The Impact of Frailty in Hepatocellular Carcinoma

Abstract

Background

Hepatocellular carcinoma (HCC) represents an increasing global health challenge, associated with poor prognosis and an epidemiological shift towards an aging and comorbid population. Frailty is a multidimensional syndrome of reduced physiological reserve. It is recognized as a robust predictor of clinical outcomes in chronic liver disease and some cancers but remains underexplored in HCC.

Summary

This narrative review synthesizes current evidence on the clinical significance of frailty in HCC, focusing on its impact on outcomes, evidence for targeted interventions, and potential role in clinical decision-making. Frailty is associated with adverse outcomes in HCC, including increased healthcare utilization, reduced treatment tolerance, and decreased survival. However, heterogeneity in frailty assessment methods and limited prospective data preclude its clinical application. Early evidence suggests that structured exercise programs are safe, feasible, and may improve frailty measures, although larger prospective studies are required to determine whether these translate into meaningful clinical benefits.

Key Messages

Frailty is an emerging prognostic marker and may inform supportive care in HCC. Standardized assessment and interventional trials are needed to support routine integration of frailty for more personalized HCC management.

Introduction

Hepatocellular carcinoma (HCC) is the leading form of primary liver cancer and remains a global public health challenge, with projections estimating a greater than 50% rise in the number of new cases and deaths by 2040 [1, 2]. HCC primarily occurs in patients with cirrhosis, and almost all remaining cases arise in those with underlying chronic liver disease (CLD). The contemporary epidemiology of HCC is shifting with an increasing proportion of cases seen in older adults over the age of 60 years [3–5]. Potential reasons for this include an aging population, improvements in CLD and HCC management such that patients are surviving longer [6], and the rise of HCC secondary to metabolic dysfunction-associated fatty liver disease, where HCC typically presents at an older age [7, 8]. Older patients also accumulate multiple comorbidities, which impact HCC management decisions [9, 10].

Despite recent therapeutic advances, the prognosis of HCC remains poor due to factors including tumoral heterogeneity, advanced tumor stage at diagnosis, and degree of liver dysfunction [11, 12]. While staging systems, including the Barcelona Clinic Liver Cancer (BCLC) classification, assist in prognostication, they fail to capture broader aspects of a patient’s condition. Notably, the assessment of performance status within the BCLC system reflects tumor-related symptoms without incorporating comorbidities, functional or cognitive impairment [13]. These factors are particularly relevant in older adults as they can significantly impact treatment options and outcomes. As such, prognostic markers encompassing these elements are critical to improve risk stratification and inform individualized treatment strategies in HCC.

In light of this, an emerging area of interest is frailty, characterized by functional decline and decreased physiologic reserve. Frailty is common in CLD, where it is already well-recognized as an important predictor of adverse clinical outcomes [14]. Given the vast majority of HCC occurs in the setting of CLD, the assessment of frailty in this setting may offer valuable insights. This narrative review summarizes current evidence on the pathophysiology, epidemiology, clinical impact, and management of frailty in HCC, as well as identifying knowledge gaps and future research priorities for patient-centered treatment decision-making.

Methods

Design and Reporting Framework

A narrative review was conducted to examine frailty in HCC. This review comprised two components (1): background sections on frailty pathophysiology, assessment tools, and epidemiology informed by a targeted search of broader CLD, geriatric, and oncology literature; and (2) a structured search examining the clinical impact of frailty as well as frailty-directed interventions in HCC. A preliminary literature search was performed to assess the feasibility of conducting a systematic review. Given the substantial heterogeneity in study designs, frailty instruments, and outcome measures, a narrative synthesis rather than a systematic review was considered the most appropriate approach. This review was conducted and reported in accordance with the Scale for the Assessment of Narrative Review Articles (SANRA) [15].

Search Strategy

For the structured literature search, Medline and Embase were searched from inception to November 1, 2025, using combined Medical Subject Headings (MeSH) and free-text terms. Keywords included a combination of “hepatocellular carcinoma,” “hepatoma,” “cirrhosis,” or “chronic liver disease” with “frailty,” “frail,” “liver frailty index,” “exercise,” “physical activity,” and “rehabilitation.” The search was limited to studies published in English within peer-reviewed journals. Additional articles were manually identified through searches of reference lists and personal libraries. The search strategy is illustrated in Figure 1.

Fig. 1.

Search strategy.

Study Screening and Selection

Observational studies and clinical trials were included if they reported frailty assessments in adults with HCC and examined clinical outcomes such as mortality, morbidity, treatment complications, and quality of life or evaluated frailty-directed interventions such as exercise or nutritional support. Given the limited evidence base, conference abstracts were only included if they reported relevant clinical outcomes data not available in full-text publications. Studies not specifically assessing frailty or frailty-related outcomes in HCC were excluded. Two reviewers (R.J. and S.S.) independently assessed study eligibility for inclusion. Studies were screened by title and abstract, with full texts reviewed where relevant.

Data Extraction and Synthesis

For each included study, information was extracted on study design, population setting, frailty assessment tool, treatment type, and clinical outcomes. For interventional studies, characteristics were extracted using frequency, intensity, time, and type principles where available. Given the significant heterogeneity in frailty assessment, study populations, and treatment modalities, a meta-analysis was not performed. Findings were synthesized narratively and organized thematically.

