GEM-DeCan: Improving tumor immune microenvironment profiling by the integration of novel gene expression and DNA methylation deconvolution signatures

Quantifying the proportion of the different cell types present in tumor biopsies remains a priority in cancer research. So far, a number of deconvolution methods have emerged for estimating cell composition using reference signatures either based on gene expression or on DNA methylation from purified cells. These two deconvolution approaches could be complementary to each other, leading to even more performant signatures, in cases where both data types are available. However, the potential relationship between signatures based on gene expression and those based on DNA methylation remains underexplored. Here we present five new deconvolution signature matrices, based on DNA methylation or RNAseq data, which can estimate the proportion of immune cells and cancer cells in a tumour sample. We test these signature matrices on available datasets for in-silico and in-vitro mixtures, peripheral blood, cancer samples from TCGA, bone marrow from multiple myeloma patients and a single-cell melanoma dataset. Cell proportions estimates based on deconvolution performed using our signature matrices, implemented within the EpiDISH framework, show comparable or better correlation with FACS measurements of immune cell-type abundance and with various estimates of cancer sample purity and composition than existing methods. Finally, using publicly available data of 3D chromatin structure in haematopoietic cells, we expanded the list of genes to be included in the RNAseq signature matrices by considering the presence of methylated CpGs in gene promoters or in genomic regions which are in 3D contact with these promoters. Our expanded signature matrices have improved performance compared to our initial RNAseq signature matrix. Finally, we show the value of our signature matrices in predicting patient response to immune checkpoint inhibitors in three melanoma and one bladder cancer cohort, based on bulk tumour sample gene expression data. We also provide GEM-DeCan: a snakemake pipeline, able to run an analysis from raw sequencing data to deconvolution based on various gene expression signature matrices, both for bulk RNASeq and DNA methylation data. The code for producing the signature matrices and reproducing all the figures of this paper is available on GitHub: https://github.com/VeraPancaldiLab/GEMDeCan.

Expression Profile of Stemness Markers CD138, Nestin and Alpha-SMA in Ameloblastic Tumours

Ameloblastic carcinoma is a rare malignant odontogenic neoplasm with a poor prognosis. It can arise de novo or from a pre-existing ameloblastoma. Research into stemness marker expression in ameloblastic tumours is lacking. This study aimed to explore the immunohistochemical expression of stemness markers nestin, CD138, and alpha-smooth muscle actin (alpha-SMA) for the characterisation of ameloblastic tumours. Six cases of ameloblastoma and four cases of ameloblastic carcinoma were assessed, including one case of ameloblastic carcinoma arising from desmoplastic ameloblastoma. In all tumour samples, CD138 was positive, whilst alpha-SMA was negative. Nestin was negative in all but one tumour sample. Conversely, the presence or absence of these markers varied in stroma samples. Nestin was observed in one ameloblastic carcinoma stroma sample, whilst CD138 was positive in one ameloblastoma case, one desmoplastic ameloblastoma case, and in two ameloblastic carcinoma stroma samples. Finally, alpha-SMA was found positive only in the desmoplastic ameloblastoma stroma sample. Our results suggest nestin expression to be an indicator for ameloblastic carcinoma, and CD138 and alpha-SMA to be promising biomarkers for the malignant transformation of ameloblastoma. Our data showed that nestin, CD138, and alpha-SMA are novel biomarkers for a better understanding of the origins and behaviour of ameloblastic tumours.

Impact of Cytological Sampling on EGFR Mutation Testing in Stage III-IV Lung Adenocarcinoma

Objectives. There have been advances in the identification and understanding of molecular subsets of lung cancer, defined by specific oncogenic aberrations. A number of actionable genetic alterations have been identified, such as the epidermal growth factor receptor (EGFR) mutation. We aimed to establish the reasons why patients were not undergoing EGFR mutation testing at the time of histological diagnosis. Methods. The records of 70 patients with advanced adenocarcinoma of the lung managed through a single multidisciplinary team at a single institution were reviewed. Data were collected on method of tumour sample collection, whether this was sent for EGFR testing, and the result. Results. Seventy patients were identified. In 21/25 (84%) cases, cytological sampling was sufficient for EGFR mutation analysis, compared with 40/45 (89%) cases with histological sampling. EGFR mutation testing was not carried out in 22/70 (31.4%) patients. There was insufficient tumour sample for EGFR testing in 9/22 (40.9%) patients. Other reasons for not testing included poor patient fitness and problems in the diagnostic pathway. Conclusions. In this series, cytological tumour sampling was not the predominant reason why cancers failed to have EGFR mutation status established.

Crambled: A Shiny application to enable intuitive resolution of conflicting cellularity estimates

It is now commonplace to investigate tumour samples using whole-genome sequencing, and some commonly performed tasks are the estimation of cellularity (or sample purity), the genome-wide profiling of copy numbers, and the assessment of sub-clonal behaviours. Several tools are available to undertake these tasks, but often give conflicting results – not least because there is often genuine uncertainty due to a lack of model identifiability. Presented here is a tool, "Crambled", that allows for an intuitive visual comparison of the conflicting solutions. Crambled is implemented as a Shiny application within R, and is accompanied by example images from two use cases (one tumour sample with matched normal sequencing, and one standalone cell line example) as well as functions to generate the necessary images from any sequencing data set. Through the use of Crambled, a user may gain insight into why each tool has offered its given solution and combined with a knowledge of the disease being studied can choose between the competing solutions in an informed manner.