Background: Patients with glioblastomas have a dismal prognosis, and there is no circulating predictive or prognostic biomarker. Circulating progastrin, hPG80, is a tumor-promoting peptide present in the blood of patients with various cancers that has been shown to have prognostic value. We evaluated the prognostic value of plasma hPG80 in patients with isocitrate dehydrogenase-wild type glioblastoma after surgery. Patients and methods: A multicentric retrospective study in glioblastoma patients treated with standard radiochemotherapy was conducted. The hPG80 levels were measured in plasma EDTA samples collected after surgery with an ELISA DxPG80.lab kit (Biodena Care, Montpellier, France), which has a detection threshold of 1.2 pM. The relationship between post-operative hPG80 plasma levels, in combination with other known prognostic factors, and patients' progression-free survival (PFS) and overall survival (OS) was evaluated. Results: Sixty-nine patients were assessable. Plasma samples were collected after tumor biopsy (B), partial resection (PR), and complete resection (CR) for 22, 25, and 22 patients, respectively. At a median concentration of 5.37 pM (interquartile range 0.00-13.90 pM), hPG80 was detected in 48 (70%) patients (hPG80+). CR was associated with significant lower values of hPG80 levels: the median value was 0.7 versus 9.1 pM for PR (P = 0.02) and 8.3 pM for B (P = 0.004). The hPG80 detection rate was also significantly lower: 50% (CR) versus 72% (PR) versus 86% (B) (P = 0.005). The median follow-up was 39 months [22.4 months-not reached]. hPG80 post-operative detection was associated with numerically shorter PFS (6.4 versus 9.4 months, P = 0.13) and OS (14.5 versus 20.9 months, P = 0.11). In multivariate analysis, hPG80 was a prognostic factor for OS (P = 0.034). Conclusions: Circulating hPG80 could serve as a new prognostic biomarker after surgery in patients with glioblastoma treated with radio-chemotherapy.

Glioblastoma is a very aggressive disease with a poor prognosis. The MGMT gene encodes an enzyme involved in DNA repair, the epigenetic silencing (by methylation) of which confers a better prognostic for GBM patients. We hypothesized that administration of folic acid could restore a methylation of MGMT promoter gene in non-methylated GBM and thus increase sensitivity to temozolomide combined with radiotherapy. FOLAGLI is a non-randomized, phase 1 open study, with dose escalation (3+3 design) and expansion cohort of folic acid in combination with temozolomide and radiotherapy for newly diagnosed glioblastoma harboring non-methylated MGMT gene promoter detected by semi quantitative methyl-specific PCR. The study assessed the tolerability, pharmacodynamics and efficacy of folic acid dose escalation from 5 to 60 mg daily, given 30 minutes before temozolomide. Between March 2013 and March 2019, 90 patients were screened of whom 66 had a MGMT promoter methylated tumor. Finally, 24 patients were enrolled and one patient withdrew his consent. Median age was 56 years old [range 42-73]. No dose limiting toxicity was reported in the phase I part. The recommended dose for the expansion phase was 60 mg. Grade 1-2 treatment-related adverse events (AEs) occurring in ≥2 patients included asthenia (n=6), lymphopenia (n=2), nausea (n=3) and vomiting (n=3), alopecia (n=2) and anemia (n=2). Grade III AEs related to treatment occurring in ≥1 patients included lymphopenia (n=2), thrombocytopenia (n=1), neutropenia (n=1), dehydration (n=1) and hypoacousia (n=1). No grade IV AES occurred. With a median follow-up of 18.3 [8.5-65.5] months, median progression free survival (PFS) was 7.9m [7.1-8.3]. The median overall survival (OS) was 17.1m [12.6-24.5] and 6, 12 and 24 months OS rate were 100%, 79.8% [54.6-91.9] and 28.2 [9.2-51.2] respectively. For the eight patients with ctDNA monitoring, the combination of folic acid with temozolomide showed restoration of methylation on of the promoter of MGMT detected in ctDNA. Preliminary safety and efficacy results suggest that folic acid combined with temozolomide and radiotherapy is well tolerated and feasible.

