PROGNOSTIC VALUE OF LOW SKELETAL MUSCLE MASS IN HEPATOCELLULAR CARCINOMA PATIENTS TREATED WITH SORAFENIB OR LENVATINIB: A META-ANALYSIS
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Original article:
PROGNOSTIC VALUE OF LOW SKELETAL MUSCLE MASS IN
HEPATOCELLULAR CARCINOMA PATIENTS TREATED WITH
SORAFENIB OR LENVATINIB: A META-ANALYSIS
Jun Guan, Qin Yang, Chao Chen, Gang Wang, Haihong Zhu*
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical
Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and
Treatment of Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of
Medicine
* Corresponding author: Haihong Zhu, State Key Laboratory for Diagnosis and Treatment
of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collabo-
rative Innovation Center for Diagnosis and Treatment of Infectious Diseases, The First Af-
filiated Hospital, Zhejiang University School of Medicine, NO.79 Qingchun Road, Hang-
zhou, Zhejiang, China. Phone: +86-571-87236579, E-mail: Zhuhh72@zju.edu.cn
http://dx.doi.org/10.17179/excli2020-3111
This is an Open Access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0/).
ABSTRACT
Growing evidence indicates that skeletal muscle depletion has a notable effect on the prognosis of hepatocellular
carcinoma (HCC) patients, though study results are still controversial. Our meta-analysis aimed at evaluating the
prognostic significance of low skeletal muscle mass (LSMM) in HCC patients treated with sorafenib or len-
vatinib.We systematically reviewed for PubMed, Cochrane, and Embase databases from their inception to August
2020 and obtained all relevant articles describing an association between LSMM and HCC patients treated with
sorafenib or lenvatinib. Demographic and characteristics of included studies, diagnostic criteria of skeletal muscle
depletion, and main outcomes (overall survival, progression-free survival, time to treatment failure) were retrieved.
Associations were expressed by calculating hazard ratios (HRs) and 95 % confidence intervals (CIs).The meta-
analysis enrolled 11 studies comprising 1148 patients. Without significant heterogeneity between studies, LSMM
was significantly associated with poor overall survival (crude HR=1.58, 95 % CI: 1.36–1.83; adjusted HR=1.83,
95 % CI: 1.46–2.29) and time to treatment failure (crude HR=1.85, 95 % CI: 1.34–2.54; adjusted HR=1.72, 95 %
CI: 1.24–2.38). However, there was no significantly association between LSMM and progression-free survival
(adjusted HR=1.44, 95 % CI: 0.95–2.20). Symmetry of distribution on the funnel plot did not show significant
publication bias.This meta-analysis supported that LSMM is significantly associated with poor overall survival
and time to treatment failure in HCC patients after sorafenib or lenvatinib administration. This negative effect was
pronounced even after adjustment for confounders. Future studies should be carried out on larger samples and
study regions based on standardized thresholds of LSMM.
Keywords: Low skeletal muscle mass, sorafenib, lenvatinib, hepatocellular carcinoma, prognosis
INTRODUCTION death globally (Ferlay et al., 2019). Especially
Hepatocellular carcinoma (HCC), charac- in Africa and East Asia, HCC has caused se-
terized by high incidence and high mortality, vere economic and health care burdens. Due
is the sixth most malignant tumor and ranks to inconspicuous symptoms of early HCC, a
fourth in the list of causes of cancer-related large majority of patients are not diagnosed
1EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
with it until advanced stages. They tend to muscle mass (LSMM) in unresectable HCC
have restricted treatment options and poor patients treated with the first-line TKIs.
outcomes.
Sorafenib, an oral kinase inhibitor, can
MATERIALS AND METHODS
simultaneously inhibit molecules and path-
ways relevant to tumor proliferation and angi- Search strategies
ogenesis (Wilhelm et al., 2004). It was firstly Electronic databases involving PubMed,
recommended as the first-line treatment for Embase, and Cochrane Library were searched
advanced HCC patients that are refractory to and browsed to obtain all eligible articles
locoregional therapy, resection, or transplan- without any restrictions on publication lan-
tation. Compared to placebo, sorafenib is ben- guage and year. The following terms were
eficial in prolonging time to progression and employed to complete search function: (“so-
median overall survival (OS) (Cheng et al., rafenib” OR “Nexavar” OR “lenvatinib” OR
2009; Keating, 2017). Later, lenvatinib, an- “lenvima” OR “tyrosine kinase inhibitors”
other tyrosine kinase inhibitor (TKI), demon- OR “TKIs”) AND (“sarcopenia” OR “skeletal
strated a comparable efficacy to sorafenib and muscle” OR “muscle depletion”) AND
was even superior in increasing progression- (“hepatocellular carcinoma” OR “liver can-
free survival (PFS) (Kudo et al., 2018), thus cer” OR “liver cell carcinoma” OR “hepa-
was approved for the second first-line drug by toma” OR “HCC”). We also examined the
the National Medical Products Administra- reference lists of satisfied publications to
tion (NMPA) in September 2018. Regardless search for more relevant citations.
