Giaccone et al
and two in the vaccination arm: radiation pneumonitis caused three deaths (two in the vaccination arm), and three deaths were caused by pancytopenia after progression and administration of second-line chemotherapy (all in the ob- servation arm).
Survival and Progression-Free Survival
This analysis is based on a cutoff date of October 2004. The median follow-up was 35.6 months, and all patients were off protocol treatment at the time of the analysis. At the time of the analysis, 72% of patients had progressed in both arms. Most patients progressed at distant sites; 46% and 40% in the observation and vaccination arms, respec- tively, had distant progression only, and 25% and 29% experienced both local and distant progression. Overall, 71% and 76% of the patients were reported dead in the observation and vaccination arms, respectively. The major cause of death was tumor progression (88% and 86%, respectively). Table 3 reports the major survival results. Figures 1 and 2 depict the overall and progression-free survival curves, respectively. There were no statistically sig- nificant differences in outcome between the two arms for both overall survival and progression-free survival when all randomly assigned patients were taken into consideration. Similar results were obtained when the ineligible patients were excluded from the analysis.
Fig 1. Kaplan-Meier overall survival curves. On the y-axis, the percentage of surviving is reported; on the x-axis, the time from randomization (months) is reported.
and a trend toward shorter survival was also observed (me- dian, 13.3 v 23.8 months; P .0857). In contrast, the progression-free and overall survival in patients with nega- tive PPD tests were essentially similar to the results obtained in the whole population.
In the small group of patients (n 55) with PPD- positive reaction, those who received vaccination had a significantly shorter progression-free survival than those in the observation arm (median, 4.9 v 9.5 months; P .0377),
Table 3. Overall Survival and Progression-Free Survival According to Treatment Arm, From Random Assignment (all patients assigned)
Progression-free survival and overall survival of pa- tients with complete remission and partial remission were not significantly different between the two arms of the study.
By univariate analysis the following variables had a significant positive impact on survival at the 5% level (Table 4): centers in North America versus rest of the world; complete versus partial response to induction; concomitant
Median PFS, months
5.6 to 7.7
5.3 to 6.6
1-year PFS, %
25.5 to 37.9
25.4 to 36.7
2-year PFS, %
Abbreviation: PFS, progression-free survival.
Median survival, months
14.6 to 20.3
13.0 to 17.7
1-year survival, %
55.3 to 67.2
52.0 to 64.1
2-year survival, %
31.5 to 43.6
29.5 to 41.4
95% CI Hazard ratio 95% CI
20.1 to 30.7
19.5 to 30.2
1.11 0.95 to 1.36
Hazard ratio 95% CI
1.12 0.91 to 1.37
Fig 2. Kaplan-Meier progression-free survival curves. On the y-axis, the percentage of surviving without progression is reported; on the x-axis, the time from randomization (months) is reported.
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on February 15, 2015. For personal use only. No other uses without permission. Copyright © 2005 American Society of Clinical Oncology. All rights reserved.