Most colitis associated carcinomas lack expression of LGR5: implications for unique pathways of carcinogenesis compared to sporadic colorectal carcinoma

Leucine-rich repeat-containing G-protein-coupled receptor 5 (LGR5), a component of the Wnt receptor complex, is thought to lineage label gastric and intestinal stem cells. LGR5 expression is increased in colorectal carcinoma (CRC) compared to normal tissue. Colitis associated colorectal adenocarcinoma (CAC) often shows distinct morphologic and molecular phenotypes compared to sporadic cases. However, the expression prole of LGR5, and by extension the potential role of an intestinal stem cell phenotype, has not been well described in a series of human CAC.


Background
In ammatory bowel disease (IBD) is a chronic relapsing-remitting disorder of the gastrointestinal tract, and for patients with long-standing colonic in ammation, colorectal adenocarcinoma (CRC) is a recognized complication, leading to routine periodic surveillance of patients with colitis for dysplasia and carcinoma. One meta-analysis estimated that the risk of colitis associated carcinoma (CAC) in patients with IBD is 2% by 10 years after initial diagnosis, and increases to 8% at 20 years and 18% at 30 years after colitis onset. [1] However more recent gures are lower than this, likely the results of better treatment reducing the risk. [2,3] Leucine-rich repeat-containing G-protein-coupled receptor 5 (LGR5), a wingless-type mouse mammary tumor virus integration site family (Wnt) target gene that functions as a receptor for Wnt agonist Rspondins (RSPOs). It was identi ed as a selective marker of crypt base columnar cells, and these LGR5positive crypt base columnar cells were demonstrated to be self-renewing, multipotent adult intestinal stem cells. [4] Increased LGR5 expression in human sporadic colorectal adenomas and cancers [5][6][7][8] and in murine models [9,10] have been well described, however, the LGR5 expression pro le has not been described in a series of CAC in human.
Recent studies have highlighted genetic, morphologic, and immunohistochemical differences between CAC and sporadic CRC, and suggested that chronic in ammation may lead to unique genomic changes that increase the risk for CAC. [11,12] Carcinogenic mutations probably rst start accumulating in the long-lived stem cell lineages of the crypt, as the lifespan of non-stem cells is too short for them to acquire the necessary mutations before being shed.
To our knowledge, a comparison of LGR5 expression in CAC and sporadic CRC has not been performed.
Here, a series of CAC cases was compared with sporadic CRC for morphologic features and the expression of LGR5.

Methods
Study approval was obtained from the Research Ethics Board at Sinai Health system. Patients who underwent neoadjuvant therapy were excluded. 10 cases of surgically resected primary CRC from patients without in ammatory bowel disease or any known hereditary cancer syndrome, hence forth termed sporadic CRC were identi ed as a control group. H&E sections were reviewed at a multiheaded microscope by four gastrointestinal pathologists (MI, HO, RR, and JC) and each invasive carcinoma was subclassi ed by consensus into one of ve morphologic subtypes: conventional, mucinous, serrated, low grade tubuloglandular (LGTG) and others. [13] One representative para n block of tumor was selected in each case for RNA in situ hybridization (ISH).
Detection of LGR5 mRNA was performed using the RNAscope® kit (Advanced Cell Diagnostics, Hayward, CA, USA) according to the manufacturer's instructions using unstained sample tissue slides. Brie y, tissue sections were pretreated by heating and protease application prior to hybridization with an LGR5-speci c probe. A detailed procedure has been described in an earlier publication. [14] Positive staining was indicated by brown punctate dots in the nucleus and/or cytoplasm. The expression level of LGR5 was quanti ed according to the ve-grade scoring system previously described [8] : 0 = no staining or less than one dot per cell; 1 = 1 to 3 dots per cell; 2 = 4 to 10 dots per cell and no or very few dot clusters; 3 = >10 dots per cell and <10% positive cells overall; and 4 = >10 dots per cell and >10% positive cells with dot clusters. For a binary analysis, LGR5 status was considered positive if the ISH score was >2. (Fig1) [15] Two of the authors (MI and JC) reviewed the RNA-ISH stains at a multiheaded microscope and reached a consensus score for each case.

