Introduction
Head and neck (HNC) constitute a substantial portion of the global cancer burden, with more than 890,000 new cases and approximately 450,000 deaths reported in 2018, ranking them among the top 10 most common cancers worldwide [
1]. These malignancies encompass a variety of tumours located in the oral cavity, pharynx, and larynx, each with distinct risk factors and epidemiological patterns. Despite advancements in diagnosis and treatment, the overall five-year survival rate remains around 50% [
2], highlighting the need for improved detection methods and therapeutic strategies.
Tobacco and alcohol consumption are widely recognised as the primary risk factors for HNC [
3], with the synergistic effect of tobacco and alcohol consumption on HNC extensively documented. However, human papillomavirus (HPV) infection has been implicated as a key determinant in the development of oropharyngeal cancers (OPC) [
4]. It is estimated that nearly 70% of OPC cases are HPV-positive, with a substantial proportion of these cases occurring in the tonsils [
5]. Approximately 85% of HPV-positive OPC cases are infected with oncolytic variants, such as HPV 16 or HPV 18 [
6]. These HPV-positive OPCs have been found to have distinct clinical characteristics when compared to HPV-negative tumours, particularly with respect to treatment response and overall survival rates [
7,
8].
The incidence rates of HPV-related OPCs are rising in both developed and developing countries. In some regions, the incidence of HPV-positive OPC has now surpassed that of HPV-negative OPC cases [
5]. Studies conducted the United States and Europe have demonstrated a sharp rise in the incidence of HPV-positive OPC, particularly in young adults. In the United States, the incidence of HPV-positive OPC has increased by 225% among young men in the last two decades [
9]. In Europe, a similar trend has been observed, with incidence rates of HPV-positive OPC increasing by approximately 60% among young men in the last 10 years [
10].
In developing countries, the incidence of HPV-related OPC is rising and is expected to continue to increase [
11]. In low- and middle-income countries within both Asia and Africa, the incidence of HPV-positive OPC has been increasing at a faster rate than in developed countries, with HPV 16 and 18 being the most common high-risk types identified [
11].
High-risk types of human papillomaviruses, including HPV 16 and HPV 18, code for several oncogenes, in particular, E6 and E7. Under normal conditions, E6 and E7 are expressed at low levels and are thought to function by creating conditions in the infected oral keratinocytes that favour replication of the virus [
12]. At higher levels, these two oncoproteins have major effects on a variety of cellular functions that may lead to uncontrolled growth [
13]. E6 is best known for its ability to bind to and mediate the degradation of the tumour suppressor p53 [
14] and other targets involved in cellular apoptotic pathways [
15]. As a consequence of these interactions, cells expressing E6 are much less likely to undergo apoptosis. E7 is known for its ability to bind to and inactivate the tumour suppressor Rb protein, disrupting its ability to regulate E2F transcription factors, resulting in disrupted cell cycle regulation [
15].
The detection of HPV is a critical component in the diagnosis and management of HPV-related OPC. There are several techniques available for HPV detection, including polymerase chain reaction (PCR)-based methods; hybrid capture (HC) assays; in situ hybridisation (ISH); and p16INK4a (p16) detection using immunohistochemistry (IHC) [
16,
17]. p16 detection is the most used method in the diagnosis of HPV-related OPC [
18]. While p16 overexpression is a marker of HPV-associated malignancy, the interpretation of p16 results can be subjective and can be affected by inter-observer variability [
19]. In addition, using the p16 detection method may produce false positive results, as p16 overexpression can occur due to other causes besides HPV infection [
20]. This underscores the complexity of using p16 as a biomarker, where its overexpression is not solely indicative of HPV involvement. Moreover, p16 overexpression is a late event in HPV-associated carcinogenesis, meaning that, it may not be present in early-stage cancers [
21].
PCR methods have been widely used for the detection of HPV16 [
22‐
28]. However, the major problem with the PCR approach is that the detection of viral DNA does not indicate an active infection. The virus may be dormant, and patients, even though they test positive for HPV, may not go on to develop cancer [
28]. Despite the various methods available for HPV detection, there is a lack of RNA-based HPV testing, which can indicate an active infection. Furthermore, most of these tests require a tissue biopsy, which may limit the scope of testing.
There has been a growing interest in using saliva as a liquid biopsy for diagnosing certain diseases [
29‐
32]. Saliva is an ideal choice as it contains genomic material and a diverse population of biological particles, making it a “mirror to the body” that reflects both local and systemic conditions [
33]. This makes it plausible that saliva may contain RNA released by head and neck cancer cells or HPV16 within the oral cavity.
