Advertisement

Nutritional status in patients with hepatocellular carcinoma: Potential relevance for clinical outcome

Open AccessPublished:July 26, 2022DOI:https://doi.org/10.1016/j.ejim.2022.07.002

      Highlights

      • Impaired nutritional status frequently occurs in hepatocellular carcinoma.
      • Baseline hand-grip strength is associated with complication-free survival.
      • Deteriorated liver function 3 months post-treatment is associated with overall and complication-free survival.

      Abstract

      Background

      Impaired nutritional status is a risk factor for unfavorable outcome in cirrhosis.

      Methods

      In this prospective cohort study in hepatocellular carcinoma patients referred for tumor-specific therapy, nutritional status was assessed before and 3 months post-treatment using 4 complementary tools: hand-grip strength (HGS), Liver Frailty Index (LFI), Patient-Generated Subjective Global Assessment (PG-SGA) and skeletal muscle index (L3-SMI). Uni- and multivariable analyses were performed using Kaplan Meier curves and Cox's regression analyses with correction for Barcelona Clinic Liver Cancer (BCLC) stage, alpha-fetoprotein and age.

      Results

      56 patients were evaluated at baseline and 38 patients 3 months post-treatment. Baseline BCLC stage was 0 in 14%, A in 27%, B in 36%, C in 21%, and D in 2%. HGS, LFI, PG-SGA and L3-SMI were impaired in 13%, 95%, 21% and 71% respectively. Of all patients, 52% died after (median, range) 373 (32–962) days. Of the nutritional assessment tools, only HGS was independently associated with complication-free survival (HR 0.304, 95%CI 0.10–0.88: p = 0.028) and, approaching significance, with overall survival (HR 0.323, 95%CI 0.103–1.008: p = 0.052). Tumor-specific therapy was administered in 50 patients (20% radiofrequency / microwave ablation, 4% resection, 74% transarterial radio- or chemoembolization, 2% sorafenib). Three months post-treatment, complete response occurred in 44%, partial response in 20%, stable disease in 20% and progressive disease in 16%. Child-Pugh scores deteriorated and such deterioration was independently associated with reduced overall and complication-free survival.

      Conclusions

      reduced baseline HGS and deteriorated post-treatment Child-Pugh score are associated with reduced overall and complication-free survival in HCC.

      Keywords

      AFP
      alpha-fetoprotein
      BCLC
      Barcelona Clinic Liver Cancer
      CFS
      complication-free survival
      CP
      Child-Pugh
      CT
      computed tomography
      EORTC-QLQ
      European Organization for Research and Treatment for Cancer Quality of Life Questionnaire
      HCC
      hepatocellular carcinoma
      HGS
      hand-grip strength
      L3-SMI
      skeletal muscle index at 3rd lumbar vertebra
      LFI
      Liver Frailty Index
      OS
      overall survival
      PFS
      progression-free survival
      PG-SGA
      patient-generated subjective global assessment

      1. Introduction

      Hepatocellular carcinoma (HCC) is the third most common cause of cancer-related mortality worldwide [
      • Sung H.
      • Ferlay J.
      • Siegel R.L.
      • Laversanne M.
      • Soerjomataram I.
      • Jemal A.
      • et al.
      Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
      ]. Impaired nutritional status is a frequent phenomenon in patients with cirrhosis with clear negative impact on clinical course [
      • Bunchorntavakul C.
      • Reddy K.R.
      Review article: malnutrition/sarcopenia and frailty in patients with cirrhosis.
      ,
      • Lai J.C.
      • Covinsky K.E.
      • McCulloch C.E.
      • Feng S.
      The liver frailty index improves mortality prediction of the subjective clinician assessment in patients with cirrhosis.
      ,
      • Wang S.
      • Whitlock R.
      • Xu C.
      • Taneja S.
      • Singh S.
      • Abraldes J.G.
      • et al.
      Frailty is associated with increased risk of cirrhosis disease progression and death.
      ,
      • Sieber C.C.
      Malnutrition and sarcopenia.
      ,
      • Huisman E.J.
      • Trip E.J.
      • Siersema P.D.
      • Van Hoek B.
      • Van Erpecum K.J.
      Protein energy malnutrition predicts complications in liver cirrhosis.
      ,
      • Montano-Loza A.J.
      Clinical relevance of sarcopenia in patients with cirrhosis.
      ]. Impaired nutritional status also frequently occurs in cirrhotic HCC patients. Despite some negative studies [
      • Kroh A.
      • Uschner D.
      • Lodewick T.
      • Eickhoff R.M.
      • Schöning W.
      • Ulmer F.T.
      • et al.
      Impact of body composition on survival and morbidity after liver resection in hepatocellular carcinoma patients.
      ,
      • Akce M.
      • Liu Y.
      • Zakka K.
      • Martini D.J.
      • Draper A.
      • Alese O.B.
      • et al.
      Impact of sarcopenia, BMI, and inflammatory biomarkers on survival in advanced hepatocellular carcinoma treated with anti-PD-1 antibody.
      ], most available data suggest that impaired nutritional status is also associated with unfavorable outcome in patients with HCC in general [
      • Marasco G.
      • Serenari M.
      • Renzulli M.
      • Alemanni L.V.
      • Rossini B.
      • Pettinari I.
      • et al.
      Clinical impact of sarcopenia assessment in patients with hepatocellular carcinoma undergoing treatments.
      ,
      • Meza-Junco J.
      • Montano-Loza A.J.
      • Baracos V.E.
      • Prado C.M.M.
      • Bain V.G.
      • Beaumont C.
      • et al.
      Sarcopenia as a prognostic index of nutritional status in concurrent cirrhosis and hepatocellular carcinoma.
      ,
      • Fujiwara N.
      • Nakagawa H.
      • Kudo Y.
      • Tateishi R.
      • Taguri M.
      • Watadani T.
      • et al.
      Sarcopenia, intramuscular fat deposition, and visceral adiposity independently predict the outcomes of hepatocellular carcinoma.
      ,
      • Chang K.V.
      • De Chen J
      • Wu W.T.
      • Huang K.C.
      • Hsu C.T.
      • Han D.S
      Association between loss of skeletal muscle mass and mortality and tumor recurrence in hepatocellular carcinoma: a systematic review and meta-analysis.
      ,
      • Begini P.
      • Gigante E.
      • Antonelli G.
      • Carbonetti F.
      • Iannicelli E.
      • Anania G.
      • et al.
      Sarcopenia predicts reduced survival in patients with hepatocellular carcinoma at first diagnosis.
      ,
      • Badran H.
      • Elsabaawy M.M.
      • Ragab A.
      • Aly R.A.
      • Alsebaey A.
      • Sabry A.
      Baseline sarcopenia is associated with lack of response to therapy, liver decompensation and high mortality in hepatocellular carcinoma patients.
      ,
      • Iritani S.
      • Imai K.
      • Takai K.
      • Hanai T.
      • Ideta T.
      • Miyazaki T.
      • et al.
      Skeletal muscle depletion is an independent prognostic factor for hepatocellular carcinoma.
      ,
      • Zhang G.
      • Meng S.
      • Li R.
      • Ye J.
      • Zhao L.
      Clinical significance of sarcopenia in the treatment of patients with primary hepatic malignancies, a systematic review and meta-analysis.
      ,
      • Schütte K.
      • Tippelt B.
      • Schulz C.
      • Röhl F.W.
      • Feneberg A.
      • Seidensticker R.
      • et al.
      Malnutrition is a prognostic factor in patients with hepatocellular carcinoma (HCC).
      ] as well as after tumor-specific treatment, such as liver resection [
      • Voron T.
      • Tselikas L.
      • Pietrasz D.
      • Pigneur F.
      • Laurent A.
      • Compagnon P.
      • et al.
      Sarcopenia impacts on short- and long-term results of hepatectomy for hepatocellular carcinoma.
      ,
      • Harimoto N.
      • Shirabe K.
      • Yamashita Y.I.
      • Ikegami T.
      • Yoshizumi T.
      • Soejima Y.
      • et al.
      Sarcopenia as a predictor of prognosis in patients following hepatectomy for hepatocellular carcinoma.
      ,
      • Takagi K.
      • Yagi T.
      • Yoshida R.
      • Shinoura S.
      • Umeda Y.
      • Nobuoka D.
      • et al.
      Sarcopenia and American society of anesthesiologists physical status in the assessment of outcomes of hepatocellular carcinoma patients undergoing hepatectomy.
      ,
      • Xu L.
      • Jing Y.
      • Zhao C.
      • Zhang Q.
      • Zhao X.
      • Yang J.
      • et al.
      Preoperative computed tomography-assessed skeletal muscle index is a novel prognostic factor in patients with hepatocellular carcinoma following hepatectomy: a meta-analysis.
      ], radiofrequency ablation [
      • Levolger S.
      • Van Vledder M.G.
      • Muslem R.
      • Koek M.
      • Niessen W.J.
      • De Man R.A.
      • et al.
      Sarcopenia impairs survival in patients with potentially curable hepatocellular carcinoma.
      ,
      • Salman A.
      • Salman M.
      • Moustafa A.
      • Shaaban H.E.-.D.
      • El-Mikkawy A.
      • Labib S.
      • et al.
      Impact of sarcopenia on two-year mortality in patients with HCV-associated hepatocellular carcinoma after radiofrequency ablation.
      ], sorafenib [
      • Antonelli G.
      • Gigante E.
      • Iavarone M.
      • Begini P.
      • Sangiovanni A.
      • Iannicelli E.
      • et al.
      Sarcopenia is associated with reduced survival in patients with advanced hepatocellular carcinoma undergoing sorafenib treatment.
      ,
      • Nishikawa H.
      • Nishijima N.
      • Enomoto H.
      • Sakamoto A.
      • Nasu A.
      • Komekado H.
      • et al.
      Prognostic significance of sarcopenia in patients with hepatocellular carcinoma undergoing sorafenib therapy.
      ], Lenvatinib [
      • Endo K.
      • Kuroda H.
      • Kanazawa J.
      • Sato T.
      • Fujiwara Y.
      • Abe T.
      • et al.
      Impact of grip strength in patients with unresectable hepatocellular carcinoma treated with lenvatinib.
      ], embolization [
      • Lanza E.
      • Masetti C.
      • Messana G.
      • Muglia R.
      • Pugliese N.
      • Ceriani R.
      • et al.
      Sarcopenia as a predictor of survival in patients undergoing bland transarterial embolization for unresectable hepatocellular carcinoma.
      ] or radiotherapy [
      • Lee J.
      • Cho Y.
      • Park S.
      • Kim J.W.
      • Lee I.J.
      Skeletal muscle depletion predicts the prognosis of patients with hepatocellular carcinoma treated with radiotherapy.
      ]. Many new therapeutic modalities have recently been introduced for HCC, but prognosis remains poor. Potential beneficial effects of early dietary interventions in these patient categories should be further explored if impaired nutritional status could be characterized in etiologic modeling studies as a risk factor causally related to unfavorable outcome.
      Examples of nutrition disorders and nutrition related conditions [
      • Lai J.C.
      • Tandon P.
      • Bernal W.
      • Tapper E.B.
      • Ekong U.
      • Dasarathy S.
      • et al.
      Malnutrition, frailty, and sarcopenia in patients with cirrhosis: 2021 practice guidance by the American association for the study of liver diseases.
      ] are sarcopenia (low muscle strength, low muscle quantity/quality and/or low physical performance) [
      • Cruz-Jentoft A.J.
      • Bahat G.
      • Bauer J.
      • Boirie Y.
      • Bruyère O.
      • Cederholm T.
      • et al.
      Sarcopenia: revised European consensus on definition and diagnosis.
      ], frailty (impaired muscle contractile function that causes increased vulnerability and decreased physiologic reserve) [
      • Lai J.C.
      • Tandon P.
      • Bernal W.
      • Tapper E.B.
      • Ekong U.
      • Dasarathy S.
      • et al.
      Malnutrition, frailty, and sarcopenia in patients with cirrhosis: 2021 practice guidance by the American association for the study of liver diseases.
      ] and malnutrition (imbalance of nutrient intake that causes adverse effects on patients’ tissue, body form or function) [
      • Lai J.C.
      • Tandon P.
      • Bernal W.
      • Tapper E.B.
      • Ekong U.
      • Dasarathy S.
      • et al.
      Malnutrition, frailty, and sarcopenia in patients with cirrhosis: 2021 practice guidance by the American association for the study of liver diseases.
      ]. Hand-grip strength (HGS) measured with a dynamometer reflects muscle strength [
      • Huisman E.J.
      • Trip E.J.
      • Siersema P.D.
      • Van Hoek B.
      • Van Erpecum K.J.
      Protein energy malnutrition predicts complications in liver cirrhosis.
      ]. CT-scan can be used to assess skeletal muscle mass [
      • Cruz-Jentoft A.J.
      • Bahat G.
      • Bauer J.
      • Boirie Y.
      • Bruyère O.
      • Cederholm T.
      • et al.
      Sarcopenia: revised European consensus on definition and diagnosis.
      ]. Frailty can be assessed with the Liver Frailty Index (LFI) [
      • Lai J.C.
      • Covinsky K.E.
      • Dodge J.L.
      • Boscardin W.J.
      • Segev D.L.
      • Roberts J.P.
      • et al.
      Development of a novel frailty index to predict mortality in patients with end-stage liver disease.
      ] and malnutrition with the Patient-Generated Subjective Global Assessment (PG-SGA) [
      • Jager-Wittenaar H.
      • Ottery F.D.
      Assessing nutritional status in cancer: role of the patient-generated subjective global assessment.
      ].
      In most previous studies in HCC patients, skeletal muscle mass was assessed by CT scan. Nevertheless, MRI-scan rather than CT-scan currently is the preferred radiological option to diagnose HCC. Also, CT-scan has a small risk of side effects, is expensive and evaluation of sarcopenia risk by CT-scan requires additional software. Therefore, other easy-to-use tools are necessary to detect nutritional risk in HCC patients.
      The primary aim of this study was to explore the prevalence of reduced hand-grip strength, frailty, malnutrition and reduced skeletal muscle mass as well as their associations with overall survival (OS), complication-free survival (CFS) and progression-free survival (PFS) in HCC patients.

