If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Department of Surgical Oncology and Gastrointestinal Surgery Erasmus MC Cancer Institute Doctor Molewaterplein 40, 3015 GD Rotterdam, the Netherlands. Tel.: +31 10 7042125.
Intensive follow-up after surgery for colorectal and breast cancer has little impact on survival.
•
Colorectal cancer recurrences are more often treated locally after intensive follow-up.
•
High quality research on follow-up for other cancers is scarce.
Abstract
Background
This meta-analysis aimed to evaluate the effectiveness of intensive follow-up after curative intent treatment for five common solid tumours, in terms of survival and treatment of recurrences.
Methods
A systematic literature search was conducted, identifying comparative studies on follow-up for colorectal, lung, breast, upper gastro-intestinal and prostate cancer. Outcomes of interest were overall survival (OS), cancer specific survival (CSS), and treatment of recurrences. Random effects meta-analyses were conducted, with particular focus on studies at low risk of bias.
Results
Fourteen out of 63 studies were considered to be at low risk of bias (8 colorectal, 4 breast, 0 lung, 1 upper gastro-intestinal, 1 prostate). These studies showed no significant impact of intensive follow-up on OS (hazard ratio, 95% confidence interval) for colorectal (0.99; 0.92–1.06), breast 1.06 (0.92–1.23), upper gastro-intestinal (0.78; 0.51–1.19) and prostate cancer (1.00; 0.86–1.16). No impact on CSS (hazard ratio, 95% confidence interval) was found for colorectal cancer (0.94; 0.77–1.16). CSS was not reported for other cancer types. Intensive follow-up increased the rate of curative treatment (relative risk; 95% confidence interval) for colorectal cancer recurrences (1.30; 1.05–1.61), but not for upper gastro-intestinal cancer recurrences (0.92; 0.47–1.81). For the other cancer types, no data on treatment of recurrences was available in low risk studies.
Conclusion
For colorectal and breast cancer, high quality studies do not suggest an impact of intensive follow-up strategies on survival. Colorectal cancer recurrences are more often treated locally after intensive follow-up. For other cancer types evaluated, limited high quality research on follow-up is available.
Most cancer survivors receive regular follow-up care after being treated with curative intent. Traditionally, follow-up is performed for a period of 5 years or longer for most types of solid tumours. Guidelines differ between tumour types, but generally advocate regular hospital visits, imaging, and serum tumour marker measurements when available [
The main rationale behind oncologic follow-up is to detect metastases or novel primary tumours early, since prompt treatment of cancer relapses is deemed important for the likelihood of cure and survival. Next to this, follow-up can be used to address patients’ needs with regards to psychosocial counselling, to evaluate treatment effects and complications, and to inform patients on their disease status and risk of recurrence [
The debate surrounding oncological follow-up practices has existed for many years. It is associated with a considerable use of hospital resources and costs, may have impact on quality of life, while the effect of follow-up intensity on survival outcomes remains equivocal [
], improvements of follow-up practices should be pursued. Many studies evaluate the effectiveness of follow-up for individual tumours types, but a broad oncological perspective remains lacking.
We therefore sought to systematically assess and meta-analyse available literature on follow-up after curative intent treatment for five types of solid tumours (colorectal, lung, breast, upper gastro-intestinal, and prostate cancers) in order to determine the impact of different follow-up strategies on survival outcomes and treatment of recurrent disease.
2. Methods
2.1 Search strategy
This study was performed in line with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis, www.prisma-statement.org) guidelines. Pubmed/MEDLINE, Embase, Web of Science, the Cochrane database, and Google Scholar were systematically searched for studies published prior to the 12th of May 2020. The search terms are provided in supplementary table A.1. Reference lists from eligible articles were also reviewed to identify additional publi-cations.
2.2 Study selection
Screening for eligible studies was performed by two authors (BG, DH), independently. Studies were included when comparing follow-up approaches after curative intent treatment for colorectal, lung, breast, upper gastro-intestinal or prostate cancer, in light of overall survival (OS) or cancer specific survival (CSS) outcomes. Treatment intent for recurrent disease (i.e. curative or palliative) was also evaluated. Both randomised controlled trials (RCTs) and observational studies (cohort and case–control) were considered. Inclusion was restricted to articles written in English. Non-original studies (e.g. reviews, editorials) were excluded, as were non-comparative studies and studies using simulation techniques (e.g. Markov modelling).
