Head and Neck Squamous Cell Carcinoma of Unknown Primary: A Diagnostic Work-Up

Abstract

The current work-up of the primary tumor site of a head and neck squamous cell carcinoma of unknown primary is not standardized and results in several time-consuming procedures that delay treatment initiation. This article seeks to consolidate contemporary strategies used to identify the primary tumor site of an unknown primary head and neck squamous cell carcinoma and offer recommendations based on current literature review.

Current work-up procedures for head and neck squamous cell carcinoma of unknown primary can result delayed treatment. This article consolidates contemporary strategies used to identify the primary tumor site of an unknown primary head and neck squamous cell carcinoma and offers recommendations based on current literature review.

Implications for Practice.

Unknown primary head and neck cancer refers to patients who present with neck lymph node metastases, for which the primary tumor site is not identified after comprehensive diagnostic work-up. The present day sequential diagnostic work-up of an unknown primary head and neck cancer remains controversial and results in several time-consuming procedures that delay treatment initiation. Due to the increased variation in diagnostic methodologies and the consequent lack of standardization, this article seeks to consolidate contemporary strategies used to diagnose the primary tumor site of patients with unknown primary squamous cell carcinoma of the head and neck and offer recommendations based on current literature review.

Introduction

Metastatic squamous cell carcinoma of an unknown primary accounts for less than 5% of tumors within the head and neck region.¹ Squamous cell carcinoma of an unknown primary in this region typically refers to patients who present with cervical lymph node metastases upon physical exam, for which the primary site cannot be ascertained despite comprehensive diagnostic work-up. Identification of the primary tumor site allows for more precise and conformal therapy, reducing the morbidity associated with wider-volume radiation therapy required to cover all head and neck mucosal sites that could harbor the hidden primary tumor in those with an unknown primary head and neck squamous cell carcinoma.² The more specific or targeted treatment for those whose primary site is identified further results in a higher probability of tumor control since failure to include the primary tumor in the treatment fields can result in local failure. Finally, identification of the primary site also allows for increased attention to the relevant anatomy during post-treatment surveillance.

The present day work-up of an unknown primary head and neck squamous cell carcinoma results in several time-consuming procedures that delay treatment initiation, which likely increase the risk of disease progression and morbidity.³ In addition, it remains controversial how to best sequence these diagnostic procedures. An example of such a diagnostic algorithm includes a thorough clinical exam with in-office nasopharyngolaryngoscopy, imaging studies such as magnetic resonance imaging (MRI), computed tomography (CT) and 18-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT), and panendoscopy with biopsy of the abnormal findings seen on exam and/or imaging, or directed biopsies of the mucosal sites likely to harbor the hidden primary tumor sites (base of tongue, tonsil, nasopharynx and/or hypopharynx).⁴ Despite this extensive workup, over 40% of primary tumors remain unidentified.^1,5

Innovative surgical techniques such as transoral laser microsurgery and transoral robotic surgery have been increasingly implemented in recent years in lieu of directed biopsies given its superior detection of the primary tumor site, especially in human papillomavirus (HPV)-related head and neck squamous cell carcinoma of unknown primary in which patients usually present with small primary tumors and large, cystic, neck lymph node metastases.^6,7

Due to the increased variation in diagnostic methodologies and the consequent lack of standardization, this article seeks to consolidate contemporary strategies used to diagnose the primary tumor sites of patients with unknown primary squamous cell carcinoma of the head and neck and offer recommendations based on current literature review (Table 1).

Table 1.

Key components of the diagnostic work-up of an unknown primary head and neck squamous cell carcinoma.

Noninvasive work-up	• History and physical examination, including in-office flexible nasopharyngolaryngoscopy with NBI • Imaging studies ◦ CECT/MRI ◦ FDG-PET/CT
Invasive work-up	• Exam under anesthesia with panendoscopy and biopsies ◦ Directed biopsy of abnormal findings ◦ Transoral robotic surgery ◦ Transoral laser microsurgery
Pathologic tissue evaluation	• p16 status • Human papillomavirus status • Epstein-Barr virus status

Abbreviations: NBI: narrow band imaging; CECT: contrast enhanced computed tomography; MRI: magnetic resonance imaging; FDG-PET/CT: fluorodeoxyglucose-positron emission tomography/computed tomography.

