Clinical Significance and Prognostic Relevance of Microsatellite Instability in Sporadic Colorectal Cancer Patients

Microsatellite instability (MSI) is a marker of the replication error phenotype. It is caused by impaired DNA mismatch repair processes (MMR), resulting in ineffectiveness of the mechanisms responsible for the DNA replication precision and postreplicative DNA repair. MSI underlies the pathogenesis of 10%–20% of colorectal cancer (CRC) cases. The data about the potential value of MMR status as a predictive factor for 5-fluorouracil (FU)-based chemotherapy remain unclear. According to National Comprehensive Cancer Network updated guidelines, MSI testing is recommended for all patients with stage II CRC because patients with MSI-H (high-frequency MSI) tumour may have a good prognosis and obtain no benefit from 5-FU-based adjuvant chemotherapy. The significance of the MSI status as a predictive factor for patients with metastatic disease was not confirmed. The association between the MSI status and the efficacy of the therapy based on anti-programmed death-1 receptor inhibitors requires further studies.


Introduction
Colorectal cancer (CRC) is a heterogeneous disease with significant differences between clinical presentation, prognosis and individual treatment response, even at the same pathological stage. CRC development is associated with various driver mutations, genetic and epigenetic tumour signatures. Epidemiological studies underline possible etiological differences according to molecular markers. All the above-mentioned factors are very important to make individually tailored treatment decisions.
Adjuvant chemotherapy based on 5-fluorouracil (5-FU) is a standard treatment for patients with node-positive CRC who underwent curative surgery. However, there is a significant group of overtreated patients who could be cured by surgery alone without adjuvant chemotherapy (about 40%) or have chemotherapy-resistant disease [1,2]. The reported five-year survival of patients with stage II CRC ranges from 75% to 87.5% [3][4][5][6] and only 3.5%-5% of this group benefits from adjuvant chemotherapy [6][7][8]. Among patients with CRC of Dukes' stage C undergoing surgery, 5-FU-based adjuvant chemotherapy improves overall survival (OS) by 10%-12% [2]. did not co-exist with BRAF V600E mutations (p < 0.01), which suggests more frequent occurrence of the Lynch syndrome in the subgroup of patients with the early onset of the disease [37]. MSI tumours are characterized by a highly upregulated expression of various immunological checkpoints that are subjects of many current clinical trials as possible therapeutic targets, i.e., PD-1, PD-L1, CTLA-4, LAG-3 or IDO [38][39][40][41][42]. The mentioned immune inhibitory signals prevent elimination of neoplastic cells by counterbalancing the active immune microenvironment of the MSI tumour [38,43]. The MSI status may be a predictive marker for immuno-modulating agents.

Microsatellite Instability (MSI) Status Assessment
According to National Comprehensive Cancer Network (NCCN) updated 2016 guidelines, screening for the Lynch syndrome is recommended for all patients with CRC diagnosed by the age of 70 years old and regardless of age if Bethesda guidelines are met [44]. Moreover, MMR or MSI testing is recommended by the NCCN for all patients with stage II CRC, based on the results obtained by Sargent et al. suggesting that patients with dMMR tumour receiving FU-based therapy may receive no benefit from adjuvant systemic therapy [7,44].
The guidelines do not refer to the assessment of MSI/MMR in patients with metastatic disease. The updated guidelines on molecular markers for CRC are being developed jointly by the American Society for Clinical Pathology, the College of American Pathologists, American Society of Clinical Oncology and the Association for Molecular Pathology and are not yet available [45].
The MSI status can be assessed by the use of reference National Cancer Institute (NCI) panel of five microsatellite markers, including two mononucleotide repeats (BAT26 and A4725) and three dinucleotide repeats (D5S346, D2S123, and D17S250). The instability of two or more of the five markers was defined by NCI as high-frequency MSI (MSI-H), whereas instability of only one marker characterizes low-frequency MSI (MSI-L) [46].
An alternative method of MMR status evaluation is the analysis of MMR gene protein products (MLH1, MSH2, MSH6 and PMS2) using the immunohistochemical (IHC) staining [32]. It is an inexpensive and increasingly available technique with high sensitivity (80%-95%) and specificity (up to 100% in most reports) [47][48][49][50]. IHC assessment of MLH1 and MSH2 protein expression has also been proven to have a high correlation degree with the MSI status estimated by polymerase chain reaction (PCR) [47].

