Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a meta-analysis

HPB

https://doi.org/10.1016/j.hpb.2018.03.019

REVIEW ARTICLE

Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a meta-analysis Zhongran Man*, Qing Pang*, Lei Zhou*, Yong Wang, Xiaosi Hu, Song Yang, Hao Jin & Huichun Liu Department of Hepatobiliary Surgery, The First Affiliated Hospital of Bengbu Medical College, Bengbu 233000, Anhui, China

Abstract Background: To date, epidemiological evidence of the association between preoperative prognostic nutritional index (PNI) and the prognosis of hepatocellular carcinoma (HCC) remains controversial. Methods: A literature search was performed in the databases of PubMed, Embase, and Web of Science. Hazard ratio (HR), odds ratio (OR), and 95% confidence interval (CI) were extracted to estimate the association of preoperative PNI with overall survival (OS), disease-free survival (DFS), and postoperative recurrence of HCC, respectively. A random-effects model was used to calculate the pooled effect size. Results: Thirteen studies with a total of 3,738 patients with HCC met inclusion criteria for this metaanalysis. It indicated that a lower level of preoperative PNI was a significant predictor of worse OS (HR = 1.82, 95%CI: 1.44-2.31) and DFS (HR = 1.49, 95% CI: 1.06-2.07). In addition, risk of postoperative recurrence was significantly higher in patients with a lower preoperative PNI (OR = 1.92, 95% CI: 1.332.76). Subgroup analysis based on therapeutic intent demonstrated a significant positive association between preoperative low PNI and worse OS for those patients undergoing surgical resection and for those undergoing TACE or non-surgical treatment. Conclusion: The current meta-analysis demonstrates that preoperative PNI is a prognostic marker in HCC. Received 24 January 2018; accepted 28 March 2018

Correspondence Huichun Liu, Department of Hepatobiliary Surgery, First Affiliated Hospital of Bengbu Medical College, 287 Changhuai Road, Bengbu, Anhui 233000, China. E-mail: [email protected]

Correspondence Hao Jin, Department of Hepatobiliary Surgery, First Affiliated Hospital of Bengbu Medical College, 287 Changhuai Road, Bengbu, Anhui 233000, China. E-mail: [email protected]

Correspondence Qing Pang, Department of Hepatobiliary Surgery, First Affiliated Hospital of Bengbu Medical College, 287 Changhuai Road, Bengbu, Anhui 233000, China. E-mail: [email protected]

Introduction Hepatocellular carcinoma (HCC) is the fifth most common malignancy and the second most frequent cause of death of

*

cancer worldwide.1 To date, surgical resection remains the main curative option for patients with HCC. However, due to the high risk of postoperative recurrence, prognosis of HCC remains unsatisfactory. Therefore, further efforts to identify the potential prognostic factors are critical to improve survival.

These authors contributed equally to this work.

HPB 2018, -, 1–8

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

2

Several factors have been reported to be associated with the prognosis of HCC, including tumor diameter, disease stage, alpha fetoprotein (AFP), vascular invasion, presence of cirrhosis, platelet count.2–4 However, there is still no consensus on the best factor for predicting the outcome of patients with HCC. There is increasing evidence supporting the role of preoperative nutritional and immunological statuses as predictors of outcome in several malignant tumors including HCC.5–7 As immunonutritional molecular biomarkers are not readily obtained in daily medical practice, several nutritional and immunological based scores derived from simple serum parameters have been proposed to serve as prognostic predictors for patients with malignancy.8,9 Several meta-analyses further confirm that nutritional and immunological indices, such as neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and the Glasgow prognostic score, can function as prognostic factors for patients with HCC.10,11 The preoperative prognostic nutritional index (PNI) reflects both the nutritional and immunological conditions of tumor patient. It is calculated using the following formula: 10 × ALB (g/dl) + 0.005 × lymphocyte count (per mm3).12 A low (the optimal cut-off value ranged from 40 to 52) preoperative PNI has been shown to be independent adverse prognostic indicator for various malignancies, such as gastric cancer, colorectal cancer, cholangiocarcinoma, and HCC.13–17 Zhao et al. conducted a meta-analysis of published studies to systematically review the association between the preoperative PNI and the prognosis of digestive system cancers.18 With regard to HCC, the authors included four studies all of a small sample size. All the four studies showed significant association between preoperative PNI and HCC survival. Recently, several additional studies have reported that preoperative PNI is not significantly related to survival in HCC.19–21 Therefore equipoise exists as to the usefulness of preoperative PNI in predicting prognosis for patients with HCC. Thus the aim of this study was to perform a meta-analysis of relevant studies regarding the association between preoperative PNI and the prognosis of HCC.

