Patient-Derived Head and Neck Cancer Organoids Recapitulate EGFR Expression Levels of Respective Tissues and Are Responsive to EGFR-Targeted Photodynamic Therapy

Abstract

Patients diagnosed with head and neck squamous cell carcinoma (HNSCC) are currently treated with surgery and/or radio- and chemotherapy. Despite these therapeutic interventions, 40% of patients relapse, urging the need for more effective therapies. In photodynamic therapy (PDT), a light-activated photosensitizer produces reactive oxygen species that ultimately lead to cell death. Targeted PDT, using a photosensitizer conjugated to tumor-targeting molecules, has been explored as a more selective cancer therapy. Organoids are self-organizing three-dimensional structures that can be grown from both normal and tumor patient-material and have recently shown translational potential. Here, we explore the potential of a recently described HNSCC-organoid model to evaluate Epidermal Growth Factor Receptor (EGFR)-targeted PDT, through either antibody- or nanobody-photosensitizer conjugates. We find that EGFR expression levels differ between organoids derived from different donors, and recapitulate EGFR expression levels of patient material. EGFR expression levels were found to correlate with the response to EGFR-targeted PDT. Importantly, organoids grown from surrounding normal tissues showed lower EGFR expression levels than their tumor counterparts, and were not affected by the treatment. In general, nanobody-targeted PDT was more effective than antibody-targeted PDT. Taken together, patient-derived HNSCC organoids are a useful 3D model for testing in vitro targeted PDT.

Keywords: EGFR; HNSCC; organoids; targeted photodynamic therapy.

Figure 1

Epidermal Growth Factor Receptor (EGFR) expression differs between patient-derived organoids from different donors and recapitulates EGFR levels of respective tissues. (a) EGFR protein expression detected by flow cytometry. EGFR expression of organoids grown in either physiological Epidermal Growth Factor (EGF) (blue peak) or high EGF (ref peak) medium is shown for two HNSCC organoid lines. An unstained control is shown in black. (b) EGFR protein expression measured by flow cytometry in 2D cell lines commonly used in in vitro EGFR-targeted PDT studies, organoid lines derived from HNSCC patients both normal and tumor, and in primary tissue samples. For organoids, the experiment was performed in technical duplicate (error bars) and biological triplicate (individual bars). EGFR expression was stable over time, as biological replicates were measured two months apart. EGFR protein levels are shown relative to HeLa cells (set at 100%). Results of tumor organoids are shown in filled bars, results of wildtype organoids are shown in clear, outlined bars. For primary tissues, each bar represents EGFR expression on a tissue sample derived from an individual patient. (c) EGFR immunohistochemical staining performed on N8 and T8 organoids. Scalebar, 100 µm.

Figure 2

Organoid response to in vitro EGFR-targeted PDT is donor-dependent and tumor-specific. (a) Schematic outline of the experimental set-up of organoid in vitro PDT. Organoids are disrupted into single cells, recovered for two days on medium containing physiological EGF, and subsequently filtered, counted, and plated into a 384 well format. A two-hour exposure to EGFR-targeting nanobody-PS or antibody-PS conjugates was followed by a light-dose activating the PS. Twenty-four hours later, cell viability was assessed. (b) EGFR-targeting conjugates used in this study did not show toxicity without activation of the PS. Here, toxicity of 7D12-9G8 in T3 and T4 is shown as an example. (c) Response to EGFR-targeted PDT of matched normal and tumor organoid pairs. Response to 7D12-PS, 7D12-9G8-PS, and cetuximab-PS is shown for N4 and T4 (orange), N5 and T5 (red), and N8 and T8 (blue). Normal organoids are depicted in a lighter shade of the same color than their tumor counterparts. (d) Quantification of organoid response to EGFR-targeted PDT shown in Figure 2c Area under the curve (AUC) is used as a readout for response to EGFR-targeted PDT. Blue indicates low AUC-values (response to therapy), whereas red indicates high AUC values (no response to therapy).

Figure 3

Organoid response to in vitro EGFR-targeted PDT correlates with EGFR expression levels. (a) Sensitivity of HNSCC organoids with variable EGFR expression to PDT using conjugates 7D12-PS (first panel), 7D12-9G8-PS (second panel), and cetuximab-PS (third panel). Color of the lines indicate the relative EGFR expression level, detected by FACS analysis (blue = highest expression, yellow = lowest expression). (b) Correlation plots showing the relation between EGFR expression on the x-axis and response to PDT therapy, as indicated by area under the curve (AUC) on the y-axis. The first panel shows the AUC for 7D12-PS, second panel for 7D12-9G8-PS, and the third panel for cetuximab-PS.

Figure 4

Induced EGFR expression enhances the response to EGFR-targeted PDT. (a) Schematic outline of lentiviral infector used to create organoid lines that can be induced to overexpress EGFR (TRE: tetracycline responsive element, CMV: cytomegalovirus promotor). The TRE and CMV promotor precede the open reading frame encoding EGFR and GFP, separated by a T2A sequencing, assuring generation of separate mRNA molecules for EGFR and GFP. (b) Merged brightfield/immunofluorescence images of organoids after two-day administration of doxycycline. Scalebar 400 μm. (c) and (d). Effect of doxycycline-mediated EGFR overexpression on EGFR protein levels in organoid lines T6 and T8, respectively. Colored peaks indicate uninduced expression, lined peaks indicate induction of EGFR expression. (e) and (f). EGFR-targeted PDT using nanobody 7D12-9G8-PS in EGFR overexpressing organoid lines T6 and T8. Organoids were either doxycycline-induced (squared symbols, dashed line) or uninduced (round symbols, solid line) and exposed to PDT as previously described.