Integrating imaging and genomic data for the discovery of distinct glioblastoma subtypes: a joint learning approach.

Glioblastoma is a highly heterogeneous disease, with variations observed at both phenotypical and molecular levels. Personalized therapies would be facilitated by non-invasive in vivo approaches for characterizing this heterogeneity. In this study, we developed unsupervised joint machine learning between radiomic and genomic data, thereby identifying distinct glioblastoma subtypes. A retrospective cohort of 571 IDH-wildtype glioblastoma patients were included in the study, and pre-operative multi-parametric MRI scans and targeted next-generation sequencing (NGS) data were collected. L21-norm minimization was used to select a subset of 12 radiomic features from the MRI scans, and 13 key driver genes from the five main signal pathways most affected in glioblastoma were selected from the genomic data. Subtypes were identified using a joint learning approach called Anchor-based Partial Multi-modal Clustering on both radiomic and genomic modalities. Kaplan-Meier analysis identified three distinct glioblastoma subtypes: high-risk, medium-risk, and low-risk, based on overall survival outcome (p < 0.05, log-rank test; Hazard Ratio = 1.64, 95% CI 1.17-2.31, Cox proportional hazard model on high-risk and low-risk subtypes). The three subtypes displayed different phenotypical and molecular characteristics in terms of imaging histogram, co-occurrence of genes, and correlation between the two modalities. Our findings demonstrate the synergistic value of integrated radiomic signatures and molecular characteristics for glioblastoma subtyping. Joint learning on both modalities can aid in better understanding the molecular basis of phenotypical signatures of glioblastoma, and provide insights into the biological underpinnings of tumor formation and progression.

Clinical measures, radiomics, and genomics offer synergistic value in AI-based prediction of overall survival in patients with glioblastoma.

Multi-omic data, i.e., clinical measures, radiomic, and genetic data, capture multi-faceted tumor characteristics, contributing to a comprehensive patient risk assessment. Here, we investigate the additive value and independent reproducibility of integrated diagnostics in prediction of overall survival (OS) in isocitrate dehydrogenase (IDH)-wildtype GBM patients, by combining conventional and deep learning methods. Conventional radiomics and deep learning features were extracted from pre-operative multi-parametric MRI of 516 GBM patients. Support vector machine (SVM) classifiers were trained on the radiomic features in the discovery cohort (n = 404) to categorize patient groups of high-risk (OS < 6 months) vs all, and low-risk (OS ≥ 18 months) vs all. The trained radiomic model was independently tested in the replication cohort (n = 112) and a patient-wise survival prediction index was produced. Multivariate Cox-PH models were generated for the replication cohort, first based on clinical measures solely, and then by layering on radiomics and molecular information. Evaluation of the high-risk and low-risk classifiers in the discovery/replication cohorts revealed area under the ROC curves (AUCs) of 0.78 (95% CI 0.70-0.85)/0.75 (95% CI 0.64-0.79) and 0.75 (95% CI 0.65-0.84)/0.63 (95% CI 0.52-0.71), respectively. Cox-PH modeling showed a concordance index of 0.65 (95% CI 0.6-0.7) for clinical data improving to 0.75 (95% CI 0.72-0.79) for the combination of all omics. This study signifies the value of integrated diagnostics for improved prediction of OS in GBM.

Acquired Cystic Kidney Disease-associated Renal Cell Carcinoma (ACKD-RCC) Harbor Recurrent Mutations in KMT2C and TSC2 Genes.