      2. Patients and methods

      2.1 Patients

      In this prospective cohort study, all consenting patients with HCC who were referred to our tertiary center for tumor-specific therapy in the period September 2018 – April 2021 were included. The local Medical Ethical Committee had no objections to the study (research protocol 18/337) and all included patients provided written and signed informed consent. HCC was diagnosed according to European Association for the Study of the Liver Clinical Practice Guidelines [
      • Galle P.R.
      • Forner A.
      • Llovet J.M.
      • Mazzaferro V.
      • Piscaglia F.
      • Raoul J.L.
      • et al.
      EASL clinical practice guidelines: management of hepatocellular carcinoma.
      ]. Baseline evaluation comprised medical history including alcohol consumption (≥3 daily consumptions in men and ≥2 daily consumptions in women were considered alcohol abuse), physical examination, laboratory tests, and magnetic resonance imaging (MRI-scan) of the liver and/or 3 phase CT-scan of the abdomen unless performed within one month from baseline. Body-mass index (BMI), World Health Organization Performance Status (PS), Child-Pugh (CP) score, Model for End-Stage Liver Disease (MELD) score, MELD-Na score and Barcelona Clinic Liver Cancer (BCLC) stage [
      • Galle P.R.
      • Forner A.
      • Llovet J.M.
      • Mazzaferro V.
      • Piscaglia F.
      • Raoul J.L.
      • et al.
      EASL clinical practice guidelines: management of hepatocellular carcinoma.
      ] were determined. Portal hypertension was defined as presence of either collaterals on radiological examination, esophageal varices by upper gastrointestinal endoscopy and/or thrombocytopenia. Quality of life was scored by the general European Organization for Research and Treatment for Cancer Quality of Life Questionnaire Core 30 (EORTC-QLQ C30) [
      • Aaronson N.K.
      • Ahmedzai S.
      • Bergman B.
      • Bullinger M.
      • Cull A.
      • Duez N.J.
      • et al.
      The European organization for research and treatment of cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
      ] and the liver cancer specific instrument (EORTC-QLQ HCC18) [
      • Blazeby J.M.
      • Currie E.
      • Zee B.C.Y.
      • Chie W.-.C.
      • Poon R.T.
      • Garden O.J.
      Development of a questionnaire module to supplement the EORTC QLQ-C30 to assess quality of life in patients with hepatocellular carcinoma, the EORTC QLQ-HCC18.
      ], in which global health status of EORTC-QLQ C30 and the summary score of EORTC-QLQ C30 and HCC18 [
      • Li L.
      • Mo F.K.F.
      • Chan S.L.
      • Hui E.P.
      • Tang N.S.L.
      • Koh J.
      • et al.
      Prognostic values of EORTC QLQ-C30 and QLQ-HCC18 index-scores in patients with hepatocellular carcinoma - clinical application of health-related quality-of-life data.
      ] were used. Follow up occurred at least every three months, including laboratory and radiological examinations (generally MRI-scan).