2.3 Data extraction and presentation
Data were extracted by two reviewers (BG, DH), independently. Studies were categorised based on the aspect of follow-up evaluated, being the frequency of testing, setting of follow-up (e.g. in-hospital or general practitioner), diagnostic modalities used, or a combination of the aforementioned categories. Data on survival (hazard ratios (HR) including 95% confidence intervals (95% CI) for OS and CSS) and the probability of treatment with curative intent for recurrent disease (relative risk (RR) including 95% CI) were collected. When no ratios were reported, data were extracted from Kaplan–Meier figures, tables, and text. Multi-layered circle plots were created to visualise all aspects in relation to outcomes and the risk of bias.
2.4 Quality assessment
Quality assessment was performed by two reviewers (BG, DH), independently. The Cochrane tools ROBINS-I (for observational studies) and RoB2 (for randomised studies) were used [
]. Studies were considered to be at low risk of bias when qualified as either ‘low’ to ‘moderate’ using ROBINS-I, or as ‘low risk’ to ‘some concerns’ using RoB2.
2.5 Quantitative assessment
A random effects meta-analysis was conducted per tumour type and stratified for study risk of bias, using the generic inverse variance method (survival) or the Mantel-Haenszel method (treatment of recurrences). Methods described by Tierney et al. were applied to calculate log HRs and corresponding standard errors, in case these were not reported [
]. Both HRs and RRs were reported using the least intensive approach (e.g. lowest frequency, non-hospital setting) as a reference. In studies with multiple groups (i.e. >2 follow-up approaches), the most intensive approaches were combined to create a single pair-wise comparison with the least intensive approach, as recommended by the Cochrane Handbook [
]. The R Project for Statistical Computing version 4.1.0 (https://www.r-project.org/) was used for both the statistical analyses and visualisation of the data (packages: meta (v4.18-1), ggplot2 (v3.3.2); circlize (v0.4.11) [
The screening and selection process is illustrated in Fig. 1. After screening 4538 studies, 167 were screened full-text. Ultimately, 63 studies were deemed eligible for inclusion [
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
The association between post-treatment surveillance testing and survival in stage II and III colon cancer patients: an observational comparative effectiveness study.
Planned FDG PET-CT scan in follow-up detects disease progression in patients with locally advanced NSCLC receiving curative chemoradiotherapy earlier than standard CT.
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. The GIVIO Investigators.
Phase II randomized clinical trial of endosonography and PET/CT versus clinical assessment only for follow-up after surgery for upper gastrointestinal cancer (EUFURO study).
The relationship of the intensity of posttreatment prostate-specific antigen surveillance and prostate cancer outcomes: results from a population-based cohort.
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
The association between post-treatment surveillance testing and survival in stage II and III colon cancer patients: an observational comparative effectiveness study.
Planned FDG PET-CT scan in follow-up detects disease progression in patients with locally advanced NSCLC receiving curative chemoradiotherapy earlier than standard CT.
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. The GIVIO Investigators.
Phase II randomized clinical trial of endosonography and PET/CT versus clinical assessment only for follow-up after surgery for upper gastrointestinal cancer (EUFURO study).
The relationship of the intensity of posttreatment prostate-specific antigen surveillance and prostate cancer outcomes: results from a population-based cohort.
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
The association between post-treatment surveillance testing and survival in stage II and III colon cancer patients: an observational comparative effectiveness study.
Phase II randomized clinical trial of endosonography and PET/CT versus clinical assessment only for follow-up after surgery for upper gastrointestinal cancer (EUFURO study).
The relationship of the intensity of posttreatment prostate-specific antigen surveillance and prostate cancer outcomes: results from a population-based cohort.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. The GIVIO Investigators.
Phase II randomized clinical trial of endosonography and PET/CT versus clinical assessment only for follow-up after surgery for upper gastrointestinal cancer (EUFURO study).
]. Table 1, Table 2 provide detailed overviews of the low and high risk of bias studies. Risk of bias assessment is provided in supplementary table A.2. Fig. 3 visualises outcomes per study. The results of the meta-analysis per tumour type, including stratified analyses are reported in Table 3.
Fig. 2Multi-layered circle plot displaying all 63 included studies by cancer type (inner circle), aspect of follow-up investigated (middle circle), and study design (outer circle).
Single chest, abdomen, and pelvis CT scan at 12–18 months if requested at study entry by hospital clinician; ∗∗ Only performed in patients at high risk for metachronous lesions (hereditary cancer, synchronous colorectal neoplasms); ∗∗∗ Classification of follow-up facility based. Expected number of scans/CEA measurements calculated through 2-level random intercept negative binomial model. High frequency facility defined as observed:expected ratio ≥1.