of their remarkable efficacy, adverse effects
cannot be neglected, such as renal toxicity, fa- Inclusion and exclusion criteria
tigue, diarrhea, hand–foot skin reaction, Eligible studies needed to meet the fol-
weight loss and hypertension, and may result lowing criteria: (1) retrospective or prospec-
in dose reduction or discontinuation under se- tive studies (2) treated with sorafenib or len-
vere conditions. These adverse events accel- vatinib rather than other kinase inhibitors (3)
erate disease progression and shorten survival the outcome was OS, PFS or time to treatment
by muscle depletion (Antoun et al., 2010). failure (TTF) and (4) provided hazard ratios
Thus, we should pay more attention to the (HRs) and 95 % confidence intervals (CIs).
changes in body composition during soraf- Case reports, review articles, duplicate litera-
enib administration. ture, and studies involving other kinase inhib-
Skeletal muscle depletion, termed as sar- itors or without any outcome of interest were
copenia, is defined by progressive and gener- excluded. When it came to studies with over-
alized loss of muscle mass and muscle func- lapped patient data, we chose the one involv-
tion (Cruz-Jentoft and Sayer, 2019), which is ing the largest sample size and the longest du-
related to aging, nutritional disorders, or some ration.
underlying diseases. Loss of skeletal muscle
mass contributes to cancer-associated ca- Data extraction and quality assessment
chexia and further seriously threatens the Two authors (JG and QY) independently
quality of life and survival. Reversing sarco- collected data using specially-designed elec-
penia markedly ameliorates the quality of life tronic forms. The following details were ex-
in breast cancer patients (Adams et al., 2016). tracted: first author’s name, publication year,
An increasing number of studies focus on the title, country, study design, enrolled numbers
relationship between skeletal muscle deple- (male vs female), HCC stage, age, prevalence
tion and poor outcomes in malignancies. of LSMM, details about measured muscle,
Therefore, our meta-analysis intended to eval- cut-off value for LSMM, outcome variables
uate the prognostic importance of low skeletal (OS, PFS and TTF), and adjustment factors.
OS was defined as the interval from the initial
2EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
date of TKIs administration to the date of text articles were assessed for eligibility. 18
death or last follow-up. PFS was defined as records didn't meet the inclusion criteria and
the interval from the initial date of TKIs ad- were discarded: 2 involving other kinase in-
ministration to the date of death, disease pro- hibitors (Nault et al., 2013, 2015), 9 with
gression or last follow-up. TTF was defined overlapped patient data (Antonelli et al.,
as the interval from treatment initiation to the 2018a; Gigante et al., 2015; Hoshino et al.,
end or last follow-up. 2015; Imai et al., 2015, 2019; Labeur et al.,
The quality evaluation of the involved 2018a, b; Okada et al., 2019; Saeki et al.,
studies was performed by using the Newcas- 2019), 5 lacking HR and 95 % CI (Mir et al.,
tle-Ottawa Scale (NOS). Studies were scored 2012; Okada et al., 2020; Saeki et al., 2018;
based on three major criteria: the selection of Uchikawa et al., 2020; Ueki et al., 2016), 1
the study groups (four items); the comparabil- without any interesting outcome (Cheng et al.,
ity of the groups (one item); and the ascertain- 2019) and 1 with low quality and faulty data
ment of either the outcome or exposure of in- (Nishikawa et al., 2017). Thus, 11 retrospec-
terest for cohort or case-control studies re- tive studies (Antonelli et al., 2018b; Endo et
spectively (three items). The maximum score al., 2020; Hiraoka et al., 2017; Imai et al.,
of the NOS was 9 points. Studies with scores 2020; Labeur et al., 2019; Naganuma et al.,
of more than 6 points were considered to be 2017; Sawada et al., 2019; Takada et al.,
of high quality; less than 4 points of low qual- 2018; Uojima et al., 2020; Wu et al., 2021;
ity; while those with scores of 4 to 6 were of Yamashima et al., 2017) were included in this
medium quality. meta-analysis, comprising 1148 patients. The
flow diagram of this study selection process
Statistical analysis is shown in Figure 1.
