D. Choy

Preliminary report of the asian-oceanian clinical oncology association randomized trial comparing cisplatin and epirubicin followed by radiotherapy versus radiotherapy alone in the treatment of patients with locoregionally advanced nasopharyngeal carcinoma†

The aim of this trial was to compare the outcome achieved with neoadjuvant chemotherapy followed by radiotherapy to that achieved with radiotherapy alone for patients with locoregionally advanced undifferentiated or poorly differentiated nasopharyngeal carcinoma (NPC) meeting one of the following criteria: Hos T3 disease, Hos N2‐N3 disease, or lymph node size ≥3 cm.

Prognostic factors of nasopharyngeal carcinoma: a review of 759 patients

The records of 759 Stage I to IV nasopharyngeal carcinoma patients seen between January 1976 and December 1983 were reviewed. There were 72 (9.5%), 162 (21.3%), 317 (41.8%) and 208 (27.4%) patients with Stage I, II, III and IV disease, respectively. Hos stage classification was found to give a reliable prognosis. The actuarial survival at 5 and 10 years for Stage I, II, III and IV disease was 80.8%, 71.5%, 40.7%, 17.7% and 33.4%, 48.4%, 30.0%, 9.5%, respectively. The significant factors affecting survival were found using Cox multivariate analysis, N and T stages, the size and degree of fixation of neck nodes, sex, age, the presence of cranial nerve palsy and ear symptoms at presentation. Bilateral neck node involvement, histology subtypes of tumour, headache and nasal symptoms at presentation, the difference in radiation dose to the primary tumour and the neck, the initial haemoglobin and white blood count were not significant factors in the present study. The N stage was found to be prognostically significant even among patient groups stratified for the size and degree of fixation of the neck nodes involved. Women had a better survival rate than men, and the survival of patients less than 40 years old was better than those who were older at 5 years, although by 10 years the survival for the two groups was comparable.

Volumetric analysis of tumor extent in nasopharyngeal carcinoma and correlation with treatment outcome

PURPOSE To investigate the variability of tumor volume in nasopharyngeal carcinoma using quantitative measurements of tumor bulk derived from computed tomography, and to study the prognostic value of tumor volume in comparison with other variables. METHODS AND MATERIALS Two hundred ninety patients with newly diagnosed nasopharyngeal carcinoma were included in the study. The primary tumor volume (PTV) and nodal tumor volume (NTV) were obtained by outlining the tumor contour followed by summation of areas in sequential pretreatment computed tomography axial scans. Total tumor volume (TTV) was obtained by adding the PTV and NTV. All patients had radiotherapy as the primary treatment, 67 patients also received cisplatin-based neoadjuvant chemotheraphy. RESULTS A large variation in tumor volume was observed, especially in advanced stage disease. The median PTV (cc) in Hos T1, T2, and T3 disease were: 6.9 (range: 0.9-42.7), 18.8 (1.6-127.9), and 52.4 (3.3-166.8). The median TTV (cc) in Hos stage I to IV disease were: 7.6 (range: 1.3-42.7), 19.8 (3.2-55.7), 40.7 (4.1-222.7), and 51.1 (3.1-274.7). Patients with a large PTV (>60 cc) were associated with significantly poorer local control (5-year local control rate: 56%) and disease-specific survival (5-year survival rate: 53%). In patients with a small PTV (< or =20 cc), there were no significant differences in local control among different T stages. Large NTV (>30 cc) was associated with significantly higher distant failure rate (5-year distant relapse-free survival rate: 54%) and lower disease-specific survival (5-year survival rate: 40%). In multivariate analysis, only PTV was found to be an independent factor in predicting local control. CONCLUSION A large variation of tumor volume was present in different T stage disease of nasopharyngeal carcinoma, and PTV represents an independent prognostic factor of local control that appears to be more predictive than Hos T stage classification.

