Prevalence and Effect of Low Skeletal Muscle Mass among Hepatocellular Carcinoma Patients Undergoing Systemic Therapy: A Systematic Review and Meta-Analysis

Simple Summary The association between low skeletal muscle mass (LSMM) and survival in HCC patients receiving systemic therapy remains inconclusive based on previous studies. Our study aimed to use meta-analysis to aggregate a large sample size and identify the association. The results confirmed that LSMM is prevalent among HCC patients undergoing systemic therapy and is associated with poorer overall survival and progression-free survival. This finding highlights the importance of evaluating muscle mass and early interventions to improve the survival of advanced HCC patients in clinical practice. Abstract Low skeletal muscle mass (LSMM) is associated with poor outcomes in hepatocellular carcinoma (HCC) patients. With the emergence of new systemic therapeutics, understanding the effect of LSMM on HCC treatment outcomes is critically important. This systematic review and meta-analysis investigates the prevalence and effect of LSMM among HCC patients undergoing systemic therapy as reported in studies identified in searches of the PubMed and Embase databases published through 5 April 2023. The included studies (n = 20; 2377 HCC patients undergoing systemic therapy) reported the prevalence of LSMM assessed by computer tomography (CT) and compared the survival outcomes [overall survival (OS) or progression-free survival (PFS)] between HCC patients with and without LSMM. The pooled prevalence of LSMM was 43.4% (95% CI, 37.0–50.0%). A random-effects meta-analysis showed that HCC patients receiving systemic therapy with comorbid LSMM had a lower OS (HR, 1.70; 95% CI, 1.46–1.97) and PFS (HR, 1.32; 95% CI, 1.16–1.51) than did those without. Subgroup analysis according to systemic therapy type (sorafenib, lenvatinib, or immunotherapy) yielded similar results. In conclusion, LSMM is prevalent among HCC patients undergoing systemic therapy and is associated with poorer survival. Early intervention or prevention strategies to improve muscle mass may be necessary for this patient population.


Introduction
Liver cancer is the fourth leading cause of cancer-related deaths globally, with hepatocellular carcinoma (HCC) being the most common form [1,2]. Identifying risk factors for the poor outcomes of HCC is crucial for choosing effective treatments and improving survival

Search Strategy
We searched PubMed and Embase for published studies addressing the prevalence of LSMM and its clinical effects on treatment outcomes in HCC patients undergoing systemic therapy, including studies published up to 5 April 2023 [28,29]. We used a free-text search with appropriate MeSH or Emtree terms related to LSMM and liver cancer. Additional articles were identified in the reference lists of pertinent original studies and relevant reviews. No language restrictions were applied to this search. The detailed search strategy is presented in Table S2.

Literature Selection and Data Extraction
Two reviewers (C.-W.T. and M.-H.K.) individually screened titles and abstracts based on the inclusion criteria and read the full-text articles for final eligibility. The agreement between two reviewers was 87% and 92% in these processes, respectively. To resolve discrepancies in the study selection, a third reviewer (C.-S.H.) was consulted to make the final decisions.
Two authors (C.-W.T. and M.-H.K.) independently collected data, including the name of the first author, publication year, country, setting, study design, treatment regimens, population, number of patients, sex ratio, age, method used for estimating muscle mass, LSMM cut-off value, the prevalence of LSMM, study period, and statistical data on the influence of LSMM on OS and PFS (including adjustment factors). All disagreements were resolved through discussions.

Assessment of Methodological Quality
The methodological quality of the included studies was assessed by two researchers (C.-W.T. and M.-H.K.) independently. The quality of the involved studies was evaluated using the Newcastle-Ottawa Scale (NOS) [30]. The studies for which the researchers differed in their assessment of quality were resolved through discussion.

Statistical Analysis
The random-effects meta-analysis of single proportions was used to estimate the pooled prevalence of LSMM in HCC patients undergoing systemic therapy. The pooled OS and PFS were compared between these patients with and without LSMM using the adjusted HR (or unadjusted HR for studies that did not report the adjusted HR) and 95% CI using a random-effects meta-analysis model. Subgroup analyses were conducted on groups defined by treatment regimen (lenvatinib, sorafenib, or immunotherapy), study region (Asian vs. non-Asian area), method used to estimate muscle mass [skeletal muscle index (SMI) vs. psoas muscle index (PMI)], and study quality (good vs. poor). To address the survival of patients with advanced HCC, we conducted one additional subgroup analysis based on observational time (over vs. under 2 years) in the OS and PFS outcomes [4]. We used I 2 to measure the statistical heterogeneity among the included studies and funnel plots/Egger's test to determine the potential publication bias. Two-sided p < 0.05 was considered statistically significant, and all calculations were performed using Comprehensive Meta-Analysis version 4.0.

