EMAST (elevated microsatellite alterations at selected tetranucleotide repeats) is a potential biomarker for many cancer types. This article compares clinical and tumor‐specific factors in cases of colorectal cancer evaluated using both EMAST markers and the Bethesda panel of markers of microsatellite instability mutations.
Keywords: EMAST, Microsatellite instability, Colorectal cancer, Immune biomarkers
Abstract
Background.
The form of microsatellite instability (MSI) affecting tetranucleotide repeats known as elevated microsatellite alterations at selected tetranucleotide repeats (EMAST) has emerged as a new potential biomarker in multiple cancers. In colorectal cancer (CRC), the correlation between EMAST and MSI mutations remain inconclusive.
Materials and Methods.
We evaluated 1,505 patients with CRC using five EMAST markers (D20S82, D20S85, D8S321, D9S242, and MYCL1) and the Bethesda panel of MSI markers. Most commonly, mutations involved in CRCs were identified by MassArray Assay, and DNA repair genes were analyzed by next‐generation sequencing. Clinical characteristics and prognostic relevance were correlated with EMAST and MSI.
Results.
Tumors that were EMAST positive and MSI high (MSI‐H) were detected in 159 (10.6%) and 154 (10.2%) of 1,505 patients with CRC. Patients were divided into four groups according to EMAST and MSI status (EMAST‐positive and MSI‐H, EMAST‐positive and microsatellite‐stable [MSS], EMAST‐negative and MSI‐H, and EMAST‐negative and MSS). The EMAST‐positive and MSI‐H group was associated with female predominance, higher prevalence of proximal colon tumors, early stage tumors, poorly differentiated tumors, mucinous histology, and higher incidence of mutations in PI3KCA, BRAF, TGFBR, PTEN, and AKT1 compared with other groups. Furthermore, compared with only EMAST‐positive tumors or only MSI‐H tumors, tumors that were both EMAST‐positive and MSI‐H had a higher frequency of MLH1, MSH3, MSH6, PMS2, and EXO1 gene mutations. Finally, the presence of EMAST‐positive and MSI‐H tumors was a good prognostic indicator in CRC.
Conclusion.
High mutations in several DNA repair genes in EMAST‐positive and MSI‐H tumors suggest that this subtype of CRC might be more suitable for treatment with immune therapy.
Implications for Practice.
Elevated microsatellite alterations at selected tetranucleotide repeats (EMAST) is a unique molecular subtype of colorectal cancer (CRC). The current study demonstrated that the EMAST‐positive and MSI‐high (MSI‐H) group was associated with female predominance, higher prevalence of proximal colon tumors, early stage tumors, poorly differentiated tumors, mucinous histology, and higher incidence of mutations in PI3KCA, BRAF, TGFBR, PTEN, and AKT1 compared with other groups. Most importantly, high mutations in DNA repair genes and MSI‐related genes in EMAST‐positive and MSI‐H tumors suggest that this subtype of CRC might be more suitable for treatment with immune therapy compared with MSI‐H tumors alone.
Introduction
Colorectal cancer (CRC) is the third‐most commonly occurring cancer worldwide and remains a formidable global health burden [1]. Prognosis and treatment decisions continue to be based on the American Joint Committee on Cancer TNM classification [2]. Although this classification is comprehensive, it cannot provide complete prognostic information, given the heterogeneous outcomes associated with cancer in patients with the same tumor stage [3], [4].
Evidence is growing to support the use of molecular subtypes to stratify prognosis and guide therapy. Recently, CRC was classified into four groups known as consensus molecular subtypes (CMSs). Tumors in the CMS1 group were microsatellite unstable and had unique features [5]. Microsatellite instability (MSI) indicates a deficient mismatch repair system responsible for the accumulation of unrepaired frameshift alterations DNA microsatellite repeats, usually in mono‐ and dinucleotides, that is associated with cancer development. MSI is detected in 15% of patients with CRC and is associated with distinct clinical features including female predominance, proximal location in the colon, BRAF mutation, lymphocytic infiltration, poorly differentiated mucinous type appearance, and favorable prognosis in early stage CRC [6], [7]. Importantly, CRC with MSI‐high (MSI‐H) status generally has good response to immunotherapy [8], and pembrolizumab is currently approved for this group of patients.