      2.2 Parameters of nutritional status

      Baseline nutrition disorders and nutrition-related conditions were assessed in all consenting patients and repeated in all available cases three months post-treatment. Skeletal muscle mass was measured with the aid of Slice-O-Matic program as cross-sectional muscle area (SMA) at the third lumbar vertebra (L3) on CT. Muscle area was defined as the pixel area between the radiodensity range of −29 and + 150 Hounsfield Units (HU) which is specific for muscle tissue. The skeletal muscle index (SMI) was calculated by correcting SMA for height and expressed in cm2/m2. L3-SMI <39 cm2/m2 for women and <50 cm2/m2 for men were considered reduced SMI [
      • Carey E.J.
      • Lai J.C.
      • Wang C.W.
      • Dasarathy S.
      • Lobach I.
      • Montano-Loza A.J.
      • et al.
      A multicenter study to define sarcopenia in patients with end-stage liver disease.
      ].
      Hand-grip strength (HGS) was determined by squeezing an analog Jamar dynamometer three times with dominant hand and full-strength, having a neutral shoulder and forearm with the elbow at 90° The highest of three attempts below the 10th centile was taken to distinguish between reduced and normal HGS on age-bound, sex-specific cutoff values [
      • Dodds R.M.
      • Syddall H.E.
      • Cooper R.
      • Benzeval M.
      • Deary I.J.
      • Dennison E.M.
      • et al.
      Grip strength across the life course: normative data from twelve British studies.
      ].
      Frailty was determined with the Liver Frailty Index (LFI) [
      • Lai J.C.
      • Covinsky K.E.
      • Dodge J.L.
      • Boscardin W.J.
      • Segev D.L.
      • Roberts J.P.
      • et al.
      Development of a novel frailty index to predict mortality in patients with end-stage liver disease.
      ]. Components of LFI are: 1. HGS, 2. five chair stands as fast as possible (one decimal accuracy and with a maximum of 60 s) and 3. Balance testing in three different positions (side-by-side, semi-tandem, tandem: with a maximum of 10 s each). The LFI calculatorhttps://liverfrailtyindex.ucsf.edu/ was used to determine whether the patient was either robust, prefrail or frail.
      Malnutrition was evaluated with the Patient-Generated Subjective Global Assessment (PG-SGA: version 3.7 NL, 2014) [
      • Jager-Wittenaar H.
      • Ottery F.D.
      Assessing nutritional status in cancer: role of the patient-generated subjective global assessment.
      ]. The PG-SGA consists of two components: the first part, regarding weight, nutritional intake, symptoms and functioning, was generated by the patient. The second part, about the patients’ medical history, metabolic stress and physical examination, was completed by the healthcare professional. Patients were divided in three groups (A = well nourished, B = moderately malnourished or suspected malnutrition, C = severely malnourished).

      2.3 Statistical analysis

      IBM SPSS Statistics 26.0.0.1 was used to perform all statistical analyses. Categorical data are expressed as absolute numbers with percentage and compared with Fisher's exact test. Continuous variables are presented as mean ± SD (range) in case of normal distribution, or as median (range) in case of non-normal distribution. Independent samples t-tests or Mann-Whitney U tests were applied for comparison between two groups as appropriate. One-way ANOVA (with Tukey's HSD as post-hoc test) or Kruskal-Wallis tests (in case of non-normal distribution) were used for comparison of three groups. Differences in patient characteristics between baseline and three months post-treatment were evaluated the paired samples t-test or Wilcoxon signed-rank test as appropriate. A two-sided p-value < 0.05 was considered statistically significant.
      Our primary aims were: 1. to explore in HCC patients referred for tumor-specific treatment, the prevalence of reduced hand-grip strength, frailty, malnutrition and reduced skeletal muscle mass and 2. to examine their associations with overall (OS), complication-free (CFS) and tumor-progression free (PFS) survival. Secondary aims were: 1. To compare nutritional status at baseline and at 3-months posttreatment and 2. to evaluate whether at 3-months follow up, deteriorated nutritional status, deteriorated liver function (Child-Pugh score) or worse tumor response (based on modified response evaluation criteria in solid tumors (mRECIST [
      • Lencioni R.
      • Llovet J.M.
      Modified RECIST (mRECIST) assessment for hepatocellular carcinoma.
      ]) were associated with reduced OS, CFS or PFS.
      Follow-up ended at the final date of evaluation (1st of May 2021) or earlier in case of death from any cause (for OS, CFS), or first complication (for CFS), whatever came first. The following events were considered as complications: ascites, (bacterial) infection, variceal bleeding, hepatic encephalopathy, (severe) icterus, stroke, heart attack or death. PFS was evaluated for those patients who received tumor-targeting treatment and started at the date of measurement at baseline and ended at the date of first diagnosed recurrence or progression of the tumor or the first newly diagnosed metastasis after treatment, or at time of death from any cause whatever came first.
      For the statistical analyses regarding our primary aims, patients were divided in subgoups, based on L3-SMI, HGS, frailty (LFI) or malnourishment (PG-SGA). Patients were divided in two groups for each parameter: reduced L3-SMI versus normal L3-SMI, reduced versus normal HGS, non-frail (i.e. robust and prefrail combined) versus frail for LFI and well-nourished (grade A) versus malnourished (grade B or C) for PG-SGA respectively. Then, the relation of the nutritional parameters (HGS, LFI and PG-SGA, L3-SMI) at baseline with OS, CFS and PFS was explored, using the Kaplan Meier method and univariable Cox's regression analyses. For our secondary aims a similar approach was used.
      We wanted to explore whether our variables of interest could be a risk factor causally related to poor outcome, in light of potential therapeutic consequences. Therefore, according to the rules for etiological modeling studies [
      • van Diepen M.
      • Ramspek C.L.
      • Jager K.J.
      • Zoccali C.
      • Dekker F.W.
      Prediction versus aetiology: common pitfalls and how to avoid them.
      ,
      • Ramspek C.L.
      • Steyerberg E.W.
      • Riley R.D.
      • Rosendaal F.R.
      • Dekkers O.M.
      • Dekker F.W.
      • et al.
      Prediction or causality? A scoping review of their conflation within current observational research.
      ], in the subsequent multivariable analyses, our variables of interest were corrected for the pre-defined variables BCLC stage, AFP and age based on the extensive pre-existing literature of the considerable impact of these three confounders on outcome in hepatocellular carcinoma.

      3. Results

      3.1 Baseline and tumor characteristics

      Of all 86 patients with HCC initially referred for tumor-specific therapy in the study period, 56 patients could have a complete baseline assessment (flowchart Supplementary Fig. 1).
      There were no significant differences in patient or tumor characteristics between these 56 patients and all 86 referred patients. Baseline characteristics of the included patients are given in Table 1.
      Table 1Baseline patient and tumor characteristics.
      Baseline tableTotal (n = 56)
      Age (years)70 (43 – 86)
      Male gender50 (89)
      Etiology

       HBV

       HCV

       NAFLD/NASH

       Alcohol

       Hemochromatosis

       Other

       Unknown


      2 (4)

      7 (12)

      17 (30)

      20 (36)

      1 (2)

      2 (4)

      7 (12)
      Cirrhotic
      Based on clinical, radiologic or histologic data.
      48 (86)
      Fibroscan
      Fibroscan was performed in 16/56 patients (29%).


       F0–2

       F3–4


      2 (13)

      14 (87)
      First treatment modality
      Four patients received additional second treatment more than three months after the primary treatment.


       Resection

       RFA/MWA

       TACE

       TARE

       Sorafenib

       Best supportive care


      2 (4)

      10 (18)

      12 (21)

      25 (45)

      1 (2)

      6 (10)
      Creatinine (µmol/L)81 ± 24 (32 – 138)
      Total bilirubin (µmol/L)16 (4 – 98)
      Alkaline phosphatase (U/L)120 (46 – 416)
      Gamma-GT (U/L)156 (35 – 1080)
      ASAT (U/L)53 (16 – 590)
      ALAT (U/L)38 (11 – 228)
      Albumin (g/L)37 ± 5 (24 – 50)
      Thrombocytes (x109/L)135 (40 – 694)
      PT-INR1.22 (1.00 – 2.08)
      Sodium (mmol/L)138 ± 3 (130 – 144)
      Alpha-fetoprotein (mcg/L)
      available in 54/56 patients (96%).
      11 (2 – 170,000)
      Portal hypertension

       Varices

       Collaterals

       Thrombocytopenia
      41 (73)

      26 (46)

      33 (59)

      31 (55)
      Ascites

       Absent

       Slight

       Moderate


      43 (77)

      10 (18)

      3 (5)
      Child-Pugh score5 (5 – 10)
      Child-Pugh class

       A (5–6)

       B (7–9)

       C (=/>10)


      46 (82)

      9 (16)

      1 (2)
      MELD-score10 (6 – 24)
      MELD-Na score12 (7 – 24)
      BCLC stage

       0

       A

       B

       C

       D


      8 (14)

      15 (27)

      20 (36)

      12 (21)

      1 (2)
      Performance score (ECOG)

       0

       1

       2


      37 (66)

      15 (27)

      4 (7)
      BMI (kg/m2)27 (20 – 46)
      Weight (kg)88 ± 17 (50 – 136)
      ∆Weight in preceding month (kg)0 (−33 – +9)
      ∆Weight in preceding 6 months (kg)−1 (−17 – +8)
      PG-SGA total score4 (1 – 14)
      PG-SGA

       A well-nourished

       B moderately malnourished


      44 (79)

      12 (21)
      Liver frailty index score3.99 ± 0.68 (2.46 – 6.33)
      LFI diagnosis

       Robust

       Prefrail

       Frail


      3 (5)

      45 (81)

      8 (14)
      Handgrip strength

       Highest (kg)

      Reduced


      34 ± 9 (16 – 52)

      7 (13)
      Chairtest
      6/56 patients (11%) were not able to perform or complete the chair test in less than 60 s and could not be included.


       Time (s)

      Impaired


      13 (7 – 29)

      9 (18)
      CT-scan L3
      CT-scan was available in 42/56 patients (75%).


