Staging Hepatocellular Carcinoma by a Novel Scoring System (BALAD Score) Based on Serum Markers-论文阅读讨论-ReadPaper - 轻松读论文 | 专业翻译 | 一键引文

DOI: 10.1016/j.cgh.2006.09.021

CAS-1 JCR-Q1 SCIE

Hidenori ToyodaTakashi KumadaYukio OsakiHiroko OkaFumihiro UranoMasatoshi KudoTakashi Matsunaga

Hidenori ToyodaTakashi KumadaYukio Osaki ...+3 Takashi Matsunaga

Clinical Gastroenterology and Hepatology

Dec 2006

90被引用

See Also

Bridges Episode 4 with Your Movie Sucks

0笔记

专业版功能更新，中文段落润色开始内测，内测期间0豆无限使用，欢迎使用~

摘要原文

Background & Aims: Previously proposed staging systems for hepatocellular carcinoma (HCC) involve clinical, imaging, or pathologic factors in the evaluation. We established and validated a novel staging system for HCC that is based on simply serum markers. Methods: The new scoring system is based on 5 serum markers: bilirubin, albumin, Lens culinaris agglutinin-reactive α-fetoprotein (AFP-L3), α-fetoprotein (AFP), and des-γ-carboxy prothrombin (DCP) and thus is termed the BALAD score. The system was validated in 2600 HCC patients from 5 institutions. The power of our system to predict patient survival and its discriminative ability were compared with those of previously reported staging systems. Results: The best tumor marker cutoff values were 400 ng/dL for AFP, 15% for AFP-L3, and 100 milli-arbitrary unit/mL for DCP. The patients were classified into 6 categories on the basis of 5 laboratory values. The categories reflected patient survival well. The discriminative ability was comparable to that of previously reported staging systems. Conclusions: The new staging system for HCC combining serum albumin, serum bilirubin, and 3 tumor markers predicts patient outcomes with excellent discriminative ability. The system is easy to use and objective. In addition, stage can be evaluated with the use of only 1 serum sample. It also allows global comparison of patients with HCC or comparison of patients from different time periods with the same standard. Background & Aims: Previously proposed staging systems for hepatocellular carcinoma (HCC) involve clinical, imaging, or pathologic factors in the evaluation. We established and validated a novel staging system for HCC that is based on simply serum markers. Methods: The new scoring system is based on 5 serum markers: bilirubin, albumin, Lens culinaris agglutinin-reactive α-fetoprotein (AFP-L3), α-fetoprotein (AFP), and des-γ-carboxy prothrombin (DCP) and thus is termed the BALAD score. The system was validated in 2600 HCC patients from 5 institutions. The power of our system to predict patient survival and its discriminative ability were compared with those of previously reported staging systems. Results: The best tumor marker cutoff values were 400 ng/dL for AFP, 15% for AFP-L3, and 100 milli-arbitrary unit/mL for DCP. The patients were classified into 6 categories on the basis of 5 laboratory values. The categories reflected patient survival well. The discriminative ability was comparable to that of previously reported staging systems. Conclusions: The new staging system for HCC combining serum albumin, serum bilirubin, and 3 tumor markers predicts patient outcomes with excellent discriminative ability. The system is easy to use and objective. In addition, stage can be evaluated with the use of only 1 serum sample. It also allows global comparison of patients with HCC or comparison of patients from different time periods with the same standard. Hepatocellular carcinoma (HCC) is one of the most common malignancies, especially in southern and eastern Asia. The incidence of HCC is also increasing in the USA.1El-Serag H.B. Mason A.C. Rising incidence of hepatocellular carcinoma in the United States.N Engl J Med. 1999; 340: 745-750Crossref PubMed Scopus (2713) Google Scholar, 2El-Serag H.B. Davila J.A. Petersen N.J. et al.The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update.Ann Intern Med. 2003; 139: 817-823Crossref PubMed Scopus (837) Google Scholar One of the most important risk factors for development of HCC worldwide3Di Bisceglie A.M. Goodman Z.D. Ishak K.G. et al.Long-term clinical and histological follow-up of chronic posttransfusion hepatitis.Hepatology. 