Human Papillomavirus antibodies and risk of subsequent head and neck cancer

Dr. Tim Waterboer
© T. Waterboer

Human papillomavirus type 16 (HPV16) infection is causally associated with a subset of oropharyngeal cancers (OPC), mainly tonsillar cancer. Similar to cervical cancer, antibodies to the E6 oncoprotein are strongly associated with HPV-driven OPC in patients after diagnosis. We currently investigate whether HPV antibodies are associated with OPC risk when measured prior to tumor diagnosis. Pre-diagnostic plasma samples from patients and controls enrolled in the EPIC (European Prospective Investigation into Cancer and Nutrition) cohort were analyzed for antibodies against multiple proteins of HPV16 and other HPV types, and Odds Ratios (ORs) and 95% Confidence Intervals (CI) were calculated, adjusting for potential confounders.
HPV16 E6 seropositivity was present in pre-diagnostic samples for 34.8% of patients with OPC and 0.6% of controls (OR 274, 95% CI 110 to 681). The increased risk of OPC among HPV16 E6 seropositive participants was independent of time between blood collection and diagnosis and was observed more than 10 years before diagnosis of OPC1(Figure). We are currently replicating these findings in other international cohort studies and investigate the potential of E6 antibody assays to serve as screening tools.


1. Kreimer AR, Johansson M, Waterboer T, Kaaks R, Chang-Claude J, ... , Pawlita M, Brennan P. Evaluation of human papillomavirus antibodies and risk of subsequent head and neck cancer. J Clin Oncol. 2013 Jul 20;31(21):2708-15.

© A.R. Kreimer et. al., Kinetics of the Human Papillomavirus Type 16 E6 Antibody Response Prior to Oropharyngeal Cancer; J Natl Cancer Inst. 2017

Figure: HPV-16 E6 antibodies as predictive biomarker for oropharyngeal cancer (OPC). Top panel: Multiplex serology enables early OPC detection. Left panel: HPV-16 E6 antibody levels (median fluorescence intensities, MFI) by years from blood draw (left) to diagnosis (right) among OPC cases. (Kreimer et al., J Natl Cancer Inst. 2017)

Development of a diagnostic tool for early detection of HPV-driven oropharyngeal cancer

Dr. Daniele Viarisio
© T. Waterboer

High risk human papillomaviruses (HR-HPV) cause carcinomas of the uterine cervix, the vulva, the vagina, the anus, the penis and the oropharynx. Patients with HPV-induced tumors frequently develop antibodies against the viral oncoprotein E6. Especially patients with HPV-induced oropharyngeal or anal carcinoma develop these antibodies years before tumor diagnosis1,2. In HPV infection and replication without malignant transformation, E6 is also expressed, but rarely induces a measurable antibody response. Thus, antibodies against HR-HPV E6 could be used as a prognostic diagnostic marker for HPV-induced oropharyngeal and anal carcinomas. We are currently developing an ELISA-based screening assay to detect the presence of specific HR-HPV E6 antibodies in human sera. The ELISA system offers a robust, cost-effective and easy to handle method that can be used in routine hospital-based laboratories. Since the conformation of purified HPV E6 proteins is highly unstable and quickly loses antigenic properties when immobilized for long periods, we are currently focusing on the improvement of assay stability over time (Figure).


1. Kreimer AR, Johansson M, Waterboer T, …, Pawlita M, Brennan P. Evaluation of human papillomavirus antibodies and risk of subsequent head and neck cancer. J Clin Oncol. 2013 Jul 20;31(21):2708-15.

2. Kreimer AR, Brennan P, Lang Kuhs KA, Waterboer T, … , Pawlita M, Johansson M. Human papillomavirus antibodies and future risk of anogenital cancer: a nested case-control study inthe European prospective investigation into cancer and nutrition study. J ClinOncol. 2015 Mar 10;33(8):877-84.