       Skeletal muscle index (cm2/m2)

       reduced skeletal muscle mass


      46.4 ± 6.7 (31.1–65.0)

      30 (71)
      EORTC-QLQ C30 Global health status75 (33 – 100)
      Summary score QoL (C30)90 (49 – 100)
      Summary score QoL (HCC18)9 (0 – 48)
      Duration of follow-up (days)373 (32 – 962)
      Data are presented as n (%), in case of parametric distribution as mean ± SD (range) or in case of nonparametric distribution as median (range).
      a Based on clinical, radiologic or histologic data.
      b Fibroscan was performed in 16/56 patients (29%).
      c Four patients received additional second treatment more than three months after the primary treatment.
      d available in 54/56 patients (96%).
      e 6/56 patients (11%) were not able to perform or complete the chair test in less than 60 s and could not be included.
      f CT-scan was available in 42/56 patients (75%).
      Median age was 70 years and 89% of the patients were male. Cirrhosis was present in 86% of patients. Most frequent underlying causes of liver disease were alcohol abuse (36%), non-alcoholic steatohepatitis (30%) and hepatitis C (12%). Portal hypertension was present in 73%, as indicated by the presence of varices (46%), collaterals (59%) and/or thrombocytopenia (55%). Baseline BCLC stage was 0 in 14%, A in 27%, B in 36%, C in 21% and D in 2%. Child-Pugh class was A in 82%, B in 16% and C in 2%.

      3.2 Baseline nutritional status

      Detailed baseline nutritional characteristics are given in Supplementary Table 1.
      L3-SMI (available in 42 cases) and HGS were reduced in 71% and 13% respectively. According to LFI, 5% of patients were robust, 81% prefrail and 14% frail. Frail and prefrail patients exhibited worse quality of life, according to EORTC-QLQ C30 Global Health Status and EORTC-QLQ C30 summary scores (Supplementary Table 1). According to PG-SGA, 79% of patients were well nourished (stage A) and 21% were moderately malnourished (stage B) at baseline. Quality of life according to EORTC-QLQ C30 Global Health Status and EORTC-QLQ HCC 18 summary score, was significantly worse in case of malnourished patients. In the 42 cases with all four nutritional parameters available, impaired nutritional status according to 0, 1, 2 or 3 parameters was found in 17%, 52%, 19%, and 12% respectively. Underlying cause of liver disease did not affect the nutritional parameters.

      3.3 Overall survival: association with baseline parameters

      During the study period, 52% of all patients died, after a median follow-up time of 373 days (range 32 – 962 days). OS was worse in case of higher Child-Pugh scores and higher BCLC stages (Supplementary Figs. 2 and 3). Concerning our confounding variables, in uni- and multivariable analyses, BCLC stage and age were independently associated with OS and AFP approached significance (P = 0.07). Concerning our main variables of interest, in univariable Cox-regression analysis, HGS was the only parameter of nutritional status significantly associated with OS (Table 2). Although patients with Child-Pugh class B or C often exhibited low HGS, there was significant overlap with Child-Pugh class A patients (Supplementary Fig. 4).
      Table 2Relation between nutritional status and overall survival.
      Overall survival (n = 56)Univariable analysisMultivariable analysis
      In multivariable analyses the following baseline confounding variables were included: BCLC stage, alpha-fetoprotein (<1000 mcg/L versus ≥1000 mcg/L) and age.
      HR (95% CI)p-valueHR (95% CI)p-value
      L3-SMI
      CT scan was available in 42 patients.
      Reduced (n = 30)ReferentReferent
      Normal (n = 12)1.540 (0.673–3.523)0.3060.441 (0.121–1.610)0.215
      HGS baseline
      reduced (n = 7)ReferentReferent
      normal (n = 49)0.304 (0.121–0.762)0.0110.323 (0.103–1.008)0.052
      LFI stage baseline
      Robust and prefrail combined, since only 3 patients were robust.
      Robust/prefrail (n = 48)ReferentReferent
      Frail (n = 8)0.946 (0.329–2.726)0.9193.017 (0.606–15.015)0.177
      PG-SGA stage baseline
      A (n = 44)ReferentReferent
      B (n = 12)1.416 (0.625–3.206)0.4041.038 (0.377–2.855)0.942
      ΔChild-Pugh score
      Child-Pugh score at 3 months post treatment available in 49 patients.
      0.0030.003
      Deterioration (n = 23)ReferentReferent
      Equal (n = 18)0.233 (0.084–0.643)0.0050.210 (0.067–0.659)0.008
      Improvement (n = 8)0.086 (0.011–0.661)0.0180.053 (0.006–0.455)0.007
      mRECIST at 3 monts
      mRECIST 3 months post-treatment available in 45 patients.
      0.0770.641
      Complete response (n = 20)ReferentReferent
      Partial response (n = 9)1.630 (0.458–5.794)0.451.782 (0.363–8.753)0.477
      Stable disease (n = 9)3.094 (0.912–10.501)0.0700.895 (0.139–5.760)0.907
      Progressive disease (n = 7)4.413 (1.331–14.636)0.0151.922 (0.421–8.784)0.399
      ΔHGS
      HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      Deteriorated (n = 3)ReferentReferent
      Equal (n = 35)1.455 (0.192–11.046)0.7170342 (0.032–3.694)0.377
      ΔLFI stage
      LFI at 3 months post-treatment available in 37 patients.
      Deterioration (n = 6)ReferentReferent
      Equal/improvement (n = 31)0.353 (0.109–1.142)0.0820.598 (0.105–3.417)0.563
      ΔPG-SGA stage
      HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      0.3830.614
      Deterioration (n = 8)ReferentReferent
      Equal (n = 26)0.656 (0.201–2.141)0.4850.554 (0.152–2.011)0.369
      Improvement (n = 4)1.616 (0.357–7.301)0.5330.966 (0.149–6.275)0.971
      a In multivariable analyses the following baseline confounding variables were included: BCLC stage, alpha-fetoprotein (<1000 mcg/L versus ≥1000 mcg/L) and age.
      b CT scan was available in 42 patients.
      c Robust and prefrail combined, since only 3 patients were robust.
      d Child-Pugh score at 3 months post treatment available in 49 patients.
      e mRECIST 3 months post-treatment available in 45 patients.
      f HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      g LFI at 3 months post-treatment available in 37 patients.
      Kaplan Meier analysis for the effect of HGS on OS is shown in Fig. 1.
      Fig. 1
      Fig. 1Kaplan-Meier curve shows reduced overall survival in patients with reduced handgrip strength (log-rank test: p = 0.007).
      In multivariable Cox-regression analysis with correction for the confounders BCLC stage, AFP and age, this association was on the border of significance (HR 0.323 (0.103–1.008: p = 0.052: Table 2).

      3.4 Complication-free survival: association with baseline parameters

      Complications occurred in 32 patients (57%). Median CFS time was 225 (range 7 – 962) days. Complications included infection (31%: 3 pneumonia, 2 erysipelas, 1 pyelonephritis, 1 spontaneous bacterial peritonitis, 1 liver abscess, 1 cholecystitis, 1 urosepsis), ascites (25%), stroke (10%), variceal bleeding (6%), hepatic encephalopathy (6%), and heart attack (3%). CFS was worse in case of higher Child-Pugh scores and higher BCLC stages (supplementary Figs. 5 and 6). Concerning our confounding variables, age was independently associated with CFS in uni- and multivariable analyses, while BCLC stage (p = 0.083) and AFP (p = 0.092) approached significance. Concerning our main variables of interest, reduced HGS was the only parameter of nutritional status significantly associated with CFS In univariable Cox regression analysis (Table 3).
      Table 3Relation between nutritional status and complication-free survival.
      Complication-free survival (n = 56)Univariable analysisMultivariable analysis
      In multivariable analyses the following confounding variables were included: baseline BCLC stage, alpha-fetoprotein (<1000 mcg/L versus ≥1000 mcg/L) and age.
      HR (95% CI)p-valueHR (95% CI)p-value
      L3-SMI baseline
      CT scan available in 42 patients.
      reduced (n = 30)ReferentReferent
      normal (n = 12)1.870 (0.843–4.152)0.1241.099 (0.353–3.426)0.870
      HGS baseline
      reduced (n = 7)ReferentReferent
      normal (n = 49)0.286 (0.111–0.738)0.0100.304 (0.105–0.881)0.028
      LFI stage baseline
      Robust and prefrail combined, since only 3 patients were robust.
      Robust/prefrail (n = 48)ReferentReferent
      Frail (n = 8)0.777 (0.272–2.217)0.6371.288 (0.307–5.401)0.729
      PG-SGA stage baseline
      A (n = 44)ReferentReferent
      B (n = 12)1.629 (0.751–3.536)0.2171.350 (0.547–3.332)0.515
      ΔChild-Pugh score
      Child-Pugh score at 3 months post treatment available in 49 patients.
      <0.0010.002
      Deterioration (n = 23)ReferentReferent
      Equal (n = 18)0.176 (0.064–0.482)0.0010.179 (0.058–0.551)0.003
      Improvement (n = 8)0.142 (0.032–0.627)0.0100.121 (0.024–0.601)0.010
      mRECIST
      mRECIST at 3 months post-treatment available in 45 patients.
      0.5300.956
      Complete response (n = 20)ReferentReferent
      Partial response (n = 9)1.193 (0.366–3.891)0.7691.308 (0.316–5.417)0.711
      Stable disease (n = 9)1.699 (0.559–5.161)0.3501.114 (0.238–5.201)0.891
      Progressive disease (n = 7)2.171 (0.722–6.529)0.1671.415 (0.334–5.991)0.637
      ΔHGS
      HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      Deteriorated (n = 3)ReferentReferent
      Equal (n = 35)2.072 (0.275–15.581)0.4790.890 (0.098–8.075)0.918
      ΔLFI stage
      LFI at baseline and 3 months post treatment available in 37 patients.
      Deterioration (n = 6)ReferentReferent
      Equal/improvement (n = 31)0.115 (0.037–0.359)<0.0010.081 (0.014–0.467)0.005
      ΔPG-SGA stage
      HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      0.5090.440
      Deterioration (n = 8)ReferentReferent
      Equal (n = 26)0.708 (0.227–2.208)0.5520.434 (0.111–1.695)0.230
      Improvement (n = 4)1.451 (0.324–6.498)0.6270.759 (0.123–4.684)0.767
      a In multivariable analyses the following confounding variables were included: baseline BCLC stage, alpha-fetoprotein (<1000 mcg/L versus ≥1000 mcg/L) and age.
      b CT scan available in 42 patients.
      c Robust and prefrail combined, since only 3 patients were robust.
      d Child-Pugh score at 3 months post treatment available in 49 patients.
      e mRECIST at 3 months post-treatment available in 45 patients.
      f HGS and PG-SGA at 3 months post-treatment available in 38 patients.
      g LFI at baseline and 3 months post treatment available in 37 patients.
      The association between HGS and CFS in univariable Kaplan Meier analysis is shown in Fig. 2.
      Fig. 2
      Fig. 2Kaplan-Meier curve shows reduced complication-free survival in patients with reduced handgrip strength (log-rank test: p = 0.006).
      In multivariable analyses with correction for the confounders for BCLC stage, AFP and age, reduced HGS remained significantly associated with CFS (HR = 0.304, 95% CI 0.105–0.881; p = 0.028: Table 3).