Literature Search and Study Selection
We identified 1232 relevant published studies. After excluding 361 duplicated studies, a total of 871 studies were screened based on titles and abstracts, leaving 166 full-text articles for further assessment. The inclusion criteria were not met by 146 of these articles, as follows: 30 studies on hepatic tumors other than HCC; 3 studies with analysis not related to sarcopenia or muscle mass; 99 studies with patients receiving a therapy other than the recommended systemic therapy; 12 studies lacking statistical data on the influence of pretreatment LSMM on OS or PFS; and 2 studies with overlapping patient data. The reasons for exclusion in the final stage are shown in Supplemental Table S3. The flow diagram for this study selection process is shown in Figure 1.

Discussion
This systematic review and meta-analysis involved 20 retrospective studies and 2377 patients. Our results demonstrate that LSMM is a common occurrence among HCC patients receiving systemic therapy, with a prevalence of 43.4%. Our findings indicate that low LSMM is associated with a reduced OS. This correlation was observed across different treatment types (such as sorafenib, lenvatinib, and immunotherapy), geographical regions (Asian and non-Asian), and measurement methods (SMI and PMI). Despite previous studies reporting no correlation between LSMM and PFS [11,15,[19][20][21][23][24][25]33], our meta-analysis showed that LSMM is indeed associated with poor PFS among patients undergoing several types of systemic therapy.
Patients undergoing systemic therapy for HCC often have LSMM, with a prevalence ranging from 15 to 65.1% in the studies included in this meta-analysis. The cumulative prevalence in this population was 43.4% (95% CI, 37.0-50.0%), which is similar to that found in a previous meta-analysis of HCC patients treated with sorafenib or lenvatinib (41%, based on data from 10 studies and 1028 patients) [18]. Compared to the overall HCC population, the prevalence of LSMM was higher among those undergoing systemic therapy. For example, a previous meta-analysis by Chuan et al. that included 42 studies with 8203 patients found that the pooled prevalence of sarcopenia among HCC patients was 39% (95% CI, 33-45%; n = 8203) [7]. Another meta-analysis by March et al. including 25 studies reported a cumulative prevalence of 38.5% for LSMM in patients with HCC [8]. Patients with advanced HCC (BCLC stage C) or cancer progression to an advanced stage typically undergo systemic treatment as the standard of care [1,3,4]. These patients often have sarcopenia as a result of the tumor burden and systemic cytokine-mediated inflammation caused by the cancer [10,37]. Furthermore, disease progression and previous cancer treatments can lead to cachexia, a condition that causes loss of muscle and weight [38]. With such a high prevalence, LMSS seems to be an important issue in this population.
Multiple meta-analyses have demonstrated an association between sarcopenia and poor prognosis for cancers such as lung, breast, and lymphoma [39][40][41], although the mechanism underlying this effect is unclear. Previous meta-analyses have also demonstrated that LSMM affects both OS and PFS in patients with HCC [7,8]. Our study, which focused on patients receiving systemic therapy with CT scans used to measure muscle mass, found that LSMM was also associated with significantly reduced OS and PFS. In summary, the presence of LSMM was found to have a negative effect on OS and PFS among HCC patients receiving systemic therapy. Our findings suggest that LSMM may be a valuable tool for guiding treatment decisions and improving patient outcomes in the management of HCC.
Further research is needed to confirm these results and to determine the optimal therapeutic approach for patients with LSMM.
The previous meta-analyses only showed that LSMM is a predictor of OS in patients receiving kinase inhibitors, such as sorafenib or lenvatinib [8,18]. The subgroup analysis in our study provided stronger evidence for the effect of LSMM on OS and also on PFS in patients receiving the same treatments. The association between LSMM and poor survival in patients treated with kinase inhibitors may result from sarcopenia patients often receiving lower initial doses of medication and experiencing more dose-limiting toxicity [10,42]. Both treatments have been found to result in grade 3-4 drug-related adverse events in about 50% of treated patients, resulting in a withdrawal rate near 15% [1]. Studies have suggested that sarcopenia can decrease a patient's ability to tolerate medication [12,14,15]. Additionally, a direct pathway may be involved, such as changes in the phosphatidylinositol-3-kinase/AKT-mammalian target of rapamycin pathway, which is essential for muscle protein synthesis [5,43]. Dysregulation of this pathway caused by medication may lead to further muscle loss [44]. These results suggest that LSMM should be considered as a factor in decision making for protein kinase inhibitor treatment in HCC patients.
