Towards reproducible research in recurrent pregnancy loss immunology: Learning from cancer microenvironment deconvolution

Abstract

The most recent international guidelines regarding recurrent pregnancy loss (RPL) exclude most of the immunological tests recommended for RPL since they do not reach an evidence-based level. Comparisons for metanalysis and systematic reviews are limited by the ambiguity in terms of RPL definition, etiological and risk factors, diagnostic work-up, and treatments applied. Therefore, cohort heterogeneity, the inadequacy of numerosity, and the quality of data confirm a not standardized research quality in the RPL field, especially for immunological background, for which potential research application remains confined in a separate single biological layer. Innovative sequencing technologies and databases have proved to play a significant role in the exploration and validation of cancer research in the context of dataset quality and bioinformatics tools. In this article, we will investigate how bioinformatics tools born for large-scale cancer immunological research could revolutionize RPL immunological research but are limited by the nature of current RPL datasets.