Epidemiology and Biostatistics Publications
Using Treatment Interruptions to Palliate the Toxicity from Concurrent Chemoradiation for Limited Small Cell Lung Cancer Decreases Survival and Disease Control
URL with Digital Object Identifier
BACKGROUND AND PURPOSE: We analyzed the impact on survival outcomes of treatment interruptions due to toxicity arising during the concurrent phase of chemotherapy/radiotherapy (ChT/RT) for our limited-stage small-cell cancer (LSCLC) population over the past 10 years.
MATERIALS AND METHODS: From 1989 to 1999, 215 patients received treatment for LSCLC, consisting of six cycles of alternating cyclophosphamide/doxorubicin or epirubicin/vincristine (CAV; CEV) and etoposide/cisplatin (EP). Thoracic RT was started with EP at either the second or third cycle (85% of patients). RT dose was either 40 Gy in 15 fractions over 3 weeks or 50 Gy in 25 fractions over 5 weeks, delivered to a target volume encompassing gross disease and suspected microscopic disease with a 2 cm margin. Treatment breaks arising during concurrent ChT+RT were used to manage severe symptomatic or hematologic toxicities. We used the interruptions in thoracic RT as the 'marker' for any concurrent break and measured 'break duration' by the total length of time (in days) RT was interrupted, since that also signaled that ChT could be re-initiated. Patient results were analyzed for the impact of interruptions/treatment prolongation on overall and disease-free survival.
RESULTS: For all patients, 2-year and 5-year overall and disease-specific survivals were 22.7 and 7.2, 27.6 and 9.3%, respectively; overall and disease-specific median survivals were 14.7 months each. A total of 56 patients (26%) had treatment breaks due to toxicity. Hematologic depression caused the majority of breaks (88%). The median duration of breaks was 5 days (range 1-18). Patients with and without interruptions were compared for a range of prognostic factors and were not found to have any significant differences. Comparing interrupted/uninterrupted courses, median survivals were 13.8 versus 15.6 months, respectively, and 5-year overall survivals were 4.2 versus 8.3%, respectively. There was a statistical difference between overall survival curves which favored the uninterrupted group (P=0.01). When comparing a series of prognostic variables, multivariable analysis found that the most significant factor influencing survival in the present study was the presence of treatment breaks (P=0.006). There was a trend for development of any recurrence in the patients with breaks (P=0.08). When controlling for the use of prophylactic cranial irradiation (PCI) in the two groups, the rate of failure in the chest was higher in the patients with RT breaks (58 vs. 33%). The rate of failure in the brain was dependent on the use of PCI only.
CONCLUSIONS: Interruptions in treatment to palliate the toxicity from concurrent chemoradiation result in poorer local control and decreased survival.