aLog-rank tests for comparisons were used.
A, The median years of follow-up for low-intensity imaging surveillance (observed to expected ratio [O:E <1]) were 5 years (interquartile range [IQR], 4.9-5.0) for patients with colon cancer and 5 years (IQR, 5.0-5.0 years) for patients with rectal cancer and for high-intensity imaging surveillance (O:E ≥1) were 5 years (IQR, 4.9-5.0 years) for patients with colon cancer and 5 years (IQR, 5.0-5.0 years) for patients with rectal cancer as derived using reversed Kaplan-Meier method. P = .85 in the comparison of low- vs high-intensity imaging surveillance among patients with rectal cancer and P = .71 among patients with colon cancer.
B, The median years of follow-up for low-intensity carcinoembryonic antigen intensity (CEA) testing surveillance (O:E <1) were 5 years (IQR, 4.8-5.0 years) for patients with colon cancer and 5 years (IQR, 5.0-5.0 years) for patients with rectal cancer and for high-intensity CEA testing surveillance (O:E ≥1) were 5 years (IQR, 5.0-5.0 years) for patients with colon cancer and 5 years (IQR, 5.0-5.0 years) for patients with rectal cancer as derived using reversed Kaplan-Meier method. P = .92 in the comparison of low- vs high-intensity CEA testing surveillance among patients with rectal cancer and P = .96 among patients with colon cancer.
aLog-rank tests were used for comparisons.
A, The median years of follow-up for low-intensity imaging surveillance (observed to expected ratio [O:E <1]) for stage I cancer were 5 years (interquartile range [IQR], 4.6-5.0 years); for stage II, 5 years (IQR, 2.9-5.0 years); and stage III, 5 years (IQR, 1.8-5.0 years) and for high-intensity imaging surveillance (O:E≥1) for stage I cancer were 5 years (IQR, 4.5-5.0 years); stage II, 5 years (IQR, 3.3-5.0 years); and stage III, 5 years (IQR, 1.9-5.0 years), as derived using reversed Kaplan-Meier method. P = .68 in the comparison of low- vs high-intensity imaging surveillance among patients with stage I cancer, P = .41 among patients with stage II cancer, and P = .27 among patients with stage III cancer.
B, The median years of follow-up for low-intensity carcinoembryonic antigen intensity (CEA) testing surveillance (O:E <1) for stage I cancer were 5 years (IQR, 4.7-5.0 years); for stage II, 5 years (IQR, 3.1-5.0 years); and stage III, 5 years (IQR, 1.7-5.0 years) and for high-intensity CEA testing surveillance (O:E≥1) for stage I cancer were 5 years (IQR, 4.5-5.0 years); stage II, 5 years (IQR, 3.2-5.0 years); and stage III, 5 years (IQR, 2.0-5.0 years), as derived using reversed Kaplan-Meier method. P = .16 in the comparison of low- vs high-intensity CEA testing surveillance among patients stage I cancer, P = .84 among patients with stage II cancer, and P = .17 among patients with stage III cancer.
A, The median years of follow-up for low-intensity imaging surveillance (observed to expected ratio [O:E <1]) were 7.8 years (interquartile range [IQR], 7.2-8.4 years) for patients with colon cancer and 7.7 years (IQR, 6.9-8.4 years) for patients with rectal cancer and for high-intensity imaging surveillance (O:E ≥1) were 7.8 years (IQR, 7.1-8.4 years) for patients with colon cancer and 7.7 years (IQR, 7.0-8.4 years) for patients with rectal cancer as derived using reversed Kaplan-Meier method. Survival curves were truncated at 7 years. P = .46 in the comparison of low- vs high-intensity imaging surveillance among patients with rectal cancer and P = .99 among patients with colon cancer.