To this end, our study aimed to develop and evaluate a probe-based biplex reverse transcriptase quantitative PCR (RT-qPCR) technique to identify viable HPV16 RNA in the saliva of patients with OPC. Detecting active virus in the saliva of OPC patients would be a valuable clinical tool that could aid in directing appropriate treatment strategies for these individuals.
Materials and methods
Cell lines
The cell lines used in this study included: squamous cell carcinoma from the cervix, SiHa (HPV16 positive); epidermoid carcinoma from the cervix, CaSki (HPV16 positive); adenocarcinoma of the cervix, HeLa (HPV18 positive); and ductal carcinoma of the mammary gland, MCF-7 (HPV negative). All cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) GlutaMAX™ (Thermo Fisher Scientific) with 1% glutamine, 10% fetal calf serum (Thermo Fisher Scientific) in a 37 °C incubator with humidified 5% CO2. Cells were passaged upon reaching approximately 80% confluency, typically every 2–3 days, to ensure optimal growth and viability. For passaging, cells were detached using 0.25% trypsin-EDTA solution (Thermo Fisher Scientific) and subsequently seeded at appropriate densities for continued culture or experimental use. All cell lines were regularly monitored for morphological consistency and tested periodically to confirm the absence of mycoplasma contamination.
Tissue specimens
Tissue specimens were retrieved from patients treated for Squamous Cell Carcinoma (SCC) of the Oropharynx at Royal Prince Alfred Hospital, Sydney, between 2002 and 2006. The study was approved by the Research Ethics Committee at Royal Prince Alfred Hospital, Sydney, Australia (Protocol number X05–0270). Informed consent was obtained for the collection of fresh tissues. Immediately after surgical resection, tissues were snap frozen on dry ice and stored at − 70 °C. The histology of tissues was assessed by hospital pathologists. For this study, six fresh-frozen tissues samples were selected from p16 positive (n = 4) and p16 negative (n = 2) OPC specimens.
Saliva specimens
All HNC saliva samples were obtained from patients through informed written consent with approval by the ethics board at Royal Prince Alfred Hospital, Sydney, between 2018 and 2022 (Ethics: X19–0195 and 2019/ETH11588). From each patient, 2 mL of unstimulated saliva was collected directly into sterile collection tubes (non-commercial kit) or a commercial kit (DNA/RNA Shield SafeCollect Saliva Collection Kit, Zymo Research). For this proof-of-principle study, we collected saliva specimens from p16 positive (n = 3) and p16 negative (n = 5) OPC patients.
RNA isolation from cell lines
5X106 cultured cells were homogenised by adding 1 mL of RNAzol® RT (Molecular Research Center). The homogenate was incubated for 5 minutes at 4 °C after the addition of 0.4 mL RNase-free water (Invitrogen, Thermo Fisher Scientific) for DNA, protein and polysaccharide precipitation, and centrifuged at 12, 000 x g for 10 minutes at 4 °C. The supernatant was then transferred to a fresh tube and 5 μL 4-bromoanisole (Molecular Research Center) was added for RNA purification. The sample was incubated for 3 minutes at 4 °C and centrifuged at 12, 000 x g for 10 minutes at 4 °C. RNA was precipitated by adding one volume of isopropanol (Sigma Aldrich) to the supernatant. The sample was incubated overnight at − 20 °C and then centrifuged at 12, 000 x g for 10 minutes. The supernatant was discarded, and the RNA pellet was washed twice with 75% ethanol (Sigma Aldrich) by centrifugation at 12,000 x g for 5 minutes. Lastly, the RNA pellet was solubilised in 20 μL of RNase-free water.
RNA isolation from tissue
100 mg of fresh frozen tissue was diced with a surgical blade, homogenised with a mortar and pestle, and rinsed with 1 mL of RNAzol® RT. 0.4 mL water was added to the sample and centrifuged at 12,000 x g for 10 minutes to precipitate the DNA and proteins. The sample was purified using BAN and centrifuged again at 12,000 x g for 10 minutes. RNA was precipitated using isopropanol according to the above protocol, and the sample was incubated at − 20 °C overnight. The RNA was then washed with 75% ethanol twice, and the RNA pellet was resuspended in 20 μL of RNase-free water.
