Abstract
Skin metastases (SM) are an unusual site for distant metastases (DM) in head and neck squamous cell carcinoma (HNSSC). We report a case of carcinoma tonsil with failure due to SM after definitive treatment with radical radiotherapy. A relevant review of literature of this uncommon presentation which has an impact on management protocols and prognosis is presented.

Introduction
Loco-regional control of HNSSC has improved with multimodality treatment options. However, this has not translated to improved survival due to failure at distant sites.¹ The lung, liver and bone are the usual sites for DM.² Isolated SM are very uncommon with an incidence ranging from 07% to 2.4%.^3,2 Oropharynx as a primary site has an intermediate propensity to SM.1 Oropharyngeal carcinomas account for 10% of all head and neck epithelial cancers and are a significant problem in our country.4 8% of patients with oropharyngeal cancers will have DM at presentation.⁵ It is important for physicians to be vigilant in identifying SM as they suggest a poor prognosis and should be taken into consideration when the treatment is planned.

Case Report
A 67 year old male patient, chronic beedi smoker, presented to our services with right cervical lymphadenopathy of 6 months duration. A fine needle aspiration cytology of the mass was reported as a metastatic squamous cell carcinoma. A head and neck evaluation revealed small, ulcerative primary in the right tonsil. The disease was staged according to the AJCC classification (2000) as squamous cell carcinoma of the right tonsil - Stage IV, T1N3M0.⁶

The patient was treated with radical radiotherapy with curative intent. The first follow up revealed a complete response at the primary site and a residual mobile nodal mass (3 cms x 3 cms) at level II on the right side. Salvage neck dissection was planned. The patient however, presented prematurely with a skin nodule over the right scapular region. Examination also revealed a skin nodule in the axilla (Figs. 1 and 2). A fine needle aspiration cytology confirmed the suspicion of metastatic skin lesion over the scapular region and the axilla (Fig. 3). Definitive treatment was deferred in view of this new development and the patient was treated with palliative intent.

Fig. 1 : Cutaneous metastatic nodule - right scapula.	Fig. 2 : Cutaneous metastatic nodule - axilla.

Fig. 3 : FNAC - Scapular region showing metastasis of a high grade squamous carcinoma (100X).

Discussion
HNSSC are distinct in their nature when compared to other cancers, in that they tend to fail loco-regionally. The emphasis on the treatment of these cancers is therefore on improving local and regional control rates. Advances in surgical techniques and chemoradiotherapy schedules have resulted in a significant improvement in loco-regional control rates. This improvement has however not translated to better overall survival rates because of failure at distant sites.¹ Improvement in survival rates will be possible by identifying and treating aggressively the subset of patients who are prone to distant failure. It is incumbent upon us as physicians to identify these lesions which will impact future management and prognosis.

DM was considered to be an uncommon entity in HNSSC with Crile, in 1923 reporting an incidence of 1%. More recent studies now suggest that distant metastases from head neck cancers are not so rare. Series that reflect clinical studies show incidence rates varying from 5.5% to 23.7% and autopsy studies that report distant metastases in 17% to 57%.⁷ The substantial increase in the incidence of DM due to better loco-regional control has forced the clinician to take cognizance of the role it plays in the management of HNSSC today.

Literature has identified several prognostic factors that have an influence on the incidence of DM. A study of 1244 patients with HNSSC who had achieved loco-regional control suggested that the hypopharynx (14%) and nasopharynx (11%) have the greatest frequency of DM. The oropharynx (7%) and supraglottis (8%) have an intermediate and the oral cavity and glottis (<1%) have the lowest frequency of DM.1 The commonest sites of metastasis from these primary cancers are the lung (70-75%), liver (17-38%) and bone (23-44%).²

Cutaneous metastases as seen in our case report are very uncommon and they account for 10% of all DM.3 The incidence of SM from HNSSC had been reported as 0.763%, 0.8% to 1.3% and 2.4% in various retrospective reviews.^3,2 SM are defined as isolated or multiple intradermal collections of tumour cells remote from the primary or loco-regional disease.² They present as solitary or multiple, discrete or confluent, dermal or subcutaneous nodules discontinuous from the epidermis, as described in our case.³ They may grow rapidly and appear in showers or they may be slow-growing and solitary. They may also ulcerate or become necrotic. It is important to differentiate these lesions from extensions from the primary cancer, scar metastasis, metastasis from other primary skin cancers, internal organs and primary cutaneous disorders.^3,8 Dermal metastases can spread by three modalities: direct spread, local spread via dermal lymphatics and distant spread via the haematogenous route.⁹ Localization of metastatic skin disease does not occur in a random pattern. The sites of SM in HNSSC include neck, chest, scalp, face, lips, axilla, areolas, back, arms and digits; the most common being the neck and chest. The majority of SM usually occur above the umbilicus.²