Pathophysiology of Frailty in HCC

Frailty is a complex syndrome defined as a state of heightened vulnerability to stressors due to multisystem physiologic decline, resulting in an impaired ability to maintain homeostasis and return to baseline [16, 17]. The concept of frailty originates from geriatric literature, with two main models being recognized. The frailty phenotype model defines frailty as a clinical syndrome characterized by exhaustion, weakness, slowness, physical inactivity, and unintentional weight loss. This model reflects impaired physiologic reserve rather than comorbidity or disability. Conversely, the deficit accumulation model measures the cumulative burden of health deficits, including medical conditions, cognitive impairment, functional decline, malnutrition, and laboratory abnormalities [18]. Although its pathogenesis is incompletely understood, frailty involves dysregulation of neurological, musculoskeletal, immune, and endocrine systems, with manifestations such as neuronal loss, sarcopenia, inflammation, and hormonal changes [17, 19]. Collectively, this impairs the body’s ability to maintain homeostasis and respond to stressors, resulting in adverse health outcomes including falls, hospitalization, and mortality [20–22].

Though frailty is increasingly recognized as multidimensional, encompassing physical, cognitive, and psychosocial domains, its conceptualization differs between specialties. Within hepatology, frailty has largely been operationalized as physical frailty or impaired muscle function, which has been shown to be a critical determinant of adverse health outcomes. On the other hand, multidimensional frailty is used in oncology populations and mainly considered in older adults with cancer. It is, however, important to note that while frailty is more prevalent in older adults, it is not synonymous with aging and can also be seen in younger adults with significant comorbidities or chronic disease [17, 23]. This reflects a key limitation of the current HCC literature, which spans two related but distinct conceptual frameworks of frailty, highlighting the urgent need for a consensus framework for frailty in HCC.

HCC is a tumor that primarily arises in the setting of CLD, and thus, frailty in this population is driven by both liver disease-related and cancer-related mechanisms. The reported prevalence of frailty in CLD ranges from 17 to 43%, with variations attributable to differences in assessment method and patient populations [14, 24–28]. Despite its age-related increase, frailty is widespread across adults with CLD, with one study reporting a prevalence of 42% in those aged 50–59, rising to over 90% among those over 70 [29]. Common systemic drivers of frailty, such as malnutrition, sarcopenia, and physical inactivity, are compounded in CLD by hepatic-specific factors. These include impaired ammonia clearance and subsequent myotoxicity, altered protein metabolism, and sequelae of portal hypertension [30], shown in Figure 2.

Fig. 2.

Contributors of CLD and HCC to frailty.

In CLD, frailty does not occur in isolation and is often seen alongside sarcopenia and malnutrition. Notably, while frailty and sarcopenia are often used interchangeably, they are distinct entities, with sarcopenia being defined as a progressive and generalized depletion in skeletal muscle mass [31]. While sarcopenia is a major contributor to frailty, the two can exist independently, though they have additive adverse prognostic effects in patients with CLD [32]. Finally, sarcopenic obesity, or the loss of muscle mass with concurrent increased adiposity, represents a further distinct phenotype contributing to frailty [33]. Drivers of sarcopenic obesity include low-grade inflammation, hormonal changes, and insulin resistance [34]. This is pertinent in HCC given the rising incidence of metabolic dysfunction-associated fatty liver disease as an underlying etiology [35, 36]; however, data examining its contribution to frailty in this population is lacking.

Frailty also plays a critical role in oncology and is prevalent across multiple tumor types. In common cancers such as lung, breast, and colorectal cancer, the pooled prevalence of frailty is reported at 45%, 43%, and 31%, respectively [37–39]. Notably, frailty affects half of older adults as well as some younger adults with cancer [40, 41]. The etiology of frailty in cancer patients is multifactorial, comprising comorbidities; treatment-induced cellular damage and senescence that accelerate aging; treatment toxicity; and cancer-related inflammation and associated cachexia. In HCC, these cancer-related mechanisms coexist with underlying CLD, amplifying the risk of frailty in this population, as shown in Figure 2.

Beyond being highly prevalent, frailty and its biological alterations may also influence treatment response, particularly to immune checkpoint inhibitors used in advanced HCC. Frailty has been associated with chronic low-grade inflammation and immune dysregulation, including elevated cytokines such as IL-6 and CRP, impaired innate and adaptive immunity, and increased inhibitory immune checkpoint receptors such as PD-1 [42, 43]. These changes are particularly relevant in HCC, where immune checkpoint inhibitors act within the liver’s inherently immune-tolerant microenvironment [44]. Cancer cachexia may further exacerbate systemic inflammation and suppress antitumor immunity. This can potentially diminish the efficacy of immune checkpoint inhibitors, a relationship that has been demonstrated across multiple solid tumors but remains poorly characterized in HCC [45]. Together, these mechanisms provide the biological rationale for frailty assessment in HCC.

Instruments Used to Measure Frailty

Several validated tools have been developed to measure frailty that differ in the domains assessed and target populations. The Fried Frailty Index (FFI) is the most widely used and is a measure of physical frailty, defined by the presence of three or more out of five components (slowness, exhaustion, low physical activity, unintentional weight loss, and reduced muscle strength) (Table 1) [46]. Limitations of the FFI include subjective responses regarding exhaustion and physical activity levels. Additionally, the assessment of weight loss may be difficult to interpret in CLD patients, as those decompensated with large-volume ascites may have significant sarcopenia despite fluid-related weight gain. In contrast, rapid weight gain due to sarcopenic obesity is also associated with frailty in the general population, suggesting that assessment of weight in isolation may be an unreliable marker of frailty [47].

Table 1.