RNA isolation from saliva
We adapted our previously published protocol for serum RNA isolation [
34] to extract total RNA from saliva samples. After retrieving the saliva samples from storage at − 30 °C, samples were centrifuged at 1600 x
g for 15 minutes at 4 °C. This process was done to separate the cellular debris. The salivary supernatant (400 μL aliquots) was homogenised with 1.5 mL Tri-Reagent RT-Liquid Samples (Molecular Research Centre) and 100 μL 4-bromoanisole, then centrifuged at 12,000 x
g for 20 minutes at 4 °C in 1.5 mL phase-lock gel tubes (5PRIME).
The RNA-containing aqueous phase was decanted into a fresh DNA Eppendorf Lo-bind tube, mixed 500 μL isopropanol and 5 μL Glycogen, and incubated overnight at − 20 °C. Following incubation, samples were centrifuged at 12,000 x g for 20 minutes at 4 °C. The supernatant was discarded, and the RNA pellet was washed twice with 1 mL of 70% ethanol, air dried, and resuspended in 20 μL of RNase-free water. For increased yield, samples from the same participant were pooled.
RNA quantification and quality control
Total RNA was quantitated using a Nanodrop™ 1000 3.7.1 UV-Vis Spectrophotometer (Thermo Fisher Scientific). After cleaning the stage with water and 70% ethanol, the instrument was blanked using 1 μL of RNase-free water. Using 1 μL of sample, the absorbance spectra were measured. RNA concentration was determined from the 260 nm peak, and purity was assessed using absorbance ratios at 280 nm (A260/ A280) and 230 nm (A260/A230). Accepted ratios for purity vary with downstream applications, however, typical A260/ A280 ratios should be between 1.8–2.2, while requirements for A260/A230 ratios are generally greater than 1.7.
cDNA synthesis
The High-Capacity cDNA Reverse Transcription Kit from Thermo Fisher Scientific was used for cDNA synthesis. cDNA synthesis was performed using a 20 μL reaction, per Table
1, and employed a range of RNA input concentrations from 50 pg to 200 ng. Tubes were then placed in a thermocycler and run using the following conditions: 10 minutes at 25 °C, 120 minutes at 37 °C, 5 minutes at 85 °C and the sample was held at 4 °C until collected.
Table 1
cDNA synthesis components per 20 μL reaction
10x Reverse Transcriptase Buffer 1.0 mL | 2.0 |
25x dNTP Mix 100 mM, 200 μL | 0.8 |
RNase Inhibitor 100 μL, 20 Units/μL | 1.0 |
RNA Input (various concentrations) | 1.0 |
10x RT Random Primer, 1.0ML | 2.0 |
MultiScribe™ Reverse Transcriptase 100 μL, 50 units/μL | 1.0 |
Nuclease-free water | 12.2 |
Total | 20.0 μL |
Reverse transcriptase quantitative polymerase chain reaction (RT-qPCR); Biplexing HPV16 oncogenes E6/E7
Following cDNA synthesis, samples were diluted 1:4 by adding 60 μL nuclease-free water. RT-qPCR was then performed in a 5 μL reaction volume, per Table
2 using the StepOnePlus™ Real-Time PCR system (Thermo Fisher Scientific, USA). Reactions we performed in triplicate. The reactions utilised the TaqMan Universal PCR Master Mix (Applied Biosystems, Thermo Fischer Scientific, USA), adhering to the cycling conditions outlined in Table
2. TaqMan assays were designed for the oncogenes
E6 and
E7, using Primer3Plus (
https://primer3plus.com/) based on sequences obtained from NCBI (
https://www.ncbi.nlm.nih.gov/refseq/). These sequences are outlined in Tables
3 and
4 below. To biplex these two oncogenes, E6 was labelled with a VIC™ and E7 was labelled with a FAM™ probe.
Table 2
RT-qPCR components per 5 μL reaction
TaqMan Universal PCR Master Mix (20X) | 2.5 |
TaqMan Assay for E6/E7 (20X) | 0.5 |
cDNA | 1.0 |
Water | 1.0 |
Total | 5.0 |
Table 3
Primer and probe sequences E6
Context sequence | TGGACAAGCAGAACCGGACAGAGCC |
Probe (VIC) | TCCGGTTCTGCTTGTCC |
Forward sequence | GCTCAGAGGAGGAGGATGAAATAGA |
Reverse sequence | GAGTCACACTTGCAACAAAAGGTT |
Table 4
Primer and probe sequences E7
Context sequence | ACCCAGAAAGTTACCACAGTTATGC |
Probe (FAM) | ACAGAGCTGCAAACAA |
Forward sequence | ACCCAGAAAGTTACCACAGTTATGC |
Reverse sequence | TGCTTGCAGTACACACATTCTAAT |
A 2-step PCR assay was employed for its enhanced specificity and flexibility, particularly beneficial for biplexing the oncogenes E6 and E7. This approach allowed for separate optimisation of reverse transcription and PCR amplification conditions, improving assay sensitivity, and reducing potential interference from RT reaction components during the PCR step. To ensure assay accuracy and prevent contamination, each RT-qPCR run included both positive and negative controls. Positive controls comprised known quantities of target cDNA (Siha and CaSki) to verify PCR efficacy, while negative controls (no-template controls) contained all reaction components except the template cDNA, serving to detect any potential contamination or non-specific amplification. These controls were systematically included in each PCR plate. Contamination prevention was rigorously addressed by employing separate workspaces for different stages of the protocol. Additionally, reagents were aliquoted to minimise exposure and reduce contamination risk.
qRT-PCR analysis for 18S, ACTB, p53, and dicer 1
Quantitative real-time PCR for 18S rRNA, ACTB, p53, and Dicer 1 was performed using specific TaqMan Gene Expression Assays. The reaction setup followed the protocol outlined in Table
2, with a total reaction volume of 5 μL comprising 2.5 μL of TaqMan Universal PCR Master Mix (2X), 0.5 μL of TaqMan Assay for each gene (20X), 1.0 μL of cDNA, and 1.0 μL of nuclease-free water.
The thermal cycling conditions are detailed above in Tables
2 and
5. To ensure assay specificity and integrity, no-template controls and no-reverse transcription controls were included in each assay and PCR plate.
Table 5
RT-qPCR cycling conditions
Denaturation | 95 °C, 15 seconds | × 40 cycles |
Annealing and elongation | 60 °C, 1 minute |
Hold | 4 °C |
Data analysis
Absolute quantitative RT-qPCR data was imported into GraphPad Prism (Version 8.2.1) and Cq values were plotted against sample using column graphs that compare both singleplex and biplex RT-qPCR results. PCR efficiencies were determined using LinRegPCR (v 2021.2) and measured on a scale between 1.0 and 2.0, with 2.0 representing 100% efficiency. PCR efficiencies above 1.5 were determined to be acceptable. A two-sided t-test was used to determine whether a significant difference was observed between singleplex and biplex RT-qPCR reactions (Cq values) and their PCR efficiencies (p < 0.05).
Discussion
The improved treatment outcomes and survival rates of HPV-positive OPC patients underline the importance of accurate HPV detection for guiding treatment strategies [
37,
38]. While HPV-encoded oncogenes E6 and E7 disrupt cell cycle regulation and are key biomarkers for HPV-associated cancers [
39‐
41], the overexpression of p16, though sensitive, is not entirely specific for high-risk HPV infection [
23]. This necessitates a more precise approach, as approximately 5–10% of all OPCs may exhibit false positives due to p16 overexpression unrelated to HPV infection [
42‐
44].
Consequently, the use of p16-IHC or HPV-specific testing alone as a reliable means of determining HPV status has been called into question, with recent studies identifying a subgroup of patients with discordant p16 and HPV positivity [
45]. Specifically, most of the discrepant cases reported to date are p16-positive but HPV DNA-PCR or DNA-ISH negative. In light of these findings, the College of American Pathologists (CAP) recommends additional HPV-specific testing at the discretion of the pathologist and/or treating clinician following p16 testing [
46]. These developments highlight the need for greater scrutiny of testing methods and the importance of accurate HPV status determination in guiding clinical decision-making.
It is for this reason that viral RNA expression has been suggested as the gold standard for a viable infection, meaning the virus is transcriptionally active. Our study contributes to this need by developing a biplex RT-qPCR method for non-invasively detecting active high-risk HPV16 in saliva. This approach, focusing on E6 and E7 mRNA, offers a potential alternative to p16-IHC and DNA-based HPV tests, enhancing the detection of transcriptionally active infections. Furthermore, this methodology is scalable and well-suited for high-throughput screening, making it an attractive option for widespread screening or HPV16 surveillance programs.
Our results demonstrated successful detection of E6 and E7 mRNA in HPV16-positive cell lines, SiHa and Caski, and in p16-positive OPC patient tissues across a range of RNA input levels. This not only highlights the assay’s sensitivity but also its efficiency in amplifying these specific mRNA targets. The specificity of our assay was further validated in clinical settings: when applied to OPC saliva samples, transcriptionally active E6 and E7 mRNA were exclusively (100%) detected in saliva from p16-positive patients. This specificity is required for the assay’s potential clinical application, ensuring it reliably identifies patients with active HPV16 infections.
