Validation of the HGG-IMMUNO-RPA model in a new cohort of patients with relapsed HGG treated by adjuvant postoperative DC vaccination

Introduction: We run the HGG-IMMUNO-2003 cohort comparison study to treat children and adults with relapsed HGG with immunotherapy after new (sub)total resection. We published recently the importance of the clinical risk profiles for prognostic patient counseling for participants to the immunotherapy trials (PubMedID 22565485). The data were derived from patients treated in cohorts A-D. We now aim to validate the HGG-IMMUNO RPA model in the E cohort of patients. Methods: Patients were included after subtotal resection of the relapsed tumor, confirmed with postoperative MRI, central pathology confirmation and quick tapering off of corticosteroids. Tumors were conserved sterile, dry and frozen at -80 °C for preparing the lysate with snap freeze/thaw cycles and subsequent 60 Gy irradiation. Dendritic cells (DC) were differentiated out of monocytes, obtained with elutriation of the leukapheresis product, in tissue bags in the presence of GM-CSF/IL- 4 for 7 days. DCs were loaded with lysate (HGG-L) and matured with IL-1b/TNF-a for two days. They were injected intradermally in imiquimod-prepared skin at weeks 1/2/3/4/8/12/16 and then each 3 months. Results: HGG-IMMUNO RPA data were available for 92 adults aged 17 – 75y (median 49). Patients were treated postoperatively with a median of 6 vaccines (1 – 23). After up to 4 weekly induction vaccines with DCm-HGG-L, boost vaccines consist of only HGG-L (weeks 8/12/6/+12). Median PFS and OS were 3.5 and 11 months with a 2-y OS of 17%. Patients were classified in HGG-IMMUNO RPA classes I (n = 8), II (n = 23), III (n = 45) and IV (n = 16). Median PFS and OS were respectively 3.7, 2.9, 4.1, 3.3 and 16,3, 14.7, 10.4 and 9.4 months (log-rank: p = 0.048). Conclusion: Although long-term follow up of the E cohort is still short, the HGG-IMMUNO RPA model could already significantly distinguish the median OS for patient subgroups with relapsed HGG treated inthe E cohort. These data together with the published data strongly support the HGG-IMMUNO RPA model to describe patient groups treated with immunotherapy, in order to appropriately evaluate the effects of immunotherapy. The model allows a more accurate interpretation of immunotherapy treatment results amongst different research groups and opens the possibility for appropriate stratification of randomised clinical trials in future. Similar modelling is urgently needed for pediatric patients with HGG.