Materials and methods Search strategy and selection criteria Two investigators (Zhongran Man and Qing Pang) independently performed a comprehensive literature search through the databases of PubMed, Embase, and Web of Science up to November 30, 2017. The main search terms were (PNI OR prognostic nutritional index) AND (hepatocellular carcinoma OR liver cancer). Additionally, a manual search was performed of the cited the references of the relevant publications. The eligible studies were screened on the basis of the following inclusion criteria: (a) designed as an observational study; (b) diagnosis of HCC was based on the pathological or clinical results; (c) assessed the effects of preoperative PNI in overall survival (OS), disease-free survival (DFS), or recurrence; (d) outcome measures could be extracted directly or calculated

HPB 2018, -, 1–8

indirectly. No restrictions by stage or treatment were placed. If there were duplicated publications based on the same population, only the larger of the duplicate studies were included. Exclusion criteria included studies that: (a) only assessed the effect of preoperative PNI on the short-term complications; (b) involved individuals with metastatic liver cancer or recurrent HCC; (c) only reported postoperative PNI; (d) outcome measures could not be calculated; (e) mean or median follow-up time less than two years (or maximum less than four years) for radical therapy, or less than one year (or maximum less than two years) for non-radical therapy; (f) were reviews, conference abstracts, case reports, or correspondence. Data abstraction Two authors (Zhongran Man and Yong Wang) independently reviewed each eligible study and extracted the following data with a standardized data-collection protocol: first author, publication year, region of the study population, study period, method of therapy, mean or median age of patient, sample size, duration of follow-up, cut-off value of preoperative PNI, HR value (or OR, adjusted value in preference) and 95% CI. Mixed therapy was defined as various treatments and included both radical and nonradical therapy in the study. The research quality of each included study was assessed by Lei Zhou with the modified Newcastle– Ottawa Scale (NOS) scores. NOS scores of greater than or equal to 7 was defined as high quality. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guideline was used throughout the reporting of the current meta-analysis. Statistical analysis The outcomes observed included OS, DFS, and recurrence. Stata 12.0 software (StataCorp, Texas, USA) was used for the pooled analysis calculations. A random-effects model was uniformly adopted to calculate pooled effect size with the eligible studies. Between-study heterogeneity studies was assessed using the Q value and I2 statistic values (25%, 50%, and 75% correspond to the cut-off values of low, moderate, and high degrees of heterogeneity, respectively). There was statistically substantial heterogeneity if P value less than 0.1 in Q statistic or I2 >50%. Afterwards, according to the factors that might explain the potential source of heterogeneity, subgroup analyses and metaregression were performed. Studies were evaluated as to whether any single study could significantly influence the pooled effect size. A fixed-effects model to calculate the pooled effect sizes, and compared them with the results from a random-effects model. Finally, publication bias was determined by Begg’s funnel plot. It was considered to be statistically significant if a P value was less than 0.05.

Results The flow chart of the process of literature retrieval and screening is shown in Fig. 1. Agreement between the two authors was good on which studies to be included (kappa = 0.919).

