Italian Validation of the LARS Score

• BACKGROUND and RATIONALE Colorectal cancer, with 49,000 new diagnoses expected in 2019 (27,000 in men and 22,000 in women) represents, in Italy, the third neoplasm in men (14%) and the second in women (12%). Among colorectal tumors, the rectum represents the site most frequently involved (about 35%).

Thanks to an earlier diagnosis and to the greater effectiveness of an implemented multidisciplinary treatment, the locoregional control and the patients' overall survival significantly improved over the last decades.

Increasing attention has been recently paid to the outcomes of surgical treatment, in terms of patient's anorectal function and quality of life (QoL). Currently, the majority of patients affected by rectal carcinoma undergo a sphincter-sparing surgery, that is, a low anterior rectal resection (LAR), therefore avoiding a permanent colostomy.

It is known that about 50% - 90% of patients undergoing LAR will have at least some degree of bowel dysfunction: for this reason, the definition of "anterior low rectal resection syndrome" (LARS) has been coined to describe this complex functional condition. The main symptoms included in this syndrome are: incontinence to gas and/or liquid or solid stools, constipation, urgency, fragmentation and frequent bowel movements. In addition, a worsening of QoL has been observed in patients with severe LARS symptoms.

Due to the importance and high prevalence of this condition, in order to identify a reliable tool for assessing its severity, the so-called "LARS score" has been introduced. The score has already been validated in several languages, including English, Chinese, Lithuanian, Swedish, Spanish, German, Danish (as consolidated international validation), and finally Dutch.

However, this important tool has not been validated in Italian language.

• AIMS The primary aim of this study will be the validation in Italian language of the LARS score in a population of Italian patients with a previous history of rectal cancer, and treated by anterior rectal resection surgery. Moreover, the study will give the possibility to investigate the convergent and discriminatory validity, and to re-test the reliability of the score.
• STUDY DESIGN The study will be a prospective observational study on patients affected by rectal cancer, and treated by anterior rectal resection surgery with total or partial excision of the mesorectum (TME, total mesorectal excision, or PME, partial mesorectal excision) in the period January 2000 - April 2018. Any Surgical Unit of the Fondazione Policlinico Universitario "Agostino Gemelli" - IRCCS of Rome that deals with rectal cancer surgery could join the study. Each Unit will identify a Study Manager, and possibly a collaborator.
• METHODS

The study will provide different phases:

• TRANSLATION OF THE QUESTIONNAIRE The validated English version of the LARS questionnaire will be translated into Italian. The translation will be performed by two independent professional translators. Translators will discuss any discrepancies between their translations until an agreed version is reached. A third native English translator will translate the Italian version into English. Subsequently, the two English versions (the initial one and the new one) will be compared and the final version in Italian will be elaborated.
• CONVERGENT VALIDITY To study the "convergent validity", the LARS score will be correlated with the single QoL question and with the EORTC-QLQ-C30 questionnaire, in order to test how the LARS score correlates with the QoL.
• DIVERGENT VALIDITY To study discriminative validity, we will evaluate the ability of the LARS score to identify different patient subgroups, which generally differ in the LARS score. Similar to previous validation studies, the clinically relevant subgroups will be based on: preoperative radiotherapy, type of surgery (TME / PME), age (cut-off at 69 years).
• TEST- RETEST RELIABILITY Test-retest reliability is one of the key measures of all health measures. To examine the test-retest reliability of the LARS score, all patients will receive a second LARS questionnaire 1-2 weeks after completing the first one, and will be invited to reply to the questionnaire again.

Each Unit could choose to assess the patients in the preferred way (clinical follow-up visit, e-mail, ordinary mail); however, the method of administration of the questionnaire must be specified in the final communication of data to the coordinating Group.

Each Unit will be responsible for the truthfulness of the data collected and provided.

• DATA COLLECTION

Each Unit joining the Study will be provided with an Excel data collection sheet which must then be completed. The collection of data and their transmission to the Study Coordinator Group must occur within 3 months from the beginning of the Study. For each patient we ask to provide:

• 2 LARS questionnaires, administered 1-2 weeks apart;
• 1 EORTC-QLQ-C30 questionnaire;
• 1 single QoL question.
• STATISTICAL ANALYSIS and DATA EVALUATION The quality control of the collected data and the statistical analysis of the results will be conducted by a specialist in Medical Statistics, who will be guaranteed co-authorship in the scientific publications derived from the study. No financial compensation will be provided for this activity.

Concerning the "convergent validity", the correspondence between the QoL categories (no impact, minor or major impact of intestinal function on QoL) and the three LARS scoring categories will be assessed. The correspondence between these two measures will be considered perfect when: patients have not reported LARS and any impact on QoL; patients have reported minor LARS and a minor impact on QoL; in case of major LARS and major impact on QoL. Correspondence will be defined as moderate in case of a difference in one category, for example no LARS but a minor impact on QoL. In case of difference in two categories, the two measures will be considered not corresponding at all.

LARS score data will be presented as median (with interquartile range) and the differences will be tested by the Kruskal-Wallis test. The same test will be used to test the null hypothesis in the study of the association between the LARS score and the five functional scales and the global QoL scale of the EORTC-QLQ-C30 questionnaire. Wilcoxon's rank test will be applied to investigate the "discriminatory validity".

The correspondence between the two tests to measure the reliability of the LARS score will be presented as proportions with 95% confidence intervals (95% CI). A Bland-Altman diagram will be calculated with 95% agreement limits and intraclass correlation (ICC). An ICC lower than 0.50 will indicate poor reliability, an ICC between 0.50 and 0.75 will reflect moderate reliability, between 0.75-0.90 good reliability and an ICC greater than 0.90 will indicate excellent reliability.

• PUBLICATION POLICY and AUTHORSHIP The results of the study will be the subject of a group publication. The co-authorship of the manager and a collaborator will be guaranteed to each Unit participating in the study. The first, second and last Author of the publication will be respectively the principal investigator and the two collaborators (junior and senior) of the promoter / coordinator Center.

Inclusion criteria

• rectal neoplasia (between 0 and 15 cm from the anal verge);
• anterior rectal resection surgery with total or partial mesorectal excision (TME or PME) in the period January 2000 - April 2018;
• if a stoma had been created, intestinal continuity must have been restored for at least 24 months (by April 2018);

Exclusion criteria

• dementia;
• metastatic or recurrent disease;
• other intestinal diseases (including Crohn's disease, ulcerative colitis);
• patients with a stoma or with intestinal continuity restored for less than 24 months;
• patients with problems of understanding the Italian language.