B, The median years of follow-up for low-intensity carcinoembryonic antigen intensity (CEA) testing surveillance (O:E <1) were 7.8 years (IQR, 7.1-8.4 years) for patients with colon cancer and 7.7 years (IQR, 6.9-8.4 years) for patients with rectal cancer and for high-intensity CEA testing surveillance (O:E ≥1) were 7.8 years (IQR, 7.2-8.4 years) for patients with colon cancer and 7.7 years (IQR, 7.0-8.3 years) for patients with rectal cancer as derived using reversed Kaplan-Meier method. Survival curves were truncated at 7 years. P = .81 in the comparison of low- vs high-intensity CEA testing surveillance among patients with rectal cancer and P = .18 among patients with colon cancer.
eTable 1. Mean observed number of imaging and carcinoembryonic antigen(CEA)tests by stage at primary diagnosis and facility-level surveillance intensity
eTable 2. Recurrence detection, resection rates, and overall survival subgroup analyses by facility-level testing intensity
eFigure 1. Bland-Altman Plot to assess facility-level agreement of intensity measured by observed/expected (O/E) ratio between imaging and carcinoembryonic antigen (CEA) testing among 1175 facilities
Customize your JAMA Network experience by selecting one or more topics from the list below.
Snyder RA, Hu C, Cuddy A, et al. Association Between Intensity of Posttreatment Surveillance Testing and Detection of Recurrence in Patients With Colorectal Cancer. JAMA. 2018;319(20):2104–2115. doi:10.1001/jama.2018.5816
Is there an association between intensity of posttreatment surveillance for stage I, II, or III colorectal cancer and time to detection of cancer recurrence?
In this retrospective cohort study that included 8529 patients with stage I, II, or III colorectal cancer, the median time to recurrence for patients in the high-intensity imaging surveillance group was 15.1 months vs 16.0 months in the low-intensity group and for patients in the high-intensity carcinoembryonic antigen surveillance group was 15.9 months vs 15.3 months in the low-intensity group. Neither difference was statistically significant.
There was no significant association between intensity of surveillance and time to detection of colorectal cancer recurrence.
Surveillance testing is performed after primary treatment for colorectal cancer (CRC), but it is unclear if the intensity of testing decreases time to detection of recurrence or affects patient survival.
To determine if intensity of posttreatment surveillance is associated with time to detection of CRC recurrence, rate of recurrence, resection for recurrence, or overall survival.
Design, Setting, and Participants
A retrospective cohort study of patient data abstracted from the medical record as part of a Commission on Cancer Special Study merged with records from the National Cancer Database. A random sample of patients (n=8529) diagnosed with stage I, II, or III CRC treated at a Commission on Cancer–accredited facilities (2006-2007) with follow-up through December 31, 2014.
Intensity of imaging and carcinoembryonic antigen (CEA) surveillance testing derived empirically at the facility level using the observed to expected ratio for surveillance testing during a 3-year observation period.
Main Outcomes and Measures
The primary outcome was time to detection of CRC recurrence; secondary outcomes included rates of resection for recurrent disease and overall survival.
A total of 8529 patients (49% men; median age, 67 years) at 1175 facilities underwent surveillance imaging and CEA testing within 3 years after their initial CRC treatment. The cohort was distributed by stage as follows: stage I, 25.0%; stage II, 35.2%; and stage III, 39.8%. Patients treated at high-intensity facilities—4188 patients (49.1%) for imaging and 4136 (48.5%) for CEA testing—underwent a mean of 2.9 (95% CI, 2.8-2.9) imaging scans and a mean of 4.3 (95% CI, 4.2-4.4) CEA tests. Patients treated at low-intensity facilities—4341 patients (50.8%) for imaging and 4393 (51.5%) for CEA testing—underwent a mean of 1.6 (95% CI, 1.6-1.7) imaging scans and a mean of 1.6 (95% CI, 1.6-1.7) CEA tests. Imaging and CEA surveillance intensity were not associated with a significant difference in time to detection of cancer recurrence. The median time to detection of recurrence was 15.1 months (IQR, 8.2-26.3) for patients treated at facilities with high-intensity imaging surveillance and 16.0 months (IQR, 7.9-27.2) with low-intensity imaging surveillance (difference, −0.95 months; 95% CI, −2.59 to 0.68; HR, 0.99; 95% CI, 0.90-1.09) and was 15.9 months (IQR, 8.5-27.5) for patients treated at facilities with high-intensity CEA testing and 15.3 months (IQR, 7.9-25.7) with low-intensity CEA testing (difference, 0.59 months; 95% CI, −1.33 to 2.51; HR, 1.00; 95% CI, 0.90-1.11). No significant difference existed in rates of resection for cancer recurrence (HR for imaging, 1.22; 95% CI, 0.99-1.51 and HR for CEA testing, 1.12; 95% CI, 0.91-1.39) or overall survival (HR for imaging, 1.01; 95% CI, 0.94-1.08 and HR for CEA testing, 0.96; 95% CI, 0.89-1.03) among patients treated at facilities with high- vs low-intensity imaging or CEA testing surveillance.
