Fluid Biomarkers in HPV Related Oropharyngeal Carcinomas: History
Please note this is an old version of this entry, which may differ significantly from the current revision.
Contributor: , , , , ,

Biomarkers are crucial in oncology, from the detection and monitoring of cancer to guiding management and predicting treatment outcomes. However, histological assessment of tissue biopsies, which is currently the gold standard for oral cancers, is technically demanding, invasive, and expensive. Liquid biopsies, through the analysis of cancer biomarkers in bodily fluids, offer promising non-invasive alternatives to patient care in oral cancers. The aim of this comprehensive literature review was to explore and discuss current markers that are detectable in various biofluids, including plasma, oropharyngeal swabs, and oral rinse.

Oropharyngeal squamous cell carcinoma is classified by the World Health Organization into HPV-positive and HPV-negative types [4], due to their significant differences in epidemiology, clinical features, histology, and prognosis. HPV-positive OPC patients generally show better prognosis and survival compared to their HPV-negative counterparts. Regarding HPV-positive OPCs, biomarkers are heavily centered around detection of HPV DNA and their associated oncoproteins, aiming to improve outcome and survival. Twenty clinical studies were identified from the systematic search and are categorized into genetics, epigenetics, extracellular vesicles, oncoproteins, and immune response-related markers.

Recent clinical trials investigating biomarkers for both HPV-positive and HPV-negative cancers have approaches from various levels and different biofluids. The majority of fluid biomarker research is still in plasma, with HPV ctDNA remaining the most-studied fluid biomarker, which has already been applied clinically to other HPV-related cancers. Promising candidates have been found that could aid in detection, staging, and predicting prognosis of OPCs, in addition to well-established factors including HPV, alcohol consumption, and smoking status. These studies also emphasize the possibility of enhancing prediction results and increasing statistical significance by multi-variate analyses. Liquid biopsies, being non-invasive methods, offer promising assistance in enhancing personalized medicine in treating cancer patients. 

  • liquid biopsies
  • fluid biomarkers
  • tongue base
  • HPV
  • oropharyngeal cancer
  • detection
  • diagnosis
  • prognosis
  • treatment monitoring
  • de-escalation

1. Genetics

1.1. DNA

Within HPV-positive oropharyngeal squamous cell carcinoma (OPSCC), the use of HPV circulating tumor DNA (ctDNA) has been a promising biomarker in various aspects of clinical management, especially in early detection, prognosis prediction, and treatment outcome monitoring.


Early detection has been associated with lower incidence and better survival. The use of ctDNA or cell free DNA (cfDNA) as biomarkers for early detection has already been established in various types of cancers, such as Epstein–Barr virus (EBV)-associated nasopharyngeal cancers [15]. However, despite the high association of HPV with OPC, many patients tend to present at an advanced stage as their disease is difficult to detect and generally asymptomatic. Nevertheless, clinical early detection may be similarly possible with liquid biopsies.
The potential use of HPV DNA in oral rinse as an early detection tool for OPC has been showcased in two studies. Twelve participants from a cohort of 660 cancer-free individuals were discovered to have positive oral HPV16-DNA and were followed-up for 24 months in a prospective study from Australia [6]. Among the 12, three cases had persistent oral HPV16 infection, with one subsequently developing a stage I OPC while another had mild tonsillar dysplastic lesion. This shows that a high persistent HPV viral load serves as a predictor for chronic infection and HPV-driven cancers including OPSCC. Despite the small sample size, this suggests the potential that a high persistent HPV load may serve as a predictor for a chronic infection and HPV-driven cancers including OPSCC, hence the possible utilization of HPV DNA to screen asymptomatic individuals for OPSCC. A similar conclusion could be drawn from another larger-scale retrospective study which validated that HPV tests on oral rinse could detect OPCs with a positive-predictive value of 94% and sensitivity of 78% [5].
Apart from oral rinse, HPV cfDNA in plasma could also serve the function of detecting OPC. HPV cfDNA could be detected in OPSCC patients with 100% specificity and 72% sensitivity, respectively [7]. Further supporting the rationale that cfDNA in plasma could be used as a diagnostic tool, viral loads in plasma were found to have a positive correlation with that in tumor tissue. However, probably due to the lack of tumor lysis and HPV DNA in the systemic circulation, cfDNA could not be detected in plasma of patients without nodal metastasis, which limits its use in early diagnosis. It was reported in another study that HPV cfDNA E6 and E7 in plasma could be used in combination to detect HPV-positive OPSCC with a specificity and sensitivity of 100% and 77%, respectively [13].
A meta-analysis reviewing 10 studies involving a total of 457 HPV-positive HNSCC patients ascertained the above conclusions. It was reported that HPV cfDNA in blood has diagnostic ability with a pooled sensitivity and specificity of 0.65 and 0.99, respectively [8]. Given that it has a pooled diagnostic odd ratio of 371.66, HPV cfDNA shows sufficient diagnostic accuracy for HNSCC. However, it was noted that the area under curve (AUC) of receiver operating characteristic (ROC) curve for the diagnostic ability of HPV cfDNA was 0.77, which suggests a certain degree of heterogeneity across results.