The outcomes for the association between
LSMM and OS, PFS or TTF were expressed Characteristics of included studies
as crude and adjusted HRs with 95 % CIs. The demographic and characteristics of
HRs and 95 % CIs were obtained directly the 11 eligible studies are shown in Table 1.
from univariate and multivariate COX regres- Out of these retrospective studies, 9 were per-
sion analyses and needed to be further con- formed in Asia (8 in Japan and 1 in China) and
verted into natural logarithm (lnHR) and the remaining 2 in Europe (Netherlands and
standard error (SE). We assessed heterogene- Italy). The vast majority of the HCC patients
ity by using Cochran's Q statistic, with p < 0.1 were in an advanced stage. The median ages
and I2 > 50 % being suggestive of meaningful of participants ranged from 64 to 72 years.
heterogeneity (Higgins et al., 2003). When Male patients accounted for the vast majority
heterogeneity was observed, the random-ef- of all the participants. All diagnoses were
fects model was selected; otherwise the fixed- made through computed tomography scans,
effects model was utilized. Potential publica- but the measured indicators and cut-off values
tion bias was evaluated by using funnel plots. for LSMM varied. In 8 studies, the total skel-
All calculations were performed using Re- etal muscle (TSM) mass was quantified at the
view Manager 5.3, and p < 0.05 was consid- third lumbar level (L3). One study measured
ered statistically significant. the psoas muscle (PM) mass at L3 and 1 study
measured transverse psoas muscle thickness
at the level of the umbilicus. Moreover, 1
RESULTS
study defined LSMM based on TSM, PM, and
Search results rectus abdominis (RA) indices respectively.
Of the 126 studies identified through da- After excluding this study without reported
tabase searching, 31 duplicated studies were number (Wu et al., 2021), the frequency of
excluded and 95 studies were screened. After LSMM patients reached 41 %. The results of
being excluded by titles and abstracts, 29 full-
3EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
the quality evaluation are demonstrated in Ta- poor prognosis (Figure 2A). Eight studies in-
ble 2, and all studies were regarded as being volving 661 patients provided the adjusted
of high quality. HRs and 95 % CIs (Antonelli et al., 2018b;
Hiraoka et al., 2017; Imai et al., 2020; Na-
Overall survival ganuma et al., 2017; Sawada et al., 2019;
The main results of the crude and adjusted Uojima et al., 2020; Wu et al., 2021; Yama-
pooled analysis are reported in Figure 2A and shima et al., 2017). One of these studies de-
2B respectively. Ten studies involving 1028 fined LSMM based on TSM, PM, and RA in-
patients provided the crude HRs and 95 % CIs dices and provided three different corre-
of the association between the LSMM and sponding HRs (Wu et al., 2021), so three ad-
HCC patients treated with sorafenib or len- justed pooled results were obtained. There
vatinib (Antonelli et al., 2018b; Endo et al., was no heterogeneity (p value = 0.9; I2 = 0 %)
2020; Hiraoka et al., 2017; Imai et al., 2020; regardless of the three different HRs, so we
Labeur et al., 2019; Naganuma et al., 2017; conducted a forest plot by applying a fixed-
Sawada et al., 2019; Takada et al., 2018; effects model. The adjusted pooled HRs were
Uojima et al., 2020; Yamashima et al., 2017). 1.83 (95 % 1.46, 2.29; p < 0.00001), 1.78
A fixed-effects model was utilized with no (95 % 1.43, 2.21; p < 0.00001), and 1.75
significant heterogeneity (p value=0.33; (95 % 1.41, 2.18; pEXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Table 1: Demographic and characteristics of included studies
First Coun- Study Enrolled Age LSMM Muscle measured Cut-off value for Out- Adjustment factors
author, try design number (years) preva- LSMM come
year (male/ lence varia-
female) ble
Antonelli Italy Retro- 96 Median: 49 % CT scan: The cross-sectional 43.0 cm2/m2 for men OS, Age, gender, BMI, complication, INR, vascu-
et al., spective (75/21) 69 M: 37 % areas of the muscle at the L3 with BMIEXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Table 1 (cont.): Demographic and characteristics of included studies
First Coun- Study Enrolled Age LSMM Muscle measured Cut-off value for Outcome Adjustment factors
author, try design number (years) prevalence LSMM variable
year (male/
female)
Endo et Japan Retro- 63 (53/10) Median: 71 35 % CT scan: The skeletal muscle mass 42.0 cm2/m2 for men None (LSMM was not included in
al., 2020 spective M:30 % at the L3 divided by the square of and 38.0 cm2/m2 for the multivariate analysis)
study the height (SMI), cm2/m2 women
Uojima et Japan Retro- 100 71.5±9.2 59 % CT scan: The skeletal muscle mass 42.0 cm2/m2 for men OS, TTF Age, gender, liver function, BW,
al., 2020 spective (75/25) M:71 % at the L3 divided by the square of and 38.0 cm2/m2 for previous therapy, refractory to
study the height (SMI), cm2/m2 women transcatheter treatment