1991; 14: 969-974Crossref PubMed Scopus (603) Google Scholar, 4Kiyosawa K. Sodeyama T. Tanaka E. et al.Interrelationship of blood transfusion, non-A, non-B hepatitis and hepatocellular carcinoma: analysis by detection of antibody to hepatitis C virus.Hepatology. 1990; 12: 671-675Crossref PubMed Scopus (1185) Google Scholar, 5Brechot C. Jaffredo F. Lagorce D. et al.Impact of HBV, HCV, and GBV-C/HGV on hepatocellular carcinomas in Europe: results of an European concerted action.J Hepatol. 1998; 29: 173-183Abstract Full Text PDF PubMed Scopus (144) Google Scholar, 6Beasley R.P. Hepatitis B virus: the major etiology of hepatocellular carcinoma.Cancer. 1988; 61: 1942-1956Crossref PubMed Scopus (1164) Google Scholar is chronic viral hepatitis. Most HCCs develop in the presence of chronic hepatitis or cirrhosis, and we often observe the coexistence of cirrhosis and HCC. Thus, 2 important factors largely affect the prognosis of patients with HCC, progression of the HCC tumor and remnant liver function. The staging of HCC has been based on the TNM classification system,7Fleming I.D. Coopers J.S. Henson D.E. et al.AJCC cancer staging manual. Lippincott-Raven, Philadelphia1997Google Scholar, 8International Union Against Cancer (UICC)Liver.in: Sobin L.H. Wittekind C.H. TNM classification of malignant tumours. 6th ed. Wiley, New York2002: 81-83Google Scholar which evaluates tumor extension. Because remnant liver function is an important factor affecting prognosis, it should be incorporated with tumor progression in staging HCC. Several staging systems for the prognosis of HCC have recently been proposed9The Cancer of the Liver Italian Program (CLIP) InvestigatorsA new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients.Hepatology. 1998; 28: 751-755Crossref PubMed Scopus (1187) Google Scholar, 10Llovet J.M. Bru C. Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification.Semin Liver Dis. 1999; 19: 329-338Crossref PubMed Scopus (2903) Google Scholar, 11Chevret S. Trinchet J.-C. Mathieu D. et al.A new prognostic classification for predicting survival in patients with hepatocellular carcinoma.J Hepatol. 1999; 31: 133-141Abstract Full Text Full Text PDF PubMed Scopus (433) Google Scholar, 12Leung W. Tang A.M. Zee B. et al.Construction of the Chinese University Prognostic Index for hepatocellular carcinoma and comparison with the TNM staging system, the Okuda staging system, and the Cancer of the Liver Italian Program staging system: a study based on 926 patients.Cancer. 2002; 94: 1760-1769Crossref PubMed Scopus (492) Google Scholar, 13Kudo M. Chung H. Haji S. et al.Validation of a new prognostic staging system for hepatocellular carcinoma: the JIS score compared with the CLIP score.Hepatology. 2004; 40: 1396-1405Crossref PubMed Scopus (314) Google Scholar that include remnant liver function as a factor in addition to progression of the HCC. These staging systems, however, also include tumor morphology, such as size, the number of tumors, and vascular invasion, which are usually assessed by imaging studies or pathologic analyses. Imaging findings, however, are strongly influenced by the methods used, skill of the examiner, and quality of the imaging apparatus. In addition, there is often discrepancy in the size or number of tumors, or especially, whether vascular invasion is present, between the results of imaging studies and those of pathologic