Development of new, non-invasive tools based on faecal bacterial signatures for the early detection of colorectal cancer

Marta Malagón Rodríguez

Dr. Xavier Aldeguer Manté

This thesis aims to develop a new non-invasive tool based on faecal bacterial markers for the early detection of colorectal cancer, which is capable of detecting precancerous lesions before the onset of clinical signs. In this work three tools have been developed: (1) RAID-CRC, based on the combination of four bacterial species with FIT, capable of reducing FIT false positive results in a 50% in a population with symptoms compatible with CRC; (2) RAID-CRC Screen, based on the combination of six bacterial species, capable of reducing I a 20% FIT false positive results in an asymptomatic population aged between 50 and 69 (current CRC screening population); and finally, (3) RAID-LS, based on the combination of three bacterial species, capable of detecting colorectal neoplastic lesions absence in Lynch syndrome carriers (subjects with CRC genetic predisposition) with a 100% sensitivity.

The doctoral thesis entitled “Development of new non-invasive tools based on fecal bacterial signatures for the early detection of colorectal cancer” is divided into 3 chapters, each of which develops a tool for the early detection of colorectal cancer (CRC) aimed at a different target population. My application proposal is focused on the RAID-CRC Screen, a tool designed to diagnose advanced colorectal neoplasia, a term that includes both precancerous lesions and CRC, non-invasively through the detection of bacterial markers in a stool sample. This tool is described in Chapter 2 of the doctoral thesis. Although CRC is the third cause of death from cancer worldwide, it is one of the easiest tumors to prevent. For this reason, governments in many countries are focusing their efforts on its prevention. One of the most widely used prevention strategies is CRC screening. The aim of screening is to reduce the incidence of CRC through the detection and elimination of precursor lesions such as colorectal adenomas and reduce mortality. CRC is a disease that can be detected through screening since a high percentage of the population (5%) will suffer from it throughout their lives, its natural history has been widely studied, methods are available for its early detection, and its treatments are more effective the sooner it is detected. In addition, it has been observed that organized CRC screening programs are more cost-effective than other screening programs such as breast or cervical cancer. Since 2000, the Catalan Institute of Oncology has pioneered a pilot program for the early detection of colon and rectal cancer in Hospitalet de Llobregat. Since then, the program has been extended throughout the Catalan territory. The CRC screening program includes participants who do not have symptoms related to CRC and who are between 50 and 69 years of age, both inclusive. The operation of this program consists of a first non-invasive test that detects the presence of hemoglobin in stool (immunochemical fecal occult blood test (FOBT), FIT for short) and, in the event of a positive result, participants undergo a colonoscopy. Colonoscopy is an invasive, high-cost technique that causes several inconveniences to patients, such as having to undergo unpleasant prior preparation, having to be sedated, absence from work, among others. It is for this reason that the first step of screening programs must have a high diagnostic capacity, in order to avoid performing colonoscopies unnecessarily. In Catalonia, the cutoff point for the FOB of 100 ng hemoglobin/mL has been established, from which a participant is considered positive. The main disadvantage of the CRC screening program is the false positive results obtained with TSOFi. These results lead to patient concern, additional costs for the healthcare system, and the performance of unnecessary colonoscopies. Between 2000 and 2010, 55.2% of positive TSOFi results turned out to be false positives. Therefore, it is necessary to improve the tools currently used by developing inexpensive and non-invasive CRC screening tools that have greater specificity than TSOFi and thus be able to reduce the costs and inconveniences derived from the false positive results it produces. Another aspect to improve is the detection of precancerous lesions, also known as advanced adenomas, since currently the sensitivity of TSOFi for their detection is only 28%. It is important to detect patients who have this type of lesion since it is the precursor lesion of CRC and it must be eliminated as soon as possible. In the doctoral thesis that is presented as a candidate for the 2020 PIONEER AWARDS, a new non-invasive tool for CRC screening based on bacterial signatures has been designed and clinically validated, which is capable of reducing false positive TSOFi results in a population that has obtained a positive TSOFi result within the framework of the Catalan screening program. Initially, in a proof of concept, an algorithm based on fecal bacterial markers was designed capable of differentiating patients with a positive TSOFi with advanced neoplasia (advanced adenoma and/or CRC) from the rest of the conditions with a sensitivity of 95% and a specificity of 26%. These data were validated in an independent external cohort where a sensitivity of 84% and a specificity of 16% were obtained. When we focus on CRC detection, the sensitivity increases to 95%. All these data translate into a reduction in the false positive rate of 16%. One of the next objectives is to be able to introduce the RAID-CRC Screen into the Catalan colorectal cancer screening program, for this reason, it is essential to ensure that the sample collection and the circuit to be followed by the patient is simple and adapts to the current one since it works well; automate and scale the sample analysis process; develop a reagent kit that can be used by hospital laboratories that participate in the CRC screening program. The company interested in the commercial development of the findings made within the framework of my doctoral thesis is GoodGut SL. It is a biotechnology company dedicated to the research and development of non-invasive systems to support the diagnosis and treatment of digestive diseases based on the intestinal microbiota. Therefore, the steps that are being carried out where the tool developed within the framework of the doctoral thesis will be applied are the following: 1. Clinical validation in the German screening population 2. Automation and scaling of the process 3. Design of the reagent kit and software for the analysis of the RAID-CRC Screen 4. Exploitation, communication and dissemination

Non-Invasive Tool; Faecal Bacterial Markers; Early Detection of Colorectal Cancer (CRC); Precancerous Lesions; RAID-CRC; Bacterial Species; FIT; Reduced False Positives; Symptomatic Population; RAID-CRC Screen; Asymptomatic Population; CRC Screening; RAID-LS; Colorectal Neoplastic Lesions; Lynch Syndrome Carriers; Genetic Predisposition; Sensitivity.