Prognostic factors of carcinoma of the nasopharynx

Abstract

Nasopharyngeal carcinoma, Prognosis, Head and neck cancer. While radiotherapy indisputably remains the primary treatment modality of nasopharyngeal carcinoma (NPC), several aspects of this unique cancer of the head and neck are controversial. There are a number of factors which may influence the treatment results: stage of the disease, histopathology, age, sex, anti-Epstein-Barr virus (EBV) antibody titers and the immune status of the patient. However, general agreement on the staging system, histopathological classification and the importance of age and sex is lacking and the prognostic value of serum IgA level remains unestablished. Among the various prognostic factors of NPC, the extent of disease on presentation is probably most important in determining the outcome of treatment. Although the staging systems of the International Union Against Cancer (UICC)” and the American Joint Committee on Cancer Staging and End-Results Reporting (AJC)’ for cancer of the head and neck in regions other than the nasopharynx have been widely adopted, the staging system for NPC has not been uniformly accepted. In fact, practically every author on this subject has his/her own version of staging classification, usually a modification of the various proposed staging systems;6”5”9.22 the paper by S.C. Huang15 in this issue is a good example. The reason for this lack of conformity to the UICC and the AJC systems is that each has its limitations. As Ho” mentioned, it is not always possible to assess the true extent of the primary tumor within the nasopharynx. Examination of the nasopharynx can be very difficult in some patients. Not infrequently, the tumor can be entirely submucosal and not visible. The diagnosis is made only after histological confirmation of random biopsy specimens. Separation of the primary lesion into T1 and T2 based on the number of sites of the nasopharynx involved may be unrealistic. Although the AJC system takes into account cranial nerve involvement, it makes no distinction between cranial nerve and base of skull involvement. Both are classified as T4. In the UICC system cranial nerve involvement was not even considered in the staging criteria for the primary tumor. The results reported by Huang15 suggest that cranial nerve involvement carries a graver prognosis than base of skull involvement (5 year survival rate of 15-28.8% vs 23.2-41.3%, depending on the extent of cervical node involvement). In the Mayo Clinic experience reported by Scalon,23 a further distinction was made between involvement of cranial nerves I-IV and VII-XII; the five year survival was 14% for the former and 25% for the latter. The prognosis of NPC in the presence of cervical lymph node metastasis is worse than in its absence. In Huang’s series,” cervical lymph node involvement was the single most important prognostic factor. In the UICC system the N staging was based on the laterality of involvement and the mobility of the nodes. In the AJC system a further subclassification was based on the size of the lymph node metastasis, although Ho’s results suggested that the level of cervical lymph node metastasis may be more important.” However, the separation of the neck into three distinct levels may be arbitrary and not always distinct in some patients. While the UICC system does not give specific recommendations for stage groupings of the various TNM categories, the stage grouping recommended by the AJC system is by no means satisfactory and has not gained universal acceptance. The staging system proposed by Ho” appears to correlate well with prognosis in his experience, however, no other publication that uses his staging is available. At the International Symposium on the Etiology and Control of Nasopharyngeal Carcinoma, which took place in Kyoto, Japan, in 1977, a new staging classification that was adopted from Ho’s classification in 1970 was proposed (Table 1 and Figure l).’ It was recommended that this system be adopted for a prospective trial of five years. While the clinical staging classification of NPC has been variable, the histopathological classification of NPC has been controversial.24,28 In general, NPC can be sepa