L’essai national randomisé multicentrique Polca vise à déterminer si le traitement des oligodendrogliomes anaplasiques nouvellement diagnostiqués par six cycles de chimiothérapie associant la procarbazine, la carmustine et la vincristine (PCV) améliore la probabilité de survie sans détérioration des fonctions cognitives (objectif primaire) comparativement au traitement standard (radiothérapie conformationnelle avec modulation d’intensité [RCMI] suivie de six cycles de PCV). Nous rapportons les résultats de la revue rétrospective des dossiers techniques des patients randomisés au 09/03/2018 afin d’évaluer le respect des recommandations du protocole. Onze institutions ont adressé le dossier technique de radiothérapie de 40 patients. Les examens d’imagerie et les dossiers techniques de radiothérapie ont été revus par au moins trois experts indépendants, membres du Granocef. L’évaluation des contours et de la dosimétrie a été réalisée à l’aide du logiciel Artiview® (Aquilab SAS, France). Trente-quatre des 40 dossiers (72,3 %) étaient évaluables. Dix-huit dossiers techniques étaient conformes aux exigences du protocole (53 %), trois ont été considérés comme acceptables (9 %), et 13 dossiers (38 %) présentaient des déviations majeures. Ces déviations majeures étaient liées à la définition des volumes cibles (dix cas), à la délinéation des organes à risque (sept cas), à l’étude dosimétrique (deux cas), aux séquences d’IRM utilisées (deux cas) et au non emploi de la RCMI (un cas). Cette analyse montre un nombre significatif (38 %) de déviations majeures au protocole de radiothérapie dans l’essai Polca. Ces déviations ont un potentiel impact sur le contrôle tumoral mais aussi sur la survenue d’effets secondaires. Il est donc proposé par souci d’amélioration un dummy-run de délinéation, la rédaction d’un document listant les principales déviations, une participation plus large des oncologues radiothérapeutes aux réunions du Granocef. Pour les futures études, une évaluation prospective individuelle devrait être réalisée avant l’irradiation.

Purpose: Reports of a glioblastoma arising in a previously irradiated field are rare in the literature, even more so in the conus medullaris of the spinal cord. Method: Case report and review of the literature reporting other radiation-induced intra-medullary glioblastomas. Results: We report a case of glioblastoma of the conus, which subsequently metastasized to the brain, arising in a 45 years old man, nine years after treatment for Hodgkin's lymphoma, which had included the administration of 41 Gy to the brain and spinal cord. Conclusion: Despite a well-conducted treatment by several lines of chemotherapy, the pronostic of radiationinduced glioblastoma is poor especially since these patients often cannot benefit from a re-irradiation due to the maximum dose supported by the spinal cord. Implications of cancer survivors: Associations are now established between therapeutic exposures and specific complications but considerable inter-individual variability is observed for a given therapeutic exposure. In this context, identification of genetic susceptibilities for specific treatment-associated late effect is a very promising future approach. Improvement in our knowledge on genetic susceptibilities may help define more personalized primary therapies that weigh treatment efficacy with the risk of late complications.

La chirurgie conservatrice, suivie d’une radiothérapie externe mammaire, est considérée comme le traitement standard du cancer du sein localisé. Elle permet un excellent taux de contrôle local à 10ans, avec un taux de rechute de 6 %. Depuis ces dernières années l’irradiation partielle accélérée s’est développée pour réduire la durée et le volume d’irradiation. La radiothérapie peropératoire en fait partie et offre une excellente adaptation au volume cible tumoral tout en épargnant de façon importante l’irradiation des tissus sains. Cette revue a pour objectif de décrire les différentes techniques de radiothérapie peropératoire, d’en évaluer l’efficacité clinique et la tolérance, de montrer sa potentielle indication pour une population de patientes très sélectionnée et d’évoquer enfin son avenir dans le cadre de la recherche clinique.

