Introduction
During the course of human fetal development, arches of tissue in the neck develop which are analoguous to the membranes (gills) of fish. In fish, these aid the extraction of oxygen from water. In humans, these arches (branchial clefts, from the Latin branchiae, meaning gills) and their intervening clefts develop not into gills but into a variety of head and neck structures, including contributions to the neck, the jaw, and the middle and external ear.Reference Benson, Dalen, Mancuso, Kerr, Cacciarelli and Mafee1 In humans, a host of different developmental anomalies may be associated with these branchial clefts, including cysts, sinuses and fistulae. Chief among these are branchial cleft cysts.Reference Golledge and Ellis2, Reference Koeller, Alamo, Adair and Smirniotopoulos3 On physical examination, branchial cleft cysts usually manifest as rounded, non-tender masses situated along the anterior border of the sternocleidomastoid muscle between the clavicle and the anterior tragus of the ear. On light microscopy, these cysts are, in the vast majority of cases, lined by stratified squamous epithelium, with less frequent contributions by ciliated, respiratory-type epithelium. The subepithelial zone is typically occupied by lymphoid tissue, often with prominent (reactive) germinal centres.Reference Koeller, Alamo, Adair and Smirniotopoulos3
For over a century and a quarter, head and neck surgeons have both defended and attacked the notion of a squamous carcinoma arising in a pre-existing branchial cleft cyst.Reference von Volkmann4–Reference Girvigian, Rechdouni, Zeger, Segall, Rice and Petrovich20 This is more than a mere academic debate, as treatment of a localised carcinoma arising within a developmental cyst and still entirely confined within that cyst would differ from treatment of a metastatic deposit derived from a (presumably occult) head and neck primary tumour situated elsewhere.
As an aside, it is interesting to note that malignant lateral cervical cysts can be related to metastatic thyroid tumours or, more rarely, to primary thyroid carcinoma arising in branchial cleft cysts.Reference Lanzafame, Caltabiano, Puzzo and Cappellani21, Reference Kreft, Hansen and Kirkpatrick22 As with presumed branchiogenic tumours, it is important to evaluate whether such a tumour is a metastasis or a papillary thyroid carcinoma arising in ectopic thyroid tissue in a branchial cleft cyst, as the thyroid gland is derived in part from the two fourth branchial pouches. Immunohistochemical stains for thyroid transcription factor one, p63 and thyroglobulin have proven to be very useful tools to determine the differential diagnosis in such circumstances.Reference Lanzafame, Caltabiano, Puzzo and Cappellani21, Reference Kreft, Hansen and Kirkpatrick22
Historical perspective
The notion of a squamous carcinoma arising in the setting of a pre-existing developmental lesion – a branchial cleft cyst – was first advanced in the latter years of the nineteenth century.Reference von Volkmann4 Even these early reports warned about the possibility of an occult primary tumour elsewhere. At that time, important differential diagnostic possibilities included ‘lymphosarcoma, tuberculous lymphoma, and actinomycosis’.Reference Warner5 A diagnosis of branchiogenic carcinoma was not routinely made pre-operatively, but, rather, only after surgery or at post-mortem examination. The concept of primary branchiogenic carcinoma gained popularity at the turn of the nineteenth century, and became a well established doctrine.Reference Warner5, Reference Crile and Kearns6 Even in the early decades of the twentieth century, it was recognised that major salivary glands, the upper aerodigestive tract, and even the oesophagus and stomach might give rise to cervical metastases which could mimic primary branchiogenic carcinoma.
It was with the 1944 and 1950 publications of Martin and colleagues that real opposition to this theory was aired, as well as opposition to the uncritical acceptance of the notion of primary branchiogenic carcinoma in many cases of cystic squamous carcinoma masses.Reference Martin and Morfit7, Reference Martin, Morfit and Ehrlich8 These authors proposed that the diagnostic criteria for primary branchiogenic carcinoma be tightened, and suggested that four criteria be met before a diagnosis of branchiogenic carcinoma could be made unreservedly: (1) the tumour must lie along the anterior border of the sternocleidomastoid muscle; (2) light microscopy must suggest that the tumour originates from a tissue type which is normally found within vestigial branchial apparatus structures; (3) no other primary tumour should be diagnosed during the five years following the putative diagnosis of a branchiogenic carcinoma; and (4) light microscopy must demonstrate that the malignancy has developed within the wall of an epithelial-lined cyst.Reference Martin and Morfit7, Reference Martin, Morfit and Ehrlich8 These papers proved to be influential, as in the latter half of the twentieth century the number of reported cases of branchiogenic carcinoma dropped substantially.
The modern debate
Some authors have adopted a deeply sceptical position and asserted that, even in the absence of a detectable primary tumour, cystic cervical deposits of squamous carcinoma will ultimately usually prove to be metastatic tumours.Reference Micheau, Klijanienko, Luboinski and Richard9, Reference Soh10, Reference Jereczek-Fossa, Casadio, Jassem, Luzzatto, Viale and Bruschini12, Reference Maturo, Michaelson and Faulkner16–Reference Thompson and Heffner19, Reference Delank, Freytag and Stoll23 These authors cite the fact that the original diagnostic criteria for branchiogenic carcinoma were incomplete, due to the absence of panendoscopy and tonsillar biopsy, limited follow up in several instances, and the administration of radiation therapy in many cases (which might have eradicated occult primary tumours and so created the impression of a true primary cervical neoplasm).Reference Micheau, Klijanienko, Luboinski and Richard9, Reference Jereczek-Fossa, Casadio, Jassem, Luzzatto, Viale and Bruschini12
In a review of 121 adult patients who presented with an initial diagnosis of lateral cervical cysts, Gourin and JohnsonReference Gourin and Johnson24 demonstrated histological evidence of metastatic squamous cell carcinoma in 12 resected cysts (9.9 per cent). The incidence of malignancy was significantly greater in patients older than 40 years (23.5 per cent). When pre-operative fine needle aspiration or frozen section histological examination showed malignancy, then panendoscopy with directed biopsies of Waldeyer's ring revealed an occult primary tumour in five of these patients.
