Comparison of Mismatch Repair Status Between Primary and Matched Metastatic Sites in Patients With Colorectal Cancer

Background: Differences between the features of primary cancer and matched metastatic cancer have recently drawn attention in research. This study investigated the concordance in microsatellite instability (MSI) and mismatch repair (MMR) status between primary and corresponding metastatic colorectal cancer (CRC). Methods: Consecutive patients with metastatic CRC who had both primary and metastatic tumors diagnosed at our institution in January 2008 through December 2016 were identified. Immunohistochemistry was used to test the MMR status of both primary and matched metastatic tumors, and PCR analysis was performed to test MSI in patients with deficient MMR (dMMR) status. Results: A total of 369 patients were included. Of the 46 patients with MSI-high primary tumors, 37 (80.4%) also had MSI-high metastatic tumors, whereas 9 (19.6%) had microsatellite stable (MSS) metastatic tumors. A high concordance was found in patients with liver, lung, or distant lymph node metastases. Interestingly, the discrepancy was more likely to be limited to peritoneal (5/20) or ovarian (4/4) metastasis (chi-square test, P<.001). These organ-specific features were also found in the pooled analysis. Along with the change of MSI-high in primary cancer to MSS in metastatic cancer, lymphocyte infiltration decreased significantly (P=.008). However, the change did not influence survival; the median overall survival of MSI-high and MSS metastatic tumors was 21.3 and 21.6 months, respectively (P=.774). The discrepancy rate was 1.6% for patients with proficient MMR primary tumors. Conclusions: For patients with dMMR primary tumors, the concordance of MSI and MMR status in primary CRC and corresponding metastatic cancer is potentially organ-specific. High concordance is found in liver, lung, and distant lymph node metastases, whereas discrepancy is more likely to occur in peritoneal or ovarian metastasis. Rebiopsy to evaluate MSI-high/dMMR status might be needed during the course of anti–PD-1 therapy in cases of peritoneal or ovarian metastasis.

Abstract

Background: Differences between the features of primary cancer and matched metastatic cancer have recently drawn attention in research. This study investigated the concordance in microsatellite instability (MSI) and mismatch repair (MMR) status between primary and corresponding metastatic colorectal cancer (CRC). Methods: Consecutive patients with metastatic CRC who had both primary and metastatic tumors diagnosed at our institution in January 2008 through December 2016 were identified. Immunohistochemistry was used to test the MMR status of both primary and matched metastatic tumors, and PCR analysis was performed to test MSI in patients with deficient MMR (dMMR) status. Results: A total of 369 patients were included. Of the 46 patients with MSI-high primary tumors, 37 (80.4%) also had MSI-high metastatic tumors, whereas 9 (19.6%) had microsatellite stable (MSS) metastatic tumors. A high concordance was found in patients with liver, lung, or distant lymph node metastases. Interestingly, the discrepancy was more likely to be limited to peritoneal (5/20) or ovarian (4/4) metastasis (chi-square test, P<.001). These organ-specific features were also found in the pooled analysis. Along with the change of MSI-high in primary cancer to MSS in metastatic cancer, lymphocyte infiltration decreased significantly (P=.008). However, the change did not influence survival; the median overall survival of MSI-high and MSS metastatic tumors was 21.3 and 21.6 months, respectively (P=.774). The discrepancy rate was 1.6% for patients with proficient MMR primary tumors. Conclusions: For patients with dMMR primary tumors, the concordance of MSI and MMR status in primary CRC and corresponding metastatic cancer is potentially organ-specific. High concordance is found in liver, lung, and distant lymph node metastases, whereas discrepancy is more likely to occur in peritoneal or ovarian metastasis. Rebiopsy to evaluate MSI-high/dMMR status might be needed during the course of anti–PD-1 therapy in cases of peritoneal or ovarian metastasis.

