EMIM 2018 ControlCenter

Online Program Overview Session: PS-15

To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).

Immuno-Oncology

Session chair: Manfred Kneiling - Tübingen, Germany; Erik Aarntzen - Nijmegen, The Netherlands
 
Shortcut: PS-15
Date: Thursday, 22 March, 2018, 1:30 PM
Room: Lecture Room 03 | level -1
Session type: Parallel Session

Abstract

Click on an contribution preview the abstract content.

1:30 PM PS-15-1

Introductory Talk by Jolanda de Vries - Nijmegen, The Netherlands

This talk provides an overview of state-of-the-art research and refers to the following presentations selected from abstract submissions.

1:50 PM PS-15-2

Positron emission tomography imaging of OX40+ activated T cells to monitor and predict response to an in situ tumor vaccine. (#69)

I. S. Alam1, 5, A. T. Mayer1, 5, I. Sagiv-Barfi2, K. Wang3, O. Vermesh1, 5, D. K. Czerwinski2, E. M. Johnson1, 4, 5, M. L. James1, 4, 5, R. Levy2, S. S. Gambhir1, 5

1 Stanford University, Radiology, Stanford, California, United States of America
2 Stanford University, Oncology, Stanford, California, United States of America
3 Harbin Medical University, Imaging Center, Harbin, China
4 Stanford, of Neurology and Neurological Sciences, Stanford, California, United States of America
5 Stanford University, Molecular Imaging Program, Stanford, California, United States of America

Introduction

Monitoring and predicting outcomes to cancer immunotherapy is challenging due to the highly varying and complex spatiotemporal dynamics of immune response in the tumor microenvironment. T cell activation is considered critical to treatment success across many classes of cancer immunotherapy and is likely more prognostic of treatment outcome than the presence of tumor infiltrating T cells alone. We report a novel PET tracer (64Cu-DOTA-mAbOX40) that enables non-invasive and longitudinal imaging of OX40, a cell surface marker expressed on antigen-specific activated T cells [1].

Methods

Anti-OX40 monoclonal antibody was DOTA-conjugated, radiolabeled with 64Cu and evaluated in vitro to assess its specificity for activated vs. resting murine T cells (supplementary Fig. A). Female BALB/c mice bearing dual A20 lymphoma tumors on the shoulders were administered low dose CpG oligonucleotide (50ug in PBS, n=7-10) or PBS alone directly in the left tumor only (vehicle control, n=7-10). PET/CT imaging of mice with 64Cu-DOTA-mAbOX40 (3.0-4.1MBq, i.v.) was performed at an early (day 2) and late (day 9) time point post-treatment initiation ( supplementary Fig. B). Flow cytometry analyses of tumor draining lymph nodes(TDLN), tumors, spleen from vehicle and CpG-treated mice was also performed at day 2 and 9 to determine OX40 expression alongside other T cell markers (CD3/4/8,44 and 25).

Results/Discussion

Early time point imaging post CpG administration revealed increased radiotracer uptake in the CpG-treated tumor [CpG 10.3±0.7; Veh 7.5±0.3 %ID/g; p<0.05] and associated TDLN [CpG 12.92 ± 1.15; Veh 10.67 ± 0.94%ID/g; p<0.01] (Fig. 1A- B). An increase in the frequency of OX40+CD3+ expression in these tissues vs untreated sites (p<0.05) and vehicle cohorts was further confirmed by FACS (Fig. 1C-D) suggesting in situ CpG vaccination triggered a local induction of cellular immune response. ViSNE, a visualization technique for high-dimensional cytometry data, showed that OX40+ cells were highly restricted to clusters associated CD4 helper T cells. Changes in OX40+ CD3 frequency preceded the increase in the overall proportion of CD3+ T cells within CpG-treated tumors. Importantly, early OX40-PET signal (mean %ID/g) in the local tumor environment was predictive of response at late time points [r2=0.746] with higher accuracy than anatomical  measurements (Supplementary Fig. C-D).