Predicting Prognosis

Apart from detection, liquid biopsy of HPV ctDNA in plasma could be applied to predict prognosis in HPV-positive OPC patients, particularly in the selection of patients who may benefit from de-intensified treatment.
HPV cfDNA E6 and E7 concentrations were found to increase with size of tumor in HPV-positive OPSCC, with the median copy number of E6 and E7 cfDNA per mL plasma being significantly higher in large tumors [13]. Similar results were obtained by Veyer et al., demonstrating that baseline HPV16 ctDNA loads were positively correlated with T, N, and M (tumor, nodes, metastasis) status in HPV-positive OPC patients [10]. Furthermore, median concentrations of HPV ctDNA increases from 1.45 to 4.8 log cp/mL from Stages I to IV, respectively, showing positive correlation with the currently used 2018 AJCC (American Joint Committee on Cancer) staging. Undetectable baseline HPV16 ctDNA is associated with lower staging, with 74% of patients presenting with stage I OPSCC. Positive trends could also be observed between baseline HPV16 ctDNA detection among HPV-positive OPSCC patients and progression-free survival (PFS), as well as mortality rates. This highlights the possibility of using HPV ctDNA as an additional tool for risk stratification apart from histological and clinical diagnosis.
The above findings are partially consistent with those from another study which ascertained the positive correlation of baseline HPV16 ctDNA levels with tumor burden among HPV-positive OPC patients [11]. However, this study further elicited that despite low pre-therapeutic levels of HPV16 ctDNA (≤200 copies/mL) being associated with low tumor burden, it is paradoxically indicative of worse prognosis. Through comparing HPV ctDNA load in plasma with HPV copy number and level of HPV integration in the biopsied tumor genome, it was found that those with lower baseline HPV ctDNA concentrations have lower HPV copy number and higher HPV integration in tumors, which were adverse tumor genomic factors leading to poorer prognosis. This showcases the possible use of HPV ctDNA in prognostication of OPC patients.
There is yet to be a broad consensus on the role of baseline HPV ctDNA for risk stratification. Easily obtainable biofluid including saliva and blood shows the potential value of providing supplementary information for treatment choice, though its clinical application as a prognostic marker awaits large-scale multicentric study to further investigate.

Treatment Monitoring

Apart from initial staging, it has been reported that HPV ctDNA concentrations could be used to predict and monitor treatment response by reflecting the dynamic tumor burden.
From a small-scale study, among six patients with post-treatment serum, a patient with negative baseline HPV ctDNA detection and another four that had significantly decreased HPV ctDNA loads post-treatment showed a complete clinical response to treatment, whereas the patient having fivefold increase in HPV ctDNA load after treatment died within days [10].
A similar conclusion could be drawn from research conducted by Haring et al. in which changes in plasma HPV16 ctDNA concentrations were shown to correlate with radiographically determined treatment responses for OPC patients on immunotherapy [9]. Results suggested that patients with a <60% increase in HPV16 ctDNA is associated with a favorable response without disease progression, whereas those with a ≥60% increase had disease progression despite therapy. Furthermore, changes in HPV16 ctDNA concentrations were found to precede determination of radiographic response, even up to 100 days in particular cases. HPV16 ctDNA could also effectively identify those who are having radiographic pseudo-progression, with its load decreasing in accordance with the molecular response. Furthermore, HPV ctDNA could also be used to predict treatment response for advanced HPV-positive OPSCC patients undergoing chemoradiotherapy (CRT). All patients that had complete radiological response demonstrated HPV DNA levels below the threshold, which was defined as 10 reads in 7 amplicons [12], reiterating its function as a dynamic marker for therapeutic response.
With more accurate predictions on treatment response, patients could be selected for de-escalation of treatment and avoid unnecessary biopsies and surgical treatment.

