Med Oncol (2014) 31:883 DOI 10.1007/s12032-014-0883-x
ORIGINAL PAPER
The inflammation-based scores to predict prognosis of patients with hepatocellular carcinoma after hepatectomy Junting Huang • Li Xu • Yaoling Luo • Fengying He • Yaojun Zhang • Minshan Chen
Received: 22 December 2013 / Accepted: 3 February 2014 / Published online: 18 February 2014 Ó Springer Science+Business Media New York 2014
Abstract The aims of this study were to compare the prognostic ability of inflammation-based prognostic scores including the Glasgow Prognostic Score (GPS), the modified Glasgow Prognostic Score (mGPS), neutrophil to lymphocyte ratio, prognostic index, and prognostic nutritional index (PNI) for patients with hepatocellular carcinoma (HCC) undergoing hepatectomy, and to propose the combination of staging systems and inflammation scores to improve the prognostic power. Data for 349 patients who underwent hepatectomy as initial treatment for HCC between 2008 and 2009 were retrieved from a prospective database. The association of inflammation scores with clinicopathological variables and overall survival (OS) was analyzed, and the concordance index (C-index) was calculated to compare the predictive ability of each inflammation scores and staging systems including Barcelona Clinic Liver Cancer (BCLC) and Cancer of the Liver Italian Program (CLIP) scores. The median follow-up period was 39 months, the 1, 2, and 3 year OS was 75.4, Junting Huang and Li Xu have contributed equally to this work. J. Huang L. Xu F. He Y. Zhang M. Chen (&) Department of Hepatobiliary Surgery, Sun Yat-sen University Cancer Center, 651 Dongfeng Road East, Guangzhou 510060, China e-mail:
[email protected] J. Huang L. Xu Y. Luo F. He Y. Zhang M. Chen State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou 510060, China Y. Luo Department of Clinical Laboratory, Sun Yat-sen University Cancer Center, 651 Dongfeng Road East, Guangzhou 510060, China
67.0, and 59.0 %, respectively, and the median OS was 39 months. All inflammation scores, except PNI, were associated with tumor size, major/microvascular invasion and clinical stages, and the GPS and mGPS had a higher C-index (0.608). Multivariate analysis showed that the GPS, BCLC, and CLIP were independently associated with OS. The combined GPS and CLIP (C-index = 0.705) were superior to CLIP alone (C-index = 0.686) or the GPS alone in prognostic ability. The prognostic ability of the GPS is superior to other inflammation scores for patients undergoing hepatectomy as initial treatment for HCC. Combining GPS and CLIP improved the prognostic power. Keywords Inflammation-based prognostic score The Glasgow Prognostic Score Hepatocellular carcinoma Hepatectomy Staging system
Introduction Hepatocellular carcinoma (HCC) is now the fifth most common cancer worldwide and the third most frequent cause of death of cancer [1]. More importantly, recent evidence has shown that both the incidence and mortality rates of HCC are rising in North America and Europe [2, 3]. Although there are various staging systems, including the American Joint Committee on Cancer (AJCC) staging system [4], Barcelona Clinic Liver Cancer (BCLC) [5], the Okuda classification [6], Cancer of the Liver Italian Program (CLIP) [7], the Japan Integrated Staging (JIS) score [8], and the Chinese University Prognostic Index (CUPI) [9], there is no worldwide consensus on which is the best system in staging and predicting the prognosis of patients with HCC. The BCLC is considered to be the most widely accepted staging system [10], since it not only considers prognostic stratification but
123
883 Page 2 of 8
Med Oncol (2014) 31:883
Table 1 Inflammation-based prognostic scores Scoring systems
Score
with HCC undergoing hepatectomy as initial treatment, and to validate the combination of staging system and inflammation score to improve the prognostic power.
