived and FCM-derived TGF values was statistically significant (P = 0.001) but marginal (R = 0.35). In both transitional and normal colonic mucosa samples, Ki-67 positivity was mainly confined to the lower half of the crypts, and their growth fraction values were significantly lower than those of tumor tissue; however, there was no difference in values between transitional and normal mucosae. Our results indicate that Ki-67-derived TGF does not correlate with known prognostic factors for colorectal carcinoma; however, long-term follow-up information will be necessary to define the clinical value of Ki-67 staining.