Prognostic Factors in Patients with Hepatocellular Carcinoma Receiving Systemic Chemotherapy Identification of Two Groups of Patients with Prospects for ProlongedSurvival * DANIEL C. IHDE, M.D. MARY J. MATTHEWS, M.D. ROBERT W. MAKUCH, Ph.D. K. ROBERT MclNTIRE, M.D.’ JOYCE L. EDDY, R.N. Betheslh, Marybnd LEONARD B. SEEFF, M.D. Washington, D. C. From the National Cancer InstiiaNavy Medical Oncology Branch, the Blostatistlcs and Data Management Sectlcn, and the Diagncsis Branch, National Cancer Institute and Naval Hospital, Be- thesda, Maryland, and the MedIcal Service, Vet- erans Adminlstration Medical Center and Georgetown University School of Medicine, Washington, D.C. The opinions or assertions contained herein are the prlvate views of the au- thors and are not to be construed as official or as reflectingthevlewsofth5~oftheNavy or the Department of Defense. Requests for re- prints should be addressed to Dr. Daniel C. Me, National Cancer Instkute-Navy Medical Oncology Branch, Building 8, Room 5108, Naval Hospital, Bethesda, Maryland 208 14. l Cwent sdhs.s: 15 West Main Street, Falmouth, Massachusetts 02450. Among 37 patients with hepatocelluiar carcinoma given systemic chemotherapy, 12 (32 percent) lived 14 to 37 months from Mtiatlon of treatment whereas the remainder died wlthln five months. Indl- vidual factors associated with improved survival included fully ambulatory performance status, lack of jaundice, response to chemotherapy, the flbrolamellar carcinoma pathologic variant, absence of cirrhosis, and normal serum alpha-fetoproteln levels. Patients llvlng longer than 12 months fell into two groups. Seven patients with fibrolamellar carcinoma lacked evidence of hepatitis B serum markers or cirrhosis and had normal alpha-fetoprotein levels and suqxislngly frequent extrahepatic metastases. All but one were Caucasians aged 25 years or less. The other five “long-term” survivors were all fully ambulatory wlthout jaundke, and the majority were older non-Caucasians with tumor confined to the liver at the time of diagnosis and with hepatitis B markers, elevated alpha- fetoprotein (evek, or cirrhosis. All patients without flbrolamellar carcinoma who were less than fully ambulatory or who had jaundke died quickly. Patients with fibrolamellar carcinoma have homoge- neous clinical features, and their disease follows a relatively indolent course. In other patients with hapatocellular carcinoma, W of ambulatory status and serum MM&in determination can ktentify those with some prospect of prolonged survival. The prognosis of patients with unresectable hepatocellular carcinoma is dismal, with median survival from diagnosis of only two to four months in most large series [l-4] in both the United States and western Europe, where the tumor is relatively uncommon, and in en- demic areas such as sub-Saharan Africa and China, where primary liver cancer is sometimes the most frequent malignant neoplasm. In nonendemic regions, however, a small number of patients will live one to two years or even longer without surgical therapy [5,6]. Although the majority of our patients with hepatocellular carcinoma treated with intravenous chemotherapy also died within a few months of diagnosis, the observation that almost one third lived more than 12 months prompted us to undertake a detailed analysis of factors as- sociated with relatively prolonged survival. We found that our patients with improved survival fell into two groups-(l) patients lacking many March 1985 The Amwkan Journal d Medklm Volume 79 399