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

3

Figure 1 Flow chart of search strategy and study selection

Article characteristics The baseline information of the thirteen articles, eleven of which published after 2015 are summarized in Supplementary Table 1. A total of 3738 patients with HCC were included for analysis, 2408 (64%) of which received surgical resection and 1143 (31%) received TACE. All were retrospective studies, three of which were performed in western countries, six from China and four from Japan. The median follow-up time (reported) ranged from 14 to 46 months. The median cut-off value of preoperative PNI was 45 (41–52). All publications except for five15,21–24 controlled for potential confounders and reported adjusted HR (or OR) values. The assessment of study quality demonstrated that the median (range) NOS score was 7. Twelve,4,17,19–28 five,4,21,25–27 and four4,15,17,19 publications were available respectively to assess the effects of preoperative PNI on OS, DFS, and recurrence of HCC. In two publications,17,26 Okamura et al. analyzed an overlapping population with different primary outcomes. Recurrence was reported in the updated one22 and DFS in the original one,24 while OS was in both. The more recent paper was used for OS analysis. According to the data of BCLC stage in seven articles,15,20,24–28 950 (40%) patients were classified as early stage (0–A) and 1424 (60%) patients were in the medium-term or advanced stage (B–D). Eight publications15,17,21,23,24,26–28 reported the specific Child–Pugh grade and 88.3% (2113/2392) of the patients were Child–Pugh A. The mean or median size of tumor ranged from 3 to 13 cm. All studies except two19,28 provided the information on the number of tumor, and there were 2235 (65%) patients with a solitary lesion, 1198 (35%) patients with multiple lesions.

HPB 2018, -, 1–8

Among the nine publications that focused on patients with surgical resection, five4,15,17,21,27 assessed the margin, size, or procedures of excision. A total of 70% (642/922) patients received anatomical hepatectomy according to the three relevant publications.15,17,27 Effect of preoperative PNI on OS for patients with HCC Eleven4,17,19–25,27,28 articles estimated the association between PNI and OS of HCC. A total of 1920 (59%) patients from eight studies4,17,19,21,22,25,27 underwent surgical resection and 1339 (41%) patients from four studies20,23,24,28 underwent TACE or mixed therapies. No significant effect of preoperative PNI on OS was observed in three studies19–21 while nine4,17,22–25,27,28 reported statistically significant results. A random-effects model was used to pool all the included studies stratified according to the therapies (Fig. 2) and demonstrated a low (the cut-off value ranged from 41 to 52) preoperative PNI was associated with a poor OS with a HR value of 1.82 (95% CI: 1.44–2.31) with a high degree of inter-study heterogeneity (I2 = 90.5%, P < 0.001). A stratified analysis demonstrated a significant association between preoperative low PNI and worse OS both in the subgroup of patients undergoing surgical resection (HR = 2.16, 95% CI: 1.39–3.34) and other therapies (HR = 1.49, 95% CI: 1.35–1.65). Effects of preoperative PNI on DFS and recurrence of HCC Five4,21,25–27 studies reported the prognostic value of preoperative PNI on DFS. All the 1518 HCC patients underwent surgical

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

4

Figure 2 Forest plots of meta-analysis on association between preoperative PNI and OS of HCC stratified according to the methods of therapy

resection. A random-effects model demonstrated that low preoperative PNI was associated with a worse DFS in patients with HCC (Fig. 3a, HR = 1.49, 95% CI: 1.06–2.07), with a high degree of inter-study heterogeneity (I2 = 90.4%, P < 0.001). A total of four relevant studies4,17,19,20 with 888 (100%) HCC patients who received surgical resection were analyzed to determine the effect of preoperative PNI on postoperative recurrence. Three4,17,20 of the four studies demonstrated patients with a lower level of preoperative PNI were associated with an increased risk of postoperative recurrence (Fig. 3b). The exploration of heterogeneity Given significant inter-study heterogeneity was observed, potential causes were explored (Table 1). After the introduction of the regression model with the seven factors, the estimate of inter-study variance decreased from 0.12 to 0.01. Thus the seven factors explain 88% of the heterogeneity. Among these factors, region of origin and type of survival analysis were found to be the significant potential sources of heterogeneity following meta-regression (P < 0.05). A similar analysis for the subgroup of patients who underwent surgical resection is. HPB 2018, -, 1–8

Sensitivity analysis and publication bias A sensitivity analysis was performed in order to validate the findings. Firstly, no significant changes were observed in all the subgroups when a fixed effect model was used (Table 1). Secondly, influence analysis was performed and demonstrated that no one study could overly affect the summary of OS risk estimate (Supplementary Fig. 1). Publication bias was assessed by funnel plot analysis and the plots showed basic symmetry (Supplementary Fig. 2). No significant publication bias was further determined by Begg’s test, with a P value of 0.304.

