Sex Differences in Mortality Among Patients With Hepatocellular Carcinoma

Taotao Yan; Ming-Lun Yeh; Jennifer Leong; Hidenori Toyoda; Hiroshi Abe; Dae Won Jun; Lewis Roberts; Yasuhito Tanaka; Myron Schwartz; Yao-Chun Hsu; Charles Landis; Phillip Vutien; Haruki Uojima; Takashi Honda; Maria Buti; Huy Trinh; Eiichi Ogawa; Satoshi Yasuda; Cheng-Hao Tseng; Mayumi Maeda; Elsa Sola; Takanori Suzuki; Chung-Feng Huang; Pei-Chien Tsai; Chia-Yen Dai; Jee-Fu Huang; Wan-Long Chuang; Ramsey Cheung; Ming-Lung Yu; Mindie H Nguyen

doi:10.1001/jamanetworkopen.2026.37313

. 2026 Oct 2;9(10):e2637313. doi: 10.1001/jamanetworkopen.2026.37313

Sex Differences in Mortality Among Patients With Hepatocellular Carcinoma

Taotao Yan ^1,^2,³, Ming-Lun Yeh ^4,^5,⁶, Jennifer Leong ⁷, Hidenori Toyoda ⁸, Hiroshi Abe ⁹, Dae Won Jun ¹⁰, Lewis Roberts ¹¹, Yasuhito Tanaka ^12,¹³, Myron Schwartz ⁷, Yao-Chun Hsu ¹⁴, Charles Landis ¹⁵, Phillip Vutien ¹⁶, Haruki Uojima ¹⁷, Takashi Honda ¹⁸, Maria Buti ^19,²⁰, Huy Trinh ²¹, Eiichi Ogawa ²², Satoshi Yasuda ⁸, Cheng-Hao Tseng ¹⁴, Mayumi Maeda ¹, Elsa Sola ¹, Takanori Suzuki ²³, Chung-Feng Huang ^4,^5,⁶, Pei-Chien Tsai ⁴, Chia-Yen Dai ^4,^5,⁶, Jee-Fu Huang ^4,^5,⁶, Wan-Long Chuang ^4,^5,⁶, Ramsey Cheung ^1,²⁴, Ming-Lung Yu ^4,^5,⁶, Mindie H Nguyen ^1,^25,^✉

¹Division of Gastroenterology and Hepatology, Stanford University Medical Center, Palo Alto, California

²Department of Infectious Diseases, The First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China

³National Regional Infectious Diseases Center, Clinical Research Center for Infectious Diseases of Shaanxi, Xi’an, China

⁴Hepatobiliary Division, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan

⁵Hepatitis Research Center, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan

⁶Center for Metabolic Disorders and Obesity, Kaohsiung Medical University, Kaohsiung, Taiwan

⁷Henry D. Janowitz Division of Gastroenterology, Mount Sinai Health System, New York, New York

⁸Department of Gastroenterology, Ogaki Municipal Hospital, Ogaki, Gifu, Japan

⁹Division of Gastroenterology and Hepatology, Shinmatsudo Central General Hospital, Chiba, Japan

¹⁰Department of Internal Medicine, Hanyang University Hospital, Hanyang University College of Medicine, Seoul, South Korea

¹¹Department of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota

¹²Department of Gastroenterology and Hepatology, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan

¹³Virology and Liver Unit, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan

¹⁴Division of Gastroenterology of Hepatology, E-Da Cancer Hospital/I-Shou University, Kaohsiung, Taiwan

¹⁵Kaiser Permanente Central Hospital, Seattle, Washington

¹⁶Division of Gastroenterology, University of Washington School of Medicine, Seattle

¹⁷Department of Gastroenterology, Internal Medicine, Kitasato University School of Medicine, Sagamihara, Japan

¹⁸Department of Gastroenterology and Hepatology, Nagoya University Graduate School of Medicine, Nagoya, Japan

¹⁹Liver Unit, Department of Internal Medicine, Hospital Universitari Vall d’Hebron and Universitat Autònoma de Barcelona, Barcelona and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas del Instituto Carlos III, Barcelona, Spain

²⁰Liver Diseases Group, Vall d’Hebron Research Institute, Barcelona, Spain

²¹San Jose Gastroenterology, San Jose, California

²²Department of General Internal Medicine, Kyushu University Hospital, Fukuoka, Japan

²³Department of Gastroenterology and Metabolism, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan

²⁴Division of Gastroenterology and Hepatology, Veterans Affairs Palo Alto Health Care System, Palo Alto, California

²⁵Department of Epidemiology and Population Health, Stanford University, Palo Alto, California

Accepted for Publication: July 31, 2026.

Published: October 2, 2026. doi:10.1001/jamanetworkopen.2026.37313

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Yan T et al. JAMA Network Open.

^✉

Corresponding Author: Mindie H. Nguyen, MD Division of Gastroenterology and Hepatology, Stanford University Medical Center, 780 Welch Rd, CJ250K, Palo Alto, CA 94304 (mindiehn@stanford.edu).

Author Contributions: Dr Yan and Prof Nguyen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Yan, Schwartz, Buti, Tsai, Dai, Yu, Nguyen.

Acquisition, analysis, or interpretation of data: Yan, Yeh, Leong, Toyoda, Abe, Jun, Roberts, Tanaka, Hsu, Landis, Vutien, Uojima, Honda, Buti, Trinh, Ogawa, Yasuda, Tseng, Maeda, Sola, Suzuki, C. Huang, J. Huang, Chuang, Cheung, Nguyen.

Drafting of the manuscript: Yan, Buti, Maeda, Nguyen.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Yan, Chuang, Nguyen.

Obtained funding: Roberts.

Administrative, technical, or material support: Yeh, Toyoda, Abe, Roberts, Tanaka, Hsu, Landis, Honda, Trinh, Maeda, Tsai, Chuang, Cheung, Nguyen.

Supervision: Jun, Dai, Yu, Nguyen.

Conflict of Interest Disclosures: Dr Roberts reported receiving grants from the National Cancer Institute during the conduct of the study; receiving grants from Exact Sciences/Abbott, Gilead Sciences, Glycotest, and Innovo Bioanalysis and personal fees from Exact Sciences/Abbott and Springer Nature outside the submitted work; and having a patent licensed to Abbott. Dr Hsu reported receiving grants from Gilead Sciences and personal fees from Gilead Sciences, GSK, Sysmex, AstraZeneca, AusperBio, AbbVie, Bristol-Myers Squibb, Gilead Sciences, Grand Marquis IVD, Grifols, Merck Sharp & Dohme, and Roche outside the submitted work. Dr Buti reported receiving grants from Gilead Sciences outside the submitted work. Dr Ogawa reported receiving grants from Gilead Sciences outside the submitted work. Dr Dai reported receiving personal fees from Roche Pharmaceuticals and AstraZeneca outside the submitted work. Dr J. Huang reported received grants and personal fees from Roche, Sysmex, Boehringer Ingelheim, Novo Nordisk, and Aligos and speaker for AbbVie, Gilead Sciences, Merck, Sysmex, Boehringer Ingelheim, and Novo Nordisk. Dr Yu reported receiving grants from the Ministry of Health and Welfare of Taiwan, AbbVie, Abbott Diagnostic, Bristol-Myers Squibb, Gilead Sciences, Merck, Novo Nordisk, and Roche diagnostics and personal fees from AbbVie, Abbott Diagnostic, Bristol-Myers Squibb, Gilead Sciences, Novo Nordisk, Roche, and Roche Diagnostics outside the submitted work. Dr Nguyen reported receiving grants from Pfizer, AstraZeneca, GSK, Delfi, Innogen, Exact Sciences, CurveBio, Gilead Sciences, Helio Genomics, the National Institutes of Health, Roche, and Aligos and personal fees from GSK outside the submitted work. No other disclosures were reported.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC13633531 PMID: 42826017

This cohort study examines the association of sex with mortality among patients with hepatocellular carcinoma.

Key Points

Question

Do sex differences exist in mortality in hepatocellular carcinoma (HCC)?

Findings

In this cohort study of 8594 patients with HCC, male sex was independently associated with an increased risk of overall mortality only in certain subgroups (Asian, Asia region, cirrhosis, and viral–related HCC subgroups). Within the nonviral subgroup, male vs female patients with alcohol-associated liver disease had increased mortality risk, while no difference was observed for those with metabolic dysfunction–associated steatotic liver disease.

Meaning

These findings support incorporation of sex-based considerations in HCC management while also accounting for differences in patient ethnicity and disease etiology.

