Objective: To determine the effectiveness of hepatic artery occlusion alone and with sequenced chemotherapy for patients with hepatic-dominant metastases of islet cell carcinomas and carcinoid tumors.
Design: Nonrandomized, observational study with follow-up from 2.5 to 10 years.
Patients: 111 ambulatory patients referred to a multidisciplinary tertiary care center who had histologically proven islet cell carcinoma or carcinoid tumor and symptomatic measurable metastatic lesions in the liver or hormonal abnormalities or both. The patients were ambulatory but were having substantial symptoms because of their endocrine syndromes or their tumors.
Intervention: All patients had hepatic artery occlusion done surgically or by catheterization and embolization. After this procedure, 71 patients were selected for chemotherapy with alternating two-drug regimens of doxorubicin plus dacarbazine and streptozocin plus fluorouracil. Main outcome measures of response to therapy were rates of tumor regression, rates of improvement in endocrine abnormalities, symptomatic improvement, and duration of favorable response.
Results: Objective regressions were observed in 60% of patients treated with occlusion alone and in 80% with chemotherapy added. Regressions were associated with substantial or complete relief from the endocrine syndromes. With occlusion alone, the median duration of regression was 4.0 months and with chemotherapy added, it was 18.0 months. Any comparative inferences about the two treatment regimens must be guarded, because this was not a randomized trial and marked differences occurred in the distribution of prognostic factors between the patient groups. Side effects of arterial occlusion included fever, nausea, pain, and abnormalities in liver function. Side effects of chemotherapy included nausea, vomiting, leukopenia, and alopecia.
Conclusions: Hepatic arterial occlusion can frequently produce major regression of neuroendocrine tumors with relief from the hormonal syndromes. Sequential chemotherapy may improve the rate and duration of the regression.