Clinical studies have shown that patients with right-sided colon cancer have reduced progression-free and overall survival compared to patients with left-sided colon cancer. At the same time, patients with right-sided colon adenomas have a higher risk of developing interval cancers after endoscopic mucosal resection. Dysbiosis and known bacterial drivers play a significant role in the carcinogenesis of colorectal cancer, and may contribute to the growth of adenomas. Clinical studies have found biofilm formation and positive staining for Fusobacterium nucleatum (FN) in almost all right-sided colon cancers and adenomas. The composition of bacteria also in the normal mucosa in these two groups of patients has been found to be different from healthy controls.

Endoscopic preoperative antibiotic local treatment of the biofilm and tumors would theoretically result in decreased tumor mass, recovery of the mucosa, and a normalization of the immune response in the treated section of the colon. The oral administration of the antibiotic metronidazole in mice has shown to lower the tumor load (FN-positive xenograft tumors) and decrease in intratumoral abundance of the pro-carcinogenic bacteria, FN.

The aim of this study is to investigate the effect of local antibiotic treatment with fosfomycin and metronidazole on tumor characteristics and the colonic biofilm in patients with right-sided colon cancer or right-sided colon adenomas. This is a clinical proof-of-concept intervention study, and the investigators are to our knowledge the first group to look into local antibiotic treatment of biofilm in patients with colon cancer or colon adenoma. It is based on a non-randomized trial design with an open label single group assignment. There are two tracks depending on the pathology of the tumor: 1) patients with right-sided colon cancer; 2a) patients with right-sided colon adenoma. The intervention is a therapeutic endoscopy where the antibiotics fosfomycin and metronidazole are sprayed throughout the right colon section from caecum to the right colon flexure (100ml of gel contains 800mg of fosfomycin and 200mg of metronidazole). Patients will be included prospectively from the surgical department of Herlev University Hospital and Zealand University Hospital once having received oral and written information, followed by signing the consent. There will be a retrospective cohort of patients with colon adenoma from 2018 (track 2b). The investigators will retrieve the archived adenoma tissue as control tissue. In Clinical Trial, Track 1 and Track 2a will be registered as two different arms that use the same intervention, but the results from each arm later on is going to be reported in separate publications.

Background Dysbiosis and known bacterial drivers play a significant role in the carcinogenesis of colorectal cancer. Among bacterial drivers, Fusobacterium nucleatum (FN) and the formation of biofilm on the surface of the colonic epithelium are considered pro-carcinogenic. The increased permeability of the mucosal layer and the invading bacteria in the colonic crypts and submucosal space result in genetic and epigenetic changes along with immune responses that are all important oncogenic factors. Clinical studies have found biofilm formation and positive staining for FN in almost all right-sided colon cancers and adenomas. Two studies compared the bacterial growth in normal mucosa from patients with and without adenoma. They found a different composition of bacteria in the normal mucosa from patients with adenomas in terms of richness, diversity and bacterial strains, thus indicating that the bacterial composition in the gut could contribute to the growth of adenomas and subsequently, adenocarcinomas.

Endoscopic preoperative local antibiotic treatment of the biofilm and tumors would theoretically result in decreased tumor mass, recovery of the mucosal layer, and a normalization of the immune response in the treated section of the colon. The oral administration of the antibiotic metronidazole in mice has already been shown to decrease tumor growth (FN-positive xenograft tumors) as well as a decrease in intratumoral FN abundance.

Investigational Medicinal Products In this study a gel containing the antibiotics fosfomycin and metronidazole will be sprayed endoscopically on the mucosa of the right hemicolon. The gel will be prepared on site by combining solutions of fosfomycin and metronidazole with a gel-forming gum.

Study objectives The aim of this study is to investigate the effect of local antibiotic treatment with fosfomycin and metronidazole on tumor characteristics and the colonic biofilm in patients with right-sided colon cancer or right-sided colon adenomas.

Study design This study is a clinical proof-of-concept intervention study. It is based on a non-randomized trial design with an open label single group assignment. There are two tracks depending on the pathology of the tumor: track 1, patients with right-sided colon cancer; track 2a, patients with right-sided colon adenoma. The intervention is a therapeutic endoscopy where the antibiotics fosfomycin and metronidazole are sprayed throughout the right colon section from caecum to the right colon flexure in a gel-forming mixture. Patients will be included from the surgical department of Herlev Hospital and Zealand University Hospital. There will be a retrospective cohort of patient with colon adenoma, track 2b.

