Evaluation of changes in the tumor microenvironment after sorafenib therapy by sequential histology and 18F-fluoromisonidazole hypoxia imaging in renal cell carcinoma

Abstract

The mechanistic dissociation of 'tumor starvation' versus 'vascular normalization' following anti-angiogenic therapy is a subject of intense controversy in the field of experimental research. In addition, accurately evaluating changes of the tumor microenvironment after anti-angiogenic therapy is important for optimizing treatment strategy. Sorafenib has considerable anti-angiogenic effects that lead to tumor starvation and induce tumor hypoxia in the highly vascularized renal cell carcinoma (RCC) xenografts. 18F-fluoromisonidazole (18F‑FMISO) is a proven hypoxia imaging probe. Thus, to clarify early changes in the tumor microenvironment following anti-angiogenic therapy and whether 18F-FMISO imaging can detect those changes, we evaluated early changes in the tumor microenvironment after sorafenib treatment in an RCC xenograft by sequential histological analysis and 18F-FMISO autoradiography (ARG). A human RCC xenograft (A498) was established in nude mice, for histological studies and ARG, and further assigned to the control and sorafenib-treated groups (80 mg/kg, per os). Mice were sacrificed on Days 1, 2, 3 and 7 in the histological study, and on Days 3 and 7 in ARG after sorafenib treatment. Tumor volume was measured every day. 18F-FMISO and pimonidazole were injected intravenously 4 and 2 h before sacrifice, respectively. Tumor sections were stained with hematoxylin and eosin and immunohistochemically with pimonidazole and CD31. Intratumoral 18F-FMISO distribution was quantified in ARG. Tumor volume did not significantly change on Day 7 after sorafenib treatment. In the histological study, hypoxic fraction significantly increased on Day 2, mean vessel density significantly decreased on Day 1 and necrosis area significantly increased on Day 2 after sorafenib treatment. Intratumoral 18F-FMISO distribution significantly increased on Days 3 (10.2-fold, p<0.01) and 7 (4.1-fold, p<0.01) after sorafenib treatment. The sequential histological evaluation of the tumor microenvironment clarified tumor starvation in A498 xenografts treated with sorafenib. 18F-FMISO hypoxia imaging confirmed the tumor starvation. 18F-FMISO PET may contribute to determine an optimum treatment protocol after anti-angiogenic therapy.

Figure 1.

Experimental protocols of (A) ex vivo histological study and (B) ARG.

Figure 2.

Tumor growth curve. Dotted arrow, treatment period.

Figure 3.

(A) Microscopy images and (B) quantitative analysis of hypoxia on Days 1, 2, 3 and 7 after treatment with vehicle or sorafenib. (A) Representative images of pimonidazole staining (brown staining in pimonidazole-positive cells). (B) Mean hypoxic fraction ± SD. ^*p<0.01 vs. control group at the same time-point.

Figure 4.

(A) Microscopy images and (B) quantitative analysis of intratumoral microvessels on Days 1, 2, 3 and 7 after treatment with vehicle or sorafenib. (A) Representative images of CD31 staining (brown staining in CD31-positive cells). (B) Mean vessel densities (MVDs) ± SD. ^*p<0.01 vs. control group at the same time-point.

Figure 5.

Quantitative analysis of necrosis on Days 1, 2, 3 and 7 after treatment with vehicle or sorafenib. Necrosis areas (%) ± SD. ^*p<0.01 vs. control group at the same time-point.

Figure 6.

(A) Representative images and (B–D) quantitative analysis of (A and B) intratumoral ¹⁸F-FMISO and immunohistochemical stainings of (A and C) pimonidazole and (A and D) CD31 on Days 3 and 7 after treatment with vehicle or sorafenib. (A) Comparison among ¹⁸F-FMISO ARG, pimonidazole IHC and CD31 IHC images. Representative ¹⁸F-FMISO ARG images, pimonidazole IHC (brown staining in pimonidazole-positive cells) and CD31 IHC (black arrowhead around microvessels, staining in CD31-positive cells). Dotted line shows muscle. (B) Mean ¹⁸F-FMISO accumulation levels ± SD. (C) Mean hypoxic fraction ± SD. (D) Mean vessel densities (MVDs) ± SD. ^*p<0.01 vs. control group at the same time-point (B–D).