Development and evaluation of nanobody tracers for noninvasive nuclear imaging of the immune-checkpoint TIGIT

Abstract

Introduction: T cell Ig and ITIM domain receptor (TIGIT) is a next-generation immune checkpoint predominantly expressed on activated T cells and NK cells, exhibiting an unfavorable prognostic association with various malignancies. Despite the emergence of multiple TIGIT-blocking agents entering clinical trials, only a fraction of patients responded positively to anti-TIGIT therapy. Consequently, an urgent demand arises for noninvasive techniques to quantify and monitor TIGIT expression, facilitating patient stratification and enhancing therapeutic outcomes. Small antigen binding moieties such as nanobodies, are promising candidates for such tracer development.

Methods: We generated a panel of anti-human or anti-mouse TIGIT nanobodies from immunized llamas. In addition, we designed a single-chain variable fragment derived from the clinically tested monoclonal antibody Vibostolimab targeting TIGIT, and assessed its performance alongside the nanobodies. In vitro characterization studies were performed, including binding ability and affinity to cell expressed or recombinant TIGIT. After Technetium-99m labeling, the nanobodies and the single-chain variable fragment were evaluated in vivo for their ability to detect TIGIT expression using SPECT/CT imaging, followed by ex vivo biodistribution analysis.

Results: Nine nanobodies were selected for binding to recombinant and cell expressed TIGIT with low sub-nanomolar affinities and are thermostable. A six-fold higher uptake in TIGIT-overexpressing tumor was demonstrated one hour post- injection with Technetium-99m labeled nanobodies compared to an irrelevant control nanobody. Though the single-chain variable fragment exhibited superior binding to TIGIT-expressing peripheral blood mononuclear cells in vitro, its in vivo behavior yielded lower tumor-to-background ratios at one hour post- injection, indicating that nanobodies are better suited for in vivo imaging than the single-chain variable fragment. Despite the good affinity, high specificity and on-target uptake in mice in this setting, imaging of TIGIT expression on tumor- infiltrating lymphocytes within MC38 tumors remained elusive. This is likely due to the low expression levels of TIGIT in this model.

Discussion: The excellent affinity, high specificity and rapid on-target uptake in mice bearing TIGIT- overexpressing tumors showed the promising diagnostic potential of nanobodies to noninvasively image high TIGIT expression within the tumor. These findings hold promise for clinical translation to aid patient selection and improve therapy response.

Keywords: TIGIT; immune checkpoint (ICP); nanobodies; noninvasive diagnosis; nuclear imaging; tracer development.

Figure 1

In vitro characterization of the purified Nbs. (A) SDS-PAGE of 5 µg purified Nbs or scFv Vibo stained with InstantBlue Coomassie protein stain. (B) Binding of 100 nM purified His₆-tagged Nbs on mouse (black) or human (blue) TIGIT overexpressing HEK293T cells, detected with PE-labeled anti-His antibody and flow cytometry. Results are shown as delta mean fluorescent intensity (ΔMFI) by subtracting the MFI of Nb binding on WT HEK293T cells from Nb binding on m/h TIGIT expressing HEK293T cells. (C) Affinity of the purified Nbs on HEK293T cells transduced to express human (left panel) or mouse TIGIT (right panel), detected by flow cytometry using anti-His antibody and a dilution range of Nbs. (D) Percentage unfolded Nbs or scFv Vibo at different temperatures, determined by the Thermofluor assay.

Figure 2

Anti-TIGIT Nbs bind to TIGIT expressed on splenocytes or PBMCs, results from the lead Nbs are shown. (A) anti-mTIGIT Nb 16988 incubated with 900 nM to CD3/28 dynabeads-activated mouse splenocytes, compared to the irrelevant ctrl Nb R3B23 and the commercially available mAb. (B) anti-hTIGIT Nb 16925 or Vibo ScFv incubated with 900 nM to CD3/28 dynabeads-activated human PBMCs, compared to the irrelevant ctrl Nb R3B23 and the commercially available mAb. Binding of Nb or scFv was detected with PE-labeled anti-His antibody on flow cytometry.

Figure 3

In vivo SPECT-CT imaging and ex vivo biodistribution of the anti-mTIGIT Nbs in immunodeficient mice bearing TC-1 tumors (n=3). (A) 3D-rendered SPECT-CT images (top) and a transversal slice at the level of the tumors (bottom) of a representative mouse bearing a mTIGIT-transduced (+) and an untransduced (-) TC-1 tumor and injected with ^99mTc-labeled anti-mTIGIT Nb 16988 (left) or control Nb R3B23 (right). (B) ex vivo biodistribution results of the control Nb and the three selected anti-mTIGIT Nbs showing percentage injected activity per gram (%IA/g) tissue in WT TC-1 and mTIGIT⁺ TC-1 tumors. Two-way ANOVA was used to calculate statistical significance. Statistical significance was set at p<0.05 (ns, not significant, ***=p<0.001, ****=p<0.0001).

