Improved Microsatellite Instability Detection In Endometrial And Prostate Cancers Using Long Mononucleotide Repeat Markers

Microsatellite instability (MSI) assays are used to screen cancer patients with mismatch repair deficiency (dMMR) for immune checkpoint inhibitor therapy. However, short mononucleotide repeat (SMR) markers in current MSI assays may sometimes display subtle shifts that are difficult to interpret in certain dMMR cancers, particular non-colorectal cancers. Long mononucleotide repeat (LMR) markers that are more prone to replication errors may improve the sensitivity of MSI assays. In this study, we compared the performance of SMR and LMR markers using two panels: (1) MSI Analysis System Version 1.2 (Promega) consisting of five SMR markers (21-27 bases in length) and (2) MSI LMR System (Promega) consisting of four SMR markers from the Version 1.2 panel and four LMR markers (52-60 bases in length). We studied 20 MMR proficient (pMMR) and 18 dMMR colorectal cancers (CRC) defined by concordant immunohistochemical staining (IHC) and the Version 1.2 MSI panel, 12 pMMR and 9 dMMR endometrial cancers (EC) defined by IHC alone, and 9 dMMR prostate cancers (PC) defined by loss of MSH2 IHC staining, loss of function MSH2 mutations, and increased mutation frequency. A marker was defined as unstable if a shift of 2 or more bases was observed in the tumor sample when compared to the normal sample. When using the Version 1.2 panel, we defined MSI high (MSI-H) as 2 or more unstable markers, MSI low (MSI-L) as 1 unstable marker, and microsatellite stable (MSS) as no shifts. With the LMR panel, we defined MSI-H as 3 or more unstable markers, MSI-L as 1 or 2 unstable markers, and MSS as no shifts. In dMMR CRC, the clinical sensitivity of the LMR and Version 1.2 panels for MSI were both 100%. In pMMR CRC, MSI-H was not observed in any cases using either panel, but MSI-L was observed in 2 and 0 cases using the LMR and Version 1.2 panels, respectively. In dMMR CRC, the average size of allelic shifts in the four LMR and SMR markers from the LMR panel were 8.7 and 22.1 bases, respectively. Germline polymorphisms were observed in 47% and 1% of LMR and SMR markers, respectively. In dMMR EC, the LMR panel showed greater sensitivity than the Version 1.2 panel (100% vs 89%). In pMMR EC, MSI-H was not observed in any cases using either panel, but MSI-L was observed more often using the LMR panel when compared to the Version 1.2 panel (2 vs 0 cases). In addition, LMR markers displayed larger shifts than SMR markers (12.1 vs 4.9 bases). Polymorphisms were also more frequent in LMR markers relative to SMR markers (60% vs 2%). Likewise, in dMMR PC, the sensitivity of the LMR panel was greater than that of the Version 1.2 panel (89% vs 44%). Furthermore, shifts in LMR markers were larger than those in SMR markers (10.3 vs 3.1 bases), and polymorphisms in LMR markers were more frequent than in SMR markers (33% vs 0%). Overall, LMR markers display greater clinical sensitivity and larger shifts, indicating that LMR markers can improve MSI detection in non-colorectal cancer. Citation Format: John H. Lin, Suping Chen, Liana B. Guedes, Emmanuel S. Antonarakis, Colin C. Pritchard, Tamara L. Lotan, James R. Eshleman. Improved microsatellite instability detection in endometrial and prostate cancers using long mononucleotide repeat markers [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 740.