Long-Term Survival After Cisplatin-Based Induction Chemotherapy and Radiotherapy for Nasopharyngeal Carcinoma: A Pooled Data Analysis of Two Phase III Trials

5524 Background: To evaluate the long-term treatment outcome in patients with advanced stage nasopharyngeal carcinoma (NPC) treated by cisplatin-based induction chemotherapy and radiotherapy (CRT) versus radiotherapy alone (RT). METHODS The updated records of two previously reported phase III studies (the Asian-Oceania Clinical Oncology Association trial and the Guangzhou trial). testing the benefit of adding induction chemotherapy to radiotherapy in NPC were reviewed and the data were pooled together for analysis. A total of 784 patients were included for analysis, with equal number of patients in both the CRT and RT arms. The induction chemotherapy consisting of 2-3 cycles of cisplatin 100 mg/m2 day 1, bleomycin 10 mg/m2 day 1 & 5, and fluorouracil 800 mg/m2 day 1-5, or cisplatin 60 mg/m2 day 1 and epirubicin 110 mg/m2 day 1. Radiotherapy was given to the nasopharynx and neck using megavoltage radiation, with a median dose of 70 Gy. Treatment compliance was 92.6% in the CRT arm and 98% in the RT arm. The median follow-up time for surviving patients was 67 months. Analysis was done by intention to treat. RESULTS The addition of induction chemotherapy to radiotherapy was associated with a decrease in relapse by 14.3% and cancer deaths by 12.9% at 5 years. The 5-year relapse-free survival rate was 50.9% in the CRT arm and 42.7% in the RT arm (p=0.014), and the 5-year disease-specific survival rate was 63.5% in the CRT arm and 58.1% in the RT arm (p=0.029). The median disease-specific survival was not yet reached in the CRT arm and it was 82 months in the RT arm. The incidence of loco-regional failure and distant metastases were reduced by 18.3% and 13.3% at 5 years respectively with induction chemotherapy. There was no significant difference in the failure patterns between the 2 arms. CONCLUSIONS The addition of cisplatin-based induction chemotherapy to radiotherapy was associated with a modest but significant improvement in survival in advanced stage NPC. No significant financial relationships to disclose.

Sensorineural hearing loss in patients treated for nasopharyngeal carcinoma: A prospective study of the effect of radiation and cisplatin treatment

PURPOSE The pattern of sensorineural hearing loss (SNHL) after primary treatment for nasopharyngeal carcinoma (NPC) was studied, and the effect of cisplatin, radiotherapy does, and fractionation were evaluated. METHODS AND MATERIALS One hundred thirty-two patients, 227 ears, and 1100 audiogram reports were analyzed. Radiotherapy dose ranged from 59.5 to 76.5 Gy. Fifty-two patients received preirradiation cisplatin, total dose 100-185 mg/m(2). Serial postirradiation bone conduction thresholds at 0.5 kHz, 1 kHz, 2 kHz, and 4 kHz were compared with pretreatment thresholds at respective frequencies. Increase of at least 15 dB was considered as significant and was further grouped as transient or persistent SNHL. Univariate and multivariate analyses were performed to identify predicting factors for persistent SNHL. RESULTS At median follow-up of 30 months, 24.2% of ears developed persistent SNHL. High frequency was more affected than low frequencies, 22 vs. 5.3%. Males were more affected than females, 29.4 vs. 15.5%, p = 0.0132. Incidence of persistent SNHL increased with age, with 0, 17.2, and 37.4% of patients aged under 30, between 30-50 and over 50 affected, respectively, p = 0.0001. High incidence was found in patient with postirradiation serous otitis media (SOM), 46.9%. Chemotherapy with cisplatin and radiation dose or fractionation had no significant effect. Multivariate analysis confirmed age, sex, and postirradiation SOM as significant prognostic factors for persistent SNHL. CONCLUSIONS Transient and persistent SNHL occurred after radiotherapy, more commonly affecting high frequency. A low dose of preirradiation cisplatin did not increase the risk. A dose fractionation effect of radiotherapy was not confirmed in this study.