It has been suggested that patients with a cystic mass located at cervical levels II or III and who are aged 40 years or older should all be investigated by computed tomography (CT) and should undergo ipsilateral tonsillectomy, even if the tonsil appears clinically normal.Reference Bath, Murty and Bradley25, Reference Flanagan, Roland and Jones26
Currently, the majority of patients who present with a neck mass undergo fine needle aspiration biopsy for cytological analysis. Fine needle aspiration cytology (FNAC) may be useful to exclude a cold abscess (i.e. tuberculosis), particularly in developing countries (Figure 1). When solid tissue is aspirated, a diagnosis is likely, but when only fluid is recovered, it is still possible to distinguish a malignant cyst from a benign cyst by the use of imaged cytometry deoxyribonucleic acid analysis, as the former will demonstrate aneuploid and the latter diploid cells.Reference Nordemar, Högmo, Lindholm, Tani, Sjöström and Auer27 With this information, it is possible to plan appropriate elective treatment for all patients aged 40 years and older.
Fig. 1 Cold abscess (tuberculosis) can be confused with other cystic masses, particularly in developing countries.
Other authors, however, are more ready to accept the possibility of malignancy arising in a pre-existing benign developmental cyst (such as a branchiogenic cyst), and thus are less opposed to the idea of a primary branchiogenic carcinoma.Reference Khafif, Prichep and Minkowitz13–Reference Skoulakis, Karametos, Fericean, Kostis, Divani and Valagiannis15, Reference Girvigian, Rechdouni, Zeger, Segall, Rice and Petrovich20, Reference Misonou, Iwabuchi, Aizawa and Murakami28–Reference Singh, Balwally, Sundaram, Har-El and Krgin30
The question remains, who is correct?
Recommendations
Cystic lateral cervical masses in adults over the age of 40 years are presumed to be malignant until proven otherwise.Reference Cinberg, Silver, Molnar and Vogl31 While adjunctive studies (including imaging and FNAC) are of some use, the fact remains – as was observed over a century ago – that surgical excision is usually the best modality to discriminate between benign and malignant cystic neck masses in this age group. Branchiogenic carcinoma is a diagnosis of exclusion. All adult patients over the age of 40 years who present with malignant cervical masses (arising in anatomical regions where branchial cleft anomalies are known to develop) should be assumed to have occult primary tumours. A diagnosis of primary branchiogenic carcinoma should be viewed at the outset with a more than healthy dose of scepticism. The diagnosis of a cystic squamous cell carcinoma in the neck should trigger a search for a presumed primary tumour, both by imaging modalities and by panendoscopy of the upper aerodigestive tract with liberal use of directed biopsies: in particular of Waldeyer's ring. Some would suggest that bilateral tonsillectomies be performed and the tonsils thoroughly examined pathologically before a diagnosis of branchiogenic carcinoma can be considered.Reference Micheau, Klijanienko, Luboinski and Richard9, Reference Soh10, Reference Marlowe, Goodman, Mobini and Dave32
The authors believe that a cautious approach to a possible prospective diagnosis of branchiogenic carcinoma should consist of the following.
First, the cystic neck mass should be demonstrated to indeed be a cystic squamous carcinoma; this may require imaging, FNAC or surgical biopsy. While malignancy can be diagnosed on cytology, it is only by surgical excision and thorough pathological examination that light microscopic features supportive of a diagnosis of branchiogenic carcinoma can be elicited.
Second, imaging of the head and neck by CT and magnetic resonance imaging should be employed to exclude an occult primary tumour. Some would even add positron emission studies to this stage of evaluation.Reference Raghavan and Bradley33, Reference Goldenberg, Sciubba and Koch34
Third, the patient should undergo panendoscopy with liberal use of directed biopsies, in the absence of detectable pathology (patients in whom panendoscopy reveals clinically suspicious areas should undergo targeted biopsy).
Fourth, bilateral tonsillectomy should routinely be performed. In those patients whose tonsils have been excised previously, the tonsillar beds should be biopsied.
Finally, if all the preceding examinations are negative, then surgical excision of the cervical mass should provide definitive pathological evidence of both a pre-existing benign developmental cyst and a malignant tumour arising therein.
In those patients in whom (against the odds) a diagnosis of primary branchiogenic carcinoma appears possible, the generally recommended treatment is rather aggressive. Wide resection of the tumour itself (with clear margins, if technically possible) should be followed by modified radical neck dissection. While the use of post-operative radiation therapy is not uniformly recommended by all authors (some suggest awaiting recurrence before resorting to adjuvant radiation therapy), we believe that post-operative radiation therapy should be routinely employed, because these tumours are aggressive in their own right, and the failure to locate another primary tumour does not mean that no occult primary tumour exists. Thus, irradiation may serve the twin purposes of treating the neck as well as the (undetected) true primary site.Reference Jereczek-Fossa, Casadio, Jassem, Luzzatto, Viale and Bruschini12, Reference Khafif, Prichep and Minkowitz13 The role of adjuvant chemotherapy remains to be settled, although some have advocated its use.Reference Katori, Nozawa and Tsukuda35
Conclusion
In summary, and to answer the title of this article, it is unlikely that one will encounter and treat a true branchiogenic carcinoma. Certainly, one should entertain this diagnosis, but only following a process of exclusion, and only after a thorough search for evidence of an occult primary carcinoma.
