Original Article Oncol Res Treat 2014;37:192–197 DOI: 10.1159/000361082
Received: August 8, 2013 Accepted: February 27, 2014 Published online: March 21, 2014
The Glasgow Prognostic Score Is an Independent Prognostic Predictor of Hepatocellular Carcinoma Following Radical Resection Qun-Xiong Pan Jian-Hua Zhang Zi-Jian Su Chong-Ren Wang Shao-Ying Ke Department of Oncosurgery, Affiliated Quanzhou First Hospital of Fujian Medical University, Quanzhou, Fujian, China
Summary Background: Some prognostic evaluation systems were developed to postoperatively predict the outcome of hepatocellular carcinoma (HCC) patients, mainly based on the cancer itself and the underlying liver diseases. However, none of these prognostic evaluation systems have so far been universally accepted. A simple and feasible scoring system is still lacking for the prediction of prognosis of HCC patients following resection. We aimed to uncover the correlation between the preoperative Glasgow Prognostic Score (GPS) and the clinical outcome of HCC patients after radical resection. Methods: The patients were separated into 3 subgroups on the basis of their GPSs. The prognostic significance of the GPS in the patient cohort was evaluated by survival analysis. Results: On univariate analysis, the levels of C-reactive protein and albumin, the Child-Pugh class, vascular invasion, tumor number, tumor size, the tumor/ node/metastasis (TNM) stage, and the GPS were associated with overall survival and time to recurrence of HCC patients after radical resection. On multivariate analysis, the tumor size, albumin level, and GPS were independently associated with the outcome of HCC postoperatively. Conclusion: The GPS is an independent biomarker for prognostic prediction of HCC following radical resection.
Q.-X.P. and J.-H.Z. contributed equally to this work.
© 2014 S. Karger GmbH, Freiburg 2296-5270/14/0374-0192$39.50/0 Fax +49 761 4 52 07 14 [email protected] www.karger.com
Accessible online at: www.karger.com/ort
Introduction Hepatocellular carcinoma (HCC), with rising incidence all over the world, is one of the most prevalent malignancies. The morbidity of HCC is over 500,000 per year worldwide, with more than half of these occurring in China alone. As one of the intractable malignancies, HCC is a major cause of mortality in cancer patients among men worldwide [1]. Until now, surgery is the most effective therapy for HCC, due to the rising rate of early detection and the improvement of surgical technologies [2]. Some prognostic evaluation systems were developed to postoperatively predict the outcome of HCC patients, mainly based on the cancer itself and the underlying liver diseases, such as the tumor/node/metastasis (TNM) classification system of the American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) [3], the Okuda staging system [4], the Cancer of the Liver Italian Program (CLIP) score [5], and the Barcelona Clinic Liver Cancer (BCLC) staging system [6]. However, none of these prognostic evaluation systems have so far been universally accepted, in part due to the complexity of the potential prognostic factors [7]. A simple and feasible scoring system is still lacking for the prediction of prognosis of HCC patients following resection. A role for inflammation in tumorigenesis is now generally accepted, and growing evidence supports the notion that an inflammatory microenvironment is an essential component of all tumors [8]. Inflammatory cytokines, which can be secreted both by inflammatory cells and tumor cells, can actively affect tumor development and progression, partially by inducing immunosuppression in the tumor milieu [9]. C-reactive protein (CRP), a plasma protein synthesized by the liver in response to stimulation by interleukin (IL)-6, is a sensitive and dynamic systemic marker of inflammation [10]. Moreover, hypoalbuminemia plays a major role in the suppression of cell-mediated immune responses [11]. It is reasonable to assume that
Dr. Qun-Xiong Pan Department of Oncosurgery Affiliated Quanzhou First Hospital of Fujian Medical University Quanzhou, Fujian 362000, China [email protected]
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Keywords Hepatocellular carcinoma · Glasgow Prognostic Score · C-reactive protein
the evaluation of the inflammation-based Glasgow Prognostic Score (GPS), derived from standard thresholds of CRP and albumin, can reflect the immunosuppression status of tumor patients. Multiple studies have shown that the GPS could be considered as an independent prognostic index in tumor patients [12–18]. However, until now, the clinical significance of the GPS in HCC, especially for HCC patients with hepatitis B virus (HBV) infection, is still undetermined. The current study aims to uncover the correlation between the preoperative GPS and the clinical outcome of HCC patients after radical resection.
Patients and Methods Patient Selection A cohort of 171 cases was selected retrospectively from those patients who underwent radical resection for HCC at the Department of Surgical Oncology, First Hospital of Quanzhou, Fujian Medical University between June 2001 and September 2007. All patients were managed by the same team of hepatic surgeons; all blood samples were collected the day before surgery. Tumor staging was determined according to the 6th edition of the TNM classification of the AJCC/UICC. Tumor differentiation was defined according to the Edmondson grading system. Data were censored at the last follow-up for living patients or for patients lost to follow-
up. Time to recurrence (TTR) and overall survival (OS) were defined as the interval from primary surgical treatment to the first recurrence or death, respectively. Ethical approval was obtained from the research ethic committee of the First Hospital of Quanzhou; informed consent was obtained from each patient. The inclusion and exclusion criteria of the patient cohorts include (a) having a distinctive pathologic diagnosis of HCC, (b) having no anticancer treatment before liver resection, (c) without distant metastasis and having curative liver resection, (d) having complete clinicopathologic and follow-up data. GPS Evaluation The GPS was constructed as described previously [16, 19, 20]. Briefly, patients with both an elevated CRP level (> 10 mg/l) and hypoalbuminemia (< 35 g/l) were allocated a score of 2. Patients with only 1 of these biochemical abnormalities were allocated a score of 1. Patients with neither of these abnormalities were allocated a score of 0. Statistics Statistical analyses were done by SPSS 13.0 for Windows. The c2 test, Fisher’s exact probability test, and Student’s t-test were used for comparison between groups. The nonparametric Mann-Whitney U test was used when variables in groups were not homogeneous. One-way analysis of variance (ANOVA) was used to compare quantitative data among 3 groups. The cumulative survival time was calculated by the Kaplan-Meier method and analyzed by the log-rank test. Univariate and multivariate analyses were based on the Cox proportional hazard regression model. Multivariate analysis was performed using a Cox regression model in a forward stepwise manner. P < 0.05 was considered statistically significant.