Takada et Japan Retro- 146 Not 58 % CT scan: The skeletal muscle mass 42.0 cm2/m2 for men OS None (LSMM was not included in
al, 2018 spective reported at the L3 divided by the square of and 38.0 cm2/m2 for the multivariate analysis)
study the height (SMI), cm2/m2 women
M: male; BMI, body mass index; CT, computed tomography; HU, Hounsfield unit; OS, overall survival; L3, third lumbar vertebra; SMI, skeletal muscle index (cm2/m2); PSI, psoas muscle index (cm2/m2);
LSMM: low skeletal muscle mass; INR: international normalized ratio; TTF: time to treatment failure; BW: body weight; PFS: progression-free survival
Table 2: Quality assessment by using The Newcastle-Ottawa Scale
First author, Is the case Representa- Selection of Definition of Comparability of cases Ascertain- Same method of as- Non-Re- Total
year definition tiveness of Controls Controls and controls on the basis ment of ex- certainment for cases sponse rate scores (*)
adequate? the cases of the design or analysis posure and controls
Antonelli et * * - * ** * * * 8
al., 2018b
Hiraoka et * * - * ** * * * 8
al., 2017
Imai et al., * * - * ** * * * 8
2020
Labeur et al., * * - * ** * * * 8
2019
Naganuma et * * - * * * * * 7
al., 2017
Sawada et * * - * ** * * * 8
al., 2019
Wu et al., * * - * * * * * 7
2020
Yamashima * * - * * * * * 7
et al., 2017
Endo et al., * * - * ** * * * 8
2020
Uojima et al., * * - * ** * * * 8
2020
Takada et al., * * - * * * * * 7
2018
* =1 point
6EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
A
B
7EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
C
D
E
Figure 2: Forest plot evaluating the association between the low skeletal muscle mass and hepatocel-
lular carcinoma patients treated with sorafenib or lenvatinib: crude (A) and adjusted (B) HRs between
low skeletal muscle mass and overall survival, adjusted HR between low skeletal muscle mass and
progression-free survival (C), crude (D) and adjusted (E) HRs between low skeletal muscle mass and
time to treatment failure.
To further investigate the association be- that LSMM could be a poor prognostic factor
tween LSMM and prognosis in HCC patients for OS in HCC patients after sorafenib admin-
with the first-line TKIs administration, we istration.
conducted subgroup analyses stratified by
types of TKIs (sorafenib or lenvatinib), study Progression-free survival
region (Europe and Asia), muscle measured Only two studies involving 202 patients
(skeletal muscle index [SMI] or others) and reported the adjusted HRs and 95 % CIs of the
whether body mass index (BMI) or under- association between the LSMM and HCC pa-
weight or body weight was involved in multi- tients treated with sorafenib or lenvatinib
variate analysis (BMI adjusted [+] or BMI ad- (Sawada et al., 2019; Wu et al., 2021). One
justed [-]). When stratifying by types of TKIs, defined LSMM based on TSM, PM, and RA
we found significantly negative impact of indices and provided three different corre-
LSMM on OS in patients treated with soraf- sponding HRs, the other is based on TSM, so
enib (pEXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
The symmetrical distribution on the funnel was utilized with no significant heterogeneity
plot indicated that there was no publication (p value > 0.5; I2 = 0 %). LSMM was signifi-
bias. The stratified analysis was not con- cantly associated with TTF with a crude
ducted owing to the limited number of studies pooled HR of 1.85 (95 % CI 1.34–2.54;
involving PFS (Figure 3C). p = 0.0002) and an adjusted pooled HR of
1.72 (95 % CI 1.24–2.38; p = 0.001) (Figure
Time to treatment failure 2D and Figure 2E). Symmetry of distribution
Likewise, only two studies involving 196 on the funnel plot supported no evidence of
patients reported the crude and adjusted HRs publication bias. The stratified analysis was
and 95 % CIs of the association between the not conducted because of the limited number
LSMM and HCC patients treated with soraf- of studies involving TTF (Figure 3D and Fig-
enib or lenvatinib (Antonelli et al., 2018b; ure 3E).
Uojima et al., 2020). A fixed-effects’ model
9EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
A
B
10EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
C
D
11EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
E
Figure 4: Forest plot of the subgroup analyses for the crude association between low skeletal muscle
mass and overall survival stratified by types of TKIs (sorafenib or lenvatinib) (A), by study region (Europe
and Asia) (B) and by muscle measured (SMI or others) (C); the adjusted association between low skel-
etal muscle mass and overall survival stratified by muscle measured (SMI or others) (D), by whether
body mass index or underweight or body weight was involved in multivariate analysis (BMI adjusted [+]
or BMI adjusted [-]) (E).
BMI, body mass index; SMI, skeletal muscle index (cm 2/m2)
DISCUSSION correlated with poor OS in multiple solid tu-
mors (involving HCC) (Shachar et al., 2016).