According to the consensus guidelines by the National Cancer Institute, the standard diagnostic testing of MSI consists of two (A/T)n and three (GT/CA)n microsatellite markers [9]. MSI status is classified as high or low, denoted MSI‐high (MSI‐H) or MSI‐low (MSI‐L), respectively. Recent research demonstrated good response to MSI‐H tumors and led to U.S. Food and Drug Administration (FDA) approval for checkpoint inhibitors for this group of cancers, regardless of the site [10]. Unlike CRC in general, MSI‐H CRC also responds favorably toward immunotherapy. However, the low prevalence of MSI‐H CRC and the low response rates of MSI‐L or microsatellite‐stable (MSS) CRC have limited the role of immunotherapy in this patient group. A distinct genotype consisting of microsatellites that harbor long tetranucleotide repeats has been proposed, termed elevated microsatellite alterations at selected tetranucleotide repeats (EMAST). EMAST have been reported in many cancer types [11], [12], [13], [14], [15], [16] and is also reported to be associated with various clinicopathologic characteristics associated with CRC, including tumor location, MSI, differentiation of tumor, and overall prognosis [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. Importantly, EMAST has a significantly higher prevalence in CRC, ranging from 30% to 60% in various studies [17], [20]. This raises an important clinical question of whether EMAST would have a similar predictive value of response to immunotherapy in CRC as MSI‐H status. However, the documented clinical characteristics of EMAST are inconsistent, and only few studies with significant case numbers have directly investigated the correlation between EMAST and MSI. To answer this question, we retrospectively analyzed a total of 1,505 patients with CRC to assess the relationship between EMAST, MSI, and clinical and tumor‐specific factors, including spectrum of mutations, DNA repair genes, genetic targets, and survival.
Materials and Methods
Patient Selection and Clinical Data Collection
A total of 1,505 patients with colorectal cancer who received surgery at the Taipei Veterans General Hospital between 2000 and 2010 were included in the current analysis [29], [30], [31]. The study was approved by the local institutional reviewer board of Taipei Veterans General Hospital (clinical study number 2017‐06‐004BC). Clinical information was collected prospectively and included patient age, sex, personal and family medical history, tumor location, TNM stage, and pathological prognostic features. Patients who died within 30 days of surgery or had preoperative radiotherapy, chemotherapy, or emergency operations were excluded. The proximal colon was defined as being from the cecum to splenic flexure, the distal colon as between the splenic flexure and rectosigmoid colon, and the rectum as within 15 cm of the anal verge. Following surgery, patients were monitored every 3 months for the first 2 years and semiannually thereafter. The follow‐up protocol included physical examination, digital rectal examination, carcinoembryonic antigen analysis, chest radiography, abdominal sonogram, and computerized tomography, if needed. Proton emission tomography or magnetic resonance imaging was arranged for patients with elevated levels of carcinoembryonic antigen but with an uncertain site of tumor recurrence.
EMAST Analysis
EMAST status was determined using five tetranucleotide microsatellite markers: D20S82, D20S85, D8S321, D9S242, and MYCL1, as described previously [19], [25]. Primer sequences for amplification of these markers were obtained from GenBank (https://blast.ncbi.nlm.nih.gov/Blast.cgi). EMAST detection was performed as previously described [19], [25]. Samples with at least two tetranucleotide markers were defined as EMAST positive, and those with zero or one MSI marker were classified as EMAST negative.
MSI Analysis
According to international criteria, five reference microsatellite markers were used to determine MSI: D5S345, D2S123, BAT25, BAT26, and D17S250. Primer sequences for amplification of these markers were obtained from GenBank. MSI detection was performed as previously described [9]. Samples with at least two MSI markers were defined as being MSI‐H, and those with zero or one MSI marker were classified as MSS.
MassARRAY‐Based Mutation Characterization
Polymerase chain reaction (PCR) and extension primers for 155 mutations were designed using MassArray Assay Design 3.1 software (Sequenom, San Diego, CA), applying default single base extension settings and default parameters; however, the maximum multiplex level input was adjusted to 15. Mutation alleles were designed manually to have lower mass than those of the reference allele through either forward or reverse extension. The resulting primer designs were run through the U.S. National Center for Biotechnology Information's Basic Local Alignment Search Tool (BLAST), and necessary modification were made to avoid pseudogene amplification. PCR was performed in a total volume of 5 μL: 1 pmol of the corresponding primers, 10 ng of genomic DNA, and HotStar reaction mix (Qiagen, Germantown, MD) in 384‐well plates. PCR conditions were as follows: 94°C for 15 minutes, followed by 40 cycles of 94°C (20 seconds), 56°C (30 seconds), 72°C (60 seconds), and a final extension of 72°C for 3 minutes. In the primer extension procedure, each sample was denatured at 94°C, followed by 40 cycles of 94°C (5 seconds), 52°C (5 seconds), and 72°C (5 seconds). The mass spectrum from time‐resolved spectra was retrieved using a MassARRAY mass spectrometer (Sequenom), and each spectrum was then analyzed using the Typer 4.0 software (Sequenom) to identify mutations. Putative mutations were further filtered by manual review.