Effect parameters For patients with right-sided colonic carcinomas and/or adenomas, the investigators hypothesize that endoscopic, preoperative antibiotic treatment in the right colon with a gel-forming spray containing fosfomycin and metronidazole will result in normalization of the colonic epithelium and immune response as well as a reduction in colonic biofilm formation often seen in these patients.

Trial structure and responsibilities The principal investigator (Astrid Bennedsen) is responsible for the preparation of the protocol, the Case Report Forms (CRFs), coordinating any meetings of the safety data monitoring committee as well as implementing decisions in the daily operations of the trial. Along with a team of co-investigators, the principal investigator will do the screening, recruitment, data collection and writing of the CRFs. The safety data monitoring committee consists of three members and will be impartial. They will overview the progress and the safety of the trial. The meetings of the safety data monitoring committee will take place every six months, but an extra meeting will be held in case of any severe adverse events or any other serious incidents during the trial. The trial will be monitored by a local GCP unit. The funding source had no role in the design of this study and will not have any role during its execution, analyses, interpretation of the data, or decision to submit results.

Stopping rules In case of serious side effects or adverse events, the data will be monitored and the trial stopped if there is any evidence of lack of safety regarding the intervention and the study drugs. Included are unexpected complications in relation to colon cancer surgery.

Study course This study has two tracks depending on the pathology of the colon tumor. When colon cancer is suspected, patients are referred to a colonoscopy during which biopsies are taken to confirm pathology indicating adenocarcinoma or benign pathology. There is a retrospective track 2b, where the investigators will retrieve FFPE adenoma tissue from up to 28 random patients operated with endoscopic mucosal resection in 2018. The trial drugs will be accounted for - both per site and per patient.

Selection and enrollment of participants Consecutive patients with endoscopically diagnosed right-sided colon adenomas or right-sided colon cancer tumors will be eligible for inclusion in the study. The investigators plan to include 14 patients with colon cancer (track 1) and 14 patients with colon adenoma (track 2a).

Also, 28 patients with colon adenoma will be included retrospectively (track 2b). The investigators will retrieve FFPE tissue from the resected mucosa bearing adenoma. They have had endoscopic mucosal resection performed in 2018 at Department of Surgery, Zealand University Hospital.

Recruitment of patients will be done consecutively according to the Research Ethics Committee's recommendations. Patients who are eligible for inclusion will receive both oral and written information about the study. At the end of the enrolment meeting the patient may sign the informed consent if they wish, and may also be given the time needed for consideration (at least 24 hours). The informed consent will be written in accordance with the Danish law. Patients will be informed if any changes in the study occur, as this might influence the patient's willingness to participate or the safety of the participating patients. The approved document will be updated accordingly.

A completed test person

A completed test person is defined as:

• Track 1: a patient who received the therapeutic endoscopy and had preoperative blood samples drawn on the operation day. Other than the interview on the day of the out-patient clinic visit (approximately postoperative day 14) there is no further study-related physical follow-up of the patient.
• Track 2a: a patient who received the therapeutic endoscopy and had preoperative blood samples drawn on the operation day. Other than the telephone interview on postoperative day 14 there is no further study-related physical follow-up on the patients.

Dropout procedure Dropout and withdrawals may happen at any point from the written consent till the preoperative blood sample on the operation day, and will not have any influence on the treatment of the patient.

Participating patients can withdraw from the study at any time for any reason. The patients will be informed of this right in accordance with the Helsinki Declaration.

The investigator can withdraw patients if the investigator feels the safety of the patient is compromised by the participation in the study. In that case, the investigator must inform the patient about the reason for withdrawal.

Patients who withdraw their consent will receive standard treatment. The reason, if stated, for the withdrawal will be noted in the CRFs. The number of dropouts and withdrawals will be reported. No further data will be collected from the patients withdrawing from the study; however, the previously collected data will be part of the final data analysis unless the patient specifically requests all data to be deleted.

