BACKGROUND: The status of tumor microsatellite instability (MSI/MSS) is a crucial parameter that influences both the prognosis and the potential response to immunotherapy. However, it is not always the case that the biomaterial collected in the context of scientific research has the results of laboratory determination of this parameter. It is acknowledged that the immunohistochemistry (IHC) and polymerase chain reaction (PCR) methods employed to ascertain this parameter exhibit a degree of inconsistency. Consequently, despite the availability of this information within the accompanying patient data, it may not align with the findings of a comprehensive transcriptome analysis. Hence, it is imperative to incorporate a mechanism for detecting microsatellite instability into the protocol for a full-transcriptome analysis of tumors AIM: The objective is to evaluate the availability and applicability of bioinformatic tools for determining MSI status on whole transcriptomes of colorectal tumors. METHODS: The efficacy of bioinformatics tools was evaluated using a local sample of whole transcriptomes of colorectal tumors from 15 patients following the resection of the primary tumor. The number of somatic mutations, MSI status, along with their correlation, were assessed. RESULTS: We found one bioinformatics tool, MIRACLE, trained on TCGA data to determine the proportion of unstable sites in the transcriptome. Repeat instability was detected in 3 out of 13 samples. The microsatellite instability status determined by it correlates with mutational load (2163 mutations in samples with MSI and 122. 9 with MSS) and unambiguously identifies samples. Known pathogenic and possibly pathogenic variants in the MSH2 and MLH1 genes causing Lynch syndrome were identified for 2 specimens. Sporadic nature of microsatellite instability due to MLH1 gene hypermethylation was suggested for one of the samples. CONCLUSION: MIRACLE tool can be readily incorporated into a comprehensive transcriptome-based tumor analysis protocol. Furthermore, it enables the generation of a biologically plausible estimate of MSI/MSS status, which correlates with tumor mutational load derived from transcriptome analysis data.
Kanygina et al. (Thu,) studied this question.