examination of the resected specimen. These recently proposed systems use the Child-Pugh classification14Pugh R.N.H. Murray-Lyon I.M. Dawson J.L. et al.Transection of the oesophagus for bleeding oesophageal varices.Br J Surg. 1973; 60: 646-649Crossref PubMed Scopus (6726) Google Scholar as the liver function variable. However, the Child-Pugh classification includes clinical factors such as the presence of ascites that are not objective, and the results can differ according to the physician performing the evaluation. We sometimes encounter patients who are reported not to have ascites on the basis of clinical examination but are shown to have ascites on the basis of ultrasonographic (US) examination. In addition, distinguishing controllable ascites from uncontrollable ascites is not always an objective assessment. In a recent report, Tateishi et al15Tateishi R. Yoshida H. Shiina S. et al.Proposal of a new prognostic model for hepatocellular carcinoma: an analysis of 403 patients.Gut. 2005; 54: 419-425Crossref PubMed Scopus (220) Google Scholar proposed a new staging system for HCC. They used only serum albumin and bilirubin values as indicators of remnant liver function. Their method allows objective and standardized evaluation of remnant liver function. Very recently, we analyzed 3 tumor markers for HCC, α-fetoprotein (AFP), Lens culinaris agglutinin A–reactive fraction of alpha-fetoprotein (AFP-L3), and des-γ-carboxy prothrombin (DCP), and showed that these 3 markers independently reflect different characteristics of HCC progression, and that the combination of results of these 3 tumor markers reflects tumor progression well, correlates well with tumor stage, and predicts patient survival well.16Toyoda H. Kumada T. Kiriyama S. et al.Prognostic significance of simultaneous measurement of three tumor markers in patients with hepatocellular carcinoma.Clin Gastroenterol Hepatol. 2006; 4: 111-117Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar In the present study, we attempted to construct and validate a novel staging system for HCC that is based simply on serum markers and does not involve imaging, pathologic, or clinical evaluations. The study group comprised 2600 patients with HCC. These patients were enrolled from 5 institutions in which a total of 3725 patients were initially diagnosed as having HCC between July 1994–December 2004, fulfilling the following criteria: 3 tumor markers of HCC (AFP, AFP-L3, and DCP) were measured at the time of diagnosis, and drugs that would influence the serum DCP level such as warfarin or vitamin K were not taken. The study protocol was approved by the Institutional Ethics Review Board at each of the institutions and was in compliance with the Declaration of Helsinki. The characteristics of patients are shown in Table 1. Mean age was 66.4 years, and there was a predominance of men (71.7%). The majority of patients were infected with HCV (74.7%). HBV infection was observed in 14.3%, and absence of hepatitis viral infection was observed in 9.3% of patients. Patients were diagnosed with HCC on the basis of histologic examination of tumor tissue taken from resected or biopsy specimens in 1143 cases (44.0%). In the remaining 1457 patients, diagnosis was made on the basis of clinical criteria,17Torzilli G. Minagawa M. Takayama T. et al.Accurate preoperative evaluation of liver mass lesions without fine-needle biopsy.Hepatology. 