      3.5 Progression-free survival: association with baseline parameters

      Of the patients who recieved tumor-specific therapy, tumor progression or death occurred in 32 cases (67%) during follow-up, with a median PFS time of 225 (range 56 –962) days. in uni- and multivariable analyses, BCLC stage and age were independently associated with PFS, while AFP (p = 0.09) was approaching significance. There was no association between any of the parameters of nutritional state at baseline and PFS, in either univariable or multivariable analyses.

      3.6 Repeated measurements three months after tumor-specific therapy

      Of all patients, 10% received best supportive care. The other 50 patients received various initial treatments, both with curative intent (20% radiofrequency ablation/ microwave ablation, 4% resection) and with palliative intent (50% transarterial radioembolization, 24% transarterial chemoembolization, 2% sorafenib). As far as transarterial radioembolization is concerned, 16 patients were treated with holmium-166 in a phase 2 study [
      • Reinders M.T.M.
      • van Erpecum K.J.
      • Smits M.L.J.
      • Braat A.J.A.T.
      • de Bruijne J.
      • Bruijnen R.C.G.
      • et al.
      Safety and efficacy of holmium-166 radioembolization in hepatocellular carcinoma - the HEPAR Primary study.
      ], the remaining patients with Yttrium-90. Four patients received additional treatment after the 3 months post-treatment evaluation: two patients underwent liver transplantation, one patient received additional transarterial chemoembolization, and one patient received additional transarterial radioembolization. Repeated nutritional measurements three months post-treatment were available in 38 of the 50 patients (76%) who received antitumor therapy (flowchart Supplementary Fig. 1). Their patient and tumor characteristics at baseline and at 3 months post-treatment follow up are given in Table 4.
      Table 4Clinical characteristics of 38 hepatocellular carcinoma patients with measurements of nutritional status available at baseline and 3 months after the first tumor treatment.
      All patients

      Baseline (n = 38)3 months after first treatment (n = 38)p-value
      Age (years)72 ± 7 (48 – 86)
      Male gender35 (92)
      Etiology

       HBV

       HCV

       NAFLD/NASH

       Alcohol

       Hemochromatosis

       Other

       Unknown


      2 (5)

      4 (11)

      12 (32)

      13 (34)

      1 (3)

      1 (3)

      5 (12)
      Cirrhotic
      Based on clinical, radiologic or histologic data.
      32 (84)
      Fibroscan
      Fibroscan was performed in 13/38 patients (34%).


       F0–2

       F3–4


      2 (15)

      11 (85)
      Treatment
      1 patient received liver transplantation 20 months after initial TACE.


       Resection

       RFA/MWA

       TACE

       TARE


      2 (5)

      6 (16)

      11 (29)

      19 (50)
      Creatinine (µmol/L)82 ± 26 (32 – 138)82 ± 27 (34 – 145)0.518
      Total bilirubin (µmol/L)14 (4 – 31)15 (4 – 46)0.959
      Alkaline phosphatase (U/L)107 (46 – 330)129 (60 – 933)0.070
      Gamma-GT (U/L)163 (35 – 1080)136 (26– 1150)0.167
      ASAT (U/L)47 (16 – 590)40 (15 – 497)0.053
      ALAT (U/L)39 (11 –224)32 (10 – 270)0.023*
      Albumin (g/L)39 ± 4 (28 – 50)36 ± 5 (23 – 44)0.002*
      Thrombocytes (x109/L)148 (40 – 694)129 (34 – 405)0.072
      PT-INR1.14 (1.00 – 1.84)1.13 (1.00 – 5.50)0.943
      Sodium (mmol/L)138 ± 3 (132 – 144)138 ± 3 (131 – 143)0.731
      Alpha-fetoprotein (mcg/L)10 (2 – 12,100)7 (2 – 4400)0.295
      Portal hypertension

       Varices

       Collaterals

       Thrombocytopenia
      26 (68)

      15 (40)

      22 (58)

      18 (47)
      Ascites

       Absent

       Slight

       Moderate

       n/a


      31 (81)

      6 (16)

      1 (3)

      0(0)


      29 (76)

      5 (13)

      3 (8)

      1 (3)


      0.470
      Hepatic encephalopathy

       No

       Grade 1–2

       n/a


      38 (100)

      0 (0)

      0 (0)


      36 (94)

      1 (3)

      1(3)


      0.317
      Dialysis

       No

       n/a


      38 (100)

      0 (0)


      37 (97)

      1(3)


      1.000
      Child-Pugh score5 (5 – 7)6 (5 – 8)0.026*
      Child-Pugh stage

       A (5–6)

       B (7–9)

       n/a


      34 (90)

      4 (10)

      0(0)


      27 (71)

      10 (26)

      1 (3)


      0.058
      MELD-score9 (6 – 20)9 (6 – 31)0.059
      MELD-Na score11 (7 – 20)11 (7 – 31)0.134
      Performance score (ECOG)

       0

       1

       2

       3


      25 (66)

      10 (26)

      3 (8)

      0 (0)


      18 (47)

      12 (32)

      6 (16)

      2 (5)


      0.026*
      mRECIST

      Complete response

      Partial response

      Stable disease

      Progressive disease


      -

      -

      -

      -


      17 (45)

      9 (24)

      8 (21)

      4 (10)
      BMI (kg/m2)28 (20 – 41)27 (22 – 41)0.232
      Weight (T0) (kg)90 ± 18 (50 – 136)89 ± 18 (54 – 133)0.189
      ∆Weight 1 month before T0 – T0 (kg)0 (−33 – +2)0.003*
      ∆Weight 6 months before T0 – T0 (kg)−2 (−17 – +8)0.939
      PG-SGA total score4 (1 – 13)5 (1 – 19)0.069
      PG-SGA stage

       A well nourished

       B moderate malnourished

       C severe malnourished


      29 (76)

      9 (24)

      0 (0)


      25 (66)

      12 (31)

      1 (3)
      0.197
      Liver frailty index score3.94 (2.46 – 6.33)4.05 (3.27 – 6.78)0.004*
      Liver frailty index

       Robust

       Prefrail

       Frail

       n/a


      3 (8)

      29 (76)

      6 (16)

      0 (0)


      0 (0)

      30 (79)

      7 (18)

      1 (3)
      0.059
      Handgrip strength

       Highest (kg)

      Reduced


      34 ± 9 (16 – 50)

      5 (13)


      31 ± 8 (11 – 43)

      8 (21)


      <0.001*

      0.083
      Chairtest
      5/38 patients (13%)at baseline and 6/38 (16%) at follow-up were not able to perform or complete the chair test in less than 60 s and were not included.


