Genomic lesions drive the metastasis of esophageal squamous cell carcinoma

Esophageal cancer ranks sixth in terms of cancer-associated mortality (1). The majority (over 80%) of esophageal cancers are squamous cell carcinomas (ESCC) (2), which is the leading cause of cancer-related death in its epidemic regions, such as northern China. Even with the advent of modern therapeutic approaches, the 5 years survival of ESCC (<20%) has not significantly improved in the past few decades. These facts necessitate the need for better characterization of the molecular basis of ESCC pathogenesis.

The dissemination and seeding of cancer cells from primary tumors to distant tissues is a complex, multi-step process. Clinically, tumor metastasis is, by and large, incurable and accounts for the majority of cancer-associated deaths. ESCC is no exception, in that the great majority of ESCC-associated deaths are caused by distal metastasis. Although the exome of more than 700 primary ESCCs have been sequenced over the years (3-8), genomic lesions associated with ESCC metastasis have not been systematic profiled. In a recent study (9), Dai et al. performed whole-exome sequencing to compare the genomic alterations between primary ESCCs and matched metastatic lesions in lymph node (LN), representing an initial effort to understand the genomic basis underlying ESCC metastatic process.

Not surprisingly, overall, more than half of the mutations in primary tumors were retained in the LN metastatic samples. These included known driven genes, including TP53, KMT2D, RB1, CDKN2A, ZNF750, NOTCH3 and TET2. This is consistent with our recent finding that the majority of these significantly mutated genes (SMGs) tend to arise as early events during clonal evolution of esophageal tumorigenesis, albeit our results were based on a multi-region profiling in primary cases and computational modeling (10). We also suggested that a few activating mutations, such as those in PIK3CA and NFE2L2, might reinforce subclonal diversification in ESCC. However, due to the limited number of LN samples, Dai et al. did not observed such features in their cohort.

Notably, Dai and colleagues observed that mutations in ZNF750 appeared more frequently in metastatic tumors than the primary ones. In addition, in silico analysis showed that ESCC individuals with ZNF750 mutations were more likely to present LN metastasis, and to have shorter disease-free survival compared with those without ZNF750 mutations. ZNF750 is a novel tumor suppressor identified by our lab and others (5,11,12), and it exhibits frequent loss-of-function mutations and genomic deletions across different types of squamous cell carcinomas in a lineage-specific fashion. In vitro experiments have shown that ZNF750 regulates the proliferation, differentiation, adhesion and migration of squamous cells, supporting its putative role in tumor metastasis.

Another interesting finding from this study is the metastatic-specific deletion of chromosome 6p22 in ESCC. The authors further showed that genes involved in nucleosomal assembly were highly significantly enriched in this genomic segment. In addition, loss of these genes appeared to correlate with the survival time of ESCC patients. Considering that a few genes involved in nucleosomal assembly were found to contribute to the metastasis in other cancer types (13,14), these results warrant further investigations to establish the biological significance of the process of nucleosomal assembly in ESCC metastasis.

In summary, Dai and colleagues demonstrated a number of noteworthy genomic lesions associated with ESCC metastasis to the LN, which provides insights into the molecular changes during regional metastasis this cancer. Clearly, both larger number of samples and epigenomic approaches (15) are needed to further establish the biological, as well as clinical significance of these alterations. More importantly, samples from matched primary and distal metastatic tumors are required to dissect the biological features associated with the distant metastasis of this aggressive malignancy.