DNA Repair‐Related Gene Sequencing
To identify mutations contributing to EMAST phenotype, we sequenced the DNA of all exons of 16 well‐known DNA repair‐related genes (EPCAM, EXO1, MLH1, MSH2, MSH3, MSH6, PMS1, PMS2, POLD1, POLE) and MSI‐targeted genes (AXIN1, AXIN2, BAX, CTNNB1, TGFBR2) using a HiSeq2500 system (Illumina, San Diego, CA). A Roche Kapa Library Preparation Kit (Kapa Biosystems, Wilmington, MA) was used to construct a sample library from 100 ng of each individual's DNA. Each DNA sample was fragmented, end‐repaired (to ensure fragments had blunt ends with 5’ or 3’ overhangs), adaptor ligated, and size selected (150~350 base pairs). Probe‐based methods were then used to enrich for the DNA of repair‐related genes. Previously designed probes were synthesized by Integrated DNA Technologies (Skokie, IL, USA) and captured according to Integrated DNA Technologies' protocol. After the probe‐based enrichment of these DNA sequences was performed, the library of each pool was subjected to 14 cycles of amplification. The amplified libraries were quantified using a quantitative PCR system and transferred to a new 1.5 mL tube containing the final pool (10 nM pooled DNA library for sequencing [Illumina HiSeq2500 sequencer, 2 × 100 base pairs]). The raw output for each individual was 250 Mb, and the mean depth of target sequencing was more than ×1,000. The sequence of each read was trimmed based on the quality score (Q30): if less than 45 base pairs, it was discarded. Reads were aligned to the sequence of human hg19 reference genome using the Burrows‐Wheeler Aligner BWA‐MEM algorithm (http://bio-bwa.sourceforge.net/), and the Genome Analysis Toolkit (GATK) Unified Genotyper (GATKLite version 2.3–9) was used for variant detection. After variant identification, the Variant Effect Predictor (http://grch37.ensembl.org/Homo_sapiens/Tools/VEP) was used to annotate the identified variants.
Statistical Analysis
The statistical endpoint for cancer‐specific survival was measured from the date of surgery to the date of death due to tumor recurrence. Patients not known to have died were censored on the date of their last follow‐up. Kaplan‐Meier survival curves were plotted and compared using a log‐rank test. A chi‐square test and two‐tailed Fisher exact procedure were used to compare the genotype frequency of clinicopathological features. Numerical values were compared using Student's t test. Data are expressed as means ± SD. Statistical significance was defined as p < .05. Statistical analyses were performed using SPSS for Windows (version 16.0).
Results
Clinicopathological Features Compared Between EMAST‐Positive and EMAST‐Negative Tumors
Age distribution, gender distribution, and clinicopathological characteristics of the 1,505 patients are presented in Table 1. The mean age of all patients was 70.0 ± 11.5 years (median, 72.2 years). Patients with EMAST‐positive tumors had female predominance (43.4% vs. 33.1%; p = .017), a higher prevalence in the proximal versus distal colon (50.9% vs. 23.7%; p < .001), a tendency toward higher prevalence of early stage disease (stage I and II, 61% vs. 50.4%; p = .020), higher prevalence of poorly differentiated tumors (16.4% vs. 4.6%; p < .001), higher prevalence of tumors with mucinous histology (17.0% vs. 7.9%; p = .001), and a higher association with MSI‐H (51.3% vs. 5.4%; p < .001).
Table 1. Clinicopathological features: Patients with EMAST‐positive versus EMAST‐negative CRC.
Abbreviations: CRC, colorectal cancer; EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability; MSI‐H, MSI‐high.
Mutational Profiling of EMAST‐Positive and EMAST‐Negative Tumors
The most common mutations occurring in patients with CRC [32] are listed in supplemental online Table 1. EMAST‐positive tumors showed a significantly higher incidence of mutations in genes PI3KCA (22.0% vs. 13.0%; p = .003), BRAF (18.2% vs. 2.7%; p < .001), TGFBR (20.8% vs. 1.2%; p < .001), PTEN (3.8% vs. 0.4%; p = .001), and AKT1 (2.5% vs. 0.7%; p = .04), whereas EMAST‐negative tumors showed a significantly higher incidence of mutations only in TP53 (30.4% vs. 18.9%; p = .002).
