Bec2/BCG Vaccination in SCLC Patients
used therapeutic strategies in this disease. In particular, by univariate and multivariate analysis, it was shown that the use of concomitant chest radiation and chemotherapy was superior to sequential chemotherapy followed by radio- therapy in terms of survival; this may explain the longer survival in North American study subjects compared to those from the rest of the world, which was observed in univariate analysis, because concomitant chemoradiother- apy is more established in North America. Furthermore, patients who received PCI also had a significantly longer survival compared with those who were not treated prophy- lactically. This is likely the result of PCI in addition to the selection of better patients for this treatment. These results, however, are in line with studies comparing concomitant versus sequential chemoradiotherapy35 and meta-analyses of PCI.36 A meta-analysis of seven randomized trials evalu- ating the value of PCI in patients in complete remission reported improvement in brain metastasis recurrence, disease-free survival, and overall survival with the addition of PCI. The 3-year overall survival was improved from 15% to 21% with PCI.36
Why this study turned out to be negative is a matter of speculation. The fact that only one third of patients devel- oped a humoral response is probably a major potential cause of failure, although this was the expected rate of immunologic response based on the previous study. The choice of adjuvant or the anti-idiotypic-vaccination ap- proach may have contributed to this. Another potential explanation is the presence of GD3 in approximately only 60% of SCLC tissues.4 This is less than 100%,3,6 and in the present study patients were neither evaluated for nor strat- ified for GD3 expression; notwithstanding the difficulty in obtaining tissue to determine GD3 expression, this might have shed some light on the issue. In view of this possible reason of failure of the present study, a multivalent vaccine, perhaps including GD3, may be a better choice for future studies. In a recent study of a wide range of doses of Bec2 given in 50 patients with melanoma, doses lower than the 2.5 mg used in the present study seemed to be more immu- nogenic; however, prolonged booster response did not in- duce or maintain antibody responses.15 Recently, chimeric monoclonal antibodies against GD3 have been developed37; in preclinical studies the monoclonal KM871 markedly suppressed tumor growth, and a phase I study with this antibody has been conducted.38 In this study in 17 patients, there was dose-limiting toxicity; no human antichimeric antibody formation was detected; and interestingly, using 111In-KM871, a 15-fold uptake was seen compared to nor- mal tissue in biopsies of tumors of at least 1.5-cm dimen- sion. In this study, a major response was also documented.
Other gangliosides are highly expressed in SCLC, in- cluding GD2. An imaging study with labeled anti-GD2 an- tibody 3F8 confirmed tumor localization in all metastatic sites except the brain.39 However, in a randomized phase III
study of 334 patients with metastatic neuroblastoma, the chimeric antibody ch14.18 did not prolong survival.40 GD2-lactone keyhole limpet hemocyanin (KLH), mixed with the adjuvant QS-21, induced antibodies against GD2 in most melanoma patients who were vaccinated.41 EMD- 273063 is a humanized antibody against GD2, coupled to two molecules of interleukin-242; a phase I trial was per- formed in patients with melanoma that demonstrated that the dose-limiting toxicities were hypoxia, hypotension, and transient elevation of transaminases.43 Fuc-GM1, also present on most SCLC cells, conjugated to the carrier protein KLH and mixed with the adjuvant QS-21, was administered to 10 patients with SCLC who responded to induction therapy.44 All patients developed a serologic response, and the vaccine was well tolerated. Another study using synthetic Fuc-GM1 conjugated to KLH reported high antibody formation against GM1 as the bovine derivative.45 Other vaccination ap- proaches, using tumor cells genetically modified to produce granulocyte-macrophage colony-stimulating factor (GVAX), have been attempted with some success in NSCLC.46
In summary, this study of Bec2/BCG vaccination as maintenance therapy in responding patients with limited- disease SCLC is essentially negative. There could be other approaches to using gangliosides as targets for vaccination therapy in SCLC. However, higher titers of anti-GD3 anti- bodies in a larger proportion of patients may be needed to improve overall survival. Also, because no one single anti- genic target is expressed on all SCLC tumors, additional antigens may be needed to form a multivalent vaccine.
We acknowledge the EORTC staff who worked on the study (Desmond Curran, Ingrid Roucloux, Delphine Dubois, Rute Carneiro, Alessandra Busato, and Sophie Van Impe) and all monitors from Quintiles, GSO mbH Free- lancer, and Merck KGaA worldwide. Furthermore, we thank Merck KGaA and ImClone personnel involved in the study and all investigators (see Appendix) who participated and the cooperative groups that took part in this study. We are grateful to Merck KGaA and Imclone for supporting this independent EORTC study. We also acknowledge critical review of the manuscript by L. Krug, MD.
Other Participants of the Study
Austria. Dr Mohn-Staudner, Pulmologisches Zentrum der Stadt Wien; Dr Pirker, Allgemeines Krankenhaus der Stadt Wien; Dr Ulsperger, Krankenhaus der Stadt Wien-Lainz; and Dr Puganigg, Krankenhaus des Landes Kaernten.
Australia. Dr Millward, Royal Prince Alfred Hospital; Dr Ransom, Royal Perth Hospital; Dr Richardson, Monash Medical
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.