The Glasgow Prognostic Score (GPS) CRP (B10 mg/L) and albumin (C35 g/L)
0
CRP (B10 mg/L) and albumin (\35 g/L)
1
CRP ([10 mg/L) and albumin (C35 g/L)
1
CRP ([10 mg/L) and albumin (\35 g/L)
2
Patients
CRP (B10 mg/L) and albumin (C35 g/L)
0
CRP (B10 mg/L) and albumin (\35 g/L)
0
CRP ([10 mg/L)
1
Patients who underwent hepatectomy as initial treatment for HCC from January 2008 to December 2009 at the Department of Hepatobiliary Surgery, Sun Yat-Sen University Cancer Canter (Guangzhou, China) were identified from our prospective database. A total of consecutive 349 patients who met all of the following criteria were included in present study as follows: (1) no previous treatment for HCC before surgery; (2) histologic confirmation of HCC; (3) R0 resection; (4) no lymph node or extrahepatic metastasis. The research was approved by the institutional review board (IRB) of Sun Yat-sen University Cancer Center, and written informed consent was obtained. Patients’ baseline characteristics were summarized in Table 2. There were 319 male (91.4 %) and 30 female (8.6 %) with a median age of 50 years (range 13–78 years) at the time of surgery. The majority of our patients had a good liver functional reserve with Child-Pugh A (96.0 %). Most of patients (90.8 %, 317/349) tested positive for hepatitis B virus (HBV), and all patients tested negative for hepatitis C virus (HCV) and human immunodeficiency virus (HIV). Hepatectomy was carried out by using the techniques we described previously [17].
Patients and methods
The modified Glasgow Prognostic Score (mGPS)
CRP ([10 mg/L) and albumin (\35 g/L)
2
Neutrophil lymphocyte ratio (NLR) Neutrophil count:lymphocyte count \ 3:1
0
Neutrophil count:lymphocyte count C 3:1
1
Plt lymphocyte ratio (PLR) Plt count:lymphocyte count \ 150:1
0
Plt count:lymphocyte count C 150:1
1
Plt count:lymphocyte count [ 300:1
2
Prognostic Index (PI) CRP (B10 mg/L) and WBC (B11 9 109/L) 9
CRP (B10 mg/L) and WBC ([11 9 10 /L)
0 1
CRP ([10 mg/L) and WBC (B11 9 109/L)
1
CRP ([10 mg/L) and WBC ([11 9 109/L)
2
Prognostic Nutritional Index (PNI) Albumin (g/L) 9 total lymphocyte count 9 109/L C 45
0
Albumin (g/L) 9 total lymphocyte count 9 109/L \ 45
1
CRP C-reactive protein, WBC white cell count
also therapeutic allocation, but the superiority of BCLC in terms of prognostic accuracy has been controversial [11]. The CLIP appears to be better in predicting survival in patients with advanced HCC, but its prognostic ability in resectable disease is still controversial [12]. There is increasing evidence supporting the role of systemic inflammation as a predictor of outcome in several human cancers including HCC [13–16]. The pathogenesis of HCC is based on inflammation. As the last and most redoubtable clinical consequence of cirrhosis, the onset of HCC is related to a myriad of pro-inflammatory stimuli, triggered by well-recognized noxae such as infection by hepatotropic viruses, iron or copper accumulation or ethanol consumption. So, we hypothesize that the combination of clinical staging system and inflammation scores might improve their prognostic power for HCC. The aims of the present study were to compare the prognostic value of inflammation-based prognostic scores [the Glasgow Prognostic Score (GPS), the modified Glasgow Prognostic Score (mGPS), neutrophil to lymphocyte ratio (NLR), platelet lymphocyte ratio (PLR), Prognostic Index (PI), and Prognostic Nutritional Index (PNI), Table 1] for patients
123
