Jane Cryan

Integrated Molecular Meta-Analysis of 1,000 Pediatric High-Grade and Diffuse Intrinsic Pontine Glioma

Summary We collated data from 157 unpublished cases of pediatric high-grade glioma and diffuse intrinsic pontine glioma and 20 publicly available datasets in an integrated analysis of >1,000 cases. We identified co-segregating mutations in histone-mutant subgroups including loss of FBXW7 in H3.3G34R/V, TOP3A rearrangements in H3.3K27M, and BCOR mutations in H3.1K27M. Histone wild-type subgroups are refined by the presence of key oncogenic events or methylation profiles more closely resembling lower-grade tumors. Genomic aberrations increase with age, highlighting the infant population as biologically and clinically distinct. Uncommon pathway dysregulation is seen in small subsets of tumors, further defining the molecular diversity of the disease, opening up avenues for biological study and providing a basis for functionally defined future treatment stratification.

Clinical implementation of integrated whole-genome copy number and mutation profiling for glioblastoma

BACKGROUND Multidimensional genotyping of formalin-fixed paraffin-embedded (FFPE) samples has the potential to improve diagnostics and clinical trials for brain tumors, but prospective use in the clinical setting is not yet routine. We report our experience with implementing a multiplexed copy number and mutation-testing program in a diagnostic laboratory certified by the Clinical Laboratory Improvement Amendments. METHODS We collected and analyzed clinical testing results from whole-genome array comparative genomic hybridization (OncoCopy) of 420 brain tumors, including 148 glioblastomas. Mass spectrometry-based mutation genotyping (OncoMap, 471 mutations) was performed on 86 glioblastomas. RESULTS OncoCopy was successful in 99% of samples for which sufficient DNA was obtained (n = 415). All clinically relevant loci for glioblastomas were detected, including amplifications (EGFR, PDGFRA, MET) and deletions (EGFRvIII, PTEN, 1p/19q). Glioblastoma patients ≤40 years old had distinct profiles compared with patients >40 years. OncoMap testing reliably identified mutations in IDH1, TP53, and PTEN. Seventy-seven glioblastoma patients enrolled on trials, of whom 51% participated in targeted therapeutic trials where multiplex data informed eligibility or outcomes. Data integration identified patients with complete tumor suppressor inactivation, albeit rarely (5% of patients) due to lack of whole-gene coverage in OncoMap. CONCLUSIONS Combined use of multiplexed copy number and mutation detection from FFPE samples in the clinical setting can efficiently replace singleton tests for clinical diagnosis and prognosis in most settings. Our results support incorporation of these assays into clinical trials as integral biomarkers and their potential to impact interpretation of results. Limited tumor suppressor variant capture by targeted genotyping highlights the need for whole-gene sequencing in glioblastoma.

Residual Convolutional Neural Network for Determination of IDH Status in Low- and High-grade Gliomas from MR Imaging

Purpose: Isocitrate dehydrogenase (IDH) mutations in glioma patients confer longer survival and may guide treatment decision making. We aimed to predict the IDH status of gliomas from MR imaging by applying a residual convolutional neural network to preoperative radiographic data. Experimental Design: Preoperative imaging was acquired for 201 patients from the Hospital of University of Pennsylvania (HUP), 157 patients from Brigham and Womens Hospital (BWH), and 138 patients from The Cancer Imaging Archive (TCIA) and divided into training, validation, and testing sets. We trained a residual convolutional neural network for each MR sequence (FLAIR, T2, T1 precontrast, and T1 postcontrast) and built a predictive model from the outputs. To increase the size of the training set and prevent overfitting, we augmented the training set images by introducing random rotations, translations, flips, shearing, and zooming. Results: With our neural network model, we achieved IDH prediction accuracies of 82.8% (AUC = 0.90), 83.0% (AUC = 0.93), and 85.7% (AUC = 0.94) within training, validation, and testing sets, respectively. When age at diagnosis was incorporated into the model, the training, validation, and testing accuracies increased to 87.3% (AUC = 0.93), 87.6% (AUC = 0.95), and 89.1% (AUC = 0.95), respectively. Conclusions: We developed a deep learning technique to noninvasively predict IDH genotype in grade II–IV glioma using conventional MR imaging using a multi-institutional data set. Clin Cancer Res; 24(5); 1073–81. ©2017 AACR.