Background

Colorectal cancer (CRC) is among the most common cancers worldwide,1 and distant metastasis is extremely common in CRC.2 Recently, different profiles between the primary CRC and corresponding metastatic cancers have been found, particularly genetic discordances between the primary and the metastatic sites.36 Moreover, oxidative DNA damage and some chemokines were significantly higher in metastases than in primary tumors,7,8 and the level of tumor-infiltrating lymphocytes and expression of PD-L1 differed between primary and metastatic tumors,9,10 with the discrepancy being site-specific in some cases.11 These differences indicate that metastases may provide a better “predictive window” for targeted therapy than the primary cancer alone.3

The mismatch repair (MMR)/microsatellite instability (MSI) pathway significantly contributes to CRC development12,13 and is associated with the immune microenvironment.14,15 However, although the distinct features of MSI-high (MSI-H) tumors have been identified,16 MSI status of the primary CRC versus its metastatic lesions has not been fully explored. Change in MSI status is important because studies have shown that only patients with deficient MMR (dMMR) or MSI-high tumors benefit from anti–PD-1 therapy.17,18 Haraldsdottir et al19 reported a 100% concordance rate in MMR status between the primary tumor and corresponding metastatic tissue in 25 patients with CRC, whereas Fujiyoshi et al20 found a discordance in 3 patients and Jung et al21 found a concordance rate of only 77%.

Our study evaluated concordance in MSI and MMR status between primary and corresponding metastatic tumors in CRC. Because prior studies examining this were small, we also performed a pooled analysis.

Methods

Patients and Ethical Concerns

This study evaluated consecutive patients with metastatic CRC diagnosed with both primary and metastatic tumors at Sun Yat-sen University Cancer Center in January 2008 through December 2016. Those with insufficient primary tumor tissue, frozen metastatic tumor tissue, and a history of receiving neoadjuvant chemoradiation before surgery were excluded (Figure 1). This study was approved by the ethics review board at Sun Yat-sen University Cancer Center, and informed consent was obtained from all patients.

Figure 1.
Figure 1.

Flowchart of patient selection.

Abbreviations: dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

Assessment of MMR Status

Formalin-fixed, paraffin-embedded (FFPE) tumor tissue was selected per case, and immunostaining was performed using standard protocols. The 4 most common MMR proteins, namely, MLH1, MSH2, MSH6, and PMS2, were analyzed using the anti-MLH1, anti-MSH2, anti-MSH6, and anti-PMS2 antibodies, respectively (1:50; Beijing Zhongshan Golden Bridge Biotechnology Co, Ltd). Methods are described in detail in eAppendix 1, available with this article at JNCCN.org.

Assessment of Lymphocyte Infiltration

Tumor microenvironment parameters were assessed as previously reported.22,23 Briefly, absent and weak increases in lymphocytes were considered low-grade infiltration, whereas moderate to severe increases were considered high-grade (see eAppendix 1).

DNA Isolation and Assessment of Tumor MSI

Genomic DNA was extracted using a QIAamp DNA FFPE Tissue Kit (QIAGEN). PCR analyses were performed to test for MSI when tumors were classified as dMMR. MSI analysis was performed using 5 microsatellite markers (BAT-25, BAT-26, D2S123, D5S346, and D17S250) recommended by NCI.

Germline MMR Mutational Analysis

Patients with MSI-high tumors underwent germline genetic testing of the MMR genes. The genetic susceptibility panel targeting all exons of MLH1, MSH2, MSH6, PMS2, and EPCAM, and proximal intronic flanking sequences (±10 base pairs [bp]) were analyzed for screening of single-nucleotide variation, including insertion and deletions shorter than 20 bp (see eAppendix 1).

BRAF V600E Mutation Analysis

In patients with MSI-high tumors, a BRAF V600E mutation analysis was performed with quantitative real-time PCR using the minor groove binder probes (see eAppendix 1).

Tumor Mutational Burden Analysis

Tumor mutational burden (TMB) was measured by counting all nonsynonymous missense mutations that had not been previously reported as germline alterations. The workflow of TMB analysis according to HaploX Biotechnology’s laboratory standardization24 is available in eAppendix 1.

MLH1 Promoter Methylation Analysis

Methylation in the region near the start codon of MLH1 was assessed using bisulfite-treated DNA. The PCR conditions were first set at 98°C for 4 minutes; following this, 20 cycles were completed at 94°C for 45 seconds, 66°C for 45 seconds, and 72°C for 1 minute. Then 20 cycles were completed at 94°C for 45 seconds, 56°C for 45 seconds, and 72°C for 1 minute, and a final extension was completed at 72°C for 8 minutes. The PCR reaction was performed with the following primers: M254-3F (5′-GTT TTG GTT TTA GTT TAG GAG TAT TGG-3′) and M254-3R (5′-CCR AAT TCA AAA ACT TAC TAT AAA CCT AC-3′).

