Introduction
In head and neck cancer (HNC) there is a 5 per cent incidence of synchronous primary cancer (SPC) in the upper aerodigestive tract (UADT).Reference Krishnatreya, Rahman, Kataki and Lahkar1–Reference Sawaf, Quereshy and Cabrera3 These most commonly occur in the head and neck, lungs and oesophagus.Reference Rennemo, Zätterström and Boysen4, Reference Strobel, Haerle, Stoeckli, Schrank, Soyka and Veit-Haibach5 Traditionally in the work up of HNC, patients would have a panendoscopy including laryngoscopy, oesophagoscopy and bronchoscopy, to screen patients for a UADT SPC.Reference Parker and Hill6 Cross-sectional computed tomography (CT) imaging of the thorax, has become routine practice in staging patients with HNC.Reference Houghton, Hughes, Garvey, Beasley, Hamilton and Gerlinger7 In recent years, bronchoscopy has largely been excluded, given the high sensitivity of CT scanning for detecting lower respiratory tract malignancy.Reference Rennemo, Zätterström and Boysen4 Out-patient (OP) flexible nasoendoscopy (FNE) with or without narrow band imaging (NBI) is used in the evaluation of the UADT. NBI uses a green light filter to narrow the bandwidth of the light delivered from the endoscope. This wavelength is absorbed by haemoglobin and results in an enhancement of blood vessels and can demonstrate abnormal neo-vascularisation which may indicate malignancy.Reference De Vito, Meccariello and Vicini8 NBI used in the OP clinic setting has a high sensitivity (97 per cent) and specificity (92.5 per cent) for detecting laryngeal malignancy.Reference De Vito, Meccariello and Vicini8
The primary outcome of this paper was to assess the rate and method of detection for UADT SPC in HNC and the ongoing use of panendoscopy.
Methods
Local Caldicott application was submitted and approved. Research ethics were not required following consultation using the online tool from the National Health Service (NHS) Health Research Authority and Medical Research Council website.Reference Research Authority9 Patients were identified retrospectively through the Regional West of Scotland Head and Neck Multidisciplinary Team (MDT) Meeting. Information was gathered on the patients using electronic records. A database of all new patients, presented at the MDT between April 2020 and August 2022, was used. Patients were included if they had synchronous primary malignancies of the UADT at the time of diagnosis of HNC. Patients who had their head and neck primary identified during investigations for another primary cancer were excluded.
Results
A total of 2325 new patients were presented at the MDT between April 2020 and August 2022; 54 patients had synchronous UADT malignancies identified at the time of diagnosis. This resulted in a 2.3 per cent rate of synchronous UADT primary malignancy. Of these patients, 47 had their original cancer originating in the head and neck. The remaining six patients had an original diagnosis of lung or lower oesophageal malignancies, and their HNC was identified during further investigations. To assess the SPC assessment in the HNC pathway, these six patients were excluded from the analysis.
Table 1 demonstrates patient and cancer demographics. The majority of primary malignancies were identified in the oral cavity and oropharynx, 13 (27.7 per cent) and 12 (25.5 per cent) respectively. The most common sites of UADT synchronous primary malignancy were the lung (57.4 per cent) and oesophagus (23.4 per cent). Data were gathered about the OP clinical assessment, the use of FNE and imaging modality (Table 1). One hundred per cent (47) of patients had CT neck and thorax, and 66 per cent (31) had FNE at OP assessment. Thirty (63.8 per cent) patients had a panendoscopy. One patient had an upper oesophageal malignancy and a second synchronous oesophageal malignancy, both of which were identified during oesophago-gastro-duodenoscopy (OGD). This patient was excluded from pathway analysis.
Table 1. Demographics of patients with synchronous UADT primary identified during head and neck cancer pathway
CT = computed tomography; FNE = flexible nasoendoscopy; HNC = head and neck cancer; MDT = Multi-Disciplinary Team; OGD = oesophago-gastro-duodenoscopy; SD = standard deviation; UADT = upper aerodigestive tract.
SPC pathway analysis
Comprehensive OP HNC assessment was defined as clinical examination including FNE, imaging of the primary site, using CT or magnetic resonance imaging (MRI), and CT of the thorax. Method of SPC identification was compared for patients with comprehensive and incomplete OP HNC assessment (Table 2). In all patients with incomplete HNC assessment, FNE was not performed or documented. Overall, 97.9 per cent (46) of the UADT synchronous primaries were identified through OP clinical examination or imaging. Only one (2.1 per cent) patient, in the incomplete OP assessment group, had their synchronous primary identified during a general anaesthesia (GA) examination under anaesthetic. This patient had a left oropharyngeal malignancy identified in palatine tonsil at OP clinic. They did not have FNE and went on to have an examination under anaesthetic (EUA). The operative findings upstaged this tumour, which extended from the palatine tonsil to the epiglottis. A synchronous right tongue base malignancy was also identified during the EUA and this was classified as a synchronous primary at the MDT. Of the total 2325 patients in this study, this was the only patient with an SPC identified through GA examination (<0.001 per cent).