SM usually suggests a very poor prognosis. The time to presentation has been reported at between 1 and 39 months, with 75% appearing within 18 months of treatment,² with a median time to occurrence of 6 months.³ Survival rates of 3 to 7 months following diagnosis have been reported.^2,3 Though some authors suggest that SM are the first indicators of loco-regional recurrences and distant failure, there is a uniform consensus that the appearance of SM is a bad prognostic sign and survival, irrespective the therapeutic intervention is dismal.

There are various factors that are prognostically important for the development of SM. Studies indicate that DM lesions can appear in the presence of absence of loco-regional control. In their series, Kotwall et al showed that 90% of patients who died with DM had uncontrolled tumour at the primary site or in the neck.10 On the other hand DM have been documented in patients with loco-regionally controlled cancers.¹ Authors agree that the presence of two or more nodes, extra capsular spread, nodes lower in the neck were found to be statistically significant in predicting SM.^3,1,11 There is a disagreement regarding T size as a prognostic factor, with some studies finding a relation between local tumour extension and DM and others not.¹ A study of DM in squamous cell carcinoma of tonsillar fossa showed no statistical significance of T stages. However, N stage was a statistically significant factor when N0/N1 stages were compared with N2/N3 stages (p < 0.01).¹² Factors of no statistical significance include stage of tumour at initial presentation, age, gender, perineural invasion, history of radiation as a primary treatment, use of induction chemotherapy.^3,1 With respect to oropharyngeal tumours, as in our case report, a prospective study of 139 patients of HNSSC has shown that loco-regional tumour control, determined by several clinical parameters including reflex otalgia, T size, smoking and predominant growth patterns to be of importance in developing distant metastasis.¹³

Treatment for SM is inconclusive. Surgical excision has been shown to improve survival rates in one study.⁴ However, surgical treatment is not the norm and should only be considered in a few highly selective cases in which the SM is the only site of disease with no other metastases in the body. In general however, the treatment offered irrespective of the form is palliative.

References

1. Leon X, Quer M, Orus C, Venegas M, Lopez M. Distant metastases in head and neck cancer patients who achieved loco-regional control. Head Neck 2000; 680-6, 2000.
   
2. Yoskovitch A, Hier MP, Okrainec A, Black MJ, Rochon L. Skin metastasis in squamous cell carcinoma of the head and neck. Otolaryngol Head Neck 2001; 124 (3) : 248-52.
   
3. Pitman KT, Johnson JT. Skin metastases from head and neck squamous cell carcinoma: incidence and impact. Head Neck 1999; 21 : 560-5.
   
4. Fletcher OH. Textbook of Radiotherapy. Philadelphia: Lea - Febiger, 1980 : 315-7.
   
5. Pugliano FA. Tumours of the Oropharynx. In Watkinson JC, Gaze MN, Wilson JA (eds): Stell and Maran’s Head and Neck Surgery. Oxford: Butterworth-Heinemann, 2000 : 319-35.
   
6. American Joint Committee on Cancer. Manual for staging of cancer, 6th ed. Heidelberg: Springer; 2002.
7. Probert JC, Thompson RW, Bagshaw MA. Patterns of spread of distant metastases in head and neck cancers. Cancer 1974; 33 : 127-33.
   
8. Schwartz RA. Cutaneous metastatic disease. Journal of American Academy of Dermatology 1995; 33 : 161-82.
   
9. Kmucha ST, Troxel JM. Dermal metastases in epidermoid carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 1993; 119 : 326-30.
   
10. Kotwall C, Sako K, Razack MS, Rao U, Bakamjian V, Shedd D. American Journal of Surgery 1987; 154 : 439-42.
    
11. Vikram B, Strong EW, Shah JP, Spiro R. Failure at distant sites of following multimodality treatment for advanced head and neck cancer. Head Neck Surgery 1984; 6 : 730-3.
    
12. Berger DS, Fletcher GH. Distant metastases following local control of squamous cell carcinoma of the nasopharynx, tonsillar fossa and base of the tongue. Radiology 1971; 100 : 141-3.
    
13. Beer KT, Greiner RH, Aebersold DM, Zbaren P. Carcinoma of the oropharynx: local failure as the decisive parameter for distant metastases and survival. Strahlenther Onkol 2000; 176 (1) : 16-21.