PATHOS: a phase II/III trial of risk-stratified, reduced intensity adjuvant treatment in patients undergoing transoral surgery for Human papillomavirus (HPV) positive oropharyngeal cancer

With the advent of minimally invasive techniques, such as Transoral Laser Microsurgery (TLM) and Transoral Robotic Surgery (TORS), there has been renewed interest in primary surgical treatment for OPSCC as these techniques result in less morbidity when compared to open surgery. There are as yet no prospective randomised data on TLM/TORS for OPSCC but retrospective studies have demonstrated excellent outcomes. A US study of 204 patients with stage III-IV OPSCC treated with TLM and neck dissection found rates of local control (LC), overall survival (OS) and disease-free survival (DFS) to be 97, 86 and 82 % respectively at 3 years, with HPV-positive OPSCC having even better outcomes [8]. Most patients had adjuvant treatment (RT/CRT), which increased toxicity. Single centre data on TLM in the UK is also encouraging. Data from Liverpool on 153 patients with T1-T3 OPSCC (66 % HPV positive) treated with TLM and neck dissection demonstrate 3 year OS of 84.5 %, disease specific survival (DSS) of 91.7 % and DFS of 78.2 %. Patients with HPV-positive OPSCC had a 71 % reduction in the risk of death. 83.6 % of patients received adjuvant treatment. (TMJ, manuscript submitted). Single institution data also show a functional advantage with upfront transoral surgery compared to primary CRT [9]. A UK study compared swallowing function between 23 patients with locally advanced OPSCC treated with TLM +/- adjuvant therapy and 33 matched patients treated with CRT from a historical cohort and reported improved early swallowing function at 3 months in the upfront surgery group, using 3 different swallowing measures, including the MD Anderson Dysphagia Inventory (MDADI) score [10].

Currently, decisions about adjuvant therapy after surgery are based on the presence of pathological risk factors established more than 20 years ago in studies that included squamous cancers from multiple head and neck anatomical subsites and that did not test for tumour HPV status [11]. These risk factors include surgical margin status, presence of perineural and vascular invasion, number of lymph node metastases and presence of extracapsular spread (ECS) of nodal disease. The relevance of these risk factors in HPV-positive disease has been questioned and the optimum adjuvant treatment protocols for HPV-positive OPSCC are yet to be determined [12].

Adjuvant RT after surgery for advanced head and neck cancers improved LC in the RTOG 73-03 trial [13]. Subsequent studies recommended a minimum dose of 57.6Gy to the primary site and involved nodal areas and doses of up to 63Gy to areas of ECS [14]. However, lower doses of adjuvant RT may be sufficient for HPV-positive OPSCC based on the following: (i) observations that HPV-positive cell lines show increased radiosensitivity compared to HPV-negative cell lines in vitro [15, 16]; (ii) phase II data (ECOG 1308) showing equivalent LC rates at 2 years with reduced dose RT (54Gy in 27 × 2Gy fractions) in patients with HPV-positive OPSCC who achieved a complete response after 3 cycles of induction chemotherapy [17]; (iii) use of lower doses of prophylactic RT (50Gy in 35 × 1.4Gy fractions, equivalent to 43Gy in 2Gy fractions) with no increase in recurrences [18].

The EORTC 22931 and RTOG 9501 Randomised Controlled Trials (RCTs) showed that adjuvant CRT improved LC and DFS compared with adjuvant RT alone in some patients with advanced head and neck cancer [19, 20]. When the results of both studies were pooled, adjuvant CRT significantly improved OS in patients with positive (or ‘involved’) surgical margins and/or presence of nodal ECS [21]. These pathological features are now widely used criteria for adjuvant CRT. There is variation in practice however, particularly around the issue of surgical margins with positive (<1 mm) and close (1-5 mm) margins being grouped together in some studies. A survey of clinical oncologists from 17 UK centres reported that all would recommend adjuvant CRT for positive surgical margins (<1 mm) whilst only 30 % would advocate its use for close margins (1–5 mm) and 88 % for ECS [22]. The relevance of the EORTC 22931 and RTOG 9501 study results should be questioned in the context of HPV-positive OPSCC. Patients in these studies had tumours from multiple head and neck anatomical subsites and the prevalence rate of HPV-positive OPSCC would have been significantly lower than the current rate [1]. Evidence for using adjuvant CRT after transoral surgery for OPSCC is also lacking. A retrospective study of TLM for advanced OPSCC showed that the addition of chemotherapy to adjuvant RT did not improve outcomes even in the presence of ECS, possibly because a high proportion of patients in that study had HPV-positive tumours with already excellent outcomes [8, 12].

There are several factors that may contribute to dysphagia after treatment. Dysphagia after RT has been shown to correlate with increased mean doses of radiation to swallowing-related organs, with a higher mean dose to superior pharyngeal constrictor muscle region and larynx being particularly associated with worse long-term swallowing outcomes [23, 24]. The sigmoidal shape of the normal tissue complication probability curve indicates that increasing mean dose to pharyngeal musculature between 50Gy and 60Gy is a critical point at which risk of long-term dysphagia begins to inflect upwards, suggesting that reducing radiation dose from 60Gy to 50Gy could impart a clinically significant improvement in long-term swallowing outcomes [23]. Adding concurrent chemotherapy to primary or adjuvant RT increases the risk of dysphagia: a systematic review of TORS for OPSCC showed clear demarcation in swallowing outcomes across a variety of outcome measures in patients who received adjuvant RT alone compared to adjuvant CRT [25].