Conclusions and Relevance
Among patients treated for stage I, II, or III CRC, there was no significant association between surveillance intensity and detection of recurrence.
clinicaltrials.gov Identifier: NCT02217865
Colorectal cancer (CRC) is the third most common cancer and second leading cause of cancer death for men and women in the United States, estimated to affect 140 250 patients in 2018.1 Approximately 80% of patients with CRC present with localized disease (American Joint Committee on Cancer [AJCC], stage I, II, or III) for which surgical therapy is curative, and as of 2014, patients with CRC represented the third largest group of cancer survivors.2
After completion of definitive treatment, surveillance is recommended with the goal of improving disease-specific and overall survival by detecting disease recurrence or a second primary cancer early, such that a patient has an opportunity for potentially curative surgery. Additionally, surveillance can be effective in monitoring long-term treatment toxicity, managing patient anxiety, and ensuring continuation of cancer survivorship care.
However, the optimal surveillance strategy is unknown, and recent data from randomized trials have not demonstrated significant survival benefit from intensive follow-up.3,4 Data to inform which testing is most beneficial and how often testing should be performed is limited, and the survival benefit of surveillance in contemporary practice is unknown. National and international consensus guidelines vary, although many still recommend frequent testing, yet adherence to these guidelines is inconsistent, resulting in both overtesting and undertesting in clinical practice.5-11 While the frequency with which intensive testing can identify recurrence early and lead to an intervention that improves outcomes is unknown, recent evidence suggests it would have little clinical effect.
The purpose of this study was to determine the association between surveillance intensity and the detection of CRC recurrence and survival within a large US population outside a clinical trial. The hypothesis was that more intensive surveillance would not be associated with earlier detection of recurrence or better overall survival.
Patients 18 years or older who had been diagnosed with stage I, II or III CRC, as defined by the AJCC, and treated with definitive surgical resection in 2006-2007 were identified from the National Cancer Data Base (Figure 1). Demographic data were obtained from predefined National Cancer Database variables,12 and stage was defined by the seventh edition of the AJCC Cancer Staging Manual.13 A random sample of up to 10 patients with CRC from each Commission on Cancer facility were selected for detailed primary data collection regarding surveillance testing, recurrence, and treatment as part of the commission’s Special Study. Predefined primary data were collected by cancer registrars from a primary chart review of records at the treating facility and physician offices. That data were then merged with corresponding National Cancer Database records. If patients sought care at separate facilities, registrars obtained records from other facilities and outpatient offices. Patient race was included in this study in order to investigate the association between race and surveillance testing intensity. The determination of patient race in this study was derived from predefined National Cancer Database data based on assignment by a Commission on Cancer registrar according to fixed categories.14
The enrollment period (2006-2007) was chosen to ensure a minimum of 5 years of follow-up for recurrence and disease-specific survival. Overall survival was assessed over a total of 7 years. Surveillance and recurrence data were collected through December 2012 with vital status through December 2014.
These data were collected via a secure web form housed at the National Cancer Database and provided to the principal investigator as deidentified data in compliance with Health Insurance Portability and Accountability Act (HIPAA). Study analysis was considered exempt by The MD Anderson Cancer Center Institutional Review Board.
The surveillance start date was defined as 90 days after curative resection. Exclusion criteria included the lack of a surveillance start date (due to recurrence, death, or loss to follow-up within 90 days of surgery), lack of tumor site or stage, and nonsurgical management. Registrars screened patients based on the above criteria. If excluded, a new randomly selected patient was assigned prior to data collection.
To assess surveillance intensity, the observed number of imaging studies—computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography, and carcinoembryonic antigen (CEA) tests—obtained for each patient during the surveillance period was recorded. Complementary imaging studies performed within 30 days, such as a CT scan of the chest plus a CT scan of the abdomen and pelvis or a CT scan of the chest plus an MRI of the abdomen and pelvis were counted as a single study.
