Variations in Metastasis Site by Primary Location in Colon Cancer.

OBJECTIVE

The purpose of this paper is to determine whether sites of distant recurrence are associated with specific locations of primary disease in colon cancer.

METHODS

A cohort including all patients (n = 947) undergoing a segmental colonic resection for colon cancer at our center (2004-2011) comparing site-specific metastatic presentation and recurrence rates, as well as their respective multivariable American Joint Committee on Cancer (AJCC) stage-adjusted hazard ratios (mHR).

RESULTS

Right-sided colectomies (n = 557) had a lower overall metastasis rate (24.8% vs. 31.8%; P = 0.017; mHR = 1.24 [95% CI: 0.96-1.60]; P = 0.011) due to significantly lower pulmonary metastasis in follow-up (2.7% vs. 9%; P < 0.001; mHR = 0.32 [95% CI: 0.17-0.58]; P = 0.001) and lower overall liver metastasis rate (15.6 vs. 22.1%; P = 0.012; mHR = 0.74 [95% CI: 0.55-0.99];P = 0.050). Left colectomies (n = 127) had higher rates of liver metastasis during follow-up (9.4% vs. 4.8%; P = 0.029; mHR = 1.64 [95% CI: 0.86-3.15]; P = 0.134). Sigmoid resections (n = 238) had higher baseline rates of liver metastasis (17.1% vs. 11.3%; P = 0.015) and higher cumulative rates of lung (12.2% vs. 5.4%; P < 0.001; mHR = 2.26 [95% CI: 1.41-3.63]; P = 0.001) and brain metastases (2.3% vs. 0.6%; P = 0.033; mHR = 4.03 [95% CI: 1.14-14.3]; P = 0.031). Other sites of metastasis, including the (retro) peritoneum, omentum, ovary, and bone, did not yield significant differences.

CONCLUSIONS

Important variations in site-specific rates of metastatic disease exist within major resection regions of colon cancer. These variations may be important to consider when evaluating options for adjuvant treatment and surveillance after resection of the primary disease.