Clinicopathological Features and Mutation Profiling Compared by EMAST and MSI Status
Our analysis in Table 1 revealed a high prevalence (>50%) of MSI‐H status in EMAST‐positive tumors. This finding prompted us to ask whether clinical features observed in EMAST‐positive tumors would also be influenced by MSI‐H status. To clarify this concern, patients were divided into four groups: the EMAST‐positive and MSI‐H group, the EMAST‐positive and MSS group, the EMAST‐negative and MSI‐H group, and the EMAST‐negative and MSS group (Tables 2 and 3). Surprisingly, EMAST‐positive and MSI‐H tumors were associated with unique clinical features, including female predominance (p = .002), higher prevalence of proximal colon tumors (p < .001), early stage tumors (p = .001), poorly differentiated tumors (p < .001), mucinous histology (p < .001), and a higher incidence of mutations in PI3KCA (p = .027), BRAF (p < .001), TGFBR (p < .001), PTEN (p < .001), and AKT1 (p = .031) compared with other groups. Our findings show that the EMAST‐positive and MSI‐H subgroup consisted of tumors with unique features that are clinically meaningful and relevant.
Table 2. Clinicopathological features according to EMAST and MSI status in patients with CRC.
Abbreviations: CRC, colorectal cancer; EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability; MSI‐H, MSI‐high; MSS, microsatellite stable.
Table 3. Mutation spectra: Patients according to EMAST and MSI CRC.
Abbreviations: CRC, colorectal cancer; EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability; MSI‐H, MSI‐high; MSS, microsatellite stable.
The Prevalence of DNA Repair Gene Mutation and the Genetic Targets of EMAST‐Positive and MSI‐H Tumors
In order to identify the DNA repair genes affected in EMAST‐positive and MSI‐H tumors, we evaluated the frequency of mutations in genes of the DNA mismatch repair pathway, including MLH1, EPCAM, MSH2, MSH6, MSH3, PMS2, PMS1, EXO1, POLD1, and POLE (Fig. 1 and Table 4), in 176 CRC tumors (79 tumors in the EMAST‐positive and MSI‐H group, 51 tumors in the EMAST‐positive and MSS group, 45 tumors in the EMAST‐negative and MSI‐H group, and 2 tumors in the EMAST‐negative and MSS group). In addition, genetic targets of MSI in CRC, including CTNNB1, BAX, TGFBR2, AXIN1, and AXIN2, were also evaluated. The mutation frequency in MSH6, MSH3, PMS2, and EXO1 genes was higher in EMAST‐positive and MSI‐H CRC than in EMAST‐positive CRC (p < .001, p = .005, p = .001, and p = .027, respectively), and the mutation frequency in MLH1, MSH6, and EXO1 genes was higher in EMAST‐positive MSI‐H tumors than in MSI‐H tumors (p = .019, p = .005, p = .046, respectively). Interestingly, the mutation prevalence in genetic targets of MSI (except AXIN2 gene) was higher in EMAST‐positive MSI‐H tumors than in either EMAST‐positive or MSI‐H tumors (p = .011, p = .009, p < .001, p < .001, p = .001, p = .017, p = .001, p = .007, respectively).
Figure 1.
Mutation frequencies in tumor samples from 176 patients. Note the clear separation between MSI and EMAST samples. Red, MSI high or EMAST positive; light blue, microsatellite stable or EMAST negative; green, missense mutation; purple, frameshift mutation; pink, stop codon (gain of function); blue, inframe (deletion); cyan, lost start codon.
Abbreviations: EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability.
Table 4. DNA repair gene mutations between patients with EMAST and MSI CRC.
Abbreviations: CRC, colorectal cancer; EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability.
The Dependence of Outcome on EMAST and MSI Status in Patients with CRC
Cancer‐specific survival was improved in patients with EMAST‐positive or MSI‐H CRC compared with patients with EMAST‐negative or MSS CRC (p < .001 and p = .037, respectively; Fig. 2A–B). In addition, patients with EMAST‐positive and MSI‐H CRC demonstrated the best outcome of all CRC groups (Fig. 2C; p < .001).
Figure 2.
Dependence of overall survival on EMAST and MSI status in patients with colorectal cancer (CRC). (A): Cancer‐specific survival in patients with EMAST‐positive or EMAST‐negative CRC. (B): Cancer‐specific survival in patients with MSI‐high or MSS CRC. (C): Cancer‐specific survival according to MSI and EMAST status in patients with CRC.
Abbreviations: EMAST, elevated microsatellite alterations at selected tetranucleotide repeats; MSI, microsatellite instability; MSS, microsatellite stable.
