Dataset Information

MRI surrogates for molecular subgroups of medulloblastoma.

ABSTRACT: Recently identified molecular subgroups of medulloblastoma have shown potential for improved risk stratification. We hypothesized that distinct MR imaging features can predict these subgroups.All patients with a diagnosis of medulloblastoma at one institution, with both pretherapy MR imaging and surgical tissue, served as the discovery cohort (n = 47). MR imaging features were assessed by 3 blinded neuroradiologists. NanoString-based assay of tumor tissues was conducted to classify the tumors into the 4 established molecular subgroups (wingless, sonic hedgehog, group 3, and group 4). A second pediatric medulloblastoma cohort (n = 52) from an independent institution was used for validation of the MR imaging features predictive of the molecular subtypes.Logistic regression analysis within the discovery cohort revealed tumor location (P < .001) and enhancement pattern (P = .001) to be significant predictors of medulloblastoma subgroups. Stereospecific computational analyses confirmed that group 3 and 4 tumors predominated within the midline fourth ventricle (100%, P = .007), wingless tumors were localized to the cerebellar peduncle/cerebellopontine angle cistern with a positive predictive value of 100% (95% CI, 30%-100%), and sonic hedgehog tumors arose in the cerebellar hemispheres with a positive predictive value of 100% (95% CI, 59%-100%). Midline group 4 tumors presented with minimal/no enhancement with a positive predictive value of 91% (95% CI, 59%-98%). When we used the MR imaging feature-based regression model, 66% of medulloblastomas were correctly predicted in the discovery cohort, and 65%, in the validation cohort.Tumor location and enhancement pattern were predictive of molecular subgroups of pediatric medulloblastoma and may potentially serve as a surrogate for genomic testing.

SUBMITTER: Perreault S

PROVIDER: S-EPMC4819007 | biostudies-literature | 2014 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

MRI surrogates for molecular subgroups of medulloblastoma.

Perreault S S Ramaswamy V V Achrol A S AS Chao K K Liu T T TT Shih D D Remke M M Schubert S S Bouffet E E Fisher P G PG Partap S S Vogel H H Taylor M D MD Cho Y J YJ Yeom K W KW

AJNR. American journal of neuroradiology 20140515 7

<h4>Background and purpose</h4>Recently identified molecular subgroups of medulloblastoma have shown potential for improved risk stratification. We hypothesized that distinct MR imaging features can predict these subgroups.<h4>Materials and methods</h4>All patients with a diagnosis of medulloblastoma at one institution, with both pretherapy MR imaging and surgical tissue, served as the discovery cohort (n = 47). MR imaging features were assessed by 3 blinded neuroradiologists. NanoString-based a ...[more]

PMID: 24831600

Similar Datasets

Project description:BACKGROUND AND PURPOSE:Distinct molecular subgroups of pediatric medulloblastoma confer important differences in prognosis and therapy. Currently, tissue sampling is the only method to obtain information for classification. Our goal was to develop and validate radiomic and machine learning approaches for predicting molecular subgroups of pediatric medulloblastoma. MATERIALS AND METHODS:In this multi-institutional retrospective study, we evaluated MR imaging datasets of 109 pediatric patients with medulloblastoma from 3 children's hospitals from January 2001 to January 2014. A computational framework was developed to extract MR imaging-based radiomic features from tumor segmentations, and we tested 2 predictive models: a double 10-fold cross-validation using a combined dataset consisting of all 3 patient cohorts and a 3-dataset cross-validation, in which training was performed on 2 cohorts and testing was performed on the third independent cohort. We used the Wilcoxon rank sum test for feature selection with assessment of area under the receiver operating characteristic curve to evaluate model performance. RESULTS:Of 590 MR imaging-derived radiomic features, including intensity-based histograms, tumor edge-sharpness, Gabor features, and local area integral invariant features, extracted from imaging-derived tumor segmentations, tumor edge-sharpness was most useful for predicting sonic hedgehog and group 4 tumors. Receiver operating characteristic analysis revealed superior performance of the double 10-fold cross-validation model for predicting sonic hedgehog, group 3, and group 4 tumors when using combined T1- and T2-weighted images (area under the curve = 0.79, 0.70, and 0.83, respectively). With the independent 3-dataset cross-validation strategy, select radiomic features were predictive of sonic hedgehog (area under the curve = 0.70-0.73) and group 4 (area under the curve = 0.76-0.80) medulloblastoma. CONCLUSIONS:This study provides proof-of-concept results for the application of radiomic and machine learning approaches to a multi-institutional dataset for the prediction of medulloblastoma subgroups.

