This article is mentioned in:

Introduction

Head and neck squamous cell carcinoma (HNSCC) is the sixth most prevalent type of cancer worldwide with 890,000 new cases and 450,000 deaths reported in 2018 (1). Based on etiological factors, HNSCC is classified into two types of disease: HPV-positive or HPV-negative. HPV-positive HNSCC occurs mainly in oropharyngeal tissues, whereas HPV-negative HNSCC is primarily found in the oral cavity and larynx (2). The occurrence of HPV-negative HNSCC is associated with the use of tobacco and excessive consumption of alcohol, and is characterized by mutations in diverse oncogenic driver genes. By contrast, HPV-positive HNSCC is related to human papillomavirus (HPV) infection, whereby viral proteins E6 and E7 cause oncogenic transformation of host cells (3–6). Compared with HPV-negative HNSCC, HPV-positive HNSCC expresses viral proteins and other neoantigens, as well as mutagenesis caused by viral infection, which may be advantageous to immunotherapy (2). In addition, HPV-positive HNSCC has better clinical outcomes with standard therapy in comparison to HPV-negative HNSCC (7). Reports have indicated that patients with HPV-positive HNSCC have a higher 5-year survival rate (~80%) than HPV-negative patients (~50%) (8); therefore, there is an urgent need for new effective therapeutics to treat HPV-negative HNSCC (9,10).

Classical therapies for patients with HNSCC without distant metastasis typically include surgical resection, radiation therapy, chemotherapy or a combination of these regimens. The specific therapeutic approach depends on various factors, such as pre-existing clinical conditions, cancer location and the Tumor-Node-Metastasis stage of the tumor (4,11). The combination of these treatments could reduce the rate of recurrence and distant metastasis for patients with metastatic disease (12). However, chemotherapy remains the primary option for patients with recurrent and distant metastatic HNSCC (13). Cisplatin is the most commonly used anticancer drug to treat advanced HNSCC; however, while a number of newly diagnosed patients with advanced HNSCC initially respond well to cisplatin-based chemotherapies, most patients either have intrinsic resistance or will eventually develop acquired resistance to cisplatin, which often leads to death within 1 year (14). Immunotherapy has been introduced to treat refractory HNSCC, but its impact has been limited (14,15). Therefore, it remains of the utmost importance to identify new therapeutic alternatives to improve the efficacy of cisplatin-based therapies, or to replace them, for the treatment of patients with advanced HNSCC.

Epidermal growth factor receptor (EGFR), a member of the ErbB kinase family, is reported to be upregulated in 90% of HNSCC cases and cell lines, according to a study that included 24 patients with SCCHN and 10 cell lines, as opposed to seven healthy control individuals (16) and serves a crucial role in the pathogenesis and clinical course of cancer (13,17,18). EGFR controls the activation of several essential pathways, such as PI3K/Akt/mTOR and RAS-RAF-MEK-ERK, which regulate cell proliferation, survival and migration (19,20). In 2006, the monoclonal EGFR antibody cetuximab was approved by the Food and Drug Administration (FDA) to treat HNSCC in combination with the standard therapy (21–23). However, the use of cetuximab has been reported to result in very limited improvement in survival rates for patients undergoing cisplatin-based therapy (24). In addition, small molecule kinase inhibitors, such as gefitinib and erlotinib, while effective in targeted therapies for non-small cell lung cancer, have not demonstrated any benefits for patients with HNSCC (25,26). Increasing evidence has demonstrated the importance of the ErbB family, which contains EGFR, HER2, HER3 and HER4, in the carcinogenesis of HNSCC and its response to therapies (27). HER2 and HER3 form heterodimers with EGFR and play a role in PI3K/Akt activation. In addition, HER2 and HER3 have been shown to be associated with resistance to EGFR and PI3K inhibitors in cancer (28,29). These results indicated that targeting the ErbB family kinases could more effectively suppress HNSCC compared with using EGFR inhibitors alone (25,30). Notably, the FDA-approved ErbB family inhibitor afatinib has shown positive results in HNSCC clinical trials and is now listed in the National Comprehensive Cancer Network (NCCN) guidelines as a third-line single agent for HNSCC treatment (25,26,31–33). Understanding the mechanisms behind resistance to afatinib and exploring methods to avoid that resistance would be beneficial.