Discussion The results of the current study demonstrated that in patients with HCC a low level of preoperative PNI was associated with a poor OS and DFS and a high risk of postoperative recurrence. The prognostic significance was not reduced by treatment intent, cut-off value of preoperative PNI, sample size, study quality, or type of survival analysis, although it was important to note that the effect did not hold true for western patients.

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

5

Figure 3 Forest plots of meta-analysis on association between preoperative PNI and DFS (a), recurrence (b) of HCC

HPB 2018, -, 1–8

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

6

Table 1 Subgroup analysis and meta-regression analysis for the studies that assessed the significance of PNI in predicting OS of HCC

Covariates

Subgroup

Overall Therapies

No. of HR (95% CI) studies Random-effects model

Fixed-effects model

Heterogeneity

Subgroup analysis

P

Q value

P

45.02

<0.001

0.099

36.6

<0.001

0.444

46.9

<0.001

0.083

83.62

<0.001

0.040

60.33

<0.001

0.580

49.76

<0.001

0.820

55.68

<0.001

0.033

I2

12

1.82(1.44–2.31)

1.07(1.04–1.10) <0.001

90.5

Surgical resection

8

2.16(1.39–3.34)

1.04(1.01–1.07) <0.001

89.8

Non-operation

4

1.49(1.35–1.65)

1.49(1.35–1.65) 0.525

0

Cut-off value of PNI

45

7

1.67(1.26–2.22)

1.05(1.02–1.08) <0.001

91.0

>45

5

2.14(1.41–3.24)

1.66(1.44–1.92) 0.014

67.8

NOS score

7

7

2.13(1.61–2.83)

2.01(1.67–2.41) 0.076

47.5

<7

5

1.50(1.13–1.98)

1.05(1.02–1.08) <0.001

93.0

Asia

9

1.89(1.55–2.30)

1.61(1.47–1.76) 0.005

64.0

Region

America/Europe 3

1.36(0.80–2.30)

1.02(0.99–1.05) 0.007

79.6

Number of patients

200

7

1.94(1.54–2.45)

1.72(1.52–1.94) 0.034

56.1

<200

5

1.53(1.11–2.12)

1.04(1.01–1.07) <0.001

90.4

Age (mean or median)

60

6

1.71(1.45–2.02)

1.66(1.46–1.88) 0.262

22.8

>60

Survival analysis Multivariable Univariate

6

1.88(1.33–2.66)

1.04(1.01–1.07) <0.001

91.6

8

2.09 (1.65–2.63)

2.01(1.70–2.38) 0.121

38.8

4

1.41(1.05–1.88)

1.05(1.02–1.08) <0.001

93.8

Meta-regression P

Bold values mean P < 0.05. PNI, prognostic nutritional index; OS, overall survival; HCC, hepatocellular carcinoma; HR, hazard ratio; TACE, transcatheter arterial chemoembolization; NOS, Newcastle–Ottawa Scale.