Conclusions

To our knowledge, this is the first report of an OX40 PET tracer that enables specific imaging of OX40+ effector T cell driven immune responses. In vivo, OX40-PET coupled with immunological and statistical techniques revealed new insights into response following in situ tumor vaccination with CpG, an adjuvant immunotherapy currently in clinical trials [2]. OX40-ImmunoPET provides a readily translatable approach for monitoring activated T cells with high sensitivity and specificity for clinical cancer immunotherapy strategies. 

References

[1] Weinberg AD, Morris NP, Kovacsovics-Bankowski M, Urba WJ, Curti BD. Science gone translational: the OX40 agonist story. Immunological reviews. 2011;244(1):218-231.

[2] Brody JD, Ai WZ, Czerwinski DK, et al. In situ vaccination with a TLR9 agonist induces systemic lymphoma regression: a phase I/II study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2010;28(28):4324-4332.

Acknowledgement

The authors would like to acknowledge the Stanford Center for Innovation in In-Vivo Imaging (SCI3). This work was supported in part by funding from the Ben & Catherine Ivy Foundation (SSG), the Canary Foundation (SSG), NCI R01 1 CA201719-02 (SSG), and the Leukemia and Lymphoma Society (RL).

Figure 1: OX40-ImmunoPET and characterization of OX40 expression in an in situ cancer vaccine model

A) PET/CT (VRT) images at day 2/day 9 post therapy (injected tumors-white arrows; distal tumor-blue arrows) B. 64Cu-DOTA-mAbOX40 uptake [mean +/- SEM, n=3-7). Two-way ANOVA with Bonferroni post-test for multiple comparisons.  C-D. OX40+ subset is restricted to CD4+ T cells and increases locally with CpG. One-way ANOVA with Bonferroni post-test. ****p < 0.0001 ***p < 0.001; ** p < 0.01; *p < 0.05. 

Supplementary Figure

A) PMA/Ionomycin or Dynabead activated T cells, showed increased OX40 expression vs. resting cells(p<0.001). Higher 64Cu-DOTA-mAbOX40 uptake in activated vs resting cells is significantly reduced in blocked & OX40-/- T cells. B) In vivo study design. C) Linear regression model: early PET signal in the local tumor is more predictive of response at late time point vs. D) Day 2 anatomic measurements.

Keywords: Positron emission tomography, OX40, T cells, cancer immunotherapy, in situ vaccine model
2:00 PM PS-15-3

MicroSPECT/CT imaging to monitor subsequent changes in tumor PD-L1 expression after radiotherapy (#195)

P. J. Wierstra1, J. D. M. Molkenboer-Kuenen1, G. Sandker1, J. Bussink2, M. Gotthardt1, E. H. J. G. Aarntzen1, S. Heskamp1

1 Radboud University Medical Centre, Radiology and Nuclear Medicine, Radboud Institute of Molecular Life Sciences, Nijmegen, Netherlands
2 Radboud University Medical Centre, Radiation Oncology, Nijmegen, Netherlands

Introduction

Immune checkpoint inhibition (ICI) therapies have proven to be effective anti-cancer treatments. However, not all patients respond to these drugs and are exposed to unnecessary side-effects while alternative therapy is delayed. Increasing tumor PD-L1 expression with radiotherapy could be a strategy to optimize ICI treatment response. However, the expression of PD-L1 is a dynamic process, which can change during disease progression and treatment[1, 2]. Here, we investigated the effect of radiotherapy on tumor PD-L1 expression using microSPECT/CT imaging in mice with syngeneic tumors.

