Immune modulation following α and β- radionuclide therapy targeting fibroblast activation protein-α in a preclinical tumor model

Abstract

α- and β^--emitting radionuclides targeting human fibroblast activation protein-α (hFAP) are under investigation for cancer therapy. In prior work, analysis of the tumor microenvironment 24 h after therapy completion indicated therapy-induced immune activation. Here, we analyzed systemic immune responses at varying timepoints during treatment to further elucidate the immune-stimulating effects of the therapy. Moreover, we analyzed end-stage tumors to gain insight in potential mechanisms of therapy resistance. Single domain antibody 4AH29 that binds hFAP was labeled with ¹³¹I or ²²⁵Ac, generating [¹³¹I]I-GMIB-4AH29 and [²²⁵Ac]Ac-DOTA-4AH29, respectively. These were used to treat C57BL/6 mice bearing subcutaneous TC-1-hFAP tumors. Blood analysis was conducted using flow cytometry, while tumor characterization was performed using flow cytometry and RNA sequencing. Given the distinct properties and doses of both radiopharmaceuticals, no head-to-head comparison was performed. Both treatments activated inflammatory responses in the tumor. Increased PD-1 expression on CD8⁺ T-cells was observed following both treatments in the tumor and periphery. In the tumor, [¹³¹I]I-GMIB-4AH29 therapy uniquely induced the expression of genes involved in tumor cell replication, TNF-α, IL-6/STAT3, IL-2/STAT5 and complement pathways, while in the blood [¹³¹I]I-GMIB-4AH29 therapy upregulated SIRPα on monocytes and TIGIT on NK cells, and downregulated CD86 expression on monocytes. Longitudinal blood immune cell analysis showed changes in composition and phenotype early in therapy, e.g. in effector and regulatory T-cells. Overall, this study corroborates the immune sensitizing capacity of α- and β^--emitting radionuclides, triggering a variety of inflammatory effector responses.

Keywords: PD-1; PD-L1; cGAS-STING; fibroblast activation protein-α; immunostimulation; immunotherapy; single domain antibody; targeted radionuclide therapy.

Figure 1.

RNA sequencing analysis of TC-1-hFAP tumors treated with [¹³¹I]I-GMIB-4AH29 or [²²⁵Ac]Ac-DOTA-4AH29, isolated at a size of 1,500 mm³. (a) PCA plot representing the first and second principal component of the 12 samples included in this analysis. The plot depicts the transcriptional profile of all three groups, clustered by color. (b-c) volcano plot displaying DEGs when adjusted p-value < 0.05 and |log2 Fold change| > 1. Red dots indicate genes upregulated in the first group compared to second. (b) [¹³¹I]I-GMIB-4AH29 compared to vehicle and (c) [²²⁵Ac]Ac-DOTA-4AH29 compared to vehicle (N = 1, n = 4). Genes were marked as DEGs when adjusted p-value < 0.05 and shrunken log2 Fold change threshold was set above or below 1 or − 1.

Figure 2.

GSEA of TC-1-hFAP lung tumors treated with [¹³¹I]I-GMIB-4AH29 or [²²⁵Ac]Ac-DOTA-4AH29. Figure summarizing the NES of MSigDB Hallmark gene sets comparing [¹³¹I]I-GMIB-4AH29 with vehicle and [²²⁵Ac]Ac-DOTA-4AH29 with vehicle (N = 1, n = 4).

Figure 3.

Gene expression profiling of TC-1-hFAP tumors treated with [¹³¹I]I-GMIB-4AH29 or [²²⁵Ac]Ac-DOTA-4AH29. (a-b) graphs showing individual normalized gene counts of CXCL10 and CCL5 (c) heatmap of IFN pathway-related genes with enrichment score calculated via GSVA in the bottom row. (d-k) graphs showing individual normalized gene counts of Ifit1, Ifit3, Ifi44, Irf7, H2-Aa, H2-Ab1, CD8A and CD274. N = 1, n = 4.

Figure 4.