Table 1. Relationship among GPS, clinicopathologic features, and blood loss during hepatectomy Age, years Gender Male Female Child-Pugh class A B CRP, mg/l Alb, g/l HbsAg status Positive Negative Vascular invasion Yes No Tumor sizea, cm Number of tumors 1 ≥2 AFP, ng/ml ALT, U/l AST, U/l Total bilirubin, mmol/l Ascites Negative Positive TNM classification I II III Blood loss, ml
Overall (n = 171)
GPS = 0 (n = 124)
GPS = 1 (n = 29)
GPS = 2 (n = 18)
P
52.70 ± 12.67
51.95 ± 12.70
53.14 ± 12.69
57.11 ± 12.16
0.267
157 14
117 7
24 5
16 2
0.109
144 27 7.17 ± 13.4 37.05 ± 5.33
110 14 2.65 ± 2.52 39.41 ± 3.12
23 6 9.96 ± 8.36 32.29 ± 4.77
11 7 33.81 ± 26.68 28.50 ± 4.48
0.008
148 23
108 16
26 3
14 4
0.481
48 123 5.75 ± 3.13
28 96 5.52 ± 3.01
11 18 6.40 ± 3.54
9 9 6.37 ± 3.24
0.023
134 37 4,130.0 ± 10,926.52 74.2 ± 125.84 81.26 ± 120.28 24.58 ± 26.41
97 27 4,645.83 ± 12,107.70 56.79 ± 63.20 70.40 ± 81.93 23.96 ± 20.62
23 6 3,088.98 ± 7,596.47 137.97 ± 251.56 127.48 ± 231.08 28.89 ± 46.83
14 4 2,253.63 ± 5,536.98 91.44 ± 125.61 81.67 ± 60.43 21.92 ± 15.63
0.990
152 19
112 12
28 1
12 6
0.004
96 56 19 624.15 ± 653.86
77 35 12 548.71 ± 568.68
14 10 5 856.90 ± 830.96
5 11 2 768.90 ± 801.30
0.039
< 0.001 < 0.001
0.270
0.588 0.006 0.070 0.601
0.044
CRP = C-reactive protein, Alb = albumin, HbsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, ALT = alanine aminotransferase, AST = aspartate aminotransferase, TNM = tumor/node/metastasis, GPS = Glasgow Prognostic Score. a Diameter.
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
Oncol Res Treat 2014;37:192–197
193
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Variable
size, tumor number, and TNM classification were also unfavorable predictors for OS.
Patient Profiles The median follow-up time was 41 months (range, 1–76 months; standard deviation (SD), 24.27 months). At the last follow-up (January 02, 2013), 158 patients (92.4%) had recurrence and 143 patients (83.6%) had died of recurrence (n = 131) or cirrhosis-related complications without recurrence (n = 12). Correlation among GPS, Clinicopathologic Features, and Blood Loss The relationships among the GPS, the clinicopathologic features, and blood loss during hepatectomy are shown in table 1. No significant differences were found when age (P = 0.267), gender (P = 0.109), tumor number (P = 0.990), tumor size (P = 0.270), a-fetoprotein (P = 0.588), aspartate aminotransferase (AST) level (P = 0.070), total bilirubin (P = 0.601), and HBV surface antigen (HBsAg) status (P = 0.481) were evaluated between different GPS subgroups. In contrast, an increasing GPS was associated with a higher alanine aminotransferase (ALT) level (P = 0.006), a higher Child-Pugh class (P = 0.008), a higher incidence of ascites (P = 0.004) and vascular invasion (P = 0.023), a higher TNM staging (P = 0.039), and a larger volume of blood loss during resection (P = 0.044). Univariate Analysis of Prognostic Factors The 3-year and 5-year survival rates of patients with a GPS of 0 were 62.1% and 42.7%, respectively. These rates were 48.3% and 13.9%, respectively, in patients with a GPS of 1 and 49.4% and 18.5%, respectively, in patients with a GPS of 2. Because there was no significant difference between the survival of patients with GPSs of 1 and 2 (P = 0.676; fig. 1), we combined the 2 subgroups when evaluating the prognostic value of the GPS. Compared with patients with a GPS of 0, the 3-year and 5-year survival rates of patients with a GPS of 1 or 2 were 48.7% and 15.7%, respectively. A significant difference exists between these 2 groups (P < 0.001; fig. 2A). The clinicopathologic features and blood loss during hepatectomy were also subjected to univariate analysis. As shown in table 2, child-Pugh class, vascular invasion, tumor
Multivariate Analysis of the Prognostic Value of the GPS Factors that showed significance by univariate analysis were used as covariates in a multivariate Cox proportional hazards model. Since over 80% of the patients were HBsAg positive, it might be crucial whether antiviral treatment was initiated or not. We included this option in the multivariate analysis. Multivariate analysis revealed that the GPS was an independent prognostic factor for OS (hazard ratio (HR), 1.995; 95% confidence interval (CI), 1.353–2.942; P < 0.001; table 2). Moreover, tumor size (HR, 1.864; 95% CI, 1.303–2.665; P = 0.001) and vascular invasion (HR, 1.594; 95% CI, 1.077–2.359; P = 0.020) also showed independent prognostic value for OS. Correlation between GPS and Recurrence Probability The 3-year and 5-year recurrence probabilities of patients with a GPS of 0 were 47.0% and 71.2%, respectively. These rates were 76.7% and 100%, respectively, in patients with a GPS of 1 or 2. A significant difference existed between the TTR of patients with a GPS of 0 and patients with a GPS of 1 or 2 (P < 0.001; fig. 2B). Multivariate analysis further confirmed that the GPS was an independent prognostic predictor for TTR (P < 0.001).
Fig. 1. Kaplan-Meier survival curves showing the relationship between GPS and survival after hepatectomy in patients with HCC (P = 0.001).
Fig. 2. Kaplan-Meier survival curves showing the survival (A) and recurrence (B) differences between 2 GPS groups (GPS = 0 vs. GPS = 1 or 2) after hepatectomy in patients with HCC.