Methods

BALB/c mice were injected with tumor cell lines with differential expression of PD-L1; colorectal cancer CT26 cells (n=12) and C57/Bl6 mice were inoculated with melanoma B16-F1 (n=12) or Lewis lung carcinoma LLC1 cells (n=12) on the right hind legs. In half of the mice, tumors were irradiated with a single dose of 10 Gy. The next day, mice were injected with 23.8 ± 1.7 MBq 111In-anti-murine PD-L1 (111In-mPD-L1, 30 µg). After 24 h, microSPECT/CT imaging and ex vivo biodistribution studies were performed; together with immunohistochemical analysis of tumor PD-L1.

Results/Discussion

Uptake of 111In-mPD-L1 was significantly increased in CT26 tumors after irradiation (26.3 ± 2.0 vs. 17.1 ± 3.1%ID/g, p = 0.003). A smaller, but significant effect was observed for LLC1 (15.7 ± 1.8 versus 12.3 ± 1.7 %ID/g, p = 0.033). For B16-F1 tumors, the difference in tracer uptake between irradiated vs. non-irradiated tumors was not significant (16.7 ± 3.5 vs. 14.9 ± 6.8 %ID/g). Uptake in draining lymph nodes of the tumor was increased in LLC1 and B16-F1 tumor bearing mice (LLC1: 11.6 ±1.7 versus 9.0 ± 0.8 %ID/g, p = 0.036, B16-F1: 13.1 ± 1.7 versus 7.6 ± 1.2 %ID/g, p = 0.002). No significant differences in splenic uptake were observed. Immunohistochemical staining showed a striking upregulation of tumor PD-L1 in CT26 tumors, and moderate increase in LLC1 tumors, which was related to increased PD-L1 expression.

Conclusions

In this study we demonstrated that radiation induced an upregulation of PD-L1 expression in CT26 and LLC1 xenografts. Also, we demonstrated that this dynamic expression of PD-L1 can be monitored and quantified non-invasively using 111In-mPD-L1 mAb imaging. Studies on the effects radiotherapy on PD-1L expression are ongoing and will allow rational design of novel combination therapies in which radiotherapy complements ICI. Visualization of PD-L1 has the potential to determine when a window of ICI treatment opportunity occurs, during which patients are likely to respond to better to ICI therapy.

References

  1. Taube, J.M., Unleashing the immune system: PD-1 and PD-Ls in the pre-treatment tumor microenvironment and correlation with response to PD-1/PD-L1 blockade. Oncoimmunology, 2014. 3(11): p. e963413.
  2. Vilain, R.E., et al., Dynamic changes in PD-L1 expression and immune infiltrates early during treatment predict response to PD-1 blockade in melanoma. Clin Cancer Res, 2017.
Comparison in 3 cell lines of radiotherapy induced changes in tumor PD-L1 expression
MicroSPECT/CT images (top panel) of mice with irradiated(10 Gy) and non-irradiated tumors 1 day post injection of 30 µg 111In-anti-mPD-L1. Middle panel shows the immunohistochemical analysis of PD-L1 expression of these tumors and the bottom panel shows the quantification of the uptake of 111In-anti-mPD-L1 in tumors, lymph nodes, and spleen.
Keywords: PD-L1, Radiotherapy, SPECT, Quantification
2:10 PM PS-15-4

Immune imaging of human PD-L1 levels in cancer using single domain antibodies (#290)

K. Broos1, Q. Lecocq1, J. Bridoux2, G. Raes3, 4, C. Xavier2, M. Keyaerts2, 5, N. Devoogdt2, K. Breckpot1

1 Vrije Universiteit Brussel, Laboratory for molecular and cellular therapy, Jette, Belgium
2 Vrije Universiteit Brussel, 2In Vivo Cellular and Molecular Imaging, Jette, Belgium
3 Vrije Universiteit Brussel, 3Cellular and Molecular Immunology, Brussels, Belgium
4 Vlaams Instituut voor biotechnologie, 4Myeloid Cell Immunology Lab, Ghent, Belgium
5 UZ Brussel, 5Nuclear Medicine Department, Jette, Belgium

Figure 1

(A) SPECT/CT images to determine the uptake of 99mTc-labeled Nb K2 in athymic nude mice bearing PD-L1 negative ( PD-L1-; left; n=4)  or PD-L1 positive (PD-L1+ ; right; n=6) 624 mel-cells (n = 6).