Abstract

Importance

Sex is a well-recognized factor associated with hepatocellular carcinoma (HCC) incidence, but its association with mortality is unclear.

Objective

To assess the association between sex and the risk of overall, liver-related, and non–liver-related mortality among patients with HCC.

Design, Setting, and Participants

This retrospective cohort study was conducted among patients with HCC in 17 centers from 5 countries and regions (2000-2023). Male and female patients with HCC who had balanced baseline characteristics via inverse probability of treatment weighting (IPTW) were included, with a propensity score–matched cohort included for sensitivity analyses. Data were analyzed between July 10 and September 8, 2025.

Exposures

Male vs female sex.

Main Outcomes and Measures

Overall, liver-related, and non–liver-related mortality.

Results

Among 8594 patients with HCC (mean [SD] age, 63.3 [10.9] years; 6214 male [72.3%]; 5644 Asian [65.7%], 458 Black [5.3%], 615 Hispanic [7.2%], and 1784 White [20.8%]), 4567 were from Asia (53.1%), 6327 had cirrhosis (73.6%), and 6617 had viral etiology (77.0%). In the unmatched cohort, males had a higher overall mortality rate per 100 person-years than females in the total cohort (15.9 [95% CI, 15.3-16.4] vs 13.3 [95% CI, 12.5-14.1]; P < .001) and in Asian (14.1 [95% CI, 13.5-14.7] vs 11.1 [95% CI, 10.3-12.0]; P < .001), Asia region (14.4 [95% CI, 13.6-15.2] vs 10.8 [95% CI, 9.9-11.8]; P < .001), cirrhosis (17.6 [95% CI, 16.9-18.3] vs 14.4 [95% CI, 13.5-15.4]; P < .001), and hepatitis B (13.6 [95% CI, 12.7-14.5] vs 10.0 [95% CI, 8.6-11.5]; P < .001) and C (15.8 [95% CI, 15.0-16.6] vs 13.4 [95% CI, 12.4-14.6]; P < .001) subgroups. In the IPTW-matched cohort, male sex was associated with increased overall mortality in the full cohort (adjusted hazard ratio [aHR], 1.14 [95% CI, 1.05-1.23]; P = .002) and in Asian (aHR, 1.22 [95% CI, 1.10-1.36]; P < .001), Asia region (aHR, 1.23 [95% CI, 1.09-1.39]; P < .001), cirrhosis (aHR, 1.18 [95% CI, 1.08-1.29]; P < .001), and viral (aHR, 1.15 [95% CI, 1.05-1.26]; P = .004) subgroups. Additionally, sex differences varied by liver disease etiology, with male sex having the highest aHR in the association with overall mortality among individuals with alcohol-associated liver disease (aHR, 1.90 [95% CI, 1.02-3.54]; P = .04), followed by those with hepatitis B (aHR, 1.22 [95% CI, 1.01-1.47]; P = .04) or C (aHR, 1.13 [95% CI, 1.02-1.26]; P = .03), but there was no association in patients with metabolic dysfunction–associated steatotic liver disease. There was no significant sex difference for non–liver-related mortality.

Conclusions and Relevance

In this study, sex disparities in HCC mortality existed but were heterogeneous across ethnic background and liver disease etiologies. These findings support incorporation of sex-based considerations in HCC management.

Introduction

Liver cancer, predominantly hepatocellular carcinoma (HCC), is the sixth most common cancer and the third leading cause of cancer-related deaths worldwide.¹ The global burden of liver cancer is projected to nearly double by 2050, with an estimated 1.52 million new cases and 1.37 million deaths.²

Sex is an important factor associated with HCC development, progression, and treatment outcomes.^{3,4,5,6,7,8,9,10,11} Males have a disproportionate incidence of HCC compared with females, at a 2:1 to 3:1 ratio across most regions.^3,4,5 Prior studies also reported significant sex disparities in liver disease etiologies, tumor characteristics, treatment receipt, and clinical outcomes in HCC.^{6,7,8,9,10,11} However, evidence regarding the association of sex with HCC mortality is sparse and inconsistent, likely due to single-center, -country, or -region design and lack of adequate adjustment for background differences between males and females.^{8,11,12,13,14} For example, a US study of 1110 patients with HCC from 2 medical centers reported a better survival outcome in females compared with males, with female sex independently associated with lower mortality (adjusted hazard ratio [aHR], 0.82; 95% CI, 0.68-0.98),⁸ while another study of a single-center US cohort of 1886 patients found no association between sex and overall survival (aHR, 1.01; 95% CI, 0.82-1.24).¹² Among studies in other regions, a cohort study of 694 patients with HCC from a tertiary French liver center reported an association between female sex and improved survival outcome (aHR, 1.39; 95% CI, 1.07-1.81),¹³ while another single-center study of 1716 patients from Singapore prospectively enrolled over 3 decades found no significant survival differences by sex (aHR for females compared with males, 0.96; 95% CI, 0.79-1.17).¹⁴ In another study of 4175 patients with HCC who underwent curative-intent hepatectomy or ablation from 12 centers in China, female sex was associated with better cancer-specific survival but only among patients who were middle aged (ages 45-59 years) and not among younger or older age groups.¹¹

To address current knowledge gaps, we comprehensively investigated the association of sex with overall, liver-related, and non–liver-related mortality in a large, multinational, clinical practice cohort of patients with HCC and long-term follow-up. We balanced baseline characteristics of male and female groups using inverse probability of treatment weighting (IPTW) in main analyses and propensity score matching (PSM) in a sensitivity analysis. These data may help identify disparities for targeted intervention and inform future treatment of patients with HCC.

Methods

Study Design and Population

This is a retrospective cohort study of adult (aged ≥18 years) patients with HCC between January 1, 2000, and December 31, 2023, in 17 centers from Japan, South Korea, Spain, Taiwan, and the US within the Real-World Evidence From the Global Alliance for the Study of HCC consortium (REAL-HCC). Details regarding HCC diagnosis, study population enrollment, and data collection are described in the eMethods in Supplement 1. Data analyses were conducted between July 10 and September 8, 2025. This study was approved by the Institutional Review Board at Stanford University and each participating study center and is reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Formal written informed consent from individual patients was waived by the Institutional Review Board at Stanford University due to the retrospective design and retrospective data collection of the study and because all data were fully anonymized prior to analysis.

Study Outcomes and Variables

Primary study outcomes were incidence of overall, liver-related, and non–liver-related mortality. Further details are described in the eMethods in Supplement 1. Race and ethnicity were self-reported by patients during clinical encounters and recorded in the electronic health record (EHR). The EHR recorded the following race and ethnicity options: Asian, Black, Chinese, Filipino, Hispanic, Japanese, Korean, Malaysian, Middle Eastern, Native American, Pacific Islander, South Asian, Southeast Asian, Taiwanese, Vietnamese, White, and mixed race. For analysis, Asian subgroups (Chinese, Korean, Japanese, Taiwanese, Vietnamese, Malaysian, Filipino, South Asian, Southeast Asian, and other Asian) were combined into a single Asian category. Remaining groups with small numbers, including Middle Eastern, Native American, Pacific Islander, and mixed race, were combined into an other category. Patients with missing race and ethnicity data were classified as unknown. Race and ethnicity were included to account for known racial and ethnic disparities in HCC mortality.¹⁵

Statistical Analysis

Primary analyses were performed in unmatched and IPTW-matched cohorts. First, baseline characteristics were described and compared by sex in the unmatched cohort. Then, IPTW was applied to balance baseline characteristics between female and male patients. Next, mortality analyses were performed in unmatched and IPTW-matched cohorts. Finally, Cox regression analyses were used to assess the association between sex and mortality. Additionally, sensitivity analyses were performed in the PSM-matched cohort. Relevant methodological details are described in the eMethods in Supplement 1. Statistical significance was defined as a 2-tailed P < .05. All analyses were conducted using R statistical software version 4.4.2 (R Project for Statistical Computing).

Results

Baseline Characteristics

In the initial cohort of 10 178 patients with HCC, we excluded 1584 patients (1150 who lacked Albumin-Bilirubin and 434 who lacked Barcelona Clinic Liver Cancer [BCLC] data), and 8594 patients (mean [SD] age, 63.3 [10.9] years; 6214 male [72.3%] and 2380 female [27.7%]; 5644 Asian [65.7%], 458 Black [5.3%], 615 Hispanic [7.2%], and 1784 White [20.8%]) met inclusion criteria and were included in the analysis (Figure 1). Among the overall population, there were 4567 patients (53.1%) from Asia (53.1%), 6327 patients (73.6%) with cirrhosis and 2267 patients without cirrhosis [26.4%], and 6617 patients with viral etiology (77.0%) and 1977 patients with nonviral etiology [23.0%]. Most HCC was viral related among Asian patients (4568 cases [80.9%]) and those who lived in the Asia region (3678 cases [80.5%]). There was substantial nonviral HCC among White (591 patients [33.1%]) and Hispanic (264 patients [43.0%]) populations. There were differences in baseline characteristics of patients in the included cohort vs those in the excluded cohort and between male and female patients of the included cohort (Table 1; eResults in Supplement 1). IPTW yielded a weighted (matched) cohort of 5898 males and 2350 females with HCC with generally balanced background (Table 1; eResults in Supplement 1).