       Time (s)

       Prolonged


      13 (7 – 29)

      5 (13)


      13 (8 – 25)

      9 (24)


      0.673

      0.102
      CT-scan L3
      Baseline CT-scan available in 28/38 patients (74%).


       Skeletal muscle index

       Sarcopenia risk


      46.5 ± 7.1 (31.1–65.0)

      20 (79)


      EORTC-QLQ C30 Global health status (C30)75 (33 – 100)75 (8 – 100)0.811
      Summary score QoL (C30)91 (49 – 100)87 (43 – 100)0.034*
      Summary score QoL (HCC18)8 (0–48)11 (0–53)0.044*
      Death during follow-up16 (42)
      Duration of follow-up (days)461 (65–962)
      Progression during follow-up24 (63)
      Data are presented as n (%), in case of parametric distribution as mean ± SD (range) or in case of nonparametric distribution as median (range). n/a not available. * p-value <0.05.
      a Based on clinical, radiologic or histologic data.
      b Fibroscan was performed in 13/38 patients (34%).
      c 1 patient received liver transplantation 20 months after initial TACE.
      d 5/38 patients (13%)at baseline and 6/38 (16%) at follow-up were not able to perform or complete the chair test in less than 60 s and were not included.
      e Baseline CT-scan available in 28/38 patients (74%).
      There were no significant differences in patient and tumor characteristics between these 38 patients and the entire group of 50 patients. For Child-Pugh scores and tumor response according to mRECIST, 3-month data were available in 49 and 45 cases.
      At the evaluation 3 months post-treatment, complete response according to mRECIST was obtained in 44%, partial response in 20%, stable disease in 20% and progressive disease in 16%. In uni- and multivariable Cox-regression analyses in all 45 patients with 3-month post-treatment mRECIST scores available, tumor response according to mRECIST at 3 months post-treatment was independently associated with PFS (p<0.001) but not with OS or CFS.
      Baseline Child-Pugh stages were A in 90% and B 10% versus A in 71% and B in 26% three months post-treatment (p = 0.058). In uni- and multivariable Cox-regression analyses in all 49 patients with 3-month post-treatment Child-Pugh scores available, deterioration of Child-Pugh score at 3 months post-treatment was independently associated with worse OS and CFS (Tables 2 and 3). Type of anti-tumor therapy (transarterial radioembolization vs other treatment) did not influence these associations. Kaplan-Meier curves showed significantly reduced OS and CFS in patients with deteriorated Child-Pugh scores (log-rank test: P<0.001: Supplementary Figs. 7 and 8).
      In most cases, repeated CT-scans three-months post-treatment were not available, precluding further analyses on (change of) L3-SMI at this time point. The other nutritional parameters deteriorated. No patient improved from reduced HGS at baseline to normal HGS post-treatment. Only one patient improved (from frail to prefrail) according to LFI between baseline and follow-up. As far as LFI is concerned, the patients were therefore divided into two groups (deterioration versus equal and improvement combined). There was no relation between change of nutritional parameters and OS (Table 2). A change in LFI score was independently associated with CFS, which was not the case for the other nutritional parameters (Table 3). Change of the nutritional parameters at 3 months post-treatment was not related to PFS.

      4. Discussion

      Despite the introduction of several new treatment modalities for HCC in the last decades, prognosis remains poor. Impaired nutritional status is a frequent phenomenon in patients with HCC. If impaired nutritional status could be identified with the aid of etiological modeling studies, as a risk factor causally related to poor outcome, further research as to whether early nutritional support could aid in improving survival and quality of life, would be indicated. In our study with etiological multivariable analyses [
      • van Diepen M.
      • Ramspek C.L.
      • Jager K.J.
      • Zoccali C.
      • Dekker F.W.
      Prediction versus aetiology: common pitfalls and how to avoid them.
      ,
      • Ramspek C.L.
      • Steyerberg E.W.
      • Riley R.D.
      • Rosendaal F.R.
      • Dekkers O.M.
      • Dekker F.W.
      • et al.
      Prediction or causality? A scoping review of their conflation within current observational research.
      ], reduced HGS was significantly associated with CFS and nearly reached significance for the association with OS (p = 0.052) (in line with a previous report [
      • Endo K.
      • Kuroda H.
      • Kanazawa J.
      • Sato T.
      • Fujiwara Y.
      • Abe T.
      • et al.
      Impact of grip strength in patients with unresectable hepatocellular carcinoma treated with lenvatinib.
      ]). Our data also suggest that HGS could be used as an easy – to – use tool that could aid in accurate predicting survival in HCC patients. In this respect it is relevant that currently, MRI-scan rather than CT-scan is the preferred imaging modality for HCC diagnosis and follow up. MRI-scan is less validated than CT-scan for assessment of sarcopenia risk. CT-scan only to determine CT L3-SMI requires patient time, commercial software, increases costs and is associated with a small risk. HGS is independently associated with mortality in various conditions including cardiovascular disease [
      • Leong D.P.
      • Teo K.K.
      • Rangarajan S.
      • Lopez-Jaramillo P.
      • Avezum A.J.
      • Orlandini A.
      • et al.
      Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study.
      ] and hemodialysis [
      • Hwang S.H.
      • Lee D.H.
      • Min J.
      • Jeon J.Y.
      Handgrip strength as a predictor of all-cause mortality in patients with chronic kidney disease undergoing dialysis: a meta-analysis of prospective cohort studies.
      ]. HGS has been suggested as a biomarker of aging across the life course and to predispose for fatal outcome in presence of various diseases [
      • Leong D.P.
      • Teo K.K.
      • Rangarajan S.
      • Lopez-Jaramillo P.
      • Avezum A.J.
      • Orlandini A.
      • et al.
      Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study.
      ,
      • Sayer A.A.
      • Kirkwood T.B.L.
      Grip strength and mortality: a biomarker of ageing?.
      ]. Unlike HGS, LFI and PG-SGA did not seem to be associated with survival. It is not surprising, that HGS, LFI, PG-SGA and L3-SMI yielded quite different outcomes. These investigations yield complementary valuable information on muscle strength, muscle quantity and quality, loss of physiological reserve with increased vulnerability and malnutrition. Interestingly, our results indicate that impaired LFI and PG-SGA were associated with worse quality of life, which was not the case for HGS or L3-SMI.
      An important finding of our study was that decrease of liver function (according to Child-Pugh score) at 3 months post-treatment was independently associated with worse overall and complication-free survival. Although decreased liver function could be due to tumor progression, potential negative effects of the anti-tumor treatment itself could also have contributed. Similarly, a recent study comparing transarterial radioembolization and sorafenib with the aid of propensity score matching analyses [
      • van Doorn D.J.
      • Hendriks P.
      • Burgmans M.C.
      • Rietbergen D.D.D.
      • Coenraad M.J.
      • van Delden O.M.
      • et al.
      Liver decompensation as late complication in HCC patients with long-term response following selective internal radiation therapy.
      ] identified decreased liver function after the radioembolization as an important negative prognostic factor. In that study, only liver decompensation occurring longer than four months post-treatment was taken into account (with the rationale to exclude earlier decompensation due to radiation-induced liver disease). Our data suggest that earlier decompensation also has a negative effect on clinical course and that this phenomenon is not restricted to transarterial radioembolization. These findings stress the importance to avoid treatment-related liver injury. New approaches for transarterial radioembolization such as personalized treatment planning based on dosimetry instead of a predefined average absorbed dose in the perfused volume could aid to avoid liver injury [
      • Garin E.
      • Tselikas L.
      • Guiu B.
      • Chalaye J.
      • Edeline J.
      • de Baere T.
      • et al.
      Personalised versus standard dosimetry approach of selective internal radiation therapy in patients with locally advanced hepatocellular carcinoma (DOSISPHERE-01): a randomised, multicentre, open-label phase 2 trial.
      ].
      Strengths of the current study are its prospective design and the baseline evaluation of nutrition disorders and nutrition-related conditions. Also, the variables of interest were corrected in the multivariable analyses of this etiologic modeling study, for BCLC stage, AFP and age, considering the established considerable impact of these three confounders on outcome in hepatocellular carcinoma. In the statistical approach of previous studies on nutritional status in HCC patients, prerequisites for etiological modeling were generally not taken into account or nor clearly reported [
      • van Diepen M.
      • Ramspek C.L.
      • Jager K.J.
      • Zoccali C.
      • Dekker F.W.
      Prediction versus aetiology: common pitfalls and how to avoid them.
      ,
      • Ramspek C.L.
      • Steyerberg E.W.
      • Riley R.D.
      • Rosendaal F.R.
      • Dekkers O.M.
      • Dekker F.W.
      • et al.
      Prediction or causality? A scoping review of their conflation within current observational research.
      ]. Limitations of our work are the heterogenous anti-tumor treatment modalities and the relatively small patients numbers. Especially for the nutritional assessments 3 months post-treatment, patient numbers were limited. We therefore cannot exclude a type II error for our finding that change of nutritional parameters was in general not associated with outcome (Tables 2 and 3). Also, in our study, CT-scan at baseline was only performed if clinically indicated and only rarely at follow-up. Therefore, definite conclusions about the value of CT-scan - derived L3-SMI cannot be drawn from the current work. Of note, we chose age-bound, sex-specific reference values for HGS, where we considered a value below the 10th percentile as reduced HGS [
      • Dodds R.M.
      • Syddall H.E.
      • Cooper R.
      • Benzeval M.
      • Deary I.J.
      • Dennison E.M.
      • et al.
      Grip strength across the life course: normative data from twelve British studies.
      ]. Higher cut-off point for reduced HGS would theoretically lead to higher sensitivity but lower specificity. Finally, the included patients were a selected group of patients, mostly in good condition, referred for potential anti-tumor therapy to a tertiary care center with special interest in transarterial radioembolisation. Early dietary intervention could be especially effective in this patient group. Our results may not be generalizable to HCC patients with end stage (BCLC-D) HCC.
      In conclusion, impaired nutritional status occurs frequently in patients with hepatocellular carcinoma. Reduced baseline hand-grip strength and deteriorated Child-Pugh score 3 months after anti-tumor treatment were associated with reduced overall and complication-free survival in HCC patients. Whereas avoiding treatment-related liver injury should certainly be pursued, the potential value of dietary interventions to improve outcome in selected HCC patients remains to be explored.