Our meta-analysis paid attention to the
Chang et al. further conducted a meta-analy-
impact of LSMM on the prognosis of HCC
sis including 13 HCC studies, and concluded
patients treated with the first-line TKIs for the
that sarcopenia was associated with increased
first time. Based on 11 studies and 1148 pa-
all-cause mortality and tumor recurrence in
tients, we found that LSMM has a negative
HCC patients (Chang et al., 2018). However,
effect on OS and TTF, but has no significant
due to the limited number of studies, the
impact on PFS. Even after adjusting for rele-
above studies did not perform stratified anal-
vant confounders, this correlation about OS
and TTF remained pronounced. Except for yses based on tumor stage. Considering that
the subgroups stratified by types of TKIs, the advanced HCC patients often present with
pooled results for the remaining subgroup skeletal muscle depletion and existing studies
analyses were not observably influenced. Our have a dispute over the relationship between
results supported that LSMM may be a prom- LSMM and prognosis after sorafenib intro-
ising poor prognosis for outcomes in HCC pa- duction, so it is of great clinical significance
tients treated with the first-line TKIs. to validate the relationship based on the latest
Reported studies demonstrated that skele- research.
tal muscle mass is associated with the prog- LSMM was prevalent in HCC patients
treated with sorafenib or lenvatinib in our ar-
nosis of multiple malignancies and postoper-
ticles, with reported prevalence rates ranging
ative complications of HCC. A meta-analysis
from 20 % to 59 %. The potential mecha-
of 38 studies demonstrated that LSMM was
nisms are as follows (Antoun et al., 2010;
12EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Nishikawa et al., 2016): (1) Insufficient gly- adverse reactions. Therefore, it is recom-
cogen storage. To compensate for glycogen mended to carry out prospective studies aim-
depletion, skeletal muscles degrade to provide ing to investigate whether sorafenib and len-
glucose and amino acids (such as branched- vatinib can benefit the survival of LSMM pa-
chain amino acids, BCAA) and result in a de- tients.
crease in blood BCAA. Thus, the function of Early evaluation and adequate interven-
BCAA as the strongest material for protein tion of high-risk factors can improve the prog-
synthesis to maintain and increase muscle nosis. Efficient treatment of LSMM includes
mass is hindered; (2) Impaired synthesis of in- exercise, nutritional support and pharmaco-
sulin growth factor 1 (IGF-1). IGF-1 aims to logical agents (Dutt et al., 2015; Nishikawa et
maintain the dynamic balance between pro- al., 2016). After exercise, IGF-1 synthesized
tein anabolism and catabolism; (3) Increased by hepatocytes and myocytes is up-regulated.
level of blood myostatin. Myostatin is a mem- Nutritional support, such as BCAA, contrib-
ber of the transforming growth factor β (TGF- utes to protein synthesis and increased skele-
β) family and can strongly inhibit skeletal tal muscle mass. The restored skeletal muscle
muscle growth; (4) Up-regulated inflamma- mass can prolong the duration of the first-line
tory cytokines and reactive oxygen species. TKIs administration, improve survival time
Tumor necrosis factor-α (TNF-α) and inter- and the quality of life of patients with ca-
leukin-6 (IL-6) can accelerate protein catabo- chexia at end-stage.
lism. Reactive oxygen species can inhibit pro- Undeniably, there are several limitations
tein anabolism; (5) Sorafenib can suppress to our meta-analysis. Firstly, the articles in-
muscle protein synthesis directly by inhibit- volved were retrospective, with limited num-
ing mTOR phosphorylation that triggers mus- bers of participants and just a few regions.
cle protein synthesis under activated condi- Retrospective assessment of the outcome
tions. Compared to controls, LSMM is signif- could be associated with selection bias and re-
icantly associated with an increased risk of porting bias. As such, a prospective study in-
mortality. High levels of tumor necrosis fac- cluding a larger sample size in multiple cen-
tor-α (TNF-α) and interleukin-6 (IL-6) play ters should be conducted. Secondly, there
an indispensable role. TNF-α, act as an im- were limited articles to enable stratified anal-
portant regulator of the tumor microenviron- ysis based on the study region and types of
ment, can promote tumor migration and inva- TKIs. Due to the difference in the cut-off
sion by the TNF-α-NF-κB-Snail pathway value and the basic characteristics of the pop-
(Wu and Zhou, 2010). Over-expression of IL- ulation, a combined analysis is not the best.