Discussion
EMAST has been previously described to be associated with a higher prevalence of CRC in the proximal colon, increased CD8+ T‐cell infiltration, and MSI‐H status [18], [22], [25]. However, the relationship of other clinical characteristics with EMAST remains controversial. In published literature, the prevalence of EMAST varied widely, from 22% to 64.8% [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. One study reported an association with poorly differentiated tumors [19], whereas another reported an association with well differentiated adenocarcinoma [18]. Furthermore, the status of EMAST as a poor prognostic factor is also controversial [18], [25], [33]. In the aforementioned studies, case numbers were typically modest, raising questions of their ability to be free from sampling bias and underpowered statistical analyses.
In this current study, we analyzed a total of 1,505 patients, which is the largest cohort to analyze EMAST in patients with CRC. In our cohort, the prevalence of EMAST is 10.6%. Consistent with previous studies, the current study identified predominance of proximal colon [18], strong association with MSI‐H [25], and a poorly differentiated and mucinous type appearance as clinicopathological features of EMAST‐positive tumors [19]. In addition, female predominance, association with early stage, and favorable prognosis in stages I–III were found to be unique features of EMAST‐positive disease (supplemental online Fig. 1). Somatic mutations in PIK3CA, BRAF, TGFBR, PTEN, and AKT1 were more frequent in EMAST‐positive tumors than in EMAST‐negative tumors (supplemental online Table 1), whereas somatic mutations in TP53 were more frequent in EMAST‐negative tumors than in EMAST‐positive tumors (supplemental online Table 1). By contrast, the frequency of TP53 mutations were greater in EMAST‐positive non‐small cell lung cancer [11]. A limitation of the genotyping in this study is that only certain prespecified mutations in a panel of CRC‐related genes were analyzed, and full sequencing was not performed.
Interestingly, several EMAST‐related clinical features and genomic aberrant profiles were very similar to those of MSI‐H tumors [5], [34], [35], [36]. Furthermore, MSI‐H CRC was highly associated with EMAST‐positive CRC (51.3%; Table 1). The results suggest that the unique genomic aberration profiles and clinical features of EMAST might originate from a common mechanism with MSI‐H. The most important finding of our study is the superior survival status in the EMAST‐positive and MSI‐H group. Interestingly, there was a significant increase in survival in the EMAST‐positive and MSI‐H group compared with the EMAST‐negative and MSI‐H group (p < .001; supplemental online Fig. 2). This suggests that the survival benefit conferred from EMAST‐positive tumors cannot be explained solely by its coexistence with MSI‐H status alone. Furthermore, it is striking that somatic mutations in KRAS, TGFBR, and PTEN (p = .022, p = .001, p = .002, respectively; supplemental online Table 4) were more frequent in the EMAST‐negative and MSI‐H group than in the EMAST‐negative and MSS group; the clinical features in these two groups were similar (supplemental online Table 3). Further experiments should be designed to elucidate whether a distinct mechanism exists as to how EMAST contributes to improved outcome.
MSI is due to a deficiency in DNA mismatch repair that leads to the accumulation of insertions and deletions in DNA repeat sequences. In genes containing coding repeats, frameshift mutations are a potential source of neoantigens recognized by the immune system, resulting in infiltration of T lymphocytes in tumors. MSI‐H is an important predictive biomarker of immune therapy, and the FDA has granted accelerated approval to immune checkpoint inhibitor for patients with MSI‐H CRC [37]. EMAST, as aforementioned, is also a microsatellite instability and is associated with CD8 T‐cell infiltration of the tumor nest and stroma [21]. Our study identified the best outcomes in the EMAST‐positive and MSI‐H group. This group of tumors theoretically is associated with increased insertions and deletions in DNA repeat sequences. Although no studies have directly assessed the relationship of EMAST status and tumor mutation burden (TMB) values, it is not unreasonable to hypothesize that EMAST‐positive tumors would possess higher TMB, which in turn would predict better response to immunotherapy [38]. Mutations in several DNA repair genes and gene targets of MSI were higher in EMAST‐positive MSI‐H tumors than in MSI‐H tumors or EMAST‐positive tumors (Fig. 1 and Table 4). Furthermore, the BRAF mutation is observed in approximately two‐thirds of microsatellite‐unstable CRCs caused by MLH1 hypermethylation and protein loss [39]. BRAF mutation was present in 28.4% of EMAST‐positive MSI‐H tumors, which suggests that MSI‐H phenotype might be due to MLH1 promotor methylation (Table 3). Therefore, a combination of MSI and EMAST might be able to identify a unique subtype of CRC that may be more suitable for treatment with immune therapy compared with MSI‐H status alone.