194
Oncol Res Treat 2014;37:192–197
Pan/Zhang/Su/Wang/Ke
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Results
Table 2. Univariate and multivariate analyses of clinicopathologic factors and blood loss in relation to the postoperative outcome Variable Gender Male (n = 157) Female (n = 14) Age (years) < 54 (n = 83) ≥ 54 (n = 88) Child-Pugh class A (n = 144) B (n = 27) CRP, mg/l ≤ 10 (n = 145) > 10 (n = 26) Alb, g/l < 35 (n = 39) ≥ 35 (n = 132) HbsAg status Positive (n = 148) Negative (n = 23) Vascular invasion Yes (n = 48) No (n = 123) Number of tumors 1 (n = 134) ≥ 2 (n = 37) Tumor sizea, cm ≤ 5 (n = 77) > 5 (n = 94) AFP, ng/ml < 400 (n = 102) ≥ 400 (n = 69) ALT, U/l ≤ 40 (n = 84) > 40 (n = 87) AST, U/l ≤ 40 (n = 69) > 40 (n = 102) Total bilirubin, mmol/l ≤ 17.1 (n = 80) > 17.1 (n = 91) Ascites Negative (n = 152) Positive (n = 19) TNM classification I (n = 96) II (n = 56) III (n = 19) GPSb 0 (n = 124) 1 or 2 (n = 47) Blood loss, ml < 420 (n = 83) ≥ 420 (n = 88)
Survival rate, %
Univariate
Multivariate HR (95% CI)
P
1.594 (1.077–2.359)
0.020
1.864 (1.303–2.665)
0.001
1.995 (1.353–2.942)
< 0.001
3-Year
5-Year
P
57.9 64.3
36.5 24.1
0.653
59 57.9
41.7 29.6
0.931
0.618 0.407
0.377 0.244
0.028
60.7 45.8
39.7 12.5
0.002
51 60.6
19.4 40.1
0.006
57.4 65.2
36.6 29.3
0.714
45.7 63.4
22.1 40.7
0.006
64.1 38.7
37.8 24.7
0.012
74 45.6
50.8 22.8
< 0.001
57.8 59.2
34.5 37.2
0.867
58.2 58.6
37 34.1
0.179
53.5 61.7
38 34
0.077
61.3 56
31.5 39.4
0.523
60.5 41.4
35.9 34.5
0.825
67.7 44.5 52.6
43.1 24.3 29.6
0.008
62.1 48.7
42.7 15.7
< 0.001
61.4 55.6
41.1 30.4
0.059
Discussion In the current study, we reported that the GPS based on the systemic inflammatory response, which was easily assessable preoperatively, was associated with the outcome of patients with HCC after radical resection. The prognostic value of the GPS was first reported by Forrest et al. [16] in nonsmall-cell lung cancer. They showed that patients with a higher GPS were associated with a poorer clinical outcome. Although
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
the mechanism by which the GPS might impact on survival of advanced cancer patients is still not well defined, cumulative evidence supports the prognostic value of the GPS in multiple solid tumors, such as in lung cancer [17, 18, 21], ovarian cancer [12], colorectal cancer [22], gastroesophageal cancer [13], renal cancer [15], and pancreatic cancer [14]. Moreover, increasing evidence supports the notion that the systemic inflammatory response is an independent prognostic predictor for cancer patients [23, 24].
Oncol Res Treat 2014;37:192–197
195
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
CRP = C-reactive protein, Alb = albumin, ALT = alanine aminotransferase, AST = aspartate aminotransferase, HbsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, TNM = tumor/node/metastasis, GPS = Glasgow Prognostic Score. a Diameter. b GPS was included in a separate multivariate analysis independent of CRP and Alb, avoiding interference with each other’s HR.
Table 3. Differences of clinicopathologic factors and blood loss during resection between the 2 GPS groups (GPS = 0 vs. GPS = 1 or 2)
complications in hepatic resection for HCC. Accordingly, we found that patients with higher GPS were prone to have larger volumes of operative blood loss. From this perspective, the preoperative GPS is able to recognize patients who are at risk of large-volume blood loss. The increase in the level of CRP, an acute-phase reactant produced by the liver in response to diverse inflammatory stimuli, may reflect the systemic inflammatory response. The prognostic value of the CRP has been recognized in multiple cancers [29, 30]. The mechanism by which cancer induces increased CRP levels is well established. Tumor cells are able to produce cytokines such as IL-6, IL-8, or tumor necrosis factor (TNF)-a, which can induce a strong production of CRP in the liver [13]. The present study also shows that the raised CRP level is a predictor of poor prognosis for HCC patients. The serum albumin level, primarily reflecting the presence of a systemic inflammatory response syndrome, has been repeatedly demonstrated to reflect the prognosis of cancer patients. It was included in staging systems such as in ChildPugh classification and CLIP staging and it also partially reflects the level of liver cirrhosis occurring in most HCC patients. The prognosis value of the serum albumin level in HCC has been widely accepted. In the current study, similar results were found.
Variable
GPS = 0 (n = 124)
GPS = 1 or 2 (n = 47)
P
Age, years Gender Male Female Child-Pugh class A B CRP, mg/l Alb, ng/ml HBsAg status Positive Negative Vascular invasion Yes No Tumor sizea, cm Number of tumors 1 ≥2 AFP, ng/ml ALT, U/l AST, U/l Total bilirubin, mmol/l Ascites Negative Positive TNM classification I (n = 96) II (n = 56) III (n = 19) Blood loss, ml
51.95 ± 12.68
54.66 ± 12.50
0.213
117 7
40 7
0.049
110 14 2.65 ± 2.52 39.41 ± 3.11
34 13 19.09 ± 21.05 30.84 ± 4.98
0.009
108 16
40 7
0.733
28 96 5.52 ± 3.01
20 27 6.39 ± 3.39
0.009
97 27 4,645.83 ± 12,107.70 56.79 ± 63.19 70.40 ± 81.93 23.96 ± 20.63
37 10 2,769.06 ± 6,828.20 120.15 ± 211.83 109.94 ± 185.37 26.22 ± 37.91
0.944
112 12
40 7
0.333
77 35 12 548.71 ± 568.68
19 21 7 823.19 ± 812.07
0.039
CRP = C-reactive protein, Alb = albumin, HBsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, ALT = alanine aminotransferase, AST = aspartate aminotransferase, TNM = tumor/node/metastasis. a Diameter.
196
Oncol Res Treat 2014;37:192–197
Pan/Zhang/Su/Wang/Ke
< 0.001 < 0.001
0.105
0.317 0.003 0.055 0.618
0.014
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Until now, the prognostic value of the GPS was mainly based on HCC patients from Europe or Japan, who were mainly hepatitis C virus-infected individuals. However, the association between the GPS and the outcome of HBV-related HCC patients in China, constituting over half of the HCC patients all over the world, was undetermined. It was supposed that, in HCC patients who often have liver dysfunction due to chronic hepatitis or liver cirrhosis caused by viral infection, the hepatocytes have weakened reactions against interleukins derived from host-versus-tumor interactions [19, 25]. So, Ishizuka et al. [25] proposed the hepatic GPS, a GPS modified by reducing the cutoff value of the serum CRP from 1.0 to 0.3, for the prediction of postoperative mortality of HCC patients. However, the present study shows that the traditional version of the GPS can discriminate between the patients and effectively predict the clinical outcome, which is in accordance with a previous study [26]. Another retrospective study conducted by Ishizuka et al. [27] showed that both the GPS and the CLIP score were effective mortality predictors for HCC patients; moreover, the GPS was able to clearly divide patients with low CLIP score into 3 independent groups. As such, the GPS can be used as a useful tool for subclass analysis when estimating the CLIP score. Fujiwara et al. [28] reported that the GPS reflects the preoperative patient status and is associated with blood transfusion and pulmonary
The association between the GPS and the clinicopathologic features showed that an increase in the GPS was accompanied by a high incidence of malignant biological behavior, such as the presence of vascular invasion, large tumors, multiple tumors, and high TNM stage. These clinicopathologic features are often regarded as prognosis predictors by clinicians. No difference was revealed between the prognosis of patients with a GPS of 1 and those with a GPS of 2, possibly due to the smaller numbers of patients with a GPS of 1 (n = 29) or 2 (n = 18) than of patients with a GPS of 0 (n = 124), which was in accordance with the previous study conducted by I shizuka et al. [25]. When combining the aforementioned 2 subgroups, we did found a significant difference between the prognosis of patients with a GPS of 0 and that of the combined group. Similarly, a high incidence of malignant biological behavior was discovered in the combined group (table 3).