There are several reasons to explain why a low preoperative PNI maybe associated with a poor prognosis in patients with HCC. Firstly, a low preoperative PNI in patients with HCC indicates a relatively poor immunonutritional status, which has been associated with advanced cancer progression, high tumor load, delayed therapy, and increased risk of postoperative complications.29,30 Secondly, oxidative stress is involved in various diseases, including chronic liver diseases and HCC.31–33 Malnutrition and immune dysfunction can induce oxidative stress and augment a cascade of molecular reactions, which play a crucial role in carcinogenesis.31,34 Thirdly, the prognostic value of preoperative PNI in HCC can be partly explained by the role of serum ALB and lymphocyte count. Serum ALB, which constitutes more than 50% of the serum proteins in healthy individuals.30 Low serum ALB level is correlated with an increased inflammatory response to the tumor.35 Recent studies have indicated that the serum ALB is associated with the prognosis of various malignancies.36–38 ALB is mainly produced by the liver parenchymal cells and thus it is widely acknowledged that ALB is affected by chronic liver disease. Due to the specific structural changes and oxidative damage, advanced cirrhosis is characterized by reduced level as well as impaired function of ALB.39 It is reported that lower level of ALB in HCC are correlated with significantly larger tumor diameters, increased tumor number, higher AFP level, and greater prevalence of portal vein invasion.40 In addition, serum ALB has been extensively used for tumor staging and further prognosis evaluation in patients with HCC.41–43 HPB 2018, -, 1–8

In contrast, as crucial components of the immune system, lymphocytes play a key role in controlling the tumor.44 Lymphocytes reflect the immunological status and can eradicate tumor cells by cytokine secretion and cytotoxic cell death.45 Elevated B lymphocytes proportion is associated with a significantly low incidence of liver nodules and carcinoma, indicating that B lymphocytes may suppress hepatic tumorigenesis.46 In contrast, T lymphocytes can efficiently promote tumor regression through the secretion of interleukin-2.47 In addition, the frequencies of CD8+ T cells, CD4 + T cells, natural killer cells, and regulatory T cells in the tumor tissue of HCC are significantly associated with survival.48 Several limitations and shortcomings are noted in the current meta-analysis. First, significant heterogeneity between the included studies was observed. Using subgroup analysis, therapeutic intent, cut-off value of preoperative PNI, study quality, region, sample size, and survival analysis were identified as the potential factors contributing to the heterogeneity. In addition how the relationship between preoperative PNI and prognosis varies according to tumor stage HCC remains unknown as few studies have reported the specific tumor stage when assessing the association between preoperative PNI and HCC. Thirdly, OS and DFS are time-dependent variables and it may be more significant to compare 1-year, 2-year, or other OS/DFS, respectively. However, none of the included studies in the current meta-analysis provided the relevant data. Fourthly, none of the included studies assessed the effect of preoperative PNI on intraoperative or postoperative short-term mortality. Whether it could help to select HCC patients to avoid futile surgery should be further

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

7

investigated. In addition, although PNI has been considered as the most consistent method to identify malnutrition in patients included on the waiting list for liver transplantation,49 the prognostic role of PNI in patients with liver transplantation remains unclear. In conclusion, preoperative PNI has been shown to be a prognostic predictive factor for OS, DFS, and recurrence in patients with HCC. Patients with a lower preoperative PNI have a worse prognosis. The improvement of preoperative nutritional conditions by means of early interventions may contribute to better outcomes.

12. Onodera T, Goseki N, Kosaki G. (1984) [Prognostic nutritional index in

Acknowledgment This work was supported by the National Natural Science Foundation of China (No.81600452).

hepatocellular carcinoma after curative hepatectomy. Hepatobiliary

gastrointestinal surgery of malnourished cancer patients]. Nihon Geka Gakkai Zasshi 85:1001–1005. 13. Zhang C, Wang H, Ning Z, Xu L, Zhuang L, Wang P et al. (2016) Prognostic nutritional index serves as a predictive marker of survival and associates with systemic inflammatory response in metastatic intrahepatic cholangiocarcinoma. Onco Targets Ther 9:6417–6423. 14. Yang Y, Gao P, Chen X, Song Y, Shi J, Zhao J et al. (2016) Prognostic significance of preoperative prognostic nutritional index in colorectal cancer: results from a retrospective cohort study and a meta-analysis. Oncotarget 7:58543–58552. 15. Liu Y, Wang ZX, Cao Y, Zhang G, Chen WB, Jiang CP. (2016) Preoperative inflammation-based markers predict early and late recurrence of Pancreat Dis Int 15:266–274. 16. Yang Y, Gao P, Song Y, Sun J, Chen X, Zhao J et al. (2016) The prognostic nutritional index is a predictive indicator of prognosis and

Conflicts of interest The authors have no funding or conflicts of interest to disclose.

postoperative complications in gastric cancer: a meta-analysis. Eur J Surg Oncol 42:1176–1182. 17. Okamura Y, Sugiura T, Ito T, Yamamoto Y, Ashida R, Uesaka K. (2017)

References

The optimal cut-off value of the preoperative prognostic nutritional

1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. (2015) Global cancer statistics, 2012. CA Cancer J Clin 65:87–108.

index for the survival differs according to the TNM stage in hepatocellular carcinoma. Surg Today 47:986–993.

