Time of Metastasis and Outcome in Colorectal Cancer.

OBJECTIVE

The aim of this study was to evaluate outcomes of metastases at various time intervals after colorectal cancer (CRC) diagnosis.

BACKGROUND

Earlier studies have indicated a short time interval between CRC diagnosis and distant metastases to be associated with poor prognosis. The majority of studies assessed outcome from CRC diagnosis or metastasis resection rather than from metastasis diagnosis and might be subject to immortal time bias.

METHODS

Patients in the population-based DACHS study were stratified: metastases at/within 1 month (immediate), 2 to 6 months (early), 7 to 12 months (intermediate), and >12 months (late) after CRC diagnosis. The primary endpoint was overall survival (OS) from metastasis diagnosis. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CI). HRs were adjusted for important confounders and immortal time.

RESULTS

A total of 1027 patients were included. T4 (P < 0.0001) and node-positive tumors (P < 0.0001) were more frequent in the immediate group. Lung metastases (P < 0.0001) and single-site metastases (P < 0.0001) were more prevalent in the late group. In multivariable analysis, immediate metastases were not associated with poor OS compared to metastases at later time points (late vs immediate: HR 1.21; 95% CI, 0.98-1.48). Subgroup analyses revealed poor OS of late versus immediate metastases for females (1.45; 1.08-1.96), proximal colon cancer (1.54; 1.09-2.16), and N0 (1.46; 1.00-2.12) or N1 disease (1.88; 1.17-3.05).

CONCLUSIONS

Immediate or early metastases are not associated with unfavorable outcome compared to late metastases. These findings challenge the current notion of poor outcome for CRC with immediate or early metastases.