6 activates hepatocarcinogenesis and deterio- Thirdly, measured muscles and cut-off values
rates liver function through p-STAT3 (Kao et that defined LSMM of all included articles
al., 2015). Sorafenib and lenvatinib, as the vary considerably between Asia and Europe.
first-line drugs for advanced liver cancer, are Cut-off values for LSMM may be gender-spe-
recognized to inhibit tumor proliferation cific and weight-specific. Differences in cut-
and angiogenesis, thus prolonging survival. off values of LSMM have an impact on the
On the other hand, sorafenib inhibits skeletal results. It is necessary to reach an interna-
muscle protein synthesis and may lead to tional consensus on the diagnostic criteria of
LSMM. The reduction of skeletal muscle LSMM as soon as possible. Lastly, the initial
mass is considered one of the criteria for di- doses of sorafenib were different between dif-
agnosing both sarcopenia and cancer cachexia ferent cohorts, and this may have caused some
(Fearon et al., 2011; Nishikawa et al., 2016). bias.
Sarcopenia and cancer cachexia may lead to
LSMM and poor prognosis. Actually, patients
with LSMM tend to have a shorter admin-
istration duration of sorafenib due to serious
13EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
CONCLUSION Antoun S, Birdsell L, Sawyer MB, Venner P, Escudier
B, Baracos VE. Association of skeletal muscle wasting
Based on this meta-analysis, we con- with treatment with sorafenib in patients with advanced
cluded that LSMM is associated with poor OS renal cell carcinoma: results from a placebo-controlled
and TTF in HCC patients treated with soraf- study. J Clin Oncol. 2010;28:1054-60.
enib or lenvatinib. This negative effect was Chang KV, Chen JD, Wu WT, Huang KC, Hsu CT, and
enhanced even after adjustment for confound- Han DS. Association between loss of skeletal muscle
ers. Shortly, we should enlarge the sample mass and mortality and tumor recurrence in hepatocel-
and study more regions based on the standard- lular carcinoma: A systematic review and meta-analy-
ized threshold of LSMM when performing sis. Liver Cancer. 2018;7:90-103.
more prospective studies. It is equally im- Cheng AL, Kang YK, Chen Z, Tsao CJ, Qin S, Kim JS,
portant to validate whether LSMM patients et al. Efficacy and safety of sorafenib in patients in the
can benefit from sorafenib or lenvatinib treat- Asia-Pacific region with advanced hepatocellular car-
ment. After all, the ultimate goal of all the cinoma: a phase III randomised, double-blind, placebo-
controlled trial. Lancet Oncol. 2009;10:25-34.
therapy is to maximize the benefits of patients
with end-stage malignancies. Cheng TY, Lee PC, Chen YT, Hou MC, Huang YH.
Sarcopenia determines post-progression outcomes in
Funding advanced hepatocellular carcinoma after sorafenib fail-
ure. J Hepatol. 2019;70:e833-4.
This study was supported by grants
awarded by the National Science and Tech- Cruz-Jentoft AJ, Sayer AA. Sarcopenia. Lancet. 2019;
nology Major Project of China (NO 393:2636-46.
2018ZX10302206), Science and Technology
Dutt V, Gupta S, Dabur R, Injeti E, Mittal A. Skeletal
Major Projects of Zhejiang Province (NO muscle atrophy: Potential therapeutic agents and their
2018C04016). mechanisms of action. Pharmacol Res. 2015;99:86-
100.
Conflict of interest
Endo K, Kuroda H, Kanazawa J, Sato T, Fujiwara Y,
All authors declare that they have no con-
Abe T, et al. Impact of grip strength in patients with
flict of interest. unresectable hepatocellular carcinoma treated with len-
vatinib. Cancers. 2020;12(8):2146 .
REFERENCES Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E,
Adams SC, Segal RJ, McKenzie DC, Vallerand JR, Fainsinger RL, et al. Definition and classification of
Morielli AR, Mackey JR, et al. Impact of resistance and cancer cachexia: an international consensus. Lancet
aerobic exercise on sarcopenia and dynapenia in breast Oncol. 2011;12:489-95.
cancer patients receiving adjuvant chemotherapy: a
Ferlay J, Colombet M, Soerjomataram I, Mathers C,
multicenter randomized controlled trial. Breast Cancer
Parkin DM, Piñeros M, et al. Estimating the global can-
Res Treat. 2016;158:497-507.
cer incidence and mortality in 2018: GLOBOCAN
Antonelli G, Gigante E, Iavarone M, Begini P, sources and methods. Int J Cancer. 2019;144:1941-53.
