Quantification of the Heterogeneity of Prognostic Cellular Biomarkers in Ewing Sarcoma Using Automated Image and Random Survival Forest Analysis

Abstract

Driven by genomic somatic variation, tumour tissues are typically heterogeneous, yet unbiased quantitative methods are rarely used to analyse heterogeneity at the protein level. Motivated by this problem, we developed automated image segmentation of images of multiple biomarkers in Ewing sarcoma to generate distributions of biomarkers between and within tumour cells. We further integrate high dimensional data with patient clinical outcomes utilising random survival forest (RSF) machine learning. Using material from cohorts of genetically diagnosed Ewing sarcoma with EWSR1 chromosomal translocations, confocal images of tissue microarrays were segmented with level sets and watershed algorithms. Each cell nucleus and cytoplasm were identified in relation to DAPI and CD99, respectively, and protein biomarkers (e.g. Ki67, pS6, Foxo3a, EGR1, MAPK) localised relative to nuclear and cytoplasmic regions of each cell in order to generate image feature distributions. The image distribution features were analysed with RSF in relation to known overall patient survival from three separate cohorts (185 informative cases). Variation in pre-analytical processing resulted in elimination of a high number of non-informative images that had poor DAPI localisation or biomarker preservation (67 cases, 36%). The distribution of image features for biomarkers in the remaining high quality material (118 cases, 104 features per case) were analysed by RSF with feature selection, and performance assessed using internal cross-validation, rather than a separate validation cohort. A prognostic classifier for Ewing sarcoma with low cross-validation error rates (0.36) was comprised of multiple features, including the Ki67 proliferative marker and a sub-population of cells with low cytoplasmic/nuclear ratio of CD99. Through elimination of bias, the evaluation of high-dimensionality biomarker distribution within cell populations of a tumour using random forest analysis in quality controlled tumour material could be achieved. Such an automated and integrated methodology has potential application in the identification of prognostic classifiers based on tumour cell heterogeneity.

Leveraging RSF and PET images for prognosis of multiple myeloma at diagnosis

“…for lung [11] or head-neck cancer [17]. However, apart from Bühnemann et al [3] who used confocal images of tissue microarrays, RSF and radiomics association has not been thoroughly investigated. We explore in this work their combination in the context of MM.…”

CXCL14, CXCR7 expression and CXCR4 splice variant ratio associate with survival and metastases in Ewing sarcoma patients

“…EWS is recognised from the onset of its original description by James Ewing as a highly vascularised tumour and amongst many other pathways, chemokine and the TGF-B pathway might play a role for this excessive vascularisation pattern [11][12][13]. Besides angiogenesis, these pathways are involved in migration that might be reflected by the high metastatic propensity of EWS [1,13,14]. In several tumour types a positive correlation between increased expression of CXCR4 and metastatic propensity was reported, but contradictory results were reported in EWS [15][16][17].…”

Biology of Bone Sarcomas and New Therapeutic Developments

“…Although Ewing sarcoma is the most homogeneous entity among bone sarcomas, composed of undifferentiated round cancer cells characterised by CD99-, FLI1-, HNK1- and CAV1-positive immunostaining associated with limited stromal components [ 36 ], recent work demonstrated in contrast their heterogeneity [ 37 – 40 ]. Previous studies highlighted only a few recurrent somatic mutations in Ewing sarcomas ( TP53, STAG2, CDKN2 ) [ 38 , 41 , 42 ]. However, more recent studies by Zhang et al used next-generation sequencing (Ion AmpliSeq™ Cancer Hotspot Panel v2) to identify a series of five new mutations ( KDR, STK11, MLH1, KRAS and PTPN11 ) related to a higher proliferation index and revealing a higher tumour heterogeneity than initially suspected [ 37 ].…”