Figure 4

In vivo SPECT-CT imaging and ex vivo biodistribution of the anti-hTIGIT Nbs in immunodeficient mice bearing TC-1 tumors (n=3). (A) 3D-rendered SPECT-CT images (top) and a transversal slice at the level of the tumors (bottom) of a representative mouse bearing a hTIGIT-transduced (+) and an untransduced (-) TC-1 tumor and injected with ^99mTc-labeled anti-hTIGIT Nb 16925 (left) or control Nb R3B23 (right). (B) ex vivo biodistribution results of the control Nb and the selected anti-hTIGIT Nbs showing percentage injected activity per gram (%IA/g) tissue in WT TC-1 and hTIGIT⁺ TC-1 tumors. Two-way ANOVA was used to calculate statistical significance. Statistical significance was set at p<0.05 (ns, not significant, ****=p<0.0001).

Figure 5

Anti-mTIGIT Nbs show high signal-to-noise ratios compared to the control Nb R3B23. Ratios of uptake of ^99mTc-labeled anti-mTIGIT Nbs in TC-1 mTIGIT⁺ tumor to (A) TC-1 WT tumor, (B) muscle, (C) blood or to (D) liver (n=3), ratios are calculated as following: uptake in TC-1 mTIGIT⁺ divided by uptake in TC-1 WT tumor, muscle, liver, or blood. One-way ANOVA was used to evaluate statistical significance. Statistical significance was set at p<0.05 (**=p<0.01, ***=p<0.001, ****=p<0.0001).

Figure 6

Anti-hTIGIT Nbs show high signal-to-noise ratios compared to the control Nb R3B23 and the scFv Vibo. Ratios of uptake of ^99mTc-labeled anti-hTIGIT Nbs in TC-1 hTIGIT⁺ tumor to (A) TC-1 WT tumor, (B) muscle, (C) blood or to (D) liver (n=3), ratios are calculated as following: uptake in TC-1 hTIGIT⁺ divided by uptake in TC-1 WT tumor, muscle, liver, or blood. One-way ANOVA was used to evaluate statistical significance. Statistical significance was set at p<0.05 and only shown for significant data (*=p<0.05, **=p<0.01, ****=p<0.0001).

Figure 7

Ex vivo biodistribution of ^99mTc-labeled anti-mTIGIT Nb 16988 in C57BL/6 mice bearing a subcutaneous MC38 tumor. (A) Ratio %IA/g of tumor, tumor draining lymph node (TdLN), cervical LN, spleen to blood or to muscle. (B). mTIGIT expression (ΔMFI) evaluated on single cells suspensions of the lymph node, spleen, and MC38 tumor on CD45⁺, CD3⁺, CD4⁺, CD8⁺ and CD25⁺ CD127^- Tregs using flow cytometry by subtracting the MFI of the fluorescence minus one (FMO) from the signal. Unparied t-test (A) or one-way ANOVA (B) was used to evaluate statistical significance. Statistical significance was set at p<0.05 (ns, not significant, *=p<0.05, **=p<0.01, ***=p<0.001, ****=p<0.0001).

Figure 8

In vivo SPECT-CT imaging and ex vivo biodistribution of ^99mTc-radiolabeled anti-mTIGIT Nb 16988 and irrelevant control Nb R3B23 in wild type C57BL/6 mice and hTIGIT KI mice. Biodistribution study with ^99mTc-Nb16988 showing uptake (%IA/g) in the thymus, spleen and lymph nodes with SPECT-CT imaging analyzed with Amide showing a transversal slice of the thymus uptake in hTIGIT KI mouse compared to WT mouse. Statistical analyses were performed using one-way ANOVA. Statistical significance was set at p<0.05 (**=p<0.01, ***=p<0.001, ****=p<0.0001).

Figure 9

In vivo SPECT-CT imaging and ex vivo biodistribution of ^99mTc-radiolabeled anti-hTIGIT Nb 16925 and irrelevant control Nb R3B23 in wild type C57BL/6 mice and hTIGIT KI mice. Biodistribution study with ^99mTc-Nb16925 showing uptake (%IA/g) in the thymus, spleen and lymph nodes with SPECT-CT imaging analyzed with Amide showing a transversal slice of the thymus uptake in hTIGIT KI mouse compared to WT mouse. Statistical analyses were performed using one-way ANOVA. Statistical significance was set at p<0.05 (ns, not significant, **=p<0.01). .