Targeting PLK1 Reduces MMP10 to Enhance Radiosensitivity in HPV− Head and Neck Cancer

Abstract

Purpose:

Radiotherapy is a key treatment for head and neck squamous cell carcinoma (HNSCC), yet local recurrence remains a challenge, especially in patients with human papilloma virus-negative (HPV−) HNSCC. Our studies identify polo like kinase 1 (PLK1) as a promising target for HNSCC. PLK1 is overexpressed in HNSCC and associated with worse survival.

Experimental Design:

Onvansertib was used to assess the effect of PLK1 inhibition on cell viability and spheroid growth in HPV− and HPV+ HNSCC cells. Cell cycle analysis was done to assess G2/M arrest when combining PLK1 inhibition with radiation. Colony formation and in vivo tumor growth assays were done to evaluate the efficacy of the combination of PLK1 inhibition with radiation. RNA-sequencing was done to identify targets of PLK1 after onvansertib treatment. Plk1 siRNA knockdown was used to confirm similar responses seen when inhibiting PLK1 with onvansertib.

Results:

PLK1 inhibition with onvansertib reduced cell viability and spheroid growth of HPV− HNSCC cells. PLK1 inhibition combined with radiation increased G2/M arrest, decreased colony formation of HPV− HNSCC cells, and reduced tumor growth in vivo. RNA-sequencing demonstrated radiation increased MMP10 expression but PLK1 inhibition reversed this effect in HPV− HNSCC cells. PLK1 inhibition reduced MMP10 activity, and Plk1 siRNA knockdown reduced MMP10 expression in HPV− HNSCC cells.

Conclusions:

These findings suggest that combining PLK1 inhibition with radiation may improve therapeutic response for HPV− HNSCC via MMP10, offering a novel approach to overcome radiation resistance.

Translational Relevance:

Patients with head and neck squamous cell carcinoma (HNSCC) treated with curative radiation therapy commonly experience local recurrence, highlighting the need for improved therapeutic strategies. Onvansertib, a highly specific polo-like kinase 1 (PLK1) inhibitor, enhances the efficacy of ionizing radiation making it a promising candidate to target, particularly in human papilloma virus negative (HPV−) HNSCC patients, a more resistant patient population. While onvansertib has shown efficacy in combination with other chemotherapies in phase 1 and 2 clinical trials, it has not been used to treat HNSCC. Moreover, onvansertib has not been combined with radiation in clinical trials to treat any cancer. Notably, our studies suggest that onvansertib reduces MMP10 expression in HPV− HNSCC. This indicates a potentially novel approach to reducing MMP10 signaling as MMP inhibitors have failed in the clinic and no MMP10 specific inhibitor currently exists.

Introduction:

Head and neck squamous cell carcinoma (HNSCC) is the 6^th most common cancer worldwide. Though incidence is rising, there have been minimal improvements in survival outcomes over the last 30 years.^1–3 HNSCC is categorized into two subgroups based on oncologic presence of the human papilloma virus (HPV): HPV-positive (HPV+) and HPV-negative (HPV−), with risk factors for HPV− HNSCC including excessive tobacco and alcohol use.² Clinically, HPV status confers a disparity in therapeutic response to ionizing radiation where patients with HPV− HNSCC demonstrate relative treatment resistance and subsequently worse survival outcomes compared to patients with HPV+ HNSCC.⁴ In addition, many patients with HNSCC are ineligible for concurrent cisplatin chemotherapy due to potential toxicity.⁴ The inferior response of patients with HPV− HNSCC to chemotherapy and radiation indicates a dire need to identify novel targeted therapies to combine with radiation to improve HNSCC survival outcomes.

Polo like kinase 1 (PLK1) is a major cell cycle regulator that controls the progression of cells through G2/M checkpoints including centrosome maturation, spindle assembly, and cytokinesis.⁵ PLK1 is a serine/threonine kinase that is activated by upstream phosphorylation of the Aurora Kinase A (AURKA) and Bora complex.⁶ Overexpression of PLK1 is associated with a poor prognosis in many cancers including breast, prostate, colorectal, and HNSCC.^{7, 8} PLK1 controls multiple processes such as proliferation, epithelial to mesenchymal transition (EMT), cell death, and immune system activation which contribute to cancer progression.^{8, 9} However, the role of PLK1 in HNSCC is not well understood. In addition, there are few studies investigating the combination of PLK1 inhibition with radiation in HNSCC although this combination has shown some efficacy in other cancer types.^10–12

PLK1 interacts with many targets that drive EMT including the matrix metalloprotease (MMP) family.^13–15 Overexpression of MMPs drives tumorigenesis by degrading the extracellular matrix of cancer cells to induce cell migration and invasion.^16–18 Specifically, MMP10 is highly expressed in HNSCC and correlates with metastasis.¹⁹ In addition, radiation induces certain MMPs at early timepoints.^20–22 However, the interaction of PLK1 and MMP10 has yet to be investigated. In lung and thyroid cancers, PLK1 suppression reduced migration and invasion and subsequently reduced MMP expression.^{23, 24} Together, these studies support a potential role for PLK1 inhibition to reduce MMP10 expression in HNSCC.

In this study, we investigated the role of PLK1 inhibition in HNSCC using a relatively new and highly specific PLK1 inhibitor (onvansertib) with existing human safety data. Onvansertib has already been reported to be well tolerated in patients with colorectal cancer.²⁵ We found that onvansertib reduced the cell viability of HPV− HNSCC cells more than HPV+ HNSCC cells. We demonstrated that the combination of onvansertib and radiation reduced colony formation and induced G2/M arrest more than either single agent alone in HPV− HNSCC cells. Importantly, the combination of onvansertib and radiation also significantly reduced tumor growth in a HPV− HNSCC in vivo model. We also identified a novel relationship between PLK1 and MMP10, as both MMP10 expression and activity were reduced upon PLK1 inhibition in HPV− HNSCC cells. Our findings suggest that the use of PLK1 inhibition, with onvansertib, in combination with radiation may be a robust approach to improve survival outcomes for patients with HPV− HNSCC .