Follow-up Patients were followed up 3 monthly for the first 2 years and then 6 monthly thereafter with physical examination, blood tests for AFP and liver function, and contrast abdominal computed tomography (CT). Chest radiography was done every 6 months to detect lung metastasis. When metastasis was suspected, CT chest, bone scintigraphy, positron emission tomography (PET), and biopsy if indicated were also performed to confirm metastasis and/or recurrence. Causes of death and sites of recurrence were determined from death certificates, medical interviews, and radiological findings. Overall survival (OS) was defined as the interval between surgery and time of either death or last follow-up. The last follow-up date for patients still alive was December 2012. Statistical analyses Comparisons between two groups were done using the student’s t test for continuous data and the chi-square
Med Oncol (2014) 31:883 Table 2 Baseline characteristics and analyses of prognostic factors for overall survival in 349 patients undergoing hepatectomy for hepatocellular carcinoma
Page 3 of 8 883
Variables
n = 349
Age (years)
50 (13–78)
0.932
Gender (M/F)
319/30
0.888
Diameter of largest lesion (cm) Number of lesions (1/[1)
5.0 (1.0–18.0) 252/97
0.002
Major Vascular invasion (absent/present)
304/45
\0.001
Micro vascular invasion (absent/present)
224/125
\0.001
WBC (9 109/L)
6.2 (3.4–14.5)
0.377
Neutrophil count (9 109/ L)
3.5 (1.2–10.8)
0.058
Lymphocyte count (9 109/ L)
1.9 (0.2–3.7)
0.686
CRP (mg/L)
2.0 (0.8–271.0) 9
AFP a-fetoprotein level, ALT alanine aminotransferase, ALP alkaline phosphatase, AST aspartate aminotransferase, CRP C-reactive protein, GPS Glasgow Prognostic Score, HBsAg hepatitis B surface antigen, HCVAb hepatitis C antibody, NLR neutrophil lymphocyte ratio, PI Prognostic Index, PLR platelet lymphocyte ratio, PNI Prognostic Nutritional Index, WBC white blood cell count
Univariate analysis P value
170 (28–449) 37.6 (8.5–418.2)
AST (l/L)
35.0 (13.4–256.3)
Albumin (g/L)
41.9 (30.3–51.0)
0.004
Total serum bilirubin (lmol/L)
13.0 (4.1–276.4)
\0.001
77.4 (28.0–307.2) 357.3 (0.61–1210000)
HBsAg (positive/negative
317/32
0.971 0.345 \0.001
0.017 \0.001 0.520
Child–Pugh grade (A/B)
335/14
GPS (0/1/2)
269/70/10
\0.001
Modified GPS (0/1/2)
278/61/10
\0.001
NLR (0/1)
283/66
\0.001
PLR (0/1/2)
301/42/6
\0.001
PI (0/1/2)
272/69/8
\0.001
PNI (0/1)
320/29
CLIP score (0/1/2/3/4)
120/135/52/24/18
BCLC stage (A/B/C)
174/124/51
test for categorical data. The OS was calculated by Kaplan–Meier method and compared by log-rank test. The prognostic varieties in predicting OS were assessed by multivariate Cox proportional hazards regression analysis. Results were given as mean ± SD. All statistical tests were two-sided, and a significant difference was considered when P \ 0.05. All the statistical except concordance index (C-index) analyses were performed using the SPSS 19.0 statistical software (SPSS Company, Chicago, Illinois, USA). The C-index was adjusted using R with the rms package of Dr. Frank Harrell and assessment being done on the same data via comparison with 150 bootstrap samples.
P value
\0.001
PLT Count (9 10 /L)
AFP (ng/ml)
Hazard ratio (95 % CI)
\0.001
ALT (l/L)
ALP (IU/L)
Multivariate analysis
0.045 1.633 (1.226–2.174)
0.001
\0.001
1.523 (1.279–1.813)
\0.001
\0.001
1.442 (1.090–1.908)
0.010
0.003
Results Inflammatory scores and their relationships with clinicopathological characteristics Among the 349 patients, 71 patients (20.3 %) had an elevated CRP level ([10 g/L), 19 (5.4 %) had hypoalbuminemia (\35 g/L), and 10 (2.9 %) had both elevated CRP level and hypoalbuminemia. Forty-eight patients (13.8 %) had PLR C 150; 66 patients had NLR C 3 (18.9 %); 28 patients (8.0 %) had PNI [ 45; and 77 patients (22.1 %) were allocated to PI 1 or 2 (Table 2).