Flowchart of hepatocellular carcinoma cohort selection and mortality counts by sex. Light gray rectangular nodes connected by thin gray arrows on a white background. Top center node: 10178 Adult patients aged greater than or equal to 18 y with H C C, 2000 to 2023. A rightward arrow from the top node points to an upper right exclusion node: 1584 Patients excluded with insufficient data; 1150 A L B I score unavailable; 434 B C L C stage unavailable. A downward arrow from the top node points to a center node: 8594 Patients with H C C included in the analysis. From this center node, a horizontal connector splits into two downward branches. Left branch to a mid left node: 2380 Female group, then a downward arrow to a lower left node: 2350 Females matched for age, race and ethnicity, cirrhosis, etiology, and B C L C stage using I P T W method. From this matched female node, three downward arrows lead to three bottom nodes, left to right: 1111 Overall mortality events with a superscript a; 859 Liver mortality events; 74 Nonliver mortality events. Right branch to a mid right node: 6214 Male group, then a downward arrow to a lower right node: 5898 Males matched for age, race and ethnicity, cirrhosis, etiology, and B C L C stage using I P T W method. From this matched male node, three downward arrows lead to three bottom nodes, left to right: 3126 Overall mortality events with a superscript a; 2379 Liver mortality events; 187 Nonliver mortality events. — ABLI indicates Albumin-Bilirubin; BCLC, Barcelona Clinic Liver Cancer; HCC, hepatocellular carcinoma; IPTW, inverse probability of treatment weighting.

^aOf 1111 all-cause deaths in the female group, 178 died from unknown cause. Of 3126 all-cause deaths in the male group, 560 died from unknown cause.

Table 1. Baseline Characteristics of Patients.

Characteristic	Patients, No. (%)
	Unmatched cohort					IPTW-matched cohort^a
	Overall (N = 8594)	Female (n = 2380)	Male (n = 6214)	SMD	P value	Female (n = 2349.7)	Male (n = 5897.8)	SMD	P value
Age, mean (SD), y	63.3 (10.9)	66.4 (10.7)	62.1 (10.8)	0.40	<.001	63.6 (10.9)	63.9 (10.4)	0.02	.45
Race and ethnicity
Asian	5644 (65.7)	1623 (68.2)	4021 (64.7)	0.11	.001	1554.2 (66.1)	3934.7 (66.7)	.02	.98
Black	458 (5.3)	137 (5.8)	321 (5.2)			131.8 (5.6)	321.8 (5.5)
Hispanic	615 (7.2)	174 (7.3)	441 (7.1)			156.9 (6.7)	399.2 (6.8)
White	1784 (20.8)	418 (17.6)	1366 (22.0)			477.5 (20.3)	1174.8 (19.9)
Other or unknown	93 (1.1)	28 (1.2)	65 (1.0)			29.3 (1.2)	67.4 (1.1)
Geographic region
Non-Asia	4027 (46.9)	1034 (43.4)	2993 (48.2)	0.10	<.001			0.03	.28
Asia	4567 (53.1)	1346 (56.6)	3221 (51.8)	0.10	<.001	1287.2 (54.8)	3146.2 (53.3)	0.03	.28
Diabetes	2577 (30.1)	679 (28.6)	1898 (30.6)	0.04	.08	635.1 (27.2)	1834.5 (31.2)	0.09	.001
Cirrhosis	6327 (73.6)	1797 (75.5)	4530 (72.9)	0.06	.02	1687.7 (71.8)	4332.4 (73.5)	0.04	.18
CTP score, mean (SD)^b	7.32 (1.69)	7.25 (1.64)	7.35 (1.71)	0.06	.05	7.38 (1.67)	7.24 (1.68)	0.08	.01
Child class^b
A	1975 (33.7)	557 (33.5)	1418 (33.7)	0.07	.09			0.13	<.001
B	2671 (45.5)	789 (47.4)	1882 (44.8)			749.8 (48.2)	1766.2 (44.0)
C	1222 (20.8)	319 (19.2)	903 (21.5)			336.5 (21.6)	797.0 (19.8)
Liver disease etiology
HBV	2425 (28.2)	483 (20.3)	1942 (31.3)	0.41	<.001	671.9 (28.6)	1653.4 (28.0)	0.03	.96
HCV	4017 (46.7)	1298 (54.5)	2719 (43.8)			1117.0 (47.5)	2869.6 (48.7)
HBV and HCV	175 (2.0)	44 (1.8)	131 (2.1)			42.7 (1.8)	120.9 (2.1)
ALD	529 (6.2)	62 (2.6)	467 (7.5)			92.9 (4.0)	224.1 (3.8)
MASLD	388 (4.5)	178 (7.5)	210 (3.4)			112.1 (4.8)	277.6 (4.7)
Other^c	139 (1.6)	67 (2.8)	72 (1.2)			42.1 (1.8)	99.4 (1.7)
Cryptogenic	921 (10.7)	248 (10.4)	673 (10.8)			271.0 (11.5)	652.9 (11.1)
BCLC stage
0/A	4789 (55.7)	1536 (64.5)	3253 (52.3)	0.25	<.001			0.01	1.00
B	1736 (20.2)	389 (16.3)	1347 (21.7)			457.4 (19.5)	1148.8 (19.5)
C	1367 (15.9)	289 (12.1)	1078 (17.3)			354.7 (15.1)	902.1 (15.3)
D	702 (8.2)	166 (7.0)	536 (8.6)			183.9 (7.8)	463.1 (7.9)
ALBI score, mean (SD)	−2.24 (0.68)	−2.23 (0.65)	−2.25 (0.69)	0.04	.15	−2.22 (0.67)	−2.27 (0.68)	0.07	.01
ALBI grade
1	2870 (33.4)	733 (30.8)	2137 (34.4)	0.11	<.001			0.07	.03
2	4721 (54.9)	1399 (58.8)	3322 (53.5)			1334.6 (56.8)	3171.7 (53.8)
3	1003 (11.7)	248 (10.4)	755 (12.1)			269.0 (11.4)	658.1 (11.2)
ALT, median (IQR), U/L	45.0 (29.0-75.0)	40.0 (27.0-67.0)	47.0 (30.0-78.0)	0.17	<.001	41.0 (27.0-67.0)	46.0 (30.0-78.0)	0.15	<.001
AST, median (IQR), U/L	55.0 (36.0-89.0)	56.0 (36.0-87.0)	55.0 (36.0-90.8)	0.08	.58	55.0 (36.0-88.0)	54.0 (36.0-89.0)	0.05	.95
TB, median (IQR), mg/dL	0.90 (0.64-1.39)	0.90 (0.60-1.30)	0.90 (0.66-1.40)	0.04	<.001	0.90 (0.60-1.30)	0.90 (0.64-1.35)	0.01	.04
Albumin, mean (SD), g/dL	3.59 (0.69)	3.55 (0.65)	3.60 (0.71)	0.07	.003	3.55 (0.67)	3.62 (0.69)	0.097	<.001
Creatinine, median (IQR), mg/dL	1.10 (1.02-1.20)	1.10 (1.01-1.20)	1.10 (1.02-1.20)	0.03	.35	1.10 (1.01-1.20)	1.10 (1.01-1.20)	0.04	.15
Platelet count, mean (SD), ×10³/μL	147.5 (89.1)	136.2 (81.6)	151.8 (91.4)	0.18	<.001	140.9 (85.6)	150.5 (90.5)	0.096	.001
Primary HCC therapy
No treatment	897 (10.4)	250 (10.5)	647 (10.4)	0.16	<.001			0.09	.10
Liver transplant	107 (1.2)	33 (1.4)	74 (1.2)			424.1 (18.1)	993.4 (16.8)
Resection	2083 (24.2)	549 (23.1)	1534 (24.7)			562.9 (24.0)	1510.7 (25.6)
Local ablation	1477 (17.2)	509 (21.4)	968 (15.6)			47.0 (2.0)	98.3 (1.7)
TACE/Y90	3699 (43.0)	965 (40.5)	2734 (44.0)			987.1 (42.0)	2514.7 (42.6)
Systemic therapy	157 (1.8)	37 (1.6)	120 (1.9)			37.1 (1.6)	64.7 (1.1)
Other or unknown	174 (2.0)	37 (1.6)	137 (2.2)			32.5 (1.4)	120.2 (2.0)

Open in a new tab

Abbreviations: ALBI, Albumin-Bilirubin; ALD, alcohol-associated liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BCLC, Barcelona Clinic Liver Cancer; CTP, Child-Turcotte-Pugh; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; IPTW, inverse probability of treatment weighting; MASLD, metabolic dysfunction–associated steatotic liver disease; SMD, standardized mean difference; TACE/Y90, transarterial chemoembolization (with or without yttrium-90 radioembolization); TB, total bilirubin.