Statistics:
Chi-square test or Fisher exact tests were used to characterize the relationship between categorical variables. Kruskal-Wallis chi-squared test and Mann-Whitney U-test were used for comparisons between LGR5 expression level scores. Staining scores are non-parametric and are thus expressed as a median score with the interquartile range. Differences were considered to be signi cant at p<0.05. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria).

Results
The 29 patients with CAC included 18 with ulcerative colitis and 11 with Crohn's disease. 1 patient with ulcerative colitis had synchronous carcinomas for a total of 30 tumors evaluated. A summary of the clinicopathologic features of the CAC cohort and sporadic controls is presented in Table 1. The median age at time of resection for patients with CAC was 54 years (range 26 -76), signi cantly younger than sporadic cases (median age 67 years; range 40 -95) (p = 0.017). CACs had a wider spectrum of tumor morphologies compared to sporadic CRCs. Only twelve of 30 (40%) CACs had conventional morphology compared to 80% of sporadic cases. None of sporadic cases were classi ed as LGTG. The one CAC case subclassi ed as "other" showed fetal enteric and hepatoid differentiation.

Discussion
Here, we evaluated the LGR5 expression pro le of 30 CAC cases and 10 sporadic CRC cases and identi ed that LGR5 was less frequently expressed in CACs than in sporadic CRCs. Further, we found CAC with non-conventional morphology showed signi cantly lower LGR5 expression levels than conventional subtype of CAC.
LGR5 has been considered one of the most reliable crypt stem cell markers of CRCs, particularly those that arise through the APC-mutant pathway of tumorigenesis. [9] LGR5 is reportedly overexpressed in human colorectal adenomas and cancers by immunohistochemistry [6,16,17] and ISH, [15] and Martin et al showed that the mRNA levels of LGR5 expressed in human CRCs is 10-fold expanded than normal intestinal crypt. [18] Jang et al recently showed that LGR5 mRNA positivity is observed in 68% of 788 human CRCs and suggested positive correlation with the chromosomal-instability pathway (characterized by left-sided location and nuclear β-catenin expression; representing an abnormal Wnt signal activation) and negative correlation with MSI, CIMP-high, and BRAF mutations. [15] In accordance with these ndings, we found positive LGR5 expression in 90% of our sporadic CRC control group. However, only 30% of CAC were LGR5 positive. The latter nding is in agreement with the previous study by Yasuda et al reporting signi cantly less expression of reliable intestinal stem cell markers such as CD133, [19] OCT4 [20] and NANOG [21] in CAC than sporadic CRC. [22] The molecular pathogenesis of CAC is different than that of sporadic CRCs with genomic changes that appear directly linked to the effects of ongoing in ammation and repeated mucosal injury in the IBD. [11,12] In keeping with this, we recently reported CAC tends to show non-conventional tumor morphology and loose intestinal markers such as SATB2 and CDX2, and show aberrant gastric mucin expression. [23] Ishibashi et al described the expression pro le of Atonal homolog-1 (ATOH1), a master transcription factor of the secretory lineage of intestinal epithelial cells, which retain their potential to revert to intestinal stem cells [24][25][26] in colitic mucosa. They identi ed both ATOH1 and LGR5 double positive tumor cells as well as LGR5 single positive tumor cells, suggesting that colitic tumors are mosaic and consisted of a heterogenous population of tumor stem cell. [27] While the exact mechanism of decreased LGR5 expression in CAC is uncertain, these data demonstrate that CACs have a unique expression pro le of intestinal stem cell markers.
We found that despite only 40% of CACs in our cohort showing a conventional morphology, 89% of LGR5 positive CAC were conventional in appearance. These also tended to be located in the rectum. This raises the possibility that the minority of CACs in our cohort that showed conventional morphology might have arisen through the conventional APC pathway. However, as the majority of tumors had non-conventional morphology and were also LGR5 negative our data provide support to other recent studies demonstrating that most CAC cases do not arise through the conventional APC pathway. [12,28] Although it did not reach statistical signi cance, conventional type CACs trended toward a lower level of LGR5 expression compared to the control group of sporadic CRCs. Thus, even in tumors arising through more conventional molecular pathways in the setting of colitis, LGR5 expression might be affected by ongoing in ammation and repeated mucosal injury. This hypothesis is supported by other studies showing that Lgr5+ stem cells are highly sensitive to epithelial injury induced by radiation or colitis. [27,29,30] Our results in human cancers do not entirely align with previous data from animal models of colitis. Kim et al reported increased Lgr5 expression in an AOM / DSS -colitis mouse model. They showed that Lgr5 expression is gradually increased as tumors developed with repeated colitis and reported all dysplastic lesions and cancers showed high Lgr5 expression. [10] This discrepancy could be due to differences between human CAC and the murine model. Interestingly, the gure showing tumor morphology in their report shows a more conventional morphology, implying that the murine model may simulate the carcinogenesis pathway in the conventional type of human CAC.
Data regarding any potential prognostic signi cance of LGR5 expression in sporadic CRC are mixed.
Recent studies using RNA-ISH for LGR5 evaluation reported that LGR5 expression is an independent predictor of favorable outcome in CRCs. [15,31]. On the contrary, a previous meta-analyses of immunohistochemical studies of LGR5 expression showed that high LGR5 expression is associated with shorter overall survival and disease free survival. [32,33] Kazama et al described the immunohistochemical expression pro le of LGR5 in ulcerative colitis cases and showed increased LGR5 expression in dysplasia and CAC.
[34] This result is contrary to our present study, but may be due to the different methods of LGR5 detection, as the reliability of antibodies against LGR5 remains uncertain, [35,36] and most recent studies have used ISH to detect LGR5 expression.
One limitation of our cohort is a limited number of cases and the unavailability of clinical follow up data. Additional study using a larger cohort will be needed to determine the relationship between LGR5 expression and clinical prognosis in human CACs.