An observed to expected ratio (O:E ratio) of testing (separately for imaging and CEA) at both the individual and facility level was used to account for differences in risk-related surveillance. The observed number of tests within 3 years was calculated for each patient. The expected number was then estimated based on fixed effects and posterior means of random effects derived from a 2-level random intercept negative binomial model in predicting the observed number, accounting for not only the random effects from hospitals but also the fixed effects including the variables indicated in Table 1.15
Comparison of the multilevel model to a nonmultilevel model (likelihood-ratio test) indicated a significant clustering effect of testing intensity by facility (P < .001). Therefore, the O:E ratio for each facility was calculated based on the sum of the individuals from that facility. The facility was categorized into high-intensity (O:E ratio ≥1) or low-intensity (O:E <1) categories for comparison. Because patients with documented recurrence undergo more testing during evaluation of recurrence, only those who were alive and disease free for at least 3 years were included for the O:E ratio estimation (n = 6279).
To minimize the influence of physician-level bias and the potential to misclassify confirmatory tests in the setting of suspected recurrence as surveillance tests, patients were assigned a surveillance intensity group (high vs low intensity) based on the facility at which they received care. After high- and low-intensity facility–based groups were empirically defined, patients who died or developed documented recurrence were reintroduced into the analysis and assigned to the intensity group based on their treatment facility. Analyses were then performed at the individual patient level using a facility-based high- vs low-intensity assignment.
The primary outcome of interest was time to detection of recurrence (locoregional or distant) and cumulative recurrence detection rates, confirmed pathologically or clinically. Prespecified secondary outcomes included resection for recurrent distant disease and overall survival.
Post hoc sensitivity analyses for internal validity were performed for (1) inclusion of only patients with stage II or III cancer in the determination of the O:E ratio estimation; (2) facilities with the highest and lowest quartiles of imaging and CEA testing; (3) hospital-level agreement for intensity by imaging vs CEA; and (4) analysis using missing or unavailable data as a separate category.
Baseline characteristics were compared across intensity groups using the χ2 test for categorical variables, t test for means, and Kruskal-Wallis test for medians of continuous variables. Differences in median time to recurrence and confidence intervals were calculated as described by Bonnet-Price.16 Cumulative recurrence rates, resection rates, and overall survival were determined using the Kaplan-Meier method and compared by log-rank tests. Adjusted analyses were performed using multivariable Cox regression, controlling for fixed-effect variables used for intensity estimation. The proportional-hazards assumptions for high vs low intensity was verified graphically using the “log-log” plot, for which the −1n [−1n (survival)] curves of the covariate vs 1n (analysis time) demonstrated reasonable parallel curves. Patients were censored at the time of death, loss to follow-up, or end of surveillance study period (5 years).
Agreement of imaging and CEA intensity was measured in a continuous form using a Bland-Altman plot. The limits of agreement were defined as the mean difference plus or minus a 1.96 standard deviation of the differences.
Two variables, perineural invasion and lymphovascular invasion, had more than 5% missing values. Multiple imputation by chained equations was used to substitute predicted values for missing values with 20 imputed values. Multivariable Cox regression was then performed with results pooled to yield statistical inferences. All tests were 2-sided, with an α of .05; because there was no adjustment for multiple comparisons, statistically significant findings for secondary end points should be interpreted as exploratory. All analyses were performed using SAS software (version 9.1.3; SAS Institute Inc) for data processing and Stata MP (version 13.1; StataCorp) for statistical analyses.
Of the 11 100 patients randomly selected for detailed primary data collection (Figure 1), 8542 had surveillance testing within 3 years after primary resection and were followed up for a median of 5 years (interquartile range [IQR], 3.75-5 years) as surveillance data collection was truncated at 5 years. Of 8542 patients who underwent surveillance testing, 6279 patients remained disease free at 3 years and were eligible for an O:E ratio estimation. The final study cohort included 8529 patients who had undergone surveillance testing and were treated at a facility allowing for an O:E ratio assignment (Figure 1). Demographic data by surveillance intensity are presented in Table 1, and the cancer characteristics are presented in Table 2. The overall survival of the cohort for 5 years was 73.7% and 65.6% for 7 years.
