Impact of biomarkers on disease survival and progression in patients treated with octreotide for advanced hepatocellular carcinoma

Abstract

Background: Current determination of prognosis for advanced hepatocellular carcinoma (HCC) is mainly based on clinical assessment. We aimed to determine the impact of biomarkers as predictive factors for HCC progression and survival during octreotide-based treatments.

Patients and methods: We included patients who had been prospectively randomised to receive either octreotide (30 mg) alone monthly (n = 39) or in combination with rofecoxib (up to 50 mg bid daily, n = 32) for a minimum of 6 months, or until death occurred.

Results: Overall median survival (154 days) and median time to progression (94 days) were not different for both treatments and the biomarkers investigated (VEGF-A, IGF-1, PGE-2, ET-A) were similarly distributed amongst treatment groups. Combined univariate group analysis revealed that survival was decreased for an uptake ratio of > 2 on initial octreoscan (P = 0.05); baseline serum VEGF-A and IGF-1 were further significantly associated with survival. On multivariate analysis, uncorrected serum VEGF-A appeared to be the most significant predictor for tumor progression and survival.

Conclusions: Biomarkers, in addition to established tumor markers, are independent predictors of tumor progression and survival in patients with advanced HCC treated with octreotide. Furthermore, the involvement of VEGF-A implies the inhibition of angiogenesis as a potential mechanism of action for this drug.

Fig. 1

Correlation between serum VEGF-A and platelets at baseline (solid circles) and month 3, (open rectangles), single values and 95%-CI regression lines: Baseline: r = 0.671, 95%-CI: 0.495–0.794, p < 0.0001, Month 3: r = 0.589, 95%-CI: 0.318–0.771, p = 0.0001

Fig. 2

Impact of baseline biomarkers on tumor progression: Kaplan-Meier Curves (logrank test. [95%-CI]) according to optimal cut-off values derived from ROC analysis for, a serum VEGF-A: chi-square 5.75 (df 1), p = 0.017, hazard ratio 0.49 [0.29–0.88], b serum IGF-1: chi-square 9.51 (df 1), p = 0.002, hazard ratio 2.5 [1.6–8.4], c serum PGE-2: chi-square 4.42 (df 1), p = 0.036, hazard ratio 0.56 [0.27–0.96], d serum ET-A: chi-square 2.26 (df 1), p = 0.133, hazard ratio 1.5 [0.86–3.0]

Fig. 3

Impact of baseline biomarkers on overall survival: Kaplan-Meier Curves (logrank test [95%-CI]) according to optimal cut-off values derived from ROC analysis for, a serum VEGF-A: chi-square 3.36 (df 1), p = 0.067, hazard ratio 0.58 [0.28–1.04], b serum IGF-1: chi-square 18.6 (df 1), p < 0.0001, hazard ratio 3.2 [2.4–10.5], c serum PGE-2: chi-square 6.0 (df 1), p = 0.014, hazard ratio 0.51 [0.26–0.86], d serum ET-A: chi-square 0.03 (df 1), p = 0.868, hazard ratio 1.05 [0.59–1.9]

Fig. 4

Impact of baseline VEGF-Platelet ratio; Kaplan-Meier Curves (logrank test [95%-CI]) according to optimal cut-off value derived from ROC analysis, a on tumor progression: chi-square 3.34 (df 1), p = 0.068, hazard ratio 0.53 [0.23–1.05], b on overall survival: chi-square 4.32 (df 1), p = 0.038, hazard ratio 0.49 [0.29–0.96], c on overall survival (paired comparison BL/Mo 3, n = 29): chi-square 4.02 (df 1), p = 0.045, hazard ratio 2.4 [1.03–9.0]

Fig. 5

Impact of Octreoscan Uptake on overall survival: Kaplan-Meier Curves (logrank test [95%-CI]) according to optimal cut-off value derived from ROC analysis, chi-square 3.86 (df1), p = 0.049, hazard ratio 0.47 [0.23–0.99]