Polymorphic reticulosis and conventional lymphomas of the nose and upper aerodigestive tract: A clinicopathologic study of 70 cases, and immunophenotypic studies of 16 cases

Seventy patients with malignant lymphomas, including the entity known as polymorphic reticulosis (PR), involving the nose, nasal sinuses, nasopharynx, oropharynx (excluding tonsil), and larynx were studied. There were 26 cases of PR, 19 cases of lymphoma with features of PR (ML[PR]) and 25 cases of conventional lymphomas. Fourteen of the 25 conventional lymphomas were due to dissemination from distant sites. For all histologic types of primary lymphoma, the presenting symptoms were similar, and the nasal cavity was more commonly involved than the nasopharynx. Patients with PR were younger, had a higher male:female ratio, and had a better overall survival rate than patients with conventional lymphomas. Cryostat section immunohistochemistry performed on 17 samples from 16 patients showed only one B lymphoma out of 11 primary lesions; the other 10 cases and three recurrent tumors at distant sites showed phenotypic markers of T lymphocytes and natural killer cells. All three secondary tumors were of B-cell type. Of eight patients with sequential biopsies, progression to a more malignant histopathologic type was found in six. In the PR and ML[PR] biopsies, angiocentricity was detected in 11%, and angioinvasion in 22%. We could not confirm identity of PR with other angiocentric immunoproliferative lesions.

Treatment outcome and prognostic factors for primary nasal lymphoma.

PURPOSE To report our experience managing a large series of Chinese patients with primary nasal lymphoma. PATIENTS AND METHODS From January 1975 to December 1993, 100 patients (median age, 50 years) with newly diagnosed primary nasal lymphoma were studied. There were four low-grade, 62 intermediate-grade, nine high-grade, and 25 unclassifiable lymphomas. Immunophenotyping was performed in 45 patients: eight B cell, 35 T cell, and two uncertain. All cases of angiocentric lymphoma that were typed were T cell. Fifty-two patients had stage I disease, 15 had stage II, four had stage III, and 29 had stage IV. Only 15 patients had B symptoms (weight loss, night sweats, and/or fever), and 11 had bulky disease. Thirty-nine patients with clinically localized stage I and II disease received local radiotherapy alone (before 1980), and the remaining 28 stage I and II patients received combination chemotherapy followed by local radiotherapy. The 33 patients with advanced stage III and IV disease were given combination chemotherapy, and additional radiotherapy was given to five of them who had bulky local disease. RESULTS Significantly higher complete remission rates were observed in patients with early stages of disease and those without B symptoms. Superior disease-free survival after complete remission was observed in patients with stage I/II disease. Univariate factors associated with a better overall survival included age less than 60 years, stage I disease, and absence of B symptoms. Survival was significantly better in the subgroup of patients with stage I disease. CONCLUSION Patients with nasal lymphoma, especially those with advanced disease, seemed to have a poor prognosis, and their clinical outcome was not improved significantly by the use of chemotherapy instead of radiotherapy or the use of doxorubicin-containing chemotherapeutic regimens.

Nasopharyngeal carcinoma: Orderly neck node spread

A prospective study of 271 consecutive patients with newly diagnosed nasopharyngeal carcinoma was undertaken to assess the pattern of cervical nodal involvement with reference to 10 cervical nodal groups and three levels of neck; 204 (75.3%) patients were found to have cervical lymphadenopathy at presentation. Fifty-four (26.5%) of these patients had right cervical lymphadenopathy, 70 (34.3%) had left cervical lymphadenopathy, and 80 (39.2%) had bilateral cervical lymphadenopathy. The occurrence of lymphadenopathy in the 10 cervical nodal groups and the mean size of nodes in these nodal groups were computed. The subdigastric and upper jugular group was involved in more than 95% of cases. The lower the position in the neck, the less frequently the nodal group was involved. The mean size of nodes was largest in the subdigastric and upper jugular region compared with the other groups. The nodes in the upper neck were generally larger than those in the lower neck. The lower two levels of neck were involved without involvement of the upper level of the ipsilateral neck in fewer than 4% of cases. The present study indicates that neck node involvement by nasopharyngeal carcinoma is by orderly spread down the neck, which explains the adverse prognostic significance of neck node involvement in the lower neck. The orderly involvement of the neck nodes suggests that prophylactic irradiation of the neck should be given at least one level beyond the clinical extent of disease, which for patients with no clinically palpable node would mean prophylactic irradiation of the upper neck.