1999; 30: 889-893Crossref PubMed Scopus (286) Google Scholar, 18Kudo M. Imaging diagnosis of hepatocellular carcinoma and premalignant/borderline lesions.Semin Liver Dis. 1999; 19: 297-309Crossref PubMed Scopus (97) Google Scholar a pertinent clinical background (association with liver cirrhosis or viral hepatitis), and typical imaging findings. Typical imaging features of HCC include a mosaic pattern with a halo on B-mode US images; hypervascularity on angiographic images; and a high-density mass on arterial phase dynamic computed tomography (CT) images with a low-density mass on portal phase dynamic CT images obtained with a helical or multidetector raw CT scanner. When typical findings for HCC were not obtained by means of dynamic CT or angiography, CT during hepatic arteriography and CT during arterial portography or T1- and T2-weighted imaging associated with superparamagnetic iron oxide–enhanced magnetic resonance imaging (MRI) were performed. In cases without typical features, biopsy was performed to confirm the diagnosis of HCC. Of the 2600 patients, 727 underwent hepatectomy, 746 were treated by locoregional ablative therapy (LAT) including ethanol injection, microwave thermocoagulation, or radiofrequency ablation, and 702 were treated by transcatheter arterial chemoembolization (TACE). Patients were followed up from 0.4–139.1 months (median follow-up period, 20.7 months). All 2600 patients were followed up at 1 of the 5 institutions.Table 1Clinical Characteristics of Study Patients (n = 2600)Age (mean ± SD, y)66.4 ± 8.9Sex ratio (female/male)737 (28.3%)/1863 (71.7%)Etiology of underlying liver disease (HBV/HCV/HBV, HCV/non-HBV, non-HCV)373 (14.3%)/1941 (74.7%)/45 (1.7%)/241 (9.3%)Child-Pugh class (A/B/C)1746 (67.1%)/683 (26.3%)/171 (6.6%)Albumin (mean ± SD, g/dL)3.50 ± 0.56Total bilirubin (mean ± SD, mg/dL)1.19 ± 1.28Diagnostic modality (histology/other)1143 (44.0%)/1457 (56.0%)Tumor stage (I/II/III/IV)553 (21.3%)/981 (37.7%)/711 (27.3%)/355 (13.7%)Tumor size (≤2 cm/ >2 and ≤5 cm/ >5 cm)850 (32.7%)/1192 (45.8%)/558 (21.5%)Number of tumors (single/multiple)1355 (52.1%)/1245 (47.9%)Vascular invasion (absent/present)2232 (85.8%)/368 (14.2%)AFP (median, ng/dL)29.7 (range,0.8–2,402,000)AFP-L3 (median, %)1.5 (range,0–95.9)DCP (median, mAU/mL)90.0 (range,2.0–1,840,000)Initial treatment No treatment168 (6.5%) Hepatectomy727 (28.0%) LAT746 (28.7%) TACE702 (27.0%) OtheraIncludes repeated arterial infusion chemotherapy and systemic chemotherapy.257 (9.8%)mAU, milli-arbitrary unit; SD, standard deviation.a Includes repeated arterial infusion chemotherapy and systemic chemotherapy. Open table in a new tab mAU, milli-arbitrary unit; SD, standard deviation. For comparison of the predictive power and discriminative ability of our staging system and other systems, patients were categorized in several ways: by tumor progression according to the TNM classification of the Liver Cancer Study Group of Japan (Table 2),19Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. English ed. Kanehara & Co, Ltd, Tokyo, Japan2003Google Scholar by remnant liver function according to the Child-Pugh classification,14Pugh R.N.H. Murray-Lyon I.M. Dawson J.L. et al.Transection of the oesophagus for bleeding oesophageal varices.Br J Surg. 1973; 60: 646-649Crossref PubMed Scopus (6726) Google Scholar and by the Cancer of the Liver Italian Program (CLIP) scoring system9The Cancer of the Liver Italian Program (CLIP) InvestigatorsA new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients.Hepatology. 1998; 28: 751-755Crossref PubMed Scopus (1187) Google Scholar and the Japan Integrated Staging (JIS) scoring system,13Kudo M. Chung H. Haji S. et al.Validation of a new prognostic staging system for hepatocellular carcinoma: the JIS score compared with the CLIP score.Hepatology. 