IMMUNOREACT 8: IMMUNE MARKERS AS PREDICTORS OF LOCAL TUMOR SPREAD IN PATIENTS UNDERGOING TRANSANAL EXCISION FOR RECTAL CANCER

Background: Local excision (LE) with transanal endoscopic microsurgery (TEM) is an established alternative to radical resection in select early rectal cancers and high-risk patients. Recent studies demonstrated adding neoadjuvant chemoradiotherapy (NACRT) before LE achieved organ preservation with similar short-term survival to radical resection for cancers up to T2N0. Study on long-term recurrence, survival, and wider application is warranted. We sought to evaluate oncologic and survival outcomes for T2 and T3 rectal cancers treated with NACRT followed by TEM. Our hypothesis was that NACRT with TEM can achieve optimal oncologic outcomes and long-term survival in T2 and T3 rectal cancers.

Methods: A prospective institutional cancer registry was reviewed for clinical stage T2 and T3 rectal adenocarcinoma patients who received NACRT then TEM for curative resection between 1997-2022. Demographics, clinical, and pathological outcomes were evaluated by univariate analysis. Kaplan-Meier analysis assessed recurrence and survival data. The main outcome measure was local recurrence rate (LR) after TEM. Secondary outcomes were disease-free survival (DFS), overall survival (OS), morbidity, and mortality.

Results: 137 cases- 81 T2 [59%] and 56 T3 [41%] were analyzed. Patients were 66.4% male (n=91) with a mean age of 67.7 (SD 12.1). Mean tumor distance from the anorectal ring was 3.2cm (SD 2.5). 88% (n=103) had a good response to NACRT (Tumor Regression Grade 1-2). R0 resection was obtained in 99.2% (n=136). The overall morbidity rate was 29.2% (n=40). There were 3 (2.2%) reoperations for wound dehiscence, intraperitoneal hemorrhage, and rectal bleeding. There was no 30-day mortality. Ten patients required total mesorectal excision (TME) after TEM, with 2 immediately for high risk disease. In these patients, the TEM functioned like a diagnostic full thickness bioipsy. Four patients required TME due to local recurrence of disease, 1 patient’s post-surgical pathology demonstrated ypT3Nx cancer, 1 developed significant stenosis, 1 elected to have an abdominoperineal resection, and 3 patients were risk averse and choose TEM to be conservative. After a median follow-up of 46.5 (IQR 15.7-94.6) months, 5 patients (3.6%) developed isolated LR (median time: 34 months) and 11 (8.0%) developed distant metastasis (median time: 25 months). The 3-year OS was 86.8% and DFS was 92.7%.

Conclusions: Using NACRT with TEM has excellent clinical and oncologic outcomes in locally advanced rectal cancers. Long-term outcomes found low rates of LR, low rate of patients that go on to TME, and survival comparable or superior to transabdominal resection rates. While controlled studies are needed, this organ-preserving option is safe, feasible, and requires consideration as an option for all patients in the shared decision-making process for clinically-staged T2/T3 rectal cancers.

Introduction: Anastomotic leak (AL) is a major complication in colorectal cancer (CRC) surgery, and is associated with increased mortality and morbidity. Recent evidence has suggested that the gut microbiota may be involved in the healing of colonic anastomoses in patients undergoing surgery for CRC. Data on the causal link between the pre-operative gut microbiota and AL in CRC patients remains scant. Our objective was to further characterize this link.

Methods: We collected fecal samples before surgery and mucosal samples during surgery in CRC patients undergoing a colorectal resection with anastomosis. The gut microbiota in these samples was analyzed using the Anchor 16 pipeline. The preoperative microbiota of patients with and without AL was transplanted into mice that then underwent a surgical colonic anastomosis. The fecal and mucosal microbiota of these mice was analyzed as well. Bacteria correlated with healing parameters, and showing different levels between patients with and without AL, were isolated, and supplemented to mice that were then subjected to colonic surgery. Their mechanism of action on key repair mechanisms was assessed in vitro.

Results: The pre-operative microbiota of patients with AL led to poor anastomotic healing in mice, to a poorly restored gut barrier function with higher bacterial translocation, and to a weakened anastomotic wound matrix with lower collagen and fibronectin. Differences in the composition of the gut microbiota of patients with and without AL were detected before surgery. These differences were transferable to mice by fecal microbiota transplantation, with 24 differentially abundant species. Two bacterial strains were strongly correlated with healing parameters, and whose role in intestinal health was unknown. These strains were isolated and labeled Pg kh35 and Ao kh33. When supplemented to mice, Pg Kh35 improved anastomotic healing, strengthened the wound matrix and restored the gut barrier, while supplementation with Ao kh33 had a deleterious effect. Most importantly, both were detected in the mucosa of mice and patients, after the administration of antibioprophylaxis and bowel preparation, and their levels were different between patients that later did or did not develop AL. Mechanistically, Pg kh35 exerted a beneficial effect by secreting a compound that upregulates the peroxisome proliferator-activated receptor gamma (PPAR-γ) in the colon, which promotes repair and alleviates inflammation, while Ao kh33 had the opposite effect.

Conclusion: The preoperative microbiota of patients with CRC is causally linked to the risk of developing AL. We unveiled how several mucosal bacteria, that are resistant to the preoperative bowel decontamination, may influence anastomotic healing. These findings pave the way toward clinical trials in which the gut microbiota may be modulated before surgery to improve outcomes.