Biondetti P, Pellicelli AM, et al. Sarcopenia predicts
Gigante E, Antonelli G, Begini P, Carbonetti F, Ianni-
survival in patients with advanced hepatocellular car-
celli E, Marchetti P, et al. Sarcopenia is associated with
cinoma treated with Sorafenib. J Hepatol. 2018a;68:
a reduced survival in patients with hepatocarcinoma
S207-8.
undergoing sorafenib treatment. J Hepatol. 2015;62:
Antonelli G, Gigante E, Iavarone M, Begini P, Sangio- S449.
vanni A, Iannicelli E, et al. Sarcopenia is associated
Higgins JP, Thompson SG, Deeks JJ, Altman DG.
with reduced survival in patients with advanced hepa-
Measuring inconsistency in meta-analyses. BMJ.
tocellular carcinoma undergoing sorafenib treatment.
2003;327:557-60.
United European Gastroenterol J. 2018b;6:1039-48.
Hiraoka A, Hirooka M, Koizumi Y, Izumoto H, Ueki
H, Kaneto M, et al. Muscle volume loss as a prognostic
marker in hepatocellular carcinoma patients treated
with sorafenib. Hepatol Res. 2017;47:558-65.
14EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Hoshino T, Naganuma A, Suzuki Y, Uehara D, Miyo- Mir O, Coriat R, Blanchet B, Durand JP, Boudou-Rou-
shi T, Sato K, et al. Skeletal muscle depletion as a poor quette P, Michels J, et al. Sarcopenia predicts early
prognostic factor in the treatment with sorafenib for dose-limiting toxicities and pharmacokinetics of soraf-
male patients with advanced hepatocellular carcinoma: enib in patients with hepatocellular carcinoma. PloS
A retrospective study. Hepatology. 2015;62:449A- One. 2012;7:e37563.
50A.
Naganuma A, Hoshino T, Suzuki Y, Uehara D, Kudo
Imai K, Takai K, Hanai T, Ideta T, Miyazaki T, Kochi T, Ishihara H, et al. Association between skeletal mus-
T, et al. Skeletal muscle depletion predicts the progno- cle depletion and sorafenib treatment in male patients
sis of patients with hepatocellular carcinoma treated with hepatocellular carcinoma: A retrospective cohort
with sorafenib. Int J Mol Sci. 2015;16:9612-24. study. Acta Med Okayama. 2017;71:291-9.
Imai K, Takai K, Miwa T, Taguchi D, Hanai T, Nault JC, Pigneur F, Costentin CE, Tselikas L, Luciani
Suetsugu A, et al. Rapid depletions of subcutaneous fat A, Laurent A, et al. Prognostic value of anthropometric
mass and skeletal muscle mass predict worse survival features assessed by CT-scan in patients with advanced
in patients with hepatocellular carcinoma treated with hepatocellular carcinoma treated by tyrosine kinase in-
sorafenib. Cancers (Basel). 2019;11(8):1206. hibitors. Hepatology. 2013;58:1262A.
Imai K, Takai K, Miwa T, Taguchi D, Hanai T, Nault JC, Pigneur F, Nelson AC, Costentin C, Tselikas
Suetsugu A, et al. Rapid depletion of subcutaneous ad- L, Katsahian S, et al. Visceral fat area predicts survival
ipose tissue during sorafenib treatment predicts poor in patients with advanced hepatocellular carcinoma
survival in patients with hepatocellular carcinoma. treated with tyrosine kinase inhibitors. Dig Liver Dis.
Cancers (Basel). 2020;12(7):1795. 2015;47:869-76.
Kao JT, Feng CL, Yu CJ, Tsai SM, Hsu PN, Chen YL, Nishikawa H, Shiraki M, Hiramatsu A, Moriya K,
et al. IL-6, through p-STAT3 rather than p-STAT1, ac- Hino K, Nishiguchi S. Japan Society of Hepatology
tivates hepatocarcinogenesis and affects survival of guidelines for sarcopenia in liver disease (1st edition):
hepatocellular carcinoma patients: A cohort study. Recommendation from the working group for creation
BMC Gastroenterol. 2015;15:50. of sarcopenia assessment criteria. Hepatol Res. 2016;
46:951-63.
Keating GM. Sorafenib: A review in hepatocellular
carcinoma. Targeted Oncol. 2017;12:243-53. Nishikawa H, Nishijima N, Enomoto H, Sakamoto A,
Nasu A, Komekado H, et al. Prognostic significance of
Kudo M, Finn RS, Qin S, Han KH, Ikeda K, Piscaglia sarcopenia in patients with hepatocellular carcinoma
F, et al. Lenvatinib versus sorafenib in first-line treat- undergoing sorafenib therapy. Oncol Lett. 2017;14:
ment of patients with unresectable hepatocellular car- 1637-47.
cinoma: A randomised phase 3 non-inferiority trial.