Summary of select frailty screening tools

Frailty score	Type	Domains assessed	Assessment method	Frailty score classification	Strengths	Limitations
Fried Frailty Index (FFI) [46]	Physical	Weight loss, exhaustion, grip strength, walking speed, physical activity	5-item physical assessment and questionnaire	Frail: ≥3	Objective, validated in large studies	Time-consuming, requires equipment, excludes cognitive and social domains
				Pre-frail: 1–2
				Robust: 0
Rockwood Clinical Frailty Scale (CFS) [48]	Clinical judgment	Function, comorbidity, cognition	9-item pictorial scale based on clinician impression	Frail: ≥5	Quick, widely used in hospitals, prognostic in many settings	Subjective, limited in non-verbal or cognitively impaired patients
Rockwood Mitnitski Frailty Index [49]	Deficit accumulation	Physical psychological, social	92-item test calculated as a ratio of deficits to total items collected	Frail: ≥0.25	Granular, continuous score, predictive of outcomes	Time-intensive
				Pre-frail: 0.12–0.25
				Robust: <0.12
Edmonton Frail Scale (EFS) [50]	Multidomain	Physical, cognitive, psychological, social, nutrition	11-item physical assessment and questionnaire	Severely frail: 12–17	Multidimensional, feasible in clinics	Requires training, limited validation in younger populations
				Moderately frail: 10–11
				Mildly frail: 8–9
				Vulnerable: 6–7
				Not frail: 0–5
Hospital Frailty Risk Score (HFRS) [51]	Comorbidity based	Comorbidity	Weighted assessment of International Classification of Diseases-10 coded comorbidities	High risk: >15	Uses existing data, reduced clinician burden, automated	Limited to hospitalized patients, relies on accurate coding, excludes cognitive and psychosocial domains
				Intermediate risk: 5–15
				Low risk: <5
Groningen Frailty Indicator (GFI) [52]	Multidomain	Physical, cognitive, social, psychological	15-item self-report questionnaire	Frail: ≥4	Easy to administer, patient-reported	Less specific, may overestimate frailty
Modified Frailty Index – 5 Item (MFI-5) [53]	Comorbidity based	Function and comorbidities	5-item binary scoring system	No unanimous classification	Quick, uses existing data	Limited, excludes cognitive and psychosocial domains
Liver Frailty Index (LFI) [54]	Performance-based (liver-specific)	Grip strength, balance, chair stands	3-item physical assessment	Frail: ≥4.5	Quick, validated in cirrhosis, predictive of decompensation and mortality	Not validated in the general population
				Pre-frail: 3.2 to <4.5
				Robust: <3.2

Thus, in CLD, frailty assessment tools focusing on physical performance are preferred. These include the Karnofsky performance status (KPS), Rockwood Mitnitski Frailty Index (FI), and Rockwood Clinical Frailty Scale (CFS) [31]. Simple bedside measures of physical function, such as the short physical performance battery, 6-min walk test (6MWT), short gait speed, and hand-grip strength (HGS), are commonly used as components of a frailty assessment. In 2018, Lai et al. [54] developed the Liver Frailty Index (LFI), a simple and efficient bedside tool based on objective performance-based metrics with validated cut-offs for frailty classification (LFI ≥4.5: frail, LFI 3.2 to <4.5: pre-frail, and LFI <3.2: robust). While major guidelines do not endorse the LFI over other frailty tools, it has been widely adopted given its feasibility across clinical settings and minimal equipment requirements [31]. Comparative studies, though limited, suggest the LFI performs similarly to other frailty assessment tools in terms of prevalence and prognostic utility [55, 56].

Conversely, in oncology, frailty is conceptualized more broadly, and routine multi-domain assessment is advocated for older adults before treatment. Validated assessment tools include the CFS, Geriatric-8 (G8) questionnaire, Senior Adult Oncology Program 3 tool, Vulnerable Elders Survey-13, and Practical Geriatric Assessment [57–59], which collectively evaluate domains such as comorbidity, cognition, physical function, nutrition, and psychosocial wellbeing.

The optimal approach to frailty assessment in HCC remains uncertain with no instrument being specifically validated in this population. Given that most patients have underlying CLD and may be younger, broader oncology metrics developed in a geriatric population may be less applicable. Preliminary data in CLD also suggest serum biomarkers including testosterone, myostatin, inflammatory markers, and growth differentiation factor 15 are associated with frailty status. While these biomarkers are not routinely used in clinical assessment, they represent potential adjuncts to frailty assessments and potential therapeutic targets [60].

Social and Psychological Frailty

To date, frailty research in HCC has focused on physical domains, although frailty also encompasses psychological and social vulnerability. Social frailty has been described as a state of increased vulnerability arising from diminished social resources, social participation, or the ability to fulfil social needs and may precede the development of physical frailty [61–63]. Evidence from broader oncology populations suggests that loneliness and social isolation are associated with increased all-cause and cancer-specific mortality [64]. In a prospective study of 181 older adults undergoing curative surgery for gastrointestinal malignancies, a high prevalence of preoperative social frailty (46.8%) was reported, which independently predicted poorer overall and cancer-specific survival. Furthermore, the coexistence of physical and social frailty was associated with substantially worse outcomes than either domain alone, with a 6.35-fold higher risk of all-cause mortality and a 7.57-fold higher risk of cancer-specific mortality [65].

Direct studies evaluating social frailty in HCC are lacking; however, population-based studies suggest that social and economic determinants may influence prognosis. In a Korean nationwide cohort, lower socioeconomic status was independently associated with increased mortality, particularly among HCC patients receiving non-curative therapies [66]. Similarly, a large USA cohort study of patients with HCC found that unmarried status was associated with poorer survival [67]. Collectively, these data suggest that current HCC frailty frameworks, which focus primarily on physical performance, may capture only part of the vulnerability experienced by patients.

Frailty in Patients with HCC

Epidemiology of Frailty in HCC

Frailty is a key determinant of adverse outcomes in both CLD and oncology, and hence its impact in patients with HCC is being increasingly recognized. The prevalence of frailty in HCC ranges from 22% to 58% across cohorts, reflecting differences in study populations, disease stage, and heterogeneity in frailty assessment tools [68–71]. In a recent Australian multicenter prospective study, 26% of patients with HCC were classified as frail (LFI >4.5) and a further 69% as pre-frail [72].