Based on the estimated O:E ratio, 613 facilities were designated low intensity and 562, high intensity for imaging, and 636 facilities were designated low intensity and 539, high intensity based CEA testing frequency. The mean number of imaging tests per patient within 3 years was 1.63 (95% CI, 1.58-1.68) for low intensity and 2.87 (95% CI, 2.79-2.94) for high intensity (mean difference, 1.23; 95% CI, 1.14-1.32; P < .001). The mean number of CEA tests per patient within 3 years was 1.63 (95% CI, 1.55-1.71) for low intensity and 4.31 (95% CI, 4.18-4.44) for high intensity (mean difference, 2.68; 95% CI, 2.52-2.82; P < .001; Table 1). The mean number of tests performed was associated with cancer stage (eTable 1 in the Supplement), but stage was not significantly associated with intensity assignment (Table 1).
There was no significant association between imaging and CEA surveillance intensity and detection of recurrence by multivariable Cox regression adjusting for patient and tumor-related factors (hazard ratio [HR], 0.99; 95% CI, 0.90-1.09 for imaging and 1.00; 95% CI, 0.91-1.11 for CEA; Table 3). Overall 5-year rates of recurrence detection did not significantly differ based on intensity of surveillance imaging (difference, −0.29%; 95% CI, −2.09% to 1.51%) or CEA testing (difference, 0.06%; 95% CI, −1.75% to 1.86%, Table 3).
The median time to any recurrence detection was 15.1 months (interquartile range [IQR], 8.2-26.3 months) for high-intensity imaging vs 16.0 months (IQR, 7.9-27.2 months) for low-intensity imaging (median difference, −0.95; 95% CI, −2.59 to 0.68; Table 3). The median time to any recurrence detection was 15.9 months (IQR, 8.5-27.5 months) for the high-intensity CEA vs 15.3 months (IQR, 7.9-25.7 months) for low-intensity CEA (median difference, 0.59; 95% CI, −1.33 to 2.51; Table 3). The median time to locoregional recurrence detection by imaging was 12.2 months for high-intensity imaging (IQR, 7.6-23.1 months) vs 13.7 months for low-intensity imaging (IQR, 7.2-25.6 months) (median difference, −1.47; 95% CI, −5.12 to 2.17) and was 11.3 months (IQR, 6.6-23.1 months) for high intensity CEA vs 14.1 months (IQR, 8.2-26.2 months) for low-intensity CEA (median difference, −2.76; 95% CI, −5.8 to 0.28). The median time to detection of distant recurrence was 16.0 months (IQR, 8.7-27.6 months) for high-intensity imaging vs 16.6 months (IQR 7.9-27.0 months) for low-intensity imaging (median difference, −0.61; 95% CI, −2.61 to 1.40) and was 17.0 months (IQR, 8.8-29.2 months) for high-intensity CEA vs 15.5 months (IQR, 7.8-26.1) for low-intensity CEA (median difference, 1.45; 95% CI, −0.63 to 3.52).
After stratification by primary tumor site, there was no significant difference in rates of recurrence detection by imaging or CEA surveillance intensity (Figure 2A and B and eTable 2 in the Supplement). In addition, 5-year rates of recurrence detection were also stratified by stage of the primary tumor, and no significant difference was observed based on intensity of imaging or CEA surveillance (eTable 2).
There was no significant difference in the proportion of patients who underwent resection for recurrence at 3 or 5 years by imaging surveillance intensity, for a difference of 0.67% (95% CI, −0.21% to 1.55%) at 3 years and a difference of 0.85% (95% CI, −0.18% to 1.87%) at 5 years nor by CEA intensity for a difference of 0.38% (95% CI, −0.50% to 1.26%) at 3 years and a difference of 0.51% (95% CI, −0.51% to 1.54%) at 5 years (Table 2). When stratified by stage, there was no significant difference in 5-year resection rates among patients with stage II or III CRC who underwent high-intensity vs low-intensity imaging for a difference of 0.89% (95% CI, −0.69% to 2.47%) at 3 years and a difference of 1.33% (95% CI, −0.80% to 3.47%) at 5 years (Figure 3A and eTable 2 in the Supplement). Similarly, high-intensity CEA testing was not associated with significantly higher resection rates between stage II or III patients for a difference of 0.48% (95% CI, −1.10% to 2.06%) at 3 years and a difference of 1.22% (95% CI, −0.91% to 3.36%) at 5 years (Figure 3B and eTable 2). Multivariable Cox regression showed no significant difference by imaging (HR, 1.22; 95% CI, 0.99-1.51) or CEA intensity (HR, 1.12; 95% CI, 0.91-1.39) (Table 3).