Identification of Studies and Eligibility Criteria

Related studies published before August 31, 2018 were searched in PubMed with the following keywords: (‘microsatellite instability’ OR ‘MSI’ OR ‘mismatch repair’ OR ‘MMR’) AND (‘cancer’ OR ‘neoplasm’ OR ‘carcinoma’ OR ‘adenocarcinoma’ OR ‘tumor’) AND (‘colorectal’ OR ‘colon’ OR ‘rectal’ OR ‘rectum’) AND (‘metastases’ OR ‘metastasis’ OR ‘metastatic’). All items were limited in [Title/Abstract]. We included only cohort or case-control studies in which the MMR or MSI status in the primary tumor and the MMR or MSI status in a paired metastatic tumor were reported. Animal studies, review articles, case reports, and duplicate studies were excluded. The study selection flowchart is summarized in supplemental eFigure 1.

Statistical Analysis

All statistical analyses were performed using SPSS Statistics, version 13.0 (SPSS Inc). Frequencies and descriptive statistics were used to report patient characteristics and chi-square test was used to detect differences. Overall survival (OS) was defined as the time from diagnosis of metastasis to the date of death or last follow-up (July 31, 2018). Survival curves were calculated using the Kaplan-Meier method, and the differences were compared using the log-rank test. A P value ≤.05 was considered significant.

Results

Patient Characteristics

In total, 369 patients with a median age at diagnosis of 59 years were evaluated (Figure 1, Table 1). Of these, 221 (59.9%) were men and 148 (40.1%) were women, and 72 (19.5%) and 297 (80.5%) had metachronous and synchronous metastases, respectively. Among those with metachronous metastases, the median time interval was 15.6 months (range, 4.0–96.7 months). A total of 112 (30.4%), 147 (39.8%), and 110 patients (29.8%) were diagnosed with right-sided colon cancer, left-sided colon cancer, and rectal cancer, respectively. Patient and tumor characteristics are shown in Table 1.

Table 1.

Patient and Tumor Characteristics

Table 1.

Patients With dMMR Primary Tumors

A total of 48 patients had dMMR primary tumors, in whom PCR analyses were performed to test for MSI; 2 patients had MSI-low tumors, and their corresponding metastatic tumors were also MSI-low. The remaining 46 patients had MSI-high primary tumors. Among these, 37 (80.4%) had MSI-high and 9 (19.6%) had microsatellite stable (MSS) metastatic tumors. The site of metastasis was the perineum in 20 patients (43.5%), the liver in 10 (21.7%), distant lymph nodes in 9 (19.6%), the ovaries in 4 (8.7%), and the lungs in 3 (6.5%). No discrepancy regarding MSI status was observed in patients with liver, lung, or distant lymph node metastasis. Surprisingly, 5 of 20 patients with peritoneal metastasis and all 4 patients with ovarian metastasis had an MSS metastatic tumor (Figure 2A, B). MSI status was more likely to be different between primary and metastatic sites in patients with MSI-high primary tumors who developed peritoneal (5/20) or ovarian (4/4) metastasis rather than other types of metastasis (P<.001; Table 2). Detailed information for these 9 patients is shown in Table 3. No significant association between the MSI-high discrepancy rate and patient age (P=.135), sex (P=.719), primary tumor location (P=.580), and synchronous versus metachronous metastasis (P=1.00) was noted (Table 4).

Figure 2.
Figure 2.

Circos diagrams of (A) MMR status between primary tumors and metastatic tumors and (B) MSI status of metastatic tumors in patients with MSI-H primary tumors.

Abbreviations: dMMR, deficient mismatch repair; MMR, mismatch repair; MSI, microsatellite instability; MSI-H, microsatellite instability-high; MSS, microsatellite stable; MT, metastatic tumor; pMMR, proficient mismatch repair; PT, primary tumor.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

Table 2.

MSI Status of Metastatic Tumors in Patients With MSI-High Primary Tumors

Table 2.
Table 3.

Characteristics of Patients With MSI-High Primary and MSS Metastatic Tumors (N=9)

Table 3.
Table 4.

Association Between MSI-High Discrepancy Rate and Characteristics of Patients With MSI-High Primary CRC

Table 4.

We also analyzed the TMB in both primary and metastatic tumors, because it has been reported to be associated with MMR status.25 Only 5 patients had acceptable DNA for analysis. Of these patients, 3 had MSI-high primary and MSS metastatic tumors (Patients 2, 5, and 8 in Table 3), and we observed a corresponding decrease in TMB from primary to metastatic tumors (Figure 3A). Two patients had MSI-high primary and metastatic tumors, and TMBs were comparable between their primary and metastatic tumors (Figure 3B). We were unable to perform a statistical analysis because of the limited number of patients.

Figure 3.
Figure 3.

Comparison of tumor mutational burden between primary and matched metastatic tumors in patients with (A) MSI-high primary tumors and MSS metastatic tumors and (B) MSI-high primary tumors and MSI-high metastatic tumors.

Abbreviations: Mb, megabase; MSI, microsatellite instability; MSS, microsatellite stable; Muts, mutations.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

We then studied whether lymphocyte infiltration reflected a corresponding change in MSI status. For the 9 patients with MSI-high primary and MSS metastatic tumors, the hematoxylin-eosin (HE)–stained tissue sections of 5 patients were reviewed (Patients 1, 2, 3, 6, and 9 in Table 3). These patients had high-grade lymphocyte infiltration in a primary tumor and low-grade lymphocyte infiltration in a metastatic tumor (P=.008; Figure 4). For the 37 patients with MSI-high primary and metastatic tumors, 32 had available HE-stained tissue sections; high-grade lymphocyte infiltration was identified in a primary tumor in 27 patients, and in a metastatic tumor in 20 patients. The lymphocyte infiltrations were similar between primary and metastatic tumors in these patients (P=.088; Figure 5).

Figure 4.
Figure 4.

Comparison of lymphocyte infiltration between primary and matched metastatic tumors in patients with MSI-high primary tumors and MSS metastatic tumors (hematoxylin-eosin, original magnification ×100). These patient numbers correspond to those in Table 3.

Abbreviations: MSI, microsatellite instability; MSS, microsatellite stable.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

Figure 5.
Figure 5.

Comparison of lymphocyte infiltration between primary and matched metastatic tumors in 3 representative patients with MSI-high primary tumors and MSI-high metastatic tumors (hematoxylin-eosin, original magnification ×100).

Abbreviation: MSI, microsatellite instability.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

For patients with MSI-high primary tumors, those with MSI-high and MSS metastatic tumors had comparable OS (median, 21.3 vs 21.6 months; P=.774; Figure 6).

Figure 6.
Figure 6.

Comparison of overall survival in patients with MSI-H primary tumors with MSS versus MSI-H metastatic tumors.

Abbreviations: MSI-H, microsatellite instability-high; MSS, microsatellite stable.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 10; 10.6004/jnccn.2019.7308

Among all 46 patients with MSI-high primary tumors, 13 (28.3%) had germline MMR gene mutations, including 10 with MLH1 mutations, 2 with MSH2 mutations, and only 1 with an MSH6 mutation. In these 13 patients, both primary and metastatic tumors were MSI-high. Only 2 patients had a BRAF V600E mutation, and both had MSI-high primary and metastatic tumors.

Of the 46 patients with MSI-high primary tumors, 30 had DNA that qualified for MLH1 promoter methylation analysis. MLH1 promoter methylation was observed in 15 patients (50%); 14 had MSI-high primary and metastatic tumors and 1 had an MSI-high primary tumor and MSS ovarian metastasis. We were unable to perform a statistical analysis and draw a final conclusion because of the limited number of patients.

Patients With pMMR Primary Tumors

Overall, 321 patients had proficient MMR (pMMR) primary tumors; of them, 316 (98.4%) had pMMR metastatic tumors and 5 (1.6%) had dMMR metastatic tumors. Among the 5 patients with dMMR metastatic tumors, 3 had MSI-low and 2 had MSI-high metastatic tumors (Figure 2A). The primary tumors were confirmed to be MSS in the 2 patients with MSI-high metastatic tumors. Both of these patients were women and had sigmoid colon cancer; one had synchronous liver metastasis and the other had synchronous peritoneum metastasis.

Patients With 2 Sites of Metastases

There were 49 patients with matched samples available from both the primary tumor and metastatic tumors from 2 different organ sites. Of these patients, 47 had both MSS primary and metastatic tumor. The remaining 2 patients had an MSI-high primary tumor; one had lung and liver metastases, with both tumors MSI-high, and the other had peritoneal and liver metastases, with the peritoneal metastasis MSS and the liver metastasis MSI-high.