© Dr. Daniele Viarisio

Figure: When preserved in the storing buffer that we are currently developing, the immobilized HPV E6 proteins do not show a measurable loss in antigenic reactivity after 16 weeks of storage, the stabilized plates are comparable to freshly prepared ones. Ongoing experiments suggest that the storage time can be further extended.

Bacterial infections associated with cancer: Whole-proteome microarrays for antigenic target identification

Katrin Hufnagel
© T. Waterboer

Published whole-proteome microarrays to systematically investigate the humoral immune response to bacterial infectious agents were so far based on individual cloning, expression and purification of hundreds or thousands of bacterial open reading frames. To overcome this limitation, and to maintain at the same time the advantages of slide-based microarrays, we have developed a novel method to generate microarrays representing an entire bacterial proteome, using Chlamydia trachomatis (Ct) as a complex model organism. It is based on a combination of multiple spotting technique and on-chip in situ protein expression. Expression constructs for each individual bacterial protein were generated by two successive PCRs using bacterial genomic DNA as a template, and proteins were expressed directly on microarray slides. We performed proteome immunoassays by incubating clinically characterized sera of Ct infected patients onto the whole-proteome arrays, and were able to provide evidence of antibody reactivity patterns that represent either Ct exposure markers, markers associated with persistent infections or Ct associated cervical cancer markers (Figure). De novo identified antigens were validated in a large population-based cervical cancer case control study using high-throughput suspension bead array serology. The combination of high-density, low-throughput proteome immunoassays and low-density, high-throughput suspension bead array serology is a powerful platform for the discovery of exposure and disease-specific biomarkers and can be easily adapted to other microorganisms in all areas of infection research as well as in e.g. autoantibody screening and epitope mapping.




1. Hufnagel K, Lueong S, ..., Hoheisel JD, Waterboer T. Immunoprofiling of Chlamydia trachomatis using whole-proteome microarrays generated by on-chip in situ expression. May 2018, Scientific Reports.

2. Hufnagel K, Reininger D, ..., Waterboer T, Hoheisel JD. In situ, Cell-free Protein Expression on Microarrays and Their Use for the Detection of Immune Responses, Feb 2019, bio-protocol.

3. Hufnagel K, Hoenderboom B , ..., Morré SA, Waterboer T. Chlamydia trachomatis Whole-Proteome Microarray Analysis of The Netherlands Chlamydia Cohort Study, Dez 2019, microorganisms.


Results of Proteome Immunoassays (PIA) using pools of five sera

© K. Hufnagel

Figure: In total 897 Ct proteins were spotted on one array. Sera from uninfected women did not show positive signals with any of the Ct proteins, while all serum pools of Ct-infected women revealed positive signals. Comparison of the antibody reactivity patterns identified antigens which reacted with all Ct seropositive pools but also antigens which reacted only with serum pools from cancer patients. Some antigens reacted only with pools of sera from women of higher age, possibly indicating persistent Ct infections. Examples for antigens associated with general Ct infections, potentially persistent infections and cervical cancer are highlighted in green, yellow, and blue boxes, respectively.

Serology of gastrointestinal bacteria and colorectal cancer risk

Dr. Julia Butt
© T. Waterboer

Colorectal cancer (CRC) is the third most common cancer worldwide and the fourth leading cause of cancer death. Apart from well-known risk factors like age, male sex, family history of CRC or obesity, research of the recent decade focused on the relation of CRC and bacterial infections. Studies addressing more specifically individual bacterial species identified Streptococcus gallolyticus subsp. gallolyticus (S. gallolyticus) and Fusobacterium nucleatum (F. nucleatum) of being associated with CRC. Moreover, the gastric cancer associated bacterium Helicobacter pylori (H. pylori) has been implicated in CRC development. There is a lack of prospective data for the observed associations, which is relevant for the identification of risk markers for CRC development. We aim to fill this knowledge gap by applying serological assays in CRC case-control studies nested within large prospective cohorts (Figure 1). Our group developed multiplex serology assays for the detection of H. pylori, S. gallolyticus and F. nucleatum, and assessed the association of antibody responses to these bacteria in collaboration with large epidemiological studies including the European Prospective Investigation into Nutrition and Cancer (EPIC) and a consortium of 10 prospective cohorts in the United States.1, 2, 3, 4