In addition, 5- and 7-year overall survival rates did not differ significantly based on imaging or CEA intensity. The overall 5-year survival rates in both imaging intensity groups were 73.7% (difference, −0.06%; 95% CI, −1.95% to 1.82%); the 7-year rates were 65.6% in the high-intensity group and 65.5% in the low-intensity group (difference, 0.04%; 95% CI, −2.03% to 2.12%; Table 3). Similarly, the overall 5-year survival rates were 74.3% in the high-intensity CEA group and 73.1% in the low-intensity group (difference, 1.20%; 95% CI, −0.69% to 3.08%); the 7-year overall survival rates were 66.4% in the high-intensity CEA group and 64.7% in the low-intensity group (difference, 1.71%; 95% CI, −0.37% to 3.78%; Table 3). Furthermore, survival rates did not differ significantly based on imaging or CEA intensity when stratifying by tumor site (Figure 4A-B and eTable 2 in the Supplement).
Multivariable Cox regression failed to demonstrate a significant association between imaging (HR, 1.01; 95% CI, 0.94-1.08) and CEA surveillance intensity (HR, 0.96; 95% CI, 0.89-1.03) and overall survival (Table 3).
Several sensitivity analyses were performed: (1) multivariable analysis with imputation using only stage II and III patients for an O:E ratio estimation (recurrence: HR, 1.06; 95% CI, 0.96-1.17 for imaging and HR, 0.97; 95% CI, 0.88-1.07 for CEA; overall survival: HR, 0.99; 95% CI, 0.93-1.07 for imaging and HR, 0.94; 95% CI, 0.89-1.01 for CEA); (2) comparison of recurrence detection and overall survival between facilities at the highest vs lowest quartiles of testing using multivariable Cox regression with imputation (recurrence: HR, 1.17; 95% CI, 1.02-1.34 for imaging and HR, 1.06; 95% CI, 0.93-1.22 for CEA; overall survival: HR, 1.09; 95% CI, 0.98-1.20 for imaging and HR, 0.99; 95% CI, 0.90-1.10 for CEA); (3) Bland-Altman plot indicated agreement of intensity measured either by imaging or CEA with only 5.9% of the hospitals outside the limits of agreement (eFigure in the Supplement)17; and (4) unknown perineural invasion and LVI modeled in multivariable Cox regression as a separate category (without imputation) (Recurrence: HR, 0.99; 95% CI, 0.90-1.09 for imaging and HR, 1.00; 95% CI, 0.91-1.10 for CEA; overall survival: HR, 1.01; 95% CI, 0.94-1.08 for imaging and HR, 0.95; 95% CI, 0.89-1.02 for CEA).
In this study of a national population of patients with stage I, II, or III CRC, intensity of follow-up testing by imaging or CEA was not associated with time to detection of disease recurrence. Additionally, no significant association was identified between surveillance testing intensity and overall survival.
These findings differ from historic data regarding the association between follow-up testing and survival after curative treatment of localized CRC. In this study, intensity was empirically defined. Test utilization at high-intensity facilities was consistent with current US guidelines for annual CT (2.9 imaging tests within 3 years) but less frequent than guidelines of every 3 to 6 months for CEA tests (4.3 CEA tests within 3 years).5,8,9,18 Low-intensity facilities performed follow-up testing at rates consistent with less intensive follow-up schedules, including at least 1 imaging test and 1 CEA test during the first 3 years of follow-up.6,7,19 Current surveillance recommendations for more intensive follow-up are derived from historical data associating intensive surveillance with an earlier time to detection of recurrence and therefore improved overall but not cancer-specific survival.20-23 However, 2 randomized3,4 trials of high-intensity surveillance vs minimal follow-up failed to demonstrate improved survival, and thus have called these traditional assumptions into question, leaving recent guidelines unable to recommend an optimal strategy.24
The Follow-up After Colorectal Surgery (FACS)3 randomized trial showed a small increase in rates of curative resection with intensive imaging and CEA testing compared with minimal follow-up. However, the absolute difference was only 5%, and combining CEA and imaging did not increase resection rates. Moreover, there was no significant difference in the number of deaths with intensive follow-up. Similarly, the CEAwatch trial,25 found a higher proportion of recurrences amenable to curative resection among patients followed up in an intensive surveillance protocol compared with standard follow-up, although overall recurrence rates in this study were low (7.5%) and overall survival was not assessed.