PI3K is one of the most important downstream effectors of the EGFR/ErbB receptor family. The genes PIK3CA, PIK3CB and PIK3CD encode three highly homologous catalytic isoforms of class IA PI3K, p110α, p110β and p110δ, respectively. These isoforms associate with any of five regulatory isoforms: p85α, its splicing variants p55α and p50α, p85β and p55γ (34). The most important PI3K-p85 complex is PI3Kα/p85α. The PI3K pathway has been reported to be the most frequently mutated oncogenic pathway, detected in 30.5% of tumors, according to whole-exome sequencing of the genes in the JAK/STAT, MAPK and PI3K pathways in 151 HNSCC tumors (35). Previous studies have also shown that PIK3CA, the gene that codes for PI3K p110α, is one of the most frequently mutated genes in HNSCC (35–37), and PIK3CA amplification and overexpression have also been identified in HNSCC (38). According to The Cancer Genome Atlas (TCGA), profiling of 279 HNSCC tumors identified amplification or mutation of PIK3CA in 34% of HPV-negative HNSCC tumors and 56% of HPV-positive tumors (39,40). PI3K activation in turn activates Akt, which then phosphorylates its substrates, such as TSC2, PRAS40, GSK3β and FOXO, to regulate multiple cellular functions that consequently control cell proliferation, survival and response to therapy. The tumor suppressor gene PTEN encodes the PTEN protein that dephosphorylates PIP3 to inhibit the PI3K pathway. Mutations that result in PTEN loss or deceased PTEN expression have been frequently observed in HPV-positive and negative HNSCC (41,42). These alterations further result in PI3K/Akt activation (43). Therefore, PI3K is an attractive target for the treatment of HNSCC (40,44,45).

Our previous study reported that co-targeting the ErbB family and PI3K, through a combination of afatnib and copanlisib, suppressed the growth of HPV-positive HNSCC (46). The present study further explored whether this combination was also effective at suppressing the growth of HPV-negative HNSCC. The present findings indicated that the combination of afatnib and copanlisib may be a promising treatment option for HPV-negative HNSCC.

Materials and methods

Cell culture

HNSCC cell lines, Cal27 and FaDu, were purchased from American Type Culture Collection, authenticated by short tandem repeat analysis and tested for Mycoplasma contamination at the Translational Core Facility at Marlene and Stewart Greenebaum Comprehensive Cancer Center, University of Maryland (Baltimore, MD, USA). All cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM glutamine, and 100 U/ml penicillin and streptomycin (all from Gibco; Thermo Fisher Scientific) at 37°C in an incubator containing 5% CO2 for 24 h before each experiment was performed.