Identification of Primary Tumor Site

History and Clinical Examination

The diagnostic algorithm for patients presenting with a cervical lymph node mass is driven by 2 objectives: (1) determine the histology of the nodal mass and (2) identify the primary tumor site.⁸ Thus, screening begins with a comprehensive history, noting for symptoms such as dysphagia, odynophagia, hoarseness, otalgia, nasal obstruction, and epistaxis; the history of tobacco/alcohol use; sexual history, in particular of multiple sexual partners and the practice of oral sex, and of previous malignancy.⁹ As an example, a young man with no history of tobacco or heavy alcohol use, who presents with a high-risk HPV positive, unknown primary head and neck squamous cell carcinoma in an anterior level II/III cervical lymph node will likely have its primary squamous cell carcinoma hidden in the oropharynx, particularly in the palatine or lingual tonsils.

The history is typically followed by a complete physical examination of the head and neck, looking and feeling for mucosal abnormalities and skin lesions, and a flexible fiberoptic endoscopy of the upper aerodigestive tract (nasopharyngolaryngoscopy).^9,10

Flexible endoscopy with narrow band imaging can complement the in-office fiberoptic exam, since it can better distinguish malignant lesions from normal mucosa by selecting for wavelengths of light that augment visualization of the microvasculature. Davaris et al demonstrated that narrow band imaging has a better sensitivity (93.3% vs. 77.0%, P < .05) and accuracy (96.3% vs. 92%, P = .001) in the identification of laryngeal malignancies when compared to flexible endoscopy with traditional white-light endoscopy.¹¹ Narrow band imaging has also been shown to identify the primary tumor site in cases of unknown primary head and neck squamous cell carcinoma where the standard diagnostic workup was ineffective with an overall detection rate of 35% (99% CI, 0.18-0.53), a sensitivity of 83% (99% CI, 0.54-0.95), and specificity 88% (99% CI, 0.55-0.97).¹² Furthermore, narrow band imaging endoscopy can also be performed transorally, in-office, and may complement the transnasal approach on the search of the unknown primary head and neck cancer. Ebisumoto et al reported a higher detection rate of oropharyngeal tumors with the combined approach, since the transnasal examination does not always provide sufficient visualization of the oropharynx.¹³

In addition, the location of the lymph node metastasis can also help identify the primary tumor site. For example, levels I-III cervical nodes metastasis often correlate with oral cavity and lip carcinomas, levels II-IV often correlates with oropharynx, hypopharynx, and larynx carcinomas, and level V with nasopharynx and cutaneous primaries, with the latter also presenting with intra-parotid lymph node metastasis.¹⁴

Neck Lymph Node Biopsy

After completing clinical examination of the head and neck, fine-needle aspiration cytology or core needle biopsy is needed to confirm the histology of the lymph node mass as squamous cell carcinoma and to determine the HPV status of the biopsied specimen.^15,16

Fine-Needle Aspiration Biopsy

Fine-needle aspiration cytology is preferable, since it is easier, safer, and cost effective, reserving core needle biopsy for those whose results are non-diagnostic.^5,17