SI conversion factors: To convert albumin to grams per liter, multiply by 10; ALT and AST to microkatals per liter, multiply by 0.0167; creatinine to micromoles per liter, multiply by 88.4; platelet count to ×10⁹ per liter, multiply by 1.0; TB to micromoles per liter, multiply by 17.104.

^{^a}

Weights were derived using age, race and ethnicity, cirrhosis, etiology, and BCLC stage as covariates.

^{^b}

Among patients with cirrhosis.

^{^c}

Includes autoimmune hepatitis, primary biliary cirrhosis, primary sclerosing cholangitis, hemochromatosis, Wilson disease, and others.

Mortality Analyses

In the total unmatched cohort, 3315 deaths (2522 liver related, 196 not liver related, and 597 of unknown cause) occurred in males after a total of 20 901.8 person-years follow-up and 1134 deaths (877 liver related, 75 not liver related, and 182 of unknown cause) occurred in females after a total of 8534.8 person-years follow-up. Males had a significantly higher overall mortality rate per 100 person-years than females (15.9 [95% CI, 15.3-16.4] vs 13.3 [95% CI, 12.5-14.1]; P < .001) (eTable 1 in Supplement 1). The 1-, 5-, and 10-year cumulative overall survival rates were lower in males compared with females (79.6% [95% CI, 78.5%-80.6%] vs 84.2% [95% CI, 82.7%-85.7%], 42.8% [95% CI, 41.4%-44.3%] vs 49.7% [95% CI, 47.4%-52.1%], and 29.5% [95% CI, 27.9%-31.2%] vs 34.5% [95% CI, 31.8%-37.5%], respectively; all P < .001) (Figure 2A), with consistent findings of a higher overall mortality rate in males compared with females by age group, presence of cirrhosis (17.6 [95% CI, 16.9-18.3] vs 14.4 [95% CI, 13.5-15.4]; P < .001), hepatitis B virus (HBV; 13.6 [95% CI, 12.7-14.5] vs 10.0 [95% CI, 8.6-11.5]; P < .001) and hepatitis C virus (HCV; 15.8 [95% CI, 15.0-16.6] vs 13.4 [95% CI, 12.4-14.6]; P < .001), and in viral and nonviral subgroups (eTable 1 and eFigures 1-3 in Supplement 1). Consistent findings for males compared with females were also observed in Asian (14.1 [95% CI, 13.5-14.7] vs 11.1 [95% CI, 10.3-12.0];, P < .001) and Asia region (14.4 [95% CI, 13.6-15.2] vs 10.8 [95% CI, 9.9-11.8];, P < .001) subgroups (Figure 3A-B; eTable 1 and eFigures 4-5 in Supplement 1). Additionally, similar disparities with higher mortality rates in males were observed for liver-related mortality rate, but no significant sex difference was observed for non–liver-related mortality rate (eTable 1 in Supplement 1; Figure 2A and Figure 3A-B). Mortality analyses data in the total IPTW-matched cohort and relevant subgroups are provided in the eResults and eTable 2 in Supplement 1 and Figure 2B.

Six-panel cumulative incidence curves by sex for overall, liver, and nonliver mortality. Two-row, three-column figure with six line charts. Top row labeled A Unmatched; bottom row labeled B Matched. Each chart has vertical axis labeled Cumulative incidence ranging from zero to zero point eight, and horizontal axis labeled Follow-up, y ranging from zero to 15. A legend in the upper left chart maps Female to an orange line and Male to a blue-gray line; both lines have semi-transparent shaded bands around them. Panel A, left chart titled Overall mortality: male curve above female across follow-up; at approximately 3 years male about zero point four five and female about zero point four; at 6 years male about zero point six and female about zero point five five; at 15 years both near zero point seven five to zero point seven eight. Text annotation P less than .001 near mid-plot. Numbers at risk beneath: Female 2380, 1019, 438, 194, 91, 32; Male 6214, 2360, 1077, 509, 261, 111. Panel A, middle titled Liver-related mortality: similar separation with male higher; around 6 years male about zero point five and female about zero point four five; around 15 years male about zero point six seven and female about zero point six five. Annotation P less than .001. Numbers at risk same as left. Panel A, right titled Non-liver-related mortality: both curves low and close, reaching about zero point one five to zero point two by 15 years; annotation P = .61. Numbers at risk same as left. Panel B repeats the three titles with similar axes and colors. Panel B left Overall mortality: male above female; annotation P less than .001; numbers at risk Female 2350, 982, 446, 216, 110, 44; Male 5898, 2307, 1034, 466, 235, 92. Panel B middle Liver-related mortality: annotation P less than .001; numbers at risk match panel B left. Panel B right Non-liver-related mortality: curves close with annotation P = .65; numbers at risk match panel B left. — Rates are given in unmatched and inverse probability of treatment weighting–matched cohorts.

Data figure with two forest plot panels of 10-year survival rates by subgroup. Two panels labeled A and B, each containing three side-by-side forest plots with horizontal axes labeled 10-y Survival rate (95% C I), percent. Orange circles denote Female and dark teal squares denote Male, with horizontal whiskers for 95% C I; a legend appears at the lower left of panel B. Panel A title Race and ethnicity. Left plot header Overall mortality with a P value column at far right; x-axis ticks at 0, 20, 40, 60. Row labels Asian, Black, Hispanic, White, each with two lines of 10-year survival rate text. Asian: Female 40.4 (36.9-44.3), Male 33.6 (31.5-35.9), P less than .001. Black: Female 13.6 (7.1-26.2), Male 17.2 (12.2-24.1), P .81. Hispanic: Female 29.4 (21.9-39.4), Male 25.5 (20.8-31.3), P .47. White: Female 21.9 (17.0-28.2), Male 23.6 (20.9-26.7), P .39. Middle plot header Liver-related mortality, same x-axis; Asian 47.6 (43.8-51.6) versus 40.9 (38.6-43.4), P less than .001; Black 19.1 (10.6-34.3) versus 30.8 (24.0-39.5), P .62; Hispanic 43.5 (34.0-55.6) versus 42.5 (36.1-50.0), P .88; White 36.9 (30.0-45.4) versus 40.5 (36.8-44.4), P .52. Right plot header Non-liver-related mortality with x-axis ticks at 40, 60, 80, 100; Asian 93.8 (91.2-96.5) versus 91.9 (89.9-94.0), P .11; Black 81.2 (62.0-100) versus 73.6 (60.0-90.4), P .63; Hispanic 87.6 (77.9-98.5) versus 89.9 (84.3-95.9), P .99; White 82.8 (74.3-92.2) versus 89.3 (85.8-93.0), P .37. Panel B title Liver disease etiology. Left plot Overall mortality with x-axis ticks at 20, 40, 60, 80; Viral 36.3 (33.2-39.6) versus 30.8 (29.0-32.6), P less than .001; Nonviral 27.2 (21.3-34.7) versus 24.1 (20.3-28.6), P .01; H B V 49.3 (43.6-55.9) versus 36.3 (33.3-39.6), P .001; H C V 32.9 (29.4-36.8) versus 27.7 (25.5-30.0), P less than .001; A L D 45.1 (30.6-66.5) versus 21.8 (16.5-28.8), P .06; M A S L D 25.5 (16.4-39.7) versus 33.8 (25.5-44.8), P .83. Middle plot Liver-related mortality, same x-axis; Viral 45.1 (41.8-48.8) versus 41.2 (39.2-43.4), P less than .001; Nonviral 38.9 (31.5-48.1) versus 37.5 (33.0-42.6), P .01; H B V 57.6 (52.0-63.8) versus 42.9 (39.7-46.4), P less than .001; H C V 41.0 (37.0-45.4) versus 40.2 (37.6-43.0), P .04; A L D 64.9 — ALD indicates alcohol-associated liver disease; HBV, hepatitis B virus; HCV, hepatitis C virus; MASLD, metabolic dysfunction-associated steatotic liver disease.