Pooled Analysis of Published Literature

Given the relatively small number of patients with MSI-high CRC, we searched the literature and performed a pooled analysis to determine the correlation between primary and metastatic CRC with respect to MMR status. A total of 7 studies were included in the final analysis (supplemental eFigure 1)1921,2629; 4 studies used immunohistochemistry to detect MMR status (supplemental eTable 1) and 3 used PCR to detect MSI (supplemental eTable 2). Overall, 354 patients had a pMMR/MSS primary tumor; among them, 343 also had pMMR/MSS metastatic cancers, with a concordance rate of 96.9%. Of the 59 patients with a dMMR/MSI-high primary CRC, 52 had a dMMR/MSI-high metastatic tumor (concordance rate, 88.1%). Among patients with a dMMR/MSI-high primary tumor, no difference in MMR/MSI status between the primary and metastatic tumors was observed in those with liver metastasis.

To further study whether the dMMR/MSI-high discrepancy rate was associated with the metastatic site, we performed a pooled analysis that included data from both the searched literature and our study. A total of 94 patients were analyzed, including 34 with peritoneum metastasis, 27 with liver metastasis, 23 with distant lymph node metastasis, 6 with lung metastasis, and 4 with ovarian metastasis. The MMR/MSI discordance was significantly more frequent between primary and metastatic sites in patients with dMMR/MSI-high primary tumors who developed peritoneal metastasis (7/34) or ovarian metastasis (4/4) compared with metastases at other sites (5/56; P<.001) (Table 5).

Table 5.

Concordance Rates of dMMR or MSI-High Status Between Primary Tumor and Corresponding Metastatic Tumors

Table 5.

Discussion

This study investigated the concordance in MSI/MMR status between primary and corresponding metastatic CRC. Although the MSI/MMR status was highly concordant, we also observed discrepancies, particularly between MSI-high primary CRC and MSS metastatic lesions, and reduced lymphocyte infiltration. Pooled analysis confirmed that the discrepancy was organ-specific and was stronger in peritoneal and ovarian metastases. To our knowledge, this is the largest study focusing on discordance in MSI/MMR status between primary and matched metastatic CRC tumors.

The exact reasons for the discrepancy in MSI status are unclear. We hypothesized that the primary tumor may be composed of distinct subclones. Bian et al30 found that the primary CRC tumor showed more complex subclonal structures than the metastatic tumor, suggesting that metastasis might be a subclone derived from the primary tumor. Xie et al5 compared 2 paired primary CRC and metastatic lesions and found 2 different metastatic patterns. These findings suggest the possibility that only MSS tumor cells develop metastasis, and this is supported by the heterogeneity of MMR defects in primary CRC.3133 A second hypothesis is that the mechanisms underlying MSI may affect tumor concordance. Many of the tumors in this study were Lynch-like; of the 46 evaluated, 13 had germline MMR gene alterations, 2 had BRAF V600E mutation, and 28 were without MLH1 methylation and germline mutation. MSI-high tumors caused by a germline mutation of MMR are unlikely to be inconsistent, because these mutations occur in a very early stage of carcinogenesis. In contrast, MSI-high tumors caused by MLH1 promoter methylation may be inconsistent. The level of methylation is targetable by drugs,34 and discrepancies in methylation between primary and metastatic tumors have been reported.35 Consistent with this assumption, we found no discrepancies in MSI-high tumors caused by germline mutations of MMR genes, whereas one patient with a MLH1 promoter methylation and MSI-high primary tumor developed MSS metastatic tumor.

Another notable observation is that only patients with peritoneal or ovarian metastases had MSI-high primary tumors with corresponding MSS metastatic tumors. Peritoneal and ovarian metastases share several other common features, including a lower incidence36 and worse prognosis.36,37 Specifically, 2 patients in this study with MSI-high primary tumors had metastasis to 2 sites, and only the peritoneal metastasis showed MSS, whereas other metastases remained MSI-high. Organ-specific susceptibility of metastasis is common in CRC and is significantly associated with primary tumor location and BRAF, KRAS, and MSI status.38

Peritoneal metastasis is associated with increased interferon-γ levels and is surrounded by natural killer (NK) cells.37 We previously found a correlation between MSI-high CRC and loss of major histocompatibility complex class I (MHC-I) expression,39 which is an activation signal for NK cells. Because MSI heterogeneity exists in primary CRC3133 and MSI-high tumor cells are more susceptible to NK cells because of loss of MHC-I,40 it might be more difficult for MSI-high tumor cells to survive peritoneal metastasis in the same patient. In a cohort of 296 female patients with CRC, none with ovarian metastases had an MSI-high tumor, whereas 6 of 96 (6.3%) with extraovarian metastases and 15 of 181 (8.3%) without any recurrence or metastasis had a MSI-high phenotype.36

Our results are valuable in that the discrepancy in MSI/MMR status may be among the mechanisms of resistance to anti–PD-1 therapy. Although peritoneal metastasis is associated with early treatment discontinuation in patients with ovarian cancer receiving anti–PD-1 treatment,41 it remains unknown whether this correlation exists in patients with CRC. Theoretically, once the MSI status was inconsistent, the immune microenvironment including tumor-infiltrating lymphocytes changed concordantly, as found in our study.