      Funding

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Declaration of Competing Interest

      The authors declare that they have no conflict of interest.

      Acknowledgement

      The authors thank C-P. L. van Erpecum for expert statistical advice.

      Appendix. Supplementary materials

      References

        • Sung H.
        • Ferlay J.
        • Siegel R.L.
        • Laversanne M.
        • Soerjomataram I.
        • Jemal A.
        • et al.
        Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
        CA Cancer J Clin. 2021; 71: 209-249https://doi.org/10.3322/caac.21660
        • Bunchorntavakul C.
        • Reddy K.R.
        Review article: malnutrition/sarcopenia and frailty in patients with cirrhosis.
        Aliment Pharmacol Ther. 2020; 51: 64-77https://doi.org/10.1111/apt.15571
        • Lai J.C.
        • Covinsky K.E.
        • McCulloch C.E.
        • Feng S.
        The liver frailty index improves mortality prediction of the subjective clinician assessment in patients with cirrhosis.
        Am J Gastroenterol. 2018; 113: 235-242https://doi.org/10.1038/ajg.2017.443
        • Wang S.
        • Whitlock R.
        • Xu C.
        • Taneja S.
        • Singh S.
        • Abraldes J.G.
        • et al.
        Frailty is associated with increased risk of cirrhosis disease progression and death.
        Hepatology. 2022; 75: 600-609https://doi.org/10.1002/hep.32157
        • Sieber C.C.
        Malnutrition and sarcopenia.
        Aging Clin Exp Res. 2019; : 793-798https://doi.org/10.1007/s40520-019-01170-1
        • Huisman E.J.
        • Trip E.J.
        • Siersema P.D.
        • Van Hoek B.
        • Van Erpecum K.J.
        Protein energy malnutrition predicts complications in liver cirrhosis.
        Eur J Gastroenterol Hepatol. 2011; 23: 982-989https://doi.org/10.1097/MEG.0b013e32834aa4bb
        • Montano-Loza A.J.
        Clinical relevance of sarcopenia in patients with cirrhosis.
        World J Gastroenterol. 2014; 20: 8061-8071https://doi.org/10.3748/wjg.v20.i25.8061
        • Kroh A.
        • Uschner D.
        • Lodewick T.
        • Eickhoff R.M.
        • Schöning W.
        • Ulmer F.T.
        • et al.
        Impact of body composition on survival and morbidity after liver resection in hepatocellular carcinoma patients.
        Hepatobiliary Pancreat Dis Int. 2019; 18: 28-37https://doi.org/10.1016/j.hbpd.2018.07.008
        • Akce M.
        • Liu Y.
        • Zakka K.
        • Martini D.J.
        • Draper A.
        • Alese O.B.
        • et al.
        Impact of sarcopenia, BMI, and inflammatory biomarkers on survival in advanced hepatocellular carcinoma treated with anti-PD-1 antibody.
        Am J Clin Oncol Cancer Clin Trials. 2021; 44: 74-81https://doi.org/10.1097/COC.0000000000000787
        • Marasco G.
        • Serenari M.
        • Renzulli M.
        • Alemanni L.V.
        • Rossini B.
        • Pettinari I.
        • et al.
        Clinical impact of sarcopenia assessment in patients with hepatocellular carcinoma undergoing treatments.
        J Gastroenterol. 2020; 55: 927-943https://doi.org/10.1007/s00535-020-01711-w
        • Meza-Junco J.
        • Montano-Loza A.J.
        • Baracos V.E.
        • Prado C.M.M.
        • Bain V.G.
        • Beaumont C.
        • et al.
        Sarcopenia as a prognostic index of nutritional status in concurrent cirrhosis and hepatocellular carcinoma.
        J Clin Gastroenterol. 2013; 47: 861-870https://doi.org/10.1097/MCG.0b013e318293a825
        • Fujiwara N.
        • Nakagawa H.
        • Kudo Y.
        • Tateishi R.
        • Taguri M.
        • Watadani T.
        • et al.
        Sarcopenia, intramuscular fat deposition, and visceral adiposity independently predict the outcomes of hepatocellular carcinoma.
        J Hepatol. 2015; 63: 131-140https://doi.org/10.1016/j.jhep.2015.02.031
        • Chang K.V.
        • De Chen J
        • Wu W.T.
        • Huang K.C.
        • Hsu C.T.
        • Han D.S
        Association between loss of skeletal muscle mass and mortality and tumor recurrence in hepatocellular carcinoma: a systematic review and meta-analysis.
        Liver Cancer. 2018; 7: 90-103https://doi.org/10.1159/000484950
        • Begini P.
        • Gigante E.
        • Antonelli G.
        • Carbonetti F.
        • Iannicelli E.
        • Anania G.
        • et al.
        Sarcopenia predicts reduced survival in patients with hepatocellular carcinoma at first diagnosis.
        Ann Hepatol. 2017; 16: 107-114https://doi.org/10.5604/16652681.1226821
        • Badran H.
        • Elsabaawy M.M.
        • Ragab A.
        • Aly R.A.
        • Alsebaey A.
        • Sabry A.
        Baseline sarcopenia is associated with lack of response to therapy, liver decompensation and high mortality in hepatocellular carcinoma patients.
        Asian Pac J Cancer Prev. 2020; 21: 3285-3290https://doi.org/10.31557/APJCP.2020.21.11.3285
        • Iritani S.
        • Imai K.
        • Takai K.
        • Hanai T.
        • Ideta T.
        • Miyazaki T.
        • et al.
        Skeletal muscle depletion is an independent prognostic factor for hepatocellular carcinoma.
        J Gastroenterol. 2015; 50: 323-332https://doi.org/10.1007/s00535-014-0964-9
        • Zhang G.
        • Meng S.
        • Li R.
        • Ye J.
        • Zhao L.
        Clinical significance of sarcopenia in the treatment of patients with primary hepatic malignancies, a systematic review and meta-analysis.
        Oncotarget. 2017; 8: 102474-102485https://doi.org/10.18632/oncotarget.19687
        • Schütte K.
        • Tippelt B.
        • Schulz C.
        • Röhl F.W.
        • Feneberg A.
        • Seidensticker R.
        • et al.
        Malnutrition is a prognostic factor in patients with hepatocellular carcinoma (HCC).
        Clin Nutr. 2015; 34: 1122-1127https://doi.org/10.1016/j.clnu.2014.11.007
        • Voron T.
        • Tselikas L.
        • Pietrasz D.
        • Pigneur F.
        • Laurent A.
        • Compagnon P.
        • et al.
        Sarcopenia impacts on short- and long-term results of hepatectomy for hepatocellular carcinoma.
        Ann Surg. 2015; 261: 1173-1183https://doi.org/10.1097/SLA.0000000000000743
        • Harimoto N.
        • Shirabe K.
        • Yamashita Y.I.
        • Ikegami T.
        • Yoshizumi T.
        • Soejima Y.
        • et al.
        Sarcopenia as a predictor of prognosis in patients following hepatectomy for hepatocellular carcinoma.
        Br J Surg. 2013; 100: 1523-1530https://doi.org/10.1002/bjs.9258
        • Takagi K.
        • Yagi T.
        • Yoshida R.
        • Shinoura S.
        • Umeda Y.
        • Nobuoka D.
        • et al.
        Sarcopenia and American society of anesthesiologists physical status in the assessment of outcomes of hepatocellular carcinoma patients undergoing hepatectomy.
        Acta Med Okayama. 2016; 70: 363-370https://doi.org/10.18926/AMO/54594
        • Xu L.
        • Jing Y.
        • Zhao C.
        • Zhang Q.
        • Zhao X.
        • Yang J.
        • et al.
        Preoperative computed tomography-assessed skeletal muscle index is a novel prognostic factor in patients with hepatocellular carcinoma following hepatectomy: a meta-analysis.
        J Gastrointest Oncol. 2020; 11: 1040-1053https://doi.org/10.21037/jgo-20-122
        • Levolger S.
        • Van Vledder M.G.
        • Muslem R.
        • Koek M.
        • Niessen W.J.
        • De Man R.A.
        • et al.
        Sarcopenia impairs survival in patients with potentially curable hepatocellular carcinoma.
        J Surg Oncol. 2015; 112: 208-213https://doi.org/10.1002/jso.23976
        • Salman A.
        • Salman M.
        • Moustafa A.
        • Shaaban H.E.-.D.
        • El-Mikkawy A.
        • Labib S.
        • et al.
        Impact of sarcopenia on two-year mortality in patients with HCV-associated hepatocellular carcinoma after radiofrequency ablation.
        J Hepatocell Carcinoma. 2021; 8: 313-320https://doi.org/10.2147/jhc.s300680