Individuals with acquired cystic kidney disease (ACKD) in the setting of end-stage renal disease (ESRD) have a high risk of developing renal cell carcinoma (RCC). ACKD-RCC is considered a distinct renal neoplasm in the International Society of Urologic Pathologists (ISUP)-World Health Organization (WHO) classification of kidney tumors which may behave aggressively. Since its original description, there have been multiple case reports and series published; however, the pathogenesis of this neoplasm is uncertain and there is limited data on the genetic aberrations of this tumor. Herein, we present our experience with ESRD kidneys, with emphasis on ACKD-RCC, associated cysts, and the somatic mutation analysis of a subset of ACKD-RCCs using next-generation sequencing. Our data on 59 cases with ESRD that underwent nephrectomy, shows that ACKD-RCC represents more than half of the tumors (25/46; 54%) developing in ESRD, followed by papillary RCC (13; 28%). History of dialysis, male sex, and African American race were potential risk factors for developing ACKD-RCCs. Further, ACKD-RCC-like cysts are possible precursors of RCCs in the ACKD setting noted in 40 of 46 (87%) cases with tumors. Next-generation sequencing analysis revealed recurrent mutations in the KMT2C gene in 4 of 5 ACKD-RCCs (80%), exclusively exhibiting cribriform "sieve-like" morphology; whereas the case negative for KMT2C mutations exhibited "type 2" papillary RCC morphology and lacked "sieve-like" growth pattern. Pathogenic mutations in TSC2 were the second common abnormality (3/5; 60%), often coexisting with KMT2C mutations. Deleterious mutations in additional genes such as CBL, PDGFRA, and SYNE1, etc. were noted but were nonrecurrent and always coexisted with mutations in KMT2C or TSC2. To conclude, our study highlights that mutations in a chromatin-modifying gene KMT2C may potentially be oncogenic drivers for the development of ACKD-RCC with classic sieve-like morphology. In addition, pathogenic mutations in TSC2 possibly play a role in the development of cysts/tumors in a subset of ACKD patients. If corroborated in larger cohorts, these findings would be useful in planning surveillance and early intervention in ESRD patients developing ACKD.

Adenomyoepitheliomas of the Breast Frequently Harbor Recurrent Hotspot Mutations in PIK3-AKT Pathway-related Genes and a Subset Show Genetic Similarity to Salivary Gland Epithelial-Myoepithelial Carcinoma.

Adenomyoepitheliomas (AME) of the breast and epithelial-myoepithelial carcinomas (EMCs) of salivary gland are morphologically similar tumors defined by the presence of a biphasic population of ductal epithelial elements mixed with myoepithelial cells. We sought to explore the molecular profile of AMEs and determine whether they might also share the PLAG1, HMGA2, and HRAS alterations seen in EMCs. Tumor tissue from 19 AMEs was sequenced and analyzed using Ion AmpliSeq Cancer Hotspot Panel v2 covering ∼2800 COSMIC mutations across 50 cancer-related genes. Cases were additionally screened by FISH for PLAG1 and HMGA2 rearrangements. Of 19 AMEs (12 benign; 7 malignant), 2 cases failed the DNA extraction. Of the remaining 17 cases, 14 had at least one nonsynonymous mutation identified. The most common mutations were in PIK3CA (6/17) and AKT1 (5/17), which were mutually exclusive. Two tumors demonstrated mutations in APC, while 1 demonstrated an STK11 mutation. Mutations in ATM, EGFR, FGFR3 or GNAS were identified in 4 cases with concurrent AKT1 mutations. HRAS mutation co-occurring with PIK3CA mutation was noted in 1 case of ER-negative malignant AME. While 2 cases harbored alterations in HMGA2, none was positive for PLAG1 rearrangement. Our findings confirm that breast AMEs are genetically heterogeneous exhibiting recurrent mutually exclusive mutations of PIK3CA and AKT1 in a majority of cases. HRAS mutations co-occur with PIK3CA mutations in ER-negative AMEs and may possibly be linked to clinically aggressive behavior. We identified hotspot mutations in additional genes (APC, STK11, ATM, EGFR, FGFR3, and GNAS). We report the presence of HMGA2 alterations in 2/16 AMEs, supporting their relationship with EMC of salivary glands in at least a subset of cases. PIK3CA, AKT1 and HRAS may serve as potential actionable therapeutic targets in clinically aggressive AMEs.

Molecular diagnosis of somatic overgrowth conditions: A single-center experience.