Project description:Recurrent medulloblastoma is a therapeutic challenge because it is almost always fatal. Studies have confirmed that medulloblastoma consists of at least four distinct subgroups. We sought to delineate subgroup-specific differences in medulloblastoma recurrence patterns.We retrospectively identified a discovery cohort of all recurrent medulloblastomas at the Hospital for Sick Children (Toronto, ON, Canada) from 1994 to 2012 (cohort 1), and established molecular subgroups using a nanoString-based assay on formalin-fixed paraffin-embedded tissues or frozen tissue. The anatomical site of recurrence (local tumour bed or leptomeningeal metastasis), time to recurrence, and survival after recurrence were assessed in a subgroup-specific manner. Two independent, non-overlapping cohorts (cohort 2: samples from patients with recurrent medulloblastomas from 13 centres worldwide, obtained between 1991 and 2012; cohort 3: samples from patients with recurrent medulloblastoma obtained at the NN Burdenko Neurosurgical Institute [Moscow, Russia] between 1994 and 2011) were analysed to confirm and validate observations. When possible, molecular subgrouping was done on tissue obtained from both the initial surgery and at recurrence.Cohort 1 consisted of 30 patients with recurrent medulloblastomas; nine with local recurrences, and 21 with metastatic recurrences. Cohort 2 consisted of 77 patients and cohort 3 of 96 patients with recurrent medulloblastoma. Subgroup affiliation remained stable at recurrence in all 34 cases with available matched primary and recurrent pairs (five pairs from cohort 1 and 29 pairs from cohort 2 [15 SHH, five group 3, 14 group 4]). This finding was validated in 17 pairs from cohort 3. When analysed in a subgroup-specific manner, local recurrences in cohort 1 were more frequent in SHH tumours (eight of nine [89%]) and metastatic recurrences were more common in group 3 and group 4 tumours (17 of 20 [85%] with one WNT, p=0·0014, local vs metastatic recurrence, SHH vs group 3 vs group 4). The subgroup-specific location of recurrence was confirmed in cohort 2 (p=0·0013 for local vs metastatic recurrence, SHH vs group 3 vs group 4,), and cohort 3 (p<0·0001). Treatment with craniospinal irradiation at diagnosis was not significantly associated with the anatomical pattern of recurrence. Survival after recurrence was significantly longer in patients with group 4 tumours in cohort 1 (p=0·013) than with other subgroups, which was confirmed in cohort 2 (p=0·0075), but not cohort 3 (p=0·70).Medulloblastoma does not change subgroup at the time of recurrence, reinforcing the stability of the four main medulloblastoma subgroups. Significant differences in the location and timing of recurrence across medulloblastoma subgroups have potential treatment ramifications. Specifically, intensified local (posterior fossa) therapy should be tested in the initial treatment of patients with SHH tumours. Refinement of therapy for patients with group 3 or group 4 tumours should focus on metastases.

Project description:PurposeMedulloblastoma comprises four distinct molecular subgroups: WNT, SHH, Group 3, and Group 4. Current medulloblastoma protocols stratify patients based on clinical features: patient age, metastatic stage, extent of resection, and histologic variant. Stark prognostic and genetic differences among the four subgroups suggest that subgroup-specific molecular biomarkers could improve patient prognostication.Patients and methodsMolecular biomarkers were identified from a discovery set of 673 medulloblastomas from 43 cities around the world. Combined risk stratification models were designed based on clinical and cytogenetic biomarkers identified by multivariable Cox proportional hazards analyses. Identified biomarkers were tested using fluorescent in situ hybridization (FISH) on a nonoverlapping medulloblastoma tissue microarray (n = 453), with subsequent validation of the risk stratification models.ResultsSubgroup information improves the predictive accuracy of a multivariable survival model compared with clinical biomarkers alone. Most previously published cytogenetic biomarkers are only prognostic within a single medulloblastoma subgroup. Profiling six FISH biomarkers (GLI2, MYC, chromosome 11 [chr11], chr14, 17p, and 17q) on formalin-fixed paraffin-embedded tissues, we can reliably and reproducibly identify very low-risk and very high-risk patients within SHH, Group 3, and Group 4 medulloblastomas.ConclusionCombining subgroup and cytogenetic biomarkers with established clinical biomarkers substantially improves patient prognostication, even in the context of heterogeneous clinical therapies. The prognostic significance of most molecular biomarkers is restricted to a specific subgroup. We have identified a small panel of cytogenetic biomarkers that reliably identifies very high-risk and very low-risk groups of patients, making it an excellent tool for selecting patients for therapy intensification and therapy de-escalation in future clinical trials.

Dataset Information

MRI surrogates for molecular subgroups of medulloblastoma.

Publications

MRI surrogates for molecular subgroups of medulloblastoma.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets