Incidence and Risk Factors for Hepatocellular Carcinoma in Cirrhosis: The Multicenter Hepatocellular Carcinoma Early Detection Strategy (HEDS) Study.

BACKGROUND & AIMS: Worldwide, hepatocellular carcinoma (HCC) is a common malignancy. We aimed to prospectively determine the incidence and risk factors of HCC in a U.S. METHODS: The multicenter Hepatocellular Carcinoma Early Detection Strategy study of the National Institutes of Health prospectively enrolled patients with cirrhosis who underwent standard surveillance for HCC. Demographics, medical and family history, etiology of liver disease, and clinical features were evaluated for associations with HCC. RESULTS: Between April 10, 2013 and December 31, 2021, 1723 patients were enrolled and confirmed eligible. During median follow-up of 2.2 years (range, 0-8.7 years), there were 109 incident cases of HCC for an incidence rate of 2.4 per 100 person-years: 88 (81%) patients with very early/early Barcelona Clinic Liver Cancer stage (0, A), 20 (18%) intermediate stage (B), and 1 (1%) unknown stage. Risk factor analyses were restricted to 1325 patients, including 95 incident HCC, with at least 6 months of follow-up. The majority were men (53.2%), obese or severely obese (median body mass index, 30.2 kg/m2), and white (86.3%); 42.0% had history of hepatitis C virus infection, 20.7% had alcoholic liver disease, and 24.9% had nonalcoholic fatty liver disease. Fourteen risk factors for HCC were significant (P < .05) in univariate analyses, and a multivariate subset was selected using stepwise logistic regression. The multivariate subset contained gender (P < .001; male; odds ratio [OR], 2.47; 95% confidence interval [CI], 1.54-4.07), years with cirrhosis (P = .004; OR, 1.06; 95% CI, 1.02-1.1), family history of liver cancer (P = .02; yes; OR, 2.69; 95% CI, 1.11-5.86), age (per 5 years; P = .02; OR, 1.17; 95% CI, 1.03-1.33), obesity (P = .02; yes; OR, 1.7; 95% CI, 1.08-2.73), aspartate aminotransferase (log(1+AST); P = .06; OR, 1.54; 95% CI, 0.97-2.42), alpha-fetoprotein (log(1+AFP); P = .07; OR, 1.32; 95% CI, 0.97-1.77), and albumin (P = .10; OR, 0.7; 95% CI, 0.46-1.07). CONCLUSIONS: Thus far, this is the largest prospective and geographically diverse study of a U.S. cohort of patients with cirrhosis that validates known risk factors for HCC (gender, age, obesity, years with cirrhosis, family history of liver cancer, baseline AFP, albumin, and AST). The incidence of HCC was 2.4% per 100 person-years.

Barriers to Surveillance for Hepatocellular Carcinoma in a Multicenter Cohort.

Importance: Hepatocellular carcinoma (HCC) surveillance is underused in clinical practice, which may be owing to patient and clinician barriers. Objective: To characterize HCC surveillance barriers and associations with clinical outcomes in a multicenter cohort of patients with cirrhosis. Design, Setting, and Participants: This retrospective, multicenter cohort study included 5 medical centers in the United States. Patients with cirrhosis and newly diagnosed HCC treated from 2014 to 2018 were included. Data were analyzed from June 2021 to February 2022. Exposure: Surveillance completion in the 36-month period prior to HCC diagnosis. Main Outcomes and Measures: Surveillance receipt was classified as semiannual, annual, or no surveillance. Multivariable logistic regression analysis was used to identify factors associated with semiannual surveillance. We conducted multivariable logistic and Cox regression analyses to characterize associations between surveillance completion with curative treatment and overall survival. Results: A total 629 eligible patients (median [IQR] age, 63.6 [56.2-71.0] years; 491 [78.1%] men) were assessed, including 7 American Indian or Alaska Native patients (1.1%), 14 Asian patients (2.2), 176 Black patients (28.0%), 86 Hispanic patients (13.1%), and 340 White patients (54.1%). Nearly two-thirds of the cohort had no surveillance prior to HCC diagnosis (mean [range by site] 63.7% [37.9%-80.4%]), with a mean (range by site) of 14.0% (5.3%-33.3%) of patients having received semiannual surveillance and 22.3% (14.3%-28.8%) of patients having received annual surveillance. The most common reasons for no surveillance were lack of surveillance orders or nonadherence (mean [range by site], 82.4% [66.7%-92.4%], although a mean (range by site) of 17.6% (10.2%-22.1%) of patients had unrecognized cirrhosis at HCC presentation. Semiannual surveillance was associated with hepatitis B infection (odds ratio [OR], 3.06 [95% CI, 1.24-7.23]) and inversely associated with Black race (OR, 0.41 [95% CI, 0.20-0.80]) and lack of cirrhosis recognition (OR, 0.14 [95% CI, 0.02-0.46]). Semiannual HCC surveillance was significantly associated with curative treatment receipt (OR, 2.73 [95% CI, 1.60-4.70]) but not overall survival (HR, 0.81 [95% CI, 0.55-1.18]). Conclusions and Relevance: In this cohort study of patients with cirrhosis, HCC surveillance was underused in more than 80% of patients and associated with failures across the screening process. Dedicated programs to improve cirrhosis detection and HCC surveillance attainment are needed.

Generalized meta-analysis for multiple regression models across studies with disparate covariate information.

Meta-analysis is widely popular for synthesizing information on common parameters of interest across multiple studies because of its logistical convenience and statistical efficiency. We develop a generalized meta-analysis approach to combining information on multivariate regression parameters across multiple studies that have varying levels of covariate information. Using algebraic relationships among regression parameters in different dimensions, we specify a set of moment equations for estimating parameters of a maximal model through information available from sets of parameter estimates for a series of reduced models from the different studies. The specification of the equations requires a reference dataset for estimating the joint distribution of the covariates. We propose to solve these equations using the generalized method of moments approach, with the optimal weighting of the equations taking into account uncertainty associated with estimates of the parameters of the reduced models. We describe extensions of the iterated reweighted least-squares algorithm for fitting generalized linear regression models using the proposed framework. Based on the same moment equations, we also develop a diagnostic test for detecting violations of underlying model assumptions, such as those arising from heterogeneity in the underlying study populations. The proposed methods are illustrated with extensive simulation studies and a real-data example involving the development of a breast cancer risk prediction model using disparate risk factor information from multiple studies.

Partial Consistency with Sparse Incidental Parameters.

Penalized estimation principle is fundamental to high-dimensional problems. In the literature, it has been extensively and successfully applied to various models with only structural parameters. As a contrast, in this paper, we first apply this penalization principle to a linear regression model with a finite-dimensional vector of structural parameters and a high-dimensional vector of sparse incidental parameters. For the estimators of the structural parameters, we derive their consistency and asymptotic normality, which reveals an oracle property. However, the penalized estimators for the incidental parameters possess only partial selection consistency but not consistency. This is an interesting partial consistency phenomenon: the structural parameters are consistently estimated while the incidental ones cannot. For the structural parameters, also considered is an alternative two-step penalized estimator, which has fewer possible asymptotic distributions and thus is more suitable for statistical inferences. We further extend the methods and results to the case where the dimension of the structural parameter vector diverges with but slower than the sample size. A data-driven approach for selecting a penalty regularization parameter is provided. The finite-sample performance of the penalized estimators for the structural parameters is evaluated by simulations and a real data set is analyzed.