Background

To date, it is still unclear whether a single metastatic subclone evolves in the primary tumor, subsequently spreading to lymph nodes (LNs) and distant sites, or whether multiple subclones in the primary tumor independently seed LNs and distant metastases (mets).

Recently, Naxerova et al. examined the evolutionary relationship between primary CRC tumors, LNs and distant mets [1]. In 65% of cases, lymphatic and distant metastases arose from independent subclones in the primary tumor. These results support the idea that LNs did not serve as a way station for the metastasizing cells, suggesting that LNs are not essential intermediaries.

Very few studies have been performed on molecular genetics and mutational heterogeneity in LNs mets. Recently, Ulintz et al. [2] found that LNs mets are polyclonal. Additionally, the authors found that single LN can harbor sub-clones from different geographic regions in the primary tumor. These findings are consistent with a model of metastasis where multiple waves of metastatic cells escape the primary tumor over time and seed a given LN during tumor progression.

Recent population-based studies on mets in CRC showed that between 40% [3] and 63% [4] of metachronous mets develop in patients without LNs mets, demonstrating that distant mets can develop independently of the presence of LNs mets [5].

The possibility to individuate beforehand those tumors which will seed LNs or distant mets would have a tremendous impact on the clinical practice. Therefore, with this study we aimed to characterize the molecular features of LNs mets from CRC. Subsequently, we compared LNs and distant mets to evaluate whether molecular differences exist between these metastatic sites.

Methods

A total of 11871 tumors that underwent comprehensive genomic profiling by Caris Life Sciences (Phoenix, AZ) were identified from a retrospective database. Molecular characteristics, MSI status, tumor mutational burden (TMB), as well as protein expression by IHC were analyzed for difference based on tumor site: primary (N = 5862), distant mets (N = 5606) and LNs (N = 403).

IHC was performed on FFPE sections. Protein staining was scored for intensity (0 = no staining; 1+ = weak staining; 2+ = moderate staining; 3+ = strong staining) and staining percentage (0-100%) by pathologists. PD-L1 testing was performed using the SP142 anti-PD-L1 clone (Ventana, Tucson, AZ).

NGS was performed on genomic DNA isolated from FFPE tumor samples using the NextSeq (592-genes)/MiSeq platform (45-gene) (Illumina, Inc., San Diego, CA). All variants were detected with greater than 99% confidence based on allele frequency and amplicon coverage, with an average sequencing depth of coverage of greater than 500 and an analytic sensitivity of 5%.

MSI was examined by counting number of microsatellite loci that were altered by somatic insertion or deletion for each sample. The threshold to determine MSI by NGS was determined to be 46 or more loci with insertions or deletions to generate a sensitivity of > 95% and specificity of > 99%.

TMB was estimated from 592 genes (1.4 megabases [MB] sequenced per tumor) by counting all non-synonymous missense mutations found per tumor that had not been previously described as germline alterations. Differences in mean TMB was assessed using ANOVA test.

Chi-square test was performed for comparative analysis using SPSS v23 (IBM SPSS Statistics), and significance was defined as p < 0.05.

Conclusions

LNs harbor a different molecular profile from both the primary colorectal cancer and distant mets. In addition, TMB-high, MSI-high and PD-L1 overexpression are more frequent in LNs than in distant mets, although similar pattern are observed in primary tumors.

TMB mean is significantly higher in LNs vs distant mets and primaries (P<.0001), independent of MSI-H status. Analyzing distant mets by location, LNs showed higher TMB compared to lung, liver and peritoneum mets (P<.0001) (figure 4).

ATM mutations may be the cause of higher TMB mean in MSS LNs.

This is one of the largest study to investigate the molecular differences between LNs, distant mets and primary tumors in metastatic CRC patients. Our data support the hypothesis that lymphatic and distant mets harbor different mutation profiles which suggests that they may arise from distinct subclones.