Antibodies and inhibitors

The following antibodies were purchased from Cell Signaling Technology, Inc.: Phosphorylated (P)-Akt-S473 (cat. no. 4058), P-Akt-T308 (cat. no. 9275), Akt (cat. no. 2938), P-HER2-Y1248 (cat. no. 2247), HER2 (cat. no. 4290), P-HER3-Y1289 (cat. no. 2842), HER3 (cat. no. 12708), caspase-3 (cat. no. 9665), cleaved-caspase-3 (cat. no. 9664) and β-actin (cat. no. 4967). HRP-labeled anti-mouse (cat. no. W4028) and anti-rabbit (cat. no. W4018) secondary antibodies were purchased from Promega Corporation. Gefitinib, erlotinib and afatinib, as well as all PI3K inhibitors (copanlisib, BKM120, GDC-0941, BYL179, AZD8186 and CAL-101) were purchased from Selleck Chemicals. Copanlisib was dissolved in 10% trifluoroacetic acid (TFA), while all other drugs were dissolved in 100% DMSO. The final concentrations of TFA and DMSO in cell culture media were 0.01% and 0.1%. For single EGFR/ErbB inhibitor treatment, cells were treated with DMSO or increasing concentrations of gefitinib (3 nM-30 µM), erlotinib (3 nM-30 µM) and afatinib (1 nM-10 µM) at 37°C for 96 h. For single PI3K inhibitor treatment, cells were treated with TFA, DMSO or increasing concentrations of copanlisib (0.01 nM-10 µM), GDC-0941 (1 nM-10 nM) and other PI3K inhibitors (3 nM-30 µM) at 37°C for 96 h. For the combination treatment, Cal27 cells were treated with the combination of various concentrations of copanlisib (2.875–92 nM) and afatinib (0.2875–9.2 nM), and FaDu cells were treated with the combination of various concentrations of copanlisib (3.375–118 nM) and afatinib (0.3125–11.6 nM) at 37°C for 96 h.

Cell lysis and western blot analysis

Cell lysis and western blot analysis were performed as previously described (46,47). Briefly, cells grown on 100-mm dishes were rinsed twice with 1X cold PBS, then lysed on ice for 30 min in 1 ml lysis buffer (M-PER™ Mammalian Protein Extraction Reagent; cat. no. 78503; Pierce; Thermo Fisher Scientific, Inc.) with phosphatase inhibitor and EDTA-free protease inhibitors (Roche Diagnostics). After centrifugation at 13,000 × g for 20 min, lysates that contained 30 µg protein were separated by SDS-PAGE on 4–12% gels, and the proteins were transferred to Pure Nitrocellulose Membranes (Bio-Rad Laboratories, Inc.), which were blocked in 5% nonfat milk at room temperature for 1 h. The membranes were then incubated with the indicated primary antibodies (1:1,000 dilutions) overnight at 4°C, and with HRP-conjugated secondary antibodies (1:4,000 dilutions) at room temperature for 1 h. Proteins were finally visualized by using Clarity™ Western ECL Substrate (cat. no. 1705060; Bio-Rad Laboratories, Inc.).

Analyzing apoptosis by Annexin V/propidium iodide (PI) staining

Apoptosis analysis was performed using the FITC Annexin V Apoptosis Detection Kit II (cat. no. 556570; BD Biosciences) combined with PI (cat. no. P1304MP; Invitrogen; Thermo Fisher Scientific, Inc.) staining as previously described (46,47). Cells were plated into 6-well plates at a density of 4.5×105/well overnight, before being treated with vehicle control, copanlisib (30 nM in Cal27 and 125 nM in FaDu), afatinib (0.5 µM in Cal27 and 1 µM in FaDu), or a combination of copanlisib and afatinib at 37°C for 48 h. Both floating and adherent cells were collected and stained with Annexin V and PI, according to the manufacturers' instructions. Samples and single stained controls were subsequently analyzed using the BD FACSCanto II flow cytometer (BD Biosciences). Data were analyzed using FCS Express 7 (Research Edition; De Novo Software). Cells stained solely with Annexin V (early apoptotic cells) or double stained with Annexin V and PI (late apoptotic cells) were considered apoptotic cells.

Cell viability assay

Cell viability was assessed using sulforhodamine B (SRB) staining as described previously (48). Briefly, cells were plated in 96-well plates at a density of 3,500 cells/well overnight before being subjected to treatment with a serial concentration of the indicated drugs or vehicle control for 96 h at 37°C. Following an overnight fixation in methanol, the cells were stained with SRB for 1 h and then dissolved with 10 mmol/l Tris base for 1 h while being shaken. The aforementioned procedures were all carried out at room temperature. Subsequently, the absorbance was read at 492 nm. The half maximal effective concentration (EC50) values were calculated using nonlinear regression analysis with variable slope in Graph-Pad Prism v8.0 (Dotmatics). The effect of the inhibitors on cell viability are represented as percentages compared with vehicle control-treated cells. Each experiment was performed in triplicate and repeated at least twice. The vehicle controls were treated with the reagents that were used to dissolve the chemicals (0.01% TFA for copanlisib and 0.1% DMSO for the other drugs). To determine synergy of the drug combination, the combination index (CI) values were determined according to the Chou-Talalay method (49) using CalcuSyn software (version 2.0; BIOSOFT).

