Transoral and Cervical Ultrasound of Patients With Suspected Oropharyngeal Cancer

Introduction The incidence of oropharyngeal cancer is rising globally. The diagnostic approach for patients with suspected oropharyngeal cancer includes clinical examination and tissue samples in the outpatient clinic (biopsies), followed by relevant cross-sectional imaging (computed tomography (CT), magnetic resonance imaging (MRI), and sometimes 18F-Fluorodeoxyglucose (FDG) positron emission tomography/CT (PET/CT)).

It can be challenging to distinguish small oropharyngeal cancers from benign unilateral tonsil hypertrophy by clinical examination and relevant cross-sectional imaging. Small tumors of the palatine- or lingual tonsils may appear as normal oropharyngeal lymphoid tissue in the clinic and on CT and MRI scans, and normal tonsillar asymmetry can resemble an oropharyngeal cancer. The enhanced metabolic activity of the lingual and palatine tonsils on a PET-CT can blur abnormal activity from a small tumor, and increased one-sided metabolic activity due to inflammation raises suspicion of malignancy. This results in expensive, and, in some cases, painful additional interventions such as biopsies, diagnostic surgeries, and additional cross-sectional imaging.

By far the most common sublocation of oropharyngeal cancers is the palatine and lingual tonsils. Advances in ultrasound (US) technology and smaller high-frequency (>15 MHz) transducers make the palatine and lingual tonsils more accessible for examination. This is done by using a transoral approach, where the transducer can reach the palatine and lingual tonsil directly with access via the oral cavity. Garset-Zamani et al. describe how transoral and cervical US of the palatine and lingual tonsils have a significantly higher diagnostic accuracy than clinical examination and MRI in patients with suspected oropharyngeal cancer.

Cervical US for oropharyngeal cancer diagnostics is demonstrated in several studies. Coquia et al. describe how cervical US can be used to visualize tumors in the lingual and palatine tonsils, advocating for its possibility to fill a gap in tumor characterization, adding information gathered by MR and CT.

US is a low-cost, easily accessible, radiation-free dynamic imaging modality. No previous randomized trials have investigated whether the addition of cervical and transoral US improves the diagnostic accuracy and has a clinical impact. Possible clinical benefits include more accurate initial diagnoses, improved biopsy accuracy, fewer unnecessary scans, fewer outpatient clinic visits, and fewer diagnostic surgeries. In summary, achieving a more precise diagnosis with fever interventions. Should this research prove successful, it could lead to the implementation of US of the palatine and lingual tonsils in the outpatient cancer clinic and substantial advancements in oropharyngeal cancer management.

Research question In patients referred to the outpatient cancer clinic suspected of oropharyngeal cancer, does a clinical examination including transoral and cervical ultrasound, compared with a standard clinical examination alone, improve diagnostic accuracy and have a clinical impact?

Methods

Study design A prospective multicenter randomized controlled trial investigates transoral and cervical US for detecting primary tumors in patients referred for suspicion of oropharyngeal cancer.

The second Surgeon-Performed Oropharyngeal Transoral UltraSonography (SPOT-US-2) is a randomized controlled study of patients seen in the outpatient cancer clinic on suspicion of oropharyngeal cancer. This study compares the standard diagnostic workup to the standard diagnostic workup supplemented by transoral and cervical ultrasound of the lingual and palatine tonsils.

Patients are included and examined at the Department of Otorhinolaryngology, Head and Neck Surgery, Rigshospitalet, Department of Otorhinolaryngology and maxillofacial Surgery, Køge, Department of Otolaryngology, North Zealand Hospital, Hillerød, and Department of Otorhinolaryngology, Head and Neck Surgery, Aarhus University Hospital.

In Denmark, patients are referred to the head and neck outpatient cancer clinic by their general practitioner, a private head and neck practitioner, or a doctor from another specialty when symptoms or objective findings suggest malignancy. All patients referred with a suspected malignancy in Denmark are entitled to a free diagnostic workup.

Patients aged 18 years or older referred to the outpatient cancer clinic on suspicion of oropharyngeal cancer can be included in the study. The suspicion of oropharyngeal cancer includes relevant symptoms, relevant objective findings, or an incidental finding of a possible oropharyngeal malignancy on a scan. Patients are excluded if they have previously had oropharyngeal cancer or radiation to the head and neck area. Patients will be screened for eligibility before their first clinical workup in the outpatient cancer clinic.

A selected group of ENT surgeons will perform the ultrasound scans. All surgeons participating in US scanning will receive verbal and hands-on education on US of the palatine and lingual tonsils.

To establish valid inclusion criteria and minimize bias, the investigators aim to screen all eligible patients before their inclusion. When screening, the investigators want to obtain information from the medical records of all patients suspected of oropharyngeal cancer.

