Localized extranodal lymphoma of the head and neck: Retrospective analysis of a series of 107 patients from a single institution

Purpose: To retrospectively analyze the outcome and patterns of relapse in localized extranodal non-Hodgkin's lymphoma of the head and neck (HN-NHL) after radiotherapy alone or combined modality treatment. Patients and methods: A retrospective analysis of 107 patients with HN-NHL was performed. Relapse patterns, overall survival (OS) and relapse-free survival (RFS) were analyzed. Only stage I (n = 50) and stage II (n = 57) patients were included with either low-grade (n = 21) or high-grade (n = 86) lymphoma. Fifty-nine patients were treated with radiotherapy (RT) alone and 48 patients received combined-modality treatment (CMT) consisting of chemotherapy (CHOP or CHOP-like) followed by radiotherapy. The volumes of irradiation included local field (n = 24), involved field (n = 13) and extended field (n = 70). The median age at diagnosis was 63 years (range, 1786 years). Results: The overall response rates (CR+PR) in the radiotherapy group and the combined modality group were 100% and 96%, respectively. With a median follow-up of 49.4 months, 29 of 59 patients after FIT alone (37%) and 30 of 48 patients after CMT (62%) were disease-free. In the whole series the projected five-year OS and RFS were 58.7% and 61.8%. At univariate analysis of clinical variables with potential impact on survival including age, stage, histology, IPI score, single or combined treatment and volumes of irradiation, only age and, to a limited extent, type of treatment influenced OS (age : 60 years 79%, > 60 years 41%, P < 0.001; RT alone 54.9%, CMT 62.8%, P = 0.0487) and RFS (<= 60 years 75%, > 60 years 50%, P < 0.001; RT alone 54%, CMT 71%, P = 0.039). Better OS and RFS rates were obtained in patients with stage 11 and high-grade disease treated with CMT (five-year OS and RFS 63% and 69%, respectively; the corresponding values for RT alone were 38% and 34%). The final model of the multivariate analysis retained only age (<= 60 years) as a significant prognostic factor for both RFS and OS (P < 0.001). In the whole series, the sites of relapse were mainly systemic (n = 32/40, 80%) and in-field relapses were rare (n = 3/40, 7.5%). Conclusion: HN-NHL is characterized by a high risk of relapse, particularly at distant sites. Older patients have a significantly worse prognosis. Radiotherapy offers a very good local control rate although combined modality treatment possibly produces better RFS and OS, especially for stage 11 and highgrade disease. Better systemic approaches are warranted for a more consistent impact on survival in this particular subset of extranodal lymphoma. However, radiotherapy alone may offer a feasible and effective modality for patients who cannot tolerate more aggressive treatments. Extended-field radiotherapy and the treatment of a larger number of uninvolved lymph nodal regions does not confer a RFS or OS advantage, either after RT alone or after CMT.