Taken together, the current study shows that the GPS is suitable for the preoperative evaluation of prognosis in Chinese patients with HCC.
Authors’ Contributions Z.-J.S. and J.-H.Z. conceived and designed the experiments. C.-R.W. and S.-Y.K. contributed to the acquisition of the data and made substantial contributions to the collected tissue samples; Q.-X.P. participated in the study design and coordination, data analysis and interpretation and drafted the manuscript. All authors have read and approved the final manuscript.
Disclosure Statement There is no conflict of interest of any of the authors in relation to the submission.
References
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
12 Sharma R, Hook J, Kumar M, Gabra H: Evaluation of an inflammation-based prognostic score in patients with advanced ovarian cancer. Eur J Cancer 2008;44:251–256. 13 Crumley AB, McMillan DC, McKernan M: Evaluation of an inflammation-based prognostic score in patients with inoperable gastro-oesophageal cancer. Br J Cancer 2006;94:637–641. 14 Glen P, Jamieson NB, McMillan DC: Evaluation of an inflammation-based prognostic score in patients with inoperable pancreatic cancer. Pancreatology 2006;6:450–453. 15 Ramsey S, Lamb GW, Aitchison M: Evaluation of an inflammation-based prognostic score in patients with metastatic renal cancer. Cancer 2007;109:205– 212. 16 Forrest LM, McMillan DC, McArdle CS: Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer 2003;89:1028–1030. 17 Pinato DJ, Mauri FA, Ramakrishnan R: Inflammation-based prognostic indices in malignant pleural mesothelioma. J Thorac Oncol 2012;7:587–594. 18 Forrest LM, McMillan DC, McArdle CS: A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-small-cell lung cancer. Br J Cancer 2005;92:1834–1836. 19 Ishizuka M, Nagata H, Takagi K: Inflammationbased prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg 2007;246:1047–1051. 20 Kobayashi T, Teruya M, Kishiki T: Elevated C-reactive protein and hypoalbuminemia measured before resection of colorectal liver metastases predict postoperative survival. Dig Surg 2010;27:285–290. 21 Forrest LM, McMillan DC, McArdle CS: Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG) in pa-
tients receiving platinum-based chemotherapy for inoperable non-small-cell lung cancer. Br J Cancer 2004;90:1704–1706. 22 Sharma R, Zucknick M, London R: Systemic inflammatory response predicts prognosis in patients with advanced-stage colorectal cancer. Clin Colorectal Cancer 2008;7:331–337. 23 Argiles JM, Busquets S, Toledo M, Lopez-Soriano FJ: The role of cytokines in cancer cachexia. Curr Opin Support Palliat Care 2009;3:263–268. 24 Roxburgh CS, McMillan DC: Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol 2010;6:149–163. 25 Ishizuka M, Kubota K, Kita J: Usefulness of a modified inflammation-based prognostic system for predicting postoperative mortality of patients undergoing surgery for primary hepatocellular carcinoma. J Surg Oncol 2011;103:801–806. 26 Horino K, Beppu T, Kuroki H: Glasgow Prognostic Score as a useful prognostic factor after hepatectomy for hepatocellular carcinoma. Int J Clin Oncol 2013;18:829–838. 27 Ishizuka M, Kubota K, Kita J: Impact of an inflammation-based prognostic system on patients undergoing surgery for hepatocellular carcinoma: a retrospective study of 398 Japanese patients. Am J Surg 2012;203:101–106. 28 Fujiwara Y, Shiba H, Furukawa K: Glasgow Prognostic Score is related to blood transfusion requirements and post-operative complications in hepatic resection for hepatocellular carcinoma. Anticancer Res 2010;30:5129–5136. 29 Hashimoto K, Ikeda Y, Korenaga D: The impact of preoperative serum C-reactive protein on the prognosis of patients with hepatocellular carcinoma. Cancer 2005;103:1856–1864. 30 Nagaoka S, Yoshida T, Akiyoshi J: Serum C-reactive protein levels predict survival in hepatocellular carcinoma. Liver Int 2007;27:1091–1097.
Oncol Res Treat 2014;37:192–197
197
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
1 Nordenstedt H, White DL, El-Serag HB: The changing pattern of epidemiology in hepatocellular carcinoma. Dig Liver Dis 2010;42(suppl 3): S206–S214. 2 Pawlik TM: Emerging therapeutic approaches to hepatocellular carcinoma. Ann Surg Oncol 2010; 17:1217–1218. 3 Lei HJ, Chau GY, Lui WY, et al.: Prognostic value and clinical relevance of the 6th Edition 2002 American Joint Committee on Cancer staging system in patients with resectable hepatocellular carcinoma. J Am Coll Surg 2006;203:426–435. 4 Okuda K, Ohtsuki T, Obata H: Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients. Cancer 1985; 56:918–928. 5 A new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients: the Cancer of the Liver Italian Program (CLIP) investigators. Hepatology 1998;28:751–755. 6 Llovet JM, Bru C, Bruix J: Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999;19:329–338. 7 Henderson JM, Sherman M, Tavill A: AHPBA/ AJCC consensus conference on staging of hepatocellular carcinoma: consensus statement. HPB (Oxford) 2003;5:243–250. 8 Grivennikov SI, Greten FR, Karin M: Immunity, inflammation, and cancer. Cell 2010;140:883–899. 9 Kessenbrock K, Plaks V, Werb Z: Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 2010;141:52–67. 10 Howren MB, Lamkin DM, Suls J: Associations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med 2009;71:171–186. 11 McMillan DC, Watson WS, O’Gorman P: Albumin concentrations are primarily determined by the body cell mass and the systemic inflammatory response in cancer patients with weight loss. Nutr Cancer 2001;39:210–213.