Introduction:
Prior studies have demonstrated declining rates of surgery for both ulcerative colitis (UC) and Crohn’s disease (CD) with the advent of biologic therapies. Although current biologic agents cannot reverse existing fibrotic strictures, limited data suggest that early control of intestinal inflammation might prevent progression toward stricturing disease. Given the unclear overall impact of biologics on stricturing disease, we hypothesized and evaluated whether nationwide rates of surgery for obstruction in the setting of CD decreased since the introduction of infliximab for CD in 1998.

Methods:
The National Inpatient Sample from 1998-2018 was queried to identify patients with CD and those who underwent bowel resection surgery using ICD-9 and ICD-10 codes in all diagnosis and procedure positions. Those who underwent surgery were further stratified into surgery for indication of obstruction. Both emergency and elective admissions for surgery were included for small bowel and colon resections. Longitudinal trends were plotted and tested for trends while adjusting for age, sex, race, comorbidity, primary payer, income quartile, hospital bed size, and hospital type.

Results:
A total of 291,692 patients with CD were admitted for surgery out of 3,472,436 total hospitalizations for CD. Those admitted for surgery were younger (44.25 vs. 49.37 years; p<0.01) and more likely to be male (46% vs. 41%; p<0.01). Table 1 further outlines differences in demographics between the surgical and non-surgical hospitalizations, with significant differences in race, Charlson comorbidity index, primary payer, income quartile, and hospital characteristics between the two groups. From 1998-2018, the proportion of all hospitalized CD patients who underwent any IBD-related surgery decreased from 12.0% to 6.5%, (p trend<0.01), while the proportion of CD surgeries for only obstruction significantly increased from 10.8% to 26.4% (p trend=0.02). In adjusted analyses, there was a trend for decrease in all surgeries (OR 0.98, 95% CI 0.97– 0.98; p trend<0.01) and increase in surgeries obstructive indication (OR 1.02, 95% CI 1.0 – 1.03; p trend=0.03) per year.

Conclusion:
In the era of biologics, there has been a significant decrease in the rate of hospitalized CD patients undergoing surgery; of those hospitalizations, however, a larger proportion underwent surgery for obstructive indication. Our findings suggest that advances in medical therapy may have decreased overall surgical rates, yet have had a limited impact on stricturing disease, leading to continued reliance on surgical treatments.

Introduction: Trans anal excision (TAE) of rectal cancer can be considered as definitive treatment if the infiltration depth is T1 or lower, and the lesion is completely included within the resection margin. Moreover, it can be used as an alternative approach in case of a complete response to neoadjuvant therapy. The immune microenvironment within the bowel mucosa plays a role in the immune surveillance mechanisms against carcinogenesis. Our aim was to analyze the immune microenvironment in healthy rectal mucosa as possible predictor of tumor infiltration depth, lateral tumor spread, and complete response to neoadjuvant therapy of rectal cancer undergoing TAE.
Methods: This study is a sub-analysis of data from the IMMUNOREACT 1 and 2 trials (NCT04915326 and NCT04917263) including all the patients who underwent TAE of rectal cancer. In this multicentric study, we collected healthy mucosa surrounding the rectal cancer. A panel of immune markers was retrospectively investigated at immunohistochemistry: CD3, CD4, CD8, CD8beta, Tbet, FoxP3, PD-L1, MSH6, and PMS2 and CD80. A prospective analysis was performed with flow cytometry to determine the proportion of epithelial cells expressing CD80, CD86, CD40, HLA ABC or HLA DR and the proportion of activated CD8+ T cells, CD4+ Th1 cell, and T reg.
Results: A total of 64 patients with rectal cancer who were treated with TAE were analyzed: 41 in the retrospective cohort and 23 in the prospective cohort. In our study group, tumor infiltration depth over T1 stage was observed in 10 patients, lateral tumor spread in 8 patients, and complete response to neoadjuvant therapy in 19 ones. Deep tumor spread (over pT1 stage) was associated with a high CD8+/FoxP3+ T cells ratio (p=0.07), low CK+CD80+ mean fluorescence intensity (MFI) (p=0.032), high CK+CD86+ cell rate (p=0.008) and high CK+HLA-I + cell rate (p=0.038). Accuracy for predicting tumor infiltration depth was AUC=0.9 for CD8+/FoxP3+ T cells ratio, AUC=0.89 for CK+CD80+ MFI, AUC=0.98 for CK+CD86+ cell rate and AUC=0.98 for CK+HLA-I + cell rate. Accuracy for predicting lateral tumor spread was AUC=0.82 for CD8+CD38+ T cell rate, AUC=0.82 for CK+CD80+ MFI, and AUC=0.96 for CK+HLA-I + cell rate. Accuracy for predicting complete response was AUC=0.90 for CK+CD86+ cell rate, and AUC=0.8, p=0.017 for CK+HLA-I + cell rate.
Conclusion: In our series, patients with locally spreading rectal cancer that should undergo further surgery to complete rectal cancer treatment show higher antigen presentation within the healthy rectal mucosa surrounding the cancer. However, lymphocyte activation and co-stimulation are lower suggesting weak immune surveillance mechanisms. Similarly, In case of lack of complete response to neoadjuvant therapy, there is a higher antigen presentation by epithelial cells but low cytotoxic T cell activation.