Molecular and Computational Methods for the Detection of Microsatellite Instability in Cancer

Abstract

Microsatellite instability (MSI) is a genomic alteration in which microsatellites, usually of one to four nucleotide repeats, accumulate mutations corresponding to deletions/insertions of a few nucleotides. The MSI phenotype has been extensively characterized in colorectal cancer and is due to a deficiency of the DNA mismatch repair system. MSI has recently been shown to be present in most types of cancer with variable frequencies (from <1 to 30%). It correlates positively to survival outcome and predicts the response to immune checkpoint blockade therapy. The different methods developed for MSI detection in cancer require taking into consideration two critical parameters which influence method performance. First, the microsatellite markers used should be chosen carefully to ensure they are highly sensitive and specific for MSI detection. Second, the analytical method used should be highly resolute to allow clear identification of MSI and of the mutant allele genotype, and should present the lowest limit of detection possible for application in samples with low mutant allele frequency. In this review, we describe all the different molecular and computational methods developed to date for the detection of MSI in cancer, how they have evolved and improved over the years, and their advantages and drawbacks.

Keywords: MSI detection method; MSI-H cancer; computational biology; microsatellite genotyping; microsatellite instability; next-generation sequencing.

Figure 1

Overview of the different methods used for MSI detection in cancer. (A) Polyacrylamide gel electrophoresis (PAGE). (B) Capillary electrophoresis fragment analysis (FA). (C) Denaturing high performance liquid chromatography (DHPLC). (D) High resolution melting analysis (HRM). (E) Next-generation sequencing (NGS).

Figure 2

Strategies for improving the limit of detection of MSI in cancer. (A) Dilution of the DNA to 0–3 genome equivalent per PCR in small-pool PCR. (B) Mutant allele enrichment in the DNA sample using sense and antisense probes complementary to WT DNA sequence and duplex specific nuclease (DSN) in NaME-Pro and NaMSIE. (C) Mutant allele enrichment during E-ice-COLD-PCR using an LNA blocker probe complementary to WT DNA sequence. LNA, Locked nucleic acid.

Figure 3

Overview of the different NGS-based computational methods developed for MSI detection in cancer. (A) Methods based on comparison of repeat length distribution of microsatellites including MSIsensor, mSINGs, MANTIS, Cortes-Ciriano method, and MSI-ColonCore. (B) Methods based on the total mutation burden in all sequences and/or the indel burden in microsatellites including MSIseq Index, MSIseq/NGS classifier, and Nowak methods. The steps 1–3 can be performed in different orders or in parallel depending on the method. The MSIseq/NGS classifier directly processes a list of somatic mutations and not raw NGS data.