2. Pang Q, Zhou L, Qu K, Cui RX, Jin H, Liu HC. (2018) Validation of

18. Zhao Y, Xu P, Kang H, Lin S, Wang M, Yang P et al. (2016) Prognostic

inflammation-based prognostic models in patients with hepatitis B-

nutritional index as a prognostic biomarker for survival in digestive

associated hepatocellular carcinoma: a retrospective observational study. Eur J Gastroenterol Hepatol 30:60–70.

system carcinomas. Oncotarget 7:86573–86583. 19. Wang Q, Blank S, Fiel MI, Kadri H, Luan W, Warren L et al. (2015) The

3. Pang Q, Qu K, Zhang JY, Song SD, Liu SS, Tai MH et al. (2015) The

severity of liver fibrosis influences the prognostic value of inflammation-

prognostic value of platelet count in patients with hepatocellular car-

based scores in hepatitis B-associated hepatocellular carcinoma. Ann

cinoma: a systematic review and meta-analysis. Medicine (Baltimore) 94:e1431.

Surg Oncol 22(Suppl. 3):S1125–S1132. 20. He CB, Lin XJ. (2017) Inflammation scores predict the survival of pa-

4. Ji F, Liang Y, Fu S, Chen D, Cai X, Li S et al. (2017) Prognostic value of

tients with hepatocellular carcinoma who were treated with transarterial

combined preoperative prognostic nutritional index and body mass

chemoembolization and recombinant human type-5 adenovirus H101.

index in HCC after hepatectomy. HPB 19:695–705.

PLoS One 12:e0174769.

5. Bhattacharya N, Yuan R, Prestwood TR, Penny HL, DiMaio MA,

21. Goh BK, Kam JH, Lee SY, Chan CY, Allen JC, Jeyaraj P et al. (2016)

Reticker-Flynn NE et al. (2016) Normalizing microbiota-induced retinoic

Significance of neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte

acid deficiency stimulates protective CD8(+) T cell-mediated immunity

ratio and prognostic nutrition index as preoperative predictors of early

in colorectal cancer. Immunity 45:641–655.

mortality after liver resection for huge (>/=10 cm) hepatocellular carci-

6. Nishida N, Kudo M. (2017) Immunological microenvironment of hepatocellular carcinoma and its clinical implication. Oncology 92(Suppl. 1): 40–49.

noma. J Surg Oncol 113:621–627. 22. Higashi T, Hayashi H, Kaida T, Arima K, Takeyama H, Taki K et al. (2015) Prognostic impact of visceral fat amount and branched-chain amino

7. Schutte K, Schulz C, Malfertheiner P. (2016) Nutrition and hepatocellular cancer. Gastrointest Tumors 2:188–194.

acids (BCAA) in hepatocellular carcinoma. Ann Surg Oncol 22(Suppl. 3): S1041–S1047.

8. Lealdini V, Trufelli DC, da Silva FB, Normando SR, Camargo EW,

23. Kinoshita A, Onoda H, Imai N, Iwaku A, Oishi M, Fushiya N et al. (2012)

Matos LL et al. (2015) Applicability of modified Glasgow Prognostic

Comparison of the prognostic value of inflammation-based prognostic

Score in the assessment of elderly patients with cancer: a pilot study.

scores in patients with hepatocellular carcinoma. Br J Cancer 107:

J Geriatr Oncol 6:479–483.

988–993.

