Effect of Incomplete Cryoablation and Matrix Metalloproteinase Inhibition on Intratumoral CD8+ T-Cell Infiltration in Murine Hepatocellular Carcinoma

Abstract

Background Image-guided tumor ablation is the first-line therapy for early-stage hepatocellular carcinoma (HCC), with ongoing investigations into its combination with immunotherapies. Matrix metalloproteinase (MMP) inhibition demonstrates immunomodulatory potential and reduces HCC tumor growth when combined with ablative treatment. Purpose To evaluate the effect of incomplete cryoablation with or without MMP inhibition on the local immune response in residual tumors in a murine HCC model. Materials and Methods Sixty 8- to 10-week-old female BALB/c mice underwent HCC induction with use of orthotopic implantation of syngeneic Tib-75 cells. After 7 days, mice with a single lesion were randomized into treatment groups: (a) no treatment, (b) MMP inhibitor, (c) incomplete cryoablation, and (d) incomplete cryoablation and MMP inhibitor. Macrophage and T-cell subsets were assessed in tissue samples with use of immunohistochemistry and immunofluorescence (cell averages calculated using five 1-μm² fields of view [FOVs]). C-X-C motif chemokine receptor type 3 (CXCR3)- and interferon γ (IFNγ)-positive T cells were assessed using flow cytometry. Groups were compared using unpaired Student t tests, one-way analysis of variance with Tukey correction, and the Kruskal-Wallis test with Dunn correction. Results Mice treated with incomplete cryoablation (n = 6) showed greater infiltration of CD206⁺ tumor-associated macrophages (mean, 1.52 cells per FOV vs 0.64 cells per FOV; P = .03) and MMP9-expressing cells (mean, 0.89 cells per FOV vs 0.11 cells per FOV; P = .03) compared with untreated controls (n = 6). Incomplete cryoablation with MMP inhibition (n = 6) versus without (n = 6) led to greater CD8⁺ T-cell (mean, 15.8% vs 8.29%; P = .04), CXCR3⁺CD8⁺ T-cell (mean, 11.64% vs 8.47%; P = .004), and IFNγ⁺CD8⁺ T-cell infiltration (mean, 11.58% vs 5.18%; P = .02). Conclusion In a mouse model of HCC, incomplete cryoablation and systemic MMP inhibition showed increased cytotoxic CD8⁺ T-cell infiltration into the residual tumor compared with either treatment alone. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Gemmete in this issue.

Graphical abstract

Figure 1:

Schematic overview of the study design. Approximately 2 million Tib-75 cells suspended in a biologically active matrix material were injected subcapsularily into the liver of 8- to 10-week-old female BALB/c mice to induce hepatocellular carcinoma (HCC). Mice were randomly divided into four groups 7 days after tumor inoculation. Incomplete cryoablation (iCryo or Cryo) was performed in groups 3 (n = 6) and 4 (n = 6) at day 0 (red diamond). Groups 2 and 4 received intraperitoneal injection of phosphate-buffered saline (PBS) (n = 6) or a broad-spectrum matrix metalloproteinase (MMP) inhibitor diluted in phosphate-buffered saline (n = 6) every 2 days for a total of 7 doses (blue shading). Animals were killed on day 17, and spleen and liver tissues were collected. Flow cytometry was performed on tumor and tumor-adjacent tissue from four different groups: control mice (no treatment), mice treated with the MMP inhibitor (n = 6), mice treated with incomplete cryoablation (n = 6), and mice treated with incomplete cryoablation and the MMP inhibitor (n = 6). Histopathologic examination (immunohistochemistry, immunofluorescence, and hematoxylin and eosin [H&E] staining) was performed on spleen and whole-liver tissue from six different groups: the aforementioned four groups, in addition to phosphate-buffered saline–treated mice (n = 6) and mice treated with incomplete cryoablation and phosphate-buffered saline (n = 6).

Figure 2:

Positive and negative control stainings (20× magnification). Staining specificity of the surface markers in BALB/c mice spleen tissue (positive control) and BALB/c mice liver tissue (negative control) was verified before staining experimental tissue. Cytotoxic T lymphocyte (CD8⁺), macrophage (CD68⁺), M2-polarized macrophage (CD206⁺), and matrix metalloproteinase 9 (MMP9⁺) (A) immunohistochemistry staining and (B) immunofluorescence staining.

Figure 3:

(A) Representative hematoxylin and eosin (H&E) staining and immunohistochemistry staining of hepatocellular carcinomas at 7 days in the control group (CTRL) (top row) and incomplete cryoablation group (iCryo) (bottom row). CD68⁺, CD206⁺, and matrix metalloproteinase 9 (MMP9⁺) staining was performed to analyze macrophage infiltration, M2-positive macrophage infiltration, and MMP9 expression, respectively. The necrotic ablation zone (NZ) is inside the dotted white line. Tumor (T) and nontumorous (L) tissue are separated with a continuous white line. Magnification, 10×. (B) Box and whisker plots compare the average number of CD68⁺, CD206⁺, and MMP9⁺ intratumoral cells at 7 days in control mice that did not receive ablation (n = 6) and mice treated with incomplete cryoablation (n = 6). Whiskers indicate maximum and minimum values. Boxes extend from the 25th to 75th percentiles. The line in the middle of the box is plotted at the median. * = P < .05 and ** = P < .01, according to an unpaired t test. FOV = field of view.