Compared to non-frail patients, frailty in HCC is more common in older patients and is associated with poorer performance status, higher Child-Pugh score, greater comorbidity burden, and rates of hepatic encephalopathy [70, 71, 73, 74]. Frail patients also demonstrate lower muscle mass, reduced physical activity, and higher rates of malnutrition [75, 76]. Additionally, they are more likely to experience non-home discharge, higher rates of health-care utilization, morbidity, and mortality (Table 2). Taken together, these findings illustrate the importance of frailty across the spectrum of HCC, with significant implications for treatment decisions and prognosis.

Table 2.

Overview of published studies of frailty in HCC

Author	Design	Sample size	Population	HCC treatment	Frailty assessment tool	Frailty (%)	Summary of main findings in frail patients
Ingawale et al. [77] (2025)a	Retrospective national	1,921	HCC	TACE, RFA, Y-90 radioablation, thermoablation	mFI-5	483 (25.1)	More likely to be discharged to care facilities rather than home
Yamada et al. [68] (2021)	Retrospective single-center	92	Age ≥75	Resection	CFS	21 (22.8)	Larger tumour diameter
							Higher rates of extrahepatic recurrence
							Reduced rates of treatment for extrahepatic recurrence
							Reduced OS, CSS, DFS
							Frailty predictor of CSS, DFS
Yang et al. [73] (2024)	Retrospective multicenter	240	Age ≥80	Resection	CFS	105 (43.8)	Higher post-operative 30-day morbidity
							Higher post-operative 30- and 90-day mortality
							Longer median post-operative LOS
							Higher 30-day readmission rate
							Reduced OS, RFS, and CSS
							Pre-operative frailty predictor of 30-day morbidity, OS, RFS, CSS
Yang et al. [74] 2024	Retrospective multicenter	497	Age ≥70	Resection	Barthel Index	157 (31.6)	Higher rates AFP >400 μg/L
							Larger tumour
							Higher macrovascular invasion
							Higher post-operative major morbidity
							Longer post-operative LOS
							Higher 30-day readmission rate
							Higher 90-day mortality
Diao et al. [78] (2024)	Retrospective multicenter	488	Age ≥70	Resection	CFS	148 (30.3)	Higher macrovascular invasion and satellite nodules
							More poor tumour differentiation
							Larger tumour size
							Longer post-operative LOS
							Higher post-operative 30-day morbidity
							Higher post-operative 30- and 90-day mortality
							Higher 30-day readmission rate
							Reduced OS, RFS
							Frailty independent predictor of OS, RFS
Okada et al. [79] (2024)	Prospective single-center	80	Age ≥65	Resection	Kihon checklist	25 (30.9)	Higher rates AFP >200 ng/mL
							More poor tumour differentiation
							Required more health care support
							Higher incidence extrahepatic recurrence
							Reduced OS
							Frailty predictor of OS
DeMaria et al. [80] (2019)	Prospective single-center	50	HCC on LT waitlist	LT	FFI	15 (30)	Longer total hospital LOS
							Longer time to first physiotherapy session
Rabei et al. [81] (2022)	Retrospective single-center	125	HCC	TACE	mFI-5	Not specified	Reduced TFS
							Frailty predictor of 30-day readmission, liver toxicity, lower TFS
Shao et al. [82] (2025)	Retrospective single-center	143	Age ≥65	TACE	mFI-5	46 (32.2)	Reduced OS
							Frailty independent predictor of OS
Loi et al. [83] (2021)	Retrospective single-center	42	Age ≥80	SBRT	G8	26 (62)	Reduced OS
van Dijk et al. [84] (2022)	Prospective single-center	56	HCC	Resection, RFA/MWA, TACE, TARE, Systemic therapy, BSC	LFI	8 (14)	Worse quality of life
							HGS but not LFI associated with improved complication-free survival and trend towards better OS
Ye et al. [72] (2025)a	Prospective multicenter	93	HCC	Not specified	LFI	24 (26)	Frailty independently associated with death
Sekiguchi et al. [85] (2022)	Retrospective single-center	101	Age ≥75	Systemic therapy	mG8	32 (31.7)	Lower OS in those taking lenvatinib but not sorafenib
							No difference in PFS
Waller et al. [71] (2024)	Prospective international multicenter	102	HCC	Systemic therapy	LFI	22 (21.6)	Higher PS
							Earlier systemic therapy discontinuation
							LFI independent predictor of OS
							No difference in adverse events, disease progression
Ramai et al. [70] (2021)	Retrospective national	10,983	Hospitalized patients with HCC	Not specified	HFRS	6,335 (57.7%)	4.5-fold increased risk of mortality
							2.3-fold increased risk of HE
							Longer LOS
							Higher total hospitalization cost

BSC, best supportive care; CFS, Clinical Frailty Scale; CSS, cancer-specific survival; DFS, disease-free survival; FFI, Fried frailty index; FHRS, G8, geriatric 8 screening tool; Hospital Frailty Risk Score, HCC, hepatocellular carcinoma; HE, hepatic encephalopathy; LFI, Liver Frailty Index; LOS, length of stay; LT, liver transplant; mFI-5, modified frailty index; mG8; modified geriatric 8 screening tool; MWA, microwave ablation; OS, overall survival; PFS, progression-free survival; PS, performance status; RFA, radiofrequency ablation, RFS, recurrence-free survival; SBRT, stereotactic body radiotherapy; TACE, transarterial chemoembolization; TARE, transarterial radioembolization; TFS, transplant-free survival; Y-90, yttrium-90.

^a Abstract form only.