PATHOS is a multicentre, open label, parallel group phase II/III RCT funded by Cancer Research UK (CRUK). Patients must have biopsy proven OPSCC and should be clinically staged T1–T3 N0-N2b M0. Their primary tumour must be considered resectable via a transoral approach as determined by the local Multidisciplinary Team (MDT). HPV-positivity will be confirmed by central testing of diagnostic biopsy specimens by p16 immunohistochemistry and high risk HPV in-situ hybridisation. Synchronous neck dissection will be undertaken as per standard protocols depending upon the volume of regional metastatic disease. Following surgery, patients will be allocated into study groups based on histological findings (Fig. 1):

Group A: Patients whose tumours have no adverse histological features will not receive any adjuvant treatment as per standard of care.

Group B: Patients with T3 tumours (or T1–T2 tumours with additional risk factors), N2a (metastasis in single ipsilateral node 31–60 mm diameter) or N2b (metastasis in multiple ipsilateral nodes <61 mm diameter), tumours with evidence of perineural and/or vascular invasion, or close margins (1–5 mm) around the primary tumour specimen but with negative marginal biopsies and no evidence of cervical lymph node ECS. Patients in this group will be randomised to either post-operative RT 60Gy in 30 fractions over 6 weeks (Control Arm B1) or post-operative RT 50Gy in 25 fractions over 5 weeks (Test Arm B2).

Group C: Patients with tumours of any T or any N stage with the following high risk pathological features: positive (<1 mm) margins around the primary tumour (but with negative marginal biopsies) and/or evidence of cervical lymph node ECS. Patients in this group will be randomised to either post-operative CRT 60Gy in 30 fractions over 6 weeks with concurrent cisplatin (Control Arm C1) or post-operative RT 60Gy in 30 fractions over 6 weeks without chemotherapy (Test Arm C2).

The phase II trial will recruit at over approximately 20 sites in the UK. Participating sites will be required to complete a registration form to confirm that they have adequate resources and experience to conduct the trial. The planned phase III trial will recruit across Europe.

Participants are eligible to enter the trial pre-operatively if they meet all of the inclusion criteria and none of the exclusion criteria (Table 1). Post-operatively, patients allocated to Groups B and C on the basis of their pathology must re-confirm their consent for the study and will be assessed for their suitability for adjuvant treatment. Patients in Group B must be fit to undergo RT. Patients in Group C must be fit to undergo CRT and meet additional criteria as per Table 2.

Patients in Groups B and C will be randomised to a trial arm using the method of minimisation with a random element. Randomisation will be performed centrally by the Wales Cancer Trials Unit (WCTU). Participants will be stratified prior to randomisation by T stage, N stage, smoking history and treating centre.

Comprehensive assessment of swallowing function requires a multidimensional panel of measures that incorporates instrumental examination of swallowing along with clinician-rated and patient reported outcomes. A functional outcomes panel for assessing swallowing function has been developed for PATHOS (Table 3). The following assessments will be conducted prior to surgery, 4 weeks post surgery and at 4 weeks, 6, 12 and 24 months post treatment: (1) MD Anderson Dysphagia Inventory score (MDADI); (2) Water swallow test (WST); (3) Performance Status Scale-Head and Neck (PSS-HN); (4) Quality of Life questionnaires (EORTC QLQ-C30 and EORTC QLQ-H&N35). Patients will also undergo a videofluoroscopy (VF) assessment prior to surgery, at 4 weeks post surgery and at 12 months post treatment. CTCAE Toxicity (v4.03) will be assessed weekly during and at the end of RT and at 4 weeks, 6, 12 and 24 months.

All surgeons will need to demonstrate evidence of suitable training in the procedures employed and/or an established surgical practice in the relevant techniques (TLM/TORS). They should have undertaken a minimum of 5 previous transoral resections for OPSCC. It is expected that over the duration of the trial, positive marginal biopsy rates for an individual surgeon will not exceed 10 %.

The Radiotherapy Quality Assurance (RTQA) programme for the trial will be coordinated by the National Radiotherapy Trials Quality Assurance group. A comprehensive RTQA guidance document has been developed to accompany the main trial protocol. In brief, this will consist of pre-accrual and on-trial components. Each site must perform a pre-accrual outlining benchmark case on one lateralised and one non-lateralised case. Sites may need to complete a pre-accrual planning exercise of a benchmark case, depending on participation in other national head and neck trials. Real time review of the first lateralised and non-lateralised patients recruited by each centre will be carried out before treatment starts, both for outlining and planning.

All swallow assessments will be conducted by speech and language therapists with the required level of competency, or appropriately trained research nurses. DVDs of the VFs will be assessed centrally by members of the research team to QA the functional endpoint data.

The trial is associated with a CR-UK funded bioresource collection – PATHOS-T. Accordingly, up to five geographically distinct biopsies from the primary tumour will be harvested prior to surgical resection. In addition, up to two samples of involved cervical lymph node tissue will also be collected. Blood samples for research will also be taken before treatment and at 6 weeks, 6, 12, 18 and 24 months post treatment. Trial participants will be asked for additional optional consent to participate in this aspect of the study.

PATHOS has ethical approval from the Wales Research Ethics Committee which is legally recognised by the UK Ethics Committee Authority for review and approval. It also has approval from the Medicines and Health Care Product Regulatory Agency to be conducted in the UK. The Wales Cancer Trials Unit, a CRUK core funded and UK Clinical Research Collaboration accredited Clinical Trials Unit, is coordinating the trial. Velindre NHS Trust is the sponsor for the trial. A Trial Steering Committee and an Independent Data Monitoring Committee has been set up to monitor the progress and safety of the study. The PATHOS Trial Management Group, including clinicians, clinical trial unit staff, patient representatives, nursing and pharmacy representatives carry out the day-to-day running of the trial. PATHOS is registered with ClinicalTrials.gov identifier: NCT02215265.