We are currently working on expanding the panel of bacterial species we can detect, as well as enabling the distinction between different classes of immunoglobulins mounted against the bacterial infection. Also, we attempt to include other potentially CRC risk associated blood markers, including auto-antigens and inflammatory markers. We are moreover interested in expanding our serological biomarker studies to young-onset gastric cancer, with a focus on H. pylori infection, autoimmune gastritis and pepsinogens I and II as a marker for resulting chronic atrophic gastritis.



1. Butt J, Jenab M, Willhauck-Fleckenstein M, et al. Prospective evaluation of antibody response to Streptococcus gallolyticus and risk of colorectal cancer. Int J Cancer. 2018;143(2):245–252. doi:10.1002/ijc.31283

2. Butt J, Blot WJ, Teras LR, et al. Antibody Responses to Streptococcus Gallolyticus Subspecies Gallolyticus Proteins in a Large Prospective Colorectal Cancer Cohort Consortium. Cancer Epidemiol Biomarkers Prev. 2018;27(10):1186–1194. doi:10.1158/1055-9965.EPI-18-0249

3. Butt J, Varga MG, Blot WJ, et al. Serologic Response to Helicobacter pylori Proteins Associated With Risk of Colorectal Cancer Among Diverse Populations in the United States. Gastroenterology. 2019;156(1):175–186.e2. doi:10.1053/j.gastro.2018.09.054

4. Butt J, Jenab M, Pawlita M, et al. Antibody Responses to Fusobacterium nucleatum Proteins in Prediagnostic Blood Samples are not Associated with Risk of Developing Colorectal Cancer. Cancer Epidemiol Biomarkers Prev. 2019;28(9):1552–1555. doi:10.1158/1055-9965.EPI-19-0313

© J. Butt

Figure: Multiplex serology of gastrointestinal bacteria to assess association with CRC risk. Antibodies to SGG and FN are detected using multiplex serology in case-control, as well as prospective studies to analyze their possible association with risk of developing CRC.

Determination of epidemiological and genetic risk factors for HPV16 E6 seropositivity in the healthy population

Nicole Brenner
© T. Waterboer

Antibodies against the oncoprotein E6 of human papillomavirus type 16 (HPV16) are considered highly sensitive and specific biomarkers for HPV16-driven oropharyngeal cancer1,2. However, approximately 0.7 % of healthy controls in previous nested case-control studies were also observed to have antibodies against HPV16 E61,2,3. To evaluate a potential application of HPV16 E6 antibodies in a future serology-based screening setting for HPV16-driven oropharyngeal cancer, the presence of antibodies against HPV16 E6 in the healthy population needs further investigation. Thus, the project aims at determining epidemiological risk factors for HPV16 E6 seropositivity and their biological meaning in the healthy population.

In four large European studies (UK Biobank, EPIC Norfolk, BGS-98, Swedish MS study) HPV16 E6 antibodies were measured by Multiplex Serology4. Risk factors for HPV16 E6 seropositivity are assessed by classical sero-epidemiology and genetic epidemiology (HLA allele analysis, GWAS see Figure; Brenner et al. in preparation). The majority of the analyzed studies are population-based and help to determine HPV16 E6 seroprevalence estimates stratified by gender and age in different European countries. This project provides insights into potential high-risk populations for HPV16 E6 seropositivity and HPV16-driven cancer. These high-risk populations may benefit from a future serology-based oropharyngeal cancer screening.  