Impact of Frailty on HCC Outcomes

Surgical Resection

Surgical resection is the preferred treatment for patients with resectable tumors and preserved liver function [86]. Despite the associated risks, advanced age is not an absolute contraindication to surgery as it is not a reliable predictor of post-operative outcomes [87]. In this setting, frailty assessment may be more beneficial as several studies have demonstrated frail patients have worse short- and long-term post-operative outcomes [73, 74, 78, 79, 88]. In the largest retrospective multicenter study of 497 patients undergoing hepatectomy for HCC, frail patients (as determined by the Barthel Index) experienced significantly higher rates of post-operative major morbidity (14% vs. 6.2%), prolonged length of stay (LOS; 40.8% vs. 14.1%), 30-day readmission (10.8% vs. 1.5%), and 90-day mortality (5.7% vs. 0.9%) [74]. Similarly, in another study of 488 patients, frail patients (as determined by the Clinical Frailty Scale) experienced a significantly higher 30-day morbidity (68.9% vs. 43.2%), 30-day mortality (6% vs. 2%), and 90-day mortality (9% vs. 3%) compared to non-frail patients [78]. Frail patients also had significantly worse median overall survival (41.6 vs. 69.6 months) and recurrence-free survival (27.6 vs. 42.7 months) compared to non-frail patients [78]. Similar findings were reported from other studies [68, 73], as well as worse cancer-specific survival (72% vs. 94.3%) [68], higher incidence of extrahepatic recurrence (30.8% vs. 3.6%) [79], higher alpha-fetoprotein (AFP) levels [68, 79], and larger tumor size [68, 74].

Liver Transplantation

Liver transplantation (LT) for HCC is advantageous as it provides definitive management of both HCC and CLD, with favorable long-term survival [89]. As LT is both a limited resource and an invasive treatment with associated morbidity, careful patient selection is critical. Frailty has been shown to be predictive of higher waitlist mortality and is associated with higher post-LT healthcare utilization and mortality [26, 90]. In a prospective study of 50 patients undergoing LT for HCC, frail patients had almost double the LOS compared to non-frail patients (14.5 days vs. 8 days), as well as increased time to first physiotherapy session (6 days vs. 4 days) [80]. There was no difference in 30-day mortality; however, the study was likely underpowered to detect this outcome [80]. Finally, while not specifically assessing LT for HCC, a large multicenter study showed that frailty remained significantly associated with post-LT mortality (HR 2.15; 95% CI: 1.40–3.30; p < 0.001) in a sensitivity analysis stratified by HCC [90]. This suggests frailty’s prognostic relevance, though evidence remains limited by small sample sizes and the absence of dedicated prospective studies. Further HCC-specific studies are needed to define its role in candidate selection and post-LT risk stratification.

Locoregional Therapy

Ablative Therapy. Percutaneous ablation for HCC is primarily performed for curative intent [87]; however, the only available data on ablative therapies and frailty comes from Ingawale et al. (published in abstract form), who studied 1921 patients with unresectable HCC. Frailty was measured by the Modified Frailty Index (mFI-5). “Ablative” treatments in this study population were heterogenous and included transarterial chemoembolization (TACE), yttrium-90 (Y-90) radioablation, radiofrequency ablation, and thermoablation [77]. Frail and severely frail patients were more likely to be discharged to care facilities rather than home; however, there was no association with major or minor complications, readmission, or reoperation [77].

Non-Ablative Therapies. Locoregional therapies for HCC disease control include TACE, selective internal radiation therapy (SIRT), or stereotactic body radiotherapy [86]. TACE is the most commonly used modality, favored for its minimally invasive nature, similar safety and efficacy across age groups, and survival benefit in appropriately selected patients [91, 92].

There are limited data on frailty and TACE; however, a retrospective study of 125 patients found that higher mFI-5 scores independently predicted 30-day readmission and severe liver toxicity, with each point increase conferring a three-fold and two-fold risk of these outcomes, respectively [81]. Patients with an mFI-5 score ≥2 (indicating frailty) also had reduced transplant-free survival (28.1 vs. 39.8 months) [81]. Similarly, another retrospective study reported significantly lower overall survival in frail patients (40 vs. 24 months), though adverse events were similar between the two groups [82]. No significant difference was observed in the overall response rate (17.53% vs. 10.87%) or disease control rate (39.18% vs. 34.78%); however, there was a trend towards better outcomes in the non-frail group [82].

While no other studies have assessed frailty in patients receiving SIRT, one study evaluating stereotactic body radiotherapy in a cohort of BCLC A and B patients used the G8 questionnaire and Charlson Comorbidity Index (CCI). A G8 score >10 (indicating lower frailty risk) was associated with improved survival, and patients with CCI ≥10 were associated with increased acute toxicity, including grade 1–2 nausea and fatigue and transient grade 2 hepatic failure with ascites [83].

Finally, a prospective study across all BCLC stages, most receiving locoregional therapy first-line, found HGS, but not LFI, to be associated with improved complication-free survival and a trend toward better overall survival [84]. This is surprising given HGS is a component of the LFI, though this may reflect the small number of frail patients (n = 8) in the cohort. In contrast, another study of frailty in 93 HCC patients across all stages found that the LFI was an independent predictor of death when adjusted for BCLC stage and AFP (aHR = 2.38, 95% CI: 1.15–4.91, p = 0.019) [72].

Systemic Therapy. Despite advances in screening programs and imaging technology, up to 40% of HCC patients present with advanced disease at diagnosis [93, 94]. Systemic therapy for these patients includes checkpoint inhibitors combined with another checkpoint inhibitor, a vasoactive agent, or a multikinase inhibitor. While immunotherapy provides a survival benefit overall, direct evidence evaluating the interaction between frailty and immunotherapy outcomes in HCC is limited.