Association Between Sex and Mortality

In the IPTW-matched cohort, multivariable Cox regression analyses (model 1) adjusted for age, region, diabetes, etiology, and BCLC stage showed that male sex was independently associated with increased risks of overall (aHR, 1.14 [95% CI, 1.05-1.23]; P = .002) and liver-related (aHR, 1.14 [95% CI, 1.04-1.24]; P = .01) mortality, while no association was observed for non–liver-related mortality (Table 2; eTable 3 and eFigure 6 in Supplement 1). However, associations between sex and mortality differed across subgroups. In younger (aged <65 years), Asian (aHR, 1.22 [95% CI, 1.10-1.36]; P < .001), Asia region (aHR, 1.23 [95% CI, 1.09-1.39]; P < .001), cirrhotic (aHR, 1.18 [95% CI, 1.08-1.29]; P < .001), and viral–related HCC (aHR, 1.15 [95% CI, 1.05-1.26]; P = .004) subgroups, male sex was independently associated with increased risk of overall mortality (aHRs ranged from 1.15 for viral etiology to 1.23 for Aisa region subgroups). Male sex was also associated with increased risk of liver-related mortality in these subgroups (aHRs ranged from 1.14 [95% CI, 1.03-1.27] for viral etiology to 1.24 [95% CI, 1.10-1.39] for Asian subgroups). No sex association was observed for either mortality outcome in older (aged ≥65 years), non-Asian, non–Asia region, noncirrhotic, and nonviral HCC subgroups. In etiology-specific subgroups, aHRs in the association between male sex and increased risk of overall mortality were highest in alcohol-associated liver disease (ALD)–related HCC (aHR, 1.90 [95% CI, 1.02-3.54]; P = .04), followed by HBV-related HCC (aHR, 1.22 [95% CI, 1.01-1.47]; P = .04) and HCV-related HCC (aHR, 1.13 [95% CI, 1.02-1.26]; P = .03). No sex association was observed in HCC related to metabolic dysfunction–associated steatotic liver disease (MASLD). Male sex was also associated with increased risk of liver-related mortality in HBV-related HCC. The overall mortality risk in males compared with females increased progressively with advancing BCLC stage, with aHRs of 1.12 (95% CI, 1.01-1.24) in stage 0/A, 1.21 (95% CI, 1.02-1.44) in stage B, and 1.31 (95% CI, 1.04-1.66) in stage D. No sex association was observed for non–liver-related mortality risk in any subgroup, except for subgroups of older patients and those with HCV-related HCC. In multivariable model 2, which was further adjusted for HCC therapy, similar findings were observed in the total cohort and across most subgroups (Table 2; eTable 3 and eFigure 6 in Supplement 1).

Table 2. Associations Between Sex and Overall Mortality in Matched Cohort.

Population	Group	Event, No.	Univariable HR (95% CI)	P value	Multivariable model 1, aHR (95% CI)^a	P value	Multivariable model 2, aHR (95% CI)^b	P value
Total	Female	1111	1 [Reference]	<.001	1 [Reference]	.002	1 [Reference]	<.001
Total	Male	3126	1.16 (1.08-1.26)	<.001	1.14 (1.05-1.23)	.002	1.16 (1.07-1.25)	<.001
Age, y
<65	Female	408	1 [Reference]	.001	1 [Reference]	.01	1 [Reference]	.01
<65	Male	1749	1.21 (1.08-1.36)	.001	1.18 (1.04-1.34)	.01	1.17 (1.03-1.32)	.01
≥65	Female	703	1 [Reference]	.03	1 [Reference]		1 [Reference]	.01
≥65	Male	1377	1.11 (1.01-1.22)	.03	1.11 (1.00-1.22)	.05	1.16 (1.04-1.28)	.01
Race and ethnicity
Asian	Female	657	1 [Reference]	<.001	1 [Reference]	<.001	1 [Reference]	<.001
Asian	Male	1818	1.24 (1.12-1.37)	<.001	1.22 (1.10-1.36)		1.21 (1.09-1.35)	<.001
Black	Female	92	1 [Reference]	.48	1 [Reference]	.50	1 [Reference]	.33
Black	Male	213	1.11 (0.83-1.48)	.48	0.91 (0.68-1.21)	.50	0.87 (0.65-1.16)	.33
Hispanic	Female	104	1 [Reference]	.69	1 [Reference]	.25	1 [Reference]	.23
Hispanic	Male	252	1.05 (0.82-1.36)	.69	1.17 (0.90-1.52)	.25	1.17 (0.91-1.52)	.23
White	Female	252	1 [Reference]	.76	1 [Reference]	.41	1 [Reference]	.14
White	Male	811	1.02 (0.88-1.19)	.76	1.07 (0.91-1.25)	.41	1.13 (0.96-1.32)	.14
Geographic region
Non-Asia	Female	610	1 [Reference]	.18	1 [Reference]		1 [Reference]	.10
Non-Asia	Male	1779	1.07 (0.97-1.19)	.18	1.07 (0.96-1.19)	.21	1.09 (0.98-1.22)	.10
Asia	Female	501	1 [Reference]	<.001	1 [Reference]	<.001	1 [Reference]	.001
Asia	Male	1347	1.26 (1.13-1.41)	<.001	1.23 (1.09-1.39)	<.001	1.23 (1.09-1.39)	.001
Cirrhosis
No	Female	197	1 [Reference]	.58	1 [Reference]	.61	1 [Reference]	.18
No	Male	622	1.05 (0.88-1.26)	.58	1.05 (0.87-1.26)		1.14 (0.94-1.37)	.18
Yes	Female	914	1 [Reference]	<.001	1 [Reference]	<.001	1 [Reference]	<.001
Yes	Male	2504	1.20 (1.10-1.30)	<.001	1.18 (1.08-1.29)	<.001	1.18 (1.08-1.29)	<.001
Viral etiology
Yes	Female	255	1 [Reference]	<.001	1 [Reference]	.004	1 [Reference]	.002
Yes	Male	635	1.17 (1.07-1.27)	<.001	1.15 (1.05-1.26)		1.16 (1.05-1.27)	.002
No	Female	856	1 [Reference]	.05	1 [Reference]	.12	1 [Reference]	.02
No	Male	2491	1.18 (1.00-1.39)	.05	1.14 (0.97-1.34)	.12	1.21 (1.03-1.42)	.02
Liver disease etiology
HBV	Female	182	1 [Reference]	.02	1 [Reference]	.04	1 [Reference]	.19
HBV	Male	839	1.24 (1.04-1.49)	.02	1.22 (1.01-1.47)		1.13 (0.94-1.36)	.19
HCV	Female	652	1 [Reference]	.01	1 [Reference]	.03	1 [Reference]	.003
HCV	Male	1579	1.13 (1.02-1.24)	.01	1.13 (1.02-1.26)	.03	1.18 (1.06-1.32)	.003
ALD	Female	12	1 [Reference]	.02	1 [Reference]	.04	1 [Reference]	.01
ALD	Male	140	2.02 (1.13-3.59)	.02	1.90 (1.02-3.54)	.04	2.35 (1.29-4.30)	.01
MASLD	Female	84	1 [Reference]	.86	1 [Reference]	.95	1 [Reference]	.57
MASLD	Male	105	0.97 (0.72-1.31)	.86	0.99 (0.72-1.36)	.95	1.10 (0.79-1.53)	.57
Other^c	Female	39	1 [Reference]	.96	1 [Reference]	.35	1 [Reference]	.67
Other^c	Male	39	1.01 (0.64-1.58)	.96	0.81 (0.52-1.26)	.35	0.90 (0.55-1.47)	.67
Cryptogenic	Female	120	1 [Reference]	.39	1 [Reference]		1 [Reference]	.49
Cryptogenic	Male	351	1.10 (0.88-1.37)	.39	1.05 (0.84-1.30)	.68	1.08 (0.87-1.34)	.49
BCLC stage
0/A	Female	617	1 [Reference]	.01	1 [Reference]	.01	1 [Reference]	.004
0/A	Male	1309	1.16 (1.05-1.29)	.01	1.12 (1.01-1.24)		1.16 (1.05-1.28)	.004
B	Female	203	1 [Reference]	.05	1 [Reference]	.03	1 [Reference]	.01
B	Male	709	1.20 (1.00-1.44)	.05	1.21 (1.02-1.44)	.03	1.25 (1.05-1.49)	.01
C	Female	184	1 [Reference]	.90	1 [Reference]	.68	1 [Reference]	.60
C	Male	730	0.99 (0.83-1.18)	.90	0.96 (0.80-1.15)	.68	0.95 (0.79-1.14)	.60
D	Female	107	1 [Reference]	.02	1 [Reference]	.02	1 [Reference]	.09
D	Male	378	1.34 (1.06-1.70)	.02	1.31 (1.04-1.66)	.02	1.23 (0.97-1.56)	.09

Open in a new tab

Abbreviations: aHR, adjusted hazard ratio; ALD, alcohol-associated liver disease; BCLC, Barcelona Clinic Liver Cancer; HBV, hepatitis B virus; HCV, hepatitis C virus; MASLD, metabolic dysfunction–associated steatotic liver disease.

