Identification of differentially distributed gene expression and distinct sets of cancer-related genes identified by changes in mean and variability

ABSTRACT There is increasing evidence that changes in the variability or overall distribution of gene expression are important both in normal biology and in diseases, particularly cancer. Genes whose expression differs in variability or distribution without a difference in mean are ignored by traditional differential expression-based analyses. Using a Bayesian hierarchical model that provides tests for both differential variability and differential distribution for bulk RNA-seq data, we report here an investigation into differential variability and distribution in cancer. Analysis of eight paired tumour–normal datasets from The Cancer Genome Atlas confirms that differential variability and distribution analyses are able to identify cancer-related genes. We further demonstrate that differential variability identifies cancer-related genes that are missed by differential expression analysis, and that differential expression and differential variability identify functionally distinct sets of potentially cancer-related genes. These results suggest that differential variability analysis may provide insights into genetic aspects of cancer that would not be revealed by differential expression, and that differential distribution analysis may allow for more comprehensive identification of cancer-related genes than analyses based on changes in mean or variability alone.

Outcomes of 27 Non-Small Cell Lung Cancer Patients Receiving Resections of Paired Primary Pulmonary and Brain Metastatic Tumours: Is EGFR Negative Mutation of Brain Metastasis Bring More Benefit?

Background: To analyze the heterogeneity and clinical outcomes of epidermal growth factor receptor (EGFR) gene mutation in primary tumour and corresponding brain metastasis(BM) in non-small cell lung cancer (NSCLC).Methods: Primary pulmonary tumours and paired metastatic brain tumours were surgically removed from twenty-seven NSCLC patients from July 1999 to November 2013 in our hospital. All brain lesions were confirmed as metastases stemming from NSCLC by pathological examination. EGFR gene (exons 18-21) mutant status was detected in matched tumour by using amplification refraction mutation system (ARMS). If inconsistency was detected, the paired tumour was evaluated again. The McNemar test was performed to compare the consistency of the paired tumour, and the Kappa test was used to quantify the agreement of both methods.Progression free survival(PFS) and overall survival(OS) were exhibited by the Kaplan-Meier.Results: In this study, of the 27 patients, nine (33.3%) cases were found to have EGFR mutations in BMs, and ten (37.0%) patients were detected positive EGFR status in primary lung tumour tissue. The rate of consistency of the matched tumour was 24/27 (88.9%, P≤0.001), and the Kappa coefficient was 0.757. Among the three cases presenting EGFR mutational heterogeneity, two patients harbored EGFR mutation in primary tumors but was negative in BMs, meanwhile, the other patient had the opposite pattern. Comparing to patients with consistent EGFR mutations(both mutant or wild),patients with inconsistent EGFR mutations tended to have better outcomes, including PFS(37.2months vs 25.0months vs 16.5months,P=0.159) and OS(53.6months vs 27.5months vs 26.8months,P=0.380), further analysis showed that two patients whose EGFR mutant-type primary tumor progressing to wild-type cerebral metastastic tumor might have longer overall survival(53.6months,37.8months) than one patient harboring reverse mutant difference(EGFR wild-type primary tumor progressing to mutant-type brain metastastic tumor) (16.7months). Also we found that patients with wild type in brain metastatic tumour had longer overall survival (OS)(mOS, 36.3 months vs 29.1months, P = 0.944).Conclusions: EGFR mutation status in NSCLC patients between primary lung tumour and paired BM was heterogeneous, patients harbored wild type EGFR mutation in BM might have better outcomes, especially for positive status transferred to wild.

Investigating the clinical, pathological and molecular profile of oncocytic adrenocortical neoplasms

Background: Malignant oncocytic adrenocortical neoplasms (OANs) are rare tumours with a distinctive biological behavior compared to conventional adrenocortical carcinoma (ACC). The current prognostic systems overestimate the malignant potential of these tumours and guidance for surveillance and treatment strategies are lacking. Aim: To evaluate the utility of clinical, pathological and molecular markers in predicting the biological behavior and outcomes of malignant OANs. Methods: A retrospective clinicopathological review of ten histologically confirmed OANs was carried out. Whole exome sequencing (WES) of germline and paired tumour samples was performed for four of the ten OAN cases and compared to WES data from five cases of conventional ACC and data from The Cancer Genome Atlas (TCGA). We reviewed all cases of malignant OAN reported in the literature and compared to our case series. Results: Eight (80%) tumours were classified as malignant, one borderline and one benign (Lin-Weiss-Bisceglia criteria: LWB). The malignant OAN were larger tumours and had higher MIB index and Helsinki scores. Molecular profiling identified a pathogenic germline variant in MSH6 in an individual in the OAN group. The OAN samples had a lower mutation burden compared to the ACC samples. Somatic driver variants were identified in OAN and ACC samples including a pathogenic missense variant in CTNNB1. Conclusion: In this study, the LWB classification demonstrated sensitivity for the differentiation of benign from malignant OAN. Molecular profiling identified dysregulation in DNA repair and Wnt signaling pathways in both OAN and ACC samples, suggesting a molecular overlap between OAN and conventional ACC.