Study interventions and procedures Intervention This is a non-randomized trial with an open label single group assignment. The intervention is a single therapeutic endoscopy during which patient´s right colon section will be sprayed with a predetermined volume of two component solutions, one containing fosfomycin and metronidazole, the other containing a gum that forms a gel on mixing with the former. During administration, the two components will be delivered through a dual channel colono-videoscope. The gel will form on the bowel wall and adhere to the mucosa.

Track 1: patients with colon cancer Preoperative procedures - blood samples On the day of the therapeutic endoscopy study, blood samples are drawn. Blood samples are analyzed for hemoglobin (Hb), leucocytes and differential counts, c-reactive protein (CRP), thrombocytes, albumin, creatinine, estimated glomerular filtration rate (eGFR), transaminases in order to have baseline information on the patients and to be able to adjust for these variables in the multivariate analyses. Blood samples will also be analyzed for tumor-specific markers: carcinoembryonic antigen (CEA). Samples are also analyzed by means of the nCounter® PanCancer Immune Profiling Panel (nanoString).

Preoperative procedures - therapeutic endoscopy At least 24 hours after the patient has given informed consent to participating in the study, the patient will be given a bowel preparation. After 48-72 hours the therapeutic endoscopy will take place. The bowel preparation procedure will follow the different sites standard guidelines of bowel preparation using Picoprep. The patient will be offered to be sedated before the endoscopy. Pictures of the colon will be taken during the endoscopy.

Before spraying the right hemicolon, biopsies will be taken from the central and peripheral part of the tumor, and biopsies from the macroscopically normal mucosa in the proximal colon at least 10 cm orally and anally from the tumor.

The proximal colon will be sprayed with a predetermined volume of the study medication through a double spray catheter from the ileum to the right flexure using a Video Colonoscope. The study medication contains fosfomycin and metronidazole, which are mixed in situ with the gel-forming components formulated on behalf of Reponex Pharmaceuticals A/S.

Preoperative procedures - pathology Some biopsies taken during the therapeutic endoscopy will all be directly fixed in Phosphate-buffered (PBS-buffered) formalin and embedded in paraffin before microscopic assessment and further molecular biological or immunologic tests are performed. One biopsy from the healthy mucosa, one biopsy from the tumor center and one biopsy from the tumor periphery will be frozen immediately following the local protocol.

Operation On the day of surgery, blood samples will be taken again. They will be analyzed using the same methods as the preoperative blood samples. Inclusion in the study does not interfere with the standard surgical protocol for patients with right-sided colon cancer.

Postoperative procedures - pathology The resected colon section will be sent unfixed to the Pathological Department at the site, where the initial macroscopic assessment is performed: measuring of the tumor, lymph node assessment etc. One tissue sample from the healthy mucosa, one tissue sample from the tumor center and one tissue sample from the tumor periphery will be frozen immediately following the local protocol. The hemicolon will then be formalin-fixed. The paraffin-embedded tumor samples will be cut according to the standard protocol before microscopic assessment and further molecular biological or immunologic tests are performed. The histological assessment will follow the standard protocol.

Discharge The discharge criteria are the same for patients included in the study as for other patients on the ward. Discharge is expected to take place on the 3rd postoperative day if no complications arise. A prolonged hospital stay will be defined as 10 days after surgery.

Postoperative out-patient visit - follow-up Patients will be seen after their postoperative out-patient visit. The investigators will assess the patient's wellbeing and to ensure that any side effects or adverse reactions to the medicine in the trial are registered.

Follow-up Patients will be contacted by telephone 12 months after surgery. There is no further study-related physical follow-up on patients, as it has no influence on the endpoints. The investigators will however, read the patient's hospital files after 12 months to see whether the therapeutic endoscopy has changed the clinical course of the patient, i.e. if staging and oncological treatment are changed because of the procedure. The patients are receiving the standard post-operative program with ambulatory follow-up according to the Danish Colorectal Cancer Group (DCCG) recommendations.

Track 2: patients with adenomatous polyps of the colon Preoperative procedures - blood samples On the day of the therapeutic endoscopy, study blood samples are drawn. They will be analyzed by the same methods as the preoperative blood samples in Track 1.

Preoperative procedures - therapeutic endoscopy The bowel preparation procedure and sedation are the same as in Track 1. Pictures of the tumor will be taken during the endoscopy.

