Colorectal cancer (CRC), as the most common type of gastrointestinal malignant tumors, ranks third in incidence and second in mortality globally (1). In the past thirty years, the incidence and mortality of CRC in China have been increasing year by year, and currently account for 18.6% and 20.1% of the cases worldwide, respectively (2, 3). According to statistics, rectal cancer comprises approximately 30% of all CRCs. Rectal cancer is predominately a disease of older individuals, with a marked increase in incidence between 40 and 50 years of age, with increasing risk as a function of advancing age (4). Because the early symptoms of CRC are not obvious, many patients at the time of diagnosis are in the locally advanced stage. Due to the complex anatomy of the rectum and the high local recurrence rate, the treatment and anal preservation in locally advanced rectal cancer (LARC) patients is facing considerable challenges. Currently, the standard treatment for LARC patients is a combination of preoperative neoadjuvant chemoradiotherapy, total mesorectal excision (TME), and postoperative adjuvant chemotherapy (5). Preoperative treatment has been developed to such an extent that the pathological complete response (pCR) rate has reached 25% at the time of surgery (5). However, it is obviously correlated with the occurrence of adverse events, including intestinal dysfunction, leukopenia, sexual dysfunction etc. (6, 7). In recent years, a new treatment combination has emerged for patients with LARC, that is, preoperative transcatheter rectal arterial chemoembolization (TRACE) combined with concurrent chemoradiotherapy followed by radical surgery. Clinical studies have confirmed that this method can significantly improve the pathological response rate of LARC, and the pCR rate can reach about 30% (8). pCR, as an efficient indicator to neoadjuvant therapy, is also an independent predictor of disease-free survival (DFS) and overall survival (OS) in patients with LARC (9), may provide these patients with an alternative to some non-surgical treatments.

According to the recent report, the link between the mismatch repair (MMR) genes and cancer has attracted much attention (10). MMR is an evolutionary conserved mechanism that corrects mutations during DNA replication and damage and plays a critical role in maintaining genomic stability (11, 12). When the MMR gene is mutated, the expression of MMR protein is decreased, not expressed or truncated. Meanwhile the deletion and insertion occurring in the process of DNA replication cannot be corrected, resulting in the instability of genomic DNA. There are as many as 12 MMR genes in humans, among which the four most important genes are MLH1, PMS2, MSH2 and MSH6. These four proteins are the main members of the MMR family, accounting for more than 90% of mutations. The loss of MMR protein expression can lead to the accumulation of mismatch during DNA replication, resulting in the occurrence of microsatellite instability high (MSI-H). About 15% of CRCs are caused by MSI pathway (13). It has been reported that some patients with metastatic CRC are closely associated with MMR protein loss (14, 15). Presently, the significance of deficient MMR (dMMR) protein in the risk assessment, diagnosis and prognosis of CRC has been reported. It is associated with 85% of hereditary non-polyposis CRC (Lynch syndrome) and 15% of sporadic CRC, which has important clinical significance. In 2015, immunotherapy was shown to provide significant clinical benefit in metastatic CRC patients with dMMR/MSI-H status, whereas no response was observed in patients with proficient mismatch repair (pMMR)/microsatellite stability (MSS) status (16–18).

Based on the significance of MMR protein, we hypothesized that preoperative TRACE would be more effective in LARC patients with MMR protein deficiency. To test this hypothesis, we designed and conducted this study. It evaluated the efficacy and safety of preoperative TRACE combined with concurrent chemoradiotherapy in LARC patients with pMMR/dMMR genes. By immunohistochemical method to detect the integrity of the MMR protein in tumor tissues, we can predict the effect of tumor regression according to the lack of MMR protein. This maybe provide a new guiding therapeutic method for patients with LARC whether to continue surgical treatment after neoadjuvant therapy.