123
883 Page 4 of 8
123
Med Oncol (2014) 31:883
Med Oncol (2014) 31:883 b Fig. 1 Kaplan–Meier survival curves for overall survival in 349
patients undergoing hepatectomy for hepatocellular carcinoma. a GPS, b modified GPS, c NLR, d PLR, e PI, f PNI, g CLIP, and h BCLC
The relationships between inflammatory scores and clinicopathological features were analyzed. All these inflammatory scores, except PNI, were associated with tumor size, major/microvascular invasion, and clinical stages (including CLIP score and BCLC stage). The GPS was associated with tumor size (P \ 0.001), tumor number (P = 0.05), major vascular invasion (P \ 0.001), microvascular invasion (P = 0.011), AST (P \ 0.001), TBIL (P = 0.021), ALP (P = 0.004), AFP (P = 0.001), ChildPugh grade (P = 0.036), CLIP score (P \ 0.001), and BCLC stage (P \ 0.001). Similar association was identified for mGPS. Survival and prognostic factors The median follow-up period was 39 months (range 3–59 months), and at the time of analysis, 153 patients died. The 1, 2, and 3 year OS was 75.4, 67.0, and 59.0 %, respectively, and the median OS was 39 months. The univariate and multivariate analyses of prognostic factors for OS in patients undergoing hepatectomy for HCC are shown in Table 2. In univariate analysis, tumor size (P \ 0.001), tumor number (P = 0.002), major vascular invasion (P \ 0.001), microvascular invasion (P \ 0.001), CRP (P \ 0.001), AST (P \ 0.001), Albumin (P = 0.004), TBIL (P \ 0.001), ALP (P = 0.017), AFP (P \ 0.001), all inflammatory scores including GPS (P \ 0.001), mGPS (P \ 0.001), NLR (P \ 0.001), PLR (P \ 0.001), PI (P \ 0.001), and PNI (P = 0.003) were associated with OS.
Page 5 of 8 883
The BCLC stage (P = 0.002) and CLIP scores (P \ 0.001) were also confirmed as significant predictors of OS (Fig. 1). Multivariate analysis showed that GPS (HR 1.633, 95 % CI 1.226–2.174, P \ 0.001), CLIP (HR 1.523, 95 % CI 1.279–1.813, P \ 0.001), and BCLC stage (HR 1.442, 95 %CI 1.090–1.908, P = 0.010) were significant prognostic factors for OS. Comparison between inflammatory prognostic scores and staging systems The prognostic power of inflammatory scores and staging systems was compared by means of Harrell’s C-index. As it was shown in Table 3, the GPS and mGPS had a higher C-index (0.608) compared to other scores. However, both the BCLC stage and CLIP had higher C-index (0.684 and 0.686, respectively) in comparison with inflammatory scores. Table 3 Comparison of the C-index between inflammatory prognostic scores and staging systems Prognostic system
C-index
95 % CI
GPS
0.608
0.572–0.644
Modified GPS
0.608
0.572–0.643
NLR
0.573
0.538–0.607
PLR
0.552
0.522–0.581
PI
0.598
0.563–0.634
PNI
0.532
0.507–0.558
CLIP
0.686
0.645–0.728
BCLC
0.684
0.644–0.723
CI confidence interval, C-index concordance index, GPS Glasgow Prognostic Score, NLR neutrophil lymphocyte ratio, PI Prognostic Index, PLR platelet lymphocyte ratio, PNI Prognostic Nutritional Index
b Fig. 1 continued
123
883 Page 6 of 8 Table 4 Combined score of the GPS and CLIP score
Med Oncol (2014) 31:883
Variables
Score 0
1
2
Child–Pugh stage
A
B
C
Tumor morphology
Uninodular and extension B 50 %
Multinodular and extension B 50 %
Multinodular or extension [ 50 %
AFP (ng/ml) Portal vein thrombosis
\400 No
C400 Yes
GPS
CRP B 10 mg/L and albumin C 35 g/L
(CRP B 10 mg/L and albumin \ 35 g/ L) or (CRP [ 10 mg/L and albumin C 35 g/L)
Fig. 2 Kaplan–Meier survival curves of combined score of the GPS and CLIP score for overall survival in 349 patients undergoing hepatectomy for HCC
Combined score of the GPS and CLIP score We proposed a combined score of the GPS and CLIP score, as it was shown in Table 4. The combined score had a higher C-index of 0.705 (95 % CI 0.665–0.746) when it compared to CLIP alone (C-index = 0.686, 95 % CI 0.645–0.728) or the GPS alone (C-index = 0.608, 95 % CI 0.572–0.644). And the survival curve showed that the combined score divides patients into subgroups more accurately (Fig. 2).