(B) Gamma counting to determine the %IA/g uptake of sdAb K2 in a PD-L1 negative (n=4) or PD-L1 positive tumor (n=6).

2:20 PM PS-15-5

Molecular imaging of anti-EGFR UniCAR immunotherapy (#242)

R. Bergmann1, S. Albert2, A. Feldmann1, N. Berndt3

1 Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiopharmaceutical Cancer Research, Dresden, Saxony, Germany
2 University Cancer Center (UCC) ‘Carl Gustav Carus’ TU Dresden, Tumor Immunology, Dresden, Saxony, Germany
3 Nationales Centrum für Tumorerkrankungen, DKFZ, Heidelberg, Baden-Württemberg, Germany

Introduction

Adoptive T-cell therapy with chimeric antigen receptor (CAR) engineered T cells (CAR T cells) has shown clinical efficacy in hematologic malignancies with more modest responses when targeting solid tumors. However, CAR T cells can lead to even life-threatening off-tumor, on-target side effects if CAR T cells cross react with healthy tissues. The UniCAR system consists of two components: (1) a CAR for an inert manipulation of T cells and (2) specific targeting modules (TMs) for redirecting UniCAR T cells in an individualized time- and target-dependent manner that will be reported.

Methods

The UniCAR-binding domain is based on the mAb anti-La 5B9 that recognizes a continuous sequence of ten amino acids (E5B9 tag) of the nuclear protein La/SS-B. The TM consists of an anti EGFR domain that was conjugated with the E5B9 tag. The in vivo killing of A431-Luc+ cells was measured by luciferase assay. Conjugation of the TM with p-SCN-Bn-NODAGA resulted in a protein with 3 attached NODAGA moieties and was 64Cu-labeled according standard methodologies reaching specific activities larger than 80 GBq/µmol TM. The 64Cu-TM were i.v. and s.c. injected without or together with A431 and/or UniCAR T-cells in NMRI nu/nu mice; and the in vivo kinetics was measured by small animal PET.

Results/Discussion

The 64Cu-TMs were fast distributed in the body after i.v. injection. Main acceptors of the 64Cu-TMs are the kidneys; after 4 h 20 ± 3%ID of the TMs were trapped in the kidney cortex, while 10 ± 9%ID were eliminated into the urine, and 8 ± 3%ID were accumulated in the liver. The 64Cu-TMs accumulated in the tumors with a peak time of 1.7 h with an activity concentration of 2.5 ± 0.8 (SUV) that was changed to 0.33 ± 0.03 after 28 h. The delivery half-lifes of the 64Cu-TMs from the subcutaneous injection sites were 1.7 h (64Cu-TM), 7.0 h (64Cu-TM & UniCAR T-cells), 15.5 h (64Cu-TM & UniCAR T cells & A431 cells), and 19.4 h (64Cu-TM & A431 cells),  the TMs delivered into the circulation were immediately trapped by the kidney; as result the activity in the kidneys increased linearly over the time. The luminescence imaging of the tumors showed the efficient killing by the UniCAR system.

Conclusions

The 64Cu-TMs have a well-defined clearance rate from the circulation, the subcutaneous injection site and from the tumors. The combination of different imaging technologies and the knowledge about the 64Cu-TM kinetics in vivo, accounting for all involved potential interactions with the respective UniCAR components, should allow defining optimal conditions for the application of UniCARs in immunotherapy, like the time of activation of the UniCAR T cells, and consequently interrupting an ongoing therapy if necessary in case of severe side effects occur.