1. A. Kreimer et al., Evaluation of human papillomavirus antibodies and risk of subsequent head and neck cancer, J. Clin. Oncol. 2013

2. A. Kreimer et al., Kinetics of the Human Papillomavirus Type 16 E6 Antibody Response Prior to Oropharyngeal Cancer, J. Natl. Cancer Inst. 2017

3. K. Lang Kuhs et al., Human Papillomavirus 16 E6 Antibodies in Individuals without Diagnosed Cancer: A Pooled Analysis, Cancer Epidemiol. Biomarkers Prev. 2015

4. Waterboer et al., Multiplex human papillomavirus serology based on in situ-purified glutathione s-transferase fusion proteins, Clin. Chem. 2005



© D. Chen et al., Genome-wide association study of HPV seropositivity; Hum. Mol. Genet. 2011

Figure: In genome-wide association analyses (GWAS), manhattan plots are used to illustrate the significance of all tested SNPs in association with the trait of interest. For each SNP, the negative logarithm of the p-value (y-axis) is plotted by chromosome and position (x-axis). Chen et al. showed a genome-wide significant association (indicated by the black line) of SNP rs9357152 (chr. 6) with seropositivity against HPV type 8 L1 antigen.

De novo identification of serological H. pylori biomarkers linked to gastric cancer

Rima Jeske
© T. Waterboer

Infection with H. pylori is the leading risk factor for development of gastric cancer. However, only a minority of infected individuals experience symptoms and even fewer develop the neoplasia. So far, the virulence factor CagA is considered the most promising serological marker to stratify patients’ individual risk. In some countries however, seropositivity to CagA is very frequent overall and therefore not suitable to identify high-risk groups. To efficiently target these patients, new biomarkers are required.

We adapted our recently published method for Chlamydia trachomatis1 and displayed the whole-proteome of H. pylori strain 26695 on high-density microarrays. The combination of a multiple spotting technique with cell-free, on-chip protein expression led to the successful display of 90% of the H.pylori proteome (Fig. left)

As a pilot experiment, we analyzed sera of gastric cancer patients and their matched controls from a European case-control study on these arrays. Results confirmed existing biomarkers (like CagA) and also revealed new candidates.2 Their potential to discriminate between gastric cancer cases and controls will be evaluated in large sero-epidemiological studies using our high-throughput Luminex platform.

To further advance the H. pylori microarray, we will expand the displayed gene set by genes from additional H. pylori strains. Thereby, we want to create a single universal platform, which can be used to detect new biomarkers regardless of the analyzed cohort, population or outcome. Gastric cancer is considered a multi-factorial disease, which results from a mismatch of host genetics, environmental factors and also H. pylori strain differences.  


1. Hufnagel, K., et al., Immunoprofiling of Chlamydia trachomatis using whole-proteome microarrays generated by on-chip in situ expression. Sci Rep, 2018. 8(1): p. 7503.
2. Jeske, R., et al., Development of Helicobacter pylori Whole-Proteome Arrays and Identification of Serologic Biomarkers for Noncardia Gastric Cancer in the MCC-Spain Study. Cancer Epidem. Biomar., 2020 (29) (11) 2235-2242

© R. Jeske

FigureH. pylori microarray displaying the whole proteome of the strain 26695. Determination of on-chip expression by dual anti-tag staining. Green signals derive from the anti-V5 (N-terminal) signal, while red signals derive from the anti-6xHis (C-terminal) antibody. Merged signals appear yellow. Negative controls (n = 56) are distributed across the slide.

Epstein-Barr virus and the risk of nasopharyngeal carcinoma

Julia Simon
© T. Waterboer

Nasopharyngeal carcinoma (NPC) is a head and neck cancer, which is endemic in Southeast Asia, Southern China, North Africa and the Arctic. In endemic regions, NPCs are uniformly associated with Epstein-Barr virus (EBV) infection. Although the majority of the world’s population is infected with EBV, only a small fraction develops NPC. IgA antibodies against the viral capsid antigen (VCAp18) and Epstein–Barr nuclear antigen 1 (EBNA1) have been shown to be particularly frequent in NPC patients, rather than in healthy EBV-infected individuals, and have been investigated for NPC screening approaches in endemic areas for many decades.

