Lack of funding

Rationale: The surgical complication of intestinal anastomotic leak remains a clear and present danger to patients despite advances in surgical technique and ever more powerful antibiotics. No surgeon is immune from this complication and leak rates have not changed in decades. The consequences of a leak (peritonitis, sepsis, death) can be so severe that in the case of rectal cancer, diverting ileostomies are routinely performed to divert the fecal stream away from the healing anastomosis. We have recently discovered that certain intestinal bacteria, with the capacity to express collagenase and cleave MMP9 (Matrix metallopeptidase 9) to its active collagen degrading form, play a key and causative role in anastomotic leak. These bacteria often escape elimination due to the failure of current antibiotic regimens and their delivery methods to remain functionally durable at anastomotic tissue sites.

Purpose: This phase II clinical trial will track, in real time, the process of anastomotic healing and its associated microbiome by performing serial endoscopic surveillance (SES) following rectal cancer resection. By capturing anastomotic images and the associated microbial and inflammatory mediators from anastomotic fluids via SES performed at three time points following rectal cancer resection, we will correlate healing to microbial composition and inflammatory mediator status. Patients will be randomized and, at each time point, will receive lavage of their anastomosis with either saline or a triple antibiotic solution (ciprofloxacin, metronidazole, neomycin). An anastomotic healing score captured during SES will be compared between the two treatment arms and correlated to microbial and inflammatory mediator analyses of fluid samples to determine how intestinal microbes influence the process of anastomotic healing.

OBJECTIVES

• Perform repeated postoperative endoscopy of anastomotic tissue sites in patients undergoing low anterior resection for rectal cancer with diverting ileostomy.
• Demonstrate that SES can reliably discriminate those anastomoses that progress along a continuum toward normal healing versus those that heal pathologically when images are evaluated by a jury blinded to treatment.
• Demonstrate that those anastomoses that are judged to progress pathologically harbor distinct microbial species and predictive inflammatory mediators.
• Demonstrate that when anastomotic tissues are lavaged with antibiotics directly via endoscopy (ciprofloxacin, metronidazole, neomycin), microbial pathogens that escape conventional antibiotics are eliminated and anastomotic healing progresses normally.
• Determine if the anastomotic healing score, microbial analyses, and/or inflammatory mediator analyses identify or predict symptomatic and clinically relevant anastomotic complications.

STUDY This is a randomized, multicenter study. All patients undergo resection of rectal cancer with a low anterior resection with anastomoses within 10cm of the anal verge and diverting ileostomy. Patients will undergo three endoscopies: one intra-operative endoscopy (POD0) and two post-operative endoscopies (POD3-7, POD12-28) (POD = post-operative day). During the SES procedure, the anastomosis will be lavaged with saline, the fluid will be collected by suction, images will be taken and then the anastomosis will be lavaged with either saline or an antibiotic solution composed of metronidazole, ciprofloxacin and neomycin. Patients' clinical courses will be followed through chart review for 6 to 7 months following surgery.

PROJECTED ACCRUAL A total of 200 patients (approximately 100 per treatment arm) will be accrued for this study within 4 years.

Inclusion Criteria

• Diagnosis of rectal cancer
• Scheduled to undergo a low anterior resection (with the planned anastomosis within 10cm of the anal verge) and a diverting ileostomy
• 18 years of age or older
• Able to provide informed consent

Exclusion Criteria

• Prior medical history of or suspected diagnosis of inflammatory bowel disease or irritable bowel syndrome
• Pregnant or breastfeeding
• Allergy to lidocaine, ciprofloxacin, metronidazole or neomycin