^{^a}

Adjusted for age, region, diabetes, etiology, and BCLC stage.

^{^b}

Adjusted for age, region, diabetes, etiology, BCLC stage, and hepatocellular carcinoma therapy.

^{^c}

Includes autoimmune hepatitis, primary biliary cirrhosis, primary sclerosing cholangitis, hemochromatosis, Wilson disease, and others.

Sensitivity analyses for association between sex and mortality were conducted in the PSM-matched cohort. Largely consistent findings were observed (eResults, eTable 4, and eTable 5 in Supplement 1).

Discussion

This large, multinational, clinical practice longitudinal cohort study examined more than 8500 patients with HCC in a population with diverse ethnic backgrounds and liver disease etiologies from both the East and West. We observed increased overall and liver-related mortality in males and no significant difference in non–liver-related morality between sexes, a finding that was consistent in matched analyses via IPTW and PSM methods. We also found heterogeneous patterns among various subgroups, with associations between male sex and increased overall and liver-related mortality among Asian, Asia region, cirrhosis, and viral etiology subgroups but not among non-Asian, non–Asia region, noncirrhosis, or nonviral etiology subgroups. The heterogeneity was also pronounced among the specific liver disease etiology subgroup, with males having an aHR of 1.90 for increased overall mortality risk among patients with ALD but only 1.22 and 1.13 among patients with HBV and HCV, respectively, and no significant difference among those with MASLD. Together, these findings highlight the importance of incorporating sex-specific considerations into individualized HCC management in addition to other factors, such as age, race and ethnicity, cirrhosis, and liver disease etiology.

Aligning with prior studies,^8,11,13,14 we also found better overall survival outcomes in female patients with HCC in unmatched and IPTW-matched cohorts. Some prior studies reported that the better survival in females was likely due to their higher adherence to HCC surveillance programs, earlier-stage diagnosis, or more favorable treatment responses.^8,9,14,16,17 In addition to evidence for behavioral and clinical factors, accumulating evidence suggests that intrinsic sex-specific mechanisms (eg, the protective effects of estrogen, function of the sex-determining region Y gene promoting an immunosuppressive microenvironment and facilitating metastasis, and sex-distinct cancer immune responses) also contribute to sex disparities in HCC prognosis.^18,19,20 In our study, associations between male sex and an increased risk of overall mortality in HCC were consistently observed in IPTW- and PS-matched cohorts after multivariable adjustment for age, region, diabetes, etiology, and BCLC stage. Compared with prior cohort studies characterized by smaller sample sizes, single-center or single-region designs, or insufficient covariate adjustment,^8,11,13 our findings based on a larger, multinational HCC cohort with rigorous covariate balancing and comprehensive multivariable adjustment expand on prior knowledge by providing robust evidence of association between sex and overall and liver-specific mortality but not non–liver-related mortality.

We also provide detailed data for several relevant subgroups, which highlight substantial variations that may be critical in informing prognosis and management of HCC. In stratified analyses of patients with and without cirrhosis, males compared with females had an aHR of 1.18 for increased risk of overall mortality among patients with cirrhotic HCC and 1.15 among those with viral-related HCC. On the other hand, approximately 25% of our cohort developed HCC on noncirrhotic or nonviral liver disease background (2267 patients without cirrhosis [26.4%] and 1977 patients with nonviral etiology [23.0%]), among whom no sex association with overall mortality was observed. Our findings of sex associations in cirrhotic HCC are partly supported by a prior study of 20 045 patients with cirrhosis without HCC, which reported a 16% higher risk of overall mortality in males.²¹ The lack of sex differences in noncirrhotic HCC may be due to preserved liver function and a relatively favorable prognosis, potentially reducing the association of sex with mortality. The higher proportions of nonviral and noncirrhotic HCC in our cohort of 23.0% and 26.4% as compared with approximately 20% by some prior reports are in line with a 2021 global study, in which nonviral HCC accounted for 32% of all HCC cases (19% ALD, 8% MASLD, and 4% other causes).^22,23 These higher proportions are likely associated with the increasing prevalence of nonviral liver diseases (ALD and MASLD) in recent years, high prevalence of noncirrhotic HCC with MASLD, and effective antiviral therapies reducing HCV-related cirrhotic HCC incidence.^24,25

By specific etiology, the greatest sex differences in overall mortality and the most pronounced sex association were observed in alcohol-associated HCC, followed by HBV- and HCV-related HCC, while no significant sex difference was seen in MASLD-related HCC. The differential magnitude of sex associations across HCC etiologies may be attributed to multiple factors. Clinically, the lack of effective interventions for ALD and the higher alcohol exposure but lower abstinence rates in males^26,27,28 may partly explain the higher aHR in the sex association observed in alcohol-associated HCC. For viral-related HCC, antiviral therapies can substantially improve survival outcomes^29,30 which may attenuate sex-related contributions, while MASLD is associated with noncirrhotic HCC,^31,32 in which associations of sex with mortality appear minimal, as reported previoulsy.³³ Biologically, sex hormones or sex-linked genes differentially influence mechanisms of tumorigenesis, tumor behavior, and immune responses in alcohol-, virus-, and metabolism-related liver diseases,^{34,35,36,37,38} which may also contribute to etiology-dependent sex disparities.

In addition to liver disease–dependent sex associations with HCC mortality, we observed distinct patterns of sex disparities across ethnic and geographical backgrounds (Asian vs non-Asian and Asia vs non–Asia region subgroups). Male vs female patients of Asian ethnicity or from the Asia region consistently had aHRs of 1.22 to 1.23 for increased risk of overall mortality, whereas no significant sex differences were observed among non-Asian patients or those from non-Asia regions. In our study, viral-related HCC accounted for most cases among Asian (4568 of 5644 cases [80.9%]) or Asia region (3678 of 4567 cases [80.5%]) patients, while nonviral etiologies accounted for a substantial proportion of HCC cases among non-Asian or non–Asia region patients, particularly among White (591 patients [33.1%]) and Hispanic (264 patients [43.0%]) patients. These differential distributions of etiologies, reflecting regional epidemiological characteristics of HCC,³⁹ may partly explain observed ethnic or geographical sex disparities given that the association of sex with mortality differs by HCC etiology. For Asian patients or those from the Asia region, certain factors may additionally contribute to the disparities in HCC mortality. These factors may include males having higher alcohol or tobacco consumption^40,41 and females having greater health awareness and knowledge on liver diseases (particularly viral hepatitis).⁴² Moreover, the development, progression, and treatment of HCC are inherently complex and are further influenced by multiple factors, including genetic differences among racial groups,^43,44,45 making it challenging to disentangle the association between sex and mortality in HCC. Nevertheless, our findings provide valuable insights on ethnic and geographical-specific sex disparities in HCC mortality, enhancing our understanding of the association of sex with HCC, particularly in clinical settings, supported by rigorous matching and comprehensive multivariable adjustments to minimize confounding.

