TY - JOUR
AU1 - Poon, R T P
AU2 - Ho, J W Y
AU3 - Tong, C S W
AU4 - Lau, C
AU5 - Ng, I O L
AU6 - Fan, S-T
AB - Abstract Background Vascular endothelial growth factor (VEGF) and endostatin stimulate and inhibit tumour angiogenesis respectively. Recent studies have demonstrated the prognostic value of serum levels of both VEGF and endostatin in patients with various types of cancer. Their significance in patients with hepatocellular carcinoma (HCC) remains unclear. Methods Serum VEGF and endostatin levels were measured by enzyme immunoassay in 108 patients with HCC before surgical resection and in 20 healthy controls. Preoperative serum VEGF and endostatin levels were correlated with clinicopathological features and long-term survival. Results Serum VEGF levels in patients with HCC were significantly higher than those in controls, but serum levels of endostatin were similar in the two groups. High serum levels of VEGF, but not endostatin, were significantly associated with venous invasion and advanced tumour stage. Patients with a serum VEGF level higher than median (over 245·0 pg/ml) had significantly worse overall and disease-free survival than those with a lower level (P = 0·012 and P = 0·022 respectively). On multivariate analysis, serum VEGF level was an independent prognostic factor (hazard ratio 1·86 (95 per cent confidence interval 1·10 to 3·92); P = 0·032). Serum endostatin levels did not have significant prognostic influence on overall or disease-free survival. Conclusion A high serum level of VEGF is a predictor of poor outcome after resection of HCC. Serum VEGF, but not endostatin, may be a useful prognostic marker in patients with HCC. Introduction Angiogenesis is essential for tumour growth and metastasis1. Vascular endothelial growth factor (VEGF) is the most potent stimulator of angiogenesis known2. Its expression has been shown to correlate with tumour progression and prognosis in many cancers3–7. The serum VEGF level may be a surrogate marker of tumour VEGF expression, and has been shown to have prognostic significance in patients with cancer8–13. Several endogenous inhibitors of angiogenesis have been identified, of which endostatin is one of the most widely studied14,15. It is a peptide derived from cleavage of collagen XVIII, and may work in a negative feedback loop to suppress endothelial cell proliferation stimulated by VEGF15. Endostatin has attracted much attention in recent years because of its therapeutic potential in cancer therapy16, although the clinical significance of tumour expression of endostatin is unclear. Recent studies have suggested that the serum endostatin level may have prognostic value in lung cancer and soft tissue sarcoma17,18, but not in vulvar cancer or non-Hodgkin's lymphoma19,20. It has been suggested that measurement of serum levels of both VEGF and endostatin may provide better prognostic information as it indicates the balance between the stimulation and inhibition of angiogenesis21. Angiogenesis plays an important role in the progression of hepatocellular carcinoma (HCC)22,23, although little is known of the significance of various angiogenic and antiangiogenic factors24. Preliminary data suggest that VEGF is an important mediator of angiogenesis in HCC25,26, there being a significant association between high serum VEGF levels and venous invasion27. Two recent studies have evaluated the clinical significance of serum endostatin in patients with HCC28,29. One found that serum endostatin levels were not significantly increased in patients with HCC compared with healthy controls and that levels did not correlate with tumour stage28. In the other study, a high serum endostatin level was significantly associated with a low angiogenic score in the tumour29. Patients with a high serum endostatin level had a tendency towards better survival than those with a low level, although the prognostic impact was not clearly defined because of small numbers of patients and short duration of follow-up. The aim of the present study was to clarify the prognostic significance of preoperative serum VEGF and endostatin levels in patients undergoing resection of HCC. Patients and methods One hundred and eight patients (76 men and 32 women; median age 57·5 (range 16–75) years) who underwent curative resection of HCC, defined as macroscopically complete removal of the tumour, between January 1998 and June 2001 were studied. Eighty-six patients (79·6 per cent) tested positive for hepatitis B surface antigen, whereas only four patients (3·7 per cent) were positive for hepatitis C antibody. No patient had received any preoperative treatment. The study was approved by the ethics committee of the authors' institution. Preoperative serum samples were collected from patients with informed consent stored at −80 °C until VEGF and endostatin levels were determined. Serum samples were also obtained from 20 healthy subjects (14 men and six women; median age 55 (range 24–70) years) without evidence of active disease. Serum VEGF levels were measured by enzyme-linked immunosorbent assay (Quantikine Human VEGF®; R&D Systems, Minneapolis, Minnesota, USA)27. The assay exhibited no significant cross-reactivity with other angiogenic factors. The sensitivity of the assay was 9·0 pg/ml, and the intra-assay and interassay coefficients of variation were less than 6·7 per cent and 8·8 per cent respectively27. Serum endostatin levels were quantified by an enzyme-linked immunosorbent assay17–21 (Accucyte Human Endostatin®; Cytimmune Sciences, College Park, Maryland, USA)29. The sensitivity of the assay was 1·95 ng/ml, and the intra-assay and interassay coefficients of variation were both less than 10 per cent. VEGF and endostatin levels were measured in each serum sample in duplicate and the mean calculated. Measurements were made by an investigator who was blinded to the patients' clinicopathological data. All clinicopathological and follow-up data were recorded prospectively in a computerized database. Histopathological examination was performed by an experienced pathologist who was unaware of the serum VEGF or endostatin results. Tumours were graded according to the criteria described by Edmonson and Steiner30, and staged according to the pathological tumour node metastasis (pTNM) classification31. All patients were monitored regularly for tumour recurrence by measurement of serum α-fetoprotein (AFP) level, chest radiography, and either ultrasonography or computed tomography every 3 months. Disease recurrence was diagnosed by the detection of any tumour with an imaging appearance typical of HCC and a raised serum AFP level. Percutaneous fine-needle aspiration cytology was performed to confirm the diagnosis in uncertain cases. Statistical analysis Continuous data were expressed as median (range), and compared between groups using the Mann–Whitney U test. The χ2 test (or Fisher's exact test where appropriate) was used for comparison of categorical variables. Correlation between continuous variables was evaluated using the Spearman rank correlation coefficient. Cumulative survival was calculated by the Kaplan–Meier method and compared between groups by the log rank test. Cox's proportional hazards model was used for multivariate analysis of prognostic factors. All statistical analyses were performed using SPSS® 10·0 for Windows (SPSS, Chicago, Illinois, USA). P < 0·050 was considered statistically significant. Results Serum levels of VEGF in patients with HCC (median 245·0 (range 12·6–1824·0) pg/ml) were significantly higher than those in healthy controls (median 180·0 (range 40·8–671·2) pg/ml) (P = 0·042). The serum endostatin levels in patients with HCC (median 35·0 (range 2·5–75·6) ng/ml) were similar to those in controls (median 31·6 (range 2·9–41·6) ng/ml) (P = 0·341). Serum VEGF and endostatin levels were significantly correlated in patients with HCC (rs = 0·224, P = 0·028) but not in controls (rs = 0·124, P = 0·456). No significant relationship was observed between serum VEGF levels and sex, age, hepatitis B surface antigen status, underlying cirrhosis or serum AFP level. A positive correlation was observed between serum VEGF level and platelet count (rs = 0·467, P < 0·001). Significantly higher serum VEGF levels were associated with tumour size greater than 5 cm, high tumour grade, presence of microscopic venous invasion, presence of microscopic satellite lesions and advanced pTNM stage (Table 1). Table 1 Preoperative serum vascular endothelial growth factor and endostatin levels categorized by clinicopathological variables Variables . No. of patients . Serum VEGF (pg/ml) . P* . Serum endostatin (ng/ml) . P* . Sex 0·856 0·335  Male 76 236·5 (12·6–1824·0) 34·0 (2·5–75·6)  Female 32 257·4 (34·5–1788·8) 37·6 (7·5–72·9) Age (years) 0·395 0·204  ≤ 65 84 221·4 (12·6–1824·0) 32·8 (2·5–75·6)  > 65 24 257·4 (8·1–1788·8) 39·8 (9·5–72·9) HBsAg 0·628 0·090  Positive 86 221·4 (12·6–1824·0) 34·0 (2·5–75·6)  Negative 22 257·4 (34·5–1728·0) 39·8 (9·5–70·0) Underlying cirrhosis 0·112 0·013  Yes 48 216·5 (12·6–1572·0) 30·3 (2·5–69·2)  No 60 262·8 (33·6–1824·0) 39·3 (3·9–75·6) Serum AFP level (ng/ml) 0·677 0·349  ≤ 20 33 214·3 (33·6–1824·0) 39·5 (9·5–70·8)  > 20 75 248·9 (8·1–1788·8) 34·0 (2·5–75·6) Tumour size (cm) 0·008 0·024  ≤ 5 45 193·4 (12·6–655·3) 30·2 (3·9–69·2)  > 5 63 354·7 (14·9–1824·0) 38·7 (2·5–75·6) Tumour number 0·817 0·483  Solitary 80 247·8 (12·6–1824·0) 34·3 (2·5–75·6)  Multiple 28 232·4 (14·9–1438·0) 39·3 (9·5–69·9) Edmonson grade 0·044 0·453  1/2 56 226·6 (12·6–1438·0) 32·6 (3·9–75·6)  