Discussion In present study, we compared the prognostic power of six inflammation-based prognostic scores, including GPS,
123
CRP [ 10 mg/L and albumin \ 35 g/L
mGPS, NLR, PLR, PI, and PNI, for patients undergoing hepatectomy as initial treatment for HCC, and proposed a combined score of the GPS and CLIP. Our results demonstrated that the GPS is an independent predictor of OS and superior to other scores. The combination of the GPS and CLIP improves the prognostic power. Several studies had shown that the inflammatory scores are associated with prognosis in patients with HCC. Ishizuka et al. [14, 18] demonstrated that the GPS is an important predictor of OS after surgical treatment for HCC. Gomez et al. [19] reported that NLR is an adverse predictor of disease-free and overall survival after curative resection for HCC. Similar results were also reported by others [20– 22]. Pinato et al. [23] indicated that PNI is an independent predictor of poor OS in patients with HCC. Our results demonstrated that elevated inflammatory scores associated with bigger tumor size, vascular invasion, advanced stages (including CLIP and BCLC stage), and poor survival, suggesting the presence of a systemic inflammatory response, are predictive of a more aggressive clinical phenotype. However, few studies compared each other inflammation-based prognostic scores. In the present study, although univariate analysis showed that GPS, mGPS, NLR, PLR, PNI, and the PI were significantly associated with OS for patients undergoing hepatectomy for HCC, comparison between inflammatory scores using Harrell’s concordance index method demonstrated that the GPS had a higher C-index than other scores, and the multivariate analysis also showed that only the GPS is an independent predictor of OS. Although based on different patient population, Kinoshita et al. [24] compared the prognostic value of these inflammatory scores in 150 patients with newly diagnosed HCC in various stages of disease and different liver functional status treated with various methods, and their results demonstrated that the GPS was an independent marker of poor prognosis and was superior to other scores in terms of prognostic ability.
Med Oncol (2014) 31:883
There are various staging systems for HCC, but none of these previous prognostic indexes are considered ideal and, despite the increasing number of published comparative studies, there is no consensus on the optimal system that should be utilized. As sustained inflammation acts as one of the main factors thought to promote development of HCC within the chronically injured liver parenchyma, we hypothesize that the combination of clinical staging system and inflammation scores might improve their prognostic power. In our study, we proposed a combined score of the GPS and CLIP score, which simply added the GPS scores into CLIP scores. Our results demonstrated that combined score had a higher C-index of 0.705 (95 % CI 0.665–0.746) when it compared to the CLIP alone (C-index = 0.686, 95 % CI 0.645–0.728) or the GPS alone (C-index = 0.608, 95 % CI 0.572–0.644), and the survival curves showed that the combined score divides patients into subgroups more accurately. We validated the concept that the combination of the GPS with the CLIP score may increase the predictive ability, suggesting the additive value of systemic inflammation to an accurate prognostic assessment in HCC. Although retrospective, the data we have presented were based on a long duration of follow-ups (median 39 months) of a large representative sample of patients who had undergone hepatic resection for HCC, as well as a systemic staging workup using sensitive and specific imaging technologies and histological confirmation in order for us to provide a valid base to evaluate the prognostic value of inflammatory scores. However, limitations in present study should be commented. It is a single institution study of a fairly homogenous population. Patient population is biased due to high prevalence of HBV infection (90.8 %). Whether these results can be applied to Western populations wherein HCV, NASH, and other etiologies of liver disease predominate requires further study and comment. Conclusion The prognostic ability of the GPS is superior to the other inflammation-based prognostic scores for patients undergoing hepatectomy for HCC. Combining the GPS and CLIP had a better prognostic power than the CLIP alone. Conflict of interest interests exist.
The authors have declared that no competing
References 1. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin. 2011;61(2):69–90. 2. Kew MC. Hepatocellular carcinoma in developing countries: prevention, diagnosis and treatment. World J Hepatol. 2012;4(3): 99–104.