Materials and Methods:

Cell Lines and Onvansertib

HPV− [HN5 (RRID:CVCL_8128), Cal27 (RRID:CVCL_1107), UMSCC1 (RRID:CVCL_7707)] and HPV+ [UMSCC47 (RRID:CVCL_7759), UDSCC2 (RRID:CVCL_E325), 93VU147T (RRID:CVCL_L895)] HNSCC cell lines were used throughout this study. HNSCC cell lines were cultured in DMEM (Corning, Cat# 10-013-CV) containing 10% FBS, 1% sodium pyruvate, 1% non-essential amino acids, 1% L-glutamine, and 1% penicillin-streptomycin. Cells were maintained at 37°C with 5% CO₂. HNSCC cells were tested every 6 months for mycoplasma, and STR profiling was done to authenticate these cells. Onvansertib was obtained from Selleckchem (Cat# S7255) for in vitro studies and generously provided by Cardiff Oncology for in vivo studies.

X-ray Irradiation

Radiation was delivered using a Precision X-ray CellRad+ cabinet irradiator (in vitro) and an Xstrahl XenX cabinet irradiator (in vivo). Radiation was delivered at 160 kV and 6.25 mA using 0.5 mm Al filtration for the Precision cabinet irradiator. Radiation was delivered at a dose rate of 5.77 cGy/min using 0.15 mm Cu filtration at 220 kV and 13 mA for the Xstrahl XenX cabinet. Quality assurance (QA) is run daily for each machine.

Acid Phosphatase (APH) Cell Viability Assay

HNSCC cells were co-incubated with onvansertib for 48 hours and then washed with 1X PBS. On the second wash, 100μl of 1X PBS was retained in the wells. PNPP tablets (Thermo Fisher, Cat# 34047) were added to APH Buffer (3M sodium acetate, 0.1% Triton X-100, ddH₂O) to make a PNPP solution (4 tablets per 10ml). 100μl of PNPP solution was added to each well in the dark for 4 hours. Absorbance was read at 405 nm and cell viability was normalized to the vehicle control.

3D Spheroid Growth Assay

Matrigel (Corning, Cat# 354230) was thawed at 4°C overnight before plating into a chilled 24-well plate. The coated plate was placed into the incubator (37°C) to solidify the Matrigel. A 2% Matrigel and DMEM solution was made by adding Matrigel to cool media and then left at room temperature. HNSCC cells were seeded at 10,000 cells per well in the 2% Matrigel DMEM overlaying the solidified Matrigel. Spheroids grew for 48 hours before onvansertib (25 nM, in 2% Matrigel DMEM) treatment. Onvansertib was replaced every other day for 7 days. Spheroids were imaged daily, with representative 10X images chosen on day 7.

Western Blot

Whole cell lysates were collected using RIPA buffer (0.5% sodium deoxycholate, 150 nM NaCl, 50nM Tris-HCL, 0.1% SDS, 1% NP-40) containing 1X Protease Cocktail Inhibitor (ThermoFisher Cat# 7844C). Lysates were centrifuged at 17,000xg for 10 minutes at 4°C. Samples were made with 6X loading buffer (1.2ml 0.5M Tris-HCl pH 6.8, 6mg bromophenol blue, 1.2g SDS, 4.7ml glycerol, 0.93g DTT, 2.1ml ddH₂O) then denatured at 95°C for 7 min. Samples were run on gradient gels (BioRad Cat# 4561094) at 85V for 1.5 hours then transferred at 100V for 1 hour on ice onto nitrocellulose membrane (Cytiva, Cat# 10600004). Membranes were blocked for 1 hour in 1X PBST with 5% BSA. Membranes were incubated in primary antibody overnight at 4°C then washed 4x with 1X PBST. Membranes were then incubated in secondary antibody at room temperature for 1 hour. Membranes were imaged using a BioRad Chemidoc MP Imager and quantified using the BioRad Image Lab 6.1 software with normalization to β-actin. Primary antibodies used (PLK1 1:1000 CST Cat# 4513 RRID:AB_2167409, pPLK1 1:500 Thermo Fisher Cat# 480028 RRID:AB_2532250, MMP10 1:500 R&D Systems Cat# MAB910 RRID:AB_2144566, β-Actin 1:1000 CST Cat# 4967 RRID:AB_330288, GAPDH 1:1000 CST Cat# 5174 RRID:AB_10622025). Secondary antibodies used (Licor: IRDye 680RD goat anti-mouse Cat# 926-68070 RRID:AB_10956588, IRDye 800CW goat anti-rabbit Cat# 926-32211 RRID:AB_621843).

Colony Formation Assay

HNSCC cells were seeded in 6-well plates at low densities (300–10,000 cells/well). HNSCC cells were treated with onvansertib (5nM) for 24 hours then irradiated (2, 4Gy). Media was changed every 2-3 days to replenish onvansertib. Colonies grew until vehicle colonies began to touch (10-14 days). Colonies were fixed with 10% neutral buffered formalin (NBF) and stained with crystal violet (0.5% crystal violet, 20% methanol) for 30 minutes each at room temperature. Colonies were manually counted and normalized to the number of cells originally plated. Treatment conditions were then normalized to vehicle.²⁶

Cell Cycle Flow Cytometry

HNSCC cells were treated with onvansertib (50nM) for 4 hours then irradiated (2, 4Gy). Cells were collected 16 hours after radiation and washed with 1X PBS. Cells were fixed by adding cold 70% ethanol while vortexing and incubated at 4°C for 30 minutes then transferred to −20°C until staining. To stain, fixed cells were centrifuged at 600xg for 5 minutes then washed 2x with 1X PBS. RNase A (100ug/ml) was added to the cell pellet to ensure only DNA was stained. Propidium iodide (BioLegend, Cat# 421301) was diluted in cell staining buffer then added to the cell pellet. Unstained cells were included as a negative control. Samples were run on a BD LSR Fortessa flow cytometer and analyzed using FlowJo software (RRID:SCR_008520).