A form of MSI that differs from MSI‐H has been routinely observed while applying the Bethesda panel to cancer diagnostics [9]. This phenotype, termed MSI‐L, was defined as instability at one microsatellite marker when using the Bethesda consensus panel, or instability at more than 30% to 40% of loci if more than five markers are analyzed [29]. MSI‐L tumors display microsatellite allele length changes primarily at dinucleotide microsatellites [40], [41]. Prior investigators have pointed out the overlap between EMAST and MSI‐L status [17]. In our data sets, the incidence of MSI‐L was 4.9% of all patients with CRC, a finding similar to that of previous studies [42]. However, a strong correlation between MSI‐L status and EMAST in our cohort was not observed (supplemental online Table 2). Therefore, our data do not support a significant overlap between EMAST and MSI‐L status in CRC.
The deficiency in DNA mismatch repair protein MutS homolog 3 (MSH3) has been proposed as the underlying mechanism because this protein is responsible for repairing mismatch repair errors at short tendon repeats of at least two nucleotides, including tetranucleotide repeats [17], [23]. Both the loss of MSH3 expression and the translocation of MSH3 protein from the nucleus to the cytoplasm have been linked to EMAST in some studies [19], whereas this observation has not been reported by others [25]. In our study, MSH3 mutation occurred in 22.5% (29 out of 129) of patients with EMAST‐positive tumors. Among these patients, the MSI‐H and EMAST‐positive group had higher MSH3 mutation compared with the EMAST‐alone group (Table 5; 30.8% vs. 9.8%, respectively; p = .005). We failed to demonstrate a correlation between MSH3 mutation and EMAST, especially in the EMAST‐alone group. It must be noted, however, that analyses in our study were limited to genetic aberrations in the MSH3 sequence, whereas MSH3 expression was not assessed. Factors including epigenetic silencing (similar to the phenomenon of MLH3 promoter methylation in MSI) can in theory lead to impaired MSH3 expression and thus affect EMAST. This has been observed in nasopharyngeal carcinoma [43]. Our study supports the rationale for further analyses, including transcriptional and epigenetic profiling of MSH3 and affected pathways, in patients with EMAST‐positive CRC.
Conclusion
We conducted a study with 1,505 patients, the largest cohort size to date, specifically probing the relationship between EMAST and MSI‐H status in CRC. We demonstrated that patients with both EMAST‐positive and MSI‐H CRC were associated with unique clinical features and mutational patterns as well as a significantly better clinical outcome. Most importantly, although the FDA has approved checkpoint inhibitors for MSI‐H tumors, the objective response rates in these patients were only 31.1% to 40% [37], [44]. In this study, the prevalence of increased mutations in several DNA repair genes in the EMAST‐positive and MSI‐H group suggest that this subtype of CRC, compared with MSI alone, might possess more promising outcomes when treated with immune therapy. Further studies can shed light on whether EMAST can serve as a biomarker and/or prognostic factor for immunotherapy regardless of MSI status. If proven true, this could significantly expand the spectrum of immunotherapy in CRC given the high prevalence of EMAST in this cancer group.
See http://www.TheOncologist.com for supplemental material available online.
Acknowledgments
This research was funded by grants from the Taipei Veterans General Hospital (V101E2‐005, V107C‐010, V107D32‐003‐MY2, V106D29‐001‐MY3); Ministry of Science and Technology, Taiwan (105‐2314‐B‐075‐010‐MY2); and Department of Health, Taipei City Government (10401‐62‐031; 10601‐62‐059). This research was also supported by Taiwan Cancer Clinic Foundation and the Yen Tjing Ling Medical Foundation.
Contributed equally.
Author Contributions
Conception/design: Ming‐Huang Chen, Shih‐Ching Chang
Provision of study material or patients: Hung‐Hsin Lin, Wen‐Yi Liang, Muh‐Hwa Yang, Yee Chao
Collection and/or assembly of data: Pei‐Ching Lin, Shung‐Haur Yang, Chun‐Chi Lin, Yuan‐Tzu Lan
Data analysis and interpretation: Chien‐Hsing Lin, Meng‐Lun Lu
Manuscript writing: Ming‐Huang Chen, Jiun‐I Lai
Final approval of manuscript: Ming‐Huang Chen, Shih‐Ching Chang, Pei‐Ching Lin, Shung‐Haur Yang, Chun‐Chi Lin, Yuan‐Tzu Lan, Hung‐Hsin Lin, Chien‐Hsing Lin, Jiun‐I Lai, Wen‐Yi Liang, Meng‐Lun Lu, Muh‐Hwa Yang, Yee Chao
Disclosures
The authors indicated no financial relationships.