Post Treatment Surveillance

Recurrences of OPSCC mostly occur within the first two years following treatment completion. HPV-positive OPSCC can recur up to five years or more after therapy [15,16]. It is crucial to have a readily obtainable biomarker to monitor conditions of patients regularly over a long period of time.
In a prospective clinical trial, HPV ctDNA in plasma demonstrated the ability to identify disease recurrence in non-metastatic HPV-positive OPSCC patients receiving CRT. With a median follow-up time of 23 months post-treatment, all 87 patients that had undetectable HPV ctDNA throughout did not develop recurrence [14]. Among the 28 patients showing positive HPV ctDNA during post-therapeutic surveillance, in which 16 of them had two consecutive plasma samples with detectable HPV DNA, 15 patients were diagnosed with biopsy-proven recurrence. This shows that recurrence could be reflected by two consecutively positive HPV ctDNA blood tests with a positive predictive value (PPV) of 94%, indicating serial HPV measurements may be useful in risk stratification with higher accuracy. Similar to its application in prognosis, detection of post-treatment HPV ctDNA positivity was also found to precede biopsy-proven recurrence by a median of 3.9 months.
This is also illustrated by Lee et al. where a patient in the study showed elevated HPV DNA level despite lack of active disease at the primary site or cervical lymph nodes [12]. Positron emission tomography (PET) showed increased 18F-fluorodeoxyglucose (18F-FDG) uptake from the liver, however, HPV DNA load demonstrated marginal reduction after resection of the liver and paradoxically started to increase continuously. PET subsequently revealed increased uptake in cervical lymph nodes 8 months after the surgical resection of liver, confirming the ability of plasma HPV DNA to detect recurrence prior to radiographic imaging. Another study specifically looking into the use of HPV cfDNA E6 and E7 ascertained the above findings. 5 patients (67.5%) with HPV-positive OPC showing two consecutive positive HPV loads after treatment subsequently developed recurrence, and HPV cfDNA was observed to increase prior to that diagnosed by routine examination [13].

1.2. RNA


The use of genetic markers for the detection of HPV-positive OPC extends to RNA apart from DNA. Presence of HPV E6/E7 mRNA indicates transcriptionally active HPV, which is involved in the carcinogenesis of OPSCC. HPV E6/E7 mRNA from fine needle aspiration (FNA) of neck lymph nodes of metastatic OPSCC patients showed an overall agreement of 88%, which is comparable to conventional immunohistochemical p16 staining of tumor tissue, allowing it to serve as a tool for diagnosis [16]. FNA samples were found to have higher accuracy as compared to oral samples.


Instead of focusing on HPV, a prospective study observed the role of oral and gut microbiomes on HPV-positive OPSCC patients after CRT through measuring rRNA of bacteria [17]. Pre-therapeutic oral and gut microbiomes were found to differ among different stages of HPV-positive OPC. For oral microbiome there was a significant enrichment in four genera (Fusobacterium, Gemella, Leptotrichia and Selenomonas) in stage III patients compared to stage I-II patients, whereas gut microbiome showed significant enrichment in two phyla: Actinobacteria and Proteobacteria in stage III patients. Further trials are required to research into whether such differences in baseline microbiome composition could be used for risk stratification in OPC patients.

Treatment Monitoring

Composition of microbiome in oropharyngeal swabs was revealed to correlate with the treatment status, with the number of species in oropharyngeal swabs decreasing significantly after therapy. A shift in taxonomic composition could also be observed with an increase in relative abundance of gut-associated obligate anaerobes, such as Bacteroides species. However, the underlying reason behind such a change is yet to be discovered.
With emerging interests on the action on microbiome, the possible implications of how novel development in the field could aid in the diagnosis and monitoring of OPC is worth further exploration.

2. Epigenetics

Studies have investigated a range of novel methylation markers aiming to identify suitable ones for diagnostic or monitoring purposes. Various genes were found to associate with OPC, warranting studies to assess the relationship between epigenetic modifications and clinical progression of the cancer.
Calmodulin-like 5 (CALML5) is a skin-specific calcium binding protein which is involved in regulation of epidermal differentiation [20]. Translocation of Yes-associated protein 1 (YAP1), an oncogenic driver of HNSCC [21], into the cytoplasm is hindered by methylation of CALML5, resulting in increased transcription in carcinogenesis [22,23]. As a membrane-bounded protein with a glycosylphosphatidylinositol anchor, lymphocyte antigen 6 complex locus D (LY6D) has an established role in the adhesion of head and neck cancer cells to endothelial cells in HNSCC patients [24]. Hypermethylation of the gene promoter region was also reported in lung cancer [25]. It was illustrated in a verification study that the methylation status of CALML5 and LY6D was associated with reduced survival with a hazard ratio of 7.01 and 10.69, respectively [18].
DNAJ heat shock protein family member C5 gamma (DNAJC5G) belongs to the DNAJ family, in which various members were revealed to have different interactions with viruses including adenovirus, vaccinia virus, HIV-1 [26], and HPV16 E7 oncoprotein [27]. Hypermethylation of CALML5, LY6D, and DNAJC5G were demonstrated to correlate with clinical condition of HPV-positive OPSCC patients, with the number of patients showing methylation of the genes significantly reduced post-treatment as compared to their pre-treatment statuses [18].
Post-translational modification of genes in the salivary rinse has also been investigated to identify markers for post-treatment surveillance of HPV-positive OPSCC. It has been shown that a combined panel of HPV DNA (HR E5L2-4) and hypermethylation of endothelin receptor type B gene (EDNRB) gave the best sensitivity (0.90) and specificity (0.81) in detecting recurrence [19]. The panel was also able detect recurrence earlier than clinical detection by 2.4 ± 1.6 months. This adds on to previous studies which established hypermethylation of EDNRB with the presence of invasive OPSCC and increased risk of locoregional recurrence [28,29].
With promising results, these liquid biopsy markers may be utilized complementary to conventional radiographic examination in long-term patient surveillance.

3. Extracellular Vesicles (EVs)

Extracellular vesicles are actively released by cells to facilitate cell-to-cell communication. They contain molecular cargo such as nucleic acids and proteins, and are present in all biological fluids, making EVs valuable and worthy of investigation. It has been established that in HPV-related cervical cancer, their contents, in particular microRNAs, are altered.

3.1. Detection

In an in vitro study on oropharyngeal tumor cells, it was revealed that HPV-positive OPSCC cells produced less EVs than HPV-negative cells. Their cargo was analyzed and revealed an abundant number of miRNAs independent of HPV status, with 9 common miRNAs found to be present in all samples. Pathway analysis showed that PI3K-Akt, Fox0, HIF-1, mTOR, and p53 signaling pathways were targeted by these miRNAs. However, 14 and 19 miRNAs were identified to be only enriched in HPV-positive and HPV-negative groups, respectively. miR-99a-5p, miR-27a-3p, and miR-27b-3p, which were previously reported as cargo of EVs from HPV18 + HeLa cervical carcinoma cells, were also found to be enriched in HPV positive OPSCC-produced EVs. Expression of miR-20b-5p, which is also enriched in HPV-positive OPSCC-derived EVs and promotes cancer cell invasion, has been proven to be dependent on the viral E6 oncoprotein. Therefore, although EVs of different origins may have overlapping miRNA profiles, certain miRNAs are dysregulated in specific subtypes, providing potential for detection of OPSCC [30].

3.2. Recurrence

One major challenge to developing clinically significant miRNA biomarker panels is the biological variability and differences in detection methodology, including collection, storage, RNA isolation, and processing, thus leading to unreproducible findings. In a study of 40 OPSCC patients, a cross-validated StaVarSel (stable variation selection) method was utilized and produced a 11-miRNA-ratio model that accurately identified HPV-positive OPSCC (90% sensitivity and 79% specificity), showing good potential for diagnosis and predicting recurrence. However, the subjects were in advanced stages of OPSCC and thus the ability of EV signatures in early detection was not tested [31].

