43P - Multiplex bead-based measurement of humoral immune responses against tumor-associated antigens in stage II melanoma patients of the EORTC 18961 trial

Date 20 November 2015
Event ESMO Symposium on Immuno-Oncology 2015
Session Welcome reception and general Poster viewing
Topics Immunotherapy
Skin Cancers
Presenter Judith Michels
Citation Annals of Oncology (2015) 26 (suppl_8): 5-14. 10.1093/annonc/mdv514
Authors J. Michels1, N. Becker2, I. Kaiser3, A. Benner2, S. Agoussi4, S. Suciu5, N. Halama3, D. Jäger3, A.M.M. Eggermont6, I. Zörnig3
  • 1Medical Oncology, Institut Gustave Roussy, 94805 - Villejuif/FR
  • 2Division Of Biostatistics, German Cancer Research Center, Heidelberg/DE
  • 3Medical Oncology, National Center for Tumor Diseases, Heidelberg/DE
  • 4Iserm U981, Institut Gustave Roussy, Villejuif/FR
  • 5Cancer, EORTC, Brussels/BE
  • 6Director, Institut Gustave Roussy, Villejuif/FR



We have previously shown, that independently of age, chemotherapy and immunotherapy, tumor associated antigen (TAA)-specific humoral immune responses correlate with shorter progression free survival (PFS) and/or overall survival (OS) in early stage melanoma patients. In this comprehensive study, the prognostic significance of serum IgG antibody responses is explored in 959 stage II melanoma patients of the randomized EORTC 18961 trial, in which the efficacy of an adjuvant ganglioside GM2-KLH/QS-21 vaccination versus observation was compared.


Patients' sera were evaluated at baseline, after 12 weeks (ws), 48 ws and at the last available time point (at recurrence/remission) (observation arm, n = 488; vaccination arm, n = 471) for the presence of serum antibody responses against a panel of 45 antigens by a bead-based multiplex serological assay.


We frequently observed serum antibodies against all tested antigens in all patients except for 22 patients, who did not have any antibody responses at all (Table 1). 52 patients had antibodies against only one antigen, whereas most patients showed antibody responses against multiple antigens. By trend, we could confirm our previous findings of spontaneous antibody responses being associated with shorter PFS and OS. A negative prognostic value was depicted for a panel of IgG antibodies at baseline against Rhod E2, CyclinB1 for OS and Rhod E2, SSX2 for PFS in a multivariate analysis (adjusted for gender, Breslow thickness, ulceration status).


The serum profiling of antibody responses against a panel of TAA will help to identify patients of dismal prognosis who may benefit from an adjuvant immunotherapy. The sequential evaluation of humoral immune responses at the available time points as well as analysing their predictive impact is planned. Table 1:

Frequency of antibody responses against a panel of 45 antigens in patients of the EORTC18961 trial

Number of positive responses Frequencies (%)
none 22 (2.3%)
1 52 (5.4%)
2-5 253 (26.4%)
6-10 257 (26.8%)
11-20 233 (24.3%)
21-30 100 (10.4%)
31-46 42 (4.4%)

Clinical trial identification

Ancillary study of the EORTC 18961 trial


All authors have declared no conflicts of interest.