The Gruppo Italiano Lavoro per la Diagnosi Anticipata (GILDA)4 trial compared intensive testing with semiannual liver ultrasound, annual colonoscopy, and chest imaging with less intensive follow-up with 2 ultrasounds within 16 months and colonoscopy at years 1 and 4 for patients with Duke stage B2 and C CRC. Although recurrence was detected at a mean 5.9 months earlier with intensive surveillance, there was no significant difference in overall survival. Earlier evidence in support of high-intensity testing are likely attributable to early recurrence from occult metastases present at diagnosis, effects mitigated by the marked improvements in preoperative staging over time.3 It is also plausible that the historical benefits from follow-up may stem, at least in part, from regular contact with a clinician, as opposed to earlier detection of recurrent disease. Additional insight from the Assessment of Frequency of Surveillance after Curative Resection in Patients with Stage II and III Colorectal Cancer (COLOFOL)26 trial, comparing the effect of high vs low intensity surveillance on 5-year disease-specific survival and overall survival is highly anticipated.
In the present study, as in recent trials, absolute rates of salvage resection were up to 1.3% higher in facilities with high-intensity imaging surveillance among patients with stage III cancer; however, this small difference is not likely to be clinically meaningful. Moreover, rates of recurrence detection did not differ between high and low intensity, suggesting that recurrences are identified through a combination of imaging, CEA, and symptom-driven evaluation. It is possible that even higher surveillance intensity or higher rates of curative resection for recurrent disease in the high-intensity group could have resulted in a greater effect. This may explain the lack of association of high-intensity CEA testing on rates of salvage surgery, which were lower than those observed with high-intensity CEA testing in the FACS3 or CEAwatch trials.25 Sensitivity analyses comparing recurrence and survival among patients treated at the highest- vs lowest-quartile facilities showed no overall differences in survival by surveillance intensity. While there was a small observed association with recurrence detection when comparing the highest vs lowest quartile for imaging intensity, this was not associated with significant differences in survival.
One explanation for the lack of overall survival effect by surveillance intensity is the low event rate of salvage surgery observed in the present study and in recently reported trials. This study may be underpowered to detect a difference in overall survival based on resection rates, although the absolute difference in overall survival would be expected to be small. Furthermore, low intensity was not equivalent to no follow-up, but rather less follow-up during the first 3 years, when the majority of recurrences occur. It is also possible that patients at highest risk of recurrence (eg, stage IIIC) might receive more intensive surveillance but may also have recurrences not amenable to curative resection. While it is generally accepted that salvage surgery can improve survival, the risk of disease relapse remains high, and the benefit may be limited for some patients. Thus, the value of earlier detection without survival effect is likely to be dependent on the personal values and preferences of the patient, highlighting the importance of individualizing surveillance plans. Surveillance testing is not without potential harm, including false-positive results leading to unnecessary tests or procedures, radiation exposure, contrast toxicity, and other testing-related complications.