4. Oncoproteins

4.1. Early Detection and Classification

Patients with HPV + OPSCC have better overall and disease-free survival. Determination of HPV status is therefore important for risk stratification. However, the surrogate immunohistochemical marker, p16(INK-4a), which is used clinically with high sensitivity, only has a specificity of 83%, leading to potential misclassification of a specific subgroup of patients who are p16(INK-4a) positive but HPV DNA-negative. These patients have been reported to have distinct clinical features, worse prognosis, and increased metastasis potential compared to both p16(INK-4a) and HPV DNA-positive patients.
Thirty-six patients with newly diagnosed OPSCC were analyzed with three serum markers, which were increased in the sera of HPV-associated OPSCCs, and were identified to differentiate HPV-associated and HPV-independent OPSCCs, including ApoF, Galactin-3-Binding protein, and complement component C7 [32]. ApoF is a lipid transfer inhibitor protein that regulates cholesterol transport. It is hypothesized to be increased due to metabolic reprogramming by malignant cells or stromal reaction by HPV per se. Galactin-3-Binding protein, a glycoprotein implicated in self-nonself discrimination, is upregulated in viral infection, with higher levels being linked to poor prognosis and progression in cancers not limited to OPSCC. Complement C7 is an important component of the membrane attack complex (MAC), and can activate oncogenic MAPK, ERK, PI3K, and Ras pathways, which are important in the progression of HPV-infected cells toward malignancy. Serum proteomes therefore have potential in providing simple and cost-effective ways to assess HPV status and assist classification of patients.
Oral testing is another focus of non-invasive biomarker development, due to the close proximity to oropharyngeal tumors in addition to the easy collection of saliva and swabs. A study compared HPV E6 oncoproteins and E6/E7 mRNA in fine-needle aspiration samples and oropharyngeal samples (saliva and oral swabs) and found that HPV E6 oncoprotein tested in such samples agreed with results of p16 or HPV tested tumors, but at a lower concentration and rate than mRNA and DNA [16]. Oral testing for oncoproteins is currently not as informative as DNA and RNA counterparts towards detecting malignancies, and further studies and development are required for sample optimization.

4.2. Predicting Prognosis

Although HPV-positive OPSCC has a more favourable prognosis, 10–25% still recur. A small single-institution study evaluated the possibility of using a liquid biopsy as a prognostic marker, measuring HPV E6E7 expression in circulating tumor cells (CTCs). Pre-treatment baseline expression of HPV E6/E7 oncogene in CTCs were associated with both progression-free survival and overall survival, with patients having significantly higher risk of relapse and death. However, the presence of E6E7 + CTCs at the end of treatment was not significant in PFS and overall OS [33]. Research is still undergoing to combine the biomarker with AJCC staging data to improve risk definition and discrimination.

5. Immune Response

Predicting Prognosis

Immunological responses to HPV may provide insights for treatment response as well. A correlation is likely between the HPV-positive status of the tumor and T cell response to HPV antigens, as demonstrated by a study which found that 80% of patients who had detectable HPV16 DNA in tumors also had CD4+ or CD8 + T cell response to HPV16 E6/E7. This study provided survival data in OPSCC stratified by cell-mediated immune response to HPV16 peptides, and demonstrated that enhanced immunoreactivity to antigen E7 was correlated with longer disease-free survival. The average disease-free survival (DFS) for all patients was 43.7 months, and was increased to 47.3 months for the HPV+ cohort. Higher average DFS (49.6 months) was observed in patients with an increased CD8 + T cell response to E7. Cox regression analysis revealed that a retained or enhanced CD8 + T cell response to E7 has significant influences on DFS [34]. Rising CD4 +/CD8 + T cell response may also indicate recurrence of disease, but requires further studies looking at a more distant time-point after treatment.
Apart from T cell response, antibodies directed against HPV antigens have also been investigated as prognostic biomarkers pre- and post-treatment. With removal of the tumor and thus the source of antigens, B cell responses should in theory become limited. A study on 77 patients revealed better survival outcomes for patients with higher levels of anti-E2 IgG. Interestingly however, the presence of IgA antibodies does not appear to be linked to survival outcomes. It was also found that levels of anti-E6 and E7 IgG antibodies differ pre- and post-treatment. However, their application on predicting recurrence is still unknown due to limitations of the cohort.

This entry is adapted from the peer-reviewed paper 10.3390/ijms232214336

This entry is offline, you can click here to edit this entry!
Video Production Service