2004; 40: 1396-1405Crossref PubMed Scopus (314) Google Scholar which incorporate both tumor progression and remnant liver function. In most cases, the maximum diameter of the tumor was determined by means of B-mode US. Vascular invasion was assessed by means of dynamic CT and angiography. Lymph node invasion and distant metastases were assessed by means of US, dynamic CT, and chest x-ray screenings. Bone scintigraphy or brain CT was performed if suggestive symptoms were present.Table 2TNM Stage Classification of the Liver Cancer Study Group of JapanT factorI. Single II. <2 cm III. No vascular involvementT1Fulfilling 3 factorsT2Fulfilling 2 factorsT3Fulfilling 1 factorT4Fulfilling 0 factorsStage IT1 N0 M0Stage IIT2 N0 M0Stage IIIT3 N0 M0Stage IV-AT4 N0 M0, or any T N1 M0Stage IV-BAny T N0-1 M1 Open table in a new tab For staging of HCC, we included only laboratory data that are measurable from serum samples. In a recent prospective study, we measured 3 tumor markers of HCC (AFP, AFP-L3, and DCP) in 685 patients at the time of initial HCC diagnosis, and we showed that the number of elevated tumor markers correlates well with tumor stage and is a predictor of survival.16Toyoda H. Kumada T. Kiriyama S. et al.Prognostic significance of simultaneous measurement of three tumor markers in patients with hepatocellular carcinoma.Clin Gastroenterol Hepatol. 2006; 4: 111-117Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar We, therefore, used the number of elevated tumor markers as tumor factors for staging. For remnant liver function factors, we used serum albumin and serum total bilirubin levels, assigning scores to these levels according to the method of Tateishi et al15Tateishi R. Yoshida H. Shiina S. et al.Proposal of a new prognostic model for hepatocellular carcinoma: an analysis of 403 patients.Gut. 2005; 54: 419-425Crossref PubMed Scopus (220) Google Scholar (Table 3). The serum albumin level was categorized as >3.5 g/dL, 2.8–3.5 g/dL, or <2.8 g/dL and scored as 0, 1, and 2, respectively. The serum total bilirubin level was categorized as <1.0 mg/dL, 1.0–2.0 mg/dL, or >2.0 mg/dL and scored as 0, 1, and 2, respectively. Liver function was then categorized by the sum of these 2 scores as A (0 or 1), B (2 or 3), or C (4). The staging score of HCC on the basis of the laboratory data was calculated simply as the sum of the tumor progression factors and liver function factors, as shown in Table 4. The categories of A, B, and C from serum total bilirubin and albumin levels correspond to 0, 1, and 2, respectively, as liver function score. The number of elevated tumor markers was tumor progression score. The staging, therefore, was of 6 categories (0–5). We referred to this as the BALAD score, based on the first letter of each of these 5 serum markers (bilirubin, albumin, Lens culinaris agglutinin A–reactive fraction of alpha-fetoprotein, alpha-fetoprotein, and des-γ-carboxy prothrombin).Table 3Scoring of Remnant Liver Function by Serum Bilirubin and Albumin0 points1 point2 pointsSerum bilirubin (mg/dL)<1.01.0–2.0>2.0Serum albumin (g/dL)>3.52.8–3.5<2.8NOTE. Liver function was categorized by the sum of these 2 points (ie, biliburin and albumin) as scores A (0–1 points), B (2–3 points), and C (4 points) (bilirubin-albumin score). Open table in a new tab Table 4Calculation of the BALAD ScoreScore0123Bilirubin-albumin scoreaAs summarized in Table 3.ABCNo. of elevated tumor markers0123NOTE. BALAD score is calculated as the sum of the remnant liver function score (bilirubin-albumin score) and tumor progression score (no. of elevated tumor markers).a As summarized in Table 3. Open table in a new tab NOTE. Liver function was categorized by the sum of these 2 points (ie, biliburin and albumin) as scores A (0–1 points), B (2–3 points), and C (4 points) (bilirubin-albumin score). NOTE. BALAD score is calculated as the sum of the remnant liver function score (bilirubin-albumin score) and tumor progression score (no. of elevated tumor markers). Albumin, bilirubin, AFP, AFP-L3, and DCP were measured in the serum sample obtained from each patient at the time of HCC diagnosis. The serum AFP level was determined by enzyme-linked immunosorbent assay with a commercially available kit (ELISA-AFP; International Reagents, Kobe, Japan). Serum AFP-L3 was measured by lectin-affinity electrophoresis coupled with antibody-affinity blotting (AFP Differentiation Kit L; Wako Pure Chemical Industries, Ltd, Osaka, Japan) and was expressed as a percentage (AFP-L3 level/Total AFP level × 100).20Taketa K. Endo Y. Sekiya C. et al.A collaborative study for the evaluation of lectin-reactive alpha-fetoprotein in early detection of hepatocellular carcinoma.Cancer Res. 1993; 53: 5419-5423PubMed Google Scholar, 21Yamash*ta F. Tanaka M. Satomura S. et al.Prognostic significance of Lens culinaris agglutinin A-reactive alpha-fetoprotein in small hepatocellular carcinoma.Gastroenterology. 1996; 111: 996-1001Abstract Full Text PDF PubMed Scopus (161) Google Scholar The serum DCP level was determined by sensitive enzyme immunoassay (Eitest PIVKA-II kit; Eisai Laboratory, Tokyo, Japan) according to the manufacturer’s instructions.22Mita Y. Aoyagi Y. Yanagi M. et al.The usefulness of determining des-gamma-carboxy prothrombin by sensitive enzyme immunoassay in the early diagnosis of patients with hepatocellular carcinoma.Cancer. 1998; 82: 1643-1648Crossref PubMed Scopus (123) Google Scholar, 23Okuda H. Nakanishi T. Takatsu K. et al.Measurement of serum levels of des-gamma-carboxy prothrombin in patients with hepatocellular carcinoma by a revised enzyme immunoassay kit with increased sensitivity.Cancer. 1999; 85: 812-818Crossref PubMed Scopus (85) Google Scholar, 24Nomura F. Ishijima M. Kuwa K. et al.Serum des-gamma-carboxy prothrombin levels determined by a new generation of sensitive immunoassays in patients with small-sized hepatocellular carcinoma.Am J Gastroenterol. 1999; 94: 650-654Crossref PubMed Google Scholar To determine the best cutoff level for each tumor marker, we tested several cutoff levels that have been proposed: 20, 100, and 400 ng/dL for AFP; 10% and 15% for AFP-L3; and 40 and 100 mAU/mL for DCP.9The Cancer of the Liver Italian Program (CLIP) InvestigatorsA new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients.Hepatology. 1998; 28: 751-755Crossref PubMed Scopus (1187) Google Scholar, 25Oka H. Tamori A. Kuroki T. et al.Prospective study of alpha-fetoprotein in cirrhotic patients monitored for development of hepatocellular carcinoma.Hepatology. 1994; 19: 61-66Crossref PubMed Scopus (355) Google Scholar, 26Koda M. Murawaki Y. Mitsuda A. et al.Predictive factors for intrahepatic recurrence after percutaneous ethanol injection therapy for small hepatocellular carcinoma.Cancer. 2000; 88: 529-537Crossref PubMed Scopus (132) Google Scholar, 27Tsukuma H. Hiyama T. Tanaka S. et al.Risk factors for hepatocellular carcinoma among patients with chronic liver disease.N Engl J Med. 1993; 328: 1797-1801Crossref PubMed Scopus (1059) Google Scholar, 28Shimizu K. Taniichi T. Satomura S. et al.Establishment of assay kits for determination of microheterogeneities of alpha-fetoprotein using lectin-affinity electrophoresis.Clin Chim Acta. 1993; 214: 3-12Crossref PubMed Scopus (60) Google Scholar, 29Taketa K. Endo Y. Sekiya C. et al.A collaborative study for the evaluation of lectin-reactive α-fetoproteins in early detection of hepatocellular carcinoma.Cancer Res. 1993; 53: 5419-5423PubMed Google Scholar, 30Okuda H. Nakanishi T. Takatsu K. et al.Clinicopathologic features of patients with hepatocellular carcinoma seropositive for α-fetoprotein-L3 and seronegative for des-gamma-carboxy prothrombin in comparison with those seropositive for des-gamma-carboxy prothrombin alone.J Gastroenterol Hepatol. 2002; 17: 772-778Crossref PubMed Scopus (40) Google Scholar, 31Weitz I.C. Liebman H.A. Des-gamma-carboxy (abnormal) prothrombin and hepatocellular carcinoma: a critical review.Hepatology. 1993; 18: 990-997Crossref PubMed Scopus (136) Google Scholar, 32Fujiyama S. Izuno K. Gohshi K. et al.Clinical usefulness of des-gamma-carboxy prothrombin assay in early diagnosis of hepatocellular carcinoma.Dig Dis Sci. 1991; 36: 1787-1792Crossref PubMed Scopus (44) Google Scholar, 33Okuda H. Nakanishi T. Takatsu K. et al.Serum levels of des-gamma-carboxy prothrombin measured using the revised enzyme immunoassay kit with increased sensitivity in relation to clinicopathological features of solitary hepatocellular carcinoma.Cancer. 2000; 88: 544-549Crossref PubMed Scopus (98) Google Scholar Values are expressed as the median ± standard deviation. Univariate survival curves were calculated by the Kaplan-Meier method,34Kaplan E.L. Meier P. Non parametric estimation for incomplete observation.J Am Stat Assoc. 1958; 53: 457-481Crossref Scopus (47897) Google Scholar and differences in survival rates between groups were analyzed by the log-rank test.35Petro R. Pike M.C. Conservation of the approximation (0-E2)/E in the log rank test for survival data on tumor incidence data.Biometrics. 1973; 29: 579-584Crossref PubMed Scopus (625) Google Scholar The Bonferroni correction for multiple comparisons was applied. The Cox proportional hazards model36Cox D. Regression models and life tables.J R Stat Soc. 1972; 34: 187-220Google Scholar was used for multivariate analysis. The likelihood ratio test was performed to examine hom*ogeneity of patients within the same group and monotonicity of the gradient between groups. The Akaike information criteria (AIC)37Akaike H. A new look at statistical model identification.IEEE Transactions on Automatic Control. 1974; AU-19: 716-722Crossref Scopus (35473) Google Scholar were used to evaluate the discriminatory ability of the given model; the smaller the value of this statistic, the better the model.38Stone M. Akaike’s criteria.in: Armitage P. Colton T. Encyclopedia of biostatistics. Wiley, Chichester, UK1998: 123-124Google Scholar All analyses were performed with SAS statistical software (version 8.2; SAS Institute, Cary, NC) or the SPSS Medical Pack for Windows (version 10.0; SPSS, Inc, Chicago, IL). All P values were derived from 2-tailed tests, and P < .05 was accepted as statistically significant. The likelihood ratios and AIC for the BALAD score with different cutoff levels for the 3 tumor markers are shown in Table 5. The likelihood ratio was highest and the AIC was lowest when AFP was 400 ng/dL, AFP-L3 was 15%, and DCP was 100 mAU/mL. Thus, these values were selected as cutoff levels.Table 5Likelihood Ratio and AIC for the BALAD Score According to Various Combinations of Tumor Marker Cutoff LevelsCut off levelLikelihood ratioAICAFPAFP-L3DCP201040515.83116001.752010100555.94015957.76201540532.47015986.462015100573.30015939.491001040544.12815979.5610010100582.48015932.781001540554.42015966.4510015100588.13215915.724001040550.53315986.7040010100586.67215932.064001540557.50015973.1940015100595.77815915.05 Open table in a new tab When the cutoff levels were set as 400 ng/dL for AFP, 15% for AFP-L3, and 100 mAU/mL for DCP, the number of patients with AFP elevation was 524 (20.2%), that of patients with AFP-L3 elevation was 701 (27.0%), and that of patients with DCP elevation was 1253 (48.2%; Figure 1). All 3 tumor markers were elevated in 268 (10.3%) patients, 2 were elevated in 406 (15.6%) patients, 1 was elevated in 862 (33.2%) patients, and none was elevated in the remaining 1064 (40.9%) patients. Remnant liver function was category A in 1718 (66.1%) patients, category B in 785 (30.2%) patients, and category C in the remaining 97 (3.7%) patients. The BALAD score was 0 in 732 (28.1%) patients, 1 in 884 (34.0%), 2 in 524 (20.2%), 3 in 336 (12.9%), 4 in 