Based on these data and the recent FACS trial, current National Comprehensive Cancer Network guideline recommendations (CT testing every 6 months for 3 years) could be considered overtesting given the absence of improvement in recurrence detection or survival. Moreover, these data suggest that the recommendation of 2 CT scans in the first 3 years and CEA testing every 6 months in the first 3 years by the National Institute for Health and Care Excellence in the United Kingdom are appropriate.19
This study has several limitations. First, this was a retrospective cohort study, and patients were not randomized, nor could individual decision making driving testing frequency be accounted for. Surveillance intensity was assigned based on treatment facility while adjusting for patient factors, but clustering by practice facility helped to control for practice facility-level variation. Nevertheless, there could have been additional within-facility variation. Second, because data were collected on up to 10 patients for each facility regardless of volume, it is possible that low-volume facilities could be overrepresented. Third, the cohort was assembled in 2006-2007 and therefore may not reflect contemporary practice. However, patients were selected from 2006 to 2007 to allow at least 5 years of follow-up data regarding surveillance and recurrence and were followed up through 2014 for vital status. Although there have been advances in systemic therapy for metastatic disease over the study period, without significant differences in the detection and treatment of recurrence, these advances similarly affect both the low-intensity and high-intensity cohorts, and therefore do not affect the primary findings of this study. Fourth, retrospectively collected data are dependent on the quality of data abstraction. It is possible that if patients completed follow-up or sought treatment at another institution, data loss could occur. Registrars were asked to follow-up with outside institutions in an effort to try to ensure data completeness, but actual data completeness was not measured. Fifth, the empirical definition of high-intensity surveillance in this study included less intensive follow-up than is practiced in some centers, for example, and a benefit of even higher intensity could not be excluded. However, the observed surveillance testing is reflective of contemporary community practice, and therefore provides relevant data for comparison.
Among patients treated for stage I, II, or III CRC, there was no significant association between surveillance intensity and detection of recurrence.
Corresponding Author: George J. Chang, MD, MS, Departments of Surgical Oncology and Health Services Research, The University of Texas MD Anderson Cancer Center, 1400 Pressler St, FCT 17.5060, Houston, TX 77030-4008 (firstname.lastname@example.org).
Accepted for Publication: April 23, 2018.
Author Contributions: Dr Chang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Snyder, Hu, Cuddy, Schumacher, You, Kozower, Greenberg, Schrag, McKellar, Winchester, Chang.
Acquisition, analysis, or interpretation of data: Snyder, Hu, Cuddy, Francescatti, Schumacher, Van Loon, You, Greenberg, Venook, Winchester, Chang.
Drafting of the manuscript: Snyder, Hu, Cuddy, Winchester, Chang.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Hu, Kozower, Schrag, Chang.
Obtained funding: Schrag, Chang.
Administrative, technical, or material support: Cuddy, Francescatti, Schumacher, You, Schrag, Venook, McKellar, Winchester.
Supervision: McKellar, Chang.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Chang reported that he has received support from Johnson & Johnson. No other disclosures were reported.
Funding/Support: Research reported in this publication was supported by award CE13-04-6855 from the Patient-Centered Outcomes Research Institute (PCORI) (Dr Chang), and grant P30CA016672 from the National Cancer Institute of the National Institutes of Health to the University of Texas, MD Anderson Cancer Center Support Grant, and U10CA180821 from the National Cancer Institute to the Alliance for Clinical Trials in Oncology.
Role of the Funder/Sponsor: The Patient-Centered Outcomes Research Institute (PCORI), National Cancer Institute (NCI), and Alliance for Clinical Trials in Oncology had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Alliance for Clinical Trials in Oncology Network Cancer Surveillance Optimization Working Group: Julia Berian, MD, MS, University of Chicago; Monica Bertagnolli, MD, Dana-Farber Cancer Institute; Ronald Chen, MD, MPH, University of North Carolina; Stephen Edge, MD, Roswell Park Comprehensive Cancer Center, Buffalo, NY; Patrick Gavin, RPh, Alliance for Clinical Trials in Oncology Network; Greer Gay, American College of Surgeons; Kay Kays, Alliance for Clinical Trials in Oncology Network, Lisa Marie Lowenstein, PhD, MPH, RD, MD, MD Anderson Cancer Center; Joellen Luca, Alliance for Clinical Trials in Oncology Network; Ryan McCabe, American College of Surgeons; Heidi Nelson, MD, Mayo Clinic; Heather Neuman, MD, MS, University of Wisconsin; Walter Peters, MD, Baylor Scott & White Medical Center; Kathryn Ruddy, MD, MPH, Mayo Clinic; Dan Sargent, PhD, Mayo Clinic; Patty Spears, University of North Carolina; Kathleen Thoburn, American College of Surgeons; and Robert Volk, PhD, Anderson Cancer Center.
Disclaimer: All statements in this publication, including, its findings, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute (PCORI), its Board of Governors or Methodology Committee, or the National Institutes of Health. JAMA Associate Editor Deborah Schrag was not involved in the review of or decision to publish this article.
Create a personal account or sign in to: