Up-regulation of platelet-derived growth factor-A is responsible for the failure of re-initiated interferon alpha treatment in hepatocellular carcinoma

Abstract

Background: Postoperative interferon-α(IFN-α) treatment delays hepatocellular carcinoma(HCC) recurrence and prolongs patient survival, and may thus be an effective form of adjuvant therapy. However, clinical observations found that HCC recurs in some patients within 8 months of IFN-α treatment being discontinued. We investigated whether HCC regrowth appears after IFN-α is discontinued, whether re-initiated IFN-α is effective, and the underlying mechanisms of IFN-α treatment.

Methods: The human HCC nude mouse model LCI-D20 was used to study the effects of IFN-α treatment, discontinued IFN-α treatment, and re-initiated IFN-α treatment on tumor growth. Tumor weight, microvessel density(MVD), serum vascular endothelial growth factor (VEGF), and tumor cell apoptosis were analyzed. Angiogenesis-related factors were studied using cDNA microarray in different tumor samples and confirmed using reverse transcription-polymerase chain reaction(RT-PCR) and Western blotting assays. Finally, imatinib was added with re-initiated IFN-α treatment to improve efficacy.

Results: IFN-α (1.5 × 107 U/kg/day for 20 days) suppressed HCC growth by 60.3% and decreased MVD by 52.2% compared with the control. However, tumor regrowth occurred after IFN-α was discontinued, and re-initiated IFN-α treatment was not effective for inhibiting tumor growth or reducing MVD compared with a saline-treated group. cDNA microarray showed VEGF was down-regulated while platelet-derived growth factor-A (PDGF-A) was up-regulated when IFN-α treatment was re-initiated. These findings were further confirmed with RT-PCR and Western blotting assay. The combination of imatinib with re-initiated IFN-α reduced HCC weight by 30.7% and decreased MVD by 31.1% compared with IFN-α treatment only (P=0.003 and 0.015, respectively).

Conclusion: Tumor regrowth occurred after IFN-α treatment was discontinued. Re-initiated IFN-α treatment was not effective and was associated with up-regulation of PDGF-A, while the VEGF remained suppressed. The combination of a PDGF-receptor inhibitor with IFN-α improved the effect of the re-initiated treatment.

Figure 1

Experimental design and grouping. Groups A and B show whether IFN-α inhibits tumor growth; groups C and D, whether tumors regrow after IFN-α is discontinued; groups E and F, whether re-initiated IFN-α treatment remains effective; groups G and H, whether the efficacy of IFN-α was influenced by tumor size at the beginning of the treatment.

Figure 2

Effects of IFN-α on (A) tumor weight, (B) serum VEGF level, (C) angiogenesis, and (D) apoptotic index. Representative photo-micrographs of immunohistochemical and TUNEL analysis of MVD and apoptotic cells performed as described in Materials and methods, scale bar = 20 μm.

Figure 3

PDGF-A expression was increased in IFN- re-initiated treatment course. (A) Endothelial cell (red:CD31 staining/DyLight 488) and apoptosis (green: TUNEL/FITC) were examined. The first IFN-α treatment resulted in marked endothelial cell apoptosis (yellow), but the re-initiated IFN-α treatment did not (scale bar = 20 μm). (B) Expression of VEGF at the mRNA level in HCC. RT-PCR for GAPDH was used to compare input mRNA integrity. (C) Effect of IFN-α on the expression of VEGF165 and PDGF-A, determined with RT-PCR and normalized to GAPDH expression. An asterisk indicates statistical significance when compared with corresponding control groups (P < 0.05).

Figure 4

The comparison of AKT phosphorylation status among different IFN- treatment groups. (A) Representative image showing Western blot analysis on HCC tissues in different treatment groups using indicated antibodies. (B) Quantification of VEGF, PDGF-A, phosphor-Y951-VEGF receptor-2, and phosphor-AKT in IFN-α treatment tumors compared with controls are shown as mean band intensity with standard deviation, *P < 0.05.

Figure 5

Combination of immatinib with IFN- improved the effect of the re-initiated treatment. The combination treatment of imatinib (100 mg/kg/day) and re-initiated IFN-α (1.5 ×10 ⁷ U/kg/day) resulted in decreased phosphorylation of PDGFR, AKT, and ERK compared with IFN treatment alone or control group (A), and decreased tumor weight and MVD (B); Double staining of CD31 (for endothelial cells, green) and α-SMA (for pericyte, red) showed fewer pericytes were found in tumor vessels in combination treatment group compared with IFN-α or normal saline treatment groups (C), scale bar = 20 μm.