Outside of endemic regions, e.g. in America and Europe, NPCs are rare and have been shown to be attributable to EBV infection in only ~65% of the cases. Human papillomavirus (HPV) infection and smoking have been shown to be additional risk factors for NPC development.

To characterize EBV antibodies in sera from NPC patients in both, endemic and low-incidence regions, we have adapted a novel antibody risk stratification signature for multiplex serology.1 The extended EBV antigen panel (n=13) allows a broad characterization of IgA and IgG antibodies, which was limited to antibodies against EBNA1, VCAp18 and early antigen diffuse (EA-D) in the last decades of EBV serology.

Research in low incidence regions is mainly focusing on NPC etiology, including EBV-positive, HPV-positive and EBV/HPV-negative NPCs, their characteristics and survival differences.2 In high incidence regions, we focus on serology with novel IgG markers. Those novel serum biomarkers, which are specific for EBV-positive NPC, and not just for EBV-infection, could improve future screening approaches in endemic regions and contribute to early diagnosis and prolonged survival of NPC patients. We aim to apply promising biomarkers from case/control studies in prospective studies, to evaluate the use for the prediction of incident NPCs.


1. Simon J., Liu Z., …, Hildesheim A., Waterboer T.: “Validation of an EBV Antibody Risk Stratification Signature for Nasopharyngeal Carcinoma using Multiplex serology”, J. Clin. Microbiol, 2020, 58 (5)

2. Simon J., Schroeder L., …, Ness A., Waterboer T.: “Epstein-Barr virus and human papillomavirus serum antibodies define the viral status of nasopharyngeal carcinoma in a low endemic country”, Int. J. Cancer, 2020, 147:461–471.

© J. Simon

Figure: Risk factors for nasopahryngeal carcinoma (NPC) development. Epstein-Barr virus and human papillomavirus are two among many other risk factors, contributing to NPC development.

RNA patterns of oncogenic HPV types: a new triage test in HPV-based cervical cancer precursor screening

Daniela Höfler
© T. Waterboer

Almost all cervical cancers are driven by human papillomavirus (HPV) infections. While primary prevention is based on vaccination against HPV, screening is an important secondary prevention strategy.

In January 2020, cervical cancer precursor screening in Germany will change: DNA testing of oncogenic HPV and cytological co-testing will be performed every three years in women above 35 years1. Since screening for HPV DNA has a low positive predictive value, an additional test is needed (i.e. a triage test) in order to discriminate women with mild lesions that are likely to regress spontaneously and thus need observation only, from women with advanced lesion that require colposcopy-directed biopsy and therapy (Figure). Changing the routine screening program will increase the need of triage tests.

One potential new triage test could be the detection of specific HPV RNA patterns. This includes the quantification of three HPV16 RNA transcripts by RT-qPCR (E6*I, E1^E4 and E1C) present at different expression levels in mild and advanced lesions2. Currently, we are validating this test in a large study.

While HPV16 is driving up to 55% of all cervical cancers, additional 26% are driven by the oncogenic HPV types 18, 31, 33, 35, 45, 52 and 58. The three informative RNA transcripts for discriminating mild and advanced lesions in HPV16 positive lesions were also identified and confirmed for these HPV types. We are currently developing RT-qPCRs for the quantification of all RNA transcripts. In patient samples, we will try to identify similar HPV RNA patterns and validate these patterns as triage test for non-HPV16-positive lesions.


1. Beschluss des Gemeinsamen Bundesausschusses über eine Änderung der Krebsfrüherkennungs-Richtlinie und eine Änderung der Richtlinie für organisierte Krebsfrüherkennungsprogramme: Programm zur Früherkennung von Zervixkarzinomen.

2. Höfler D, Böhmer G, von Wasielewski R, Neumann H, Halec G, Holzinger D, Dondog B, Gissmann L, Pawlita M, Schmitt M. HPV16 RNA patterns defined by novel high-throughput RT-qPCR as triage marker in HPV-based cervical cancer precursor screening. Gynecol Oncol. 2015;183(3):676-82.