Moreover, the assay’s performance in terms of PCR efficiency was robust. We observed mean PCR efficiencies of approximately 1.8, indicative of optimal amplification. This efficiency was consistent across both singleplex and biplex methods for E6 and E7 assays. The slightly better performance of the E6 assay compared to the E7 assay, with mean Cq values of 31.5 and 34.4, respectively, aligns with existing studies suggesting more consistent expression of E6 in various HPV types and stages of infection [
47,
48].
Saliva testing offers advantages over blood and tissue-based testing due to its non-invasiveness and ease of sample collection, allowing for a time- and cost-effective diagnosis. Two primary considerations drove the use of tissue biopsies as a comparison in this study. Firstly, tissue biopsies are widely recognised as the ‘gold standard’ in the diagnosis and characterisation of OPCs. By comparing our saliva-based test with tissue samples, we aimed to benchmark our method against the established standard in clinical practice. Secondly, our access to these specific tissue samples, with their well-documented histological profiles, provided an opportunity to validate our saliva test. While brush biopsies are less invasive and easier to collect, they were unavailable for this study. Furthermore, our focus was to compare our saliva-based diagnostic approach with the most rigorous diagnostic method currently available, which, in this case, was tissue biopsy analysis.
Several studies have sought to use saliva for oral cancer detection but very few studies to date have used RNA to detect viable infections [
49‐
56]. In one Australian study, it was demonstrated that saliva rinses could be used to detect key HPV-DNA oncogenic targets with 92.9% sensitivity. It was shown that 39 of 42 oral rinses from p16-positive patients had detectable HPV16-DNA [
51]. Another study using oral rinses from 110 patients employed nested PCR to detect low copy numbers and showed a sensitivity rate of 75% [
54]. Furthermore, antibodies specific to HPV16 E6 and E7 were present in serum at a similar rate of 51.4%. Although the sensitivity rates were low, it suggests that HPV detection in oral rinses may be comparable with the gold standard method of p16 testing in tumour tissues [
53].
The primary limitation of our approach is that RT-qPCR only detects RNA, raising the potential for false negatives, particularly if the virus is dormant and not transcriptionally active. To enhance diagnostic accuracy, this salivary RT-qPCR could be used alongside p16 staining. A dual positive result from both tests might provide more clinical insight that relying on p16 staining alone. Future enhancements of this assay could include additional HPV16 targets to address these limitations.
One possible approach is to identify HPV16 genes associated with viral dormancy and include these targets along with E6 and E7. The E2 gene is frequently overexpressed during viral latency and a key regulator of both E6 and E7 [
57]. The triumvirate of E2, E6, and E7 targets might be able to discern between viable and latent viral infection. Another strategy is to detect both the presence of viral DNA and RNA in the same RT-qPCR assay. Other viral DNA targets could include the L1 and L2 genes which are highly conserved [
58,
59] or non-transcribe regions of the HPV16 genome. The latter would be an ideal RT-qPCR target.
In this study, we focused on demonstrating the feasibility of detecting HPV16 RNA in saliva, a significant step toward non-invasive diagnostics. We acknowledge that the determination of limit of detection (LOD) and limit of quantitation (LOQ) was not performed, a decision shaped by our aim to primarily assess qualitative detection capabilities. While our results highlight the assay’s potential sensitivity, the absence of LOD and LOQ analyses represents a limitation. Future research should address this by quantifying the assay’s sensitivity and expanding its clinical applicability, thereby expanding our understanding of HPV16’s detection dynamics in saliva and its implications for early diagnosis and monitoring.
We also acknowledge that the sample size used for the salivary testing is limited and a larger cohort will be required to further assess the sensitivity and specificity of this salivary RT-qPCR method. An additional hurdle in utilising salivary samples is the absence of universally recognised standards for the collection and handling of such specimens. A consistent collection protocol and a reliable approach for extracting genomic material must be established to address this issue [
36]. Presently, only a limited number of techniques are available that can extract both DNA and RNA from a single salivary sample [
60,
61].
Overall, continued efforts towards standardisation and optimisation of saliva-based testing will be important for advancing the field of liquid biopsy and improving patient diagnosis. Despite these challenges, the use of saliva in HPV16 testing continues to show promise. Ongoing efforts to standardise salivary collection, processing, and inclusion of other RNA/DNA targets, will be critical in developing a robust RT-qPCR liquid assay for HPV detection.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.