This study has several limitations. First, it was a retrospective study conducted at a single institution. Second, MSI was tested by analyzing 5 microsatellite markers recommended by NCI, and more advanced modalities may have yielded more accurate results. Next-generation sequencing–based tests can scan thousands of loci, allowing for a more thorough assessment.42 Third, although we found a stronger discrepancy in MSI-high status in peritoneal and ovarian metastases, only a small number of patients were evaluated.

Conclusions

Although we found a high concordance rate for MSI/MMR status between primary CRCs and their matched metastatic lesions, some discrepancies were noted, particularly a change from MSI-high in primary CRC to MSS in corresponding peritoneal and ovarian metastases. Rebiopsy to evaluate MSI-high/dMMR status and attentive evaluation of treatment response in metastatic lesions at these 2 sites might be needed during the course of anti–PD-1 therapy. Further studies that include a larger population are needed to validate these findings.

Acknowledgments

The authors have engaged the services of Editage, which provides English editing services, to improve language in this manuscript.

References

  • 1.

    Siegel RL, Miller KD, Fedewa SA, . Colorectal cancer statistics, 2017. CA Cancer J Clin 2017;67:177–193.

  • 2.

    Hugen N, van de Velde CJ, de Wilt JH, . Metastatic pattern in colorectal cancer is strongly influenced by histological subtype. Ann Oncol 2014;25:651–657.

  • 3.

    Vermaat JS, Nijman IJ, Koudijs MJ, . Primary colorectal cancers and their subsequent hepatic metastases are genetically different: implications for selection of patients for targeted treatment. Clin Cancer Res 2012;18:688–699.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Lee SY, Haq F, Kim D, . Comparative genomic analysis of primary and synchronous metastatic colorectal cancers. PLoS One 2014;9:e90459.

  • 5.

    Xie T, Cho YB, Wang K, . Patterns of somatic alterations between matched primary and metastatic colorectal tumors characterized by whole-genome sequencing. Genomics 2014;104:234–241.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Brannon AR, Vakiani E, Sylvester BE, . Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. Genome Biol 2014;15:454.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.

    van der Waals LM, Jongen JMJ, Elias SG, . Increased levels of oxidative damage in liver metastases compared with corresponding primary colorectal tumors: association with molecular subtype and prior treatment. Am J Pathol 2018;188:2369–2377.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Hu D, Du C, Xue W, . The expression of chemokine receptors CCR6, CXCR2 and CXCR4 is not organ-specific for distant metastasis in colorectal cancer: a comparative study. Histopathology 2013;63:167–173.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Schweiger T, Berghoff AS, Glogner C, . Tumor-infiltrating lymphocyte subsets and tertiary lymphoid structures in pulmonary metastases from colorectal cancer. Clin Exp Metastasis 2016;33:727–739.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Wang HB, Yao H, Li CS, . Rise of PD-L1 expression during metastasis of colorectal cancer: Implications for immunotherapy. J Dig Dis 2017;18:574–581.

  • 11.

    Kim MJ, Lee HS, Kim JH, . Different metastatic pattern according to the KRAS mutational status and site-specific discordance of KRAS status in patients with colorectal cancer. BMC Cancer 2012;12:347.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12.

    Germano G, Lamba S, Rospo G, . Inactivation of DNA repair triggers neoantigen generation and impairs tumour growth. Nature 2017;552:116–120.

  • 13.

    Yim KL. Microsatellite instability in metastatic colorectal cancer: a review of pathology, response to chemotherapy and clinical outcome. Med Oncol 2012;29:1796–1801.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Hamada T, Soong TR, Masugi Y, . TIME (Tumor Immunity in the MicroEnvironment) classification based on tumor CD274 (PD-L1) expression status and tumor-infiltrating lymphocytes in colorectal carcinomas. OncoImmunology 2018;7:e1442999.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Liu S, Kong P, Wang X, . Tumor microenvironment classification based on T-cell infiltration and PD-L1 in patients with mismatch repair-proficient and -deficient colorectal cancer. Oncol Lett 2019;17:2335–2343.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16.