Before spraying the right hemicolon, biopsies from the macroscopically normal mucosa will be taken at least 10 cm orally and anally from the polyp.

The spraying procedure is the same as in Track 1.

Preoperative procedures - pathology Three of the four biopsies of the normal mucosa taken during the therapeutic endoscopy will all be directly fixed in PBS-buffered formalin and embedded in paraffin before microscopic assessment and further molecular biological or immunologic tests are performed. One biopsy of the normal mucosa will be frozen immediately following the local protocol.

Resection On the day of resection, blood samples will be taken again. They will be analyzed by the same methods as the preoperative blood samples. During the endoscopic mucosal resection, biopsies from the healthy colon will be taken: two biopsies 10 cm orally and two biopsies 10 cm anally from the polyp.

Postoperative procedures - pathology Three of the four biopsies of the normal mucosa taken during the operation will all be fixed directly in PBS buffered formalin. One biopsy of the normal mucosa will be frozen immediately following the local protocol. The resected mucosa with the adenoma will be send non-fixed to the Pathological Department at the site, where the initial macroscopic assessment is performed, measurement of the polyp in three dimensions. The mucosa and the biopsies will be fixed with formalin, and will be cut according to the standard protocol and paraffin embedded before microscopic assessment and further molecular biological or immunologic tests are performed. The histological assessment will follow the standard protocol.

Follow-up On postoperative day 14+/-2, an investigator will call the patient to assess the patient's wellbeing and to ensure that any side events or adverse reactions to the medicine in the trial are registered. The Follow-up is as in Track 1.

Archived FFPE tissue from patients with colon adenoma FPPE tissue blocks from colon adenomas will be retrieved from the archive of Department of Pathology, Zealand University Hospital. The tissue blocks will be cut according to standard protocol. They will be analyzed as the tissue from the resected mucosa of the patients included in track 2a.

Pathology procedures for both tracks The paraffin blocks will be sectioned in series. In each step, the section 1 is stained with standard hematoxylin and eosin (H&E), section 3 is stained immunohistochemically for CD8 and CD3 (following the local immunoscore protocol), and section 5 will be stained for CD68 (macrophages). Sections 7, 9 and 11 will be analyzed for bacterial biofilm using a general bacterial PNA FISH probe. Section 13 will be prepared for PanCancer IO 360TM Gene Expression Panel (nanoString) analysis following the nanoString protocol. The remaining sections (2, 4, 6, 8, 10 and 12) remain unstained, and will be kept at -80°C.

Microbiology procedures for both tracks In this study tissue samples will be analyzed for bacteria and biofilm using Fluorescence in situ hybridization and Metagenome sequencing. These analyzes will be performed at Costerton Biofilm Center.

Safety assessments In this study, patients are administered the study drugs in a standard single dose of 800 mg of fosfomycin and 200 mg of metronidazole. The general dosage guideline for adults treated with fosfomycin infusions is 6 g*3 per day (in patients with renal insufficiency and a creatinine clearance of 10 ml/min the suggested dose is 1 g*2 per day). The general dosage guideline for treatment with metronidazole is 1500 mg as a single infusion.

Blood samples The drawing of blood samples will result in some discomfort and may risk bruising at the site. A total of 8.5mL will be drawn at the sites: Zealand University Hospital, Roskilde, and Herlev Hospital (3.0 mL Gel + LiHep Tubes + 3.0 mL EDTA Tubes + 2.5 mL PAXgene Blood RNA Tubes). A total of 11.5 mL will be drawn at the site: Zealand University Hospital, Køge (3.0 mL Gel + LiHep Tubes + 3.0 mL Gel + LiHep Tubes + 3.0 mL EDTA Tubes + 2.5 mL PAXgene Blood RNA Tubes).