Kinetics of [64Cu]Cu-α-EGFR TM (pro) (NODAGA)1.2
Kinetics of [64Cu]Cu-α-EGFR TM (pro) (NODAGA)1.2 in selected organs and tissus and orthogonal images of the TM accumulation in the NMR Foxn1 nu/nu mouse bearing a A431 tumor after 90 min.
Retargeting Therapy of EGFR-positive tumor cells
Luminescence images of the subcutaneous injection sites of the control and treatment groups in the UnCAR anti-EGFR therapy.
Keywords: Immunotherapy, PET, half-life, UniCAR, Luminescence, Target module
2:30 PM PS-15-6

In vivo tracking of CAR-T by [18F]BF4-PET/CT in human breast cancer xenografts reveals differences in CAR-T tumour retention. (#112)

E. Kurtys1, L. Lim1, F. Man1, A. Volpe1, G. O. Fruhwirth1, 2

1 King's College London, 1. Department of Imaging Chemistry and Biology, School of Biomedical Engineering and Imaging Sciences, London, United Kingdom
2 King's College London & UCL, Comprehensive Cancer Imaging Centre King's College London & UCL, London, United Kingdom

Introduction

Genetically modified T cells are emerging anti-cancer immunotherapeutics. Challenges for T cell immunotherapy of solid tumours include on-target off-site toxicities against healthy tissues, in vivo relocalization reducing on-site efficacy, and cytokine storm. Non-invasive cell tracking can characterise in vivo cell therapy distribution and relocalization kinetics as well as identify off-site targets.

Our goal was to quantify tumour retention and potential relocalization of an anti-ErbB family chimeric antigen receptor T cell (CAR-T) in mouse models of triple-negative human breast cancer (TNBC.)

Methods

The T1E28z CAR-T1 was made in vivo traceable by PET using the human sodium iodide symporter (NIS) as a reporter, which was additionally fused to red fluorescent protein (RFP)2 to aid preclinical studies. NIS-RFP:CAR-T were generated by lentiviral transduction, selectively expanded via co-expressed IL-4:IL-2/15 receptor chimera3, and fully characterized in vitro for reporter (expression, localization, radiotracer uptake) and CAR function (IFNγ release, tumour cell killing). MDA-MB-436 and MDA-MB-231 breast tumours were established orthotopically in young-adult female immunodeficient NSG mice. NIS-RFP:CAR-T was intratumorally injected. Tumour retention was quantified by PET/CT over weeks. Subsequent to animal culling, tissues were collected for ex-vivo analyses (γ-counting, histology).

Results/Discussion

In vivo traceable NIS-RFP:CAR-T were found to be fully functional (reporter and CAR) without negative impacts of radiolabelling (Fig.1). NIS-RFP:CAR-T were successfully detected at various concentration levels using [18F]BF4- PET/CT imaging. Only live NIS-RFP:CAR-T were detectable due to NIS function being coupled to an intact Na+/K+-gradient. Intratumourally administered NIS-RFP:CAR-T were alive in vivo for at least 15 days. In MDA-MB-436 tumours (Fig.2), [18F]BF4- uptake significantly increased 7d post administration (13.0±1.9, N=5 vs. 9.0±3.1, N=9, P=0.0194) suggesting on-site expansion of traceable NIS-RFP:CAR-T cells. After 15d, [18F]BF4- uptake was comparable to day 1 levels (8.86±2.8, N=9, P=0.9493). Interestingly, the [18F]BF4- uptake was significantly lower after 15d in highly metastatic MDA-MB-231 cells (20.3±9.6% of day 1; P=0.0069; N=4), suggesting relocalization or on-site cell therapy death. Ex vivo γ-counting and histology confirmed in vivo imaging results.

Conclusions

NIS reporter gene imaging by [18F]BF4- -PET is a sensitive tool for monitoring anti-cancer T cell therapy in breast cancer. PET imaging informed on T cell tumour presence, which is a prerequisite for their therapeutic effect. In vivo imaging allowed quantification of NIS-RFP:CAR‑T retention in primary tumours and identified tumour model-specific differences. Further research is required to elucidate the molecular reasons for the observed differences of anti-ErbB CAR-T retention in TNBC models.