CT-only: CT chest, abd, pelvis every 6 m Y1-2, every 12 m Y3-5 CT + CEA: protocols combined
Single chest, abdomen, and pelvis CT scan at 12–18 months if requested at study entry by hospital clinician; ∗∗ Only performed in patients at high risk for metachronous lesions (hereditary cancer, synchronous colorectal neoplasms); ∗∗∗ Classification of follow-up facility based. Expected number of scans/CEA measurements calculated through 2-level random intercept negative binomial model. High frequency facility defined as observed:expected ratio ≥1.
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Clin exam/CEA and CA 19-9: every 4 m Y1-2, every 6 m Y3-4, every 12 m Y5 CXR/colonoscopy: every 12 m Y1-5 Liver US: @ 4, 8, 12, 16, 24, 16, 48 and 60 m
Clin exam/CEA: every 4 m Y1-2, every 6 m Y3-4, every 12 m Y5 Colonoscopy: @ 12 and 48 m Liver US: @ 4 and 16 m
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. The GIVIO Investigators.
Phase II randomized clinical trial of endosonography and PET/CT versus clinical assessment only for follow-up after surgery for upper gastrointestinal cancer (EUFURO study).
The relationship of the intensity of posttreatment prostate-specific antigen surveillance and prostate cancer outcomes: results from a population-based cohort.
Abd = abdominal, CA 19-9 = carbohydrate antigen 19-9, CEA = carcinoembryonic antigen, clin exam = clinical examination, CRC = colorectal cancer, CT = computed tomography, CXR = chest x-ray, GOJ = gastro-oesophageal junction, m = months, MRI = magnetic resonance imaging, NR = not reported, NSCLC = non-small cell lung cancer, PET-CT = positron emission tomography – computed tomography, PSA = prostate specific antigen, US = ultrasound, w = weeks, y = years.
a Single chest, abdomen, and pelvis CT scan at 12–18 months if requested at study entry by hospital clinician; ∗∗ Only performed in patients at high risk for metachronous lesions (hereditary cancer, synchronous colorectal neoplasms); ∗∗∗ Classification of follow-up facility based. Expected number of scans/CEA measurements calculated through 2-level random intercept negative binomial model. High frequency facility defined as observed:expected ratio ≥1.
The association between post-treatment surveillance testing and survival in stage II and III colon cancer patients: an observational comparative effectiveness study.
High risk patients Clin exam/CEA: every 3 m Y1-2, every 4 m Y3, every 6 m Y4-5 Abd and pelvic US: every 6 m 1-3Y, every 12 m Y4-5 CXR/rectosigmoidoscopy
For rectal and sigmoid cancer patients only. ∗∗ Only in patients with symptoms possibly related to bone metastases.
: every 12 m Y1-5 Low risk patients Clin exam/CEA: every 6 m Y1-2, every 12 m Y3-5 Abd and pelvic US: every 6 m 1-2Y, every 12 m Y3-5 CXR: every 12 m Y1-5 Rectosigmoidoscopy
Clin exam/CEA: every 3 m Y1-2, every 6 m Y3-5 CXR + abd US/CT + colonoscopy: every 6 m Y1-5 (alternating per 6 m) Rectoscopy @ 3, 6 and 9 months or Non-standardised follow-up in hospital or at GP
Follow-up within clinical trial Clin exam/blood analysis/CXR: every 2/3 m Y1-2, every 6 m thereafter CT chest and upper abd/MRI brain: every 6 m Y1-3, every 12 m thereafter
Non-trial follow-up Clin exam/blood analysis/CXR: every 3 m Y1-2/3, every 6 m Y3/4-5
Planned FDG PET-CT scan in follow-up detects disease progression in patients with locally advanced NSCLC receiving curative chemoradiotherapy earlier than standard CT.
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
WeChat app-based education and rehabilitation program - including disease related education (once a week for 12w), rehabilitation exercise guidance (once a week for 40w), daily activity supervision (once a week for 12 m), and psychosocial support (every 2 weeks for 12 m)
Intensive follow-up, any of the following criteria: ≥4 visits and/or blood tests and/or CXR per year, ≥1 CT per year, or any bronchoscopy and/or sputum cytology
Non-intensive follow-up (none of the criteria met)
Fig. 3Multi-layered circle plot summarising the reported effect in all 63 included studies, stratified by risk of bias. The inner circle represents cancer type, the middle circle study design, and the outer circle the effect of the intervention on overall or cancer specific survival, and treatment intent.