Locally Recurrent Nasopharyngeal Carcinoma: Treatment Results for Patients with Computed Tomography Assessment

PURPOSE To study the treatment outcome in patients with locally recurrent nasopharyngeal carcinoma as restaged by computed tomography (CT). PATIENTS AND METHODS One hundred forty patients with CT restaged locally recurrent nasopharyngeal carcinoma were reviewed. Patients were restaged at recurrence according to the AJCC stage classification with the following distribution: T1-T2:30%, T3:19%, T4:51%. Ninety-seven patients received reirradiation; among these 62 had external irradiation, 34 had brachytherapy, and 1 had both. Twelve patients received surgery. Thirty-one patients were treated with palliative intent and received either chemotherapy or supportive treatment only. Overall survival (OAS) and performance-adjusted survival (PAS, defined as surviving with a Karnofsky performance score [KPS] > 50) were calculated by Kaplan-Meier method. Multivariate analysis was performed using the Cox model. RESULTS The median survival for all patients was 23.8 months. After reirradiation, the 3-yr and 5-yr OAS rates were 46% and 36%, respectively. The corresponding PAS rates were 40% and 28%. The 3-yr OAS rates for recurrent T1-2, T3, and T4 disease after reirradiation were 71%, 42%, and 30%; the corresponding 5-yr OAS rates were 57%, 42%, 17%. The 3-yr and 5-yr OAS rates in patients receiving palliative treatments only were 19% and 0%, respectively. The 3-yr OAS rate after surgery was 42%. In the multivariate analysis, older age, recurrent T3-4 disease, and palliative treatment were unfavorable factors in predicting overall survival, whereas recurrent T3-4 disease, baseline KPS < 70, and palliative treatment were unfavorable factors in predicting PAS. A high complication rate was observed after reirradiation, with 34% of patients developing neurological sequel. CONCLUSION Aggressive treatment for locally recurrent nasopharyngeal carcinoma is warranted especially for those with disease confined to the nasopharynx. Survival after retreatment for more extensive disease remains poor but was still superior to supportive treatment only. Early diagnosis of local recurrence allows prompt administration of treatment and is associated with better outcome. Future studies should aim at improving the therapeutic ratio in the retreatment of recurrent disease especially in patients with more extensive local recurrence.

THE EFFECT OF LOCO-REGIONAL CONTROL ON DISTANT METASTATIC DISSEMINATION IN CARCINOMA OF THE NASOPHARYNX: AN ANALYSIS OF 1301 PATIENTS

PURPOSE This study evaluated the effect of loco-regional control on incidence of distant metastases in patients with nasopharyngeal carcinoma. METHODS AND MATERIALS Retrospective analysis was performed on 1301 patients with nasopharyngeal carcinoma treated from 1976-1989 in our hospital. The effect of different prognostic indicators on distant metastases free survival (DMFS) was analyzed, including T stage, N stage, sex, and loco-regional control. The significance of loco-regional control in distant metastatic dissemination was further studied with period and subgroup analysis. Cox regression was performed to identify the independent prognostic variables. RESULTS Patients with loco-regional relapse had significantly higher distant metastases rate than patients with loco-regional control (time-adjusted distant metastases rate at 5 years was 40.7% vs. 29.4%, p = 0.0012). By period analysis, the effect of loco-regional control on distant metastases was found to be significant in the first 2 years from diagnosis. Subgroup analysis showed that loco-regional relapse was associated with significantly higher distant metastases rate in patients with T1, N0, and N1 disease (p = 0.001, 0.001, 0.0226, respectively). The Cox regression model also confirmed loco-regional control as an independent prognostic indicator of distant metastases (p = 0.0001) besides T stage (p = 0.0006) and N stage (p = 0.0001). CONCLUSION Loco-regional relapse is a significant risk factor for development of distant metastases. Further effort should be made to improve loco-regional control and to eradicate distant metastases.