Received: August 8, 2013 Accepted: February 27, 2014 Published online: March 21, 2014
The Glasgow Prognostic Score Is an Independent Prognostic Predictor of Hepatocellular Carcinoma Following Radical Resection Qun-Xiong Pan Jian-Hua Zhang Zi-Jian Su Chong-Ren Wang Shao-Ying Ke Department of Oncosurgery, Affiliated Quanzhou First Hospital of Fujian Medical University, Quanzhou, Fujian, China
Summary Background: Some prognostic evaluation systems were developed to postoperatively predict the outcome of hepatocellular carcinoma (HCC) patients, mainly based on the cancer itself and the underlying liver diseases. However, none of these prognostic evaluation systems have so far been universally accepted. A simple and feasible scoring system is still lacking for the prediction of prognosis of HCC patients following resection. We aimed to uncover the correlation between the preoperative Glasgow Prognostic Score (GPS) and the clinical outcome of HCC patients after radical resection. Methods: The patients were separated into 3 subgroups on the basis of their GPSs. The prognostic significance of the GPS in the patient cohort was evaluated by survival analysis. Results: On univariate analysis, the levels of C-reactive protein and albumin, the Child-Pugh class, vascular invasion, tumor number, tumor size, the tumor/ node/metastasis (TNM) stage, and the GPS were associated with overall survival and time to recurrence of HCC patients after radical resection. On multivariate analysis, the tumor size, albumin level, and GPS were independently associated with the outcome of HCC postoperatively. Conclusion: The GPS is an independent biomarker for prognostic prediction of HCC following radical resection.
Q.-X.P. and J.-H.Z. contributed equally to this work.
© 2014 S. Karger GmbH, Freiburg 2296-5270/14/0374-0192$39.50/0 Fax +49 761 4 52 07 14 [email protected] www.karger.com
Accessible online at: www.karger.com/ort
Introduction Hepatocellular carcinoma (HCC), with rising incidence all over the world, is one of the most prevalent malignancies. The morbidity of HCC is over 500,000 per year worldwide, with more than half of these occurring in China alone. As one of the intractable malignancies, HCC is a major cause of mortality in cancer patients among men worldwide [1]. Until now, surgery is the most effective therapy for HCC, due to the rising rate of early detection and the improvement of surgical technologies [2]. Some prognostic evaluation systems were developed to postoperatively predict the outcome of HCC patients, mainly based on the cancer itself and the underlying liver diseases, such as the tumor/node/metastasis (TNM) classification system of the American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) [3], the Okuda staging system [4], the Cancer of the Liver Italian Program (CLIP) score [5], and the Barcelona Clinic Liver Cancer (BCLC) staging system [6]. However, none of these prognostic evaluation systems have so far been universally accepted, in part due to the complexity of the potential prognostic factors [7]. A simple and feasible scoring system is still lacking for the prediction of prognosis of HCC patients following resection. A role for inflammation in tumorigenesis is now generally accepted, and growing evidence supports the notion that an inflammatory microenvironment is an essential component of all tumors [8]. Inflammatory cytokines, which can be secreted both by inflammatory cells and tumor cells, can actively affect tumor development and progression, partially by inducing immunosuppression in the tumor milieu [9]. C-reactive protein (CRP), a plasma protein synthesized by the liver in response to stimulation by interleukin (IL)-6, is a sensitive and dynamic systemic marker of inflammation [10]. Moreover, hypoalbuminemia plays a major role in the suppression of cell-mediated immune responses [11]. It is reasonable to assume that
Dr. Qun-Xiong Pan Department of Oncosurgery Affiliated Quanzhou First Hospital of Fujian Medical University Quanzhou, Fujian 362000, China [email protected]
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Keywords Hepatocellular carcinoma · Glasgow Prognostic Score · C-reactive protein
the evaluation of the inflammation-based Glasgow Prognostic Score (GPS), derived from standard thresholds of CRP and albumin, can reflect the immunosuppression status of tumor patients. Multiple studies have shown that the GPS could be considered as an independent prognostic index in tumor patients [12–18]. However, until now, the clinical significance of the GPS in HCC, especially for HCC patients with hepatitis B virus (HBV) infection, is still undetermined. The current study aims to uncover the correlation between the preoperative GPS and the clinical outcome of HCC patients after radical resection.
Patients and Methods Patient Selection A cohort of 171 cases was selected retrospectively from those patients who underwent radical resection for HCC at the Department of Surgical Oncology, First Hospital of Quanzhou, Fujian Medical University between June 2001 and September 2007. All patients were managed by the same team of hepatic surgeons; all blood samples were collected the day before surgery. Tumor staging was determined according to the 6th edition of the TNM classification of the AJCC/UICC. Tumor differentiation was defined according to the Edmondson grading system. Data were censored at the last follow-up for living patients or for patients lost to follow-
up. Time to recurrence (TTR) and overall survival (OS) were defined as the interval from primary surgical treatment to the first recurrence or death, respectively. Ethical approval was obtained from the research ethic committee of the First Hospital of Quanzhou; informed consent was obtained from each patient. The inclusion and exclusion criteria of the patient cohorts include (a) having a distinctive pathologic diagnosis of HCC, (b) having no anticancer treatment before liver resection, (c) without distant metastasis and having curative liver resection, (d) having complete clinicopathologic and follow-up data. GPS Evaluation The GPS was constructed as described previously [16, 19, 20]. Briefly, patients with both an elevated CRP level (> 10 mg/l) and hypoalbuminemia (< 35 g/l) were allocated a score of 2. Patients with only 1 of these biochemical abnormalities were allocated a score of 1. Patients with neither of these abnormalities were allocated a score of 0. Statistics Statistical analyses were done by SPSS 13.0 for Windows. The c2 test, Fisher’s exact probability test, and Student’s t-test were used for comparison between groups. The nonparametric Mann-Whitney U test was used when variables in groups were not homogeneous. One-way analysis of variance (ANOVA) was used to compare quantitative data among 3 groups. The cumulative survival time was calculated by the Kaplan-Meier method and analyzed by the log-rank test. Univariate and multivariate analyses were based on the Cox proportional hazard regression model. Multivariate analysis was performed using a Cox regression model in a forward stepwise manner. P < 0.05 was considered statistically significant.
