Completeness and accuracy of the registration of recurrences in the Swedish Colorectal Cancer Registry (SCRCR) and an update of recurrence risk in colon cancer

Abstract

Background: The completeness and accuracy of the registration of synchronous metastases and recurrences in the Swedish Colorectal Cancer Registry has not been investigated. Knowing how accurate these parameters are in the registry is a prerequisite to adequately measure the current recurrence risk. Methods: All charts for patients diagnosed with stage I-III colorectal cancer (CRC) in two regions were reviewed. In one of the regions, all registrations of synchronous metastases were similarly investigated. After the database had been corrected, recurrence risk in colon cancer was calculated stratified by risk group as suggested by ESMO in 2020. Results: In patients operated upon more than five years ago (N ¼ 1235), there were 20 (1.6%) recurrences not reported. In more recent patients, more recurrences were unreported (4.0%). Few synchronous metastases were wrongly registered (3.6%) and, likewise, few synchronous metastases were not registered (about 1%). The five-year recurrence risk in stage II was 6% for low-risk, 11% for intermediate risk, and 23% for high-risk colon cancer patients. In stage III, it was 25% in low-and 45% in highrisk patients. Incorporation of risk factors in stage III modified the risks substantially even if this is not considered by ESMO. Adjuvant chemotherapy lowered the risk in stage III but not to any relevant extent in stage II. Conclusion: The registration of recurrences in the registry after 5 years is accurate to between 1 and 2% but less accurate earlier. A small number of unreported recurrences and falsely reported recurrences were discovered in the chart review. The recurrence risk in this validated and updated patient series matches what has been recently reported, except for the risk of recurrence in stage II low risk colon cancers which seem to be even a few percentage points lower (6 vs. 9%).

KRAS-G12C Mutation in One Real-Life and Three Population-Based Nordic Cohorts of Metastatic Colorectal Cancer

“…In the Uppsala region cohort, all mCRC patients were prospectively identified in a biobank initiative Uppsala-Umeå Comprehensive Cancer Consortium (UCAN)) ( 30 ) since April 2010, and the remaining patients with mCRC since January 2010 were retrospectively identified using a hospital-based registry and the Swedish ColoRectal Cancer Registry (SCRCR) ( 31 ). After a validation study against the medical records of all patients with a diagnosis of CRC since January 2010, this cohort can be considered 100% complete ( 32 ).…”

Accurate population-based model for individual prediction of colon cancer recurrence

“…Here, the estimates are more precise which may be because the use of adjuvant chemotherapy in stage II patients have been rather restricted in Sweden. Non-registered recurrences may explain a small part of the deviation in high-risk patients; a recent investigation of the SCRCR estimated that 1-2% of recurrences at 5 years are missing in the registry [35]. Variables significant in univariable analysis were initially included.…”

“…However, the calibration was better in low-intermediate risk patients than the other nomograms but got progressively worse with better-observed prognosis in high-risk patients. This may be an effect of good survival outcomes in Norway [36], regression to the mean or recurrences not reported to the registry (94% conformity between patient records and registry [37], lower than in Sweden [35]). A weakness is that we did not externally validate the predictions in stage I patients, however, it is not difficult to predict the outcome since recurrence rates are very low for most if not all stage I patients.…”

Finding Waldo: The Evolving Paradigm of Circulating Tumor DNA (ctDNA)—Guided Minimal Residual Disease (MRD) Assessment in Colorectal Cancer (CRC)

“…However, accumulating data indicate that risk stratification based on the clinicopathologic criteria is imprecise in identifying patients with MRD, resulting in overtreatment and undertreatment of a significant number of patients [10][11][12]. To elaborate further, it is well known that approximately 50% of the stage III patients and 74% of low-risk stage III patients are cured by surgery alone [10,13]. However, current guidelines [14,15] recommend ACT for all such patients, resulting in unnecessary chemotherapy administration in a large proportion of patients, causing a myriad of short-and long-term toxicities.…”

“…Conversely, 5% of stage I CRC and 10-25% of stage II patients recur; however, for stage I and average-risk stage II patients, there is no available risk-stratification tool to identify patients who are destined to recur after definitive surgery [16]. Furthermore, adjuvant trial results suggest that the survival benefit with oxaliplatin-based ACT is modest, adding approximately 17% absolute 5-year disease-free survival (DFS) benefit over surgery alone in stage III patients (i.e., oxaliplatinbased ACT increases 5-year DFS from around 50% with surgery alone to 67% with surgery plus ACT [10,13,[17][18][19]). It is also important to emphasize that the benefit of ACT after modern surgery may be less than quoted above, as the improvements in surgical techniques possibly cure more patients than before [20][21][22].…”