In Vivo Xenograft Studies

The University of Cincinnati’s (IACUC) institutional review boards approved all animal experiments. Cal27 cells (1x10⁶) were harvested and injected into the right flank of athymic nude mice (male, 5 week old, homozygous J:Nu, Jackson Laboratories, Strain# 007850 RRID:IMSR_JAX:007850). Cells were injected in saline in a 1:1 ratio with Matrigel to form tumors. Once palpable tumors were formed, mice were sorted into treatment groups to ensure similar tumor size among all groups (~75 mm³). The treatment groups were as follows (10 mice per group): vehicle, onvansertib alone, radiation alone, and onvansertib + radiation. Oral gavage treatment started on day 1, when vehicle and radiation alone groups received vehicle (0.5% methylcellulose, 0.1% Tween-80) and onvansertib and onvansertib + radiation groups received onvansertib (60 mg/kg suspension in vehicle). Treatments were given 6 days per week. Starting on day 2, tumors in the radiation alone and onvansertib + radiation groups were focally irradiated at 2 Gy for 5 consecutive days (10 Gy total). Onvansertib dose (50 mg/kg) was decreased two weeks after treatment began. Tumor volumes and mouse weights were measured every 2-3 days. Tumor volume was calculated using the formula “volume = [(width²) × length]/2”. Tumors were collected and fixed in 10% NBF then weighed and paraffin embedded. UMSCC47 cells (1x10⁶) were injected in a similar manner as the Cal27 cells. Treatment groups were also similar. As UMSCC47 cells are HPV+, they demonstrate increased sensitivity to radiation in our in vitro studies and by others.²⁷ Thus, UMSCC47 tumors were focally irradiated to 2 Gy every other day (6 Gy total). Tumor volumes and mouse weights were measured in a similar manner to Cal27 studies.

Immunohistochemistry (IHC)

Tumor tissue from in vivo studies was paraffin embedded and sectioned . Tissue slides were baked at 60°C for 20 min then deparaffinized in xylenes. Tissue slides were then rehydrated. Antigen retrieval was done in 10mM sodium citrate solution (pH 6, 0.05% Tween-20) in a vegetable steamer for 20 minutes. Slides were cooled to room temperature then peroxidase quenching was done in a 0.6% H₂O₂ solution for 15 minutes. Tissues were blocked (10% normal goat serum in 1X PBST) for 30 minutes. Primary antibody (Ki67, 1:500, CST Cat# 9027 RRID:AB_2636984) was diluted in green antibody diluent (Thermo Fisher, Cat# 3118) and incubated on tissues overnight at 4°C. Secondary antibody (biotinylated goat anti-rabbit IgG, 1:200, Vector Laboratories, Cat# BA-1000-1.5 RRID:AB_2313606) was diluted in 3% normal goat serum and 1X PBS and added to tissues for 1 hour. The ABC kit (Vector Laboratories, Cat# PK-6100) was added to tissue for 30 minutes. The DAB kit (Vector Laboratories, Cat# SK-4100) was made immediately before adding to tissue. Once the tissue was slightly brown, the DAB was removed, and slides were rinsed in tap water. Slides were quickly dipped in Gills3 hematoxylin (Sigma, Cat# GHS316) then thoroughly rinsed with tap water. Tissue was dehydrated. Permount (Fisher, Cat# SP15) was then added to tissue and a coverslip was placed on top. Tissue slides were imaged on a brightfield microscope and representative 10X images were obtained. ImageJ was used for quantification of Ki67 positive cells on each tumor tissue slide.

Tissue Microarray (TMA) Quantification

TMAs were stained by the Integrated Pathology Research Facility at Cincinnati Children’s Hospital and Medical Center. TMAs consisted of HNSCC tissue at biopsy from a previously completed head and neck clinical trial.²⁸ Tissue was stained with MMP10 and blindly quantified on a scale from 0-3 based on intensity of staining as well as given a percentage of the number of cells stained. These two scores were multiplied to give a final MMP10 score by a board-certified pathologist with expertise in head and neck cancer (S.B.).

RT-qPCR

HNSCC cells were treated with onvansertib (100nM) for 4 hours then irradiated (4, 8Gy). RNA was collected 20 hours after radiation using the manufacturer’s protocol for the Qiagen RNA isolation kit (Qiagen, Cat# 74104). RNA concentrations were measured using a Nanodrop. RNA was transcribed into cDNA (iScript Reverse Transcription Supermix, BioRad, Cat# 1708840). cDNA was diluted (1:5) in nuclease free ddH₂O then used for RT-qPCR. Primers were obtained from Integrated DNA Technologies (IDT): Plk1 forward 5’-GCACAGTGTCAATGCCTCCAAG-3’, reverse 5’-GCCGTACTTGTCCGAATAGTCC-3’; MMP10 forward 5’-GCATGTTCTGTGACTGAAGAAGA-3’, reverse 5’-CACATCTATGAAAATACATTCTCTCAC-3’; TBP forward 5′-AGTTCTGGGATTGTACCGCA-3′, reverse 5′-TGTGCACACCATTTTCCCAG-3′. RT-qPCR was performed using iTaq Universal SYBR Green Supermix (BioRad, Cat# 1725120), and the BioRad CFX96 Opus was used to evaluate real-time mRNA expression. To analyze RT-qPCR data, the ΔΔCt method was used and treatments were normalized to vehicle. TBP was used as a housekeeping gene.

Plk1 siRNA Transfection

HNSCC cells were seeded 24 hours before transfection. Plk1 siRNA (10nM siRNA ‘A’ or ‘B’, Origene, Plk1 Cat# SR321347) was prepared by diluting the siRNA in 1X transfection buffer then adding siTran 2.0 siRNA transfection reagent (Origene, Cat# TT320001). The siRNA mixture or scrambled negative control (0.1nM NC, Origene, Cat# SR30004) was incubated at room temperature for 15 minutes then added dropwise to each well. RNA was collected 24 hours after transfection for RT-qPCR.

MMP10 Activity Assay

HNSCC cells were treated with onvansertib (100nM) for 4 hours then irradiated (4, 8Gy). Lysates were collected 20 hours after radiation using the provided assay buffer (SensoLyte 520 MMP10 Assay Kit, AnaSpec, Cat# AS-72024) with 0.1% Triton X-100. Samples were prepared in assay buffer to contain a final protein concentration of 150ug. Recombinant human MMP10 (R&D Systems, Cat# 910-MP) was used as a positive control. Samples were incubated with APMA (1mM) for 1 hour at 37°C then added in duplicate to a black 96-well plate with MMP10 substrate. Assay buffer was used as a blank. The plate was incubated at 37°C for 1 hour in the dark. Fluorescence was measured at excitation/emission of 490/520 + 20nm. The blank was subtracted from all samples and technical replicates were averaged and normalized to vehicle.