Complications related to fine-needle aspiration cytology are rare and generally limited to vasovagal reactions and slight discomfort following aspiration. In 2008, Tandon et al published a systematic review and meta-analysis evaluating the diagnostic efficacy of fine-needle aspiration cytology, involving 30 studies and 3459 aspirates.¹⁸ This study reported a sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of 89.6%, 96.5%, 93.1%, 96.2%, and 90.3%, respectively, which were similar to those identified in their 10-year institutional review of 2702 head and neck aspirates, substantiating reports of fine-needle aspiration cytology’s reliability.¹⁸ Factors such as inadequate sampling, insufficient material, faulty interpretation, and lack of triage necessary for ancillary studies have long been known to weaken the diagnostic capacity of fine-needle aspiration cytology.^18,19 However, the recent disparities in fine-needle aspiration cytology efficacy may be explained by a dwindling number of cytopathologist-performed biopsies. In a comparative study of fine-needle aspiration cytology performed by cytopathologists versus clinicians, Wu et al demonstrated that 83% of cytopathologist-performed biopsies were diagnostic compared to 24% of clinicians-performed biopsies.²⁰ Ninety-four percent of cytopathologist-performed biopsies also indicated agreement with final surgical pathology in comparison to only 67% of clinicians-performed biopsies. A recent study examining the impact of Coronavirus disease 2019 (COVID-19) on pathologist, versus surgeon-performed fine-needle aspiration cytology demonstrated similar results to the study reported by Wu et al.²¹ Compared to the pre-COVID-19 era, COVID-19 lockdown required more surgeon-performed than pathologist-performed biopsies, and this change resulted in a corroborating increase in the fine-needle aspiration cytology nondiagnostic rate.²¹ Overall, studies have shown that fine-needle aspiration cytology reliability is improved when conducted in tandem by specialized physicians and experienced cytopathologists.

In the last half century, palpation-guided fine-needle aspiration cytology has been increasingly replaced with ultrasound-guided fine needle aspiration cytology due to its superior diagnostic ability.²² Specifically, ultrasound-guided fine-needle aspiration cytology is better able to target smaller and less-palpable lesions, and more accurately avoid neighboring anatomical structures. A systematic review published by Conrad et al revealed that ultrasound-guided fine-needle aspiration cytology produced lower non-diagnostic rates compared to palpation-guided fine needle aspiration cytology (6.7% vs. 20.7%, respectively).²² Ultrasound-guided fine-needle aspiration cytology also gave rise to significantly lower rates of discordance with final surgical pathology compared to palpation-guided fine needle aspiration cytology (5.4% vs. 12.8%, respectively).

Core Needle Biopsy

Core needle biopsy is a minimally invasive technique that has been increasingly utilized in cases of non-diagnostic or inconclusive fine-needle aspiration cytology. There are concerns that core needle biopsy tracts may increase the risk of extranodal disease spread or tumor cell seeding, and bleeding, but these risks are rare and related to the anatomical location and the size of the needle used for the biopsy. The need to obtain a definitive diagnosis far outweighs these potential concerns.^23-25

In 2011, Novoa et al published a systematic review and meta-analysis of 16 studies with a total of 1291 cervical lesions, demonstrating core needle biopsy’s exceptional ability in assessing lymphadenopathies.²⁶ Specifically, core needle biopsy displayed a 96% accuracy in its ability to identify malignancy in suspicious lymph nodes and provided a correct specific diagnosis in 87% of the cases. In samples histologically verified by excisional biopsy, core needle biopsy achieved greater accuracy (96% vs. 93%), specificity (99% vs. 96%), and negative predictive value (95% vs. 90%) compared to fine-needle aspiration cytology, although differences in sensitivity and positive predictive value were not statistically significant.²⁶ In addition, while core needle biopsy is typically used in conjunction with ultrasonography, a study reported by Ferreira et al determined that core needle biopsy can be performed independently in an outpatient setting, with an accuracy, sensitivity, and specificity of 94%, 92%, and 100%, respectively.²⁷ Despite these excellent results, ultrasonography-guided biopsy is still preferable, if available, to minimize the risk of complications.

Virus-Related Head and Neck Cancer

High-risk HPV positive squamous cell carcinoma of the oropharynx accounts for the majority of unknown primary squamous cell carcinoma, with common involvement of levels II/III cervical nodes, usually presenting as large cystic lymph nodes on imaging, and as non-keratinizing squamous cell carcinoma metastasis on biopsy.^1,5,28,29 Detection of high-risk HPV in the lymph node of the unknown primary head and neck squamous cell carcinoma aids in the localization of the primary tumor site, since it is often located in the palatine or lingual tonsillar crypts.^30,31 Although polymerase chain reaction for identification of high-risk HPV can be performed on fine-needle aspiration cytology specimen, a core needle biopsy is generally required for ribonucleic acid (RNA) in situ hybridization to definitively identify transcriptionally active HPV.