MSI Status and Colorectal Cancer (CRC) Prognosis
Numerous studies established the value of the MSI status as a prognostic factor in all stages of CRC regardless of age [37,[51][52][53]. The results of a meta-analysis including 7642 patients indicated that MSI tumours corresponded with significantly improved prognosis compared to MSS CRCs (OS associated with MSI: hazard ratio (HR) 0.65 (95% confidence interval (CI): 0.59-0.71)) [53].
There are reports suggesting a reduced rate of lymph node involvement [34,[54][55][56][57][58] and distant metastases in MSI tumours compared with MSS tumours [57][58][59][60]. However, MacQuarie et al. noted no differences in the number of all lymph nodes or the count of negative lymph nodes in patients with MSI and MSS tumours in stage III [61].
Mohan et al., in a single centre study including 1250 CRC patients, noted improved disease-free survival (DFS) in patients with MSI CRCs in stage I and II disease in comparison with MSS tumours. However, in CRC patients in stage III disease, MSI tumours were characterised by more prominent lymphovascular and perineural invasion and the DFS in this subgroup was shorter than in MSS patients in stage III [57].
On the other hand, the results obtained by Venderbosch et al. suggest an association between MSI and reduced OS in patients with advanced CRC [31]. The authors performed a pooled analysis of 4 phase III studies in the first-line treatment of metastatic CRC focusing on MMR and BRAF status and suggesting that the worse prognosis in dMMR mCRCs may be driven by the BRAF mutation. The study included 3063 patients with a low prevalence of dMMR (5%) and BRAF mutation (8.2%) in the whole studied population, however in patients with dMMR tumours, impaired BRAF was noted more often than in proficient MMR (pMMR) tumours (34.6% vs. 6.8% respectively, p < 0.001). Patients with MSI had significantly reduced progression-free survival (PFS; HR 1.33 (95% CI: 1.12-1.57)) and OS (HR 1.35 (95% CI: 1.13-1.61)). Interestingly, the authors observed decreased OS and PFS in mCRC patients with pMMR tumour and BRAF mutation while in the dMMR subgroup no association between BRAF status and survival was proven. On the other hand, survival effect in patients with BRAF mutation did not differ between pMMR and dMMR subgroup [31].

MSI Status and 5-Fluorouracil (FU) Chemotherapy Response
5-FU is an analogue of uracil and its cytotoxic activity is the cause of the misincorporation of fluoronucleotides into nucleic acids by the active metabolites of 5-FU and in the inhibition of thymidylate synthase-an enzyme taking part in the synthesis of nucleotides [62]. 5-FU cytotoxic effect seems to be determined by several enzymatic mechanisms belonging to DNA repair systems, resulting in removing of 5-FU from DNA: the base excision repair (BER; by excision of 5-FU or uracil by uracil DNA glycosylases) [63] and MMR (by removing of mismatched nucleotides) [64,65]. 5-FU and 5-fluorouracil-2 -deoxyuridine-5 -triphosphate excision from DNA may exacerbate the cytotoxicity of the drug by increasing its intracellular concentration [64].
The results of in vitro studies suggested the ineffectiveness of 5-FU in human colorectal cell lines with MSI [66][67][68]. The findings of the early clinical trials did not confirm these observations, suggesting a possible better survival in the group of patients with MSI tumours receiving 5-FU compared to the MSS CRC group [52,[69][70][71]. However, it is explained by some authors as potential enhancement of better prognosis noted frequently in patients with MSI CRC [1]. Other available data suggest a potential similar benefit from 5-FU regardless of the presence of MSI or MSS [1,29,72]. On the other hand, several authors indicated a lack of benefit from 5-FU-based adjuvant therapy in patients with MSI CRC [7,[73][74][75][76][77][78][79], including the suggestion that the presence of MSI may be associated with shorter survival [7,75]. The data about the potential value of the MMR status as a predictive factor for 5-FU-based adjuvant chemotherapy are contradictory.
Guastadisegni et al. [64] conducted a meta-analysis in order to assess the clinical significance of the MSI status for CRC patients receiving 5-FU-based chemotherapy. The study included 12,782 patients with various tumour stages examined in 31 studies reporting survival data. Authors confirmed the association between MSI and improved outcome in terms of DFS (summary OR of 0.58 (95% CI: 0.47-0.72), p < 0.0001) and OS (summary OR of 0.6 (95% CI: 0.53-0.69), p < 0.0001) irrespective of the tumour stage. The analysis of the potential survival effect of 5-FU adjuvant chemotherapy depending on the MSI status was based on the evidence from seven studies in which survival data were stratified separately for MSI-H and MSS patients. MSS patients have been shown to obtain benefit from 5-FU-based chemotherapy, whereas the survival effect in the MSI group did not reach statistical significance. The statistical analysis showed the high inter-study heterogeneity in the MSI subgroup.