MMP10 Zymography

HNSCC cells were treated with onvansertib (100nM) for 4 hours in serum-free media then irradiated (8Gy). Supernatant was collected 20 hours after radiation and spun at 15,000xg for 20 minutes at 4°C in a 10 kDa molecular weight cutoff protein concentrator (ThermoFisher Cat# 88513). Supernatant samples were incubated with APMA (1mM) for 1 hour at 37°C. 2X loading buffer (0.5M Tris-HCl pH 6.8, Glycerol, 10% SDS, Bromophenol blue) was added to samples and then loaded into a 10% acrylamide gel containing 0.1% casein. Gels were run at 85V for 1.5 hours on ice. Gels were washed 2x in renaturing solution (2.5% Triton X-100 in ddH₂O) for 30 minutes each. Gels were then incubated in developing buffer (50mM Tris-HCl pH 7.5, 200mM NaCl, 5mM CaCl₂, 0,02% Brij 35) for 30 minutes. Developing buffer was replaced and gels were incubated for 18 hours at 37°C. Gels were washed in ddH₂O. Coomassie blue (Bio-Rad, Cat# 1610436) was added for 1 hour then removed by washing with destaining solution (40% MeOH, 10% acetic acid, 50% ddH₂O). Gels were imaged using a BioRad Chemidoc MP Imager.

Migration Assay

HNSCC cells were seeded in serum-free media containing onvansertib (50nM) in Transwell inserts (Corning, Cat# 353097) placed in compatible 24-well plates (Corning, Cat# 353504). Media containing serum below the insert served as a chemoattractant. After 24 hours, cells were irradiated (4Gy). 48 hours after irradiation, media was removed from the insert, and the inside was cleaned with a cotton q-tip. Inserts were fixed with methanol then stained with 0.2% crystal violet. Images were obtained using a Leica MZ16FA stereo microscope. The number of migrated cells was counted using ImageJ (RRID:SCR_003070) with a Phalanskar threshold (k = 0.25, r = 0.5) to distinguish migrated cells from the background.

Scratch Assay

HNSCC cells were seeded and irradiated (4, 8Gy) 24 hours later. Wells were scratched with a p1000 pipette tip and cells were treated with onvansertib (25nM) 24 hours after radiation. Images were taken immediately after scratching for a 0 hour timepoint. Representative 4X images taken with a brightfield microscope at 48 hours after onvansertib treatment started are shown. Quantification represents the 48 hour scratch widths normalized to 0 hour.

MMP10 Crispr-Cas9 Knockout

Guide RNAs (gRNA) for MMP10 were combined with a Trac-ATTO 550 RNA (IDT, Cat# 1075927) and incubated at 95°C for 5 minutes. Cas9 (IDT, Cat# 1081058), Cas9 PLUS (ThermoFisher, Cat# CMAX00003), and Opti-MEM were then added, and the mixture was incubated for 5 minutes. Lipofectamine was then added and incubated for 20 minutes. HNSCC cells were plated in a 96-well plate and the transfection mixture was added dropwise on top of the cells. Single cells were sorted into collagen-coated 96-well plates. Sorting was done on cells with a high transfection efficiency (98% fluorescence).

TCGA and CPTAC Analysis

The Cancer Genome Atlas (TCGA) gene expression analysis utilized the cBioPortal database.^29–31 From cBioPortal, the dataset “Head and Neck Squamous Cell Carcinoma (TCGA, PanCancer Atlas)” was selected and queried by gene (Plk1 or MMP10). Gene expression was categorized by HPV status by choosing “Subtype” in the offered tracks. mRNA expression data was evaluated as the log₂ gene expression for HPV− and HPV+ patients. Kaplan Meier survival analysis was done using the cBioPortal and the Human Protein Atlas databases.^32–34 The fragments per kilobase of transcript per million mapped reads (FPKM) values for gene expression were classified into two expression groups (high and low) and the correlation between gene expression and patient survival was evaluated. A “best expression cut off” was analyzed indicating the maximal difference regarding survival when assessing FPKM values. The Clinical Proteomic Tumor Analysis Consortium (CPTAC) database was used to assess mRNA and protein expression of PLK1 and MMP10 in patients with HNSCC compared to normal tissue controls. Cox proportional hazard models with survival in months were used to evaluate Plk1 expression (FPKM) in the HNSCC cohort from the TCGA. Hazard ratios reflect a standard deviation increase of 1 in expression. Analyses were stratified by HPV status and multivariate analysis focused on HPV− patients with all variables available for analysis. Ties were handled with the Breslow method. Proportional hazard assumptions were assessed using Schoenfield residuals. Two-sided p-values <0.05 and 95% confidence intervals are reported.

RNA-Seq and Analysis

HN5 (HPV−) and UMSCC47 (HPV+) HNSCC cells were treated with onvansertib (25nM) for 24 hours then irradiated (4Gy). RNA was collected 24 hours after radiation using the Qiagen RNA isolation kit (Qiagen, Cat# 74104). Three biological replicates were submitted for each cell line including all four treatment groups: vehicle, onvansertib (Onv), radiation (4Gy), and onvansertib + radiation (Onv + Gy). With assistance from the Biostatistics and Bioinformatics core at UC, sequence reads were aligned to the human reference genome (hg38) using the STAR aligner (RRID:SCR_004463)³⁵, and the reads aligned to each known gene were counted based on the latest GENCODE definitions of gene features.³⁶ The quality control of raw and aligned reads was performed using FastQC³⁷, RNA-SeQC³⁸, and summarized using MultQC³⁹ software. Differentially expressed genes were identified based on the FDR-adjusted p-values⁴⁰ obtained by fitting a generalized linear model based on negative-binomial distribution of read counts as implemented in the edgeR (RRID:SCR_012802) Bioconductor (RRID:SCR_006442) package.⁴¹ The functional characteristics of DEGs were studied by Gene Set Enrichment Analysis (GSEA)⁴² as implemented in the R package fgsea.⁴³ Gene sets analyzed included hallmark gene sets in MSigDb.⁴⁴ Significance was determined as p<0.1.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism 10.2.3 software (RRID:SCR_002798). Data was expressed as the mean of biological replicates with error bars represented as the standard error of the mean (+SEM). Comparisons of 2 groups were analyzed by an unpaired t-test with Welch’s correction. Comparisons of more than 2 groups were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons test. When appropriate, outliers were removed by Grubb’s test. A p-value less than 0.05 (p<0.05) was considered statistically significant for all experiments except RNA-seq where significance was defined as p<0.1.

Data Availability

The bulk RNA sequencing data can be found on the Gene Expression Omnibus (GEO) (RRID:SCR_005012) database: GSE306710. Raw data used to construct the figures in this study are available upon request to the corresponding author.