References

1Davies et al., Mol Med, 2012; 18:565-76.

2Diocou et al, Sci Rep 2017; 7(1):946

3Wilkie S. et al. J Biol Chem 2010; 285(33):25538-44.

Acknowledgement

We would like to thank Dr Maher for the T1E28z CAR, and Cancer Research UK, EPSRC, Worldwide Cancer Research and King’s Health Partners for financial support.

Fig 1 In vitro expansion and validation of NIS-RFP:CAR-T cells.
Fig.2 PET imaging and ex-vivo biodistribution of [18F]BF4- in the MDAMB436 xenograft tumour model.
Keywords: Immunotherapy, PET, CAR T cells, brest cancer
2:40 PM PS-15-7

Metabolic profiling of secondary lymphatic organs by 18F-FDG-Positron Emission Tomography (PET)/CT in checkpoint inhibitor immunotherapy (CIT) patients with metastatic melanomas as a novel tool to predict the immune response (#401)

J. Schwenck1, 2, B. F. Schörg1, F. Fiz2, K. Wistuba-Hamprecht3, A. Forschner3, T. Eigentler3, B. Weide3, C. Garbe3, M. Röcken3, C. Pfannenberg4, B. J. Pichler1, C. la Fougère2, M. Kneilling1, 3

1 Eberhard Karls University, Werner Siemens Imaging Center, Tübingen, Baden-Württemberg, Germany
2 Eberhard Karls University, Department of Nuclear Medicine and Clinical Molecular Imaging, Tübingen, Baden-Württemberg, Germany
3 Eberhard Karls University, Department of Dermatology, Tübingen, Baden-Württemberg, Germany
4 Eberhard Karls University, Department of Diagnostic and Interventional Radiology, Tübingen, Baden-Württemberg, Germany

Introduction

CIT prolongs the overall survival in the majority of patients with metastatic melanoma but still a large percentage of patients does not benefit for unknown reasons. Although the knowledge about the exact mechanisms of CIT is limited, there is evidence that a systemic immune response is needed to enable an effective CIT-induced anti-tumor response. Aim of our retrospective study was to identify CIT responders by analysis of CIT-specific alterations of the glucose metabolism in secondary lymphoid organs such as the spleen of metastatic melanoma patients by 18F-FDG-PET.

Methods

In experimental tumor models we were able to differentiate between effective and non-effective immunotherapies by analyzing the 18F-FDG-uptake in the spleen. Thus, we  investigated retrospectivly 18F-FDG-PET/CT scans of 38 patients with metastatic melanoma pre- and post-therapy with CTLA-4 or PD-1 Ab (21 responder: 5x nivolumab; 7x pembrolizumab; 9x ipilimumab; 17 non-responder: 2x nivolumab; 11x pembrolizumab; 4x ipilimumab). Regions of interest (ROI) in the spleen were defined in the CT images and copied to the coregistered PET for semiquantitative analysis. Total lesion glycolysis (TLG) was calculated by multiplication of the spleen volume and the SUVmean.

Results/Discussion

We observed no significant differences between responders and non-responders in the baseline 18F-FDG-PET/CT-scans before CIT) neither by analyzing the spleen volume (221±18 cm3 vs. 209 ±22 cm3) nor the spleen 18F-FDG-uptake (SUVmean: 1,74±0,06vs.1,72±0,05; TLG: 384±37 vs. 359±36).

After onset of CIT the follow up 18F-FDG-PET/CT-scans provided a comparable increase in spleen volume in responders (+8±6%) and non-responders (+7±5%), but 15 out of 21 responders showed an a higher 18F-FDG uptake in spleen when compared to the baseline 18F-FDG-PET/CT-scans. The mean standard uptake values in the spleen of responders increased by +10±9% SUVmean, while hardly any change were observed in non-responders (SUVmean -1,3±2,6%). Furthermore, the total lesion glycolysis (TLG) in the CIT-responders increased stronger (+25±22%) than in non-responders (+6±6%).