Tumor xenograft formation in mice

In vivo experiments were conducted with the approval of the Institutional Animal Care and Use Committee (IACUC) of University of Maryland (IACUC #1021001) and within an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited vivarium. Female 6-week-old nude mice (n=45), purchased from Inotiv, Inc., were used for the animal studies. All mice were housed under the following conditions: Ambient temperature (21–22°C), regulated humidity (40–50%) ad libitum access to food and water, 12-h light/dark cycle. The animals were treated and handled in accordance with the guidelines set by the IACUC of the University of Maryland. A total of 45 nude mice received injection of 0.5×106 FaDu cells into the right flank. The mice were split into four groups (n=7/group) 13 days after cell injection, so that the mean tumor volume was similar (~150 mm3). Notably, 45 mice were injected but only 28 mice entered the study; mice were excluded from entering the study based on irregular shape of the tumors (difficult to measure). One mouse per group was excluded from analysis due to necrosis at termination of the study; mice were sacrificed as soon as necrosis was observed. The day of sorting occurred on the first day of dosing and was defined as Day 1. The four treatment groups (n=6/group at termination) were: Vehicle control, copanlisib (6 mg/kg, intraperitoneal, 5 times/week, Monday to Friday), afatinib (6 mg/kg, oral, 5 times/week, Monday to Friday), and a combination of copanlisib and afatinib. Tumor volume was measured twice per week using electronic calipers and the animals were weighed five times per week. Tumor volume was calculated as (L × W2)/2, where W is the smaller dimension and L is tumor length. At the end of the study (32 days), mice were euthanized by CO2 asphyxiation (30% vol/min) followed by cervical dislocation as noted in the IACUC-approved protocol, and the tumors were excised, weighed and cut in half. Half of the tumors were fixed in formalin (4%) at room temperature for 24 h and then transferred to 70% ethanol (half), whereas the other half were frozen at −80°C. Tumors were preserved by fixation and freezing for future evaluation. All mice were euthanized 32 days after the start of dosing. No animal who entered the study was found dead; however, as aforementioned, one mouse from each group was removed from the study and euthanized due to necrosis. The study was terminated based on tumor volume (~2,000 mm3) as listed in the IACUC-approved protocol as a humane endpoint. No analgesics were administered during the study since flank tumor studies are considered painless. Mice were briefly anesthetized with 2.5% isoflurane during the subcutaneous injection of cells.

Statistical analysis

In vitro data were repeated at least three times unless specifically described in figure legends. The data were presented as the mean ± SD and animal data are shown as the mean ± SEM. Statistical analysis was performed using GraphPad Prism version 10.3.1 (Dotmatics). One-way ANOVA was used to statistically analyze multiple groups, followed by Tukey's post hoc analysis. P<0.05 was considered to indicate a statistically significantly difference.

Results

HPV-negative HNSCC cell lines are sensitive to afatinib

The EC50 values of EGFR/ErbB family inhibitors, gefitinib, erlotinib and afatinib, were determined in two HPV-negative HNSCC cell lines, Cal27 and FaDu. Both cell lines were relatively resistant to erlotinib and sensitive to gefitinib (Fig. 1A and B). However, they were revealed to be very sensitive to afatinib, although Cal27 cells were more sensitive to afatinib in comparison to FaDu cells. These results suggested that afatinib was the more effective small molecule inhibitor for the treatment of HPV-negative HNSCC in comparison to other EGFR inhibitors.