Information from medical records include

• Central personal registry number
• Age
• Sex
• Clinical diagnosis: suspected oropharynx cancer. Based on what of the following:
• Tonsil asymmetry
• Visible tumor tissue
• Ulcus, erosion, exudative lesions
• NBI dots
• A firm tonsil palpated compared to the contralateral side
• Suspected pathologic lymph node on US of the neck
• A scan (MR, CT, PET/CT) suggesting oropharyngeal malignancy
• Patient symptoms
• Pain in the oropharynx
• A sense of increased amount of tissue in the oropharynx
• The given international classification of diseases (ICD) Code
• Included/ excluded.
• If not included reasons for this:
• Not offered inclusion.
• Did not wish to participate.
• Inclusion criteria are met.
• Final pathology diagnosis in freeze section, formalin, and cytology
• Histopathologic tumor type (keratinizing/non-keratinizing squamous cell carcinoma, salivary gland carcinoma, lymphoma, other types), HPV-type, and p16 status of the primary tumor
• Tumor location:
• Palatine tonsil left/right
• Lingual tonsil left/right
• Uvula
• Ventral soft palate
• Posterior and lateral pharyngeal walls at the oropharyngeal level
• Epiglottic vallecula
• Time from the patient is seen at the outpatient clinic until the beginning of treatment
• Number of standard image modalities ordered for diagnostics
• Symptomatology and time to admission
• Alcohol and tobacco consumption
• Imaging results (MRI, CT, PET/CT)
• TNM classification
• Treatment plan
• Previous cancer diagnosis

The tumor type will be specified as benign or malignant, and subtyped according to histological type. The final diagnosis will be the binary histopathology diagnosis, as the presence or absence of an oropharyngeal cancer. The location of the cancer, HPV, P16 status, and T-stage will be noted. All patients diagnosed with the absence of oropharyngeal cancer will have their medical charts reviewed six months after inclusion to confirm their initial benign diagnosis. Clinical T-stage in the control and intervention groups will be compared. The clinical T stage will be compared to MRI. The gold standard in T-stage evaluation is post-surgical histopathology or staging at MDT conferences for non-surgical patients.

Statistics Sample size: A non-randomized study by Garset-Zamani et al. found a diagnostic accuracy of 86% using transoral US compared to 68% by clinical examination. The power calculation is performed in R. The power is 80%, and the significance level is 0.05.

A sample size of 170 (85*2) patients is estimated. The inclusion period for 170 patients is set at three years, as patients will be included only by ENT surgeons with relevant transoral US experience.

The final analysis of the data will be conducted using RStudio. Descriptive statistics for the intervention and control groups will be generated and compared. Descriptive statistics for the medical records of patients screened but not included in the study will also be compared with those of patients included in the study. This includes sex, age, smoking, initial symptoms, reason for referral, final diagnosis, and cancer location. Measures of diagnostic accuracy include sensitivity, specificity, positive and negative predictive values, and overall accuracy. Sensitivity is defined as the true positives divided by the number of patients with a verified histopathologic tumor diagnosis. The true positives are defined as patients with verified histopathologic tumor diagnosis and US with tumor suspicion. Specificity is the number of true negatives divided by the number of patients evaluated to be cancer-free. True negatives are defined as US results in patients assessed as cancer-free without tumor suspicion at six months follow-up. Fisher's exact test will be used for categorical variables, including diagnostic accuracy, sensitivity, specificity, the number of correct diagnoses in the outpatient clinic, the number of patients with initial positive biopsies, the number of patients with standard imaging performed, and the number of patients with diagnostic surgery performed. A two-sample t-test will be used to compare the total number of biopsies performed, the tumor size, and the time from initial examination to confirmed diagnosis.

The investigators will perform a subgroup analysis of patients seen on suspicion of cancer in the palatine tonsils, the lingual tonsil, referrals due to an incidental finding on a scan, and tonsil asymmetry without other suspicious findings.

Ethics US is a non-invasive diagnostic imaging modality and is considered safe. Transoral US may cause momentary discomfort, which is reduced by local anesthetic spray. The study will be conducted in accordance with the Declaration of Helsinki. The protocol is approved by the Medical Ethics Committees (MREC), trial number 2501276.

Conclusion This study examines whether the implementation of transoral and cervical US of the palatine and lingual tonsils in patients with suspected oropharyngeal cancer will have a clinical impact. This concerns more accurate initial diagnoses, more correct biopsies, fewer unnecessary scans, fewer diagnostic surgeries and fewer visits to the outpatient clinic. A key strength is the randomized controlled trial design, which allows us to assess the effects of adding US to the diagnostic workup of patients with suspected oropharyngeal cancer.

If our hypothesis is confirmed, these findings could lead to significant improvements in the diagnostic workup of patients with suspected oropharyngeal cancer. This study can determine if transoral US should be implemented as a routine examination for these patients.