Results:

High Plk1 Expression is Associated with Reduced Survival in Patients with HNSCC

Overexpression of PLK1 has been associated with the progression of multiple cancers including HNSCC.^{8, 45} However, the role of PLK1 expression and activation has not been thoroughly investigated in HNSCC, neither HPV− nor HPV+. Analysis of The Cancer Genome Atlas (TCGA) revealed that high expression of Plk1 is associated with worse survival in all patients with HNSCC including both HPV− and HPV+ (Fig 1A). When delineating between patients with HPV− and HPV+ HNSCC with high and low Plk1 expression, patients with HPV− HNSCC with high Plk1 expression experience the worst survival outcomes among all patients with HNSCC (Fig 1B). Notably, amongst patients with HPV− HNSCC, Plk1-low patients have significantly higher survival than Plk1-high patients. In patients with HPV+ HNSCC, survival does not differ by Plk1 levels. Total Plk1 mRNA expression is slightly higher in patients with HPV+ HNSCC compared to patients with HPV− HNSCC (Fig 1C). On average, PLK1 protein expression is also slightly higher in HPV+ HNSCC cell lines compared to HPV− but not significantly (Fig 1D). Consistent with published data in other cancer types, it is important to note that PLK1 expression is also higher in all HNSCC cell lines compared to the normal oral keratinocyte (NOKs) cell line (Fig 1D).⁴⁵ The higher expression of Plk1 in HNSCC relative to normal tissue was also confirmed using the TCGA and Clinical Proteomic Tumor Analysis Consortium (CPTAC) databases. Moreover, in matched samples from the same patient, HNSCC tissue demonstrated higher Plk1 mRNA expression than normal tissue (Supp Fig 1A–B), and PLK1 protein expression (Supp Fig 1C) was significantly higher in HNSCC patient tissue compared to normal tissue. Plk1 was also assessed as a continuous and multivariate variable in HNSCC patients (Supp Fig 1D–E) and higher Plk1 expression was associated with reduced overall survival. These findings support that PLK1 is overexpressed in patients with HNSCC and HNSCC cell lines. Differences in survival and total Plk1 expression in patients with HPV− and HPV+ HNSCC indicate a potential difference in PLK1 signaling in these two disease subtypes, with Fig 1B suggesting that PLK1 plays a more important role in the progression of HPV− HNSCC.

Figure 1. High Plk1 expression is associated with reduced survival in patients with HNSCC.

(A) In patients with HNSCC (HPV+ and HPV−), overall survival is reduced in patients with high Plk1 mRNA expression (high Plk1 n = 182, low Plk1 n = 317) using data from The Cancer Genome Atlas (TCGA) database, cBioPortal, and Protein Atlas. (B) When stratified by Plk1 mRNA expression levels and HPV status, TCGA data demonstrates a significant reduction in overall survival only in patients with HPV− HNSCC (HPV+ high Plk1 n = 37, HPV+ low Plk1 n = 32, HPV− high Plk1 n = 145, HPV− low Plk1 n = 280) (Kaplan Meier Survival curves, significance when p<0.05). (C) Plk1 mRNA expression is slightly higher in patients with HPV+ (n = 72) vs HPV− (n = 415) HNSCC (t-test, ** p<0.01) (D) Immunoblot and quantification of total PLK1 expression in representative HPV− and HPV+ HNSCC cell lines notes no significant difference in PLK1 protein levels. Normal oral keratinocytes (NOKs) are used as a normal cell line control. GAPDH was used as a loading control (n = 3, t-test).

PLK1 inhibition with onvansertib inhibits PLK1 phosphorylation and reduces cell viability of HPV− HNSCC cells

Onvansertib is a highly selective PLK1 inhibitor with safety data in humans.²⁵ Yet, onvansertib has not previously been investigated as a targeted therapy for patients with HNSCC. As it is known that onvansertib inhibits PLK1 kinase activity, we initially confirmed that onvansertib inhibits PLK1 phosphorylation in HPV− and HPV+ HNSCC cells.⁴⁶ Two HPV− (HN5, Cal27) and two HPV+ (UMSCC47, UDSCC2) HNSCC human cell lines with relatively similar doubling times were used to increase rigor. Treatment with onvansertib reduced PLK1 phosphorylation in both HPV− and HPV+ HNSCC cell lines at early timepoints (Fig 2A). As PLK1 regulates cell cycle progression, inhibition of PLK1 stalls cells at the G2/M checkpoint to induce cell death.^{47, 48} Interestingly, we found that PLK1 inhibition with onvansertib significantly reduced cell viability of HPV− HNSCC cells more than HPV+ HNSCC cells (Fig 2B–C). Using a 3D model, to better represent the tumor microenvironment, onvansertib also reduced growth of HPV− HNSCC spheroids more than HPV+ (Fig 2D). We also assessed the effect of onvansertib on normal cell growth by treating NOKs. Onvansertib reduced cell viability of the NOKs but required a higher drug concentration than was needed in our HPV− HNSCC cell lines (Supp Fig 1F). The increased sensitivity of HPV− HNSCC cells to PLK1 inhibition indicates a potential difference in downstream PLK1 signaling between HPV− and HPV+ HNSCC cells. These findings also implicate PLK1 inhibition using onvansertib as a promising therapy for the more treatment resistant HPV− HNSCC patient population.

Figure 2. PLK1 inhibition reduces HPV− HNSCC cell viability.

(A) PLK1 phosphorylation (pPLK1) was reduced in HNSCC cells after onvansertib treatment (100 nM; 0, 2, 4 hrs). Total PLK1 expression was assessed and β-actin was used as a loading control (n = 2). (B) HNSCC cells were treated with onvansertib for 48 hrs and cell viability was assessed by measuring acid phosphatase activity (SEM, n = 3). (C) Onvansertib significantly reduced HPV− HNSCC cell viability compared to HPV+ when data points were combined from (B) (t-test, *p<0.05, ***p<0.001, ****p<0.0001). (D) Representative images (10X) of HPV− (HN5, Cal27) and HPV+ (UMSCC47, UDSCC2) spheroids after 7 days of growth with noticeably smaller and deformed HPV− HNSCC spheroids after onvansertib (25nM) treatment (n = 2, scale bars represent 100 μm).

PLK1 Inhibition Combined with Radiation Enhances Cell Cycle Stalling and Reduces Tumor Growth of HPV− HNSCC

While therapies targeting PLK1 have shown promise in both pre-clinical and clinical studies, few studies have evaluated combining PLK1 inhibition with the clinically relevant treatment of ionizing radiation.^{8, 46} There are no current clinical trials combining onvansertib with radiation in any cancer, including HNSCC. Given the role of PLK1 in cell cycle progression and G2/M phase arrest, we sought to rationally combine onvansertib with radiation in HPV− and HPV+ HNSCC, in vitro and in vivo. We first performed colony formation assays to determine the LD50 dose for radiation of our HPV− and HPV+ HNSCC cell lines. The LD50 dose for our HPV− HNSCC cell lines was ~4Gy and for our HPV+ HNSCC cell lines was ~2Gy (Supp Fig 2A). These results are consistent with previously published literature and clinical responses noting HPV+ HNSCC is more sensitive to radiation.^{49, 50} These LD50 doses were used to inform experiments throughout this study. Colony formation assays revealed that the combination of onvansertib with radiation significantly reduced the ability of HPV− HNSCC cells to form colonies compared to either single agent alone but not to the same extent in HPV+ cells. (Fig 3A, Supp Fig 2B). Additional colony formation assays also confirmed the combination of onvansertib with radiation at 4Gy was an optimal dose to observe an additive effect with combination treatment in HPV− HNSCC cells (Supp Fig 2C). We next investigated G2/M cell cycle stalling with the combination of onvansertib and radiation as both treatments are known to induce cell cycle blockade.^{10, 51, 52} Again, we found that the combination of onvansertib and radiation induced G2/M stalling more than either single agent in three HPV− HNSCC cell lines, but the combination effect was less pronounced in our HPV+ HNSCC cells (Fig 3B, Supp Fig 2D).

Figure 3. The combination of PLK1 inhibition and radiation reduces colony formation, increases G2/M stalling, and reduces tumor growth in HPV− HNSCC models.

(A) Colony formation was reduced in HPV− HNSCC cells treated with onvansertib (5 nM) and radiation (2, 4 Gy) compared to single treatments (n = 3, one-way ANOVA, ***p<0.001, ****p<0.0001). (B) G2/M stalling was increased after onvansertib (50 nM, 4hrs) and radiation (2, 4 Gy, 16hrs) treatment in HPV− HNSCC cells (n = 3, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 (significance of G2/M stalling compared to onvansertib alone or radiation alone is represented)). (C) Schematic depicting the heterotopic flank xenograft model used to assess Cal27 and UMSCC47 tumor growth in vivo. Mice were treated with onvansertib (60 mg/kg) 6 days a week. Tumors were irradiated at 2 Gy for 5 consecutive days (Cal27, total 10 Gy) or every other day (UMSCC47, total 6 Gy) (Created in BioRender. Korns, J. (2025) https://BioRender.com/m4atr3d). (D) Cal27 tumor volume ((vehicle n = 7, onv n = 6, Gy n = 9, onv + Gy n = 7); one-way ANOVA, *p<0.05, **p<0.01 (veh vs. onv + gy)) and tumor weight (one-way ANOVA, *p<0.05) were reduced upon combination treatment compared to vehicle. (E) UMSCC47 tumor volume ((vehicle n = 8, onv n = 10, Gy n = 9, onv + Gy n = 9); one-way ANOVA, *p<0.05, **p<0.01 (veh vs. gy or veh vs. onv + gy)) and tumor weight (one-way ANOVA, *p<0.05, **p<0.01) showed no difference when radiation was added.

To increase robustness and investigate a more clinically relevant model, we treated HPV− and HPV+ HNSCC tumors with onvansertib and radiation in vivo. Briefly, we implanted either Cal27 (HPV−) or UMSCC47 (HPV+) HNSCC cells into the flanks of nude mice and allowed palpable tumors to form. We then started onvansertib treatment one day before focal irradiation of tumors began. We irradiated the tumors with fractionated 2Gy doses to a total of 10Gy (Cal27) and 6Gy (UMSCC47) and continued onvansertib treatment 6 days per week (Fig 3C). Similar to in vitro trends, the combination of onvansertib and radiation significantly reduced HPV− HNSCC tumor growth compared to vehicle. Yet, single agent onvansertib or radiation did not significantly reduce tumor growth. (Fig 3D, Supp Fig 3A). The combination of onvansertib and radiation also significantly reduced HPV− HNSCC tumor weight (Fig 3D). Consistent with in vitro data, in the HPV+ tumor model, onvansertib did not reduce tumor volume or tumor weight alone or in combination with radiation (Fig 3E, Supp Fig 3A).⁵³ Mouse weight remained stable in all experiments indicating that onvansertib and radiation treatments were well tolerated (Supp Fig 3B). Ki67 staining to evaluate proliferation was also reduced by onvansertib in HPV− tumors but not HPV+ tumors (Supp Fig 3C). These findings, both in vitro and in vivo, provide strong rationale for combining PLK1 inhibition, using onvansertib, with radiation in patients with HPV− HNSCC to improve radiation treatment efficacy. To better understand the effect of the combination therapy in HPV− HNSCC cells, we next explored potentially disparate downstream targets of PLK1 to identify a difference in PLK1 signaling between HPV− and HPV+ HNSCC.

PLK1 Inhibition Reduces MMP10 Expression and Activity in HPV− HNSCC cells

We performed bulk RNA-seq analysis of HN5 (HPV−) and UMSCC47 (HPV+) cells treated with vehicle, onvansertib alone, radiation alone, or the combination of both. We first confirmed that onvansertib and radiation treatments were targeting known pathways in our HNSCC cells. As expected, onvansertib increased G2/M checkpoint signaling while radiation induced inflammation and apoptosis signaling in HN5 and UMSCC47 cells (Supp Fig 4A–B). We next assessed genes that were upregulated after radiation that could drive radioresistance in patients with HNSCC. We also wanted to determine if any of these same genes were downregulated by onvansertib. One candidate gene differentially expressed was matrix metalloprotease 10 (MMP10) (Fig 4A). Notably, when onvansertib was combined with radiation in HN5 cells, MMP10 was significantly reduced compared to radiation alone (Fig 4B) in bulk RNA-seq analysis. However, UMSCC47 cells exhibited much lower levels of MMP10 compared to HN5 cells, and onvansertib did not reduce MMP10 expression in these HPV+ HNSCC cells (Supp Fig 4C).

Figure 4. PLK1 inhibition reduces MMP10 expression and activation when combined with radiation in HPV− HNSCC cells.