BACKGROUND: Somatic overgrowth conditions, including Proteus syndrome, Sturge-Weber syndrome, and PIK3CA-related overgrowth spectrum, are caused by post-zygotic pathogenic variants, result in segmental mosaicism, and give rise to neural, cutaneous and/or lipomatous overgrowth. These variants occur in growth-promoting pathways leading to cellular proliferation and expansion of tissues that arise from the affected cellular lineage. METHODS: We report on 80 serial patients evaluated for somatic overgrowth conditions in a diagnostic laboratory setting, including three prenatal patients. In total, 166 tissues from these 80 patients were subjected to targeted sequencing of an 8-gene panel capturing 10.2 kb of sequence containing known pathogenic variants associated with somatic overgrowth conditions. Deep next-generation sequencing was performed with the IonTorrent PGM platform at an average depth typically >5,000×. RESULTS: Likely pathogenic or pathogenic variants were identified in 36 individuals and variants of unknown significance in four. The overall molecular diagnostic yield was 45% but was highly influenced by both submitted tissue type and phenotype. In the prenatal setting, two patients had pathogenic variants identified in cultured amniocytes but in a third patient, the pathogenic variant was only present in post-natal tissues. Finally, expanding the test to include full gene sequencing of PIK3CA in contrast to targeted sequencing identified likely pathogenic variants in 3 of 7 patients that tested negative on the original panel. CONCLUSION: Next-generation sequencing has enabled sensitive detection of somatic pathogenic variants associated with overgrowth conditions. However, as the pathogenic variant allele frequency varies by tissue type within an individual, submission of affected tissue(s) greatly increases the chances of a molecular diagnosis.

A comparison of Illumina and Ion Torrent sequencing platforms in the context of differential gene expression.

BACKGROUND: Though Illumina has largely dominated the RNA-Seq field, the simultaneous availability of Ion Torrent has left scientists wondering which platform is most effective for differential gene expression (DGE) analysis. Previous investigations of this question have typically used reference samples derived from cell lines and brain tissue, and do not involve biological variability. While these comparisons might inform studies of tissue-specific expression, marked by large-scale transcriptional differences, this is not the common use case. RESULTS: Here we employ a standard treatment/control experimental design, which enables us to evaluate these platforms in the context of the expression differences common in differential gene expression experiments. Specifically, we assessed the hepatic inflammatory response of mice by assaying liver RNA from control and IL-1ß treated animals with both the Illumina HiSeq and the Ion Torrent Proton sequencing platforms. We found the greatest difference between the platforms at the level of read alignment, a moderate level of concordance at the level of DGE analysis, and nearly identical results at the level of differentially affected pathways. Interestingly, we also observed a strong interaction between sequencing platform and choice of aligner. By aligning both real and simulated Illumina and Ion Torrent data with the twelve most commonly-cited aligners in the literature, we observed that different aligner and platform combinations were better suited to probing different genomic features; for example, disentangling the source of expression in gene-pseudogene pairs. CONCLUSIONS: Taken together, our results indicate that while Illumina and Ion Torrent have similar capacities to detect changes in biology from a treatment/control experiment, these platforms may be tailored to interrogate different transcriptional phenomena through careful selection of alignment software.

Effect of heartfulness meditation on burnout, emotional wellness, and telomere length in health care professionals.