CI = confidence interval, CSS = cancer specific survival, HR = hazard ratio, N = number of studies reporting on outcome, NA = not applicable, OS = overall survival, RR = relative risk, Upper GI = upper gastrointestinal.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
The 33 colorectal cancer studies comprised 50,431 patients in total (Table 1, Table 2). Across all studies, intensive follow-up led to improved OS (HR 0.82, 95% CI 0.73–0.91) and an increased probability of curative intent treatment for recurrences (RR 1.60, 95% CI 1.21–2.11). An equally large, but non-significant, impact on CSS (HR 0.80, 95% CI 0.63–1.01) was observed. Considerable heterogeneity was present (I2 66–85% for the three outcomes) (Table 3).
In the eight studies (24%) considered to be at low risk of bias, including seven RCTs [
Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial.
Effect of more vs less frequent follow-up testing on overall and colorectal cancer-specific mortality in patients with stage II or III colorectal cancer: the COLOFOL randomized clinical trial.
], no significant impact on OS (HR 0.99, 95% CI 0.92–1.06) and CSS (0.94, 95% CI 0.77–1.16) was observed with little to no heterogeneity (I2 7% and 0%). All low risk studies evaluated frequency of follow-up, of which three evaluated a symptom-based approach without use of diagnostics [
]. Although survival was not significantly impacted by follow-up strategy, intensive follow-up remained significantly associated with the probability of curative intent treatment for recurrences (RR 1.30, 95% CI 1.05–1.61) in low risk studies (I2 28%).
Twenty-five studies (76%) were deemed to be at high risk of bias, the majority being observational (N = 16, 64%) [
The association between post-treatment surveillance testing and survival in stage II and III colon cancer patients: an observational comparative effectiveness study.
]. Pooled effect estimates in high risk colorectal cancer studies were larger for all outcomes evaluated, with considerable heterogeneity (Table 3).
3.3 Lung cancer
Within the thirteen lung cancer studies, 26,162 patients were included (Table 2). All of the studies identified were considered to be at high risk of bias, including two RCTs [
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
Planned FDG PET-CT scan in follow-up detects disease progression in patients with locally advanced NSCLC receiving curative chemoradiotherapy earlier than standard CT.
The impact of WeChat app-based education and rehabilitation program on anxiety, depression, quality of life, loss of follow-up and survival in non-small cell lung cancer patients who underwent surgical resection.
]. Follow-up did not significantly impact OS (HR 0.94, 95% CI 0.84–1.05) (Table 3). Heterogeneity was moderate (I2 49%). Only one study reported on CSS, in which no significant survival difference was obtained [
In total 10,585 breast cancer patients were included in eleven studies (Table 1, Table 2). Across all studies no significant impact of intensive follow-up on OS (HR 0.80, 95% CI 0.54–1.18) or CSS (HR 0.52, 95% CI 0.27–1.02) was observed (Table 3). Heterogeneity was considerable for both outcomes (I2 92% and 94%). None of the studies reported on the (curative) treatment for local recurrence or metastatic disease.
Four studies (36%) were considered to be at low risk of bias, including two RCTs [
Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer follow-up.
Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. The GIVIO Investigators.
Intensive vs clinical follow-up after treatment of primary breast cancer: 10-year update of a randomized trial. National Research Council Project on Breast Cancer Follow-up.
]. All low risk studies compared different modalities used in the same frequency, generally every 3–6 months. When pooling the effects of individual studies, no impact on OS was observed (pooled HR 1.06, 95% CI 0.92–1.23), with no heterogeneity (I2 0%). None observed significant additional value of using multiple diagnostics (e.g. liver ultrasonography, chest radiography, laboratory tests) next to clinical examinations and mammography's. None of the studies reported on CSS.
Most of the seven high risk studies were observational (71%) [
]. In contrast to the low risk studies, all of the observational studies evaluated the frequency of follow-up, while the RCTs evaluated setting of follow-up. The randomised studies found no impact on OS when follow-up was performed by the family physician (HR 1.05, 95% CI 0.60–1.84) or the nurse practitioner (HR 1.22, 95% CI 0.58–2.57), compared to the standard hospital-based physician-led approach [
]. The studies evaluating the frequency of follow-up all assessed the impact of receiving one or more diagnostic evaluations (i.e. mammography or multiple diagnostics) to a nihilistic approach [