Novel plasma Human Papillomavirus DNA assay as a predictor of residual disease after chemo-radiotherapy for locally advanced head and neck cancer

We are planning a multicentre biological sample collection study to validate circulating human papilloma virus (HPV) DNA as a marker of residual disease following potentially curative treatment in HPV positive oropharyngeal (throat) cancer.
Human papilloma virus (or HPV) affects the skin and the mucosa. The mucosa is the moist membrane that lines the inside of certain body parts eg mouth and throat. HPV infection is very common and over 100 different types of HPV have been identified. Although HPV can increase the risk of developing some types of cancer, most people who have HPV won't develop cancer.
Approximately 2000 patients are diagnosed with oropharyngeal (throat) cancer caused by HPV infection every year in the UK. The standard treatment for this type of cancer is up-front chemo-radiotherapy followed by surgical removal of any cancer left behind. The surgery is undertaken based on the results of a scan known as an 18F-FDG PET-CT. However, of the patients who undergo surgery, only 20% have viable residual cancer on microscopic pathological examination.
Therefore these patients (300 every year) undergo unnecessary surgical procedures. Approximately 25% of patients proceeding to surgery have severe complications immediately after surgery and all of them suffer with significant permanent side effects such as pain, shoulder dysfunction with altered quality of life. Therefore a more reliable marker of presence of residual cancer after chemo-radiotherapy is required to guide management decisions and avoid unnecessary surgical procedures. Furthermore, patients who develop HPV related throat cancer are young (range 47-65 yrs. Average age 55) and due to the excellent treatment outcomes with the current treatments (>90% 5 year survival) are expected to carry the burden of treatment related toxicity for life.
Oropharyngeal cancers caused by HPV release material from the tumour (containing the HPV DNA) in to the blood stream. Presence of HPV DNA in the blood stream can serve as a potential marker of the residual (remaining) cancer. We have developed an ultra-sensitive and specific assay 'HPV-detect', which is a non-invasive way to measure the HPV DNA levels after chemo-radiotherapy using a simple blood test to help guide management decisions and avoid any unnecessary surgical procedures.
The HPV-detect test was developed and validated in a pilot study carried out by a team of researchers at the Institute of Cancer Research. The pilot study comprised test and blinded validation groups and demonstrated that the HPV-detect test had a sensitivity and specificity in excess of 90%. The research also showed that the plasma HPV DNA levels correlated with the pathological response (ie no active cancer cells present) as assessed following chemo-radiotherapy.
The current proposal is designed to validate the relevance and usefulness of the test in patient care (clinical utility) in a large prospective multi-centre study, in order to establish its potential to predict the absence of residual disease.
To prove specificity of at least 85% will require the recruitment of 143 HPV+ oropharyngeal cancer patients undergoing chemo-radiotherapy and 48 HPV negative patients as negative controls.
Plasma HPV DNA levels will be measured using HPV-detect and at 12 weeks following treatment will be correlated with the 18F-FDG PET-CT results. Laboratory tests will be blinded to HPV status and clinical data. In parallel with the validation study in collaboration with the NHR-Diagnostic Evidence Cooperative at Imperial College London, we will study barriers to adoption of the test and how to address these. We will also undertake a detailed health economic analysis to study the cost benefits of implementing the test in the UK healthcare system.

Following curative chemo-radiation (CRT), approx. 20-30% patients with locally advanced human papillomavirus related (positive) oropharyngeal cancer (HPV+OPC) undergo unnecessary neck dissection (ND) or biopsies based on results of post-CRT 18F-FDG PET-CT (PET-CT): the positive predictive value of which is sub-optimal. Therefore a more reliable marker of 'true' residual disease is required as a means of guiding management decisions. Our newly developed ultrasensitive and specific assay 'HPV-detect' can non-invasively measure the post-CRT plasma HPV DNA levels, and guide management decisions avoiding unnecessary surgical procedures. In a single centre prospective pilot study comprising test and blinded validation cohorts, we have demonstrated that HPV-detect has a sensitivity and specificity in excess of 90% in detecting HPV DNA in pre-CRT plasma, when correlated to tumour biopsy HPV status and that plasma HPV DNA levels at 3-months post-CRT correlate with pathological response. The current proposal is designed to validate the clinical utility of HPV-detect in a large prospective multi-centre study, in order to establish its potential to predict absence of residual disease. To prove specificity of 85% or above compared with PET-CT will require accrual of 143 HPV+OPC patients undergoing CRT; 48 HPV- patients will be recruited as negative controls. Plasma HPV DNA levels will be measured using HPV-detect and at 12 weeks post-CRT will be correlated with the PET-CT. Following the validation study, if HPV-detect demonstrates a specificity of 85% or above we will design a phase III randomised study comparing PET-CT versus an HPV-detect guided treatment algorithm. In parallel, in collaboration with NIHR-Diagnostic Evidence Cooperative, we will
study stakeholder perceptions, barriers to adoption of the test and formulate mitigation strategies. In addition we will undertake an economic analysis to study the benefits of implementing the test in the UK healthcare system.

The principal beneficiaries of this research will be patients with HPV+OPC and the National Health Service. In the UK 2000 patients per year are diagnosed with HPV+OPC and this is projected to rise by approximately 230% over the next 20 years. Majority of these patients undergo CRT with curative intent and approximately 600 patients/year have a positive or equivocal 18F-FDG PET-CT scan post CRT. Approximately 75% of these will undergo neck dissection, with only 25% of those having viable tumour at pathology. Therefore 300 patients with HPV+OPC undergo unnecessary ND following curative treatment. Following clinical validation and integration of our assay into decision making algorithm, these patients will avoid further interventions. This will prevent patient anxiety, immediate and permanent side-effects related to the surgical procedures and the resulting deterioration in QOL. This group of patients are young (range 47-65 yrs.) without other medical conditions and are expected to have a near normal life expectancy following treatment (greater than 90% 5 year survival/cure rates). Avoiding surgery and subsequent toxicity will also help mitigate the adverse physical and psychological impact on the cancer survivorship experience.
Incorporation of HPV-detect into clinical practice will have a positive impact in terms of saving costs and better utilisation of surgical resources. Preliminary cost-effectiveness analyses that we have undertaken, demonstrate that preventing 300 patients from having a ND using our assay provides a cost saving of £5 million and 7.5 quality adjusted life years over a 5-year period. We have assumed that all HPV+OPC patients undergoing CRT will require HPV-detect assay at baseline and 3 months. With the incidence of HPV+OPC predicted to rise by >200% over the next 15 years, this will mean significant cost savings for the NHS.