Lancet. 2018;391:1163-73. Okada M, Nakanishi H, Kurosaki M, Kirino S, Osawa
L, Watakabe K, et al. Myopenia as a significant prog-
Labeur T, Van Vugt J, Cate DT, Takkenberg B, Koer- nostic factor in BCLC-B intermediate-stage hepatocel-
kamp BG, Man RD, et al. Sarcopenia is not a predictor lular carcinoma treated with sorafenib. J Clin Oncol.
of survival or sorafenib toxicity in advanced hepatocel- 2019;37(15, Suppl):e15639.
lular carcinoma: A Dutch multicenter study. J Hepatol.
2018a;68:S434. Okada M, Nakanishi H, Kurosaki M, Inada K, Kirino
S, Yamashita K, et al. The impact of skeletal muscle
Labeur T, van Vugt J, ten Cate D, Takkenberg B, Groot loss for hepatocellular carcinoma treated with len-
Koerkamp B, de Man R, et al. Body composition is as- vatinib. J Clin Oncol. 2020;38(4, Suppl):493.
sociated with poor overall survival and accelerated
time to progression in HCC patients treated with soraf- Saeki I, Yamasaki T, Maeda M, Kawano R, Hisanaga
enib. HPB. 2018b;20:S263. T, Iwamoto T, et al. No muscle depletion with high vis-
ceral fat as a novel beneficial biomarker of sorafenib
Labeur TA, Van Vugt JLA, Ten Cate DWG, Takken- for hepatocellular carcinoma. Liver Cancer. 2018;7:
berg RB, Ijzermans JNM, Groot Koerkamp B, et al. 359-71.
Body composition is an independent predictor of out-
come in patients with hepatocellular carcinoma treated Saeki I, Yamasaki T, Maeda M, Hisanaga T, Iwamoto
with sorafenib. Liver Cancer. 2019;8:255-70. T, Matsumoto T, et al. Effect of body composition on
survival benefit of hepatic arterial infusion chemother-
apy for advanced hepatocellular carcinoma: A compar-
ison with sorafenib therapy. PloS One. 2019;14:
e0218136.
15EXCLI Journal 2021;20:1-16 – ISSN 1611-2156
Received: November 02, 2020, accepted: December 14, 2020, published: January 04, 2021
Sawada K, Saitho Y, Hayashi H, Hasebe T, Nakajima Uojima H, Chuma M, Tanaka Y, Hidaka H, Nakazawa
S, Ikuta K, et al. Skeletal muscle mass is associated T, Iwabuchi S, et al. Skeletal muscle mass influences
with toxicity, treatment tolerability, and additional or tolerability and prognosis in hepatocellular carcinoma
subsequent therapies in patients with hepatocellular patients treated with lenvatinib. Liver Cancer. 2020;
carcinoma receiving sorafenib treatment. JGH Open. 9:193-206.
2019;3:329-37.
Wilhelm SM, Carter C, Tang L, Wilkie D, McNabola
Shachar SS, Williams GR, Muss HB, Nishijima TF. A, Rong H, et al. BAY 43-9006 exhibits broad spec-
Prognostic value of sarcopenia in adults with solid tu- trum oral antitumor activity and targets the
mours: A meta-analysis and systematic review. Eur J RAF/MEK/ERK pathway and receptor tyrosine ki-
Cancer. 2016;57:58-67. nases involved in tumor progression and angiogenesis.
Cancer Res. 2004;64:7099-109.
Takada H, Kurosaki M, Nakanishi H, Takahashi Y,
Itakura J, Tsuchiya K, et al. Impact of pre-sarcopenia Wu CH, Liang PC, Hsu CH, Chang FT, Shao YY,
in sorafenib treatment for advanced hepatocellular car- Ting-Fang Shih T. Total skeletal, psoas and rectus ab-
cinoma. PloS One. 2018;13:e0198812. dominis muscle mass as prognostic factors for patients
with advanced hepatocellular carcinoma. J Formos
Uchikawa S, Kawaoka T, Namba M, Kodama K, Ohya Med Assoc. 2021;120:559-66.
K, Morio K, et al. Skeletal muscle loss during tyrosine
kinase inhibitor treatment for advanced hepatocellular Wu Y, Zhou BP. TNF-alpha/NF-kappaB/Snail path-
carcinoma patients. Liver Cancer. 2020;9:148-55. way in cancer cell migration and invasion. Br J Cancer.
2010;102:639-44.
Ueki H, Hiraoka A, Kawasaki H, Ninomiya T, Hirooka
M, Koizumi Y, et al. Muscle wasting associated with Yamashima M, Miyaaki H, Honda T, Shibata H, Mi-
poor outcome in patients with hepatocellular carci- uma S, Taura N, et al. Significance of psoas muscle
noma undergoing sorafenib treatment. Gastroenterol- thickness as an indicator of muscle atrophy in patients
ogy. 2016;150:S514. with hepatocellular carcinoma treated with sorafenib.
Mol Clin Oncol. 2017;7:449-53.
16You can also read