Background: Burnout poses significant challenges during training years in residency and later in the career. Meditation is a tool to treat stress-related conditions and promote wellness. Telomere length may be affected by burnout and stress. However, the benefits of meditation have not been fully demonstrated in health care professionals. Objective: We assessed the effects of a 12-week 'Heartfulness Meditation' program on burnout, emotional wellness, and telomere length in residents, faculty members, and nurses at a large community teaching hospital during the 2015-16 academic year. Methods: All subjects completed a baseline Maslach Burnout Inventory (MBI) and Emotional Wellness Assessment (EWA) at the beginning of the study. Meditators received instructions in Heartfulness Meditation. At week 12, subjects completed a follow up MBI and EWA scores. Salivary telomere length was measured at baseline and week 12. Results: Twenty-seven out of a total 155 residents (17.4%) along with eight faculty physicians and 12 nurses participated in the study. Thirty-five enrolled as meditators and 12 as controls. At 12 weeks, the meditators had statistically significant improvement in all measures of burnout and in nearly all attributes of EWA. Controls showed no statistically significant changes in either burnout or emotional wellness scores. Relative telomere length increased with statistical significance in a younger subset of meditators. Conclusion: Our results indicate that meditation offers an accessible and efficient method by which physician and nurse burnout can be ameliorated and wellness can be enhanced. The increased telomere length is an interesting finding but needs to be confirmed with further research. Abbreviations: EWA: Emotional wellness assessment; MBI: Maslach burnout inventory; EE: Emotional exhaustion; DP: Depersonalization; PA: Personal accomplishment; PI: Prinicipal investigator; JT: Jayaram Thimmapuram.

Improved Methods to Generate Spheroid Cultures from Tumor Cells, Tumor Cells & Fibroblasts or Tumor-Fragments: Microenvironment, Microvesicles and MiRNA.

Diagnostic and prognostic indicators are key components to achieve the goal of personalized cancer therapy. Two distinct approaches to this goal include predicting response by genetic analysis and direct testing of possible therapies using cultures derived from biopsy specimens. Optimally, the latter method requires a rapid assessment, but growing xenograft tumors or developing patient-derived cell lines can involve a great deal of time and expense. Furthermore, tumor cells have much different responses when grown in 2D versus 3D tissue environments. Using a modification of existing methods, we show that it is possible to make tumor-fragment (TF) spheroids in only 2-3 days. TF spheroids appear to closely model characteristics of the original tumor and may be used to assess critical therapy-modulating features of the microenvironment such as hypoxia. A similar method allows the reproducible development of spheroids from mixed tumor cells and fibroblasts (mixed-cell spheroids). Prior literature reports have shown highly variable development and properties of mixed-cell spheroids and this has hampered the detailed study of how individual tumor-cell components interact. In this study, we illustrate this approach and describe similarities and differences using two tumor models (U87 glioma and SQ20B squamous-cell carcinoma) with supporting data from additional cell lines. We show that U87 and SQ20B spheroids predict a key microenvironmental factor in tumors (hypoxia) and that SQ20B cells and spheroids generate similar numbers of microvesicles. We also present pilot data for miRNA expression under conditions of cells, tumors, and TF spheroids.

Enhanced sensitivity for detection of low-level germline mosaic RB1 mutations in sporadic retinoblastoma cases using deep semiconductor sequencing.

Sporadic retinoblastoma (RB) is caused by de novo mutations in the RB1 gene. Often, these mutations are present as mosaic mutations that cannot be detected by Sanger sequencing. Next-generation deep sequencing allows unambiguous detection of the mosaic mutations in lymphocyte DNA. Deep sequencing of the RB1 gene on lymphocyte DNA from 20 bilateral and 70 unilateral RB cases was performed, where Sanger sequencing excluded the presence of mutations. The individual exons of the RB1 gene from each sample were amplified, pooled, ligated to barcoded adapters, and sequenced using semiconductor sequencing on an Ion Torrent Personal Genome Machine. Six low-level mosaic mutations were identified in bilateral RB and four in unilateral RB cases. The incidence of low-level mosaic mutation was estimated to be 30% and 6%, respectively, in sporadic bilateral and unilateral RB cases, previously classified as mutation negative. The frequency of point mutations detectable in lymphocyte DNA increased from 96% to 97% for bilateral RB and from 13% to 18% for unilateral RB. The use of deep sequencing technology increased the sensitivity of the detection of low-level germline mosaic mutations in the RB1 gene. This finding has significant implications for improved clinical diagnosis, genetic counseling, surveillance, and management of RB.