Although less common, high-risk HPV has also been identified in other head and neck mucosal sites such as the oral cavity, nasopharynx, hypopharynx, and larynx, so attention to those areas is recommended, particularly when the diagnostic work-up fails to reveal an oropharyngeal primary tumor.^32,33

The tumor suppressor p16 is often overexpressed in high-risk HPV tumors, so p16 evaluation by immunohistochemistry (IHC) is currently used as a surrogate marker for HPV-positivity in both the primary oropharyngeal squamous cell carcinoma site and the lymph node metastasis.^34,35 Of note, p16 can be overexpressed in other cancers such as in cutaneous squamous cell carcinoma and lung cancer, which can also present with neck lymph node metastases, so p16-positive unknown primary squamous cell carcinoma may not be entirely suitable as an independent marker of HPV-related oropharyngeal squamous cell carcinoma.^36-39 Ultimately, polymerase chain reaction (PCR)-based HPV deoxyribonucleic acid (DNA) testing in conjunction with p16 testing by IHC is a reliable method of determining HPV status in unknown primary head and neck squamous cell carcinoma.⁴⁰

Given submucosal and infiltrative characteristics of nasopharyngeal carcinoma, HPV-negative unknown primary head and neck carcinoma should also be screened for Epstein-Barr Virus (EBV) through EBV encoding region (EBER) in situ hybridization, as this may help localize the primary tumor to the nasopharynx.^41,42 Of note, a negative EBER testing does not exclude the nasopharynx as the primary tumor site, particularly in non-endemic areas, like the United States where the most common nasopharyngeal carcinoma is the keratinizing subtype (World Health Organization type 1) which is tobacco and alcohol-related.⁴³

Finally, there is limited data supporting the presence of an ultraviolet light-related tumor phenotype to distinguish cutaneous from mucosal head and neck squamous cell carcinoma using next generation tumor sequencing.⁴⁴ However, this has yet to enter widespread clinical practice.

Imaging

Contrast-enhanced CT (CECT) of the neck is usually the initial imaging modality in the diagnostic workup of unknown primary head and neck cancer given its ability to assess all tissues of the neck rapidly, its widespread availability, and its relatively lower cost in comparison to MRI or FDG-PET or FDG-PET/CT.^8,42,45 MRI has better soft-tissue contrast than CECT and can be useful for detection of small oropharyngeal and nasopharyngeal tumors. FDG-PET or FDG-PET/CT has been increasingly used in the detection of an unknown primary head and neck cancer, since its metabolic uptake through the glucose metabolism of different tissues can aid on the identification of the primary tumor site and reveal other sites of disease that could have been missed by the anatomic imaging modalities. In patients with significant dental artifacts caused by dental restorations that can obscure CECT and MRI images, FDG-PET/CT is preferred.

Rusthoven et al performed a comprehensive analysis of 16 studies with a total of 302 unknown primary patients with head and neck cancer evaluating the efficacy of FDG-PET in the detection of primary tumors after being undetected by conventional work-up, consisting of history and physical and CT and/or MRI of the head and neck in all studies, and panendoscopy in 10 studies. The detection rate of the primary site by FDG-PET was 24.5% (sensitivity 88.3%; specificity 74.9%; diagnostic accuracy 78.8%).⁴⁶ The tonsil accounted for the highest false-positive rate (39.3%) likely due to its physiologic uptake, while the highest false negative rate was for the base of tongue (sensitivity, 79.3%), likely due to the high baseline level of uptake and misinterpretations as a benign physiologic uptake. In addition, FDG-PET detected the primary below the clavicle in 24.3% and led to the detection of unrecognized metastatic disease in 27.1% of the patients (regional, 15.9%; distant, 11.2%) which affected treatment recommendations.