Operative ubiquitin-specific protease 22 deubiquitination confers a more invasive phenotype to cholangiocarcinoma

Oncogenic ubiquitin-specific protease 22 (USP22) is implicated in a variety of tumours; however, evidence of its role and underlying molecular mechanisms in cholangiocarcinoma (CCA) development remains unknown. We collected paired tumour and adjacent non-tumour tissues from 57 intrahepatic CCA (iCCA) patients and evaluated levels of the USP22 gene and protein by qPCR and immunohistochemistry. Both the mRNA and protein were significantly upregulated, correlated with the malignant invasion and worse OS of iCCA. In cell cultures, USP22 overexpression increased CCA cell proliferation and mobility, and induced epithelial-to-mesenchymal transition (EMT). Upon an interaction, USP22 deubiquitinated and stabilized sirtuin-1 (SIRT1), in conjunction with Akt/ERK activation. In implantation xenografts, USP22 overexpression stimulated tumour growth and metastasis to the lungs of mice. Conversely, the knockdown by USP22 shRNA attenuated the tumour growth and invasiveness in vitro and in vivo. Furthermore, SIRT1 overexpression reversed the USP22 functional deficiency, while the knockdown acetylated TGF-β-activated kinase 1 (TAK1) and Akt. Our present study defines USP22 as a poor prognostic predictor in iCCA that cooperates with SIRT1 and facilitates tumour development.

Spatial Distribution of Private Gene Mutations in Clear Cell Renal Cell Carcinoma

Intra-tumour heterogeneity is the molecular hallmark of renal cancer, and the molecular tumour composition determines the treatment outcome of renal cancer patients. In renal cancer tumourigenesis, in general, different tumour clones evolve over time. We analysed intra-tumour heterogeneity and subclonal mutation patterns in 178 tumour samples obtained from 89 clear cell renal cell carcinoma patients. In an initial discovery phase, whole-exome and transcriptome sequencing data from paired tumour biopsies from 16 ccRCC patients were used to design a gene panel for follow-up analysis. In this second phase, 826 selected genes were targeted at deep coverage in an extended cohort of 89 patients for a detailed analysis of tumour heterogeneity. On average, we found 22 mutations per patient. Pairwise comparison of the two biopsies from the same tumour revealed that on average, 62% of the mutations in a patient were detected in one of the two samples. In addition to commonly mutated genes (VHL, PBRM1, SETD2 and BAP1), frequent subclonal mutations with low variant allele frequency (<10%) were observed in TP53 and in mucin coding genes MUC6, MUC16, and MUC3A. Of the 89 ccRCC tumours, 87 (~98%) harboured private mutations, occurring in only one of the paired tumour samples. Clonally exclusive pathway pairs were identified using the WES data set from 16 ccRCC patients. Our findings imply that shared and private mutations significantly contribute to the complexity of differential gene expression and pathway interaction and might explain the clonal evolution of different molecular renal cancer subgroups. Multi-regional sequencing is central for the identification of subclones within ccRCC.

Profiling of NUMTs in Gingivobuccal Oral Cancer

Very little information exists on the NUclear MiTochondrial sequences (NUMTs) in oral squamous cell carcinoma-gingivobuccal (OSCC-GB). We analysed whole genome sequences obtained from paired tumour and blood DNA from 25 patients of OSCC-GB for detecting NUMTs. Nine potential somatic NUMTs and 78 germline NUMTs were identified in these patients. None of the somatic NUMTs could be confirmed by PCR assay. None of the germline NUMTs were found to be specific for OSCC-GB. Although there have been recent reports on detection of somatic NUMTs in other cancer types, our results suggest NUMTs, both somatic and germline, are not involved in OSCC-GB.