103 (4.0%), and 5 in 21 (0.8%) patients. Table 6 shows the initial treatment for HCC applied for patients in each category of BALAD score. In BALAD scores 0 and 1, more than half of patients underwent curative treatment such as hepatectomy or LAT, the percentage of which decreased with the increase in the score. In contrast, the percentage of patients who did not receive treatment increased in higher BALAD scores.Table 6Comparison of a Treatment for HCC According to the BALAD ScoreHepatectomy (n = 727)LAT (n = 746)TACE (n = 702)OthersaIncludes repeated arterial infusion chemotherapy, systemic chemotherapy. (n = 257)None (n = 168)BALAD score 0 (n = 732)235 (32.1)315 (43.0)150 (20.5)24 (3.3)8 (1.1)BALAD score 1 (n = 884)278 (31.4)281 (31.8)236 (26.7)58 (6.6)31 (3.5)BALAD score 2 (n = 524)127 (24.2)104 (19.8)176 (33.6)71 (13.6)46 (8.8)BALAD score 3 (n = 336)77 (22.9)37 (11.0)109 (32.4)66 (19.7)47 (14.0)BALAD score 4 (n = 103)10 (9.7)6 (5.8)30 (29.1)33 (32.1)24 (23.3)BALAD score 5 (n = 21)03 (14.3)1 (4.8)5 (23.8)12 (57.1)NOTE. Percentage in parentheses.a Includes repeated arterial infusion chemotherapy, systemic chemotherapy. Open table in a new tab NOTE. Percentage in parentheses. Figure 2 shows patient survival according to the elevation of each tumor marker. Patients with elevated levels of AFP (>400 ng/mL), AFP-L3 (>15%), and DCP (>100 mAU/mL), respectively, exhibited significantly lower survival rate in comparison with those without the elevation of each tumor marker (all, P < .0001). We compared the discriminative ability of 2 systems that focus on tumor progression to predict patient survival, the TNM tumor classification of the Liver Cancer Study Group of Japan19Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. English ed. Kanehara & Co, Ltd, Tokyo, Japan2003Google Scholar and the number of elevated tumor markers in our new system. As is shown in Figure 3, both staging systems discriminated patient survival well (both P < .0001); patient survival categorized by the number of elevated tumor markers was comparable to that categorized by tumor stage. We also compared the discriminative ability of 2 systems that focus on remnant liver function to predict patient survival, the Child-Pugh classification and the serum albumin and serum total bilirubin levels of our new systems. Discrimination of patient survival categorized by serum albumin and total bilirubin levels was comparable to that by the Child-Pugh classifications (Figure 4; both P < .0001). When we analyzed discrimination of patient survival categorized on the basis of staging systems that incorporate tumor progression factors and liver function factors, ie, CLIP scoring system, JIS scoring system, and BALAD scoring system, all 3 scoring systems showed good prediction and discrimination of patient survival (Figure 5, Figure 6). When we compared the AIC of these 3 staging systems, the AIC of the BALAD score (15,915.05) was higher than that of the JIS score (15,782.09) but lower than that of the CLIP score (15,963.40).Figure 3Patient survival according to tumor progression. (A) According to TNM tumor stage by Liver Cancer Study Group of Japan; (B) according to number of elevated tumor markers. Both systems predicted patient survival well, and the discriminative

展开全部

机器翻译

AI理解论文&经典十问

参考文献

被引用

社区问答

参考文献

发布时间 · 被引用数 · 默认排序

被引用

发布时间 · 被引用数 · 默认排序

社区问答

领域

· Gastroenterology· Internal medicine· Radiology· Oncology· Biomarker· Scoring system· Multivariate analysis· Confidence interval· Predictive value of tests· Cirrhosis· Area under the curve· Receiver operating characteristic· Hepatocellular carcinoma· Medicine

Staging Hepatocellular Carcinoma by a Novel Scoring System (BALAD Score) Based on Serum Markers-论文阅读讨论-ReadPaper - 轻松读论文 | 专业翻译 | 一键引文 | 图表同屏 (2024)