Page 7 of 8 883 3. Lodato F, Mazzella G, Festi D, et al. Hepatocellular carcinoma prevention: a worldwide emergence between the opulence of developed countries and the economic constraints of developing nations. World J Gastroenterol. 2006;12(45):7239–49. 4. Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol. 2010;17(6):1471–4. 5. Forner A, Reig ME, de Lope CR, Bruix J. Current strategy for staging and treatment: the BCLC update and future prospects. Semin Liver Dis. 2010;30(1):61–74. 6. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients. Cancer. 1985;56(4):918–28. 7. Llovet JM, Bruix J. Prospective validation of the Cancer of the Liver Italian Program (CLIP) score: a new prognostic system for patients with cirrhosis and hepatocellular carcinoma. Hepatology. 2000;32(3):679–80. 8. Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology. 2000;32(6):1224–9. 9. Chan SL, Mo FK, Johnson PJ, et al. Prospective validation of the Chinese University Prognostic Index and comparison with other staging systems for hepatocellular carcinoma in an Asian population. J Gastroenterol Hepatol. 2011;26(2):340–7. 10. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42(5):1208–36. 11. Vitale A, Saracino E, Boccagni P, et al. Validation of the BCLC prognostic system in surgical hepatocellular cancer patients. Transpl Proc. 2009;41(4):1260–3. 12. Huitzil-Melendez FD, Capanu M, O’Reilly EM, et al. Advanced hepatocellular carcinoma: which staging systems best predict prognosis? J Clin Oncol. 2010;28(17):2889–95. 13. Forrest LM, McMillan DC, McArdle CS, et al. A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-smallcell lung cancer. Br J Cancer. 2005;92(10):1834–6. 14. Ishizuka M, Kubota K, Kita J, et al. Usefulness of a modified inflammation-based prognostic system for predicting postoperative mortality of patients undergoing surgery for primary hepatocellular carcinoma. J Surg Oncol. 2011;103(8):801–6. 15. Hashimoto K, Ikeda Y, Korenaga D, et al. The impact of preoperative serum C-reactive protein on the prognosis of patients with hepatocellular carcinoma. Cancer. 2005;103(9):1856–64. 16. Kinoshita A, Onoda H, Takano K, et al. Pretreatment serum C-reactive protein level predicts poor prognosis in patients with hepatocellular carcinoma. Med Oncol. 2012;29(4):2800–8. 17. Shi J, Lai EC, Li N, et al. Surgical treatment of hepatocellular carcinoma with portal vein tumor thrombus. Ann Surg Oncol. 2010;17(8):2073–80. 18. Ishizuka M, Kubota K, Kita J, et al. Impact of an inflammationbased prognostic system on patients undergoing surgery for hepatocellular carcinoma: a retrospective study of 398 Japanese patients. Am J Surg. 2012;203(1):101–6. 19. Gomez D, Farid S, Malik HZ, et al. Preoperative neutrophil-tolymphocyte ratio as a prognostic predictor after curative resection for hepatocellular carcinoma. World J Surg. 2008;32(8):1757–62. 20. Halazun KJ, Hardy MA, Rana AA, et al. Negative impact of neutrophil–lymphocyte ratio on outcome after liver transplantation for hepatocellular carcinoma. Ann Surg. 2009;250(1):141–51. 21. Huang ZL, Luo J, Chen MS, et al. Blood neutrophil-to-lymphocyte ratio predicts survival in patients with unresectable hepatocellular carcinoma undergoing transarterial chemoembolization. J Vasc Interv Radiol. 2011;22(5):702–9. 22. Chen TM, Lin CC, Huang PT, Wen CF. Neutrophil-to-lymphocyte ratio associated with mortality in early hepatocellular
123
883 Page 8 of 8 carcinoma patients after radiofrequency ablation. J Gastroenterol Hepatol. 2012;27(3):553–61. 23. Pinato DJ, North BV, Sharma R. A novel, externally validated inflammation-based prognostic algorithm in hepatocellular carcinoma: the prognostic nutritional index (PNI). Br J Cancer. 2012;106(8):1439–45.
123
Med Oncol (2014) 31:883 24. Kinoshita A, Onoda H, Imai N, et al. Comparison of the prognostic value of inflammation-based prognostic scores in patients with hepatocellular carcinoma. Br J Cancer. 2012; 107(6):988–93.