Nanopore sequencing enables comprehensive transposable element epigenomic profiling

We apply long-read nanopore sequencing and a new tool, TLDR (Transposons from Long Dirty Reads), to directly infer CpG methylation of new and extant human transposable element (TE) insertions in hippocampus, heart, and liver, as well as paired tumour and non-tumour liver. Whole genome TLDR analysis greatly facilitates studies of TE biology as complete insertion sequences and their epigenetic modifications are readily obtainable.

Comparison of somatic variant detection algorithms using Ion Torrent targeted deep sequencing data

Background The application of next-generation sequencing in cancer has revealed the genomic landscape of many tumour types and is nowadays routinely used in research and clinical settings. Multiple algorithms have been developed to detect somatic variation from sequencing data using either paired tumour-blood or tumour-only samples. Most of these methods have been developed and evaluated for the identification of somatic variation using Illumina sequencing datasets of moderate coverage. However, a comprehensive evaluation of somatic variant detection algorithms on Ion Torrent targeted deep sequencing data has not been performed. Methods We have applied three somatic detection algorithms, Torrent Variant Caller, MuTect2 and VarScan2, on a large cohort of ovarian cancer patients comprising of 208 paired tumour-blood samples and 253 tumour-only samples sequenced deeply on Ion Torrent Proton platform across 330 amplicons. Subsequently, the concordance and performance of the three somatic variant callers were assessed. Results We have observed low concordance across the algorithms with only 0.5% of SNV and 0.02% of INDEL calls in common across all three methods. The intersection of all methods showed better performance when assessed using correlation with known mutational signatures, overlap with COSMIC variation and by examining the variant characteristics. The Torrent Variant Caller also performed well with the advantage of not eliminating a high number of variants that could lead to high type II error. Conclusions Our results suggest that caution should be taken when applying state-of-the-art somatic variant algorithms to Ion Torrent targeted deep sequencing data. Better quality control procedures and strategies that combine results from multiple methods should ensure that higher accuracy is achieved. This is essential to ensure that results from bioinformatics pipelines using Ion Torrent deep sequencing can be robustly applied in cancer research and in the clinic.

PIPA: A phase Ib study of selective ß-isoform sparing phosphatidylinositol 3-kinase (PI3K) inhibitor taselisib (T) plus palbociclib (P) in patients (pts) with advanced solid cancers—Safety, tolerability, pharmacokinetic (PK), and pharmacodynamic (PD) analysis of the doublet combination.

3087 Background: Oncogenic hyperactivation of the PI3K pathway is common in multiple cancers, with preclinical data showing that CDK4/6 inhibitors sensitise PIK3CA mutant cancers to PI3K inhibitors. We report the activity of the P+T in solid tumors with PI3K pathway activation, along with the PD biomarker analysis. Methods: We previously reported the dose escalation phase identifying 125mg P given 3-weeks-on, 1-week-off in combination with T 2mg as the recommended phase 2 dose (R2PD, Lim, ASCO 2017). We report the results in solid tumors with confirmed activating mutations (mts) in the PI3K pathway, from dose escalation and expansion, with no prior exposure to CDK4/6 or PI3K pathway inhibitors. PD studies include analyses of platelet-rich plasma (PRP) and paired tumour biopsies. Results: 20 pts (median age 61, range 34-72) were treated at the doublet RP2D, M/F 7/13, with a median 4 prior treatments (range 2-11). Tumour types included colorectal, breast, lung, endometrial,oligodendroglioma and head and neck cancers. Durable disease control occurred in 3 patients with ER+ advanced breast cancer with responses lasting >6 months including 1 pt with a H1047R PIK3CAmt with an ongoing RECIST PR>36 cycles, 2 pts with PIK3CAmt colorectal cancer had RECIST SD for >5 months, and 1 patient with a PIK3CGmt anaplastic oligodendroglioma had clinical and radiological benefit lasting 5.5 months. Treatment was well tolerated with predictable G1-2 adverse events (AEs). G3 toxicities of neutropenia (n=6), thrombocytopenia (2), rash (2), mucositis (1) and raised transaminases (1 each) were all transient with no G4/5 AEs. Significant decreases in tumour pRb, and pAKT and pGSK3ß in PRP, confirmed modulation of both CDK4/6 and PI3K pathways at R2PD. Conclusions: Doublet P+T is well tolerated at the combination RP2D, with PD evidence of PI3K and CDK4/6 modulation in both plasma and tumor. Promising preliminary anti-tumor activity is seen in a mixed histology cohort selected for activating PIK3 mutations. Clinical trial information: NCT02389842.