Table 1. Relationship among GPS, clinicopathologic features, and blood loss during hepatectomy Age, years Gender Male Female Child-Pugh class A B CRP, mg/l Alb, g/l HbsAg status Positive Negative Vascular invasion Yes No Tumor sizea, cm Number of tumors 1 ≥2 AFP, ng/ml ALT, U/l AST, U/l Total bilirubin, mmol/l Ascites Negative Positive TNM classification I II III Blood loss, ml
Overall (n = 171)
GPS = 0 (n = 124)
GPS = 1 (n = 29)
GPS = 2 (n = 18)
P
52.70 ± 12.67
51.95 ± 12.70
53.14 ± 12.69
57.11 ± 12.16
0.267
157 14
117 7
24 5
16 2
0.109
144 27 7.17 ± 13.4 37.05 ± 5.33
110 14 2.65 ± 2.52 39.41 ± 3.12
23 6 9.96 ± 8.36 32.29 ± 4.77
11 7 33.81 ± 26.68 28.50 ± 4.48
0.008
148 23
108 16
26 3
14 4
0.481
48 123 5.75 ± 3.13
28 96 5.52 ± 3.01
11 18 6.40 ± 3.54
9 9 6.37 ± 3.24
0.023
134 37 4,130.0 ± 10,926.52 74.2 ± 125.84 81.26 ± 120.28 24.58 ± 26.41
97 27 4,645.83 ± 12,107.70 56.79 ± 63.20 70.40 ± 81.93 23.96 ± 20.62
23 6 3,088.98 ± 7,596.47 137.97 ± 251.56 127.48 ± 231.08 28.89 ± 46.83
14 4 2,253.63 ± 5,536.98 91.44 ± 125.61 81.67 ± 60.43 21.92 ± 15.63
0.990
152 19
112 12
28 1
12 6
0.004
96 56 19 624.15 ± 653.86
77 35 12 548.71 ± 568.68
14 10 5 856.90 ± 830.96
5 11 2 768.90 ± 801.30
0.039
< 0.001 < 0.001
0.270
0.588 0.006 0.070 0.601
0.044
CRP = C-reactive protein, Alb = albumin, HbsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, ALT = alanine aminotransferase, AST = aspartate aminotransferase, TNM = tumor/node/metastasis, GPS = Glasgow Prognostic Score. a Diameter.
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
Oncol Res Treat 2014;37:192–197
193
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Variable
size, tumor number, and TNM classification were also unfavorable predictors for OS.
Patient Profiles The median follow-up time was 41 months (range, 1–76 months; standard deviation (SD), 24.27 months). At the last follow-up (January 02, 2013), 158 patients (92.4%) had recurrence and 143 patients (83.6%) had died of recurrence (n = 131) or cirrhosis-related complications without recurrence (n = 12). Correlation among GPS, Clinicopathologic Features, and Blood Loss The relationships among the GPS, the clinicopathologic features, and blood loss during hepatectomy are shown in table 1. No significant differences were found when age (P = 0.267), gender (P = 0.109), tumor number (P = 0.990), tumor size (P = 0.270), a-fetoprotein (P = 0.588), aspartate aminotransferase (AST) level (P = 0.070), total bilirubin (P = 0.601), and HBV surface antigen (HBsAg) status (P = 0.481) were evaluated between different GPS subgroups. In contrast, an increasing GPS was associated with a higher alanine aminotransferase (ALT) level (P = 0.006), a higher Child-Pugh class (P = 0.008), a higher incidence of ascites (P = 0.004) and vascular invasion (P = 0.023), a higher TNM staging (P = 0.039), and a larger volume of blood loss during resection (P = 0.044). Univariate Analysis of Prognostic Factors The 3-year and 5-year survival rates of patients with a GPS of 0 were 62.1% and 42.7%, respectively. These rates were 48.3% and 13.9%, respectively, in patients with a GPS of 1 and 49.4% and 18.5%, respectively, in patients with a GPS of 2. Because there was no significant difference between the survival of patients with GPSs of 1 and 2 (P = 0.676; fig. 1), we combined the 2 subgroups when evaluating the prognostic value of the GPS. Compared with patients with a GPS of 0, the 3-year and 5-year survival rates of patients with a GPS of 1 or 2 were 48.7% and 15.7%, respectively. A significant difference exists between these 2 groups (P < 0.001; fig. 2A). The clinicopathologic features and blood loss during hepatectomy were also subjected to univariate analysis. As shown in table 2, child-Pugh class, vascular invasion, tumor
Multivariate Analysis of the Prognostic Value of the GPS Factors that showed significance by univariate analysis were used as covariates in a multivariate Cox proportional hazards model. Since over 80% of the patients were HBsAg positive, it might be crucial whether antiviral treatment was initiated or not. We included this option in the multivariate analysis. Multivariate analysis revealed that the GPS was an independent prognostic factor for OS (hazard ratio (HR), 1.995; 95% confidence interval (CI), 1.353–2.942; P < 0.001; table 2). Moreover, tumor size (HR, 1.864; 95% CI, 1.303–2.665; P = 0.001) and vascular invasion (HR, 1.594; 95% CI, 1.077–2.359; P = 0.020) also showed independent prognostic value for OS. Correlation between GPS and Recurrence Probability The 3-year and 5-year recurrence probabilities of patients with a GPS of 0 were 47.0% and 71.2%, respectively. These rates were 76.7% and 100%, respectively, in patients with a GPS of 1 or 2. A significant difference existed between the TTR of patients with a GPS of 0 and patients with a GPS of 1 or 2 (P < 0.001; fig. 2B). Multivariate analysis further confirmed that the GPS was an independent prognostic predictor for TTR (P < 0.001).
Fig. 1. Kaplan-Meier survival curves showing the relationship between GPS and survival after hepatectomy in patients with HCC (P = 0.001).
Fig. 2. Kaplan-Meier survival curves showing the survival (A) and recurrence (B) differences between 2 GPS groups (GPS = 0 vs. GPS = 1 or 2) after hepatectomy in patients with HCC.
194
Oncol Res Treat 2014;37:192–197
Pan/Zhang/Su/Wang/Ke
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Results
Table 2. Univariate and multivariate analyses of clinicopathologic factors and blood loss in relation to the postoperative outcome Variable Gender Male (n = 157) Female (n = 14) Age (years) < 54 (n = 83) ≥ 54 (n = 88) Child-Pugh class A (n = 144) B (n = 27) CRP, mg/l ≤ 10 (n = 145) > 10 (n = 26) Alb, g/l < 35 (n = 39) ≥ 35 (n = 132) HbsAg status Positive (n = 148) Negative (n = 23) Vascular invasion Yes (n = 48) No (n = 123) Number of tumors 1 (n = 134) ≥ 2 (n = 37) Tumor sizea, cm ≤ 5 (n = 77) > 5 (n = 94) AFP, ng/ml < 400 (n = 102) ≥ 400 (n = 69) ALT, U/l ≤ 40 (n = 84) > 40 (n = 87) AST, U/l ≤ 40 (n = 69) > 40 (n = 102) Total bilirubin, mmol/l ≤ 17.1 (n = 80) > 17.1 (n = 91) Ascites Negative (n = 152) Positive (n = 19) TNM classification I (n = 96) II (n = 56) III (n = 19) GPSb 0 (n = 124) 1 or 2 (n = 47) Blood loss, ml < 420 (n = 83) ≥ 420 (n = 88)
Survival rate, %
Univariate
Multivariate HR (95% CI)
P
1.594 (1.077–2.359)
0.020
1.864 (1.303–2.665)
0.001
1.995 (1.353–2.942)
< 0.001
3-Year
5-Year
P
57.9 64.3
36.5 24.1
0.653
59 57.9
41.7 29.6
0.931
0.618 0.407
0.377 0.244
0.028
60.7 45.8
39.7 12.5
0.002
51 60.6
19.4 40.1
0.006
57.4 65.2
36.6 29.3
0.714
45.7 63.4
22.1 40.7
0.006
64.1 38.7
37.8 24.7
0.012
74 45.6
50.8 22.8
< 0.001
57.8 59.2
34.5 37.2
0.867
58.2 58.6
37 34.1
0.179
53.5 61.7
38 34
0.077
61.3 56
31.5 39.4
0.523
60.5 41.4
35.9 34.5
0.825
67.7 44.5 52.6
43.1 24.3 29.6
0.008
62.1 48.7
42.7 15.7
< 0.001
61.4 55.6
41.1 30.4
0.059
Discussion In the current study, we reported that the GPS based on the systemic inflammatory response, which was easily assessable preoperatively, was associated with the outcome of patients with HCC after radical resection. The prognostic value of the GPS was first reported by Forrest et al. [16] in nonsmall-cell lung cancer. They showed that patients with a higher GPS were associated with a poorer clinical outcome. Although
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
the mechanism by which the GPS might impact on survival of advanced cancer patients is still not well defined, cumulative evidence supports the prognostic value of the GPS in multiple solid tumors, such as in lung cancer [17, 18, 21], ovarian cancer [12], colorectal cancer [22], gastroesophageal cancer [13], renal cancer [15], and pancreatic cancer [14]. Moreover, increasing evidence supports the notion that the systemic inflammatory response is an independent prognostic predictor for cancer patients [23, 24].
Oncol Res Treat 2014;37:192–197
195
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
CRP = C-reactive protein, Alb = albumin, ALT = alanine aminotransferase, AST = aspartate aminotransferase, HbsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, TNM = tumor/node/metastasis, GPS = Glasgow Prognostic Score. a Diameter. b GPS was included in a separate multivariate analysis independent of CRP and Alb, avoiding interference with each other’s HR.
Table 3. Differences of clinicopathologic factors and blood loss during resection between the 2 GPS groups (GPS = 0 vs. GPS = 1 or 2)
complications in hepatic resection for HCC. Accordingly, we found that patients with higher GPS were prone to have larger volumes of operative blood loss. From this perspective, the preoperative GPS is able to recognize patients who are at risk of large-volume blood loss. The increase in the level of CRP, an acute-phase reactant produced by the liver in response to diverse inflammatory stimuli, may reflect the systemic inflammatory response. The prognostic value of the CRP has been recognized in multiple cancers [29, 30]. The mechanism by which cancer induces increased CRP levels is well established. Tumor cells are able to produce cytokines such as IL-6, IL-8, or tumor necrosis factor (TNF)-a, which can induce a strong production of CRP in the liver [13]. The present study also shows that the raised CRP level is a predictor of poor prognosis for HCC patients. The serum albumin level, primarily reflecting the presence of a systemic inflammatory response syndrome, has been repeatedly demonstrated to reflect the prognosis of cancer patients. It was included in staging systems such as in ChildPugh classification and CLIP staging and it also partially reflects the level of liver cirrhosis occurring in most HCC patients. The prognosis value of the serum albumin level in HCC has been widely accepted. In the current study, similar results were found.
Variable
GPS = 0 (n = 124)
GPS = 1 or 2 (n = 47)
P
Age, years Gender Male Female Child-Pugh class A B CRP, mg/l Alb, ng/ml HBsAg status Positive Negative Vascular invasion Yes No Tumor sizea, cm Number of tumors 1 ≥2 AFP, ng/ml ALT, U/l AST, U/l Total bilirubin, mmol/l Ascites Negative Positive TNM classification I (n = 96) II (n = 56) III (n = 19) Blood loss, ml
51.95 ± 12.68
54.66 ± 12.50
0.213
117 7
40 7
0.049
110 14 2.65 ± 2.52 39.41 ± 3.11
34 13 19.09 ± 21.05 30.84 ± 4.98
0.009
108 16
40 7
0.733
28 96 5.52 ± 3.01
20 27 6.39 ± 3.39
0.009
97 27 4,645.83 ± 12,107.70 56.79 ± 63.19 70.40 ± 81.93 23.96 ± 20.63
37 10 2,769.06 ± 6,828.20 120.15 ± 211.83 109.94 ± 185.37 26.22 ± 37.91
0.944
112 12
40 7
0.333
77 35 12 548.71 ± 568.68
19 21 7 823.19 ± 812.07
0.039
CRP = C-reactive protein, Alb = albumin, HBsAg = hepatitis B virus surface antigen, AFP = a-fetoprotein, ALT = alanine aminotransferase, AST = aspartate aminotransferase, TNM = tumor/node/metastasis. a Diameter.
196
Oncol Res Treat 2014;37:192–197
Pan/Zhang/Su/Wang/Ke
< 0.001 < 0.001
0.105
0.317 0.003 0.055 0.618
0.014
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
Until now, the prognostic value of the GPS was mainly based on HCC patients from Europe or Japan, who were mainly hepatitis C virus-infected individuals. However, the association between the GPS and the outcome of HBV-related HCC patients in China, constituting over half of the HCC patients all over the world, was undetermined. It was supposed that, in HCC patients who often have liver dysfunction due to chronic hepatitis or liver cirrhosis caused by viral infection, the hepatocytes have weakened reactions against interleukins derived from host-versus-tumor interactions [19, 25]. So, Ishizuka et al. [25] proposed the hepatic GPS, a GPS modified by reducing the cutoff value of the serum CRP from 1.0 to 0.3, for the prediction of postoperative mortality of HCC patients. However, the present study shows that the traditional version of the GPS can discriminate between the patients and effectively predict the clinical outcome, which is in accordance with a previous study [26]. Another retrospective study conducted by Ishizuka et al. [27] showed that both the GPS and the CLIP score were effective mortality predictors for HCC patients; moreover, the GPS was able to clearly divide patients with low CLIP score into 3 independent groups. As such, the GPS can be used as a useful tool for subclass analysis when estimating the CLIP score. Fujiwara et al. [28] reported that the GPS reflects the preoperative patient status and is associated with blood transfusion and pulmonary
The association between the GPS and the clinicopathologic features showed that an increase in the GPS was accompanied by a high incidence of malignant biological behavior, such as the presence of vascular invasion, large tumors, multiple tumors, and high TNM stage. These clinicopathologic features are often regarded as prognosis predictors by clinicians. No difference was revealed between the prognosis of patients with a GPS of 1 and those with a GPS of 2, possibly due to the smaller numbers of patients with a GPS of 1 (n = 29) or 2 (n = 18) than of patients with a GPS of 0 (n = 124), which was in accordance with the previous study conducted by I shizuka et al. [25]. When combining the aforementioned 2 subgroups, we did found a significant difference between the prognosis of patients with a GPS of 0 and that of the combined group. Similarly, a high incidence of malignant biological behavior was discovered in the combined group (table 3).