Figure 4:

Matrix metalloproteinase (MMP) inhibition (MMPI) decreases intratumoral MMP9 expression and tumor-associated macrophage (TAM) infiltration. (A) Representative low-magnification (10×) hematoxylin and eosin (H&E) images show Tib-75–induced tumors in BALB/c mice after MMP inhibition with the broad-spectrum inhibitor batimastat (bottom row) relative to control mice receiving injection with the vehicle phosphate-buffered saline (PBS) (top row). L = healthy liver, T = tumor. Representative images of immunofluorescence costaining of intratumoral MMP9 expression and infiltration of TAMs (CD206⁺). Magnification, 20×. DAPI = 4',6-diamidino-2-phenylindole. (B, C) Box and whisker plots show quantification of intratumoral (B) MMP9-expressing cells and (C) CD206⁺ cells in Tib-75 induced–hepatocellular carcinomas. A decrease in intratumoral MMP9⁺ cells and CD206⁺ cell infiltration is seen in mice treated with the MMP inhibitor (n = 6) relative to phosphate-buffered saline–treated control mice (n = 6 per group) and in mice treated with incomplete cryoablation (iCryo) and the MMP inhibitor (n = 6) versus mice treated with incomplete cryoablation and phosphate-buffered saline (n = 6). Whiskers indicate maximum and minimum values. Boxes extend from the 25th to 75th percentiles. The line in the middle of each box is plotted at the median. * = P < .05 and *** = P < .001, according to a Kruskal-Wallis test with Dunn correction for multiple comparisons. FOV = field of view.

Figure 5:

Combination treatment with incomplete cryoablation (iCryo) and matrix metalloproteinase (MMP) inhibition (MMPI) modulate lymphocyte populations in the tumor microenvironment. (A) Representative low-magnification (10×) hematoxylin and eosin (H&E) images show the necrotic ablation zone (NZ) in Tib-75–induced tumors in BALB/c mice after treatment with incomplete cryoablation and the vehicle phosphate-buffered saline (PBS) (top row) and after treatment with incomplete cryoablation and the MMP inhibitor batimastat (bottom row). Representative images of immunofluorescence costaining of tumor-infiltrating cytotoxic T cells (CD8⁺) (magnification, 20×). DAPI = 4',6-diamidino-2-phenylindole, L = healthy liver, T = tumor. (B) Representative flow cytometry plots (n = 1) for CD3 and CD8 coexpression in CD45.2 (hematopoietic) gate show higher infiltration of CD8⁺ cells in mice treated with incomplete cryoablation and the MMP inhibitor relative to no treatment (CTRL), the MMP inhibitor only, and incomplete cryoablation only. (C) Box and whisker plot shows the average number of CD8⁺ cells in five random fields of view (FOVs) from the tumor center in immunofluorescence images under 20× magnification (n = 6 per group). (D) Box and whisker plot shows the percentage of CD3⁺/CD4⁺ T helper lymphocytes and CD3⁺/CD8⁺ cytotoxic lymphocytes in CD45.2 gate as quantified at flow cytometry (n = 6 per group). Whiskers indicate maximum and minimum values. Boxes extend from the 25th to 75th percentiles. The line in the middle of each box is plotted at the median. * = P < .05 and ** = P < .01, according to a Kruskal-Wallis test with Dunn correction for multiple comparisons.

Figure 6:

Matrix metalloproteinase (MMP) inhibition (MMPI) promotes interferon γ (IFNγ)– and chemokine receptor 3 (CXCR3)–expressing CD8⁺ T cells in tumors treated with incomplete cryoablation (iCryo). (A) Representative flow cytometry plot (n = 1) of CD8⁺IFNγ⁺ T cells in CD45.2/CD3 gate in tumors treated with incomplete cryoablation and the MMP inhibitor compared with mice treated with incomplete cryoablation only. (B) Box and whisker plot compares the percentage of CD8⁺IFNγ⁺ T cells (gate on CD45.2⁺/CD3⁺ cells) in four treatment groups: control (CTRL) mice (no treatment), mice treated with the MMP inhibitor (n = 6), mice treated with incomplete cryoablation (n = 6), and mice treated with incomplete cryoablation and the MMP inhibitor (n = 6). * = P < .05 and *** = P < .001, according to a Kruskal-Wallis test with Dunn correction for multiple comparisons. (C) Representative flow cytometry plot (n = 1) of CD8⁺CXCR3⁺ T cells in CD45.2/CD3 gate in tumors treated with incomplete cryoablation and the MMP inhibitor compared with mice treated with incomplete cryoablation only. (D) Box and whisker plot compares the percentage of CD8⁺CXCR3⁺ T cells (gate on CD45.2⁺/CD3⁺ cells) in the aforementioned four treatment groups. * = P < .01 and ** = P < .001, according to a one-way analysis of variance with Tukey correction for multiple comparisons.