Our study is the first to propose that MMR can predict pathological responses. In this study involving MMR proteins, patients with LARC were treated with neoadjuvant TRACE combined with concurrent chemoradiotherapy. Only four weeks after treatment, biopsies were taken by endoscopy or pathological examination of surgically resected tissue. Some patients can achieve pCR, while the pCR rate of dMMR patients and pMMR patients is 43.2% and 10%, respectively. Although there was no difference in DFS and OS between the two groups in our study, significant differences could be seen in pathological response. So, we put forward a new experimental study. After the neoadjuvant treatment of preoperative TRACE combined with concurrent chemoradiotherapy in patients with LARC, the MMR protein expression status can be used to guide whether it is necessary to continue surgical treatment. That is, under the guidance of MMR proteins, the pCR rate of patients may be improved, so that patients can choose to wait for treatment instead of radical resection. Moreover, the study provides preliminary evidence.

In NICHE’s study, pCR was achieved in all CRC patients with dMMR protein, 95% of which were primary cancers (≤ 10% of viable residual tumors in surgical specimens), including 12 (approximately 60%) pCR cases (23). One retrospective analysis has demonstrated that high rates of pCR can be achieved with checkpoint inhibition in pretreated dMMR metastatic CRC patients (24). In our study, although both groups received the same neoadjuvant therapy, rectal cancer patients in the dMMR group produced a higher pCR rate and were well tolerated without safety problems. Radiotherapy and surgery may have long-term effects on the bowel and bladder function and immune function of patients (6, 7, 25, 26). These adverse events have a significant impact on the daily life of patients, so if the MMR protein loss can be used to guide the need for surgery in patients with LARC, it may bring greater benefits to this population.

Studies have shown that MMR genes correlate with clinicopathological features of LARC and can be biomarkers that contribute to the prognosis of patients. Meanwhile, MMR gene is an important part of the DNA damage response pathway, which maintains the integrity of the genome and reduces primary mutations. It is mainly composed of a series of repair proteins that specifically repair DNA mismatches, including MLH1, PMS2, MSH2, MSH6, etc. (27). Inactivation of any of these genes can lead to the loss of function of MMR proteins, which may be related to the occurrence and development of endometrial cancer, CRC and ovarian cancer. The relationship between the MMR genotype and pCR rates has increased the interest in options of rectal cancer treatment. As we all know, neoadjuvant immunotherapy has been tested in several solid tumors (28–31), and the level of activity in other tumor types is far less than the degree of activity we have observed in LARC patients with dMMR (32). Therefore, dMMR protein tumors respond better to immunotherapy (33). At the same time, we observed that patients with MMR protein deficiency in our study were more sensitive to TRACE and were prone to achieve pCR.

Neoadjuvant chemoradiotherapy has been a standard treatment for LARC. Traditionally, radical surgery is scheduled 6 to 10 weeks after the end of a long course of chemoradiotherapy. Whereas, the tumor was resected 4 weeks after TRACE. All patients received adjuvant chemotherapy using mFOLFOX6 or CapeOx regimens for 4–6 months. Study shows that regardless MSI, 5-fluoruracile (5-FU) have been the backbone in CRC treatment, being combined with others antitumor drugs such as platinum alkylating agents (34). Therefore, we also chose 5-FU as one of the main anti-tumor drugs after surgery. Studies have shown that both proficient and deficient CSC/MMR showed high 5-FU chemoresistance regardless to MMR status (35). Previous studies have shown that adding neoadjuvant chemoradiotherapy can reduce the tumor stage, reduce distant metastasis, and improve pCR rates in LARC patients; however, there was no difference in DFS or OS compared with the group of adjuvant only (36, 37). Also in our study, there were no statistically significant differences in DFS and OS between the pMMR and dMMR groups. In another study, the overall pCR rate was 25.3% in patients treated with preoperative chemotherapy alone and those treated with both preoperative and postoperative chemotherapy (38). Our study added the MMR proteins on the basis of neoadjuvant therapy. Meanwhile, our research is the first to show that MMR proteins can predict pathological responses. The addition of TRACE before surgery provides better pathological remission rate of more LARC patients versus standard treatment with a similar safety profile (8).In other words, preoperative TRACE provided better pathological remission rate versus standard treatment and did not increase the risk of perioperative and postoperative complications. In particular, preoperative TRACE with concurrent chemoradiotherapy has potential advantages in preventing distant metastasis of LARC. Besides, it also provides an opportunity for the implementation of non-surgical, watch-and-wait strategies to improve quality of life. However, compared with conventional neoadjuvant chemotherapy treatment modalities, TRACE has a higher cost, and patients also have to bear the risk of intestinal necrosis caused by surgery. Based on this study, if the MMR gene can be combined with the treatment mode of TRACE, preoperative TRACE combined with chemoradiotherapy may be more accurate in patients with LARC with pMMR or dMMR and better pathological response.