    Venderbosch S, Nagtegaal ID, Maughan TS, . Mismatch repair status and BRAF mutation status in metastatic colorectal cancer patients: a pooled analysis of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res 2014;20:5322–5330.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Dudley JC, Lin MT, Le DT, . Microsatellite instability as a biomarker for PD-1 blockade. Clin Cancer Res 2016;22:813–820.

  • 18.

    Le DT, Uram JN, Wang H, . PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 2015;372:2509–2520.

  • 19.

    Haraldsdottir S, Roth R, Pearlman R, . Mismatch repair deficiency concordance between primary colorectal cancer and corresponding metastasis. Fam Cancer 2016;15:253–260.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Fujiyoshi K, Yamamoto G, Takahashi A, . High concordance rate of KRAS/BRAF mutations and MSI-H between primary colorectal cancer and corresponding metastases. Oncol Rep 2017;37:785–792.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Jung J, Kang Y, Lee YJ, . Comparison of the mismatch repair system between primary and metastatic colorectal cancers using immunohistochemistry. J Pathol Transl Med 2017;51:129–136.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Klintrup K, Mäkinen JM, Kauppila S, . Inflammation and prognosis in colorectal cancer. Eur J Cancer 2005;41:2645–2654.

  • 23.

    Park JH, Powell AG, Roxburgh CS, . Mismatch repair status in patients with primary operable colorectal cancer: associations with the local and systemic tumour environment. Br J Cancer 2016;114:562–570.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    GeneFuse: detection and visualization of target gene fusions from DNA sequencing data. Int J Biol Sci 2018;14:843–848.

    • Crossref
    • Export Citation
  • 25.

    Zhuang W, Ma J, Chen X, . The tumor mutational burden of Chinese advanced cancer patients estimated by a 381-cancer-gene panel. J Cancer 2018;9:2302–2307.

  • 26.

    Murata A, Baba Y, Watanabe M, . Methylation levels of LINE-1 in primary lesion and matched metastatic lesions of colorectal cancer. Br J Cancer 2013;109:408–415.

  • 27.

    Ishimaru G, Adachi J, Shiseki M, . Microsatellite instability in primary and metastatic colorectal cancers. Int J Cancer 1995;64:153–157.

  • 28.

    Larsen NB, Heiberg Engel PJ, Rasmussen M, Rasmussen LJ. Differential expression of hMLH1 in sporadic human colorectal cancer tumors and distant metastases. APMIS 2009;117:839–848.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Melloni G, Doglioni C, Bandiera A, . Prognostic factors and analysis of microsatellite instability in resected pulmonary metastases from colorectal carcinoma. Ann Thorac Surg 2006;81:2008–2013.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30.

    Bian S, Hou Y, Zhou X, . Single-cell multiomics sequencing and analyses of human colorectal cancer. Science 2018;362:1060–1063.

  • 31.

    Tachon G, Frouin E, Karayan-Tapon L, . Heterogeneity of mismatch repair defect in colorectal cancer and its implications in clinical practice. Eur J Cancer 2018;95:112–116.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Chapusot C, Martin L, Bouvier AM, . Microsatellite instability and intratumoural heterogeneity in 100 right-sided sporadic colon carcinomas. Br J Cancer 2002;87:400–404.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33.

    Hühns M, Krohn S, Murua Escobar H, . Genomic heterogeneity in primary colorectal carcinomas and their metastases: born bad or brought up a villain? Hum Pathol 2018;74:54–63.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34.

    Yu G, Wu Y, Wang W, . Low-dose decitabine enhances the effect of PD-1 blockade in colorectal cancer with microsatellite stability by re-modulating the tumor microenvironment. Cell Mol Immunol 2018;16:401–409.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Moarii M, Pinheiro A, Sigal-Zafrani B, . Epigenomic alterations in breast carcinoma from primary tumor to locoregional recurrences. PLoS One 2014;9:e103986.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 36.

    Mori Y, Nyuya A, Yasui K, . Clinical outcomes of women with ovarian metastases of colorectal cancer treated with oophorectomy with respect to their somatic mutation profiles. Oncotarget 2018;9:16477–16488.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Seebauer CT, Brunner S, Glockzin G, . Peritoneal carcinomatosis of colorectal cancer is characterized by structural and functional reorganization of the tumor microenvironment inducing senescence and proliferation arrest in cancer cells. OncoImmunology 2016;5:e1242543.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Prasanna T, Karapetis CS, Roder D, . The survival outcome of patients with metastatic colorectal cancer based on the site of metastases and the impact of molecular markers and site of primary cancer on metastatic pattern. Acta Oncol 2018;57:1438–1444.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39.