IDH1/2 mutations and 1p/19q codeletion occur frequently in anaplastic gliomas and are prognostic factors. We combined these two biomarkers to stratify patients treated for anaplastic oligodendroglioma (AO). 43 consecutive WHO AO were selected. We combined immunohistochemistry (IHC) with the monoclonal antibody mIDH1R132H and DNA sequencing of IDH1 and IDH2 genes. Fluorescence in situ hybridization was carried out to evaluate 1p/19q codeletion. These biomarkers were correlated with progression-free survival (PFS) and overall survival (OS). IDH1/IDH2 mutations occurred in 23/43 (54 %) patients: 20/43 IDH1-R132H mutation in IHC, 2/43 IDH1-R132G mutation and 1/43 IDH2-R172K mutation identified by DNA sequencing. 1p/19q codeletion was detected for 23/43 patients. With median follow-up of 19 months (range 1.4–128), median PFS and OS were 22 and 35 months respectively. IDH1/IDH2 mutations were strongly associated with improved PFS and OS: 5-year PFS was 86 versus 6 % and 5-year OS was 91 versus 9 % for patients with IDH1/IDH2 mutations versus wild-type IDH respectively. In multivariate analyses, IDH1/IDH2 mutations and 1p/19q loss were independent prognostic factors. Three groups with distinct prognostic features were identified: patients with IDH1/2 mutations and 1p/19q loss (median PFS, median OS not reached), patients with IDH1/2 mutations or 1p/19q loss (median PFS: 22 months, median OS: 30 months), and patients without IDH1/2 mutations nor 1p/19q loss with a bad prognosis (median PFS: 8.6 months, median OS: 9.9 months). Combining two biomarkers, IDH1/2 and 1p/19q codeletion, makes it possible to stratify AO in three groups with very distinct prognostic features.

2020 Background: IDH1 mutations have been indentified in about 70% of WHO grade 2 and 3 astrocytomas and oligodendrogliomas, conferring a better prognosis. The most frequent mutation leads to a specific amino-acid change from arginine to histidine at codon 132 (c.395G>A, p.R132H). We compared mIDH1 R132H immunohistochemistry, allele specific PCR and DNA sequencing for determination of IDH1 status. Methods: We performed a retrospective study of 91 WHO grade 2 (n=43; mean 40.6 years) and grade 3 (n=48; mean 50 years) oligodendrogliomas, diagnosed in the Neurosurgery Department of Nantes. Immunohistochemistry was performed with mouse monoclonal mIDH1 R132H (gift von Dr von Deimling, Heidelberg, Germany). DNA was extracted from FFPE sections following macrodissection using an iPrep system. A fragment of exon 4 (122 bp) spanning the sequence encoding the catalytic domain of IDH1 including codon 132 was amplified and sequenced using standard conditions. Allele specific amplification was performed using forward primers with variations in their 3′ nucleotides such that each was specific for the wild type or the mutated variant. Results: DNA could be amplified in 90 cases. IDH1 mutations were found in 55/90 patients (61.1%) by direct sequencing. Among these patients, R132H mutations was found in 47/55 patients (85.4%). The results of the allele specific PCR was totally correlated to DNA sequencing. Other mutations (p.R132C, p.R132S and pR132G) were found by DNA sequencing in 3, 3 and 2 tumors, respectively (8/55 patients, 14.6%). mIDH1 R132H immunostaining was found in the 47 patients presenting the R132H mutation (sensitivity 47/47 = 100% for this mutation). None of the tumors presenting a wild type IDH1 gene were stained (specificity 35/35 = 100%). Conclusions: Immunohistochemistry with mIDH1 R132H antibody is feasible, even in case of small biopsies, highly sensitive and specific. Because it only recognizes the most frequent mutation of IDH1 gene, immunohistochemical negative cases must then be tested by allele-specific PCR or DNA sequencing.

2002 Background: Recent studies have shown that IDH1 mutations occur frequently in WHO grade 2 and 3 gliomas. However, their prognostic and predictive impact remains unclear. Methods: Search for IDH1 mutation by mIDH1 R132H immunohistochemistry (IHC) and by DNA sequencing were performed in a series of 43 WHO grade 3 oligodendrogliomas. IHC was performed with mouse monoclonal mIDH1 R132H (gift von Dr von Deimling, Heidelberg, Germany). DNA was extracted from FFPE sections following macrodissection using an iPrep system. A fragment of exon 4 (122 bp) spanning the sequence encoding the catalytic domain of IDH1 including codon 132 was amplified and sequenced using standard conditions. The impact of IDH1 mutations on progression free survival (PFS) and overall survival (OS) was analyzed. Results: Mean age at diagnosis was 51 years, ratio male/female (28/15) and Karnovsky Performans Status (KPS): [<70] n=3, [70-100] n=40. Diagnosis was obtained by surgical biopsy (n=13, 30%), suboptimal resection (n=20, 47%) and optimal resection (n=10, 23%). IDH1 R132H mutation was identified by IHC in 20/43 (47%) patients and R132G mutation by DNA sequencing in 2 (9%) of the 23 remaining patients. Finally, IDH1 mutation was identified in 22/43 (51%) patients. First-line treatments following surgery were chemotherapy alone (n=5), radiotherapy alone (n=5) and sequential or concomitant radio-chemotherapy (n=33) according to PCV regimen (n=29) or Stupp regimen (n=4). With a median follow up of 19 months (range 1.4-128), median PFS and OS were 22 and 35 months. IDH1 mutation was statistically associated with prolonged PFS and OS in univariate analysis: 1-year PFS rate 100% vs 31%, 3-years PFS rate 86% vs 6%, 1-year OS rate 100% vs 46% and 3-years OS 92% vs 9% for IDH1 mutated patients versus wild type respectively. Median OS was 108 vs 12 months (p=0.00002, Hazard Ratio 0.1 IC95 0-0.3) for IDH1 mutated patients vs wild type respectively. Conclusions: In this treated group of grade 3 oligodendrogliomas, the presence of IDH1 mutation assessed by IHC and DNA sequencing is found to carry a very strong predictive impact on PFS and OS.

