RESPONSE: Colorectal Cancer Survivors' Follow-up Care - Now Digital and Need-based

Background:

Colorectal cancer (CRC) screening was implemented in Denmark in 2014 and has effectively shifted the CRC stage at the time of diagnosis from late stage (III and IV) to earlier stages (I and II)[1]. Consequently, more patients are offered curative intended treatment, which increases the number of survivors in postoperative follow-up care: In 2020, 65% of Danish patients with CRC, potentially eligible for follow-up care, had stage I-II disease[1].

The current follow-up for CRC survivors is recurrence-focused, with computed tomography (CT) imaging at 12 and 36 months[2] as early recurrence detection is critical to increase the possibility of curative treatment: The 5-year survival rate for patients treated for recurrence with curative intent is ~40% compared to <10% for patients managed with palliative or best supportive care[3-6]. However, the risk of recurrence strongly depends on the CRC stage: The 3-year cumulative recurrence rate is only 4.5%-7.9% for stage I and 10%-16% for stage II[7-9]. Consequently, the resources allocated to CRC follow-up in Denmark are primarily dissipated on patients who will never experience a recurrence. Hence, the challenge remains to distinguish between high- and low-risk patients, i.e., tailor the follow-up program to the personal risk of recurrence instead of "one-size-fits-all".

A promising and novel surveillance method for CRC recurrence is to screen longitudinally collected blood samples for the presence of circulating tumor DNA (ctDNA). Serial ctDNA analyses detect recurrence with high sensitivity (88%) and specificity (97%) independent of the stage (hazard ratio (HR)=40.7; 95% confidence interval (CI): 11.6-143) and with a median lead-time of 7-10 months compared to current standard-of-care follow-up[10-13]. Thus, serial ctDNA analyses have the potential to efficiently identify the 4.5%-16% of stage I-II patients, who should be offered CT imaging, whereas the remaining 84-95% of patients are spared unnecessary CT imaging.

CRC survivors with a low risk of recurrence may perceive other challenges than CRC recurrence as equally or more important in everyday life. Such challenges may include the psychological distress related to the CRC diagnosis, exaggerated fear of cancer recurrence (FCR) regardless of the actual risk of recurrence, and the presence of treatment-related organ-specific late effects, which may negatively impact their quality of life (QoL). Independently of stage, 13% of patients with CRC report persistently low QoL and/or high levels of psychological distress, e.g., impaired emotional well-being and/or high FCR[14].

Approximately half of CRC survivors suffer from organ-specific late effects, e.g., bowel, urinary, or sexual dysfunction[15-18]. In a recent study, 20% of colon cancer patients and 30% of rectal cancer patients expressed a wish for help managing their organ-specific late effects[19]. Besides organ-specific late effects, many CRC survivors experience one or more persistent general symptoms and late effects after their cancer treatment including psychological distress, depression, anxiety, insomnia, fatigue, pain, and impaired cognitive function. While the management of these so-called biopsychosocial late effects has received only little attention until recently, a growing body of evidence suggests that these issues can be treated effectively with cognitive behavioral approaches[20-24]. However, none of these challenges are addressed by today's recommended follow-up care program.

To improve the management of follow-up care, recent studies have demonstrated the benefit of high patient satisfaction with electronic Patient-Reported Outcome Measures (ePROMs)[25]. Furthermore, it has been shown that 80% of Danish patients with CRC respond to ePROMs and that those with organ-specific late effects and/or biopsychosocial late effects can effectively be identified using ePROMs[19]. Hence, ePROMs have the potential to help clinicians stratify CRC survivors to postoperative surveillance or interventions for treating both organ-specific late effects and biopsychosocial late effects.

New technology further facilitates the management of follow-up care: the use of digital care-guides has become increasingly popular in the Danish Health care system. One example is a framework based on a smartphone app that enables implementation of a comprehensive digital care guide in the follow-up program for CRC (Emento)[26]. This app can help maintain patient autonomy, acting as both a reference work and a timed tool to inform, educate, and guide the patient through the follow-up program.

RESPONSE proposes to use each of the elements described above in a new, individualized follow-up program for CRC. All elements have already been tested and have shown their great potential in separate efficacy trials[11-13,19,23,25]. However, the impact of combining all four elements in a single follow-up program has never previously been investigated.

Aim:

The overall aim of this study was to investigate whether the combination of the above elements in recurrence surveillance, could improve health-related QoL (HRQoL), without compromising overall survival (OS) and recurrence-free survival (RFS) or increasing costs. Thus, our study objective was to design a surveillance program fulfilling these criteria. Further, the objective was to conduct a trial where this program could be compared to standard-of-care recurrence surveillance.