Therapeutic endoscopy:

In this study a full colonoscopy is necessary to deliver the study medication to the mucosa of the right hemicolon. Also, to assess the effect of the local antibiotic treatment it is necessary to take biopsies from the healthy mucosa, and if the patient is included in track 1, also the colon cancer tumor. Forty-eight-72 hours before a colonoscopy the patient needs to be given a bowel preparation (15). There are risks associated with a colonoscopy: intestinal bleeding, bowel perforation, localized infection, adverse reaction to the sedatives and generalized bowel infection. Of these, intestinal bleeding and iatrogenic bowel perforation are the most prominent adverse events associated with a significant mortality rate and the need for surgical intervention. The risk of accidental bowel perforation has been estimated to 0.85/1000, and the risk of bleeding has been estimated to 1.64/1000 (16). However, the complications are less common with colonoscopies that do not involve polypectomy, such as this study. The colonoscopy will be performed by a trained specialist surgeon in the out-patient clinic. Patients unable to be sedated due to known allergies or comorbidities have already been excluded from study participation.

Adverse events During and after administration of fosfomycin and metronidazole any adverse events or intercurrent illnesses will be reported. Patients who have known allergy to fosfomycin and or metronidazole are excluded from study participation. Treatment with metronidazole is known to give rise to Antabuse (disulfiram)-like reactions in patients who continue to drink alcohol. Patients who are currently heavy drinkers are excluded from study participation. Patients will be orally instructed not to drink alcohol from the day before the therapeutic endoscopy during the enrollment meeting. This information is included in the Patient information. A summary of the product characteristics of fosfomycin and metronidazole will function as reference documents in evaluation of adverse effects. The investigators will monitor adverse events from the day of the intervention until postoperative day 14 (approximately 19 days in total). Safety of the intervention will be measured by adverse events.

All serious adverse events (SAE) will be registered and reported to the sponsor by investigator within 24 hour after the event has been recorded. All adverse events will be reported to the Danish Medicines Agency and the local Ethics Committee in an annual report with a report on patient safety of the trial. All serious adverse reactions (SAR) will be reported in an annual report together with a report on patient safety to the Danish Medicines Agency. All SAR will be judged through the product summary if they are a possible suspected unexpected serious adverse reaction (SUSAR). The investigators are responsible for reporting all lethal or life-threatening SUSAR to the Danish Medicines Agency and the regional Ethical Committee as soon as possible and no later than 7 days after knowledge of such reaction has been received. No later than after 8 days after the reporting, will the Danish Medicines Agency be informed of the follow-up. All other SUSAR will be reported to the Danish Medicines Agency and the regional Ethics Committee no later than 15 days after knowledge has been received by the sponsor and the investigators. The patient will in these situations be followed until the reaction has terminated either via the outpatient clinic at the hospital or via contact with the sponsor or the investigators.

Statistical considerations The investigators plan to include 14 consecutive patients with right-sided colon adenoma and 14 consecutive patients with right-sided colon cancer tumor. The trial will end by the last visit of the last patient. The retrospective cohort of 28 patients with colon adenoma (track 2b) will be matched 2:1 with the prospective 14 patients with right-sided colon adenoma (track 2a). The investigators estimate a dropout rate of 15%. Therefore inclusion will stop by the 24th fully evaluable patient. This is a clinical proof-of-concept intervention study, and the investigators are to our knowledge the first group to look into local antibiotic treatment of biofilm in patients with colon cancer or colon adenoma. the investigators expect this number of patients to be sufficient to provide the necessary experience in implementing the technique and form a basis for future studies, including their sample size calculation.

Data will be analyzed using non-parametric statistics or parametric statistics depending on the distribution of the data. P-values <0.05 are considered significant. Correction for multiple comparisons will be performed if needed. Descriptive statistics will be used for baseline characteristics. Both a principal intention-to-treat analysis and a per-protocol analysis will be performed, along with a standard sensitivity analysis. The effect of any missing data will be assessed with sensitivity analyses. Missing data will be addressed with a creation of clinically responsible imputations. The investigators will focus on molecular signatures discovery using supervised analyses (statistical tests like t-test or Wilcoxon test) and unsupervised analyses like PCA (principal component analysis) and ICA (independent component analysis). Statistics will be performed in collaboration with biostatistician Nicolas Derian.

Patients included in track 2a, who are diagnosed with cancer after the postoperative pathological assessment of the polyp, will be included in the intention-to-treat analysis.

Data collection and quality assessment In accordance with Danish law on personal data, the investigators will seek approval for this trial from the Danish Data Protection Agency. Data handling will be in accordance with the act on processing of sensitive personal information, the General Data Protection Regulation (GDPR) and the Data Protection Act. The investigators will also seek approval for the trial from The Research Ethics Committee and the Danish Medicines Agency.