(A) Volcano plots of RNA-seq data identifying MMP10 as a significantly downregulated gene after onvansertib treatment and upregulated gene after radiation in HN5 cells. (B) MMP10 mRNA expression is downregulated with onvansertib in HN5 cells shown by RNA-seq analysis. HNSCC cells were treated with onvansertib (25nM, 24hr) then irradiated (4Gy) and RNA was collected 24hrs later (n = 3, ****p<0.001). (C-E) HN5 and Cal27 cells were treated with onvansertib (100nM) for 4hrs then irradiated (8Gy) and samples were collected 20hrs later. (C) RT-qPCR (n > 3, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001), (D) Immunoblotting for total MMP10 expression (n = 2), and (E) Zymography (n = 3) all show decreased MMP10 activity upon onvansertib treatment compared to radiation alone in HPV− HNSCC cells. (F) HN5 cells were treated with Plk1 siRNA for 24hrs which reduced Plk1 and MMP10 mRNA levels assessed by RT-qPCR (n = 3, one-way ANOVA, ***p<0.001, ****p<0.0001). (G) MMP10 expression was assessed by IHC in patients with HNSCC indicating higher MMP10 expression in higher Grade (1-2: n = 27, 3: n = 4, t-test, *p<0.05) and T stage (1-2: n = 4, 3-4: n = 27, t-test) tumor tissues.

To validate our RNA-seq findings, we performed RT-qPCR to assess MMP10 mRNA expression in HPV− and HPV+ HNSCC cells after onvansertib and radiation treatment. Similar to RNA-seq analyses, onvansertib significantly reduced MMP10 expression in HPV− HNSCC cells but not HPV+. The combination of onvansertib with radiation also significantly reduced MMP10 expression compared to radiation alone in HPV− HNSCC cells (Fig 4C, Supp Fig 4D). We saw similar trends in protein expression in that onvansertib reduced MMP10 expression in HPV− HNSCC cells but not HPV+ (Fig 4D, Supp Fig 4E). To specifically assess MMP10 activity after onvansertib and radiation treatment we performed zymography. Zymography demonstrates the ability of MMPs to cleave specific substrates in a polyacrylamide gel indicating its activity. Importantly, we found that onvansertib alone reduced MMP10 activity compared to vehicle, and the addition of onvansertib reduced MMP10 activity when combined with radiation in HPV− HNSCC cells. When MMP10 is knocked out, we see no MMP10 activity (Fig 4E). To increase rigor, we further examined MMP10 activity in HPV− and HPV+ HNSCC cells after onvansertib and radiation treatment using a fluorescent kit. After treatment with onvansertib alone, MMP10 activity was significantly reduced compared to vehicle in HPV− HNSCC cells but not HPV+. We also saw a reduction in MMP10 activity when onvansertib was combined with radiation in HPV− HNSCC cells but not significant when compared to radiation alone. In HPV+ HNSCC cells, radiation enhanced MMP10 activity even when combined with onvansertib (Supp Fig 4F). To evaluate the effect of reduced Plk1 on MMP10 expression, HPV− HN5 cells were treated with siRNA against Plk1. In cells with reduced Plk1, MMP10 expression was also significantly reduced (Fig 4F). This indicates Plk1 specifically regulates MMP10 expression independent of the drug effect we see with onvansertib. Overall, inhibiting PLK1 with onvansertib reduces MMP10 expression and activity in HPV−, but not HPV+, HNSCC cells indicating a potential difference in MMP10 signaling in these two HNSCC subtypes.

MMP10 is known to be overexpressed in multiple cancers including HNSCC. We assessed MMP10 expression by immunohistochemistry in pre-treatment biopsy specimens of patients with HPV− HNSCC treated at our institution. MMP10 expression was higher in patients with grade 3 tumors and more advanced T stage at diagnosis (Fig 4G). This indicates that MMP10 is associated with HPV− HNSCC tumors that are more aggressive and generally less responsive to treatment. To increase rigor, with a broader dataset, patient data from the TCGA and CPTAC databases was analyzed. MMP10 mRNA and protein expression were significantly higher in HNSCC specimens compared to normal tissue; MMP10 mRNA was also higher in tissue specimens from matched normal specimens acquired from the same patient (Supp Fig 5A–C). In addition, MMP10 expression was significantly higher in patients with HNSCC when compared to all other cancer types (Supp Fig 5D). Similar to our RNA-seq results, we also found that total MMP10 mRNA expression is significantly higher in patients with HPV− HNSCC compared to HPV+ (Supp Fig 5E). Importantly, high MMP10 expression is also associated with worse survival in patients with HNSCC (Supp Fig 5F). Currently, there are no MMP10 specific targeted therapies available. Other MMP inhibitors that have been used in clinical trials have been unsuccessful due to dose-limiting toxicity.⁵⁴ These data combined posit MMP10 as a novel downstream effector of PLK1 that could be targeted with onvansertib to improve survival outcomes in patients with HPV− HNSCC.

PLK1 Inhibition Combined with Radiation Reduces Migration of HPV− HNSCC Cells

MMPs are known to contribute to the process of epithelial to mesenchymal transition (EMT), which ultimately leads to metastasis.⁵⁵ Patients with HPV− HNSCC typically experience worse clinical outcomes than those with HPV+ driven disease, presumably due to persistent cell migration and invasion despite aggressive treatment. It is also known that radiation induces the migration and invasion of HNSCC cells.⁵⁶ From our RNA-seq, we found that onvansertib reduces EMT signaling in HN5 cells (Supp Fig 5G). From these findings, we wanted to determine if combining PLK1 inhibition with radiation could reduce HPV− HNSCC migration, an MMP10 associated phenotype. To evaluate migration, we performed a scratch assay and found that the combination of onvansertib with radiation significantly reduced the ability of HPV− HNSCC cells to migrate into the scratched space (Fig 5A). Onvansertib combined with radiation also significantly reduced migration of HPV− HNSCC cells compared to vehicle and single agent treatments in a Transwell assay (Fig 5B). We also generated MMP10 Crispr-Cas9 knockout HPV− HNSCC cells to determine if loss of MMP10 mitigates the efficacy of onvansertib to reduce migration. We confirmed MMP10 knockout by RT-qPCR (Fig 5C). Using a Transwell migration assay, we found that loss of MMP10 in HN5 cells significantly reduced migration compared to wildtype cells. Importantly, PLK1 inhibition had minimal effect on migration in HN5 MMP10 knockout cells (Fig 5D). These findings suggest PLK1 inhibition combined with radiation reduces the migratory phenotype of HPV− HNSCC cells at least partially through MMP10.