Figure 1. Afatinib more effectively inhibits the viability of human papillomavirus-negative head and neck squamous cell carcinoma cells compared with gefitinib and erlotinib. (A) Cal27 and (B) FaDu cells were treated with DMSO or increasing concentrations of gefitinib (3 nM-30 µM), erlotinib (3 nM-30 µM) and afatinib (1 nM-10 µM) at 37°C for 96 h. Cell viability was measured using the sulforhodamine B cell viability assay and EC50 values were determined using GraphPad Prism version 10.3.1. The experiments were performed in triplicate and repeated at least three times. EC50, half maximal effective concentration.

Copanlisib is the most effective PI3K inhibitor to suppress HPV-negative HNSCC viability

To identify the most effective PI3K inhibitor, the EC50 values of six PI3K inhibitors were determined, including three pan-PI3K inhibitors, copanlisib, BKM120 and GDC-09410; the PI3Kα inhibitor Byl719; the PI3Kβ inhibitor AZD8186; and the PI3Kδ inhibitor Cal-101 (Fig. 2). Cal27 cells were strongly resistant to the PI3Kδ inhibitor, and relatively resistant to PI3Kα and PI3Kβ inhibitors. However, they were much more sensitive to the pan-PI3K inhibitors copanlisib, GDC-0941 and BKM120. Moreover, the EC50 value of copanlisib was much lower in comparison to those of BKM120 and GDC-0941. Similar results were found in FaDu cells (Fig. S1). These results suggested that copanlisib may be the most effective small molecule PI3K inhibitor to treat HPV-negative HNSCC.

Figure 2. Copanlisib more effectively inhibits cell viability compared with other PI3K inhibitors in Cal27 cells. Cal27 cells were treated with DMSO or increasing concentrations of copanlisib (0.01 nM-10 µM) and other PI3K inhibitors (0.01 nM-100 µM) at 37°C for 96 h. Cell viability was measured using the sulforhodamine B assay and the viability curves are shown. The experiments were performed in triplicate and repeated at least three times. The associated EC50 values of the inhibitors were determined using GraphPad Prism version 10.3.1.

Synergistic inhibition of cell viability using a combination of afatinib and copanlisib

The present study aimed to assess whether simultaneous inhibition of ErbB family and PI3K pathways could more effectively inhibit HPV-negative HNSCC cell viability. Based on the aforementioned data, afatinib and copanlisib were selected for the combination therapy. Similar to the results shown in Figs. 1 and 2, afatinib or copanlisib alone inhibited cell viability; however, the combination caused enhanced inhibition of viability in both Cal27 (Fig. 3A) and FaDu cells (Fig. 3B). Furthermore, the related CI value for each combination was calculated according to the Chou-Talalay method (49) using CalcuSyn software. The CI values for all combinations were <1.0, which indicated a synergistic effect in the combination of afatinib and copanlisib (Fig. 3C and D). These data indicated that afatinib and copanlisib may synergistically inhibit HPV-negative HNSCC cell viability.

Figure 3. Synergistic inhibition of cell viability by the combination of afatinib and copanlisib in vitro. (A) Cal27 and (B) FaDu cells were treated with two-fold diluted concentrations of afatinib, copanlisib or their combination at 37°C for 96 h with the half maximal effective concentration values in the middle. Cell viability was measured using the sulforhodamine B cell viability assay. The experiments were performed in triplicate and repeated twice due to the consistent results each time. (C) CI values of the combination of various concentrations of copanlisib (2.875–92 nM) and afatinib (0.2875–9.2 nM) in Cal27 cells were determined using CalcuSyn version 2.0. (D) CI values of the combination of various concentrations of copanlisib (3.375–118 nM) and afatinib (0.3125–11.6 nM) in FaDu cells were determined using CalcuSyn version 2.0. CI, combination index.

