Individual Follow-up After Rectal Cancer - Focus on the Needs of the Patient (FURCA)

With an increasing number of rectal cancer (RC) survivors, we see a rising attention to the late adverse effects following treatment of RC. Late adverse effects that are highly prevalent and negatively impact patients' symptom burden and quality of life are: bowel-, urological and sexual dysfunctions; psychological distress; fear of recurrence. The current follow-up program primarily focuses on detection of recurrence, with less attention to late adverse effects. In a patient-led follow-up program, the surveillance for recurrent disease is combined with detection and treatment of late adverse effects and supportive survivorship care. The follow-up involves a high degree of patient-involvement, aiming at meeting the individual patient's needs. The patient-led follow-up programme is based on a standardized patient-education in order to enforce the patients to assess and respond sufficiently to symptoms and health problems. In case of symptoms and concerns, the patients are instructed to consult a health professional for adequate assistance and intervention. The intervention is tested in a multicenter randomized trial, comparing the patient-led follow-up to standard routine follow-up, involving prescheduled outpatient visits.

Recruitment: Patients are included from colorectal surgical departments in Aarhus, Randers, Herning and Aalborg. These centres cover one third of all Danish rectal cancer patients. Patients are approached at the postoperative visit in the outpatient clinic where they are presented for the pathology results, approximately 2 weeks after primary surgery. The doctor provides short oral information about the study and hand out written information. Following that, a research nurse makes contact with the patient by telephone a few days later, in order to clarify whether the patient wants to participate or not. Participation requires written informed consent and baseline information from the patient prior to randomisation. Sample size: An initial estimation of sample size showed, that a total of 334 participants were needed in order to obtain statistically significant results (alpha 0.05, beta 0.20, SD 13, minimally important difference 4 points on the TOI-score). The assumed standard deviation (SD) used for this estimation was based on studies where setting and populations differ substantially from ours. Therefore, an interim analysis of the main-outcome (Trial Outcome Index - TOI) has been performed in order to determine a more precise SD. In an updated sample size estimation, the following preconditions are listed: 10.4 SD, 0.05 alpha-value, 0.20 beta-value, an expected difference of 4 points on the TOI-scale and an expected total drop-out of 30%. Furthermore, additionally 15% are added to the sample size, due to the assumption that the outcome data will be non-parametric. Thus the conclusion from the updated sample size estimation is that a total of 324 participants are needed in order to obtain statistically significant results. Drop-out is defined by: verified metachronous recurrence, metastasis or other primary cancer, death, emigration from the participating regions or if a participant leaves the study. Data: Most data is collected using questionnaires, administered by the patient (Patient-reported outcome measures -PROM's). Patients are asked to complete a questionnaire at baseline (time of inclusion), and then at 12 and 36 months after surgery. The questionnaires consists of questions regarding symptom burden and quality of life, specific symptoms, patient activation, self-efficacy, patient involvement, patient information and needs in the follow-up period, and how they have been met by the Health care system. Data regarding recurrence of cancer and mortality is collected from the Medical charts, while data regarding socioeconomic factors is derived from national registers and data on comorbidity is extracted from the national clinical database for colorectal cancer surgery. Finally, data on health care utilisation in the follow-up period will be collected from national registers, in combination with information from patient records (limited access). Statistical analysis: Outcomes will be analyzed using methods for repeated measurements, and survival analysis. Any difference between the two allocation groups will be significance tested, and adjusted for covariates, using multiple regression analysis. Differences in demographics between the two groups will be calculated and significance tested using a chi2-test (dichotomous and categorical data) and t-test (numerical data). Quality-adjusted life years will be calculated using standard health economic methods and in collaboration with an expert in health economics. Study timeline: Development of the intervention was performed during the second half of 2015. Inclusion of patients in the RCT was initiated in February 2016, and is expected to be complete by the 31st of July 2018. With a three-year follow-up period, the final results from the study will be analyzed and ready for publication by the end of 2021. Ethical considerations: The risks and ethical concerns related to the project are limited, but there are some potential concerns. Removing routine follow-up could lead to a delay in detecting asymptomatic local recurrences. However, patients having symptoms with the opportunity of self-referral may get an earlier detection of recurrence. Answering the questionnaire could remind participants about their disease and lead some patients to unpleasant or stressful situations. Patients agreeing to participate should feel secure in whatever group they are randomized to follow. The study follows the ethical principles in the Helsinki Declaration, and is reported and approved by The National Committee on Health Research Ethics as required. Data is handled and stored, according to national law and only anonymised results will be published. The study is reported and approved by the Danish Data Protection Agency.