Figure 5. PLK1 inhibition combined with radiation reduces migration of HPV− HNSCC cells.

(A) HN5 and Cal27 cells treated with onvansertib (25nM) and radiation migrated significantly slower than single agents into the scratched space. Quantification was normalized to 0 hr scratch widths. Representative images (4X) and quantification of one replicate are shown (scale bar represents 500 μm, n = 2, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). (B) HN5 and Cal27 cells were treated with onvansertib (50nM) for 24 hrs then irradiated (4 Gy). Transwell inserts were fixed 24 hrs later. Combination treatment reduced migration of HPV− HNSCC cells (n = 3, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001). (C) MMP10 mRNA expression (RT-qPCR) was reduced in HN5 Crispr-Cas9 MMP10 knockout cells (MMP10 KO #1, #2) compared to HN5 wildtype (WT) cells (n = 3, one-way ANOVA, ****p<0.0001). (D) HN5 WT and MMP10 KO cells were treated with onvansertib (50nM) for 48 hrs. Onvansertib did not reduce migration in MMP10 KO cells (n = 5, two-way ANOVA, *p<0.05, ***p<0.001).

Discussion

HNSCC, particularly HPV− HNSCC, is a prevalent disease that has shown minimal improvement in survival outcomes over the last several decades. Treatment algorithms have remained largely unchanged, with radiation therapy and concurrent cisplatin chemotherapy forming the backbone of treatment. Perhaps more concerning, many patients with HNSCC are medically ineligible for platinum-based chemotherapies, due to high rates of ototoxicity and nephrotoxicity.⁵⁷ Therefore, there is a need not only to identify more robust concurrent regimens to cisplatin, but also to identify potential alternative approaches to enhance the efficacy of definitive radiation.

Identifying druggable targets in HNSCC with targeted therapies that are highly specific and with limited toxicities is critical. In addition, most patients with HNSCC will receive radiation therapy in combination with these targeted therapies, so it is important to find drugs that enhance the cell death response when combined with radiation. Our studies have identified that targeting PLK1 with onvansertib enhances radiation efficacy to reduce HPV− HNSCC tumor growth. Onvansertib is a highly specific small molecule PLK1 inhibitor exhibiting thousand-fold specificity for PLK1 over other PLK family members. Onvansertib has also been well tolerated as an oral therapy; even in combination with FOLFIRI/bevacizumab chemotherapy in a Phase 1b clinical trial in colorectal cancer patients with prior oxaliplatin exposure only 15% of patients experienced Grade 3 and 4 adverse events.²⁵

Previously, other PLK1 inhibitors have been shown to induce similar anti-mitotic events as onvansertib. Yet, specificity in inhibiting PLK1 alone was lacking, presumably driving toxicities for patients in clinical trials.⁵⁸ Onvansertib was also efficacious in overcoming cisplatin resistance in a pre-clinical lung adenocarcinoma study.^{25, 59} Only one study has indicated that onvansertib could decrease cell viability of HNSCC cells that were resistant to radiation or cisplatin. Though this study was limited to only one HPV− HNSCC cell line, which was then made radioresistant, this study does provide corroborating evidence from another cell line in support of our studies. Moreover, this group did not evaluate a mechanism for how PLK1 inhibition was inducing cell death.⁴⁵ However, these studies also warrant further investigation into the combination of onvansertib and cisplatin as another potential avenue to treat patients with HNSCC. Our studies have minor limitations as the tools available to investigate MMP10 are limited. The MMP10 activity kit used to corroborate our zymography results is non-specific as MMPs other than MMP10 can induce a fluorescent signal. However, our studies may enhance the translation of onvansertib to the clinic for patients with HNSCC as we have used multiple HPV− and HPV+ HNSCC cell lines to determine if PLK1 inhibition induces the same response in each disease subtype. To enhance our pre-clinical studies, we focally irradiated the HPV− and HPV+ flank tumors with fractionated doses in combination with onvansertib which is reflective of daily dosing of patients in the clinic. By one approach to evaluate synergy, the combination of onvansertib and radiation reduced tumor growth in an additive manner in the HPV− model.⁵³ We also provide substantial data in support of MMP10 serving as a novel downstream effector of PLK1 in HPV− HNSCC cells.

In addition to its primary role in cell cycle progression, PLK1 has been identified as a driver of EMT processes including metastasis.⁶⁰ PLK1 has also been associated with regulating MMP2 and MMP9 expression in HNSCC.²⁴ MMP10 has been shown to be associated with nodal metastasis in HNSCC.⁶⁰ Yet, no connection has been made between PLK1 and MMP10 in cancer. Here we show that PLK1 inhibition with onvansertib reduced MMP10 expression and activity in HPV− HNSCC cells. Plk1 knockdown by siRNA reduced MMP10 expression in HPV− HNSCC cells. When onvansertib was combined with radiation, MMP10 expression and migration were both reduced, reversing an MMP-driven phenotype. These data suggest that PLK1 regulates MMP10 expression and activity and that PLK1 inhibition may reduce cell viability and tumor growth of HPV− HNSCC cells through MMP10, yet further studies are still warranted to solidify this signaling network (Fig 6).

Figure 6. Proposed schema:

PLK1 inhibition with onvansertib counters radiation mediated induction of MMP10. Therefore, combination treatment of onvansertib with radiation slows HPV− HNSCC progression. (Created in BioRender. Korns, J. (2025) https://BioRender.com/qpgm1w0)

Currently, patients with HPV− and HPV+ HNSCC are treated similarly in the clinic, though they respond differently to therapy. Our findings support continuing studies that investigate the mechanistic differences between HPV− and HPV+ HNSCC as we have identified PLK1 inhibition to be more efficacious in HPV− HNSCC models. Furthermore, there are no clinical trials investigating the use of onvansertib in HNSCC nor are there any clinical trials that combine onvansertib with radiation in any cancer, including HNSCC. Thus, our findings highlight the potential for onvansertib to be combined with radiation to reduce PLK1 and MMP10 signaling in Phase I clinical trials. We provide preclinical data in support of combination therapy to ultimately advance treatment paradigms and improve survival outcomes for patients with HPV− HNSCC.

Abbreviations List:

PLK1

Polo like kinase 1

HPV

human papilloma virus

HNSCC

head and neck squamous cell carcinoma

MMP10

matrix metalloprotease 10