        • Antonelli G.
        • Gigante E.
        • Iavarone M.
        • Begini P.
        • Sangiovanni A.
        • Iannicelli E.
        • et al.
        Sarcopenia is associated with reduced survival in patients with advanced hepatocellular carcinoma undergoing sorafenib treatment.
        United Eur Gastroenterol J. 2018; 6: 1039-1048https://doi.org/10.1177/2050640618781188
        • Nishikawa H.
        • Nishijima N.
        • Enomoto H.
        • Sakamoto A.
        • Nasu A.
        • Komekado H.
        • et al.
        Prognostic significance of sarcopenia in patients with hepatocellular carcinoma undergoing sorafenib therapy.
        Oncol Lett. 2017; 14: 1637-1647https://doi.org/10.3892/ol.2017.6287
        • Endo K.
        • Kuroda H.
        • Kanazawa J.
        • Sato T.
        • Fujiwara Y.
        • Abe T.
        • et al.
        Impact of grip strength in patients with unresectable hepatocellular carcinoma treated with lenvatinib.
        Cancers. 2020; 12 (Basel): 2146https://doi.org/10.3390/cancers12082146
        • Lanza E.
        • Masetti C.
        • Messana G.
        • Muglia R.
        • Pugliese N.
        • Ceriani R.
        • et al.
        Sarcopenia as a predictor of survival in patients undergoing bland transarterial embolization for unresectable hepatocellular carcinoma.
        PLoS ONE. 2020; 15e0232371https://doi.org/10.1371/journal.pone.0232371
        • Lee J.
        • Cho Y.
        • Park S.
        • Kim J.W.
        • Lee I.J.
        Skeletal muscle depletion predicts the prognosis of patients with hepatocellular carcinoma treated with radiotherapy.
        Front Oncol. 2019; 9: 1075https://doi.org/10.3389/fonc.2019.01075
        • Lai J.C.
        • Tandon P.
        • Bernal W.
        • Tapper E.B.
        • Ekong U.
        • Dasarathy S.
        • et al.
        Malnutrition, frailty, and sarcopenia in patients with cirrhosis: 2021 practice guidance by the American association for the study of liver diseases.
        Hepatology. 2021; 74: 1611-1644https://doi.org/10.1002/hep.32049
        • Cruz-Jentoft A.J.
        • Bahat G.
        • Bauer J.
        • Boirie Y.
        • Bruyère O.
        • Cederholm T.
        • et al.
        Sarcopenia: revised European consensus on definition and diagnosis.
        Age Ageing. 2019; 48: 601https://doi.org/10.1093/ageing/afy169
        • Lai J.C.
        • Covinsky K.E.
        • Dodge J.L.
        • Boscardin W.J.
        • Segev D.L.
        • Roberts J.P.
        • et al.
        Development of a novel frailty index to predict mortality in patients with end-stage liver disease.
        Hepatology. 2017; 66: 564-574https://doi.org/10.1002/hep.29219
        • Jager-Wittenaar H.
        • Ottery F.D.
        Assessing nutritional status in cancer: role of the patient-generated subjective global assessment.
        Curr Opin Clin Nutr Metab Care. 2017; 20: 322-329https://doi.org/10.1097/MCO.0000000000000389
        • Galle P.R.
        • Forner A.
        • Llovet J.M.
        • Mazzaferro V.
        • Piscaglia F.
        • Raoul J.L.
        • et al.
        EASL clinical practice guidelines: management of hepatocellular carcinoma.
        J Hepatol. 2018; 69: 182-236https://doi.org/10.1016/j.jhep.2018.03.019
        • Aaronson N.K.
        • Ahmedzai S.
        • Bergman B.
        • Bullinger M.
        • Cull A.
        • Duez N.J.
        • et al.
        The European organization for research and treatment of cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
        J Natl Cancer Inst. 1993; 85: 365-376https://doi.org/10.1093/jnci/85.5.365
        • Blazeby J.M.
        • Currie E.
        • Zee B.C.Y.
        • Chie W.-.C.
        • Poon R.T.
        • Garden O.J.
        Development of a questionnaire module to supplement the EORTC QLQ-C30 to assess quality of life in patients with hepatocellular carcinoma, the EORTC QLQ-HCC18.
        Eur J Cancer. 2004; 40: 2439-2444https://doi.org/10.1016/j.ejca.2004.06.033
        • Li L.
        • Mo F.K.F.
        • Chan S.L.
        • Hui E.P.
        • Tang N.S.L.
        • Koh J.
        • et al.
        Prognostic values of EORTC QLQ-C30 and QLQ-HCC18 index-scores in patients with hepatocellular carcinoma - clinical application of health-related quality-of-life data.
        BMC Cancer. 2017; 17: 8https://doi.org/10.1186/s12885-016-2995-5
        • Carey E.J.
        • Lai J.C.
        • Wang C.W.
        • Dasarathy S.
        • Lobach I.
        • Montano-Loza A.J.
        • et al.
        A multicenter study to define sarcopenia in patients with end-stage liver disease.
        Liver Transpl. 2017; 23: 625-633https://doi.org/10.1002/lt.24750
        • Dodds R.M.
        • Syddall H.E.
        • Cooper R.
        • Benzeval M.
        • Deary I.J.
        • Dennison E.M.
        • et al.
        Grip strength across the life course: normative data from twelve British studies.
        PLoS ONE. 2014; 9e113637https://doi.org/10.1371/journal.pone.0113637
        • Lencioni R.
        • Llovet J.M.
        Modified RECIST (mRECIST) assessment for hepatocellular carcinoma.
        Semin Liver Dis. 2010; 30: 52-60https://doi.org/10.1055/s-0030-1247132
        • van Diepen M.
        • Ramspek C.L.
        • Jager K.J.
        • Zoccali C.
        • Dekker F.W.
        Prediction versus aetiology: common pitfalls and how to avoid them.
        Nephrol Dial Transplant. 2017; 32 (Off Publ Eur Dial Transpl Assoc Eur Ren Assoc): ii1-ii5https://doi.org/10.1093/ndt/gfw459
        • Ramspek C.L.
        • Steyerberg E.W.
        • Riley R.D.
        • Rosendaal F.R.
        • Dekkers O.M.
        • Dekker F.W.
        • et al.
        Prediction or causality? A scoping review of their conflation within current observational research.
        Eur J Epidemiol. 2021; 36: 889-898https://doi.org/10.1007/s10654-021-00794-w
        • Reinders M.T.M.
        • van Erpecum K.J.
        • Smits M.L.J.
        • Braat A.J.A.T.
        • de Bruijne J.
        • Bruijnen R.C.G.
        • et al.
        Safety and efficacy of holmium-166 radioembolization in hepatocellular carcinoma - the HEPAR Primary study.
        J. Nucl. Med. 2022; (Online ahead of print.PMID: 35589409)https://doi.org/10.2967/jnumed.122.263823
        • Leong D.P.
        • Teo K.K.
        • Rangarajan S.
        • Lopez-Jaramillo P.
        • Avezum A.J.
        • Orlandini A.
        • et al.
        Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study.
        Lancet. 2015; 386: 266-273https://doi.org/10.1016/S0140-6736(14)62000-6
        • Hwang S.H.
        • Lee D.H.
        • Min J.
        • Jeon J.Y.
        Handgrip strength as a predictor of all-cause mortality in patients with chronic kidney disease undergoing dialysis: a meta-analysis of prospective cohort studies.
        J Ren Nutr. 2019; 29 (Off J Counc Ren Nutr Natl Kidney Found): 471-479https://doi.org/10.1053/j.jrn.2019.01.002
        • Sayer A.A.
        • Kirkwood T.B.L.
        Grip strength and mortality: a biomarker of ageing?.
        Lancet. 2015; 386: 226-227https://doi.org/10.1016/S0140-6736(14)62349-7
        • van Doorn D.J.
        • Hendriks P.
        • Burgmans M.C.
        • Rietbergen D.D.D.
        • Coenraad M.J.
        • van Delden O.M.
        • et al.
        Liver decompensation as late complication in HCC patients with long-term response following selective internal radiation therapy.
        Cancers. 2021; 13 (Basel): 5427https://doi.org/10.3390/cancers13215427
        • Garin E.
        • Tselikas L.
        • Guiu B.
        • Chalaye J.
        • Edeline J.
        • de Baere T.
        • et al.
        Personalised versus standard dosimetry approach of selective internal radiation therapy in patients with locally advanced hepatocellular carcinoma (DOSISPHERE-01): a randomised, multicentre, open-label phase 2 trial.
        Lancet Gastroenterol Hepatol. 2021; 6: 17-29https://doi.org/10.1016/S2468-1253(20)30290-9