Participants are randomized (1:1) into either intervention or control group. This is done by block randomization, stratifying by center, sex and treatment type (± postoperative oncological treatment and ± temporary ileal stoma).

Participants in the experimental arm are enrolled in a patient-led follow-up program, based on patient-education and self-referral, in addition to recommendations from national guidelines for follow-up.

Participants in this control group follow standard care for follow-up after rectal cancer treatment, as described in local and national guidelines.

Patient-education, with information about signs of recurrent disease (alarm-symptoms), potential late effects, and how these can be managed. The patients are instructed to contact a specialist nurse in case of alarm-symptoms, concerns or other problems related to the cancer disease throughout the entire follow-up period. The specialist nurse responds to the referrals, according to a standardized algorithm. In addition, all patients receive: CEA and CT of the chest, abdomen and pelvis at 1 and 3 years after surgery perioperative clean colon colonoscopy, and then every 5 years until the age of 75 years Only for patients with a stoma: - access to stoma care by specialist nurses

Patients with sphincter-preserving resection: - outpatient visits incl. rectoscopy at 6, 12, 18, 24 and 36 months Patients with rectal amputation and a permanent stoma: outpatient visits at 3, 12 and 36 months access to stoma care by specialist nurses All patients: CEA and CT of the chest, abdomen and pelvis at 1 and 3 years after surgery perioperative clean colon colonoscopy, and then every 5 years until the age of 75 years

Full name: The Trial Outcome Index (TOI), which is a subscale within the Functional Assessment of Cancer Therapy - Colorectal (FACT-C) questionnaire. Construct: The FACT-C measures health related quality of life. Five subscales can be derived from the questionnaire: physical wellbeing subscale, family/social wellbeing subscale, emotional well-being subscale, functional well-being subscale and a colorectal cancer-specific subscale. The TOI combines the physical, functional and cancer-specific subscales, which are considered likely to change by an intervention being tested. Scale range (minimum and maximum scores): 0-84 Which values are considered to be a better or worse outcome: The higher the score, the better the QOL. Indicate how subscales are combined (summed, averaged etc.): The TOI is calculated by summing the scores from the physical, functional and colorectal cancer-specific subscales.

Full name: The Low Anterior Resection Syndrome Score (LARS-score). Construct: Assessment of bowel dysfunction following a low anterior resection for rectal cancer Scale range (minimum and maximum scores): 0-42 Which values are considered to be a better or worse outcome: The higher the score, the more severe symptoms of bowel dysfunction Indicate how subscales are combined (summed, averaged etc.): no subscales are used.

Full name: The Bristol Stool Scale Construct: Assessment tool of the form of human faeces, and classification into seven stool types. Scale range (minimum and maximum scores): type 1-7 Which values are considered to be a better or worse outcome: Types 1 and 2 indicate constipation. Types 3 and 4 indicate the ideal stools. Type 5 indicates slight diarrhea. Types 6 and 7 indicate diarrhea. Indicate how subscales are combined (summed, averaged etc.): no subscales are used.

Full name: 'The colostomy impact score'. Construct: measures impact from colostomy dysfunction on quality of life Scale range (minimum and maximum scores): The colostomy impact score ranges from 0 to 38 points. Which values are considered to be a better or worse outcome: A score of ≥ 10 indicates major colostomy impact. Indicate how subscales are combined (summed, averaged etc.): no subscales are used

Full name: 'The male/female lower urinary tract symptoms' (ICIQ-FLUTS/ICIQ-MLUTS) Construct: Evaluates male/female lower urinary tract symptoms and impact on quality of life Scale range (minimum and maximum scores): The ICIQ-FLUTS It is scored on a scale from 0-16 for symptoms of filling, 0-12 for voiding symptoms and 0-20 for incontinence symptoms. Range of the total score is 0-48. The ICIQ-MLUTS is scored on a scale from 0-20 for voiding symptoms and 0-24 for incontinence symptoms. The range of the total score is 0-44. Which values are considered to be a better or worse outcome: Higher scores indicate greater impact of individual symptoms for the patient Indicate how subscales are combined (summed, averaged etc.): For both the ICIQ-FLUTS and the ICIQ-MLUTS, the total score is calculated by summing the scores from the subscales.