Immune cell composition in TC-1-hFAP tumors after treatment with [¹³¹I]I-GMIB-4AH29 or [²²⁵Ac]Ac-DOTA-4AH29. (a) Percentage of CD45.2⁺ immune cells within living cells. (b-h) percentage of B cells (CD45.2⁺ CD3⁻ CD19⁺), T-cells (CD45.2⁺ CD3⁺), NK cells (CD45.2⁺ NK1.1⁺ CD3⁻), NKT-cells (CD45.2⁺ NK1.1⁺ CD3⁺), CD4⁺ T-cells (CD45.2⁺ CD3⁺ CD4⁺), Tregs (CD45.2⁺ CD3⁺ CD4⁺ CD25⁺ CD127⁻) and CD8⁺ T-cells (CD45.2⁺ CD3⁺ CD8⁺) within CD45.2⁺ immune cells. (i) Percentage of E7-specific cells within CD8⁺ T-cells. (j-n) percentage of neutrophils (CD45.2⁺ CD11b⁺ Ly6G⁺), monocytes (CD45.2⁺ CD11b⁺ MHCII⁻ Ly6c⁺), MHC-II^high macrophages (CD45⁺ CD11b⁺ MHC-II⁺ F4/80⁺), cDC1 (CD45.2⁺ MHC-II^high CD11c⁺ CD11b⁺) and cDC2 (CD45.2⁺ MHC-II^high CD11c⁺ CD11b⁻) within CD45.2⁺ immune cells. N = 1, n = 4–5.

Figure 5.

Impact of [¹³¹I]I-GMIB-4AH29 and [²²⁵Ac]Ac-DOTA-4AH29 on phenotypic markers of co-inhibition in the TME. (a) Mean fluorescence intensity of PD-1 on CD8⁺ T-cells. (b-c) Mean fluorescence intensity of LAG-3 on CD4⁺ T-cells and Tregs. (d) Mean fluorescence intensity of SIRPα on cDC2. (e-h) Mean fluorescence intensity of PD-L1, LAG-3, SIRPα and CD86 on MHC-II^high macrophages. N = 1, n = 4–5.

Figure 6.

Impact of [¹³¹I]I-GMIB-4AH29 and [²²⁵Ac]Ac-DOTA-4AH29 on myeloid cells in the blood of TC-1-hFAP tumor-bearing mice. (a-c) graphs showing changes in (a) inflammatory DCs (CD45.2⁺ CD11c⁺ Ly6c⁺) within CD45.2⁺ immune cells and their expression of (b) PD-L1 and (c) CD86. (d-f) graphs showing changes in (d) monocytes (CD45⁺CD11b⁺ MHCII⁻ Ly6c⁺) within CD45.2⁺ immune cells and their expression of (e) SIRPα and (e) CD86 (N = 1, n = 4–5). Statistical significance is shown with a red and blue asterisk for [²²⁵Ac]Ac-DOTA-4AH29 vs. vehicle and [¹³¹I]I-GMIB-4AH29 vs. vehicle, respectively.

Figure 7.

Impact of [¹³¹I]I-GMIB-4AH29 and [²²⁵Ac]Ac-DOTA-4AH29 on lymphoid cells in the blood of TC-1-hFAP tumor-bearing mice. (a-b) graphs showing changes in (a) CD4⁺ T-cells (CD45.2⁺ CD3⁺ CD4⁺) within CD45.2⁺ immune cells and their expression of (b) PD-1. (c-e) graphs showing changes in (c) CD8⁺ T-cells (CD45.2⁺ CD3⁺ CD8⁺) within CD45.2⁺ immune cells and their expression of (d) PD-1 and (e) LAG-3. (f-g) graphs showing changes in (f) Tregs (CD45.2⁺ CD3⁺ CD25⁺ CD127⁻) within CD45.2⁺ immune cells and their expression of (g) PD-1. (h-i) graphs showing changes in (h) NK cells (CD45.2⁺ NK1.1⁺) within CD45.2⁺ immune cells and their expression of (i) TIGIT (N = 1, n = 4–5). Statistical significance is shown with a red and blue asterisk for [²²⁵Ac]Ac-DOTA-4AH29 vs. vehicle and [¹³¹I]I-GMIB-4AH29 vs. vehicle, respectively.