While FDG-PET alone is a validated technique in the diagnostic work-up of unknown primary head and neck cancer, FDG-PET/CT exhibited a considerably higher primary detection rate (36.8% vs. 15.4%; P = .033) and positive prediction rate (93.3% vs. 46.2%; P = .01) in the head and neck; this detection rate increased to 47.4% with the addition of panendoscopy.⁴⁷ The detection rates of second primary below the clavicles or distant metastasis were 18.4% for FDG-PET/CT and 20.5% for FDG-PET.

Waltonen et al performed a retrospective chart review of 183 patients diagnosed with unknown primary head and neck squamous cell carcinoma to assess the efficacy of different diagnostic modalities in detecting the occult primary site.¹ In general, directed biopsies of the nasopharynx, tonsil, base of tongue, and hypopharynx were performed if imaging and exam under anesthesia with panendoscopy failed to identify the primary site. The rate of successful identification of the primary tumor site was 9.6% (14 of 146 patients) for neck CT, 0% (0 of 13 patients) for neck MRI, 14.6% (6 of 41 patients) for FDG-PET, and 44.2% (23 of 52 patients) for FDG-PET/CT (P = .001). The rate of false negative results for CT was 34.9% (51 of 146 patients), for MRI was 38.4% (5 of 13 patients), for FDG-PET was 19.5% (8 of 41 patients), and for PET/CT was 15.3% (8 of 52 patients). The authors concluded that the highest yield in identifying the primary tumor sites was with the FDG-PET/CT plus panendoscopy with directed biopsies with or without tonsillectomy: 59.6% (31 of 52 patients).¹

In contrast, Cianchetti et al reported their experience with the diagnostic evaluation in 236 patients with unknown primary head and neck squamous cell carcinoma.⁵ The occult primary site was detected in 53.4% of the cases, including 21 of the 72 patients who had no suspicious findings on physical exam or imaging studies (CT and/or MRI). None of the 21 patients in their series had the primary detected solely on FDG-PET and FDG-PET/CT, suggesting that the incremental benefit of FDG-PET and FDG-PET/CT over physical examination and CT/MRI is limited. In addition, both FDG-PET and FDG-PET/CT had high false-positive rates in the base of tongue and tonsils, which were the 2 most likely sites for an occult primary cancer.⁵

Faisal et al reported the results of a multicenter, retrospective study, analyzing the value of different diagnostic methods in detecting the primary tumor site in 124 patients of carcinoma of unknown primary, and the impact of primary tumors on their clinical outcome.⁴ The diagnostic procedure comprised of nasopharyngoscopy, imaging, tonsillectomy, and panendoscopy. Despite extensive diagnostic work-up, the primary site remained unknown in 75.8%. The detection rates of CT, MRI, FDG-PET/CT, tonsillectomy, and panendoscopy were 10.1%, 4.8%, 6.5%, 14.9%, and 15.2% respectively, suggesting that the clinical examination in combination with tonsillectomy and panendoscopy was superior to imaging in detecting the primary tumor. Furthermore, the PET/CT was not superior to the conventional imaging in the detection of the primary tumor site.⁴

Based on the above data, the added value of the FDG-PET/CT in the identification of the primary tumor site is debatable, but recommended (if available), given its ability to identify other sites of disease that will likely impact on treatment recommendations. In addition, FDG-PET/CT should be performed prior to panendoscopy and mucosal biopsies to decrease the risk of false-positive results related to the inflammation caused by the invasive diagnostic procedure.

Recent studies have highlighted the usage of FDG-PET/MRI, as opposed to FDG-PET/CT, in the work-up of unknown primary head and neck cancers. In a study of 20 patients, Ruhlmann et al found that FDG-PET/MRI and FDG-PET/CT found comparable results in the identification of the primary (11/20) and diagnostic confidence (2.7 ± 0.5), with better evaluation of cervical lesions with FDG-PET/MRI and lung lesions with FDG-PET/CT.⁴⁸ FDG-PET/MRI also offers a considerably lower dose of ionizing radiation and may provide an advantage of FDG-PET/CT in long-term evaluation of patients, although its cost and availability limit its application.

