BACKGROUND: Primary clear cell carcinoma of the liver (PCCCL) is an infrequent variant of primary hepatocellular carcinoma (HCC), we retrospectively performed a large population-based cohort study to elucidate the relationships between demographic, carcinoma- and therapy-specific variables and overall survival (OS).

MATERIAL AND METHODS: The Surveillance, Epidemiology and End Results (SEER) database was queried to extract data on 419 patients with pathologically confirmed PCCCL from 1988 to 2015. A nomogram with good accuracy was formulated to predict long-term survival of PCCCL patients.

RESULTS: The OS for PCCCL patients was 25.6 months (95% confidence interval [CI]: 22.2–29 months), the overall 1-year, 3-year, and 5-year survival rates were 59.5%, 39.3%, and 29.9%, respectively. Log-rank analysis revealed that there was no statistically significant discrepancy in clinical outcome between PCCCL and common-type HCC after propensity-matched analysis. Multivariate Cox analysis confirmed that larger lesions (>96 mm), distant metastases and elevated alpha-fetoprotein (AFP) levels were independent prognostic factors for undesirable outcome. Conversely, surgery was an independent protective factor (hazard ratio [HR]=0.23, 95% CI 0.17–0.31), which significantly boosted OS by virtually 35 months (47.3 months versus 12.7 months, P<0.001). Radiotherapy or chemotherapy was not associated with OS for PCCCL patients (both P>0.05). The nomogram incorporated 4 independent prognostic factors and its concordance index for predicting survival was 0.761.

CONCLUSIONS: The prognosis of PCCCL resembled that of common-type HCC. Larger lesions, distant metastases, and enhanced AFP levels were associated with unsatisfactory prognosis. Surgery fulfill favorable prognosis while radiotherapy or chemotherapy exerted no significant effects on survival.

Keywords: Adenocarcinoma, Clear Cell, Carcinoma, Hepatocellular, nomograms, SEER Program, Aged, 80 and over, Algorithms, Cohort Studies, Databases, Genetic, Liver, Liver Neoplasms, Proportional Hazards Models, Survival Rate