Table 2. Methods of detection for synchronous primary
GA = general anaesthesia; HNC = head and neck cancer; OP = out-patient; SPC = synchronous primary cancer.
Fisher exact test was used to determine statistical significance comparing comprehensive and incomplete HNC assessment, but this finding was not significant (0.36).
Discussion
Primary outcome
There was a 2.3 per cent rate of synchronous UADT primary malignancy in this retrospective cohort study of 2325 patients, lower than previous reported rates of synchronous primary malignancy in the UADT.Reference Coca-Pelaz, Rodrigo, Suárex, Nixon, Mäkitie and Sanabria10 Coca-Pelaz et al.Reference Coca-Pelaz, Rodrigo, Suárex, Nixon, Mäkitie and Sanabria10 carried out a systematic review to investigate the rate of second primary malignancy in HNC. This review included 61 articles published between 1979 and 2019, and results found a mean incidence of 5.3 per cent synchronous primary tumour. Over this 40-year period, the aetiology of HNC has changed, and this is shown in the lower rates of SPC in this article’s study population from 2020 to 2022. In this cohort, over half (25, 53.2 per cent) of the patients’ initial primary malignancy originated in the oral cavity or oropharynx. Rates of SPC have declined in this population, likely secondary to the rise in human papillomavirus (HPV) -related cancers.Reference Jain, Sikora, Baxi and Morris11 Slaughter et al.Reference Slaughter, Southwick and Smejkal12 introduced the concept of “field cancerisation”, that suggested a regional carcinogenic exposure to the mucosa of the UADT increases the risk of multiple malignancies within this area. Traditionally, in the UADT, carcinogenesis has been associated with smoking and alcohol.Reference Krishnatreya, Rahman, Kataki and Lahkar1 Our results suggest the risk of SPC has decreased, as the HPV-related HNC has risen.
Comprehensive out-patient head and neck cancer assessment
The authors have defined comprehensive OP HNC assessment as OP clinical examination with FNE and imaging. Fleming et al.Reference Fleming, Al-Radhi, Kurian and Mitchell13 compared the use of OP flexible endoscopy with rigid endoscopy under GA in patients with HNC. For the patients with OP flexible nasoendoscopy performed, there were no new malignancies identified on GA examination. In HPV-related oropharyngeal malignancy, SPCs are most likely to occur in the head and neck, rather than other areas of the UADT.Reference Jain, Sikora, Baxi and Morris11 The authors have found these mucosal surfaces are more amenable for OP examination, than the lung and oesophagus, through headlight inspection of the oral cavity and FNE. The results of this study demonstrate that comprehensive OP HNC assessment was effective in detecting all UADT SPCs, including those in the head and neck region. The only patient who had a new SPC identified under GA, did not have FNE used at OP assessment. This SPC was located in the tongue base, which is challenging to exam in the OP setting without the use of endoscopy. In all patients who had FNE and CT examination, the SPC was identified without the need for GA endoscopy.
Imaging for synchronous primary cancers
CT of the neck and thorax was performed in all patients. At the authors’ institution, this is the preferred method of staging and imaging assessment for SPC during initial work. In oropharyngeal malignancy, UK guidance recommends that MRI can improve staging and soft tissue assessment.Reference Homer, Winter, Abbey, Aga, Agrawal, Dafydd and Arunjit14 MRI may also provide superior cross-sectional assessment for SPCs in the head and neck region. Due to availability of MRI, this imaging modality is not routinely used in initial assessment at our institution, but is used when clinically indicated, such as streak artefact created by dental amalgam. In patients with suspected SPC detected on initial imaging, positron emission tomography (PET)–CT can be used to further delineate the aetiology and metabolic activity of the suspicious lesion. However, it is not routinely used for screening patients for SPCs at initial work-up.
The role of panendoscopy
In this cohort, 63.8 per cent (30) of patients had a panendoscopy, or examination under GA. When indicated, panendoscopy has an important role, in the diagnosis and assessment of HNC. EUA can provide valuable information in the assessment of tumours and aid in biopsy and planning for surgical resection. However, the authors would propose that there is a limited role of panendoscopy in the routine screening of HNC patients for UADT SPC, if they have had a comprehensive OP HNC assessment. The results of this study found 100 per cent (30) of patients with comprehensive HNC assessment had their SPC identified. Of the 2325 patients in this study, less than 0.001 per cent (1) of patients had a synchronous primary identified through GA examination.
For diagnostic purposes, some lesions will not be amenable to biopsy under local anaesthesia (LA) and require a GA to gain tissue diagnosis. In particular, sites like the glottis and subglottis can be challenging to biopsy in the OP setting. For sites that are easier to access at OP clinic, including the oral cavity and oropharynx, representative biopsies can be taken under LA. The results of this study found, despite the majority of primary malignancies originating from the oral cavity or oropharynx, 63.8 per cent of patients had a panendoscopy. This high rate of GA endoscopy in tumours that are potentially accessible in clinic, suggests the rate of panendoscopy may be reduced if clinic-based LA biopsy is utilised.
Lung malignancy was the most common synchronous primary and 100 per cent of these were detected through CT. Rigid bronchoscopy has fallen out of use and is no longer routinely included in panendoscopy. CT has a high sensitivity for detecting lung cancerReference Toyoda, Nakayama, Kusunoki, Iso and Suzuki15 and is the gold-standard method of imaging recommended by National Institute for Health and Care Excellence (NICE) for diagnosing lung cancer.16 In this study, synchronous oesophageal malignancies were identified through CT or OGD. Use of rigid oesophagoscopy under GA has traditionally been used to screen for upper oesophageal or pharyngeal malignancies. This procedure can carry an added risk of oesophageal perforation, and in our cohort of patients did not detect any new cancers. If a patient is symptomatic, or a suspicious area has been highlighted in imaging, rigid oeosphagoscopy can be used for further assessment.
Out-patient diagnostic head and neck cancer pathway
HNC pathway times are increasing and represent some of the longest delays in commencing treatment across all cancers.16, 17 Limited OP clinics, imaging capacity and access to theatres cannot accommodate the growing number of urgent suspected HNC referrals. In 2020, 9 per cent of all urgent suspected referrals were for HNC. NHS England have introduced a faster diagnosis standard (FDS) that outlines a 28-day best practice pathway from referral to diagnosis.18 This was in response to poor adherence to cancer pathway targets, with only 61 per cent of HNC patients meeting their 62-day target from referral to treatment between 2018 and 2020.17 The FDS includes LA biopsy at a one-stop clinic, and only advises EUA/panendoscopy/GA biopsy if required. LA pathways have been found to reduce HNC pathways, in comparison to those patients requiring GA.Reference Lim, Zahra, Moen and Montgomery19 In Scotland, the Optimal Diagnostic Pathway also promotes the use of LA pathway and recommends only utilising GA if required.20
A national survey was carried out to understand the current practice in the United Kingdom for investigating HNC and the use of OP local anaesthetic biopsy. Only 48 per cent of respondents to the survey reported that they would use oral forceps and channelled endoscopy under LA in OPs. Respondents were asked about disadvantages of LA biopsy, and 19 per cent reported they were concerned about missing an SPC.Reference Lim, Moen and Montgomery21 Despite national recommendations and growing evidence for the safety and efficacy of OP LA biopsy,Reference Lim, Rogers, Owusu-Ayim, Ranjan, Manickavasagam and Montgomery22 there is a hesitation to move towards an OP diagnostic HNC pathway. Ongoing concern for missing an SPC may contribute to the high rates of panendoscopy in HNC investigation.
• In head and neck cancer (HNC) there is a recognised risk of synchronous primary cancer (SPC), due to the common carcinogens involved in these cancers.
• Our cohort of new patients with HNC had a 2.3% rate of SPC.
• 63.8% (30) of patients had a panendoscopy.
• All patients who had flexible nasoendoscopy and cross-sectional imaging had their SPC detected without the need for a panendoscopy.
• Panendoscopy does not play a role in screening for SPC in HNC.
Conclusion
With the changing aetiology of HNC, the rate of SPC has decreased. This cohort study has found comprehensive OP HNC assessment, with FNE and imaging, is effective in the detection of UADT SPCs. Wider use of local anaesthetic diagnostic pathways may improve cancer waiting times, while reducing the requirement for theatre space and a general anaesthetic.
Competing interests
Authors declare no conflict of interest or financial support/funding.
Author contributions
All authors contributed to the conceptualisation, data curation and analysis, methodology and draft writing and editing of the paper.