Full name: 'The Sexual Function - Vaginal Changes Questionnaire' (SVQ) Construct: measures sexual and vaginal problems in gynaecological cancer, although it is assumed that rectal cancer patients experience similar problems due to changes in pelvic physiology and in the nervous innervation. Scale range (minimum and maximum scores): The measure includes 20 core-items and 7 change-items. Out of these, 13 questions are only filled-out if the responder is sexually active. A total score will not be reported. Items 1-3 and 21-23 are combined into a subscale evaluating sexual interest and relations, with a range from 6-21. Which values are considered to be a better or worse outcome: Higher values indicate better outcome Indicate how subscales are combined (summed, averaged etc.): The subscale for sexual interest and relations is calculated by summing the scores from the items 1-3 and 21-23.

Full name: 'The International Index for Erectile Function' (IIEF) for males. Construct: measures the effects from erection problems on sex life Scale range (minimum and maximum scores): 6-75 Which values are considered to be a better or worse outcome: higher scores indicate better outcome. Indicate how subscales are combined (summed, averaged etc.): no subscales are used

Full name: 'The Chronic Pain Score after Rectal Cancer Treatment' Construct: The score measures chronic pain in the pelvic area following treatment for rectal cancer. Scale range (minimum and maximum scores): 0 to 46 points. Which values are considered to be a better or worse outcome: Scores from 0-4 indicate no pain; 5-20 indicate minor pain; ≥ 21 indicate major pain Indicate how subscales are combined (summed, averaged etc.): No subscales are used

Full name: The Multidimensional Fatigue Inventory (MFI-20) Construct: evaluates five dimensions of fatigue: general fatigue, physical fatigue, reduced motivation, reduced activity, and mental fatigue. Scale range (minimum and maximum scores): 20-100 Which values are considered to be a better or worse outcome: Higher total scores correspond with more acute levels of fatigue. Indicate how subscales are combined (summed, averaged etc.): no subscales are used

Full name: The Fear of Cancer Recurrence Inventory (FCRI) Construct: multidimensional measure for fear of recurring cancer. Includes seven subscales, of which one (severity subscale) is considered a short-form subscale. Scale range (minimum and maximum scores): 0-168 Which values are considered to be a better or worse outcome: higher values indicate worse outcome Indicate how subscales are combined (summed, averaged etc.): The short-form subscale is calculated by summing values from items 9-17. The total score is summed by all item values.

Full name: 'The Hospital Anxiety and Depression Scale' (HADS) Construct: measures anxiety and depression in a general medical population of patients Scale range (minimum and maximum scores): 0-21 Which values are considered to be a better or worse outcome: Scores of 0-7 are considered normal, 8-10 borderline and 11 or over indicate clinical 'caseness'. Indicate how subscales are combined (summed, averaged etc.): The score is divided into a depression scale and an anxiety scale - both scales range from 0-21.

Full name: The patient activation measure (PAM) Construct: measures the patient's knowledge, skills and confidence for self-management. Scale range (minimum and maximum scores): The 13 items have a calibrated scale range from 38.6 to 53.0 (on a theoretical 0-100 point scale) Which values are considered to be a better or worse outcome: PAM segments people into one of four progressively higher levels of activation. Indicate how subscales are combined (summed, averaged etc.): no subscales are used.

Full name: Patient involvement Construct: measures the patient's experience of being involved by health care professionals during follow-up Scale range (minimum and maximum scores): 6-60 Which values are considered to be a better or worse outcome: higher scores indicate better outcome. Indicate how subscales are combined (summed, averaged etc.): no subscales are used

Full name: Patient information and sense of security Construct: measures the patient's experience of receiving adequate information and help during follow-up Scale range (minimum and maximum scores): 5-25 Which values are considered to be a better or worse outcome: higher scores indicate worse outcome. Indicate how subscales are combined (summed, averaged etc.): no subscales are used

QALY's derived from 'The EuroQol 5-dimensional health state measurement'. Data from national health care registries are also utilised

Inclusion Criteria: rectal resection for primary rectal adenocarcinoma R0/R1 resection (radical resection of the tumour) Exclusion Criteria: metastatic disease synchronous cancer not able to understand Danish language severe cognitive deficit, i.e. dementia residual life expectancy less than two years concurrent participation in other scientific studies which affect frequency and content of the follow-up program

Hovdenak Jakobsen I, Juul T, Bernstein I, Christensen P, Jensen FS, Johansen C, Lindhardt Larsen S, Laurberg S, Madsen MR, Thorlacius-Ussing O, Vind Thaysen H. Follow-up after rectal cancer: developing and testing a novel patient-led follow-up program. Study protocol. Acta Oncol. 2017 Feb;56(2):307-313. doi: 10.1080/0284186X.2016.1267400. Epub 2017 Jan 9.