Synergistic inhibition of xenograft tumor growth via the combination of afatinib and copanlisib in mice

The present study also evaluated the antitumor activity of the afatinib and copanlisib combination in vivo using a mouse xenograft model. FaDu cells were inoculated into the mice, and when the tumors reached ~200 mm3, the mice were randomly split into four groups, which were treated with a vehicle control, copanlisib, afatinib or their combination. The original aim was to treat the mice for more than 6–8 weeks; however, the experiment was terminated on day 32 when tumor necrosis was observed in the majority of the mice in the control group. The average tumor volume in groups treated with either copanlisib or afatinib was lower than that of the control group, but there were no significant differences in tumor volume among these three groups, which may be due to tumor necrosis-induced decreases in tumor volumes in the control group (Fig. 4A). However, tumor volumes in the combination treatment group were significantly lower than in the single treatment groups, and they were more significantly lower than those in the control group by the end of the study.

Figure 4. Inhibition of head and neck squamous cell carcinoma growth by a combination of afatinib and copanlisib in vivo. FaDu cells were inoculated into mice. When tumors reached ~150 mm3, mice were randomized to one of four treatment groups (n=6/group): Vehicle control, copanlisib (6 mg/kg, intraperitoneal, 5 times/week Monday to Friday), afatinib (6 mg/kg, oral, 5 times/week, Monday to Friday), or a combination of copanlisib and afatinib. The treatments were performed for 32 days. (A) Xenograft tumor volumes and images, (B) final average weights of tumors, and (C) average body weight changes of mice in each group were compared by GraphPad Prism 10.3.1 software. After one-way ANOVA, Tukey's post hoc test was performed for multiple comparisons. ***P<0.001; **P<0.01; *P<0.05.

Similarly, after the tumors were excised and weighed at the end of the study, there was no significant difference in tumor weight (mg) between the groups treated with either copanlisib or afatinib (Fig. 4B). In addition, there were no significant differences in tumor weight (mg) between the control group and the single agent treatment groups due to individual differences. However, tumor weight in the combination treatment group was significantly lower compared with that in the control agent treated group. In summary, while afatinib or copanlisib alone had modest inhibitory effects on tumor growth, they did not reach statistical significance, whereas the combination of the two drugs significantly inhibited tumor growth.

Notably, all doses of copanlisib and afatinib in single and combination treatments were well-tolerated, as indicated by no significant weight loss observed during the study (Fig. 4C). These results indicated the feasibility of treatment with a combination of afatinib and copanlisib to suppress HPV-negative HNSCC.

A combination of afatinib and copanlisib induces apoptosis

The present study assessed whether a combination of afatinib and copanlisib could induce more apoptosis compared with either single treatment. Cal27 cells were treated with afatinib (0.5 µM), copanlisib (30 nM) or their combination for 48 h before an apoptosis assay was performed. Afatinib, but not copanlisib alone, induced modest cell apoptosis, whereas their combination significantly increased cell apoptosis (Figs. 5A and S2). Since FaDu cells exhibited higher EC50 to afatinib and copanlisib in comparison to Cal27 cells (Figs. 1, 2 and S1), higher concentrations of afatinib and copanlisib were chosen for the treatments. Treatment with copanlisib (125 nM) significantly enhanced apoptosis, whereas treatment with afatinib (1.0 µM) did not significantly increase apoptosis. However, a combination of copanlisib (125 nM) and afatinib (1.0 µM) led to significantly increased apoptosis compared with either single treatment (Figs. 5B and S3). These data indicated that afatinib and copanlisib may cooperate to induce apoptosis in HPV-negative HNSCC.

Figure 5. A combination of afatinib and copanlisib increases cell apoptosis compared with afatinib or copanlisib treatment alone. (A) Cal27 and (B) FaDu cells were treated with vehicle control, copanlisib, afatinib or a combination for 48 h, and cell apoptosis was analyzed by Annexin V/propidium iodide staining. The experiments were performed in triplicate and repeated at least three times. Early and late-stage apoptotic, and dead cells were counted, and statistical analysis was performed by GraphPad Prism 10.3.1 software. After ANOVA, Tukey's post hoc test was performed for multiple comparisons. ****P<0.0001; ***P<0.001; **P<0.01.