Spinal epidural angiolipoma.

A 37-year-old female patient was referred to the hospital with pelvic pain and progressive lower limb weakness of 6 months’ duration. This weakness began during the final trimester of her pregnancy and had continued after the birth of her child. There was no associated sphincter disturbance. On questioning, the patient reported experiencing some right lower extremity weakness approximately 12 months previously, for which no cause was found despite normal brain magnetic resonance imaging and cervical spine imaging. Physical examination revealed positive Babinski signs and significant reduced power of hip flexion bilaterally, with no loss of power detected in other lower limb muscle groups. A sensory level was found at the T7 and T8 levels. A 3 Tesla magnetic resonance imaging examination of the cervical and thoracic spine, with and without contrast, was performed (Figs. 1–2). This demonstrated an avidly enhancing tumor mass in the posterior epidural space, extending craniocaudally from the level of T6 to T9 for a distance of 7 cm. Enhancing epidural soft tissue was detected, extending into the left T7–T8 neural exit foramen. There was spinal cord compression, most marked at T7 and T8. The tumor mass was surgically resected. Histopathologic evaluation confirmed a spinal angiolipoma (Figs. 3–4). Spinal angiolipoma accounts for 0.04%–1.2% of all spinal axis tumors. It predominantly occurs in the mid-thoracic region. Pregnancy is an aggravating factor in some patients. Prognosis post-resection is excellent [1].

A PROGRESSIVE MULTIFOCAL NEUROLOGICAL SYNDROME IN A 42-YEAR-OLD WOMAN

This is the case of a 42-year-old female who presented with transient dizziness. Her symptoms and signs progressed to include dysarthria, ataxia and cognitive decline over 2 years, such that she was unable to care for herself. She died 4 years after first presentation without a diagnosis. Investigations revealed a normochromic normocytic anaemia. Cerebrospinal fluid was normal. Serial computed tomography brain showed a wedge-shaped frontal infarct but no progressive changes. Examination at autopsy showed discoloration of the gray and white matter of the brain and spinal cord.Microscopy of leptomeningeal and parenchymal vessels showed they were filled with atypical B lymphocytes confined to the intravascular space with multiple infarcts in the brain, cerebellum and spinal cord. A diagnosis of intravascular B cell lymphoma was made and is discussed.

Machine learning: a useful radiological adjunct in determination of a newly diagnosed glioma’s grade and IDH status

IntroductionMachine learning methods have been introduced as a computer aided diagnostic tool, with applications to glioma characterisation on MRI. Such an algorithmic approach may provide a useful adjunct for a rapid and accurate diagnosis of a glioma. The aim of this study is to devise a machine learning algorithm that may be used by radiologists in routine practice to aid diagnosis of both: WHO grade and IDH mutation status in de novo gliomas.MethodsTo evaluate the status quo, we interrogated the accuracy of neuroradiology reports in relation to WHO grade: grade II 96.49% (95% confidence intervals [CI] 0.88, 0.99); III 36.51% (95% CI 0.24, 0.50); IV 72.9% (95% CI 0.67, 0.78). We derived five MRI parameters from the same diagnostic brain scans, in under two minutes per case, and then supplied these data to a random forest algorithm.ResultsMachine learning resulted in a high level of accuracy in prediction of tumour grade: grade II/III; area under the receiver operating characteristic curve (AUC) = 98%, sensitivity = 0.82, specificity = 0.94; grade II/IV; AUC = 100%, sensitivity = 1.0, specificity = 1.0; grade III/IV; AUC = 97%, sensitivity = 0.83, specificity = 0.97. Furthermore, machine learning also facilitated the discrimination of IDH status: AUC of 88%, sensitivity = 0.81, specificity = 0.77.ConclusionsThese data demonstrate the ability of machine learning to accurately classify diffuse gliomas by both WHO grade and IDH status from routine MRI alone—without significant image processing, which may facilitate usage as a diagnostic adjunct in clinical practice.