Examination Under Anesthesia and Mucosal Biopsy

Panendoscopy (direct laryngoscopy, bronchoscopy, and esophagoscopy) examination under anesthesia is commonly performed in patients with neck lymph node metastasis whose primary site has not been identified through clinical exam and radiological studies.⁹ A retrospective study evaluating the utility of panendoscopy in the detection of the primary tumor site in a cohort of unknown primary head and neck squamous cell carcinoma patients, identified 32 out of 103 primaries (18 in the tonsil and 8 in the base of tongue), otherwise not discovered on FDG-PET/CT.⁴⁹ Examination under anesthesia is used to evaluate high-risk sites, such the nasal cavity, nasopharynx, oral cavity, oropharynx, larynx and hypopharynx with directed biopsies targeting sites with suspicious ulceration, color change, asymmetry, fullness, or irregular mucosal bleeding.^9,10

While random biopsies have traditionally been performed, they are no longer recommended with the advent of modern imaging due to low diagnostic yield.^50,51 However, in the event that examination under anesthesia with directed biopsies fail to yield positive results, ipsilateral tonsillectomy, and base of tongue mucosectomy (or lingual tonsillectomy) are recommended.^52-54 In a study of 236 patients, the tonsillar fossa and base of tongue were identified as the most common primary sites of unknown primary head and neck squamous cell carcinoma with a rate of 44.7% and 43.9%, respectively.⁵ In patients whose primary tumor site has not been identified on clinical exam and imaging studies, Cianchetti et al demonstrated that examination under anesthesia with directed biopsy and/or ipsilateral tonsillectomy identified the primary tumor site in 44.3% of patients.⁵ In addition, Waltonen et al found that ipsilateral tonsillectomy was superior to deep tonsillar biopsy (29.6% vs. 3.2%, P = .0002) in its ability to detect the occult primary site.⁵⁵ Similarly, Tanzler et al showed that within the palatine tonsils, a radical ipsilateral tonsillectomy improves the chances of identifying an occult tumor (39%) than biopsy alone (13%).⁵⁰ While the merits of ipsilateral versus bilateral tonsillectomy have been widely debated, recent studies are in favor of bilateral tonsillectomy, as there seems to be a small risk of an occult malignancy in the contralateral tonsil, and the morbidity associated with the procedure is not significantly increased.^10,56

Transoral robotic surgery and transoral laser microsurgery are minimally invasive techniques utilized in the diagnostic work-up of the unknown primary head and neck squamous cell carcinoma.^52-54,57 A survey completed by 359 otolaryngologists found that the superior visualization of the surgical field, improved quality of life-related outcomes, and shorter periods of hospitalization were the primary benefits of transoral robotic surgery.⁵⁸ While 74% of responders identified cost as a barrier, several studies found that transoral robotic surgery, as a method of both diagnosis and treatment, is cost-effective.^59,60

Fu et al reported that transoral robotic surgery and transoral laser microsurgery identified the primary in 67% of patients whom the clinical, radiological, and panendoscopy with directed biopsies examinations were otherwise negative.⁶¹ Similarly, Farooq et al reported that tongue base mucosectomy through transoral robotic surgery and transoral laser microsurgery identified a base of tongue primary cancer in 78% of patients who had negative CT/MRI imaging, FDG-PET/CT, and examination under anesthesia with tonsillectomy.⁶² More significantly, transoral robotic surgery and transoral laser microsurgery are superior in their ability to detect the primary tumor site in patients with HPV-positive unknown primary head and neck squamous cell carcinoma.⁵⁴ Graboyes et al reported a detection rate of 89% in an observational cohort study of 65 p16-positive, unknown primary head and neck squamous cell carcinoma.⁵⁴ Several meta-analyses have substantiated this finding, reporting high identification rates in p16-positive unknown primary head and neck squamous cell carcinoma, especially in the base of tongue.^6,63

Whether to perform the exam under anesthesia, and the palatine and lingual tonsillectomies concurrently or sequentially, is a matter of debate.⁶⁴ Although the concurrent approach may minimize treatment initiation delay, procedure related-complications may lead to prolonged hospital stay and an increase in overall costs.⁶⁵