Strengths and Limitations

Our study has several strengths. First, we included more than 8500 patients with HCC in clinical settings with multiethnic backgrounds and diverse etiologies from Eastern and Western regions. These patients had detailed clinical data and long-term follow-up (2000-2023), enhancing the generalizability of our findings. Second, we applied 2 approaches, IPTW and PSM, to balance baseline characteristics between study groups and minimize the influence of potential confounders, with further multivariable adjustment to strengthen the validity of observed associations. Third, we expanded the sex differences in HCC mortality, previously reported mainly for overall or viral-related HCC, across all specific major etiologies (HBV, HCV, ALD, and MASLD). Given that the underlying etiology of HCC is rapidly changing from viral to nonviral liver diseases with the successful use of curative antiviral treatment for HCV, our findings of increased mortality risk in males with ALD-related HCC and no significant sex difference in patients with MASLD-related HCC are particularly valuable. With the worse prognosis for male patients with HCC, especially those with ALD, additional clinical interventions should be considered. Male patients with viral-related HCC and ALD-associated HCC should be counseled regarding their increased mortality risk to encourage better adherence to HCC surveillance, clinical monitoring, and treatment of HCC and the underlying liver disease. Care professionals should also be well informed of the poorer prognosis in male patients with HCC to counsel and monitor their male patients accordingly. Because HCC prognosis is highly dependent on HCC treatment types, which are in turn dependent on liver function and tumor staging, early HCC diagnosis is paramount. However, only 52.3% of males in this study had early HCC (BCLC 0/A) at diagnosis compared with 64.5% of females. This findings suggest that more efforts are needed by clinicians and patients to adhere to HCC surveillance to detect new HCC cases, as well as HCC recurrences in established HCC cases in all patients, especially male patients. Additionally, clinicians and female patients with MASLD-related HCC should be aware of the lack of significant sex differences in the mortality of patients with MASLD-related to avoid potential bias against female patients with MASLD-related HCC due to perceived higher risk in their male counterparts.

We also acknowledge several limitations. First, despite rigorous matching and multivariable adjustments, residual confounding from certain unmeasured factors (eg, adherence of cancer recurrence surveillance after HCC therapies, background differences between patients included in study analyses and those excluded due to missing data, type of health care delivery system, management practices, socioeconomic status, patient preference, and availability of therapies at different centers and their changes over time) still exist. Second, given the relatively smaller sample size of patients who were not Asian, were not cirrhotic, or had nonviral HCC compared with Asian patients, those who were cirrhotic, or those with viral-related HCC, further studies are needed to validate our findings for these smaller subgroups. Nevertheless, we had more than 2000 patients with noncirrhotic or nonviral HCC in our cohort, and their proportions are consistent with previous reports,^23,31,32 reflecting typical clinical characteristics of HCC, in which cirrhotic or viral-related HCC remains the predominant form. The predominance of Asian patients in our study was likely due to the large number of participating centers from Asian regions, as well as the location of many US centers in the Western US, where there are large Asian populations, an ethnicity disproportionately affected by HCC. Third, due to the relatively small sample size of patients receiving systemic therapy, we did not perform further stratified analyses for this subgroup or different systemic regimens (eg, sorafenib and immune checkpoint inhibitors). Additional studies are therefore needed to fill this data gap. Fourth, potential heterogeneity and selection bias are inherent in the retrospective multicenter design. However, we attempted to mitigate for these by using a standardized study variable definition and data-collection form via individual patient record review at each participating center.

Conclusions

In this cohort study of multinational HCC with longitudinal data, our findings provided robust clinical evidence for notable sex disparities in mortality risk among patients with HCC, driven mainly by differences in liver-related mortality. Male sex was independently associated with an increased risk of overall mortality (aHR, 1.14); however, this sex association remained only in certain subgroups, including Asian, Asia region, cirrhotic, and viral–related HCC subgroups. Additionally, male patients within the nonviral subgroup with ALD had an increased mortality risk compared with females (aHR, 1.90), while no significant difference was observed for patients with MASLD. These findings of sex disparities in HCC mortality support the incorporation of sex-based considerations in HCC management while also accounting for differences in patient ethnicity and disease etiology.

Supplement 1.

eMethods.

eResults.

eReferences.

eTable 1. Survival rates for (A) overall, (B) liver-related, and (C) non–liver-related mortality in patients with HCC in the unmatched cohort

eTable 2. Survival rates for (A) overall, (B) liver-related, and (C) non–liver-related mortality in patients with HCC in the IPTW-matched cohort

eTable 3. Associations between sex and (A) liver-related and (B) non–liver-related mortality of patients with HCC in the IPTW-matched cohort

eTable 4. Baseline characteristics of patients with HCC in the PSM cohort by sex

eTable 5. Associations between sex and (A) overall, (B) liver-related, and (C) non–liver-related mortality in patients with HCC in the PSM cohort

eFigure 1. Cumulative overall, liver-related, and non–liver-related mortality rates in patients with HCC, subgroup analysis by age, in the unmatched cohort

eFigure 2. Cumulative overall, liver-related, and non–liver-related mortality rates in patients with HCC, subgroup analysis by race and ethnicity, in the unmatched cohort

eFigure 3. Cumulative overall, liver-related, and non–liver-related mortality rates in patients with HCC, subgroup analysis by region, in the unmatched cohort

eFigure 4. Cumulative overall, liver-related, and non–liver-related mortality rates in patients with HCC, subgroup analysis by cirrhosis status, in the unmatched cohort

eFigure 5. Cumulative overall, liver-related, and non–liver-related mortality rates in patients with HCC, subgroup analysis by viral status, in the unmatched cohort

eFigure 6. Associations between sex and (A) liver-related and (B) non–liver-related mortality in patients with HCC in the IPTW-matched cohort

jamanetwopen-e2637313-s001.pdf^{(2.3MB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2637313-s002.pdf^{(24.8KB, pdf)}