References
- 1.Torre LA, Bray F, Siegel RL et al. Global cancer statistics, 2012. CA Cancer J Clin 2015;65:87–108. [DOI] [PubMed] [Google Scholar]
- 2.Weitz J, Koch M, Debus J et al. Colorectal cancer. Lancet 2005;365:153–165. [DOI] [PubMed] [Google Scholar]
- 3.Zlobec I, Lugli A. Prognostic and predictive factors in colorectal cancer. J Clin Pathol 2008;61:561–569. [DOI] [PubMed] [Google Scholar]
- 4.Gunderson LL, Jessup JM, Sargent DJ et al. Revised TN categorization for colon cancer based on national survival outcomes data. J Clin Oncol 2010;28:264–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Guinney J, Dienstmann R, Wang X et al. The consensus molecular subtypes of colorectal cancer. Nature Med 2015;21:1350–1356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology 2010;138:2073–2087 e2073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lin CC, Lai YL, Lin TC et al. Clinicopathologic features and prognostic analysis of MSI‐high colon cancer. Int J Colorectal Dis 2012;27:277–286. [DOI] [PubMed] [Google Scholar]
- 8.Asaoka Y, Ijichi H, Koike K. PD‐1 blockade in tumors with mismatch‐repair deficiency. N Engl J Med 2015;373:1979. [DOI] [PubMed] [Google Scholar]
- 9.Boland CR, Thibodeau SN, Hamilton SR et al. A National Cancer Institute workshop on microsatellite instability for cancer detection and familial predisposition: Development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998;58:5248–5257. [PubMed] [Google Scholar]
- 10.Boyiadzis MM, Kirkwood JM, Marshall JL et al. Significance and implications of FDA approval of pembrolizumab for biomarker‐defined disease. J Immunother Cancer 2018;6:35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ahrendt SA, Decker PA, Doffek K et al. Microsatellite instability at selected tetranucleotide repeats is associated with p53 mutations in non‐small cell lung cancer. Cancer Res 2000;60:2488–2491. [PubMed] [Google Scholar]
- 12.Danaee H, Nelson HH, Karagas MR et al. Microsatellite instability at tetranucleotide repeats in skin and bladder cancer. Oncogene 2002;21:4894–4899. [DOI] [PubMed] [Google Scholar]
- 13.Singer G, Kallinowski T, Hartmann A et al. Different types of microsatellite instability in ovarian carcinoma. Int J Cancer 2004;112:643–646. [DOI] [PubMed] [Google Scholar]
- 14.Arai H, Okudela K, Oshiro H et al. Elevated microsatellite alterations at selected tetra‐nucleotide (EMAST) in non‐small cell lung cancers–a potential determinant of susceptibility to multiple malignancies. Int J Clin Exp Pathol 2013;6:395–410. [PMC free article] [PubMed] [Google Scholar]
- 15.Burger M, Denzinger S, Hammerschmied CG et al. Elevated microsatellite alterations at selected tetranucleotides (EMAST) and mismatch repair gene expression in prostate cancer. J Mol Med (Berl) 2006;84:833–841. [DOI] [PubMed] [Google Scholar]
- 16.Stoehr C, Burger M, Stoehr R et al. Mismatch repair proteins hMLH1 and hMSH2 are differently expressed in the three main subtypes of sporadic renal cell carcinoma. Pathobiology 2012;79:162–168. [DOI] [PubMed] [Google Scholar]
- 17.Haugen AC, Goel A, Yamada K et al. Genetic instability caused by loss of MutS homologue 3 in human colorectal cancer. Cancer Res 2008;68:8465–8472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yamada K, Kanazawa S, Koike J et al. Microsatellite instability at tetranucleotide repeats in sporadic colorectal cancer in Japan. Oncol Rep 2010;23:551–561. [PMC free article] [PubMed] [Google Scholar]
- 19.Lee SY, Chung H, Devaraj B et al. Microsatellite alterations at selected tetranucleotide repeats are associated with morphologies of colorectal neoplasias. Gastroenterology 2010;139:1519–1525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Devaraj B, Lee A, Cabrera BL et al. Relationship of EMAST and microsatellite instability among patients with rectal cancer. J Gastrointest Surg 2010;14:1521–1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lee SY, Miyai K, Han HS et al. Microsatellite instability, EMAST, and morphology associations with T cell infiltration in colorectal neoplasia. Dig Dis Sci 2012;57:72–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Goel A. Colorectal cancer: Sailing with a T‐cell EMAST. Dig Dis Sci 2012;57:1–3. [DOI] [PubMed] [Google Scholar]
