Giaccone et al
A total of 2,271 (1,146 in the observation arm and 1,125 in the vaccination arm) QoL forms were collected, of which only 1,362 (723 in the observation arm and 639 in the vaccination arm) were used in the analyses. A total of 798 forms (388 in the observation arm and 410 in the vaccina- tion arm) could not be assigned to a time window. An additional 181 (82 in the observation arm and 99 in the vaccination arm) forms belonged to patients who were not included in the analysis population. The clinical baseline characteristics of the 334 patients of the analysis population (176 observation and 158 vaccination) were comparable with the rest of the intent-to-treat population. The compli- ance for the analysis population was similar in the two arms up to approximately 1 year and was 80% or higher at the 6-month follow-up. QoL scores between the two treatment arms were not statistically different. There seems to have been a drop in global QoL in both arms at the 6-week time point, but this effect was of short duration; the scores re- turned to the baseline level at the next time point (Fig 4).
and matrix metalloprotease inhibitors such as marimas- tat.31 All these approaches failed to improve survival con- vincingly. A similar trial to the marimastat study, using the metalloproteinase BAY 12-9566, was closed early because a shorter survival was detected in the investigational arm. Other studies are ongoing with inhibitors of angiogenesis (ZD6474) in patients who responded to induction therapy.
A common characteristic of these studies is to attack minimal residual-disease SCLC after induction therapy to deal with the smallest tumor burden possible. The present study is a prototype of a vaccination strategy in patients with minimal residual-disease SCLC. Bec2 is an anti- idiotypic antibody that mimics GD3, a ganglioside often expressed in tumors of neuroectodermal origin, including SCLC. In previous studies in melanoma patients, anti-GD3 antibody responses were detected in 20% to 33%12,15; how- ever, the outcome did not seem to be improved by the treatment. In a pilot study of 15 SCLC patients, performed at Memorial Sloan-Kettering Cancer Center, promising re- sults were observed,10 which stimulated the initiation of this large phase III trial.
Patients with limited disease display a high response to chemoradiotherapy, but only approximately 15% to 25% can be considered cured after combined-modality thera- py.16,17 Because SCLC is a disease that is sensitive to chemo- therapy and radiotherapy, several approaches to try and reduce the relapse rate have been studied.2 There is no clear evidence from reported data that maintenance chemother- apy improves survival duration,18-21 although some studies have shown a prolonged time to progression.
A number of other approaches have been attempted to prolong tumor control in patients responding to initial standard treatment, including high-dose chemotherapy late intensification,22,23 interferon alfa24-27 and gamma,28-30
Unfortunately, none of the major end points of the present study were achieved, and no improvement in sur- vival in patients who were vaccinated was observed overall or in any patient subgroups. QoL also was not improved in this study. Furthermore, substantial local toxicity and flu- like symptoms were accompanied by the administration of the vaccine, and some patients declined treatment because of these toxicities. Interestingly, the humoral response ob- tained in approximately one third of patients did seem to point in the direction of an improvement of survival. How- ever, although a prospectively planned analysis, this com- parison of nonrandomized groups of patients should be interpreted with caution, because imbalances of patient characteristics may be present; in fact, it seemed that a substantial number of patients in the responder group had received PCI, which in the multivariate analysis of the whole study imparted a significant positive influence on survival. In a study by Grant et al,10 one third (five of 15) of patients also developed anti-GD3 antibodies. The exceptionally good results observed in this small pilot study probably can be explained by patient selection. Another interesting ob- servation in our study is that patients with a PPD-positive skin reaction did significantly worse after vaccination. In the 1980s, intrapleurally administered adjuvant BCG was evaluated in patients with radically resected NSCLC. These studies not only failed to show a benefit,32 but some of them suggested a worse outcome in patients receiving BCG.33,34 Thus, it is conceivable that BCG might have had a negative impact in our study, at least in a subgroup of patients.
Fig 4. Means and standard deviations of overall health/quality of life QoL in both arms of the study.
Our study, represents the first large-scale phase III trial of an antiganglioside vaccine as adjuvant therapy in SCLC. It provides a large database of patients with limited-disease SCLC, which adds additional validation to currently
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.