Conclusions

The results of our retrospective study imply an association of the CIT-induced anti-tumor therapy effect with metabolic changes in secondary lymphatic organs, which was detectable by non-invasive 18F-FDG-PET/CT. Whether this may represent a novel powerful tool to monitor CIT-induced systemic immune responses, has to be uncovered by preclinical research focusing on the exact mode of action of the CIT-induced systemic immune response and prospective clinical studies to evaluate the prognostic value.

2:50 PM PS-15-8

Visualizing hypoxia-mediated immunosuppression in bone metastasis (#458)

B. Weigelin1, 2, M. Giampetraglia1, W. Tian1, E. Dondossola1, D. Hutmacher3, X. Lu4, R. de Pinho4, C. Logothetis1, P. Friedl2, 1

1 MD Anderson Cancer Center, Genitourinary Medical Oncology and Koch Center, Houston, United States of America
2 Radboud University Medical Center, Cell Biology, Nijmegen, Netherlands
3 Queensland University of Technology, Brisbane, Australia
4 MD Anderson Cancer Center, Cancer Biology, Houston, United States of America

Introduction

For many cancer types, immunotherapy has the potential to reach long-lasting remission in patient subsets. However, bone metastases typically resist immunotargeting, implicating the bone stroma as contributor to cancer resistance. Despite high vascular density, local oxygen tension in the bone marrow is overall low and heterogeneous (0.5 – 5 % in mice). Using advanced microscopy, we here tested whether hypoxia, via activation of hypoxia-inducible factor-1 alpha (Hif-1a) contributes to local cancer resistance towards T cell-mediated effector function in bone.

Methods

B16F10 melanoma and PPSM prostate cancer cells expressing the ovalbumin antigen were confronted with activated OT-1 CTL in vitro and in vivo. By combining a collagen-based organotypic assay and an automated image segmentation workflow, we developed a 3D medium-throughput screening assay to identify immunosuppressive conditions derived from the bone microenvironment. Results were validated in bone metastatic lesions, relating CTL position and function to vascularization, hypoxia and Hif-1a expression using bone clearing (Bone CLARITY) and 3D reconstruction of the mouse tibia by multiphoton microscopy. To monitor CTL killing efficacy in live tissue, cancer cells were implanted in a subcutaneous tissue-engineered bone ossicle which allows for intravital imaging at single-cell resolution.

Results/Discussion

Using the 3D cytotoxicity assay, we found that prostate cancer cells develop resistance towards CTL-mediated killing at oxygen levels of 1.5 % while lower oxygen levels (0.5 %) were required to induce resistance in melanoma cells. Resistant cancer cell subsets showed nuclear accumulation of Hif-1a and re-sensitization was achieved by blocking Hif-1a activation using glyceryl trinitrate. Resistance was not associated with upregulation of PD-L1 and CTL remained negative for the immune checkpoint molecules PD1 and VISTA. Monitored through an imaging window or by 3D reconstruction after tissue clearing, PPSM/OVA cells were poorly targeted by CD8 CTL. CTL infiltration was dependent on the size of the lesion and the antigen-specificity of the CTL. Large, established lesions typically excluded CD8 T cells while small, early microlesions were efficiently infiltrated. Both, large and small lesions were resistant despite systemic application of OVA-specific adoptively-transferred OT1 CTL.

Conclusions

By combining in vitro screening, ex vivo whole-organ 3D reconstruction using tissue clearing, and in vivo dynamic intravital imaging, this workflow will serve to identify the cellular mechanisms that compromise immune-effector function and provide rationales for therapeutic combinations to enhance immunotherapy in bone.

Keywords: Cytotoxic T cells, intravital microscopy, hypoxia, bone metastasis