Two studies suggest frailty influences treatment tolerance and survival during systemic therapy. A multicenter prospective cohort of 102 patients undergoing systemic therapy (both checkpoint inhibitors and multikinase inhibitors) demonstrated that frailty was an independent predictor of survival and was associated with earlier discontinuation of therapy [71]. Furthermore, an optimal cut-off of LFI to predict overall survival at 1 year was found to be 4.2, with patients scoring above this experiencing a halved median survival (8.5 vs. 16.4 months) [71]. Despite these findings, the mixed treatment cohort limits subgroup-specific conclusions, and the identified LFI cut-off requires further validation. Another study of elderly patients receiving multikinase inhibitor therapy reported that non-frail patients, identified by the modified G8 score, achieved significantly longer survival with lenvatinib compared to their frailer counterparts (20.0 vs. 7.7 months). In contrast, no difference was found in those treated with sorafenib [85]. Neither study identified Eastern Cooperative Oncology Group (ECOG) performance status or age as predictors of overall survival, reinforcing their limitations as surrogates of physiological reserve and the added prognostic value of frailty assessment. Incorporating frailty measures into clinical decision-making may better identify patients most likely to benefit from systemic therapy while sparing those too vulnerable from unnecessary harm.

Synthesis and Quality of Evidence

Overall, the available evidence demonstrates a consistent association between frailty and adverse clinical outcomes across the HCC treatment spectrum, supporting its role as a clinically relevant prognostic marker. However, this body of evidence is limited by several key factors. Although the surgical literature provides the most robust evidence base, supported by several relatively large multicenter cohorts with broadly consistent findings, the data is predominantly retrospective. In LT, evidence is limited to a single, small prospective study, with additional support from a multicenter analysis that did not specifically investigate HCC. Locoregional therapies are likewise mainly informed by retrospective single-center studies with relatively small sample sizes. Systemic therapy data, while comprising only two studies, includes a multicenter prospective cohort, offering a comparatively stronger, though still limited, level of evidence.

Interpretation of findings across all treatment modalities is further complicated by substantial heterogeneity in both patient populations and methods of frailty assessment. In HCC, where frailty arises from both liver disease and cancer-specific mechanisms, the optimal approach to frailty assessment remains uncertain. Studies included a range of assessment methods, from performance-based to deficit accumulation indices which capture distinct components of vulnerability. Thus, direct comparisons between studies are challenging, and the prognostic performance of different frailty tools in HCC remains unclear. In the systemic therapy studies, both relied on a single baseline frailty assessment, leaving unanswered whether serial frailty assessment better predicts treatment tolerance, immune-related toxicity, and survival. Nevertheless, the consistency of adverse outcomes despite significant methodological variability suggests that frailty, irrespective of the instrument used, captures meaningful impairment beyond conventional prognostic markers. Future prospective studies using standardized frailty assessment tools are required to validate findings and to determine how frailty can be integrated into clinical decision-making.

Modification of Frailty in HCC

Although frailty has been shown to predict morbidity and mortality in HCC, the extent to which it can be modified through intervention and whether such improvement translates to better outcomes remains uncertain. While early work in CLD populations suggests that physical frailty can be improved through nutritional support and structured exercise programs [27, 95, 96], evidence in HCC remains scarce. In oncology populations, managing multidimensional frailty has been shown to reduce chemotherapy toxicity, improve treatment adherence and health-related quality of life [97–100]. Beyond improving physical frailty, exercise interventions in CLD have demonstrated other benefits such as improving aerobic function, muscle strength, health-related quality of life, and hospital LOS [101–103]. Existing studies in HCC are relatively small, heterogeneous, and limited to short exercise-based inpatient interventions, underscoring the need for robust trials in this population.

Exercise Interventions

There have been three studies exploring exercise interventions in HCC specifically (Table 3). The first was a retrospective observational study of 181 patients (139 pre-frail/frail) undergoing any treatment for HCC. Patients in the exercise group (n = 114) undertook a mixture of aerobic and resistance exercises (20–40 min daily) the day following HCC treatment, for a median time of 4 days. This resulted in statistically significant improvements in the LFI during hospitalization compared with the non-exercise group (ΔLFI –0.17 vs. −0.02) [104]. On multivariable analysis, exercise (OR 2.38) and female sex (OR 2.09) were independent predictors of LFI improvement [104]. Innovative outpatient models are also being explored to expand access to frailty interventions. A 10-week group-based home exercise intervention for older HCC patients delivered via videoconference (n = 19; 7 frail, 12 pre-frail), significantly improved mean LFI scores (−0.34, 95% CI: –0.46 to −0.23) but faced engagement challenges, with recruitment at 48%. Qualitative feedback demonstrated good acceptability, with most patients reporting perceived physical and mental health benefits [105]. A 12-week mobile application-based program (n = 37) with high adherence (84%) found significant improvements in HGS, 30-s chair stand, 6MWT, muscle mass, and quality of life pain scores, without biochemical deterioration [106].

Table 3.