3/4 52 344·1 (34·5–1824·0) 36·3 (2·5–69·2) Tumour capsule 0·317 0·783  No 71 257·4 (12·6–1824·0) 36·3 (3·9–75·6)  Yes 37 216·5 (14·9–1788·8) 32·6 (2·5–64·1) Venous invasion 0·004 0·613  No 52 201·6 (14·9–1788·0) 35·7 (2·5–75·6)  Yes 56 342·5 (12·6–1824·0) 36·3 (8·0–72·9) Microsatellite lesions 0·016 0·379  No 59 201·6 (12·6–1728·0) 34·0 (2·5–75·6)  Yes 49 344·1 (34·5–1824·0) 38·7 (7·5–67·0) pTNM stage 0·001 0·134  I/II 66 211·1 (12·6–1390·6) 32·7 (2·5–70·8)  IIIA/IIIB 42 411·7 (14·9–1824·0) 36·8 (9·5–75·6) Variables . No. of patients . Serum VEGF (pg/ml) . P* . Serum endostatin (ng/ml) . P* . Sex 0·856 0·335  Male 76 236·5 (12·6–1824·0) 34·0 (2·5–75·6)  Female 32 257·4 (34·5–1788·8) 37·6 (7·5–72·9) Age (years) 0·395 0·204  ≤ 65 84 221·4 (12·6–1824·0) 32·8 (2·5–75·6)  > 65 24 257·4 (8·1–1788·8) 39·8 (9·5–72·9) HBsAg 0·628 0·090  Positive 86 221·4 (12·6–1824·0) 34·0 (2·5–75·6)  Negative 22 257·4 (34·5–1728·0) 39·8 (9·5–70·0) Underlying cirrhosis 0·112 0·013  Yes 48 216·5 (12·6–1572·0) 30·3 (2·5–69·2)  No 60 262·8 (33·6–1824·0) 39·3 (3·9–75·6) Serum AFP level (ng/ml) 0·677 0·349  ≤ 20 33 214·3 (33·6–1824·0) 39·5 (9·5–70·8)  > 20 75 248·9 (8·1–1788·8) 34·0 (2·5–75·6) Tumour size (cm) 0·008 0·024  ≤ 5 45 193·4 (12·6–655·3) 30·2 (3·9–69·2)  > 5 63 354·7 (14·9–1824·0) 38·7 (2·5–75·6) Tumour number 0·817 0·483  Solitary 80 247·8 (12·6–1824·0) 34·3 (2·5–75·6)  Multiple 28 232·4 (14·9–1438·0) 39·3 (9·5–69·9) Edmonson grade 0·044 0·453  1/2 56 226·6 (12·6–1438·0) 32·6 (3·9–75·6)  3/4 52 344·1 (34·5–1824·0) 36·3 (2·5–69·2) Tumour capsule 0·317 0·783  No 71 257·4 (12·6–1824·0) 36·3 (3·9–75·6)  Yes 37 216·5 (14·9–1788·8) 32·6 (2·5–64·1) Venous invasion 0·004 0·613  No 52 201·6 (14·9–1788·0) 35·7 (2·5–75·6)  Yes 56 342·5 (12·6–1824·0) 36·3 (8·0–72·9) Microsatellite lesions 0·016 0·379  No 59 201·6 (12·6–1728·0) 34·0 (2·5–75·6)  Yes 49 344·1 (34·5–1824·0) 38·7 (7·5–67·0) pTNM stage 0·001 0·134  I/II 66 211·1 (12·6–1390·6) 32·7 (2·5–70·8)  IIIA/IIIB 42 411·7 (14·9–1824·0) 36·8 (9·5–75·6) Values are median (range). VEGF, vascular endothelial growth factor; HBsAg, hepatitis B surface antigen; AFP, α-fetoprotein; pTNM, pathological tumour node metastasis. * Mann–Whitney U test. Open in new tab Table 1 Preoperative serum vascular endothelial growth factor and endostatin levels categorized by clinicopathological variables Variables . No. of patients . Serum VEGF (pg/ml) . P* . Serum endostatin (ng/ml) . P* . Sex 0·856 0·335  Male 76 236·5 (12·6–1824·0) 34·0 (2·5–75·6)  Female 32 257·4 (34·5–1788·8) 37·6 (7·5–72·9) Age (years) 0·395 0·204  ≤ 65 84 221·4 (12·6–1824·0) 32·8 (2·5–75·6)  > 65 24 257·4 (8·1–1788·8) 39·8 (9·5–72·9) HBsAg 0·628 0·090  Positive 86 221·4 (12·6–1824·0) 34·0 (2·5–75·6)  Negative 22 257·4 (34·5–1728·0) 39·8 (9·5–70·0) Underlying cirrhosis 0·112 0·013  Yes 48 216·5 (12·6–1572·0) 30·3 (2·5–69·2)  No 60 262·8 (33·6–1824·0) 39·3 (3·9–75·6) Serum AFP level (ng/ml) 0·677 0·349  ≤ 20 33 214·3 (33·6–1824·0) 39·5 (9·5–70·8)  > 20 75 248·9 (8·1–1788·8) 34·0 (2·5–75·6) Tumour size (cm) 0·008 0·024  ≤ 5 45 193·4 (12·6–655·3) 30·2 (3·9–69·2)  > 5 63 354·7 (14·9–1824·0) 38·7 (2·5–75·6) Tumour number 0·817 0·483  Solitary 80 247·8 (12·6–1824·0) 34·3 (2·5–75·6)  Multiple 28 232·4 (14·9–1438·0) 39·3 (9·5–69·9) Edmonson grade 0·044 0·453  1/2 56 226·6 (12·6–1438·0) 32·6 (3·9–75·6)  3/4 52 344·1 (34·5–1824·0) 36·3 (2·5–69·2) Tumour capsule 0·317 0·783  No 71 257·4 (12·6–1824·0) 36·3 (3·9–75·6)  Yes 37 216·5 (14·9–1788·8) 32·6 (2·5–64·1) Venous invasion 0·004 0·613  No 52 201·6 (14·9–1788·0) 35·7 (2·5–75·6)  Yes 56 342·5 (12·6–1824·0) 36·3 (8·0–72·9) Microsatellite lesions 0·016 0·379  No 59 201·6 (12·6–1728·0) 34·0 (2·5–75·6)  Yes 49 344·1 (34·5–1824·0) 38·7 (7·5–67·0) pTNM stage 0·001 0·134  I/II 66 211·1 (12·6–1390·6) 32·7 (2·5–70·8)  IIIA/IIIB 42 411·7 (14·9–1824·0) 36·8 (9·5–75·6) Variables . No. of patients . Serum VEGF (pg/ml) . P* . Serum endostatin (ng/ml) . P* . Sex 0·856 0·335  Male 76 236·5 (12·6–1824·0) 34·0 (2·5–75·6)  Female 32 257·4 (34·5–1788·8) 37·6 (7·5–72·9) Age (years) 0·395 0·204  ≤ 65 84 221·4 (12·6–1824·0) 32·8 (2·5–75·6)  > 65 24 257·4 (8·1–1788·8) 39·8 (9·5–72·9) HBsAg 0·628 0·090  Positive 86 221·4 (12·6–1824·0) 34·0 (2·5–75·6)  Negative 22 257·4 (34·5–1728·0) 39·8 (9·5–70·0) Underlying cirrhosis 0·112 0·013  Yes 48 216·5 (12·6–1572·0) 30·3 (2·5–69·2)  No 60 262·8 (33·6–1824·0) 39·3 (3·9–75·6) Serum AFP level (ng/ml) 0·677 0·349  ≤ 20 33 214·3 (33·6–1824·0) 39·5 (9·5–70·8)  > 20 75 248·9 (8·1–1788·8) 34·0 (2·5–75·6) Tumour size (cm) 0·008 0·024  ≤ 5 45 193·4 (12·6–655·3) 30·2 (3·9–69·2)  > 5 63 354·7 (14·9–1824·0) 38·7 (2·5–75·6) Tumour number 0·817 0·483  Solitary 80 247·8 (12·6–1824·0) 34·3 (2·5–75·6)  Multiple 28 232·4 (14·9–1438·0) 39·3 (9·5–69·9) Edmonson grade 0·044 0·453  1/2 56 226·6 (12·6–1438·0) 32·6 (3·9–75·6)  3/4 52 344·1 (34·5–1824·0) 36·3 (2·5–69·2) Tumour capsule 0·317 0·783  No 71 257·4 (12·6–1824·0) 36·3 (3·9–75·6)  Yes 37 216·5 (14·9–1788·8) 32·6 (2·5–64·1) Venous invasion 0·004 0·613  No 52 201·6 (14·9–1788·0) 35·7 (2·5–75·6)  Yes 56 342·5 (12·6–1824·0) 36·3 (8·0–72·9) Microsatellite lesions 0·016 0·379  No 59 201·6 (12·6–1728·0) 34·0 (2·5–75·6)  Yes 49 344·1 (34·5–1824·0) 38·7 (7·5–67·0) pTNM stage 0·001 0·134  I/II 66 211·1 (12·6–1390·6) 32·7 (2·5–70·8)  IIIA/IIIB 42 411·7 (14·9–1824·0) 36·8 (9·5–75·6) Values are median (range). VEGF, vascular endothelial growth factor; HBsAg, hepatitis B surface antigen; AFP, α-fetoprotein; pTNM, pathological tumour node metastasis. * Mann–Whitney U test. Open in new tab No significant relationship was observed between serum levels of endostatin and sex, age, hepatitis B surface antigen status or serum AFP level. Serum endostatin levels were significantly lower in patients with cirrhosis than in those without (P = 0·013). A positive correlation was also observed between serum endostatin level and platelet count (rs = 0·295, P = 0·002). When serum levels of VEGF and endostatin were expressed with respect to platelet count there was no significant correlation between serum VEGF level per platelet and serum endostatin level per platelet (rs = 0·013, P = 0·892). There was a significant association between high serum endostatin levels and tumour size greater than 5 cm. No significant relationship was observed between serum endostatin levels and other pathological features (Table 1). The median duration of follow-up of the 108 patients with HCC was 42 (range 22–62) months. When stratified into those with high and low serum VEGF levels, and those with high and low endostatin levels, using the median concentrations as cut-off values, 1- and 3-year overall survival rates were 83 and 71 per cent in those with low serum levels of VEGF (less than or equal to 245·0 pg/ml), and 69 and 51 per cent respectively in patients with high levels (more than 245·0 pg/ml) (P = 0·012) (Fig. 1). The disease-free survival of patients with low serum VEGF levels was also significantly better than that of patients with high levels (median disease-free survival 21·5 versus 8·2 months; P = 0·022). Fig. 1 Open in new tabDownload slide Cumulative overall survival curves of patients with high (more than 245·0 pg/ml) and low (less than or equal to 245·0 pg/ml) serum levels of vascular endothelial growth factor (VEGF). P = 0·012 (log rank test) There were no significant differences between patients with low serum levels of endostatin (less than or equal to 35·0 ng/ml) and those with high levels (more than 35·0 ng/ml) in either overall survival (P = 0·124) (Fig. 2) or disease-free survival (median disease-free survival 19·5 versus 9·8 months; P = 0·198). Fig. 2 Open in new tabDownload slide Cumulative overall survival curves of patients with high (more than 35·0 ng/ml) and low (less than or equal to 35·0 ng/ml) serum levels of endostatin. P = 0·124 (log rank test) Analysis based on further stratification into four subgroups according to both VEGF and endostatin levels (low VEGF and low endostatin, 33 patients; low VEGF and high endostatin, 21; high VEGF and low endostatin, 21; high VEGF and high endostatin, 33), revealed significantly better survival in the first two groups with low VEGF levels compared with the latter two groups with high VEGF levels. There were no significant differences in survival between the first two groups or between the last two groups. Univariate analysis of 12 clinicopathological features that might influence prognosis according to previous studies32 was performed. Six of these factors had significant prognostic influence in the univariate analysis (Table 2). Table 2 Significant prognostic factors of overall survival by univariate analysis . No. of patients . 3-year survival (%) . P* . Cirrhosis 0·027  No 60 72·7  Yes 48 49·9 Serum AFP (ng/ml) 0·047  ≤ 20 33 74·9  > 20 75 50·4 Tumour number 0·032  Solitary 80 65·8  Multiple 28 46·8 Venous invasion 0·007  No 52 73·3  Yes 56 49·9 Microsatellite lesions 0·006  No 59 74·2  Yes 49 48·1 pTNM stage 0·001  I or II 66 75·6  IIIA or IIIB 42 47·9 . No. of patients . 3-year survival (%) . P* . Cirrhosis 0·027  No 60 72·7  Yes 48 49·9 Serum AFP (ng/ml) 0·047  ≤ 20 33 74·9  > 20 75 50·4 Tumour number 0·032  Solitary 80 65·8  Multiple 28 46·8 Venous invasion 0·007  No 52 73·3  Yes 56 49·9 Microsatellite lesions 0·006  No 59 74·2  Yes 49 48·1 pTNM stage 0·001  I or II 66 75·6  IIIA or IIIB 42 47·9 AFP, α-fetoprotein; pTNM, pathological tumour node metastasis. * Log rank test. Open in new tab Table 2 Significant prognostic factors of overall survival by univariate analysis . No. of patients . 3-year survival (%) . P* . Cirrhosis 0·027  No 60 72·7  Yes 48 49·9 Serum AFP (ng/ml) 0·047  ≤ 20 33 74·9  > 20 75 50·4 Tumour number 0·032  Solitary 80 65·8  Multiple 28 46·8 Venous invasion 0·007  No 52 73·3  Yes 56 49·9 Microsatellite lesions 0·006  No 59 74·2  Yes 49 48·1 pTNM stage 0·001  I or II 66 75·6  IIIA or IIIB 42 47·9 . No. of patients . 3-year survival (%) . P* . Cirrhosis 0·027  No 60 72·7  Yes 48 49·9 Serum AFP (ng/ml) 0·047  ≤ 20 33 74·9  > 20 75 50·4 Tumour number 0·032  Solitary 80 65·8  Multiple 28 46·8 Venous invasion 0·007  No 52 73·3  Yes 56 49·9 Microsatellite lesions 0·006  No 59 74·2  Yes 49 48·1 pTNM stage 0·001  I or II 66 75·6  IIIA or IIIB 42 47·9 AFP, α-fetoprotein; pTNM, pathological tumour node metastasis. * Log rank test. Open in new tab When serum VEGF level (more than 245·0 pg/ml versus less than or equal to 245·0 pg/ml) was entered into a Cox regression analysis together with the six factors in Table 2, serum VEGF level, underlying cirrhosis and advanced pTNM stage were identified as independent prognostic factors of overall survival (Table 3). Table 3 Significant prognostic factors of overall survival by Cox regression analysis . Hazard ratio . P . Serum VEGF level (> 245·0 versus < 245·0 pg/ml) 1·86 (1·10, 3·92) 0·032 Underlying cirrhosis (present versus absent) 2·15 (1·16, 3·60) 0·015 Tumour pTNM stage (IIIA or IIIB versus I or II) 2·40 (1·52, 3·46) 0·001 . Hazard ratio . P . Serum VEGF level (> 245·0 versus < 245·0 pg/ml) 1·86 (1·10, 3·92) 0·032 Underlying cirrhosis (present versus absent) 2·15 (1·16, 3·60) 0·015 Tumour pTNM stage (IIIA or IIIB versus I or II) 2·40 (1·52, 3·46) 0·001 Values in parentheses are 95 per cent confidence intervals. VEGF, vascular endothelial growth factor; pTNM, pathological tumour node metastasis. Open in new tab Table 3 Significant prognostic factors of overall survival by Cox regression analysis . Hazard ratio . P . Serum VEGF level (> 245·0 versus < 245·0 pg/ml) 1·86 (1·10, 3·92) 0·032 Underlying cirrhosis (present versus absent) 2·15 (1·16, 3·60) 0·015 Tumour pTNM stage (IIIA or IIIB versus I or II) 2·40 (1·52, 3·46) 0·001 . Hazard ratio . P . Serum VEGF level (> 245·0 versus < 245·0 pg/ml) 1·86 (1·10, 3·92) 0·032 Underlying cirrhosis (present versus absent) 2·15 (1·16, 3·60) 0·015 Tumour pTNM stage (IIIA or IIIB versus I or II) 2·40 (1·52, 3·46) 0·001 Values in parentheses are 95 per cent confidence intervals. VEGF, vascular endothelial growth factor; pTNM, pathological tumour node metastasis. Open in new tab Discussion Previous studies revealed a significant relationship between high circulating VEGF levels and the presence of venous invasion or metastasis in patients with HCC27,33,34, but did not clarify whether serum VEGF had any independent prognostic significance. In the present study, serum VEGF level was a predictor of long-term survival independently of pathological variables such as venous invasion and tumour stage. This may have clinical implications. The pTNM stage, which embraces pathological variables such as tumour multiplicity and venous invasion, has been shown previously to provide prognostic information after resection of HCC35. However, there is substantial variation in prognosis among patients within the same stage. The serum VEGF level might provide additional prognostic information and allow selection of patients at high risk of tumour recurrence for neoadjuvant or adjuvant therapy. This study also evaluated the prognostic significance of the serum endostatin level. Serum endostatin was not a useful prognostic marker in patients with HCC. In agreement with Yamagata et al.28, there was no significant difference in serum endostatin levels between patients with HCC and healthy controls. In the only other report of serum endostatin levels in patients with HCC, no comparison was made with healthy controls29. There was no significant correlation between serum endostatin levels and tumour invasiveness or tumour stage, in agreement with previous findings28,29. Dhar et al.29 reported that patients with high serum endostatin levels had a tendency towards longer survival, suggesting that the serum endostatin level might reflect the antiangiogenic activity of endostatin in the tumour. Other studies have shown that high serum or plasma endostatin levels predicted worse rather than better survival17,18,36,37. Circulating endostatin levels may reflect the tumour burden rather than antiangiogenic activity in the tumour36,37. The relationship between serum endostatin levels and tumour size demonstrated in the present study may also reflect an association between serum endostatin levels and tumour burden. Serum endostatin levels had no significant influence on overall or disease-free survival in the present study. The reason for the discrepancy in the findings between the present study and that of Dhar et al.29 is not clear. Most of the present HCCs were related to hepatitis B, whereas most in the study of Dhar et al.29 were associated with hepatitis C. In the latter study, patients with hepatitis C-related HCCs had significantly lower serum endostatin levels than those with hepatitis B-related tumours. Further studies are required to clarify any difference in the role of endostatin in HCCs related to hepatitis B virus and those related to hepatitis C virus. In addition, Dhar et al.29 did not evaluate the relationship between serum endostatin levels and underlying cirrhosis. In the present study patients with cirrhosis had significantly lower serum endostatin levels than those without. Angiogenesis also plays a role in the development of liver cirrhosis38. The role of endostatin in the pathophysiological mechanism of cirrhosis merits further study. Both serum VEGF and serum endostatin levels correlated positively with platelet count. A significant correlation was also observed between serum VEGF and endostatin levels, although the correlation disappeared when serum concentrations were expressed with respect to the platelet count. This suggests that the correlation between serum VEGF and endostatin levels was mainly indirect and related to the platelet count. Both VEGF and endostatin are stored in platelets, and platelets may play a role in scavenging tumour-derived angiogenic and antiangiogenic factors to prevent unwanted effects in normal tissues8,39,40. Because serum VEGF levels represent both free circulating VEGF and that stored in platelets and released during clotting, there has been some debate as to whether serum or plasma VEGF level is the better surrogate marker of tumour angiogenesis8. Most previous studies that demonstrated a significant prognostic value of circulating VEGF measured levels in serum8. A recent study that measured matched preoperative plasma and serum VEGF levels in patients with colorectal carcinoma showed that a high serum, but not plasma, VEGF level was an independent predictor of survival10. As the platelet load of VEGF may reflect VEGF released from the tumour and sequestered in platelets, it seems reasonable to measure serum VEGF levels as a reflection of tumour angiogenic activity41. Measurement of circulating angiogenic factors has obvious advantages over evaluation of angiogenic activity in tumour specimens: it is technically simpler, no tumour specimen is required and it can provide prognostic information before resection of the tumour. In this study, as in others12,13,20, the median serum VEGF level was taken as the cut-off value, rather than using the upper limit of serum VEGF levels in the healthy controls. The small number of control subjects and the wide variation in serum VEGF levels made it difficult to define a reliable cut-off value from the control group. It is unclear why some healthy people have a high serum VEGF level, although physiological angiogenesis, such as the reproductive cycle in women, may contribute to the variation of VEGF levels among healthy controls42. Serum VEGF may be a useful prognostic marker in patients with HCC. Acknowledgements This study was supported by a grant from the Committee on Research and Conference Grants of the University of Hong Kong References 1 Folkman J . Clinical applications of research on angiogenesis . N Engl J Med 1995 ; 333 : 1757 – 1763 . Google Scholar OpenURL Placeholder Text WorldCat 2 Toi M , Matsumoto T, Bando H. Vascular endothelial growth factor: its prognostic, predictive, and therapeutic implications . Lancet Oncol 2001 ; 2 : 667 – 673 . Google Scholar OpenURL Placeholder Text WorldCat 3 Takahashi Y , Tucker SL, Kitadai Y, Koura AN, Bucana CD, Cleary KR et al. Vessel counts and expression of vascular endothelial growth factor as prognostic factors in node-negative colon cancer . Arch Surg 1997 ; 132 : 541 – 546 . Google Scholar OpenURL Placeholder Text WorldCat 4 Linderholm B , Tavelin B, Grankvist K, Henriksson R. Vascular endothelial growth factor is of high prognostic value in node-negative breast carcinoma . J Clin Oncol 1998 ; 16 : 3121 – 3128 . Google Scholar OpenURL Placeholder Text WorldCat 5 Imoto H , Osaki T, Taga S, Ohgami A, Ichiyoshi Y, Yasumoto K. Vascular endothelial growth factor expression in non-small-cell lung cancer: prognostic significance in squamous cell carcinoma . J Thorac Cardiovasc Surg 1998 ; 115 : 1007 – 1114 . Google Scholar OpenURL Placeholder Text WorldCat 6 Saito H , Tsujitani S, Kondo A, Ikeguchi M, Maeta M, Kaibara N. Expression of vascular endothelial growth factor correlates with hematogenous recurrence in gastric carcinoma . Surgery 1999 ; 125 : 195 – 201 . Google Scholar OpenURL Placeholder