- 23.Campregher C, Schmid G, Ferk F et al. MSH3‐deficiency initiates EMAST without oncogenic transformation of human colon epithelial cells. PLoS One 2012;7:e50541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Tseng‐Rogenski SS, Chung H, Wilk MB et al. Oxidative stress induces nuclear‐to‐cytosol shift of hMSH3, a potential mechanism for emast in colorectal cancer cells. PLoS One 2012;7:e50616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Venderbosch S, van Lent‐van Vliet S, de Haan AF et al. EMAST is associated with a poor prognosis in microsatellite instable metastatic colorectal cancer. PLoS One 2015;10:e0124538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hamaya Y, Guarinos C, Tseng‐Rogenski SS et al. Efficacy of adjuvant 5‐fluorouracil therapy for patients with EMAST‐positive stage II/III colorectal cancer. PLoS One 2015;10:e0127591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Watson MM, Lea D, Rewcastle E et al. Elevated microsatellite alterations at selected tetranucleotides in early‐stage colorectal cancers with and without high‐frequency microsatellite instability: Same, same but different? Cancer Med 2016;5:1580–1587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lee HS, Park KU, Kim DW et al. Elevated microsatellite alterations at selected tetranucleotide repeats (EMAST) and microsatellite instability in patients with colorectal cancer and its clinical features. Curr Mol Med 2016;16:829–839. [DOI] [PubMed] [Google Scholar]
- 29.Lan YT, Jen‐Kou L, Lin CH et al. Mutations in the RAS and PI3K pathways are associated with metastatic location in colorectal cancers. J Surg Oncol 2015;111:905–910. [DOI] [PubMed] [Google Scholar]
- 30.Chang SC, Lin PC, Lin JK et al. Mutation spectra of common cancer‐associated genes in different phenotypes of colorectal carcinoma without distant metastasis. Ann Surg Oncol 2016;23:849–855. [DOI] [PubMed] [Google Scholar]
- 31.Lin JK, Chang SC, Yang YC et al. Loss of heterozygosity and DNA aneuploidy in colorectal adenocarcinoma. Ann Surg Oncol 2003;10:1086–1094. [DOI] [PubMed] [Google Scholar]
- 32.Cancer Genome Atlas Network . Comprehensive molecular characterization of human colon and rectal cancer. Nature 2012;487:330–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Garcia M, Choi C, Kim HR et al. Association between recurrent metastasis from stage II and III primary colorectal tumors and moderate microsatellite instability. Gastroenterology 2012;143:48–50.e41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mlecnik B, Bindea G, Angell HK et al. Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of patient survival than microsatellite instability. Immunity 2016;44:698–711. [DOI] [PubMed] [Google Scholar]
- 35.Abubaker J, Bavi P, Al‐Harbi S et al. Clinicopathological analysis of colorectal cancers with PIK3CA mutations in Middle Eastern population. Oncogene 2008;27:3539–3545. [DOI] [PubMed] [Google Scholar]
- 36.Lin EI, Tseng LH, Gocke CD et al. Mutational profiling of colorectal cancers with microsatellite instability. Oncotarget 2015;6:42334–42344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Le DT, Uram JN, Wang H et al. PD‐1 blockade in tumors with mismatch‐repair deficiency. N Engl J Med 2015;372:2509–2520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Rizvi NA, Hellmann MD, Snyder A et al. Cancer immunology. Mutational landscape determines sensitivity to PD‐1 blockade in non‐small cell lung cancer. Science 2015;348:124–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Palomaki GE, McClain MR, Melillo S et al. EGAPP supplementary evidence review: DNA testing strategies aimed at reducing morbidity and mortality from Lynch syndrome. Genet Med 2009;11:42–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Dietmaier W, Wallinger S, Bocker T et al. Diagnostic microsatellite instability: Definition and correlation with mismatch repair protein expression. Cancer Res 1997;57:4749–4756. [PubMed] [Google Scholar]
- 41.Goel A, Arnold CN, Niedzwiecki D et al. Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res 2003;63:1608–1614. [PubMed] [Google Scholar]
- 42.de la Chapelle A, Hampel H. Clinical relevance of microsatellite instability in colorectal cancer. J Clin Oncol 2010;28:3380–3387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Ni H, Jiang B, Zhou Z et al. Inactivation of MSH3 by promoter methylation correlates with primary tumor stage in nasopharyngeal carcinoma. Int J Mol Med 2017;40:673–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Overman MJ, McDermott R, Leach JL et al. Nivolumab in patients with metastatic DNA mismatch repair‐deficient or microsatellite instability‐high colorectal cancer (CheckMate 142): An open‐label, multicentre, phase 2 study. Lancet Oncol 2017;18:1182–1191. [DOI] [PMC free article] [PubMed] [Google Scholar]