Data from each patient will be recorded continuously and written in forms (CRF) prepared before the trial begins. During the recruitment meeting, the investigators will use a paper work sheet for baseline data and inclusion/exclusion criteria. During the therapeutic endoscopy, the investigators will use paper work sheets for data collection; afterwards data will be entered into an electronic database using the Easytrial platform. All other data will be recorded directly to the electronic Easytrial database. The paper work sheets will be stored identifiable in a locked cabinet at the respective research sites. All collected data will have source documentation.

Data will be stored as personally identifiable for 10 years, after which they are deleted the 1st of February 2029. Data will be stored in compliance with GDPR and at Teamsite (secure data storage).

If a patient withdraws from the study, the collected data will be anonymized. If the patient requests the collected data to be deleted, this will be done.

Good Clinical Practice and quality assessment This trial is performed in accordance with the Danish Medicines Act §88 section 2 and the order on good clinical practice number 744 of 29th of June 2006. The trial will be monitored by a Good Clinical Practice (GCP) unit, Capital Region of Denmark. The existing procedures for quality assurance and control in intervention studies will be followed through a scheduled monitoring agreement at the beginning, mid-term and at the end of the project. At auditing, monitoring and/or inspection from the Danish Medicines Agency, GCP-unit or from other relevant authorities, these will all be allowed relevant access to the data and information stored.

Ethical considerations The study is performed in accordance with the Helsinki Declaration and Danish law 593 of 14th June 2011 on the Research Ethics Committee's practices regarding health research trials. The project has been approved by the department management, the Danish Data Protection Agency (REG-010-2019), the Regional Research Ethics Committee (number SJ-753), and the Danish Medicines Agency (2019-000131-51). The study is registered at the EudraCT (number 2019-000131-51) and will be registered at www.clinicaltrials.gov. The trial will be monitored by the local GCP Unit. Any modification of the protocol which may impact on the conduct of the study, potential benefit of the patient or may affect patient safety, including changes of study objectives, study design, patient population, sample sizes, study procedures, or significant administrative aspects will require a formal amendment to the protocol. An amendment will be agreed upon by the study group and will need the approval of the Ethics Committee as well as other authorities prior to implementation, if suited.

The study will be conducted in the Surgical Department, Herlev Hospital, and Zealand University Hospital. It will contribute to the clarification of whether local antibiotic treatment of the colon tumor and the surrounding normal mucosa has a place in the treatment of colon cancer and colon adenomas.

Study patients are included after they have received oral and written information and after written consent has been obtained. At any time, the patients can withdraw their consent without impairing their relationship to the investigator or the doctors involved and without prejudicing their treatment at the Department. All patients can require further information about the project from the clinically responsible doctor.

The patients will be informed of any additional health information found in the study.

Funding and insurance DMSc Ismail Gögenur (sponsor), MD Hans Raskov (investigator), MD Mahdi Alamili (investigator) and MD Astrid Bennedsen (principal investigator) initiated this project. The investigators and the sponsor have no financial interests or conflicts of interest in this study. Reponex Pharmaceuticals A/S (Slotsmarken 12, 1.th.DK-2970 Hørsholm) will distribute the trial drugs and colonoscopes to the participating departments as well as cover salary expenses for MD Hans Raskov and MD Astrid Bennedsen and the running costs of the trial. The funding source had no role in the design of this study and will not have any role during its execution, analyses, interpretation of the data, or decision to submit results.

Information about external funding is included in the written information for the patients.

Insurance Patients are covered by the already existing insurance for hospitalized patients in the Surgical Department, Herlev Hospital and Zealand University Hospital (Patient Insurance Scheme).

Publication of research findings Positive, negative and inconclusive research results will be published in international peer-reviewed journals. The Research Ethics Committee and the Danish Medicines Agency will be informed within 90 days when the last patient has had the last visit, and a final report will be reported within 12 months in the EUDRA database.

