Influence of microsatellite instability and KRAS and BRAF mutations on lymph node harvest in stage I-III colon cancers

Abstract

Lymph node (LN) harvest is influenced by several factors, including tumor genetics. Microsatellite instability (MSI) is associated with improved node harvest, but the association to other genetic factors is largely unknown. Research methods included a prospective series of stage I-III colon cancer patients undergoing ex vivo sentinel-node sampling. The presence of MSI, KRAS mutations in codons 12 and 13, and BRAF V600E mutations was analyzed. Uni- and multivariate regression models for node sampling were adjusted for clinical, pathological and molecular features. Of 204 patients, 67% had an adequate harvest (≥ 12 nodes). Adequate harvest was highest in patients whose tumors exhibited MSI (79%; odds ratio [OR] 2.5, 95% confidence interval [CI] 1.2-4.9; P = 0.007) or were located in the proximal colon (73%; 2.8, 1.5-5.3; P = 0.002). In multiple linear regression, MSI was a significant predictor of the total LN count (P = 0.02). Total node count was highest for cancers with MSI and no KRAS/BRAF mutations. The independent association between MSI and a high LN count persisted for stage I and II cancers (P = 0.04). Tumor location in the proximal colon was the only significant predictor of an adequate LN harvest (adjusted OR 2.4, 95% CI 1.2-4.9; P = 0.01). An increase in the total number of nodes harvested was not associated with an increase in nodal metastasis. In conclusion, number of nodes harvested is highest for cancers of the proximal colon and with MSI. The nodal harvest associated with MSI is influenced by BRAF and KRAS genotypes, even for cancers of proximal location. Mechanisms behind the molecular diversity and node yield should be further explored.

Figure 1

LN counts for MSI status and tumor location in the colon. (A) Total node counts according to MSI status (MSS, MSI-L or MSI-H). The asterisks indicate statistical significance. (B) The total LN counts for MSI status according to location in either the proximal or distal colon. The box-and-whiskers plot shows the medians, interquartile ranges and 95% CIs, and the circles/asterisks indicate outliers. (C) Adequate LN harvest improved significantly within the MSS cancers for proximal cancers. Adequate harvest proportion increased significantly with addition of MSI to either location (P = 0.001 for trend).

Figure 2

Proportion of patients with adequate node harvest according to MSI and KRAS/BRAF genotypes. The rates of adequate LN sampling are shown for MSI/MSS tumors with or without KRAS mutations (A) and MSS/MSI tumors with or without BRAF V600E mutations (B).

Figure 3

Proportion of adequate node harvest according to age, location and pT-stage. (A) Proportion of adequate LN harvest according to age-groups, stratified for MSI status (P < 0.05 for trend). (B) Proportion of adequate LN harvest according to age and tumor location (P < 0.05 for trend). (C) Proportion of adequate LN harvest according to pT-stage and MSI, P < 0.05 for trend. (D) Distribution of pN categories according to MSI status.

Figure 4

Cancer-specific survival for stage (A), MSI and MSS (B) and MSI/MSS and BRAF wt/mut genotypes (C). Hashes on the survival curves denote censored patients.