9. Templeton AJ, Ace O, McNamara MG, Al-Mubarak M, Vera-Badillo FE,

24. Liu C, Li L, Lu WS, Du H, Yan LN, Yang JY et al. (2017) Neutrophil-

Hermanns T et al. (2014) Prognostic role of platelet to lymphocyte ratio

lymphocyte ratio plus prognostic nutritional index predicts the outcomes

in solid tumors: a systematic review and meta-analysis. Cancer

of patients with unresectable hepatocellular carcinoma after transarterial

Epidemiol Biomarkers Prev 23:1204–1212.

chemoembolization. Sci Rep 7:13873.

10. Zheng J, Cai J, Li H, Zeng K, He L, Fu H et al. (2017) Neutrophil to

25. Wu SJ, Lin YX, Ye H, Li FY, Xiong XZ, Cheng NS. (2016) Lymphocyte to

lymphocyte ratio and platelet to lymphocyte ratio as prognostic pre-

monocyte ratio and prognostic nutritional index predict survival out-

dictors for hepatocellular carcinoma patients with various treatments: a

comes of hepatitis B virus-associated hepatocellular carcinoma pa-

meta-analysis and systematic review. Cell Physiol Biochem 44:967–981.

tients after curative hepatectomy. J Surg Oncol 114:202–210.

11. Chen H, Hu N, Chang P, Kang T, Han S, Lu Y et al. (2017) Modified

26. Okamura Y, Ashida R, Ito T, Sugiura T, Mori K, Uesaka K. (2015) Pre-

Glasgow prognostic score might be a prognostic factor for hepatocel-

operative neutrophil to lymphocyte ratio and prognostic nutritional index

lular carcinoma: a meta-analysis. Panminerva Med 59:302–307.

HPB 2018, -, 1–8

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019

HPB

8

predict overall survival after hepatectomy for hepatocellular carcinoma.

38. Chen Z, Shao Y, Wang K, Cao W, Xiong Y, Wu R et al. (2016) Prognostic role of pretreatment serum albumin in renal cell carcinoma: a systematic

World J Surg 39:1501–1509. 27. Chan AW, Chan SL, Wong GL, Wong VW, Chong CC, Lai PB et al. (2015) Prognostic nutritional index (PNI) predicts tumor recurrence of very early/early stage hepatocellular carcinoma after surgical resection.

review and meta-analysis. OncoTargets Ther 9:6701–6710. 39. Spinella R, Sawhney R, Jalan R. (2016) Albumin in chronic liver disease: structure, functions and therapeutic implications. Hepatol Int 10:124–132. 40. Carr BI, Guerra V. (2017) Serum albumin levels in relation to tumor

Ann Surg Oncol 22:4138–4148. 28. Pinato DJ, North BV, Sharma R. (2012) A novel, externally validated inflammation-based prognostic algorithm in hepatocellular carcinoma: the prognostic nutritional index (PNI). Br J Cancer 106:1439–1445. 29. Yang L, Xia L, Wang Y, Hong S, Chen H, Liang S et al. (2016) Low prognostic nutritional index (PNI) predicts unfavorable distant metastasis-free survival in nasopharyngeal carcinoma: a propensity score-matched analysis. PLoS

parameters in hepatocellular carcinoma patients. Int J Biol Markers 32: e391–e396. 41. Toyoda H, Kumada T, Tada T, Sone Y, Kaneoka Y, Maeda A. (2015) Tumor markers for hepatocellular carcinoma: simple and significant predictors of outcome in patients with HCC. Liver Cancer 4:126–136. 42. Kim H, Ahn SW, Hong SK, Yoon KC, Kim HS, Choi YR et al. (2017) Survival benefit of liver resection for Barcelona Clinic Liver Cancer stage

One 11:e0158853. 30. Zhang Z, Pereira SL, Luo M, Matheson EM. (2017) Evaluation of blood biomarkers associated with risk of malnutrition in older adults: a systematic review and meta-analysis. Nutrients 9. 31. Saha SK, Lee SB, Won J, Choi HY, Kim K, Yang GM et al. (2017) Correlation between oxidative stress, nutrition, and cancer initiation. Int