The new follow-up program includes: 1) serial ctDNA monitoring to identify individuals with high risk of recurrence, 2) serial ePROMs monitoring to identify 'organ-specific late effects' and 'biopsychosocial late effects', 3) planned and systematic management/intervention of recurrence and late effects, and 4) personalized self-managed follow-up by a digital care guide as a smartphone application.

Study design:

This new program will be compared to the standard-of-care imaging-based recurrence surveillance in a Danish multicenter, interventional effectiveness trial, including 392 patients from 11 surgical centers. The patients will be divided into two arms: the intervention group (IG) and the standard-of-care group (SG).

IG patients (n=196) will receive all the following at 3-,12-,24-, and 36-months post-surgery:

1. Recurrence risk stratification by plasma ctDNA.

   1. If ctDNA becomes positive, CT imaging of the thorax and abdomen will be performed. This enables the CT imaging resources to be directed at the high-risk individuals (=ctDNA positive) only.

   2. The results of the CT imaging are discussed at the usual MDT at the responsible surgical department, where pathologists, oncologists, CRC surgeons, and radiologists are present. The MDT decides whether further diagnostic initiatives should be taken, e.g., endoscopy or further imaging.

   3. If recurrent disease is detected, the patient is treated according to the national Danish guidelines, and the outcome is registered in the RESPONSE trial.

   4. If a recurrence is NOT detected by imaging or subsequent clinical examinations, the patient returns to the RESPONSE trial with increased ctDNA testing frequency every four months.

   5. If longitudinal ctDNA tests become negative, the patient returns to the default ctDNA test frequency.

   6. If longitudinal ctDNA tests are repeatedly positive, CT imaging will be prompted and discussed at MDT until a site of recurrence can be confirmed.

2. Personalized self-managed follow-up care, using a digital platform with longitudinal collection of ePROMs to identify

   1. whether patients suffer from organ-specific late effects and/or biopsychosocial late effects

   2. whether this impacts patients' overall HRQoL.

3. Intervention for organ-specific late effects and/or biopsychosocial late effects if needed.

SG patients (n=196) will receive standard follow-up with CT imaging at 12- and 36-months post-surgery, at the surgical departments according to Danish national guidelines. Any local variation/addition to the standard follow-up program will be allowed.

All SG patients will have longitudinal blood samples collected at the same time points as IG patients but only analyzed after the end of the trial, to enable comparison of ctDNA vs. CT imaging as a recurrence predictor. Furthermore, SG patients will receive similar ePROMs as IG patients to collect information at the same time points. However, these will only be analyzed after the end of the trial.

Outcomes and power calculation:

The primary study outcome will be the difference in HRQoL between groups. This will be calculated as the difference in EORTC-QLQ-C30 (global health/QoL domain) between IG and SG at 36 months.

A mean global score of 61 points in the SG is assumed. A score difference of 7 or more between groups will be considered as the minimal clinical important difference (MID). Thus, 170 patients are required in each group to detect an increase of 7 for a mean global score of 68 in the IG with 80% power and 5% significance level. Expecting a drop-out rate of 10%, 189 patients need to be included in each group.

For the secondary objective, OS and RFS, the difference in cumulative RFS and OS between groups will be calculated at 3 and 5 years. Expectantly the cumulative RFS will be 93% in the SG and 92% in the IG. With 196 patients in each group, a decrease in RFS of 9% in IG can be shown with 80% power at a 5% significance level. Thus, the number of included patients is increased to 196 in each group. Non-inferiority will be declared if the difference in RFS is within this limit. The mean OS in the groups is expectantly 77,5%. With 196 patients in each group, an increase of 10% in OS for the IG can be shown with 70% power at a 5% significance level.

Data analyses and statistics:

All data will be presented using descriptive statistics. The ePROM scores will be calculated according to the scoring guidelines for each of the questionnaires. Missing PRO data in form of missing items within an ePROM assessment will be imputed according to the questionnaire's scoring guideline.

The incremental cost-effectiveness ratio (ICER) will be calculated as ∆cost/∆effect. QALY will be calculated as life expectancy x HRQoL, as determined by the Danish value set for EQ-5D-5L. Kaplan-Meier estimates will be used for the estimation of median times to clinical recurrence, disease, or death, and their confidence intervals stratified according to follow-up intensity. The difference in clinical recurrence versus molecular recurrence will be compared using paired t-tests and regression analyses.

Data will be analyzed as intention-to-treat and per-protocol. Interim analyses will be made when 50% of patients in each group have been included, and when 100% in each group have been included and have completed a one-year follow-up.

Quality insurance and ethics:

Data completeness and quality will be monitored by the RESPONSE steering committee, and The General Data Protection Regulation, the Danish Data Protection Act, the Health Act, and the Helsinki II declaration will be complied with unconditionally.

The results of the RESPONSE study are expected to be published in international scientific journals. The reporting will follow the CONSORT guidelines for reporting randomized controlled trials.