    Liu SS, Yang YZ, Jiang C, . Comparison of immunological characteristics between paired mismatch repair-proficient and -deficient colorectal cancer patients. J Transl Med 2018;16:195.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40.

    Konjević G, Vuletić A, Mirjačić Martinović K. Natural killer cell receptors: alterations and therapeutic targeting in malignancies. Immunol Res 2016;64:25–35.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41.

    Boland JL, Zhou Q, Martin M, . Early disease progression and treatment discontinuation in patients with advanced ovarian cancer receiving immune checkpoint blockade. Gynecol Oncol 2019;152:251–258.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 42.

    Middha S, Zhang L, Nafa K, . Reliable pan-cancer microsatellite instability assessment by using targeted next-generation sequencing data. JCO Precis Oncol 2017:2017.

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Submitted September 25, 2018; accepted for publication April 5, 2019.Author contributions: Study concept and design: He, Hu, Shao, L.P. Xia. Data acquisition, analysis, or interpretation: He, Hu, Wang, Jiang, Lu, Zhang, Ding, Shao, L.P. Xia. Statistical analysis: He, Hu, Wang, L.P. Xia. Acquisition of funding: X.J. Xia, Shao, L.P. Xia. Administrative, technical, or material support: Rong, L. Yang, Xie, Y.Z. Yang, Qiu, Lu. Supervision: X.J. Xia, Shao, L.P. Xia. Drafting of manuscript: He, X.J. Xia, L.P. Xia. Critical revisions: all authors.Disclosures: The authors have disclosed that they have not received any financial considerations from any person or organization to support the preparation, analysis, results, or discussion of this article. The data in the study were deposited in the Research Data Deposit system of Sun Yat-sen University Cancer Center [RDDA2019001007].Funding: This work was supported by grants from the Natural Science Foundation of Guangdong, China (2015A030313010), Science and Technology Program of Guangzhou, China (1563000305), the Guangdong Innovative and Entrepreneurial Research Team Program (2016ZT06S638), the Science and Technology project of Guangdong Province (2017A020215031), and National Natural Science Foundation of China (81773051, 81272641, and 81572409).Correspondence: Liang-Ping Xia, MD, and Jian-Yong Shao, MD, Sun Yat-sen University Cancer Center, 651 Dongfeng East Road, Guangzhou, Guangdong 510060, P. R. China. Email: xialiangping@163.com; shaojy@sysucc.org.cnView associated content

Supplementary Materials

  • View in gallery

    Flowchart of patient selection.

    Abbreviations: dMMR, deficient mismatch repair; pMMR, proficient mismatch repair.

  • View in gallery

    Circos diagrams of (A) MMR status between primary tumors and metastatic tumors and (B) MSI status of metastatic tumors in patients with MSI-H primary tumors.

    Abbreviations: dMMR, deficient mismatch repair; MMR, mismatch repair; MSI, microsatellite instability; MSI-H, microsatellite instability-high; MSS, microsatellite stable; MT, metastatic tumor; pMMR, proficient mismatch repair; PT, primary tumor.

  • View in gallery

    Comparison of tumor mutational burden between primary and matched metastatic tumors in patients with (A) MSI-high primary tumors and MSS metastatic tumors and (B) MSI-high primary tumors and MSI-high metastatic tumors.

    Abbreviations: Mb, megabase; MSI, microsatellite instability; MSS, microsatellite stable; Muts, mutations.

  • View in gallery

    Comparison of lymphocyte infiltration between primary and matched metastatic tumors in patients with MSI-high primary tumors and MSS metastatic tumors (hematoxylin-eosin, original magnification ×100). These patient numbers correspond to those in Table 3.

    Abbreviations: MSI, microsatellite instability; MSS, microsatellite stable.

  • View in gallery

    Comparison of lymphocyte infiltration between primary and matched metastatic tumors in 3 representative patients with MSI-high primary tumors and MSI-high metastatic tumors (hematoxylin-eosin, original magnification ×100).

    Abbreviation: MSI, microsatellite instability.

  • View in gallery

    Comparison of overall survival in patients with MSI-H primary tumors with MSS versus MSI-H metastatic tumors.

    Abbreviations: MSI-H, microsatellite instability-high; MSS, microsatellite stable.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1483 1483 93
PDF Downloads 478 478 31
EPUB Downloads 0 0 0