B hepatocellular carcinoma. Br J Surg 104:1045–1052. 43. Deng Y, Pang Q, Miao RC, Chen W, Zhou YY, Bi JB et al. (2016) Prognostic significance of pretreatment albumin/globulin ratio in patients with hepatocellular carcinoma. OncoTargets Ther 9:5317–5328. 44. Mendes F, Domingues C, Rodrigues-Santos P, Abrantes AM, Goncalves AC, Estrela J et al. (2016) The role of immune system

J Mol Sci 18. 32. Wang Z, Li Z, Ye Y, Xie L, Li W. (2016) Oxidative stress and liver cancer: etiology and therapeutic targets. Oxid Med Cell Longev 2016,

exhaustion on cancer cell escape and anti-tumor immune induction after irradiation. Biochim Biophys Acta 1865:168–175. 45. Man YG, Stojadinovic A, Mason J, Avital I, Bilchik A, Bruecher B et al.

7891574. 33. Cardin R, Piciocchi M, Bortolami M, Kotsafti A, Barzon L, Lavezzo E et al. (2014) Oxidative damage in the progression of chronic liver disease to hepatocellular carcinoma: an intricate pathway. World J

(2013) Tumor-infiltrating immune cells promoting tumor invasion and metastasis: existing theories. J Cancer 4:84–95. 46. Wang K, Nie X, Rong Z, Fan T, Li J, Wang X et al. (2017) B lymphocytes repress hepatic tumorigenesis but not development in Hras12V trans-

Gastroenterol 20:3078–3086. 34. Hunsche C, Hernandez O, Gheorghe A, Diaz LE, Marcos A, De la Fuente M.

genic mice. Int J Cancer 141:1201–1214.

(2018) Immune dysfunction and increased oxidative stress state in diet-

47. Antony PA, Piccirillo CA, Akpinarli A, Finkelstein SE, Speiss PJ,

induced obese mice are reverted by nutritional supplementation with

Surman DR et al. (2005) CD8+ T cell immunity against a tumor/self-

monounsaturated and n-3 polyunsaturated fatty acids. Eur J Nutr 57:

antigen is augmented by CD4+ T helper cells and hindered by naturally occurring T regulatory cells. J Immunol 174:2591–2601.

1123–1135. 35. Liu J, Dai Y, Zhou F, Long Z, Li Y, Liu B et al. (2016) The prognostic role

48. Ma W, Wu L, Zhou F, Hong Z, Yuan Y, Liu Z. (2017) T cell-associated

of preoperative serum albumin/globulin ratio in patients with bladder

immunotherapy for hepatocellular carcinoma. Cell Physiol Biochem 41:

urothelial carcinoma undergoing radical cystectomy. Urol Oncol 34:

609–622. 49. Garcia-Rodriguez MT, Lopez-Calvino B, Pinon-Villar MDC, Otero-

484.e1–484.e8. 36. Tokunaga R, Sakamoto Y, Nakagawa S, Miyamoto Y, Yoshida N, Oki E

Ferreiro A, Suarez-Lopez F, Gomez-Gutierrez M et al. (2017) Concor-

et al. (2015) Prognostic nutritional index predicts severe complications,

dance among methods of nutritional assessment in patients included on

recurrence, and poor prognosis in patients with colorectal cancer un-

the waiting list for liver transplantation. J Epidemiol 27:469–475.

dergoing primary tumor resection. Dis Colon Rectum 58:1048–1057. 37. Gupta D, Lis CG. (2010) Pretreatment serum albumin as a predictor of

Appendix A. Supplementary data

cancer survival: a systematic review of the epidemiological literature.

Supplementary data related to this article can be found at https://doi.org/10.

Nutr J 9:69.

1016/j.hpb.2018.03.019.

HPB 2018, -, 1–8

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Man Z, et al., Prognostic significance of preoperative prognostic nutritional index in hepatocellular carcinoma: a metaanalysis, HPB (2018), https://doi.org/10.1016/j.hpb.2018.03.019