Summary of studies addressing frailty modification in HCC

Author	Design	Sample size	Setting	Intervention	Findings
Koya et al. [107] (2017)	Prospective cohort	54	Inpatient, all-stage HCC	F: daily	Preserved 6MWT and CP-class. Improved HR variability during 6MWT. SMI decline not prevented. Branched chain amino acid distinguishing factor for minimizing muscle atrophy
				I: 2.5–4.0 metabolic equivalents (METs)
				T: 20 min
				T: Supervised stretching, strength, balance, and endurance training
Koya et al. [108] (2019)	Case-control	209	Inpatient post-TACE	F: daily	Exercise independently predicted improved SMI
				I: Median 2.5 METs
				T: 20–40 min
				T: Supervised stretching, strength, balance, and endurance training
Hashida et al. [109] (2020)	Prospective cohort	152	Inpatient post-TACE	F: daily	Improved SMI and survival
				I: Median 2.5 METs
				T: 20–40 min/day
				T: Supervised stretching, strength, balance, and endurance training
Narao et al. [110] (2020)	Retrospective single-center chart review	19	Inpatient, all-stage HCC	F: Resistance – 2–3 or 5×/week, stretching 3×/week, aerobic - daily	Improved muscle strength, no hepatic decompensation
				I: Resistance – Bodyweight, stretching – painless strength, aerobic – Borg 11–13
				T: 20–60 min
				T: Supervised stretching, strength, balance, and endurance training
Kim et al. [106] (2020)	Single-arm intervention	37	Outpatient, variable time post HCC treatment	F: daily (via app) for 12 weeks	Increased muscle strength, mobility, and quality of life. Adherence rate 84%
				I: Individualised according to fitness
				T: Individualised
				T: Stretching, aerobic and strength training
Tsuchihashi et al. [104] (2021)	Multicenter prospective	181	Inpatient post-TACE	F: 5×/week	Exercise independent predictor of improvement in LFI
				I: Individualised
				T: 20–40 min
				T: Supervised stretching, resistance, balance and aerobic training
Hallsworth et al. [105] (2024)	Single-arm intervention	40	Outpatient all-stage HCC ≥60 years old	F: 2×/week for 10 weeks	LFI improved. Recruitment rate 48%
				I: Individualised
				T: 45 minutes
				T: Group-based virtual home-based aerobic and resistance exercises

CP, Child-Pugh; LFI, Liver Frailty Index; HCC, hepatocellular carcinoma; HR, heart rate; METs, metabolic equivalents; 6MWT, 6-min walk test; SMI, skeletal muscle index.

Sarcopenia, defined as reduced muscle mass, is a key contributor to frailty and can thus be targeted to complement frailty-focused interventions [31]. Two studies have investigated the safety and efficacy of inpatient exercise-based interventions on modifying sarcopenia in all-stage HCC. In a retrospective review of 19 patients, a median 10-day program of resistance and aerobic exercise was safe and improved functional independence and muscle strength without precipitating hepatic decompensation [110]. Similarly, another study found that 7–13 days of inpatient physiotherapy preserved 6MWT performance and improved heart rate variability, though skeletal muscle atrophy was not completely prevented [107].

Larger Japanese studies have evaluated inpatient sarcopenia-focused interventions in post-TACE patients. A case-control study of 209 inpatients found 20–40 min of daily physiotherapist-led exercise significantly improved skeletal muscle index (SMI) compared to controls (HR 2.13, 95% CI: 1.215–3.846), with exercise being a predictor of improvement despite participants being significantly older [108]. Similarly, a prospective study of 152 patients undergoing inpatient rehabilitation following TACE demonstrated significant SMI improvement compared to controls and was independently associated with survival on regression analysis (HR 1.760) [109]. Survival was also longer in the rehabilitation group compared to controls (median 552 vs. 424 days), remaining significant after propensity matching (median 529 vs. 369 days) [109].

Collectively, these studies demonstrate that exercise interventions in HCC are safe, feasible, and can improve physical performance, muscle mass, and frailty measurements. The challenge remains, however, to move beyond short inpatient interventions and establish structured, scalable outpatient programs with meaningful clinical benefits. Future implementation will need to emphasize outpatient, home-based, and hybrid programs, as illustrated by the group-based intervention reported by Hallsworth et al. [105], although recruitment and long-term engagement remain barriers to broader uptake. Until HCC-specific evidence becomes available, rehabilitation should be guided by established principles from oncology and cirrhosis rehabilitation.

Non-Exercise Interventions

Nutritional Supplementation

There are no studies on interventions addressing non-physical components of frailty in HCC. While nutritional support improves sarcopenia in CLD, its impact on frailty has been less frequently studied. In CLD, data on branched-chain amino acid (BCAA) supplementation have been mixed. While two short-term (12 and 16 week) randomized trials have shown that BCAA supplementation improved muscle mass and LFI scores [111, 112], a 12-month study of 150 patients with cirrhosis (74 BCAA, 76 control) demonstrated no improvement in muscle strength, mass or physical frailty [113]. The differences in these results may be explained by the control group receiving an equivalent whey protein supplement, which likely reduced confounding from differences in total protein intake - a factor not controlled for in other studies. This further affirms the importance of meeting protein goals in CLD patients. Finally, another trial of 100 patients with decompensated cirrhosis found that a 6-month, dietitian-supported intensive home nutrition program improved LFI more than standard care (0.8 vs. 0.4) [114]. On subgroup analysis of patients adherent to total calorie and protein targets at 6 months, significant improvements were noted in liver disease scores and survival (p < 0.05) [114].

Hormonal Supplementation

Low testosterone is common in men with CLD and is associated with sarcopenia and frailty [115]. In a randomized trial of 101 men with cirrhosis and low serum testosterone levels, replacement therapy increased muscle mass [116]. Despite this, data on improvements in clinical outcomes are limited [117]. Although earlier concerns about testosterone increasing HCC risk appear overstated [115], guidelines list a history of HCC as a relative contraindication [31] and provide no recommendations regarding its use in patients with active disease.

Other Interventions

To date, no studies have assessed psychosocial or cognitive interventions for frailty in CLD or HCC, highlighting a key gap. Collectively, these findings support the need for multimodal interventions that extend beyond physical activity and nutrition to address broader dimensions of frailty.

Treatment-Specific Considerations

Despite the lack of direct evidence for treatment-specific frailty interventions in HCC, existing data can be extrapolated to inform potential approaches. In the setting of surgery (resection or LT), “prehabilitation” is a proactive, multimodal intervention which can include nutrition therapy, physical activity, smoking cessation, and psychological support. Such programs have demonstrated post-operative benefits in patients undergoing major abdominal surgery for both cancer and non-cancer indications. This approach could plausibly improve post-operative outcomes in HCC, though this requires validation in a frail HCC cohort. For patients undergoing TACE, the post-procedural exercise interventions summarized above suggest a window for targeted intervention to mitigate treatment-related decline in muscle mass, strength, and function. Whether this approach is generalizable to other locoregional therapies for HCC and whether the intervention is most effective when delivered post-treatment, pre-treatment, or a combination of both remains unknown and requires further investigation. Finally, in systemic therapy, frailty assessments could help identify patients who are unlikely to tolerate or benefit from systemic therapy in advanced HCC, enabling timely redirection to supportive care. It raises the question of whether a “futility threshold” could be defined beyond which systemic therapy may be more harmful than helpful. For patients who are deemed suitable, targeting frailty while on treatment may be beneficial as has been demonstrated in oncology literature where frailty intervention improved treatment tolerance and completion rates [98, 99].

Synthesis and Quality of Evidence

The evidence of frailty modification in HCC remains in its early stages. HCC-specific data only exists for exercise interventions and is limited to a small number of short-term, single-center studies which have marked differences in study populations, intervention protocols, and outcome measures. While studies demonstrate exercise interventions are safe and feasible, most are underpowered to study important clinical endpoints, including healthcare utilization, treatment tolerance, and survival. Non-exercise interventions targeting multidimensional components of frailty in HCC remain unstudied. While broader CLD and oncology literature supports a combination of exercise and nutritional interventions to improve frailty, these findings may not be generalizable to HCC patients. These limitations highlight that while frailty may be modifiable in principle, well-designed prospective trials in HCC are urgently needed to define optimal multimodal management.

Frailty in Clinical Decision Making

Current evidence supports frailty as an independent prognostic marker; however, whether it should influence HCC treatment selection or primarily guide supportive care remains uncertain. The paucity of studies examining its impact on treatment decision-making limits the ability to draw firm conclusions. With more studies, frailty assessment may ultimately be integrated alongside tumor stage, liver function, and patient goals to guide supportive care, rehabilitation, and shared decision-making. Until prospective validation is available, frailty may best be considered complementary to established treatment selection tools rather than a determinant of treatment eligibility.

If future studies confirm that frailty is modifiable, targeted rehabilitation incorporating exercise, nutrition, and psychosocial support could potentially increase a patient’s capacity to receive and tolerate HCC treatment. Rehabilitation may have a role as a bridge to therapy, supporting recovery following decompensation, and improving quality of life. Rather than representing a single pre-treatment assessment, frailty could be considered a dynamic construct, with reassessment informing rehabilitation needs as treatment progresses. Although current evidence is insufficient to recommend treatment-specific rehabilitation protocols, rehabilitation goals are likely to differ according to treatment intent. Figure 3 presents a proposed conceptual framework for integrating frailty assessment and multimodal rehabilitation across the HCC care continuum, noting this requires validation.

Fig. 3.

Proposed framework for integrating frailty assessment and frailty-focused intervention into hepatocellular carcinoma management. BCLC, Barcelona Clinic Liver Cancer; HCC, hepatocellular carcinoma; MDT, multidisciplinary team; ALBI, Albumin-Bilirubin score.

Several implementation questions remain unanswered. It is unclear whether rehabilitation should be delivered universally or targeted to patients with established frailty; whether interventions are most effective before, during, or after treatment or throughout the disease course; and whether supervised inpatient programs provide additional benefit over outpatient, home-based, or telehealth models. The optimal nutritional strategy for patients with HCC also remains undefined, as does the need for treatment-specific rehabilitation protocols and their cost-effectiveness. The next phase of HCC frailty research should move beyond prognostic studies to determine whether frailty-focused interventions improve treatment tolerance and patient outcomes.

Future Directions and Conclusion

Frailty is clearly clinically relevant in HCC, though significant limitations of current evidence include heterogeneity, small sample sizes, and a paucity of prospective studies. Broadly across the HCC spectrum, frailty is associated with adverse outcomes, including higher surgical risk, increased healthcare utilization, reduced treatment tolerance, and survival. Importantly, preliminary interventional studies suggest frailty in HCC may be reversible through exercise and nutritional strategies, though evidence addressing its broader, non-physical dimensions remains absent.

Future research should first and foremost seek to address a consensus framework for frailty and its assessment in the context of HCC. With numerous frailty tools in circulation, efforts should focus toward validating a small set of practical and reproducible measures. A standardized approach will facilitate comparison across studies and be easier to implement into clinical workflows. Mechanistic studies are also required to clarify the biological basis of frailty in HCC, alongside prospective trials to characterize frailty trajectories and determine how dynamic changes relate to treatment tolerance and outcomes. Second, there is an urgent need to define optimal strategies of frailty management in HCC, extending beyond the physical domain to determine whether broader multimodal approaches can improve frailty and ultimately enhance treatment tolerance, quality of life, and overall survival. Finally, given different therapeutic goals and toxicities across HCC treatment modalities, future studies should prioritize investigation of treatment-tailored frailty management rather than adopting a ’one-size-fits-all’ approach.

Taken together, routine incorporation of frailty assessment has the potential to improve prognostication of treatment tolerance and guide therapy selection to minimize harm. Frailty-focused care has the potential to reshape HCC management towards a more personalized and holistic approach that balances therapeutic efficacy with preserved physiological reserve and function.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Rachael Jacob and Shirin Salimi: writing – original draft, review, and editing and visualization. Dinuli Kamaladasa, Karen Waller, Emily He, Mark Danta, Simone I Strasser, Geoff W McCaughan, and Jacob George: writing – original draft, review, and editing. Ken Liu: conceptualization and writing – original draft, review, and editing. Guarantor of article: Ken Liu. All authors have approved the final draft submitted.

Funding Statement

This study was not supported by any sponsor or funder.