While transoral robotic surgery and transoral laser microsurgery have a substantial impact on HPV-related unknown primary head and neck squamous cell carcinoma, they have a lower yield for the identification of the primary tumor site in the HPV-negative unknown primary head and neck squamous cell carcinoma. Specifically, Al-Lami et al, Kubik et al, and Ryan et al studies reported detection rates of 12%, 13%, and 26%, respectively in HPV-negative unknown primary head and neck squamous cell carcinoma, suggesting that these tumors are less likely to originate from the oropharynx.^66-68 Although transoral robotic surgery and transoral laser microsurgery typically have a low risk of complications, hemorrhage, postoperative dysphagia, and tongue swelling, and the potential need for tracheostomy and/or gastrostomy can happen.^61,62,67 Although, the use of transoral robotic surgery and transoral laser microsurgery may delay treatment initiation and exacerbate the cost-of-care in these patients, it is now considered the preferred invasive diagnostic method in the identification of the HPV-related unknown primary head and neck squamous cell carcinoma.⁶⁷

Conclusions

In summary, unknown primary head and neck squamous cell carcinoma accounts for less than 5% of all head and neck squamous cell carcinomas. The diagnostic work-up is highly variable among different institutions and despite a thorough diagnostic work-up, many patients with a metastatic lymph node to the neck end up with an unidentified primary tumor. Upon initial evaluation of a patient with an unknown primary head and neck squamous cell carcinoma, a good clinical history should be obtained, addressing the patient’s symptoms and risk factors; a thorough physical examination, looking and feeling for oral cavity, oropharynx and head and neck skin abnormalities, in addition to the flexible endoscopy exam in office, complemented by narrow band imaging (if available) are recommended. Following history and physical examination (Fig. 1), diagnostic imaging with a contrast-enhanced neck CT and a fusion FDG-PET/CT covering at least the areas from the mid-skull to the thigh can aid on the identification of not only the primary tumor site, but other regional and distant disease sites, which will affect treatment recommendations. A contrast-enhanced neck MRI should be considered in those who have contra-indications for the iodine contrast used on CT imaging, or for those whose CT images are affected by dental artifacts. Directed biopsy of the abnormal clinical or imaging findings is recommended for histologic confirmation. If no abnormal findings are seen or felt on clinical exam or seen on imaging studies, an exam under anesthesia with ipsilateral palatine and lingual tonsillectomy through transoral robotic surgery or transoral laser microsurgery are recommended, particularly if HPV or p16-positive unknown primary head and neck squamous cell carcinoma. If bilateral neck disease is identified on clinical exam or imaging studies, consideration can be made for palatine and lingual tonsillectomy of the site with more advanced neck nodal disease. Contralateral palatine or lingual tonsillectomy should be considered in those with abnormal findings felt or seen on clinical exam or imaging studies.

Figure 1.

Diagnostic algorithm of head and neck squamous cell carcinoma of unknown primary (HNSCCUP). Abbreviations: NBI: narrow band imaging; US: ultrasonography; FNAC: fine needle aspiration cytology; CNB: core needle biopsy; LN: lymph node; SCC: squamous cell carcinoma; HR-HPV: high risk-human papillomavirus; CECT: contrast enhanced computed tomography; MRI: magnetic resonance imaging; EUA: exam under anesthesia; TORS: transoral robotic surgery; TLM: transoral laser microsurgery. *If bilateral neck disease, consider palatine and lingual tonsil biopsies of the site with more advanced neck disease.

Finally, it is not always possible to identify the primary tumor site. However, it is only after the above diagnostic work-up has been completed that these patients can be appropriately diagnosed with unknown primary head and neck squamous cell carcinoma.

Funding

The author indicated no financial relationships.

Conflict of Interest

The authors indicated no financial relationships.

Author Contributions

Conception/design: G.R. Provision of study material or patients: All authors.

Collection and/or assembly of data: All authors. Data analysis and interpretation: All authors.

Manuscript writing and final approval of manuscript: All authors.

Data Availability

No new data were generated or analyzed in support of this research.