© D. Höfler

Figure: HPV infects cervical cells and can lead to morphological changes histologically diagnosed as cervical intraepithelial neoplasia (CIN) grade 1 to grade 3 and cancer. Cytological diagnoses are low-grade squamous intraepithelial lesions (LSIL) or high-grade lesions (HSIL). Red and green arrows indicate progression and regression and the arrow size indicates probability of progression or regression. Transient/mild lesion need to be observed only while advanced lesions need treatment. While HPV DNA is present in infection and all later stages, a good triage test should detect advanced lesions only.

HPV as prognostic biomarker in head and neck cancers

Lea Schroeder
© T. Waterboer

Besides alcohol and tobacco, viral infections are an established risk factor for some head and neck cancers. Of these, the most frequent and rising subset is oropharyngeal cancer (OPC) driven by human papillomavirus (HPV), which represents a distinct clinical and molecular entity. Although frequently diagnosed at advanced stage, HPV-driven OPC are characterized by a better treatment response and survival compared to HPV-negative OPC.

We identified HPV antibody patterns present in patients with HPV-driven OPC and associated with favorable prognosis.1,2 These patterns were also detectable in patients with head and cancer of unknown primary with a similar prognostic effect.3 High sensitivity and specificity for HPV-driven cancers were determined by assessing the presence of HPV nucleic acids in tumor tissue of these patients.1,2,3

We have recently confirmed our findings in Head and Neck 5000, a large prospective clinical cohort from the UK (; Figure).4 It includes >1500 participants with OPC, about 1000 with oral cavity or laryngeal cancer, respectively, and also representative numbers of the rarer sites, like hypopharynx and salivary glands. This will enable us to study HPV antibodies and tissue markers in head and neck subsets for which the prevalence and prognostic impact of HPV is still under discussion. HPV16 is by far the most frequent type, but the role of other high-risk types is less clear.

While the detection of HPV RNA in tumor biopsies is the research gold standard for the identification of HPV-driven cancers, analysis of minimal-invasive liquid biopsies is of growing interest for clinical application. Thus, we are currently developing methods for the detection of cell-free HPV DNA for screening, diagnosis and monitoring of HPV-driven head and neck cancers.


1. Holzinger, D., G. Wichmann, ..., M. Pawlita and T. Waterboer (2017). "Sensitivity and specificity of antibodies against HPV16 E6 and other early proteins for the detection of HPV16-driven oropharyngeal squamous cell carcinoma." Int J Cancer 140(12): 2748-2757.
2. Broglie, M. A., W. Jochum, ..., M. Pawlita and D. Holzinger (2017). "Evaluation of type-specific antibodies to high risk-human papillomavirus (HPV) proteins in patients with oropharyngeal cancer." Oral Oncol 70: 43-50.
3. Schroeder, L., G. Wichmann, ..., T. Waterboer and D. Holzinger (2018). "Antibodies against human papillomaviruses as diagnostic and prognostic biomarker in patients with neck squamous cell carcinoma from unknown primary tumor." Int J Cancer 142(7): 1361-1368.
4. Schroeder, L., M. Pring, ..., T. Waterboer and A. R. Ness (2020). "HPV driven squamous cell head and neck cancer of unknown primary is likely to be HPV driven squamous cell oropharyngeal cancer." Oral Oncol 107: 104721.

© Schroeder, L., M. Pring, …, T. Waterboer and A. R. Ness (2020). "HPV driven squamous cell head and neck cancer of unknown primary is likely to be HPV driven squamous cell oropharyngeal cancer." Oral Oncol 107: 104721.

Figure: Kaplan Meier plots for Head and Neck 5000 participants with oropharyngeal cancer (OPC) and cancer of unknown primary (CUP) stratified by HPV status.

to top
powered by webEdition CMS