References

1.Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-263. doi: 10.3322/caac.21834 [DOI] [PubMed] [Google Scholar]
2.Chan SL, Sun HC, Xu Y, et al. The Lancet Commission on addressing the global hepatocellular carcinoma burden: comprehensive strategies from prevention to treatment. Lancet. 2025;406(10504):731-778. doi: 10.1016/S0140-6736(25)01042-6 [DOI] [PubMed] [Google Scholar]
3.Vogel A, Meyer T, Sapisochin G, Salem R, Saborowski A. Hepatocellular carcinoma. Lancet. 2022;400(10360):1345-1362. doi: 10.1016/S0140-6736(22)01200-4 [DOI] [PubMed] [Google Scholar]
4.Collaborators GBDC; GBD 2023 Cancer Collaborators . The global, regional, and national burden of cancer, 1990-2023, with forecasts to 2050: a systematic analysis for the Global Burden of Disease Study 2023. Lancet. 2025;406(10512):1565-1586. doi: 10.1016/S0140-6736(25)01635-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Abboud Y, Ismail M, Khan H, et al. Hepatocellular carcinoma incidence and mortality in the USA by sex, age, and race: a nationwide analysis of two decades. J Clin Transl Hepatol. 2024;12(2):172-181. doi: 10.14218/JCTH.2023.00356 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Tran TT, Be HT, Hoang KD, Dong HD. Sex and gender disparities in hepatocellular carcinoma: insights into risk, diagnosis, and therapeutic outcomes. Clin Transl Oncol. 2025;27(12):4301-4315. doi: 10.1007/s12094-025-03965-3 [DOI] [PubMed] [Google Scholar]
7.Phipps M, Livanos A, Guo A, et al. Gender matters: characteristics of hepatocellular carcinoma in women from a large, multicenter study in the United States. Am J Gastroenterol. 2020;115(9):1486-1495. doi: 10.14309/ajg.0000000000000643 [DOI] [PubMed] [Google Scholar]
8.Rich NE, Murphy CC, Yopp AC, Tiro J, Marrero JA, Singal AG. Sex disparities in presentation and prognosis of 1110 patients with hepatocellular carcinoma. Aliment Pharmacol Ther. 2020;52(4):701-709. doi: 10.1111/apt.15917 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Cullaro G, Rubin J, Mehta N, Yao F, Verna EC, Lai JC. Sex-based disparities in hepatocellular carcinoma recurrence after liver transplantation. Transplantation. 2021;105(11):2420-2426. doi: 10.1097/TP.0000000000003575 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Cron DC, Mazur RD, Bhan I, Adler JT, Yeh H. Sex and size disparities in access to liver transplant for patients with hepatocellular carcinoma. JAMA Surg. 2024;159(11):1291-1298. doi: 10.1001/jamasurg.2024.3498 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Pang C, Li JM, Wang Z, et al. Age-dependent female survival advantage in hepatocellular carcinoma: a multicenter cohort study. Clin Gastroenterol Hepatol. 2024;22(2):305-314. doi: 10.1016/j.cgh.2023.07.029 [DOI] [PubMed] [Google Scholar]
12.Ladenheim MR, Kim NG, Nguyen P, et al. Sex differences in disease presentation, treatment and clinical outcomes of patients with hepatocellular carcinoma: a single-centre cohort study. BMJ Open Gastroenterol. 2016;3(1):e000107. doi: 10.1136/bmjgast-2016-000107 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Busso C, Nault JC, Layese R, et al. Prolonged survival in women with hepatocellular carcinoma: a French observational study. Clin Res Hepatol Gastroenterol. 2024;48(10):102498. doi: 10.1016/j.clinre.2024.102498 [DOI] [PubMed] [Google Scholar]
14.Liou WL, Tan TJ, Chen K, Goh GB, Chang JP, Tan CK. Gender survival differences in hepatocellular carcinoma: is it all due to adherence to surveillance: a study of 1716 patients over three decades. JGH Open. 2023;7(5):377-386. doi: 10.1002/jgh3.12910 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Singal AG, Llovet JM, Yarchoan M, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78(6):1922-1965. doi: 10.1097/HEP.0000000000000466 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Liang T, He Y, Mo S, et al. Gender disparity in hepatocellular carcinoma recurrence after curative hepatectomy. Ann Hepatol. 2022;27(3):100695. doi: 10.1016/j.aohep.2022.100695 [DOI] [PubMed] [Google Scholar]
17.Fa A, Danos DM, Maniscalco L, et al. Is there really a difference in outcomes between men and women with hepatocellular cancer? Cancers (Basel). 2023;15(11):2892. doi: 10.3390/cancers15112892 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Hassan MM, Botrus G, Abdel-Wahab R, et al. Estrogen replacement reduces risk and increases survival times of women with hepatocellular carcinoma. Clin Gastroenterol Hepatol. 2017;15(11):1791-1799. doi: 10.1016/j.cgh.2017.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Song L, Sun L, Ren Y, Wang X, Xian L. Sex disparities in hepatocellular carcinoma immunotherapy: hormonal and genetic influences on treatment efficacy. Front Immunol. 2025;16:1607374. doi: 10.3389/fimmu.2025.1607374 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Zhang W, Chen LF, Lu Q, et al. Sex disparity of liver cancer recurrence following surgery: sex-determining region Y gene promotes immunosuppressive microenvironment and favors metastasis. Int J Surg. 2025;111(11):7666-7676. doi: 10.1097/JS9.0000000000002900 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Mazumder NR, Celaj S, Atiemo K, et al. Liver-related mortality is similar among men and women with cirrhosis. J Hepatol. 2020;73(5):1072-1081. doi: 10.1016/j.jhep.2020.04.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Sato-Espinoza K, Valdivia-Herrera M, Chotiprasidhi P, Diaz-Ferrer J. Hepatocellular carcinoma in patients without cirrhosis. World J Gastroenterol. 2025;31(23):107100. doi: 10.3748/wjg.v31.i23.107100 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Tan EY, Danpanichkul P, Yong JN, et al. Liver cancer in 2021: Global Burden of Disease Study. J Hepatol. 2025;82(5):851-860. doi: 10.1016/j.jhep.2024.10.031 [DOI] [PubMed] [Google Scholar]
24.Mi K, Ye T, Zhu L, Pan CQ. Risk-stratified hepatocellular carcinoma surveillance in non-cirrhotic patients with MASLD. Gastroenterol Rep (Oxf). 2025;13:goaf018. doi: 10.1093/gastro/goaf018 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Mathur K, Mazhar A, Patel M, et al. Changing trends of cirrhotic and noncirrhotic hepatocellular carcinoma in the era of directly-acting antiviral agents. Clin Transl Gastroenterol. 2021;12(11):e00420. doi: 10.14309/ctg.0000000000000420 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Jophlin LL, Singal AK, Bataller R, et al. ACG clinical guideline: alcohol-associated liver disease. Am J Gastroenterol. 2024;119(1):30-54. doi: 10.14309/ajg.0000000000002572 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.White AM. Gender differences in the epidemiology of alcohol use and related harms in the United States. Alcohol Res. 2020;40(2):01. doi: 10.35946/arcr.v40.2.01 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Zywiak WH, Stout RL, Trefry WB, et al. Alcohol relapse repetition, gender, and predictive validity. J Subst Abuse Treat. 2006;30(4):349-353. doi: 10.1016/j.jsat.2006.03.004 [DOI] [PubMed] [Google Scholar]
29.Kanneganti M, Al-Hasan M, Bourque S, et al. Older age but not comorbidity is associated with worse survival in patients with hepatocellular carcinoma. Clin Gastroenterol Hepatol. 2025;23(8):1377-1386.e2. doi: 10.1016/j.cgh.2024.10.015 [DOI] [PubMed] [Google Scholar]
30.Singal AG, Rich NE, Mehta N, et al. Direct-acting antiviral therapy for hepatitis C virus infection is associated with increased survival in patients with a history of hepatocellular carcinoma. Gastroenterology. 2019;157(5):1253-1263.e2. doi: 10.1053/j.gastro.2019.07.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.van Meer S, van Erpecum KJ, Sprengers D, et al. Hepatocellular carcinoma in cirrhotic versus noncirrhotic livers: results from a large cohort in the Netherlands. Eur J Gastroenterol Hepatol. 2016;28(3):352-359. doi: 10.1097/MEG.0000000000000527 [DOI] [PubMed] [Google Scholar]
32.Dong Y, Wang WP, Lee WJ, et al. Hepatocellular carcinoma in the non-cirrhotic liver. Clin Hemorheol Microcirc. 2022;80(4):423-436. doi: 10.3233/CH-211309 [DOI] [PubMed] [Google Scholar]
33.Bengtsson B, Stål P, Wahlin S, Björkström NK, Hagström H. Characteristics and outcome of hepatocellular carcinoma in patients with NAFLD without cirrhosis. Liver Int. 2019;39(6):1098-1108. doi: 10.1111/liv.14087 [DOI] [PubMed] [Google Scholar]
34.Natri HM, Wilson MA, Buetow KH. Distinct molecular etiologies of male and female hepatocellular carcinoma. BMC Cancer. 2019;19(1):951. doi: 10.1186/s12885-019-6167-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Ruggieri A, Gagliardi MC, Anticoli S. Sex-dependent outcome of hepatitis B and C viruses infections: synergy of sex hormones and immune responses? Front Immunol. 2018;9:2302. doi: 10.3389/fimmu.2018.02302 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Su L, Luo H, Yan Y, et al. Exploiting gender-based biomarkers and drug targets: advancing personalized therapeutic strategies in hepatocellular carcinoma. Front Pharmacol. 2024;15:1433540. doi: 10.3389/fphar.2024.1433540 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Taniai M. Alcohol and hepatocarcinogenesis. Clin Mol Hepatol. 2020;26(4):736-741. doi: 10.3350/cmh.2020.0203 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Toniutto P, Shalaby S, Mameli L, et al. ; Special Interest Group Gender in Hepatology of the Italian Association for the Study of the Liver (AISF) . Role of sex in liver tumor occurrence and clinical outcomes: a comprehensive review. Hepatology. 2024;79(5):1141-1157. doi: 10.1097/HEP.0000000000000277 [DOI] [PubMed] [Google Scholar]
39.Liu J, Xu T, Wang Y, Ji F, Zhang L. The spatio-temporal trends and determinants of liver cancer attributable to specific etiologies: a systematic analysis from the Global Burden of Disease Study 2021. Glob Health Res Policy. 2025;10(1):22. doi: 10.1186/s41256-025-00416-y [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Park E, Kim YS. Gender differences in harmful use of alcohol among Korean adults. Osong Public Health Res Perspect. 2019;10(4):205-214. doi: 10.24171/j.phrp.2019.10.4.02 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Im PK, Millwood IY, Guo Y, et al. ; China Kadoorie Biobank (CKB) collaborative group . Patterns and trends of alcohol consumption in rural and urban areas of China: findings from the China Kadoorie Biobank. BMC Public Health. 2019;19(1):217. doi: 10.1186/s12889-019-6502-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Lee MH, Ahn SH, Chan HLY, et al. Contextual and individual factors associated with knowledge, awareness and attitude on liver diseases: a large-scale Asian study. J Viral Hepat. 2022;29(2):156-170. doi: 10.1111/jvh.13636 [DOI] [PubMed] [Google Scholar]
43.Nathani P, Gopal P, Rich N, et al. Hepatocellular carcinoma tumour volume doubling time: a systematic review and meta-analysis. Gut. 2021;70(2):401-407. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Wagle NS, Park S, Washburn D, et al. Racial, ethnic, and socioeconomic disparities in curative treatment receipt and survival in hepatocellular carcinoma. Hepatol Commun. 2022;6(5):1186-1197. doi: 10.1002/hep4.1863 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Tsilimigras DI, Stecko H, Moris D, Pawlik TM. Next-generation sequencing demonstrates racial and sex differences in genomic profiling of hepatocellular carcinoma patients: an AACR GENIE project analysis. HPB (Oxford). 2025;27(3):371-376. doi: 10.1016/j.hpb.2024.12.008 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials