The fraction of patients who are cancer-free survivors 5 years after curative-intended surgery for colorectal cancer (CRC) is increasing, suggesting that extending surveillance beyond 5 years may be indicated. Here we estimate the incidence of late recurrence, metachronous CRC, and second primary cancers 5–10 years postoperative. All patients resected for UICC stage I–III CRC in Denmark through 2004–2013 were identified. Through individual-level linkage of nationwide health registry data, recurrence status was determined using a validated algorithm. Cancer-free survivors 5 years after surgery, were included. Cumulative incidence functions (CIF) of late recurrence, metachronous CRC, and second primary cancer 5–10 years postoperative were constructed. Subdistribution hazards ratios (sHR) were computed using Fine-Gray regression. Among 8883 patients, 370 developed late recurrence (5–10-year CIF = 4.1%, 95%CI: 3.7%–4.6%), 270 metachronous CRC (5–10-year CIF = 3.0%, 95%CI: 2.7%–3.4%), and 635 a second primary cancer (5–10-year CIF = 7.2%, 95%CI: 6.7%–7.7%). The risk of late recurrence was reduced for patients operated in 2009–2013 compared to 2004–2008 (2.9% vs. 5.6%, sHR = 0.52, 95% CI: 0.42–0.65). The risk of metachronous CRC was likewise reduced from 4.1% to 2.1% (sHR = 0.50, 95%CI: 0.39–0.65). While the risk of second primary cancer did not change between 2009–2013 and 2004–2008 (7.1% vs. 7.1%, sHR = 0.98, 95% CI: 0.84–1.15). Using nation-wide 10-year follow-up data, we document that the incidences of late recurrence and metachronous CRC are low and decreasing from 2004 to 2013. Thus, despite increasing numbers of long-term cancer survivors, the data do not advocate for extending CRC-specific surveillance beyond 5 years.

Graphical Abstract

What's new?

Recent advancements in colorectal cancer (CRC) management have reduced the risk of late recurrence and metachronous CRC 5–10 years after curative surgery for non-metastatic CRC—a timeframe with no surveillance according to current guidelines. Despite more patients are cancer-free survivors at the 5-year mark, late events remain uncommon. Our study does not support extending current surveillance protocols, as late recurrence or metachronous CRC don't impact more patients than late second non-CRC primary cancers.

Abbreviations

ASCO: American Society of Clinical Oncology
CIF: cumulative incidence function
CRC: colorectal cancer
DCCG: Danish Colorectal Cancer Group
DCR: Danish Cancer Registry
DNPR: Danish National Patient Registry
DPR: Danish Pathology Registry
ESMO: European Society for Medical Oncology
ICD-10: International Classification of Diseases 10th Revision
NMSC: non-melanoma skin cancer
OS: overall survival
sHR: subdistribution hazard ratio
SNOMED: systematized nomenclature of medicine
STROBE: strengthening the reporting of observational studies in epidemiology
UICC: Union for International Cancer Control
95%CI: 95% interval of confidence

1 INTRODUCTION

Colorectal cancer (CRC) is ranked third in global incidence and second in cancer-specific mortality worldwide.¹ Death from CRC is primarily associated with disease recurrence. Despite curative-intended treatment, every fifth patient with stage I–III disease will develop recurrence with >80% of the recurrences diagnosed within 3 years of primary surgery.^{2, 3} Correspondingly, international guidelines recommend 5 years of postoperative surveillance to diagnose recurrence and metachronous CRC.^{4, 5}

Over the last two decades, many countries have introduced different initiatives to reduce CRC recurrence and improve survival.⁶ Improvements in recurrence and survival within 5 years following primary treatment has been reported following the implementation of these initiatives.³ Furthermore, more patients are diagnosed with CRC at an earlier disease stage,⁷ and the life expectancy in the general population is increasing.⁸ Consequently, more patients are recurrence-free survivors at the end of standardized 5-year surveillance. Hence a higher proportion of CRC patients become at risk of late recurrence as well as metachronous CRC and other second primary cancers during a timeframe with no surveillance. Previous studies have reported late CRC recurrence (here defined as recurrence >5 years after primary resection) to occur in 1%–6% of patients treated from 1985 to 2006.^9-11 However, new and preferably nationwide studies are needed to assess the risk of late recurrence following modern treatment of CRC. Also, the burden of late metachronous CRC and second primary cancers should be taken into account when considering prolonged surveillance specific for CRC late recurrence. If the incidence of late recurrence does not exceed that of second primary cancers, increased health costs to prolong surveillance specific for CRC recurrence may not be justified.

The aim of this study is to determine the incidence of late recurrence, metachronous CRC, and second primary cancer in a nationwide cohort of stage I-III CRC patients who underwent surgical resection with curative intent. Late recurrence, metachronous CRC and second primary cancer was defined as events occurring after completed postoperative surveillance 5 years after surgery. Furthermore, we aimed to describe the prognosis following diagnosis of late recurrence, metachronous CRC and second primary cancer with late recurrence as reference.

2 MATERIALS AND METHODS

2.1 Design and study population

This study was conducted as a nationwide cohort study and reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.¹² All Danish patients younger than 80 years undergoing surgical resection of first-time Union for International Cancer Control (UICC) TNM stage I–III CRC between January 2004 and December 2013 were eligible for inclusion. Patients were identified from the Danish Colorectal Cancer Group (DCCG) database.^{13, 14}

The cohort included all patients at risk of first time recurrence, metachronous CRC, and second primary cancers 5 years after surgery. The cohort was followed until 10 years postoperative or January 1, 2023, whichever came first.

2.2 Data sources and identification of recurrence

Individual-level data were obtained from Danish health care registries¹⁵ including the Danish Cancer Registry (DCR),¹⁶ the Danish National Patient Registry (DNPR),¹⁷ and the Danish Pathology Registry (DPR).¹⁸ Data records were linked using the unique 10-digit civil registration number assigned by the Danish Civil Registration System¹⁹ to all Danish citizens.

Recurrence status was determined using a validated algorithm previously described in detail.^{20, 21} The algorithm defined recurrence as at least one of the following four indicators:

International Classification of Diseases 10th Revision (ICD-10) codes of metastasis (ICD-10 DC76-DC80) in DNPR or DCR;
ICD-10 codes of cytostatic treatment (BWHA1-2, BOHJ17 or BOHJ19B1) registered by an oncological department in DNPR 60 or more days from last cytostatic therapy code;
Systematized Nomenclature of Medicine (SNOMED) combinations in DPR; or
ICD-10 code specific to CRC local recurrence in DNPR (DC189X, DC209X or DC991).

The date of recurrence was defined as the earliest date on which at least one of the four indicators of the algorithm were met, and all other codes meeting an indication within the following 60 days were classified as belonging to the same recurrence event, acknowledging that there is some delay in getting the diagnostic workup of the recurrence registered in the national registries.

2.3 Definition of metachronous CRC

Metachronous CRC was defined as at least one of the following indicators:

An ICD-10 code of CRC (DC18 or DC20) registered in the DCR;
Recurrence as defined by the algorithm followed by an ICD-10 code of CRC (DC18 or DC20) registered in DCR within 60 days from date of recurrence (meaning the recurrence was reclassified from CRC recurrence to metachronous CRC); or
A mapping of the remaining bowel at time of recurrence for each patient was constructed from surgical procedure codes in the DNPR (available from 1996 and onwards, Table S1). Recurrence as defined by the algorithm was reclassified from CRC recurrence to metachronous CRC in the following scenarios based on SNOMED:
1. If specimens registered in the DPR contained adenocarcinoma with SNOMED topography code corresponding to colon or rectum (without P30755; “supplementary analysis on previously assessed material”) in a remaining bowel segment at time of recurrence, or
2. If specimens registered in the DPR contained adenocarcinoma with SNOMED topography code corresponding to colon or rectum (without P30755) within 2 weeks after a procedure involving polypectomy, biopsy or mucosal resection.

2.4 Definition of second primary cancer

Second primary cancer was defined as ICD-10 code other than CRC (DC18 or DC20) or non-melanoma skin cancer (NMSC; DC44) registered in the DCR. Furthermore, recurrence as defined by the algorithm (other than DC189X, DC209X or DC991) was reclassified from CRC recurrence to secondary primary cancer if a secondary primary cancer was registered in DCR within 60 days after the date of recurrence.

2.5 Studied covariates

Eligible patients were categorized by sex, age-group at surgery or event (<50, 50–59, 60–69, 70–79 or ≥80 years old), UICC stage of the primary CRC (I, II or III according to UICC TNM classification 5th edition²²), Charlson comorbidity score (0, 1, 2 or ≥3),²³ location of primary CRC tumor (colon and rectum), and surgical priority (elective or emergency). Time of primary treatment was grouped in two 5-year calendar periods (2004–2008 and 2009–2013).

2.6 Statistical method

Patients were followed from 5 years after primary surgery until date of recurrence (event), metachronous CRC (event), second primary cancer other than NMSC (event), death (competing event), emigration, or end of follow-up (January 1st 2023) whichever came first.

Descriptive characteristics were presented for all patients at risk 5 years after primary surgery stratified by calendar period of CRC surgery.

Risk of late recurrence, metachronous CRC, and second primary cancer were presented as a 5–10-year cumulative incidence function (CIF), treating death as a competing event. Cumulative incidence curves were constructed using the Aalen-Johansen estimator for visualization. The association between calendar period of primary treatment and the risk of late recurrence, metachronous CRC and second primary cancer was estimated using the Fine-Gray competing risk method treating death as a competing event, and was reported as subdistribution hazard ratios (sHR) with 95% interval of confidence (95%CI).²⁴ Both a univariate analysis and a multivariable analysis adjusted for all covariates were calculated using sHR.

To evaluate the overall survival (OS) following late events (recurrence, metachronous CRC, and second primary cancer), the patients who developed late recurrence, metachronous CRC or second primary cancer between 5 and 10 years postoperative were followed from the time of event and until death, 5 years, or end of follow-up (January 1, 2023) whichever came first. The OS was plotted for visualization using the Kaplan–Meier method. Cox proportional hazards regression was used to compare mortality following metachronous CRC and second primary cancer with late recurrence as reference. These analyses were adjusted for age at event, sex and comorbidity at event.

Analyses were performed using R version 4.3.1 (https://www.r-project.org/) between January and October 2023. No hypothesis testing was executed in this observational study.

3 RESULTS

We identified 20,012 patients undergoing surgery for first-time UICC TNM stage I–III CRC between 2004 and 2013 in the DCCG database. Of these, 8883 were alive and event-free 5 years after surgery, and therefore at risk of a late event, Figure 1. Characteristics of the cohort are provided in Table 1.

Details are in the caption following the image — **FIGURE 1**
Open in figure viewer PowerPoint

Flowchart of cohort selection from patients aged <80 years undergoing intended curative surgery for UICC stage I to III Colorectal Cancer in Denmark from 2004 to 2013. Patients were excluded if they had a history of CRC or unspecified cancer more than 180 days before CRC diagnosis in DCCG, cancers other than CRC or non-melanoma skin cancer (NMSC) any time prior to primary CRC surgery, and if they emigrated, died, or were diagnosed with a new primary cancer other than CRC or NMSC, within 180 days from the index CRC surgery. Patients with recurrence, metachronous CRC, secondary cancer or who died between 180 days and 5 years after surgery were excluded prior to analysis of the patients at risk of outcomes 5 years after surgery. *Patients may be excluded by more than one criterion. Categories with <5 are reported in aggregate to reduce identification of individuals in the data. DCCG, Danish Colorectal Cancer Group; CRC, colorectal cancer.

TABLE 1. Patient characteristics of the study population stratified by calendar period of curative CRC surgery.

Variable	Overall N = 8883^a	Calendar period of CRC surgery
Variable	Overall N = 8883^a	2004–2008 N = 4136	2009–2013 N = 4747
Gender, n (%)
Female	4122 (46%)	1934 (47%)	2188 (46%)
Male	4761 (54%)	2202 (53%)	2559 (54%)
Age group at CRC surgery, n (%)
<50	648 (7%)	304 (7%)	344 (7%)
50–59	1482 (17%)	737 (18%)	745 (16%)
60–69	3442 (39%)	1574 (38%)	1868 (39%)
70–79	3311 (37%)	1521 (37%)	1790 (38%)
Charlson comorbidity index score at CRC surgery, n (%)
CCI 0	6406 (72%)	3049 (74%)	3357 (71%)
CCI 1	1717 (19%)	756 (18%)	961 (20%)
CCI 2	415 (4.7%)	177 (4%)	238 (5%)
CCI ≥ 3	345 (3.9%)	154 (4%)	191 (4%)
CRC tumor location, n (%)
Colon	5691 (64%)	2600 (63%)	3091 (65%)
Rectum	3192 (36%)	1536 (37%)	1656 (35%)
UICC stage, n (%)
I	1961 (22%)	890 (22%)	1071 (23%)
II	4078 (46%)	1927 (47%)	2151 (45%)
III	2844 (32%)	1319 (32%)	1525 (32%)
Surgical priority, n (%)
Elective	8447 (95%)	3913 (95%)	4534 (96%)
Emergency	434 (5%)	222 (5%)	212 (4%)
Adjuvant chemotherapy, n (%)	2414 (27%)	829 (20%)	1585 (33%)
Neoadjuvant therapy,^b n (%)	1022 (12%)	541 (13%)	481 (10%)

Abbreviation: CRC, colorectal cancer.
^a Including both chemo- and/or radiotherapy.
^b n (%).

3.1 Risk of late recurrence and metachronous CRC

Among the 8883 included patients, late recurrence occurred in 370 patients (305 [82.4%] histologically verified in the DPR) and metachronous CRC in 270 patients (265 [98.1%] histologically verified in the DPR). The recurrences were categorized as local recurrence in 18 patients (4.9% of recurrences).

Between years 5 and 10, the CIF of late recurrence was 4.1% (95%CI: 3.7%–4.6%), Figure 2A. The corresponding 5–10 year CIF of metachronous CRC was 3.0% (95%CI: 2.7%–3.4%).

Specific CIF for late recurrence, metachronous CRC, and second primary cancers according to clinicopathological characteristics are presented in Table 2.

TABLE 2. Risk factors of late recurrence, metachronous CRC and second primary cancers 5–10 years from CRC surgery.

Characteristic	Late recurrence					Metachronous CRC					Second primary cancers
	Cumulative incidence^a	Univariate analysis^a		Multivariable analysis^a		Cumulative incidence^b	Univariate analysis^b		Multivariable analysis^b		Cumulative incidence^c	Univariate analysis^c		Multivariable analysis^c
	5–10 year CIF	sHR	CI	sHR	CI	5–10 year CIF	sHR	CI	sHR	CI	5–10 year CIF	sHR	CI	sHR	CI
Exposure variable
Calendar period of CRC surgery
2004–2008	5.6% (4.9%, 6.3%)	1	Reference	1	Reference	4.1% (3.5%, 4.7%)	1	Reference	1	Reference	7.1% (6.4%, 8.0%)	1	Reference	1	Reference
2009–2013	2.9% (2.4%, 3.4%)	0.50	0.40, 0.62	0.52	0.42, 0.64	2.1% (1.7%, 2.5%)	0.52	0.41, 0.67	0.50	0.39, 0.65	7.1% (6.4%, 7.9%)	0.99	0.85, 1.16	0.98	0.84, 1.15
Confounder variables
Gender
Female	4.0% (3.5%, 4.7%)	1	Reference	1	Reference	2.9% (2.5%, 3.5%)	1	Reference	1	Reference	6.4% (5.7%, 7.2%)	1	Reference	1	Reference
Male	4.2% (3.7%, 4.8%)	1.08	0.88, 1.32	1.07	0.87, 1.32	3.1% (2.7%, 3.6%)	1.04	0.82, 1.32	1.05	0.83, 1.34	7.8% (7.1%, 8.6%)	1.20	1.03, 1.41	1.21	1.03, 1.42
Age group at CRC surgery
<50	5.1% (3.6%, 7.0%)	1	Reference	1	Reference	3.9% (2.6%, 5.6%)	1	Reference	1	Reference	3.5% (2.2%, 5.1%)	1	Reference	1	Reference
50–59	4.8% (3.8%, 5.9%)	0.94	0.62, 1.42	0.92	0.61, 1.40	2.7% (2.0%, 3.7%)	0.70	0.42, 1.15	0.70	0.42, 1.16	5.4% (4.3%, 6.7%)	1.58	0.99, 2.53	1.58	0.99, 2.53
60–69	4.0% (3.4%, 4.7%)	0.79	0.54, 1.16	0.79	0.54, 1.16	2.9% (2.4%, 3.5%)	0.76	0.49, 1.17	0.78	0.50, 1.20	8.0% (7.1%, 8.9%)	2.40	1.55, 3.70	2.34	1.52, 3.61
70–79	3.8% (3.2%, 4.5%)	0.74	0.50, 1.09	0.72	0.48, 1.08	3.1% (2.6%, 3.8%)	0.81	0.52, 1.25	0.84	0.53, 1.31	7.8% (6.9%, 8.8%)	2.37	1.53, 3.66	2.27	1.47, 3.52
Charlson comorbidity index score at CRC surgery
CCI 0	4.1% (3.6%, 4.6%)	1	Reference	1	Reference	3.1% (2.7%, 3.6%)	1	Reference	1	Reference	6.8% (6.2%, 7.5%)	1	Reference	1	Reference
CCI 1	4.3% (3.4%, 5.4%)	1.04	0.81, 1.35	1.11	0.86, 1.45	2.5% (1.9%, 3.4%)	0.84	0.61, 1.17	0.87	0.62, 1.20	7.9% (6.7%, 9.2%)	1.16	0.96, 1.41	1.07	0.88, 1.30
CCI 2	4.6% (2.9%, 6.9%)	1.12	0.70, 1.78	1.24	0.77, 2.00	3.0% (1.6%, 5.1%)	0.94	0.52, 1.67	0.97	0.54, 1.76	10% (7.4%, 13%)	1.49	1.08, 2.05	1.32	0.95, 1.83
CCI ≥ 3	3.5% (1.9%, 5.9%)	0.84	0.47, 1.50	0.89	0.49, 1.62	4.4% (2.6%, 6.9%)	1.43	0.84, 2.41	1.47	0.86, 2.51	6.8% (4.4%, 9.8%)	0.99	0.65, 1.50	0.87	0.57, 1.33
CRC tumor location
Colon	4.0% (3.5%, 4.5%)	1	Reference	1	Reference	3.2% (2.8%, 3.7%)	1	Reference	1	Reference	7.5% (6.9%, 8.3%)	1	Reference	1	Reference
Rectum	4.4% (3.8%, 5.2%)	1.11	0.90, 1.37	1.02	0.82, 1.28	2.7% (2.2%, 3.4%)	0.84	0.65, 1.09	0.88	0.68, 1.16	6.5% (5.7%, 7.4%)	0.88	0.74, 1.03	0.86	0.73, 1.03
UICC stage
I	3.0% (2.3%, 3.8%)	1	Reference	1	Reference	2.6% (1.9%, 3.3%)	1	Reference	1	Reference	7.5% (6.3%, 8.7%)	1	Reference	1	Reference
II	4.1% (3.5%, 4.7%)	1.34	1.00, 1.80	1.38	1.02, 1.87	3.2% (2.7%, 3.7%)	1.25	0.90, 1.73	1.17	0.85, 1.63	7.2% (6.4%, 8.0%)	0.95	0.78, 1.16	0.92	0.75, 1.13
III	5.0% (4.2%, 5.8%)	1.66	1.23, 2.24	1.93	1.38, 2.71	3.2% (2.6%, 3.9%)	1.25	0.88, 1.76	1.05	0.70, 1.56	7.0% (6.1%, 8.0%)	0.93	0.75, 1.15	0.96	0.74, 1.25
Adjuvant chemotherapy
No	4.3% (3.8%, 4.8%)	1	Reference	1	Reference	2.9% (2.5%, 3.4%)	1	Reference	1	Reference	7.3% (6.7%, 7.9%)	1	Reference	1	Reference
Yes	3.8% (3.1%, 4.6%)	0.90	0.71, 1.14	0.75	0.56, 1.01	3.3% (2.6%, 4.1%)	1.14	0.88, 1.48	1.29	0.92, 1.81	6.9% (5.9%, 7.9%)	0.93	0.78, 1.11	0.95	0.75, 1.20

Abbreviations: CI, confidence interval; CIF, cumulative incidence function; sHR, subdistribution hazard ratio.
^a 370 events.
^b 270 events.
^c 635 events.

The 5–10 year CIF of late recurrence decreased over time from 5.6% (95%CI: 4.9%–6.3%) in patients operated in 2004–2008 to 2.9% (95%CI: 2.4–3.4%) in 2009–2013, Table 2. Thus, estimates of the Fine-Grey regression adjusted for confounding covariates showed that the risk of late recurrence was reduced for patients operated in 2009–2013 (Figure 2B) compared to 2004–2008 (Figure 2C, adjusted sHR = 0.52, 95% CI: 0.42–0.64). For metachronous CRC the corresponding 5–10 year hazard in 2009–2013 compared to 2004–2008 patients also displayed a decrease in the adjusted analysis (adjusted sHR = 0.50, 95%CI: 0.39–0.65) as the 5–10 year CIF decreased from 4.1% (95%CI: 3.5%–4.7%) in 2004–2008 to 2.1% (95%CI: 1.7–2.5%) in 2009–2013.

3.2 Second primary cancers 5–10 years after surgical resection of CRC

Among the 8883 included patients, 635 patients were diagnosed with a second primary cancer (CIF = 7.2%, 95%CI: 6.7%–7.7%). Histological verification was achieved in 611 (96.2%) of the cases. The most frequently occurring second primary cancers in women (n = 267) were breast cancer (34%) and lung cancer (20%), Figure S1. Correspondingly, the most frequent second primary cancers in men (n = 368) were prostate cancer (31%) and lung cancer (17%).

In adjusted Fine-Gray regression, the risk of second primary cancer was similar in 2009–2013 compared to 2004–2008 with 5–10 year CIFs of 7.1% (6.4%–7.9%) vs. 7.1% (6.4%–8.0%) resulting in an adjusted sHR of 0.98 (95% CI: 0.84–1.15).

3.3 Overall survival following late events

Characteristics of the patients who developed late recurrence, metachronous CRC or second primary cancer between 5 and 10 years postoperative are provided in Table S2.

Of the 370 patients with late recurrence, 192 died within 5 years of diagnosis, Table 3. Correspondingly, among the 270 metachronous CRC patients, 72 died within 5 years, and among the 635 patients with secondary primary cancers, 316 died within 5 years.

TABLE 3. Hazard ratios of all-cause mortality after late recurrence, metachronous CRC and second primary cancers diagnosed 5–10 years after surgery for stage I–III CRC.

Characteristic	No. of patients	PY	No. of deaths	5-years OS	Crude HRs		Adjusted HRs
Characteristic	No. of patients	PY	No. of deaths	5-years OS	HR	CI	aHR^a	CI
Event
Late recurrence	370	1050	192	46% (41%, 52%)	1	Reference	1	Reference
Metachronous CRC	270	1040	72	72% (66%, 78%)	0.41	0.31, 0.53	0.37	0.28, 0.49
Second primary cancer	635	1864	316	48% (44%, 52%)	0.92	0.77, 1.10	0.78	0.65, 0.93

Abbreviations: aHR, adjusted hazard ratio; CI, confidence interval; HR, hazard ratio; OS, overall survival; PY, person years at risk from event.
^a Adjusted for gender, age at event, and Charlson Comorbidity Index Score at diagnosis of late event.

The 5-year OS was higher following metachronous CRC (5-year OS = 72%, 95% CI: 66%–78%) compared to late recurrence (5-year OS = 46%, 95% CI: 41%–52%), with an adjusted HR of 0.37 (95% CI: 0.28–0.49), Table 3 and Figure S2. In comparison, the 5-year OS was 48% (95% CI: 44%–52%) following second primary cancer, with an adjusted HR of 0.78 (95% CI: 0.65–0.93) compared to patients diagnosed with late recurrence.

4 DISCUSSION

International guidelines recommend 5 years of recurrence surveillance after curative intent treatment for CRC.^{4, 5} However, over the last decades the fraction of CRC patients who are cancer-free survivors at 5 years has been continuously increasing,³ which made us raise the question if extended recurrence surveillance beyond 5 years is indicated. To address this, we have investigated the risk of late events including late recurrence, metachronous CRC and second primary cancer after 5-years of surveillance in a nationwide, contemporary cohort of CRC patients with 10 years of follow-up. In patients operated from 2004 to 2013 the risk of late recurrence from CRC was low, affecting approximately one in 25 patients. Furthermore, the risk of recurrence was nearly halved for patients operated during 2009–2013 compared to 2004–2008. An older French cohort (1985–2000) reported higher rates of late recurrence, affecting one of 12 colon cancer patients and one of 13 rectal cancer patients.^{10, 11} Similar to our findings, they report decreasing rates of late recurrence over time. Thus, the lower and decreasing rates of late recurrence observed in our contemporary cohort indicate a continued decrease during recent decades.

Second primary cancers are common following primary CRC and have been found to affect up to 8.6% of the patients, with 20% of the cancers being metachronous CRC.^{25, 26} In our cohort, late occurring metachronous CRC (5–10 years after primary CRC) was observed for 3.0% of patients. A bit surprising, the incidence of late occurring metachronous CRC decreased from 2004 to 2013 and affected only 2.1% of patients operated from 2009 to 2013. We speculate if the lowered risk of metachronous CRC is due to increased focus on adenoma removal and updated recommendations to include colonoscopies in standardized postoperative surveillance within 5 years after primary surgery.^{4, 27, 28} Excluding late metachronous CRC, we report late occurring second primary cancers to affect 7.2% of the patients, and we found second primary cancers to occur at cancer sites similar to what has been reported by others.²⁶ In contrast to late recurrence and late metachronous CRC the risk of late second primary cancers did not change during our study period.

The American Society of Clinical Oncology (ASCO) endorsement of Cancer Care Ontario guidelines²⁹ recommend a surveillance colonoscopy approximately 1 year postoperative and repeated every 5 years to identify metachronous CRC. Similarly, European Society for Medical Oncology (ESMO) guidelines^{4, 5} recommend a colonoscopy at 1 year postoperative, and thereafter every 3 to 5 years. Consequently, colonoscopy is the only current surveillance modality, which extend beyond 5 years. However, as we show that the risk of late metachronous CRC is only 2.1% for 2009–2013 patients with a flattening cumulative incidence curve from 5 to 10 years, we speculate if long-term endoscopic follow-up is at all justified if the colon and rectum appears normal 5 years after surgery.³⁰ Also given that our survival analysis showed the survival associated with late metachronous CRC was high with a 5-year overall survival of 72%.

Strengths of the current study include its design and the nationwide cohort, reflecting the CRC management at a national level and not the performance of individual sites or clinical trials. A study demanding a minimum of 5 and preferably 10 years of follow-up requires a design with consistent registration of exposures and patient characteristics, and consistent identification of outcomes throughout the study period. This is possible in Denmark due to our extensive health care registries. In this study, we identified recurrence using an algorithm validated on Danish health care registry data from 2001 throughout 2018. Previous validations have shown great performance in identifying recurrence.²¹ However, the algorithm as originally designed, classifies all events as recurrence and does not distinguish between recurrence and metachronous CRC. Misclassification of recurrence as metachronous CRC and vice versa might therefore impact our findings. We took precautions to limit misclassification by mapping the remaining segments of the colon and rectum for each patient to be able to see if the recurrence (as defined by the algorithm) was in a segment not previously resected at time of recurrence. In addition, if there is any misclassification at all, we expect it to affect the 2004–2008 and the 2009–2013 periods similarly. Another limitation of our approach is our ability to distinguish between distant and local recurrence. We report lower rates of local recurrence compared to other studies,^{10, 11} and therefore refrained from performing local recurrence specific analyses. Finally, we chose a maximum age of 80 years at surgery to limit the potential detection bias among elderly patients, as they are not always offered recurrence surveillance according to guidelines.

In conclusion, our findings do not advocate for extending CRC-specific recurrence surveillance beyond 5 years. The burden of late recurrence and metachronous CRC has decreased to a low level, while the burden of second primary cancers has remained unchanged. Thus, symptoms or suspicion of a cancer occurring 5–10 years from primary CRC treatment, is more likely to represent a non-CRC cancer (7.1%), than late recurrence (2.9%) or metachronous CRC (2.1%). Furthermore, the survival following second primary cancers were comparable to the post-recurrence survival.

AUTHOR CONTRIBUTIONS

All the authors made significant contributions to the completion of this study. Jesper Nors: conceptualization, methodology, formal analysis, investigation, data curation, writing—original draft, visualization, project administration. Kåre Andersson Gotschalck: conceptualization, methodology, investigation, writing—review & editing, supervision. Rune Erichsen: conceptualization, methodology, writing—review & editing, supervision. Claus Lindbjerg Andersen: conceptualization, investigation, resources, data curation, writing—review & editing, supervision, project administration, funding acquisition. All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Jesper Nors and Claus Lindbjerg Andersen had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The work reported in the paper has been performed by the authors, unless clearly specified in the text.

ACKNOWLEDGEMENT

We would like to extend our gratitude to the Danish Colorectal Cancer Group (DCCG) for providing data from the national colorectal cancer database, and to CONNECT at Aarhus University Hospital for assistance in applying for data from the Danish Health Data Authority.

FUNDING INFORMATION

This study was funded by a PhD scholarship from Institute of Clinical Medicine, Aarhus University, Denmark [Nors] and support from the Novo Nordisk Foundation [grant number NNF17OC0025052 (Andersen), NNF22OC0074415 (Andersen)], Innovation Fund Denmark [grant number 9068-00046B (Andersen)] and the Danish Cancer Society [grant numbers R231-A13845, R257-A14700 (Andersen)]. The funding organizations and entities listed above had no influence on the study's design, data collection, analysis, interpretation of the data, manuscript preparation and review, or decision to submit for publication.

CONFLICT OF INTEREST

The author(s) have no affiliations or financial involvement with any organization or entity with a financial interest or financial conflict with the study subject or materials discussed in the manuscript.

ETHICS STATEMENT

This study was approved by the Danish Colorectal Cancer Group and the Danish Data Protection Agency (Central Denmark Region, J. no. 1-16-02-274-19) and adheres to the General Data Protection Regulations. The study is based on anonymized registry data and therefore does not require approval from the National Research Ethics Committee.

Open Research

DATA AVAILABILITY STATEMENT

The data and code for data cleaning and analysis that support the findings of this study are available from Danish Health care registries and upon reasonable collaboration with a Danish Research Institution. Please see https://sundhedsdatastyrelsen.dk/da/english/health_data_and_registers/research_services for further details. Further information is available from the corresponding author upon request.

Supporting Information

REFERENCES

1Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021; 71(3): 209-249.
10.3322/caac.21660
PubMed Web of Science® Google Scholar
2van Gestel YRBM, de Hingh IHJT, van Herk-Sukel MPP, et al. Patterns of metachronous metastases after curative treatment of colorectal cancer. Cancer Epidemiol. 2014; 38(4): 448-454.
10.1016/j.canep.2014.04.004
PubMed Web of Science® Google Scholar
3Nors J, Iversen LH, Erichsen R, Gotschalck KA, Andersen CL. Incidence of recurrence and time to recurrence in stage I to III colorectal cancer: a nationwide Danish cohort study. JAMA Oncol. 2023; 16: 54-62. doi:10.1001/jamaoncol.2023.5098
10.1001/jamaoncol.2023.5098
Google Scholar
4Glynne-Jones R, Wyrwicz L, Tiret E, et al. Rectal cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2018; 29(Suppl 4):iv263.
10.1093/annonc/mdy161
CAS PubMed Web of Science® Google Scholar
5Argilés G, Tabernero J, Labianca R, et al. Localised colon cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2020; 31(10): 1291-1305.
10.1016/j.annonc.2020.06.022
CAS PubMed Web of Science® Google Scholar
6Iversen LH, Green A, Ingeholm P, Østerlind K, Gögenur I. Improved survival of colorectal cancer in Denmark during 2001-2012: the efforts of several national initiatives. Acta Oncol. 2016; 55(Suppl 2): 10-23.
10.3109/0284186X.2015.1131331
PubMed Web of Science® Google Scholar
7Larsen MB, Njor S, Ingeholm P, Andersen B. Effectiveness of colorectal cancer screening in detecting earlier-stage disease: a nationwide cohort study in Denmark. Gastroenterology. 2018; 155(1): 99-106.
10.1053/j.gastro.2018.03.062
PubMed Web of Science® Google Scholar
8 Health at a Glance: Europe. State of Health in the EU Cycle. 2022.
Google Scholar
9Seo SI, Lim SB, Yoon YS, et al. Comparison of recurrence patterns between ≤5 years and >5 years after curative operations in colorectal cancer patients. J Surg Oncol. 2013; 108(1): 9-13.
10.1002/jso.23349
CAS PubMed Web of Science® Google Scholar
10Bouvier AM, Launoy G, Bouvier V, et al. Incidence and patterns of late recurrences in colon cancer patients. Int J Cancer. 2015; 137(9): 2133-2138.
10.1002/ijc.29578
CAS PubMed Web of Science® Google Scholar
11Cottet V, Bouvier V, Rollot F, et al. Incidence and patterns of late recurrences in rectal cancer patients. Ann Surg Oncol. 2015; 22(2): 520-527. doi:10.1245/s10434-014-3990-1
10.1245/s10434-014-3990-1
PubMed Web of Science® Google Scholar
12von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Epidemiology. 2007; 18(6): 800-804.
10.1097/EDE.0b013e3181577654
PubMed Web of Science® Google Scholar
13Sørensen HT, Pedersen L, Jørgensen J, Ehrenstein V. Danish clinical quality databases: an important and untapped resource for clinical research. Clin Epidemiol. 2016; 8: 425-427.
10.2147/CLEP.S113265
PubMed Web of Science® Google Scholar
14Ingeholm P, Gögenur I, Iversen LH. Danish colorectal cancer group database. Clin Epidemiol. 2016; 8: 465-468.
10.2147/CLEP.S99481
PubMed Web of Science® Google Scholar
15Schmidt M, Schmidt SAJ, Adelborg K, et al. The Danish health care system and epidemiological research: from health care contacts to database records. Clin Epidemiol. 2019; 11: 563-591.
10.2147/CLEP.S179083
PubMed Web of Science® Google Scholar
16Gjerstorff ML. The Danish cancer registry. Scand J Public Health. 2011; 39(7_suppl): 42-45. doi:10.1177/1403494810393562
10.1177/1403494810393562
PubMed Web of Science® Google Scholar
17Schmidt M, Schmidt SAJ, Sandegaard JL, Ehrenstein V, Pedersen L, Sørensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol. 2015; 7: 449-490.
10.2147/CLEP.S91125
PubMed Web of Science® Google Scholar
18Erichsen E. Existing data sources for clinical epidemiology: the Danish National Pathology Registry and data Bank. Clin Epidemiol. 2010; 2: 51-56. doi:10.2147/clep.s9908
10.2147/CLEP.S9908
PubMed Google Scholar
19Pedersen CB. The Danish civil registration system. Scand J Public Health. 2011; 39(7 Suppl): 22-25.
10.1177/1403494810387965
PubMed Web of Science® Google Scholar
20Lash TL, Riis AH, Ostenfeld EB, Erichsen R, Vyberg M, Thorlacius-Ussing O. A validated algorithm to ascertain colorectal cancer recurrence using registry resources in Denmark. Int J Cancer. 2015; 136(9): 2210-2215. doi:10.1002/ijc.29267
10.1002/ijc.29267
CAS PubMed Web of Science® Google Scholar
21Nors J, Mattesen TB, Cronin-Fenton D, et al. Identifying recurrences among non-metastatic colorectal cancer patients using National Health Data Registries: validation and optimization of a registry-based algorithm in a modern Danish cohort. Clin Epidemiol. 2023; 15: 241-250.
10.2147/CLEP.S396140
PubMed Web of Science® Google Scholar
22Sobin LH, Fleming ID. TNM classification of malignant tumors, fifth edition (1997). Union Internationale Contre le cancer and the American joint committee on cancer. Cancer. 1997; 80(9): 1803-1804.
10.1002/(SICI)1097-0142(19971101)80:9<1803::AID-CNCR16>3.0.CO;2-9
CAS PubMed Web of Science® Google Scholar
23Charlson ME, Pompei P, Ales KL, Ronald MKC. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40(5): 373-383. doi:10.1016/0021-9681(87)90171-8
10.1016/0021-9681(87)90171-8
CAS PubMed Google Scholar
24Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999; 94(446): 496-509. doi:10.1080/01621459.1999.10474144
10.1080/01621459.1999.10474144
Web of Science® Google Scholar
25Halamkova J, Kazda T, Pehalova L, et al. Second primary malignancies in colorectal cancer patients. Sci Rep. 2021; 11(1): 2759.
10.1038/s41598-021-82248-7
CAS PubMed Web of Science® Google Scholar
26Liang LA, Tseng YJ, Tanaka LF, Klug SJ. Second primary cancer among 217702 colorectal cancer survivors: an analysis of national German cancer registry data. Int J Cancer. 2023; 153: 1459-1471. doi:10.1002/ijc.34638
10.1002/ijc.34638
CAS PubMed Web of Science® Google Scholar
27van Lanschot MCJ, van Leerdam ME, Lansdorp-Vogelaar I, et al. Yield of surveillance colonoscopies 1 year after curative surgical colorectal cancer resections. Clin Gastroenterol Hepatol. 2019; 17(11): 2285-2293.
10.1016/j.cgh.2019.02.026
PubMed Web of Science® Google Scholar
28Rex DK, Kahi CJ, Levin B, et al. Guidelines for colonoscopy surveillance after cancer resection: a consensus update by the American Cancer Society and US multi-society task force on colorectal cancer. CA Cancer J Clin. 2006; 56(3): 160-167. quiz 185–186.
10.3322/canjclin.56.3.160
PubMed Web of Science® Google Scholar
29Meyerhardt JA, Mangu PB, Flynn PJ, et al. Follow-up care, surveillance protocol, and secondary prevention measures for survivors of colorectal cancer: American Society of Clinical Oncology clinical practice guideline endorsement. J Clin Oncol. 2013; 31(35): 4465-4470.
10.1200/JCO.2013.50.7442
PubMed Web of Science® Google Scholar
30Bouvier AM, Latournerie M, Jooste V, Lepage C, Cottet V, Faivre J. The lifelong risk of metachronous colorectal cancer justifies long-term colonoscopic follow-up. Eur J Cancer. 2008; 44(4): 522-527.
10.1016/j.ejca.2008.01.007
PubMed Web of Science® Google Scholar

Volume154, Issue11

1 June 2024

Pages 1890-1899

This article also appears in:

Wiley Colorectal Cancer Awareness Collection

Incidence of late recurrence and second primary cancers 5–10 years after non-metastatic colorectal cancer

Abstract