No meta-analysis has addressed the association between LSMM and the effectiveness of immunotherapy in patients with HCC. The major studies have found no significant correlation between LSMM and survival (OS and PFS) in patients receiving immunotherapy [20,24,25]. Two studies found a correlation between LSMM and survival [19,26]. After pooling these studies, this meta-analysis revealed that patients receiving immunotherapy who had LSMM had a shorter OS and PFS than did those without LSMM. Although the relationship between LSMM and reduced effectiveness of immunotherapy is unclear, it may be explained by the tumor microenvironment (inflammation and immunity) and cytokine activity. Skeletal muscle is now considered an organ with immune regulatory properties [45]. It produces and releases important proteins known as myokines, which counteract the harmful effects of proinflammatory adipokines and contribute to the tumor microenvironment [10,[45][46][47]. However, when muscle wasting occurs, these counteracting effects disappear, resulting in systemic inflammation and immune disturbances [47]. Studies have shown that inflammatory factors and LSMM are associated with survival in HCC patients receiving immunotherapy [20,21,24]. On the other hand, skeletal muscle cells modulate immune function by signaling through different soluble factors, cell surface molecules, or cell-to-cell interactions [45]. The interaction between muscle wasting and immune senescence appears to be bidirectional. Cytokines such as transforming growth factor-β [48] and interleukin-6 [49] may play a role in the development of sarcopenia and contribute to T-cell exhaustion, which can lead to a poor response to immunotherapy [10,50]. Skeletal muscle depletion also causes a decrease in myokine levels, which may result in a poor response to immunotherapy [45,46]. Therefore, LSMM may affect immune regulation and increase resistance to immunotherapy, leading to poorer outcomes among HCC patients.
The key strength of this study is the use of updated data from a variety of studies that included HCC patients receiving systemic therapy. These additional data increased the statistical power of this study, enabling subgroup analyses of the effect of LSMM on OS and PFS. As a result, this study can aid physicians in understanding the effect of LSMM in patients receiving immunotherapy. Furthermore, while most previous studies reported no association between LSMM and progression-free survival among HCC patients receiving systemic therapy, our meta-analysis revealed a correlation due to the larger sample size polled. Third, this is the first meta-analysis to address the association between LSMM and the effectiveness of immunotherapy in patients with HCC. Further research is needed to better understand the underlying mechanisms and potential interventions.
However, we acknowledged several limitations in this systematic review and metaanalysis. First, the variation between studies in the methods and cutoff values used to determine LSMM through CT images may have resulted in the varying prevalence rates. However, the subgroup analyses suggests that the association between LSMM and survival is not affected by these differences, indicating that LSMM has a negative effect regardless of method and cutoff value. For future research, it is important to use standardized assessment methods and cutoff values, as well as conducting prospective studies. Second, most studies only included Asian patients, primarily from Japan, leading to uncertainty about the applicability of these findings to patients from other regions. Additional studies that include patients from other regions are needed to address this limitation. Third, the patient cohorts used to study sarcopenia among HCC patients vary with respect to etiology and Child-Pugh scores. To obtain a more precise evaluation of the effect of sarcopenia on mortality, a new prediction system that considers LSMM is needed. Fourth, some included studies in this meta-analysis had small sample sizes, but the results from these included studies were consistent with the pooled results of the meta-analysis. In addition, the sensitivity analysis using the one-study removal method indicated that the pooled results were robust. Therefore, the impacts from the inclusion of small sample sizes on our findings were relatively minor.

Conclusions
LSMM is prevalent among HCC patients undergoing systemic therapy with drugs such as sorafenib, lenvatinib, and new immunotherapeutic agents and is associated with poor OS and PFS. Early intervention or prevention strategies to improve muscle mass may be necessary for this patient population.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cancers15092426/s1, Table S1. PRISMA checklist; Table S2. Search strategy; Table S3. Excluded studies and reasons; Table S4. Quality of the included cohort studies as assessed by Newcastle-Ottawa Scale; Table S5. Subgroup analysis of the prevalence of LSMM among HCC patients undergoing systemic therapy; Table S6. Subgroup analysis of HR and 95% confidence interval of overall survival among HCC patients undergoing systemic therapy with and without LSMM; Table S7. Subgroup analysis HR and 95% confidence interval of progression free survival among HCC patients undergoing systemic therapy with and without LSMM; Figure S1. Publication bias; Figure S2. Sensitivity analysis.