No later than 12 months after trial termination, an article including the obtained results will be submitted for publication in an internationally recognized peer-reviewed journal with high impact. The investigators expect to get three publications from the study; a feasibility publication with a detailed description of the methods, a publication of the results of the intervention on patients with colon cancer and a publication on patients with colon adenoma. Authorship eligibility will be based on the guidelines of the International Committee of Medical Journal Editors (ICMJE).

Inclusion Criteria:

Inclusion criteria for track 1 and 2a:

• Male and female patients above 18 years of age with either:

right-sided colon cancer tumor with adenocarcinoma histologically verified scheduled for open or laparoscopic resection at the Department of Surgery, Herlev Hospital or Zealand University Hospital for track 1.

Right-sided adenomas ≥2cm in diameter endoscopically verified scheduled for endoscopic mucosal resection at the Department of Surgery, Herlev Hospital or Zealand University Hospital for track 2a.

• ASA I,II or III
• The caecum must be reached by the endoscope.

Inclusion criteria for retrospective controls (track 2b):

Male and female patients above 18 years of age who were operated for colon adenoma ≥2cm in diameter in the right hemicolon in 2018 at Department of Surgery, Zealand University Hospital. They will be matched 2:1 with the patients included in track 2a (cases) based on age and gender and pathology of the tumor.

Exclusion Criteria:

Exclusion criteria for track 1 and track 2a

1. Patients with previous allergic reaction to fosfomycin and/or metronidazole

2. Patients under current antibiotic treatment or patient who had the last dose of antibiotics 30 days prior to inclusion.

3. Patients with a non-passable tumor or patients where a part of the tumor is not visible during endoscopy (Track 1)

4. Patients with neoadjuvant chemotherapy or radiation 12 months prior to the resection.

5. Patients with a history of familial adenomatous polyposis (FAP) or hereditary nonpolyposis colorectal cancer (HNPCC)

6. Patients with a history of inflammatory bowel disease (IBD)

7. Patients under current treatment with warfarin (Marevan) and phenprocoumon (Marcoumar), or NOAK such as dabigatran (Pradaxa®), rivaroxiban (Xarelto®), edoxaban (Lixiana®) or apixaban (Eliquis®)

8. Patients under current treatment with Fenemal (Phenobarbital)

9. Patients who previously have received a fecal transplantation

10. Patients who have previously had colorectal cancer, and are now presenting with a secondary colon tumor.

11. Patients with a current alcohol use disorder (AUD): defined as a patient who are currently drinking 8 or more drinks/week for women and 15 or more drinks/week for men.

12. Predictable poor compliance (psychiatric disease, not speaking fluent Danish, mentally, impaired etc)

13. Patients with an American Society of Anaesthesiologists physical status

14. Classification (ASAscore) of IV.

15. Patients unable to be sedated

16. Pregnancy or lactation (fertile women must have a negative serum or urine pregnancy test to participate)

17. Fertile women who do not use safe contraception during the study period

18. Following contraceptive methods are acceptable when used consistently and in accordance, with both the product label and the instructions of the physician are:

    • Oral contraceptive, either combined or progestogen alone
    • Injectable progestogen
    • Implants of levonorgestrel
    • Estrogenic vaginal ring
    • Percutaneous contraceptive patches
    • Intrauterine device or intrauterine system with a documented failure rate < 1% per year
    • Male partner sterilization (vasectomy with documented azoospermia) prior to female patient ́s entry into the study, and this male is the sole partner for that patient.
    • Double barrier method: condom with spermicidal agent (foam/gel/film/cream/suppository), condom and occlusive cap (diaphragm or cervical/vault cap) with vaginal spermicidal agent (foam/gel/film/cream/suppository).

Exclusion criteria for retrospective controls (track 2b):

• Patients under the age of 18 years during the resection for colon adenoma
• Patients who were under antibiotic treatment during the EMR or patients who had the last dose of antibiotics 30 days prior to EMR
• Patients with neoadjuvant chemotherapy or radiation 12 months prior to the resection
• Patients with a history of familial adenomatous polyposis (FAP) or hereditary nonpolyposis colorectal cancer (HNPCC)
• Patients with a history of inflammatory bowel disease (IBD)
• Patients who previously have received a fecal transplantation
• Patients who have previously had colorectal cancer
• Patients registered in "Vævsanvendelsesregistret"
• Patients with insufficient tissue samples