Taken together, the current study shows that the GPS is suitable for the preoperative evaluation of prognosis in Chinese patients with HCC.
Authors’ Contributions Z.-J.S. and J.-H.Z. conceived and designed the experiments. C.-R.W. and S.-Y.K. contributed to the acquisition of the data and made substantial contributions to the collected tissue samples; Q.-X.P. participated in the study design and coordination, data analysis and interpretation and drafted the manuscript. All authors have read and approved the final manuscript.
Disclosure Statement There is no conflict of interest of any of the authors in relation to the submission.
References
GPS as a Prognostic Predictor of Hepatocellular Carcinoma
12 Sharma R, Hook J, Kumar M, Gabra H: Evaluation of an inflammation-based prognostic score in patients with advanced ovarian cancer. Eur J Cancer 2008;44:251–256. 13 Crumley AB, McMillan DC, McKernan M: Evaluation of an inflammation-based prognostic score in patients with inoperable gastro-oesophageal cancer. Br J Cancer 2006;94:637–641. 14 Glen P, Jamieson NB, McMillan DC: Evaluation of an inflammation-based prognostic score in patients with inoperable pancreatic cancer. Pancreatology 2006;6:450–453. 15 Ramsey S, Lamb GW, Aitchison M: Evaluation of an inflammation-based prognostic score in patients with metastatic renal cancer. Cancer 2007;109:205– 212. 16 Forrest LM, McMillan DC, McArdle CS: Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer 2003;89:1028–1030. 17 Pinato DJ, Mauri FA, Ramakrishnan R: Inflammation-based prognostic indices in malignant pleural mesothelioma. J Thorac Oncol 2012;7:587–594. 18 Forrest LM, McMillan DC, McArdle CS: A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-small-cell lung cancer. Br J Cancer 2005;92:1834–1836. 19 Ishizuka M, Nagata H, Takagi K: Inflammationbased prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg 2007;246:1047–1051. 20 Kobayashi T, Teruya M, Kishiki T: Elevated C-reactive protein and hypoalbuminemia measured before resection of colorectal liver metastases predict postoperative survival. Dig Surg 2010;27:285–290. 21 Forrest LM, McMillan DC, McArdle CS: Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG) in pa-
tients receiving platinum-based chemotherapy for inoperable non-small-cell lung cancer. Br J Cancer 2004;90:1704–1706. 22 Sharma R, Zucknick M, London R: Systemic inflammatory response predicts prognosis in patients with advanced-stage colorectal cancer. Clin Colorectal Cancer 2008;7:331–337. 23 Argiles JM, Busquets S, Toledo M, Lopez-Soriano FJ: The role of cytokines in cancer cachexia. Curr Opin Support Palliat Care 2009;3:263–268. 24 Roxburgh CS, McMillan DC: Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol 2010;6:149–163. 25 Ishizuka M, Kubota K, Kita J: Usefulness of a modified inflammation-based prognostic system for predicting postoperative mortality of patients undergoing surgery for primary hepatocellular carcinoma. J Surg Oncol 2011;103:801–806. 26 Horino K, Beppu T, Kuroki H: Glasgow Prognostic Score as a useful prognostic factor after hepatectomy for hepatocellular carcinoma. Int J Clin Oncol 2013;18:829–838. 27 Ishizuka M, Kubota K, Kita J: Impact of an inflammation-based prognostic system on patients undergoing surgery for hepatocellular carcinoma: a retrospective study of 398 Japanese patients. Am J Surg 2012;203:101–106. 28 Fujiwara Y, Shiba H, Furukawa K: Glasgow Prognostic Score is related to blood transfusion requirements and post-operative complications in hepatic resection for hepatocellular carcinoma. Anticancer Res 2010;30:5129–5136. 29 Hashimoto K, Ikeda Y, Korenaga D: The impact of preoperative serum C-reactive protein on the prognosis of patients with hepatocellular carcinoma. Cancer 2005;103:1856–1864. 30 Nagaoka S, Yoshida T, Akiyoshi J: Serum C-reactive protein levels predict survival in hepatocellular carcinoma. Liver Int 2007;27:1091–1097.
Oncol Res Treat 2014;37:192–197
197
Downloaded by: NYU Medical Center Library 128.122.253.212 - 4/28/2015 7:00:28 AM
1 Nordenstedt H, White DL, El-Serag HB: The changing pattern of epidemiology in hepatocellular carcinoma. Dig Liver Dis 2010;42(suppl 3): S206–S214. 2 Pawlik TM: Emerging therapeutic approaches to hepatocellular carcinoma. Ann Surg Oncol 2010; 17:1217–1218. 3 Lei HJ, Chau GY, Lui WY, et al.: Prognostic value and clinical relevance of the 6th Edition 2002 American Joint Committee on Cancer staging system in patients with resectable hepatocellular carcinoma. J Am Coll Surg 2006;203:426–435. 4 Okuda K, Ohtsuki T, Obata H: Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients. Cancer 1985; 56:918–928. 5 A new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients: the Cancer of the Liver Italian Program (CLIP) investigators. Hepatology 1998;28:751–755. 6 Llovet JM, Bru C, Bruix J: Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999;19:329–338. 7 Henderson JM, Sherman M, Tavill A: AHPBA/ AJCC consensus conference on staging of hepatocellular carcinoma: consensus statement. HPB (Oxford) 2003;5:243–250. 8 Grivennikov SI, Greten FR, Karin M: Immunity, inflammation, and cancer. Cell 2010;140:883–899. 9 Kessenbrock K, Plaks V, Werb Z: Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 2010;141:52–67. 10 Howren MB, Lamkin DM, Suls J: Associations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med 2009;71:171–186. 11 McMillan DC, Watson WS, O’Gorman P: Albumin concentrations are primarily determined by the body cell mass and the systemic inflammatory response in cancer patients with weight loss. Nutr Cancer 2001;39:210–213.
