Prognosis of glioblastoma patients improves significantly over time interrogating historical controls

BACKGROUND

Glioblastoma (GBM) is the most common devastating primary brain cancer in adults. In our clinical practice, median overall survival (mOS) of GBM patients seems increasing over time.

METHODS

To address this observation, we have retrospectively analyzed the prognosis of 722 newly diagnosed GBM patients, aged below 70, in good clinical conditions (i.e. Karnofsky Performance Status -KPS- above 70%) and treated in our department according to the standard of care (SOC) between 2005 and 2018. Patients were divided into two groups according to the year of diagnosis (group 1: from 2005 to 2012; group 2: from 2013 to 2018).

RESULTS

Characteristics of patients and tumors of both groups were very similar regarding confounding factors (age, KPS, MGMT promoter methylation status and treatments). Follow-up time was fixed at 24 months to ensure comparable survival times between both groups. Group 1 patients had a mOS of 19 months ([17.3-21.3]) while mOS of group 2 patients was not reached. The recent period of diagnosis was significantly associated with a longer mOS in univariate analysis (HR=0.64, 95% CI [0.51 - 0.81]), p < 0.001). Multivariate Cox analysis showed that the period of diagnosis remained significantly prognostic after adjustment on confounding factors (adjusted Hazard Ratio (aHR) 0.49, 95% CI [0.36-0.67], p < 0.001).

CONCLUSION

This increase of mOS over time in newly diagnosed GBM patients could be explained by better management of potentially associated non-neurological diseases, optimization of validated SOC, better management of treatments side effects, supportive care and participation in clinical trials.

Molecular and clinical diversity in primary central nervous system lymphoma

BACKGROUND

Primary central nervous system lymphoma (PCNSL) is a rare and distinct entity within diffuse large B-cell lymphoma presenting with variable response rates probably to underlying molecular heterogeneity.

PATIENTS AND METHODS

To identify and characterize PCNSL heterogeneity and facilitate clinical translation, we carried out a comprehensive multi-omic analysis [whole-exome sequencing, RNA sequencing (RNA-seq), methylation sequencing, and clinical features] in a discovery cohort of 147 fresh-frozen (FF) immunocompetent PCNSLs and a validation cohort of formalin-fixed, paraffin-embedded (FFPE) 93 PCNSLs with RNA-seq and clinico-radiological data.

RESULTS

Consensus clustering of multi-omic data uncovered concordant classification of four robust, non-overlapping, prognostically significant clusters (CS). The CS1 and CS2 groups presented an immune-cold hypermethylated profile but a distinct clinical behavior. The 'immune-hot' CS4 group, enriched with mutations increasing the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) and nuclear factor-κB activity, had the most favorable clinical outcome, while the heterogeneous-immune CS3 group had the worse prognosis probably due to its association with meningeal infiltration and enriched HIST1H1E mutations. CS1 was characterized by high Polycomb repressive complex 2 activity and CDKN2A/B loss leading to higher proliferation activity. Integrated analysis on proposed targets suggests potential use of immune checkpoint inhibitors/JAK1 inhibitors for CS4, cyclin D-Cdk4,6 plus phosphoinositide 3-kinase (PI3K) inhibitors for CS1, lenalidomide/demethylating drugs for CS2, and enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) inhibitors for CS3. We developed an algorithm to identify the PCNSL subtypes using RNA-seq data from either FFPE or FF tissue.

CONCLUSIONS

The integration of genome-wide data from multi-omic data revealed four molecular patterns in PCNSL with a distinctive prognostic impact that provides a basis for future clinical stratification and subtype-based targeted interventions.

P11.09.B Chemotherapy is an efficient treatment in primary CNS MALT lymphoma

Background

Mucosae-associated lymphoid tissue (MALT) lymphomas are a rare and still poorly understood form of primary central nervous system lymphoma (PCNSL). The aim of this study was to better describe these tumors and their management, and to better appreciate their long-term prognosis after treatment.

Material and Methods

Adult immunocompetent patients with primary CNS MALT lymphoma (PCNSML) were retrospectively selected from the database on PCNSL of the Pitié-Salpêtrière Hospital. All cases had double-read for pathology.

Results

Eleven PCNSML representing 1.7% of the 662 PCNSL included in the database were selected, There were 9 women and two men, all immunocompetent. Their median age was 56 years (min 29; max 78). The median time from first symptoms to diagnosis was 13 months and the median KPS at diagnosis was 90. Location was dural in 8 cases, suggestive of meningioma in 7 cases and of subdural hematoma in 1 case, and parenchymal in 3 cases. The disease was unifocal/localized in 4 cases and multifocal/diffuse in 7 cases. In 3 cases, it was the second reading that allowed making the diagnosis. In first-line treatment, all patients received chemotherapy (CT): “systemic” CT, i.e usually used in systemic MALT lymphomas, in 7 cases (n=7) and “CNS” CT, i.e usually used in CNS lesions in 4 cases. CT was preceded by surgery in 4 cases. No patient received radiotherapy in first line. According to the IPCG (International PCNSL Collaborative Group) criteria, the overall response rate was 7/11 (64%). At latest news, 5 patients had a persistent contrast enhancement, stable without any treatment since a median of 57 months, raising the question of complete response despite persisting contrast enhancement. No patient developed neurotoxicity except for one of the 2 patients who subsequently received radiotherapy. The median follow-up was 109 months. The median progression-free survival was 78.0 months (95% CI [43.4;NA[) and the 10-year overall survival rate was 90% (95% CI [0.7;1[).

