Background: Microsatellite instability (MSI) occurs in some tumors from defects in mismatch repair genes. Immune checkpoint inhibition is approved for treatment of MSI-high (MSI-H) tumors. Plasma-based genotyping assays are being used more commonly in breast cancer to identify targetable mutations. The objective of this study was to evaluate MSI-H detection by plasma-based genotyping in metastatic breast cancer (MBC) and understand the co-existing genomic landscape. In selected cases, correlations of MSI-H with clinical characteristics were determined.

Methods:Patients who had MSI-H detected by cell-free DNA (cfDNA) analysis via Guardant360 testing (next-generation sequencing (NGS), up to 74 gene panel) between 9/27/2018 and 3/12/2020 with a diagnosis of breast cancer reported on the test requisition form were identified from a de-identified database. MSI detection is based on plasma NGS of 90 microsatellite sites and combining observed read sequences and molecular barcoding information in a probabilistic model (Willis, 2019). A retrospective review was conducted to identify demographic and genomic characteristics of patients with MSI-H findings. For 11 patients, clinical data was provided by the treating physician for evaluation of demographics and response to immunotherapy.

Results:Of 7824 patients with breast cancer with alterations in plasma NGS, 40 (0.5%) were MSI-H. Of these 40 patients, 40 (100%) were female. The median age was 61 (range 39-92). The median number of genomic alterations per sample was 13 (range 4-69), and the median maximum allelic fraction (MAF) was 15.9% (range 0.92-54.2), compared to 4 and 2.7%, respectively, in all breast cancer samples reported in this time period. Table 1 depicts the most common co-existing non-synonymous mutations, which included DNA damage repair genes (ATM and BRCA2).

Clinical data was available for 11/40 MSI-H patients with MBC. Of these 11 patients, 4 (36%) had triple-negative MBC (mTNBC), and 7 (64%) had hormone receptor positive (HR+)/HER2- MBC. None had known Lynch syndrome. Seven patients received treatment with an immune checkpoint inhibitor (2 atezolizumab/nab-paclitaxel, 3 pembrolizumab, 1 pembrolizumab/capecitabine followed by pembrolizumab/eribulin, and 1 nivolumab). Treatment duration was available for 5 patients, and the median duration of treatment was 108 days (range 65-273 days). Two patients had durable benefit (1 with stable disease for 10 cycles, and another on treatment for > 152 days), both of whom had mTNBC and were treated in the first-line setting with atezolizumab/nab-paclitaxel.

Conclusions:Plasma-based genotyping assays can identify the presence of MSI-H in breast cancer, including in patients with mTNBC and HR+/HER2- MBC. MSI-H breast cancers had a higher number of somatic alterations and MAF, suggesting higher tumor burden and genomic instability. The co-existing genomic landscape is heterogeneous, and mutations in TP53, PI3KCA, ESR1, RB1, NOTCH1, and ARID1A, and DNA damage repair genes (ATM and BRCA2) may be present. Since plasma based genotyping is increasingly being utilized to identify actionable mutations including PI3KCA, the ability to detect additional genomic alterations such as MSI-H extends the potential clinical application. However, the clinical utility of MSI detection by cfDNA needs to be determined and further prospective research is needed to validate the use of immunotherapy in cfDNA detected MSI-H MBC.

Table 1.

GeneNumber (%) of MSI-H patients with ≥ 1 mutation
TP53 31 (78) 
PI3KCA 25 (63) 
ESR1 20 (50) 
RB1 16 (40) 
NOTCH1 15 (38) 
ARID1A 15 (38) 
ATM 14 (35) 
BRAF 14 (35) 
EGFR 13 (33) 
PDGFRA 11 (28) 
PTEN 11 (28) 
APC 11 (28) 
MET 10 (25) 
KIT 10 (25) 
FGFR2 10 (25) 
BRCA2 10 (25) 
GeneNumber (%) of MSI-H patients with ≥ 1 mutation
TP53 31 (78) 
PI3KCA 25 (63) 
ESR1 20 (50) 
RB1 16 (40) 
NOTCH1 15 (38) 
ARID1A 15 (38) 
ATM 14 (35) 
BRAF 14 (35) 
EGFR 13 (33) 
PDGFRA 11 (28) 
PTEN 11 (28) 
APC 11 (28) 
MET 10 (25) 
KIT 10 (25) 
FGFR2 10 (25) 
BRCA2 10 (25) 

Citation Format: Neelima Vidula, Katherine Hesler, Kristin Price, Andrew Lipman, Georges Azzi, Ahmed Elkhanany, Tarek Sabagh, Dejan Juric, Estelamari Rodriguez, Shumei Kato, Joyce O'Shaughnessy, Aditya Bardia. Microsatellite instability high (MSI-H) detection utilizing targeted plasma based genotyping in metastatic breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS18-18.