Temporal stability of MGMT promoter methylation in glioblastoma patients undergoing STUPP protocol

Epigenetic silencing of O-6-methylguanine-DNA methyltransferase (MGMT) promoter via methylation in a glioblastoma (GBM), has been correlated with a more favourable response to alkylating chemotherapeutic agents such as temozolomide. The use of global methylation surrogates such as Long Interspersed Nucleotide Element 1 (LINE1) may also be valuable in order to fully understand these highly heterogeneous tumours. In this study, we analysed both original and recurrent GBMs in 22 patients (i.e. 44 tumours), for both MGMT and LINE1 methylation status. In the 22 patients: 14 (63.6%) displayed MGMT methylation stability in the recurrent GBM versus 8 (36.4%), with instability of methylation status. No significant differences in overall and progression free survival was evident between these two groups. LINE1 methylation status remained stable for 12 (54.5%) of recurrent GBM patients versus 9 (41%) of the patients with instability in LINE1 methylation status (p = 0.02), resulting in an increase in overall survival of the stable LINE1 group (p = 0.04). The results obtained demonstrated major epigenetic instability of GBMs treated with temozolomide as part of the STUPP protocol. GBMs appear to undergo selective evolution post-treatment, and have the ability to recur with a newly reprogrammed epigenetic status. Selective targeting of the altered epigenomes in recurrent GBMs may facilitate the future development of both prognostic biomarkers and enhanced therapeutic strategies.

An Irish case of limb-girdle muscular dystrophy 2I with structural eye involvement.

1. Medina PJ, Adams VR. PD-1 Pathway inhibitors: immuno-oncology agents for restoring antitumor immune responses. Pharmacotherapy 2016;36: 317–334. 2. Loochtan AI, Nickolich MS, Hobson-Webb LD. Myasthenia gravis associated with ipilimumab and nivolumab in the treatment of small cell lung cancer. Muscle Nerve 2015;52:307–308. 3. Shirai T, Sano T, Kamijo F, Saito N, Miyake T, Kodaira M, et al. Acetylcholine receptor binding antibody-associated myasthenia gravis and rhabdomyolysis induced by nivolumab in a patient with melanoma. Jpn J Clin Oncol 2016;46:86–88. 4. Johnson DB, Saranga-Perry V, Lavin PJM, Burnette WB, Clark SW, Uskavitch DR, et al. Myasthenia gravis induced by ipilimumab in patients with metastatic melanoma. J Clin Oncol 2015;33:122–124. 5. Liao B, Shroff S, Kamiya-Matsuoka C, Tummala S. Atypical neurological complications of ipilimumab therapy in patients with metastatic melanoma. Neuro Oncol 2014;16:589–593. 6. Zimmer L, Goldinger SM, Hofmann L, Loquai C, Ugurel S, Thomas I, et al. Neurological, respiratory, musculoskeletal, cardiac and ocular side-effects of anti-PD-1 therapy. Eur J Cancer 2016;60:210–225.

Ten years on: genetic screening for mitochondrial disease in Ireland.

Mitochondrial DNA (mtDNA) analysis is centralized in the Department of Neuropathology, Beaumont Hospital. Services offered include analysis of common mtDNA point mutations, large scale mtDNA deletions/rearrangements, and sequencing of the nuclear gene POLG. The aims of this study were to audit the mtDNA diagnostic service over a 10-year period, to determine appropriate use of the service, and to improve efficient use of the service by devising a requisition form that includes diagnostic algorithms. Between July 2002 and October 2013, 716 samples were received for analysis. Overall, the number of confirmed mutations was low. Lack of diagnostic algorithms may result in expansive, unrefined requests, leading to costly investigations. We introduced a requisition form that extracts clinical, biochemical, and pathological data prior to analysis. With this information, diagnostic algorithms can be applied to select the most relevant mutations for initial analysis and also to increase the incidence of mutation detection.

Two Cases of Intraventricular Hemorrhage in Young Patients

We present 2 cases of intraventricular hemorrhage (IVH) in young patients. Case 1 is a 33-year-old man who presented with collapse after acute onset occipital headache. On admission to hospital brainstem activity was absent. CT brain (figure 1a) showed a large acute intraventricular hemorrhage, underlying intraventricular mass and obstructive hydrocephalus. Autopsy examination confirmed massive intraventricular hemorrhage. Coronal section of the brain (Figure 1b) shows destruction of the diencephalon and underlying tumor fragments. Case 2 is a 28-year-old woman who presented with recent onset headache. T1W MR post contrast (Figure 1c) showed an enhancing tumor involving the 3 ventricle and right thalamus. She underwent burrhole biopsy during which there was significant intra-tumoral hemorrhage. She did not regain consciousness post-operatively and imaging confirmed a large thalamic hemorrhage.