Text WorldCat 7 Smith BD , Smith GL, Carter D, Sasaki CT, Haffty BG. Prognostic significance of vascular endothelial growth factor protein levels in oral and oropharyngeal squamous cell carcinoma . J Clin Oncol 2000 ; 18 : 2046 – 2052 . Google Scholar OpenURL Placeholder Text WorldCat 8 Poon RT , Fan ST, Wong J. Clinical implications of circulating angiogenic factors in cancer patients . J Clin Oncol 2001 ; 19 : 1207 – 1225 . Google Scholar OpenURL Placeholder Text WorldCat 9 Miyake H , Hara I, Yamanaka K, Gohji K, Arakawa S, Kamidono S. Elevation of serum level of vascular endothelial growth factor as a new predictor of recurrence and disease progression in patients with superficial urothelial cancer . Urology 1999 ; 53 : 302 – 307 . Google Scholar OpenURL Placeholder Text WorldCat 10 Werther K , Christensen IJ, Nielsen HJ. Prognostic impact of matched preoperative plasma and serum VEGF in patients with primary colorectal carcinoma . Br J Cancer 2002 ; 86 : 417 – 423 . Google Scholar OpenURL Placeholder Text WorldCat 11 Karayiannakis AJ , Syrigos KN, Polychronidis A, Zbar A, Kouraklis G, Simopoulos C et al. Circulating VEGF levels in the serum of gastric cancer patients: correlation with pathological variables, patient survival, and tumor surgery . Ann Surg 2002 ; 236 : 37 – 42 . Google Scholar OpenURL Placeholder Text WorldCat 12 Teknos TN , Cox C, Yoo S, Chepeha DB, Wolf GT, Bradford CR et al. Elevated serum vascular endothelial growth factor and decreased survival in advanced laryngeal carcinoma . Head Neck 2002 ; 24 : 1004 – 1011 . Google Scholar OpenURL Placeholder Text WorldCat 13 Cooper BC , Ritchie JM, Broghammer CL, Coffin J, Sorosky JI, Buller RE et al. Preoperative serum vascular endothelial growth factor levels: significance in ovarian cancer . Clin Cancer Res 2002 ; 8 : 3193 – 3197 . Google Scholar OpenURL Placeholder Text WorldCat 14 O'Reilly MS , Boehm T, Shing Y, Fukai N, Vasios G, Lane WS et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth . Cell 1997 ; 88 : 277 – 285 . Google Scholar OpenURL Placeholder Text WorldCat 15 Sasaki T , Hohenester E, Timpl R. Structure and function of collagen-derived endostatin inhibitors of angiogenesis . IUBMB Life 2002 ; 53 : 77 – 84 . Google Scholar OpenURL Placeholder Text WorldCat 16 Thomas JP , Arzoomanian RZ, Alberti D, Marnocha R, Lee F, Friedl A et al. Phase I pharmacokinetic and pharmacodynamic study of recombinant human endostatin in patients with advanced solid tumors . J Clin Oncol 2003 ; 21 : 223 – 231 . Google Scholar OpenURL Placeholder Text WorldCat 17 Suzuki M , Iizasa T, Ko E, Baba M, Saitoh Y, Shibuya K et al. Serum endostatin correlates with progression and prognosis of non-small cell lung cancer . Lung Cancer 2002 ; 35 : 29 – 34 . Google Scholar OpenURL Placeholder Text WorldCat 18 Feldman AL , Pak H, Yang JC, Alexander HR Jr, Libutti SK. Serum endostatin levels are elevated in patients with soft tissue sarcoma . Cancer 2001 ; 91 : 1525 – 1529 . Google Scholar OpenURL Placeholder Text WorldCat 19 Hefler L , Tempfer C, Kainz C, Obermair A. Serum concentrations of endostatin in patients with vulvar cancer . Gynecol Oncol 1999 ; 74 : 151 – 152 . Google Scholar OpenURL Placeholder Text WorldCat 20 Bertolini F , Paolucci M, Peccatori F, Cinieri S, Agazzi A, Ferrucci PF et al. Angiogenic growth factors and endostatin in non-Hodgkin's lymphoma . Br J Haematol 1999 ; 106 : 504 – 509 . Google Scholar OpenURL Placeholder Text WorldCat 21 Shaarawy M , El-Sharkawy SA. Biomarkers of intrinsic angiogenic and anti-angiogenic activity in patients with endometrial hyperplasia and endometrial cancer . Acta Oncol 2001 ; 40 : 513 – 518 . Google Scholar OpenURL Placeholder Text WorldCat 22 Tanigawa N , Lu C, Mitsui T, Miura S. Quantitation of sinusoid-like vessels in hepatocellular carcinoma: its clinical and prognostic significance . Hepatology 1997 ; 26 : 1216 – 1223 . Google Scholar OpenURL Placeholder Text WorldCat 23 Poon RT , Ng IO, Lau C, Yu WC, Yang ZF, Fan ST et al. Tumor microvessel density as a predictor of recurrence after resection of hepatocellular carcinoma: a prospective study . J Clin Oncol 2002 ; 20 : 1775 – 1785 . Google Scholar OpenURL Placeholder Text WorldCat 24 Poon RT , Fan ST, Wong J. Clinical significance of angiogenesis in gastrointestinal cancers: A target for novel prognostic and therapeutic approaches . Ann Surg 2003 ; 238 : 9 – 28 . Google Scholar OpenURL Placeholder Text WorldCat 25 Li XM , Tang ZY, Zhou G, Lui YK, Ye SL. Significance of vascular endothelial growth factor mRNA expression in invasion and metastasis of hepatocellular carcinoma . J Exp Clin Cancer Res 1998 ; 17 : 13 – 17 . Google Scholar OpenURL Placeholder Text WorldCat 26 Ng IO , Poon RT, Lee JM, Fan ST, Ng M, Tso WK. Microvessel density, vascular endothelial growth factor and its receptors Flt-1 and Flk-1/KDR in hepatocellular carcinoma . Am J Clin Pathol 2001 ; 116 : 838 – 845 . Google Scholar OpenURL Placeholder Text WorldCat 27 Poon RT , Ng IO, Lau C, Zhu LX, Yu WC, Lo CM et al. Serum vascular endothelial growth factor predicts venous invasion in hepatocellular carcinoma: a prospective study . Ann Surg 2001 ; 233 : 227 – 235 . Google Scholar OpenURL Placeholder Text WorldCat 28 Yamagata M , Shiratori Y, Dan Y, Shiina S, Takayama T, Makuuchi M et al. Serum endostatin levels in patients with hepatocellular carcinoma . Ann Oncol 2000 ; 11 : 761 – 762 . Google Scholar OpenURL Placeholder Text WorldCat 29 Dhar DK , Ono T, Yamanoi A, Soda Y, Yamaguchi E, Rahman MA et al. Serum endostatin predicts tumor vascularity in hepatocellular carcinoma . Cancer 2002 ; 95 : 2188 – 2195 . Google Scholar OpenURL Placeholder Text WorldCat 30 Edmonson HA , Steiner PE. Primary carcinoma of the liver: a study of 100 among 48 900 necropsies . Cancer 1954 ; 7 : 462 – 503 . Google Scholar OpenURL Placeholder Text WorldCat 31 Liver including intrahepatic bile ducts. In American Joint Committee on Cancer Staging Manual (AJCC) (6th edn), Greene FL, Page DL, Fleming ID, Fritz AG, Balch CM, Haller DG et al. (eds). Springer : New York , 2002 ; 131 – 144 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 32 Poon RT , Fan ST, Wong J. Risk factors, prevention and management of postoperative recurrence after resection of hepatocellular carcinoma . Ann Surg 2000 ; 232 : 10 – 24 . Google Scholar OpenURL Placeholder Text WorldCat 33 Jinno K , Tanimizu M, Hyodo I, Nishikawa Y, Hosokawa Y, Doi T et al. Circulating vascular endothelial growth factor (VEGF) is a possible tumor marker for metastasis in human hepatocellular carcinoma . J Gastroenterol 1998 ; 33 : 376 – 382 . Google Scholar OpenURL Placeholder Text WorldCat 34 Li XM , Tang ZY, Qin LX, Zhou J, Sun HC. Serum vascular endothelial growth factor is a predictor of invasion and metastasis in hepatocellular carcinoma . J Exp Clin Cancer Res 1999 ; 18 : 511 – 517 . Google Scholar OpenURL Placeholder Text WorldCat 35 Poon RT , Fan ST. Evaluation of the new AJCC/UICC staging system for hepatocellular carcinoma after hepatic resection in Chinese patients . Surg Oncol Clin N Am 2003 ; 12 : 35 – 50 . Google Scholar OpenURL Placeholder Text WorldCat 36 Feldman AL , Alexander HR Jr, Bartlett DL, Kranda KC, Miller MS, Costouros NG et al. A prospective analysis of plasma endostatin levels in colorectal cancer patients with liver metastases . Ann Surg Oncol 2001 ; 8 : 741 – 745 . Google Scholar OpenURL Placeholder Text WorldCat 37 Lai R , Estey E, Shen Y, Despa S, Kantarjian H, Beran M et al. Clinical significance of plasma endostatin in acute myeloid leukemia/myelodysplastic syndrome . Cancer 2002 ; 94 : 14 – 17 . Google Scholar OpenURL Placeholder Text WorldCat 38 Rosmorduc O , Wendum D, Corpechot C, Galy B, Sebbagh N, Raleigh J et al. Hepatocellular hypoxia-induced vascular endothelial growth factor expression and angiogenesis in experimental biliary cirrhosis . Am J Pathol 1999 ; 155 : 1065 – 1073 . Google Scholar OpenURL Placeholder Text WorldCat 39 George ML , Eccles SA, Tutton MG, Abulafi AM, Swift RI. Correlation of plasma and serum vascular endothelial growth factor levels with platelet count in colorectal cancer: clinical evidence of platelet scavenging? Clin Cancer Res 2000 ; 6 : 3147 – 3152 . Google Scholar OpenURL Placeholder Text WorldCat 40 Ma L , Hollenberg MD, Wallace JL. Thrombin-induced platelet endostatin release is blocked by a proteinase activated receptor-4 (PAR4) antagonist . Br J Pharmacol 2001 ; 134 : 701 – 704 . Google Scholar OpenURL Placeholder Text WorldCat 41 Salgado R , Vermeulen PB, Benoy I, Weytjens R, Huget P, Van Marck E et al. Platelet number and interleukin-6 correlate with VEGF but not with bFGF serum levels of advanced cancer patients . Br J Cancer 1999 ; 80 : 892 – 897 . Google Scholar OpenURL Placeholder Text WorldCat 42 Heer K , Kumar H, Speirs V, Greenman J, Drew PJ, Fox JN et al. Vascular endothelial growth factor in premenopausal women—indicator of the best time for breast cancer surgery? Br J Cancer 1998 ; 78 : 1203 – 1207 . Google Scholar OpenURL Placeholder Text WorldCat Copyright © 2004 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2004 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.
TI - Prognostic significance of serum vascular endothelial growth factor and endostatin in patients with hepatocellular carcinoma
JF - British Journal of Surgery
DO - 10.1002/bjs.4594
DA - 2004-09-17
UR - https://www.deepdyve.com/lp/oxford-university-press/prognostic-significance-of-serum-vascular-endothelial-growth-factor-khkv0xuOKb
SP - 1354
EP - 1360
VL - 91
IS - 10
DP - DeepDyve
ER -