ABSTRACT
A relationship between the rate of regression of lymphomas treated with chemotherapy and long-term outcome has been observed. This study was undertaken to determine if the rate of tumor regression during radiotherapy for mediastinal Hodgkin's disease is a predictor of in-field recurrence. Twenty-nine patients with early-stage Hodgkin's disease treated with radiotherapy alone as part of an NCI randomized trial had both a non-massive mediastinal component of disease and all requisite simulation and port films available for analysis. The histology was nodular sclerosis in all patients. Stage distribution was as follows: IA-1; IIA-17; IIB-8; IIIA1-3. The median age was 27 years and the median radiation dose was 4470 cGy. A mediastinal mass ratio was calculated from each patient's simulation and weekly port films by dividing the width of the mediastinal mass by the intrathoracic diameter at the level of the carina. Histopathologic correlation was also done to quantify the degree of tumor vs. sclerosis in the specimens. Univariate analysis and Cox proportional hazards analysis were used to study the association between several covariates (stage, sex, symptoms, extra-lymphatic disease, initial mediastinal mass ratio, age, dose, percent tumor in the specimen, and cumulative percentage of tumor regression) and time to in-field recurrence, as well as probability of any failure. Univariate analysis indicates that lower dose, higher percent tumor in the specimen, and lower cumulative percent regression are statistically significant predictors for in-field recurrence, as well as for any failure. By Cox regression analysis, cumulative percent regression is the sole factor independently associated with in-field recurrence (two-tailed P=0.04). The percent tumor in the specimen is the only factor similarly identified for time to any failure (two-tailed P=0.02). Histopathologic correlation suggests that patients with early stage mediastinal Hodgkin's disease who demonstrate a high percent tumor in the specimen may be at increased risk of failure. Patients with a low cumulative percent regression during radiotherapy appear to be at an increased risk of in-field recurrence.