Combination of afatinib and copanlisib completely inhibits ErbB and PI3K pathways, which results in induction of caspase-3 cleavage

Previous studies have demonstrated that PI3K inhibitors can induce HER2 and HER3 phosphorylation, thus conferring resistance to PI3K inhibitors (50,51). Our recent study showed that copanlisib induced an increase in P-HER2 Y1248 and P-HER3 Y1289 in HPV-positive HNSCC cells, whereas a combination of afatinib and copanlisib blocked the phosphorylation of HER2 and HER3 (46). The present study further tested whether copanlisib induced P-HER2 Y1248 and P-HER3 Y1289 in HPV-negative Cal27 (Fig. 6A) and FaDu cells (Fig. 6B). Similarly, copanlisib markedly inhibited the phosphorylation of Akt and induced phosphorylation of HER2 Y1248 and HER3 Y1289, whereas the combination of copanlisib and afatinib completely blocked phosphorylation of Akt, HER2 and HER3. In addition, increased caspase-3 cleavage was induced by the combination compared with the single treatments.

Figure 6. A combination of copanlisib and afatinib effectively blocked the phosphorylation of HER2, HER3 and Akt, and enhanced caspase-3 cleavage. (A) Cal27 and (B) FaDu cells were treated with increasing concentrations of copanlisib, 0.5 µM afatinib or their combination for 24 h before lysis. The indicated proteins were detected by western blot analysis. C-, cleaved; P-, phosphorylated.

Discussion

Afatinib has shown positive results in HNSCC clinical trials and is currently listed in the NCCN guidelines as a third-line single agent for HNSCC treatment (25,26,31–33). It is important to determine the mechanisms by which patients with HNSCC develop resistance to afatinib in order to identify new afatinib-based combination therapies for patients with advanced HNSCC. The present study assessed the efficacy of co-inhibiting the ErbB family and PI3K through a combination of afatinib and copanlisib in the treatment of HPV-negative HNSCC. The results showed that the combination of afatinib and copanlisib induced a marked inhibition of cell viability and suppressed cell survival in vitro in comparison to either treatment alone. Notably, the combination also led to a significant inhibition of xenograft tumor growth without affecting the body weight of the mice. These results suggested that the combination of afatinib and copanlisib may have clinical potential for the treatment of HPV-negative HNSCC.

HNSCC is a heterogenous disease, and its gene mutations/alterations are associated with EGFR inhibitor resistance (5,40,52). In a previous study by Cheng et al (53), the genomic and transcriptomic changes for 15 HPV-negative and 11 HPV-positive HNSCC cell lines were characterized and compared with data from 279 tumors from TCGA. This study identified a number of genetic alterations, including in TP53, PTEN, PIK3CA and FAT1, in HPV-negative and HPV-positive HNSCC cells. The EGFR/ErbB and PI3K/Akt/mTOR pathways are considered the most attractive pathways to target for treatment of HNSCC due to overexpression or activating mutation of PIK3CA, and loss of function mutations of PTEN (5,54–57). It has been reported that constitutive activation of the PI3K/Akt/mTOR pathway due to the alterations in PIK3CA, PTEN, Akt or mTOR may be associated with resistance to EGFR inhibitors (5). The present study used two cell lines: Cal27 and FaDu cells. Both cell lines have TP53 mutations, whereas FaDu cells also have PIK3CA amplification, but Cal27 cells do not (58,59). The present study showed that treating both Cal27 and FaDu cells with afatinib alone blocked Akt phosphorylation at Thr308, but only partially blocked the phosphorylation of Akt at Ser473. Furthermore, it has been reported that PI3K inhibition can lead to increased phosphorylation and total levels of HER3, which confer resistance to PI3K inhibitors (50,51,60–62). The present data showed that copanlisib increased P-HER3 (Y1289), which was counteracted by the addition of afatinib. Notably, the combination of copanlisib and afatinib induced caspase-3 cleavage in addition to the complete inhibition of ErbB and PI3K/Akt pathways. These results provide a rationale for the co-inhibition of ErbB and PI3K as a method to treat HNSCC with or without PIK3CA amplification. However, it is important to test the efficacy of this combination in more HNSCC cell lines with more genetic alterations.