Conclusion

This series confirms the classical clinical and radiological presentation of PCNSML. Compared to the literature where the treatment was based on radiotherapy in almost two thirds of cases, the therapeutic approach presented here appears unconventional. This is the first series demonstrating that chemotherapy is an efficient treatment in PCNSML, with an excellent long-term outcome and the absence of neurotoxicity, and calling into question the relevance of the IPCG criteria for the evaluation of their therapeutic response.

PL1.2 Multiparametric assessment of factors influencing 2 HG accumulation in diffuse brain gliomas

BACKGROUND

2-hydroxyglutarate (2HG) can be detected non-invasively in IDH-mutant gliomas by in vivo magnetic resonance spectroscopy (MRS). We investigated factors affecting 2 HG accumulation and explored the prognostic value of 2 HG detection in IDH mutant gliomas and 2 HG variations during anti-cancer therapies.

MATERIAL AND METHODS

We prospectively scanned by MEGA-PRESS 70 glioma patients (24 before surgery and 46 IDH mutant operated glioma) and measured 2HG. CRLB cut-off was 50%. We followed up 9 IDH mutant patients during radiotherapy and chemotherapy.We analyzed radiological parameters (tumor and cystic/necrotic volumes, fractions of VOI filled with tumor, spectroscopic profile, “infiltrative” versus “expansive” morphology, contrast-enhancement) and genetic profile (IDH1, IDH2, 1p19q codeletion). 2HG concentrations in plasma, urine, and surgical obtained samples were measured by gas chromatography-mass spectrometry (GC-MS).

RESULTS

MEGA-PRESS sequence detected 2HG with a sensitivity of 95% in untreated patients, and of 69% in pre-treated patient. Positive predictive value was 100% in both groups. 2HG was lower in pre-treated IDH mutant gliomas (1.1 versus 2.3 mM, P=0.02) and decreased rapidly during radiotherapy and chemotherapy before any radiological change. 2HG detection was positively correlated with tumor volume (P=0.02), choline measurements (r=0.58 P<0.0001), cellular density (measured by restricted diffusivity, Pearson r -0.40 P=0.01), “expansive” presentation, mutated reads/total reads ratio by NGS and was inversely correlated with Myo-inositol (Pearson R -0.29 P=0.03) and cystic/necrotic areas (P=0.04). 2HG by MRS positively correlated with urine 2HG (r=0.80 P=0.003). 2 HG was higher in IDH2 mutant (4.7 versus 2.4 Mm, P=0.02) and lower in non R132H IDH1 mutant (1.12 mM P=0.004). Among IDH mutant glioma patients, 2 HG detection was associated with longer survival (HR 0.09; 95%CI 0.018–0.52).

CONCLUSION

Tumor volume, cellular density, previous radio- and chemotherapy and genetic features determine 2 HG detection in IDH mutant gliomas. 2 HG detection is associated with better outcome and can be reliably monitored during anti-cancer treatments.

OS9.2 Radiomics analysis of lower-grade gliomas, a POLA Network study

BACKGROUND

Lower-grade gliomas (LGG) are divided into three histo-molecular groups: i) IDH-wildtype, ii) IDH mutant and 1p19q intact and iii) IDH mutant and 1p19q co-deleted. The current classification has improved the clinical stratification and its reproducibility. However, LGGs are still associated with an important degree of clinical heterogeneity. We sought to analyze the cross-talk between the spatial distribution and the quantitative imaging features (radiomics) with the clinical evolution and their molecular background (radiogenomics).

MATERIAL AND METHODS

We performed a retrospective multicentric study from 4 cohorts of high-grades gliomas (POLA Network, TCGA, REMBRANDT and LGG-1p19q), totaling 900 gliomas. We performed N4 and WhiteStripe imaging corrections to standardize MRI intensities. We used ITK-SNAP to obtain a mask of the different habitats of the tumor. Then we used PyRadiomics to obtain 2616 radiomic features per sample. We used plsRcox for fitting several Cox model in high-dimensional settings. We assessed the performance of the difference Cox model with the Harrel’s concordance index. We used a Sparse Canonical Correlation analysis to analyze the spatial distribution of the tumors.

RESULTS

Radiomics features allow identification in an unsupervised manner IDH-mutant gliomas with a median AUC of 0.96 [0.92–0.98]. Interestingly, in the analysis of survival, radiomics features provided additional information to clinical or genetics covariates and the model with only radiomics features obtained a C-Index of 0.78 [0.72–0.82]. In addition, survival model with the best performance in the prediction of overall survival was the one combining radiomics, clinics and genetics features with a C-Index 0.85 [0.82–0.92] and was validated in the other cohorts. The analysis of spatial distribution showed a very strong distribution of 1p19q co-deleted oligodendrogliomas in the frontal lobes.

CONCLUSION

Radiomics features may provide additional relevant clinical information by improving the prognosis of LGG. Radiomics allow non-invasive prediction of the most relevant molecular alterations of LGG.