Medulloblastoma is a very rare disease in adult population. Standard treatment include craniospinal irradiation (usually 36 Gy) whatever the risk level is. In children population risk-adapted radiation doses is employed to reduce the neurotoxicity. Because any clinical protocol for adult medulloblastoma is available, toxicity of cranio-spinal irradiation is not yet evaluated in this population. Among 253 adult patients treated for medulloblastoma this multicentric study assesses the social integration level after craniospinal radiation. Among 253 patients irradiated from 1978 to 2000, only patients free of disease up to 7 years after treatment were included in this study and retrospectively reviewed. Data were collected by file reviewing or mail enquiry. Patients from the 5 largest centers were reviewed. 72 were free of disease up to 7 years after radiation and available for social integration level data. At last follow-up 23 patients (32%) worked with no modification of employment. Finally half of them stopped definitively working because of late neurocognitive sequelae (median 10 years). 31% of patients got adapted job after treatment. Three patients become dependent on medical care. Hearing loss occurred in 43% of patients. Among 72 patients, 12 received serial neurocognitive testing. This study is one of the first work able to assess neurocognitive toxicity due to radiation therapy in adult medulloblastoma. This high incidence of toxicity suggested that current practice may require modifications for adults medulloblastoma treatment.

7688 Objective: To evaluate efficacy and toxicity of HDR BT in non operable endobronchial carcinoma from a retrospective multicentric study. Patients and Methods: Criteria for selection: non small cell carcinoma accessible to fiberoptic bronchoscopy, no extrabronchial extension on CT, contraindication to surgery and external radiation therapy (ERT). Statistical analysis: survival curves calculated with the Kaplan-Meier method and compared with the Logrank test; Cox model to evaluate in uni and multivariate analysis the impact on survival and complications of these parameters: location of tumor: lobar or segmental vs main stem bronchus, previous ERT vs no, total dose:= 30 Gy vs less, dose per fraction:= 5 Gy vs more, number of catheter(s):1 vs = 2. Results: Between 1991 and 2006, 226 patients from 9 radiotherapy departments were included. Main characteristics of tumors: squamous-cell histology: 96%, stage Tis: 60, T1: 153, T2: 9, Tx 4; lobar or segmental location: 91%. 51 patients (22.5%) had received ERT for previous lung cancer(s). Characteristics of HDR BT were: total dose = 30 Gy: 70%, dose per fraction = 5 Gy: 66%, 1 catheter: 46%. Dose was prescribed at 1 cm from the radius. Mean follow-up was 30.4 months (9- 116). Histologic evaluation was performed at 3 months in 137 patients. 92% had a complete response. 128 patients were died: intercurrent disease 45, local failure 35, complications 13. Two and 5-year survival: overall: 57%, 29%; specific (death of lung cancer) 81%, 56%; local- relapse free (LRF) 68%, 50%. Toxicity included 1.3% pneumothorax, hemoptysis 6.6% (5% fatal), bronchitis 20%. In univariate analysis: overall, specific and LRF survival were better for lobar or segmental location vs main stem bronchus (p=0.0001), overall and specific survival were higher with no previous ERT (p=0.006). In multivariate analysis, lobar or segmental location was associated with improved overall (p=0.0001) and LRF (p=0.003) survival. LRF survival was better in patients treated with = 2 catheters (p=0.007). No factor influence frequency of complications. Conclusion: This large retrospective study confirmed that HDRBT is efficient and safe in medically inoperable patients particulary with lobar or segmental endobronchial carcinoma. No significant financial relationships to disclose.