Our previous study reported that the combination of afatinib and copanlisib effectively suppressed HPV-positive HNSCC. The combination therapy blocked both ErbB and PI3K/Akt pathways, which was accompanied by deceased E6 and E7, and the induction of apoptosis, which indicated the increased efficacy of this combination in HPV-positive HNSCC (46). Milewska et al (63) reported that cell lines from multiple types of cancers, including HNSCC with PIK3CA mutations, are sensitive to the combination of afatinib and copanlisib. As the basal level of PI3K/Akt is also high in HPV-negative HNSCC and serves essential roles in the regulation of growth, metastasis and sensitivity to chemotherapy and targeted therapies (5,37,40,64), it would be reasonable to predict that this combination would also be beneficial in HPV-negative HNSCC with upregulated PI3K/Akt signaling.

One of the challenges in treating patients with PI3K pathway inhibitors is the associated toxicity from on-target and off-target effects (65). The most common side-effects observed with PI3K pathway inhibitors in patients are hyperglycemia, rash, stomatitis, diarrhea, nausea and fatigue (65,66). Due to the limitation of the xenograft model, this information is lacking in this study. Moreover, it is important to analyze the toxicity of the combination of the two drugs to the liver and kidney of mice to illustrate the feasibility of the combination; the lack of this analysis is another limitation of the present study. Thus, to accelerate its transition to treatment for patients with HNSCC, it is important to further determine the drug dosage and administration method of the combination in additional preclinical models.

Cisplatin-based chemotherapy is the primary option for the treatment of advanced HNSCC. It is interesting to compare the efficacy and side effects of co-targeting ErbB and PI3K, or in combination with cisplatin in preclinical models. We are currently performing these experiments using HPV-negative HNSCC cell lines with different PIK3CA and PTEN alterations.

Immunotherapy, which includes immune checkpoint blockade that targets PD-L1/PD-1 using PD-L1 or PD-1 inhibitors, has been an important advancement in the treatment of advanced HNSCC. Afatinib modulates PD-L1 expression in multiple types of cancer, including gastric cancer (67). In addition, it has been reported that PI3K inhibitors, such as BKM120, can decrease the expression of PD-L1 in HNSCC cells (68). It would be interesting to determine the effects of afatinib, copanlisib and their combination on the expression of PD-L1 in HNSCC cells, in order to determine the impact of the combination of afatinib and copanlisib on immune checkpoints.

In conclusion, the present results suggested that co-targeting the ErbB family kinases and PI3K using a combination of afatinib and copanlisib may have clinical potential for the treatment of patients with HNSCC.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

This research was supported, in part, by grants from the National Cancer Institute and National Institute of Dental and Craniofacial Research to HD (grant nos. R00CA149178, R01CA212094 and R56DE030423) and the Orakowa Foundation. This research was also supported by funds through the National Cancer Institute-Cancer Center Support Grant (grant no. P30CA134274) and the Maryland Department of Health's Cigarette Restitution Fund Program (grant no. CH-649-CRF).

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

XG contributed to the study conception and design, performed the experiments and data analysis, wrote the original draft, and reviewed and edited the manuscript. SA performed the experiments and conducted data analysis. ZY, RGL and XF contributed to data analysis. YT and RM were involved in data analysis, and reviewing and editing the manuscript. KJC contributed to conceptualization, data analysis, supervision, review, editing the manuscript and funding acquisition. HD contributed to conceptualization, data analysis, supervision, writing, review, editing the manuscript and funding acquisition. XG, SA and HD confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

In vivo experiments were conducted with the approval of the Institutional Animal Care and Use Committee of University of Maryland.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References