Reliable Pan-Cancer Microsatellite Instability Assessment by Using Targeted Next-Generation Sequencing Data

Abstract

Purpose: Microsatellite instability (MSI)/mismatch repair (MMR) status is increasingly important in the management of patients with cancer to predict response to immune checkpoint inhibitors. We determined MSI status from large-panel clinical targeted next-generation sequencing (NGS) data across various solid cancer types.

Methods: The MSI statuses of 12,288 advanced solid cancers consecutively sequenced with Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets clinical NGS assay were inferred by using MSIsensor, a program that reports the percentage of unstable microsatellites as a score. Cutoff score determination and sensitivity/specificity were based on MSI polymerase chain reaction (PCR) and MMR immunohistochemistry.

Results: By using an MSIsensor score ≥ 10 to define MSI high (MSI-H), 83 (8%) of 996 colorectal cancers (CRCs) and 42 (16%) of 260 uterine endometrioid cancers (UECs) were MSI-H. Validation against MSI PCR and/or MMR immunohistochemistry performed for 138 (24 MSI-H, 114 microsatellite stable [MSS]) CRCs, and 40 (15 MSI-H, 25 MSS) UECs showed a concordance of 99.4%. MSIsensor also identified 68 MSI-H/MMR-deficient (MMR-D) non-CRC/UECs. Of 9,591 non-CRC/UEC tumors with MSS MSIsensor status, 456 (4.8%) had slightly elevated scores(≥3 and <10) of which 96.6% with available material were confirmed to be MSS by MSI PCR. MSI-H was also detected and confirmed in three non-CRC/UECs with low exonic mutation burden (< 20). MSIsensor correctly scored all 15 polymerase ε ultra-mutated cancers as negative for MSI.

Conclusion: MSI status can be reliably inferred by MSIsensor from large-panel targeted NGS data. Concurrent MSI testing by NGS is resource efficient, is potentially more sensitive for MMR-D than MSI PCR, and allows identification of MSI-H across various cancers not typically screened, as highlighted by the finding that 35% (68 of 193) of all MSI-H tumors were non-CRC/ UEC.

Fig 1.

Distribution of scores for a colorectal cancer/uterine endometrioid cancer (CRC/UEC) validation data set and receiver operating characteristic analysis. Box plots and scatter plots show the distribution of MSIsensor scores for the CRC/UEC samples used in validation. (A) Green points indicate samples where orthogonal methodology indicated microsatellite stable (MSS) phenotype; red points indicate samples where a microsatellite instability–high (MSI-H) phenotype is observed. (B) Receiver operating characteristic analysis shows sensitivity and false-positive rate (1-specificity) analysis across various cutoff values for the validation data set. In rare cases, MSIsensor demonstrated higher sensitivity than MSI polymerase chain reaction for mismatch repair deficiency detection. (C) A poorly differentiated CRC (hematoxylin and eosin [HE] stain) with (D) MLH1 and (E) PMS2 loss only had one unstable locus, (F) Mono-27 (arrows), and was not MSI-H by polymerase chain reaction, whereas (G) MSIsensor detected MSI-H status with ≥ 10% of loci with instability. Arrows indicate the range of instability in the tumor and normal samples.

Fig 2.

MSIsensor score and polymerase chain reaction/immunohistochemistry (IHC) status across tumor types. Of 11,553 samples from 10,900 patients, 204 from 193 patients were microsatellite instability high (MSI-H) by MSIsensor. Validation showed high concordance, with only two cancers with MSI-H/mismatch repair deficiency status by polymerase chain reaction/IHC that were < 10 by MSIsensor. A single uterine endometrioid cancer with an MSIsensor score of 16 was equivocal by mismatch repair IHC. MMR-D, mismatched repair deficient; MMR-P, mismatched repair proficient; MSS, microsatellite stable.

Fig 3.

Tumor mutation burden (TMB) versus MSIsensor score and signature analysis. (A) For cancers with high TMB and microsatellite stable results on the basis of MSIsensor, (B) other signatures were identified, including UV exposure, APOBEC deficiency, smoking, temozolomide (TMZ) treatment, and polymerase ε (POLE) deficiency. (C) On the other hand, mismatch repair-deficient (MMR-D)/microsatellite instability–high (MSI-H) samples showed similar rates of TMB. Mb, megabase.

Fig 4.

Survival analysis of microsatellite stable (MSS) versus microsatellite instability–high (MSI-H) status. Although colorectal cancer (CRC) showed a trend of better survival for patients with MSI-H status (P = .057), other tumor types did not show a significant association between MSI status and survival. UEC, uterine endometrioid cancer.

Fig A1.

MSIsensor scores differ with matched versus pooled normal DNA. Ten tumors that were microsatellite stable by polymerase chain reaction/immunohistochemistry were analyzed with MSIsensor by using their matched normal and an unrelated pool of normals as comparator. Unmatched analysis significantly inflated the MSIsensor scores (P < .001), which rendered the current MSIsensor score cutoff of 10 as incorrect.

Fig A2.

Distribution of MSIsensor scores for 11,553 samples from 10,900 patients in the cohort. Inset shows the distribution of MSIsensor scores that were ≥ 10.

Fig A3.

Number of microsatellites evaluated by version of Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) sequencing data. MSK-IMPACT does not cover all genomic microsatellites, yet the number of covered microsatellites has increased with more recent versions.

Fig A4.

Cutoff score analysis for unmatched tumors. Distribution of MSIsensor scores from the unmatched analysis. Samples are classified as stable versus unstable on the basis of the MSIsensor score cutoff of 25.6 on the x-axis, and colors indicate the orthogonal validation status.