Predictive Value of MSI Status in II and III CRC
According to the NCCN guidelines, MSI testing is recommended for all patients with stage II CRC, which is based on the assumption that these individuals obtain no benefit from 5-FU-based adjuvant therapy [44] based on the results obtained by Sargent et al. [7]. Unlike individuals with MSS tumours, patients with stage II and III MSI CRC did not benefit from 5-FU chemotherapy compared to patients undergoing surgical treatment alone (DFS: HR 1.10 (95% CI: 0.42-2.91)). The data were pooled with further 570 cases from a previous study [75]. In the combined dataset, in patients with MSI CRC in stage II, adjuvant chemotherapy was associated with decreased OS (HR 2.95 (95% CI: 1.02-8.54)).
Taking into account that the NCCN recommendations are based on one study and considering the finding of the two large meta-analyses by Guastadisegni

Predictive Value of MSI Status in Metastatic CRC
There are insufficient data about the significance of MSI in the mCRC patients. Most of the available studies were conducted without randomization, based on small groups and without clear discrimination of sporadic and hereditary CRCs [35]. It may be an effect of a decreased metastatic potential of MSI tumours [57][58][59] resulting in a reduced incidence of MSI CRC among patients with advanced disease [58].
A meta-analysis undertaken by Des Guetz et al. [82] focused on potential predictive significance of MSI assessment for patients with metastatic disease. The analysis included 6 studies representing 964 patients receiving 5-FU-based chemotherapy (30%) or combinations of 5-FU or capecitabine with oxaliplatin and/or irinotecan (70%). While OS and RFS data were unavailable, the parameter used to assess benefit of chemotherapy was response rate (RR) ratio according to the RECIST criteria. No significant difference in effect of treatment in terms of RR ratio was observed for the MSI-H compared with the MSS patients (RR ratio 0.82 (95% CI: 0.65-1.03; p = 0.09)). The possible cause may be a considerable statistically significant heterogeneity probably arising from different chemotherapy regiments and an unequal distribution of patients in the studies.
A recent phase II study by Le et al. [43] evaluated the efficacy of anti-programmed death 1 (PD-1) receptor inhibitor pembrolizumab depending on the MSI status. The analysed population consisted of 41 patients with progressive metastatic disease, including 13 dMMR CRC, 25 pMMR CRC and 10 MMR-deficient non-colon cancer. The immune-related objective RR and 20-week-immune-related PFS rate were 40% (95% CI: 12%-74%) and 78% (95% CI: 40%-97%) respectively, for dMMR CRC and 0% (95% CI: 0%-20%) and 11% (95% CI: 1%-35%) for pMMR CRCs. The median PFS and OS in the pMMR CRC cohort were 2.2 and 5.0 months, respectively, while in the dMMR CRC group PFS and OS have not been reached yet. The results suggest that patients with dMMR tumours are more responsive to pembrolizumab than pMMR mCRC patients. However, the results need confirmation in further well-designed studies in a larger population.
On the other hand, Kim et al. [60] analysed the outcomes in patients with colorectal cancer recurrence depending on the MSI status. The study group included 2940 patients with stage I-II CRC after curative resection. MSI-H was observed in 8.9%. Although MSI-H was associated with significantly longer OS in patients without recurrence compared to MSI-L or MSS, among individuals with recurrence MSI-H CRCs was associated with worse OS from diagnosis to death (HR: 1.363 (95% CI: 1.022-1.819), p = 0.035) and OS from recurrence to death (HR: 2.667 (95% CI: 1.541-4.616), p < 0.001). The most frequent treatment strategy after recurrence in patients with MSI-H tumours was palliative chemotherapy without resection (74.0% compared to 44.1% in MSI-L/MSS patients, p = 0.01). Patients in this group underwent reoperation less often compared to MSI-L/MSS patients with recurrence (20.0% vs. 51.6%, p = 0.01). However, there was a major disproportion between the number of MSI-H and MSI-L/MSS patients with CRC recurrence (20 vs. 374 patients, respectively).
The main characteristic of available meta-analysis on prognostic value of microsatellite instability (MSI) are summarised in Table 1.

Conclusions
Colorectal cancer is a heterogeneous disease with varied biological and genetic features. The heterogeneity of CRC leads to different prognosis and clinical management. Treatment decisions should be based in each case individually on various clinical factors, including driver mutations, and genetic and epigenetic signatures identified in the tumour.
MSI is a promising factor that could potentially enable predicting patient response to the treatment based on the reports postulating no benefit from adjuvant 5-FU chemotherapy in patients with stage II and III CRC if MSI was detected. MSI may also be associated with favourable prognosis and a better overall survival in patients with advanced disease. However, the role of the MSI status as a predictive marker remains controversial.
In our opinion, the individually tailored treatment decision and the consideration of more aggressive approach may be needed in the BRAF mutated tumours and alternative chemotherapy in the MSI tumours.
The preliminary studies on the potential role of MSI in targeted immunotherapy have been promising and further investigation on this topic is warranted. Author Contributions: Angelika Copija essentially contributed to the theme and concept of the work, gathering and interpretation of the literature, and preparing the manuscript; Dariusz Waniczek gathered and interpreted the literature, and prepared the manuscript; Andrzej Witkoś gathered and interpreted the literature, and prepared the manuscript; Katarzyna Walkiewicz gathered and interpreted the literature, and prepared the manuscript; and Ewa Nowakowska-Zajdel essentially contributed to the theme and concept of the work, analysed the literature, and revised the article content.

Conflicts of Interest:
The authors declare no conflict of interest.