One question worth considering is why rectal tumors deficient in MMR proteins respond better. Recent evidence suggests that a combination regimen with a single dose of a programmed death 1 blocking agent or cytotoxic T lymphocyte-associated antigen-4 inhibitors is highly effective against advanced MSI-H/dMMR tumors but not against pMMR tumors (16, 39). The exact mechanisms by which MSI-H/dMMR LARC patients develop resistance to immunotherapy are unknown but may be explained by the biodiversity of the host immune system and tumor biology. Drug resistance associated with intra-tumor heterogeneity in MSI-H/dMMR status has been reported (40, 41). In LARC, most hypermutated tumors are MSI-H/dMMR.

MMR proteins may be another potential predictive biomarker. Many preoperative and perioperative treatment strategies have achieved good results in gastrointestinal tumors. Therefore, we propose MMR proteins as a potentially effective tool to guide therapeutic strategies in locally advanced MSI/dMMR and MSS/PMMR CRC populations. Further research should focus on developing novel imaging biomarkers in combination with molecular markers to provide a more accurate assessment, or even prediction, of response to neoadjuvant therapy. Given that our study was an exploratory study with a small sample size, we recommend that the findings be interpreted with caution. Furthermore, patients with dMMR seem more likely to benefit from our neoadjuvant study strategy than patients with pMMR.

Although our results are promising, especially the pCR rate of 42% in the dMMR group, the sample size of this study was small and only represents the results of a single institution. Another significant limitation was that larger-scaled RCTs should have been performed. More importantly, if colonoscopy indicates good pathological regression after neoadjuvant therapy, some patients choose to refuse surgical treatment, and their long-term prognosis cannot be compared. We should add some observation indicators in subsequent studies to follow up the long-term prognostic effect. If confirmed in subsequent large studies, the lack of MMR protein may provide guidance for neoadjuvant therapy in patients with LARC, which can make most patients avoid surgical resection, save the anus, and improve the quality of life. In our preliminary study, TRACE combined with concurrent chemoradiotherapy had a better pathological response rate and short-term efficacy (8, 42). These results lead us to believe positively that in the background of preoperative interventional chemoembolization, disease remission will account for a larger proportion. As the data become more complete and more mature, we envision that MMR proteins will be used in the context of preoperative interventional chemoembolization to evaluate other MMR protein-deficient tumors, such as gastric and prostate cancers. This may open up a new world of neoadjuvant chemotherapy for different types of tumor.

In conclusion, the prognosis can be predicted according to the absence of MMR proteins such as MLH1, MSH2, MSH6 and PMS2 in clinical treatment of rectal cancer. Then the MMR genotype was used to make the next targeted treatment plan for the patient, such as waiting treatment without surgery. All in all, we found that patients with MMR protein deficiency in our study were more sensitive to TRACE and were prone to achieve pCR. Further prospective studies are needed to verify the efficacy and safety of preoperative interventional chemoembolization in patients with LARC guided by MMR.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

Our research was a retrospective and single-center study, and approved by the Ethics Committee of Daping Hospital.

YG: writing-original draft preparation, and funding acquisition. HX: immunohistochemistry and reading of pathological sections. JL: data analysis and graphics. QC and GZ: formal analysis and investigation. WM: participation in the discussion, supervision, review, and editing. CL: conceptualization and funding acquisition; LB: conceptualization, funding acquisition, and supervision. All authors read and approved the final manuscript. All authors contributed to the article and approved the submitted version.