Hepatocellular carcinoma (HCC) is a leading cause of cancer death worldwide, the incidence of which is increasing in the United States. In this report, we analyze predictors of overall survival (OS) for a large cohort of patients diagnosed with HCC at an urban academic cancer center.
Nitin Ohri, MD, Chandan Guha, MD, PhD, Andreas Kaubisch, MD, Santiago Aparo, MD, Jonathan M. Schwartz, MD, Madhur K. Garg, MD; Montefiore Medical Center, Albert Einstein College of Medicine
Background: Hepatocellular carcinoma (HCC) is a leading cause of cancer death worldwide, the incidence of which is increasing in the United States. In this report, we analyze predictors of overall survival (OS) for a large cohort of patients diagnosed with HCC at an urban academic cancer center.
Methods: We queried our institution’s tumor registry to identify all patients diagnosed with HCC in the years 2000–2011. Patients for whom adequate demographic, clinical, and treatment data were available were included in this analysis. An institutional database was referenced to obtain a quantitative estimate of each patient’s socioeconomic status (SES) using “neighborhood” information, such as median household income, housing unit values, education level, and occupation statistics. SES was reported as a z-score compared with the national average and dichotomized at the median value for statistical analyses. Bivariate and multivariate Cox proportional hazards modeling was used to identify predictors of OS.
Results: A total of 683 patients met the eligibility criteria for this analysis. Median SES for the cohort was –3.39, with an interquartile range of –6.35 to –1.18. Median follow-up for living patients was 16.0 months. Median survival for all patients was 11.4 months, with 1-, 2-, and 3-year actuarial survival rates of 49%, 34%, and 28%, respectively. On bivariate analyses, advanced age, advanced American Joint Committee on Cancer (AJCC) stage, high Model for End-Stage Liver Disease (MELD) score, decreased serum albumin, decreased serum sodium, and lack of local/locoregional therapy were all associated with decreased OS. In the final multivariate model, statistically significant independent predictors of OS were age (hazard ratio [HR] = 1.17 per decade; 95% confidence interval [CI], 1.08–1.27; P < .001), male gender (HR = 1.34; 95% CI, 1.07–1.67; P = .010), stage (HR = 1.45 for stage 3–4 vs stage 1–2; 95% CI, 1.18–1.77; P < .001), MELD score (HR = 1.04 per unit increase; 95% CI, 1.02–1.05; P < .001), albumin (HR = 0.73 per g/dL; 95% CI, 0.63–0.84; P < .001), use of local/locoregional therapy (HR = 0.48; 95% CI, 0.38–0.60; P < .001), and SES (HR = 1.24 for patients below median value; 95% CI, 1.03–1.51; P = .027).
Conclusion: In addition to established prognostic factors, we have identified SES as an independent predictor of survival following HCC diagnosis. Additional study to understand this phenomenon is warranted.