Conclusion
we demonstrate that, unlike sporadic CRC, LGR5 expression is less frequent in CAC, particularly those with non-conventional morphology. It is more commonly expressed in CAC with conventional morphology and in tumors located in the rectum, indicating that a limited subset of CACs may arise through the conventional APC pathway. While the precise mechanisms for the low frequency of LGR5 expression in CAC remains unde ned, the wider spectrum of tumor morphology in CAC may be associated with absence of a LGR5-expressing intestinal stem cell phenotype.

Abbreviations
LGR5: leucine-rich repeat-containing G-protein-coupled receptor 5; CRC: colorectal carcinoma; CAC: Colitis associated colorectal adenocarcinoma; LGTG: low grade tubuloglandular Declarations Ethics approval and consent to participate Study approval was obtained from the Research Ethics Board at Sinai Health System. The investigation was conducted in compliance with the Helsinki Declaration.

Consent for publication
Not applicable.

Availability of data and materials
All data generated and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests" in this section.

Funding
The authors did not receive any funding

Author Contributions
Each author has participated su ciently in the work to take public responsibility for appropriate portions of the content: MI contributed to the conception and design of the study. MI, HO and JC analyzed histopathological features and drafted the manuscript. RR provided valuable advice and suggestions as a histopathologic consultant. TN and TU contributed to part of the RNA-ISH study. All authors read and approved the nal manuscript.    Figure 1 LGR5 expression level was quanti ed according to the ve-grade scoring system: score 2; a few dots per cell and no or very few dot; LGR5 negative (a), score 3; >10 dots per cell and <10% positive cells overall; LGR5 positive, (b, c), and score 4; >10 dots per cell and >10% positive cells; LGR5 positive (d). Expression level of LGR5 in control sporadic colorectal adenocarcinoma, conventional type colitis associated adenocarcinoma and non-conventional type colitis associated adenocarcinoma. *P = 0.034, ***P < 0.001