NEW YORK (GenomeWeb) – A new genomic study spanning several cancer types has uncovered a set of common driver mutations within different metastatic tumors from each patient, despite broader genomic diversity in the tumors.
"[W]e observed a mix of overlapping and differing driver mutations," first author Johannes Reiter, an early cancer detection and evolutionary dynamics researcher affiliated with Stanford and Harvard Universities, said in a statement. "But through computational analyses, we inferred that the driver mutations that were most likely to contribute to cancer development were shared among all metastases in each patient."
Reiter and his colleagues from Stanford, Harvard, and elsewhere turned to cancer phylogenetic analyses to retrace mutation events in primary tumor and untreated metastatic tumor samples from 20 individuals with skin, breast, colorectal, endometrial, gastric, lung, pancreatic, or prostate cancers, distinguishing between so-called MetTrunk drivers shared between metastases and MetBranch mutations found in just a subset.
Based on whole-genome- or exome sequencing data for more than three-dozen samples from primary tumors and 76 of the metastatic tumor samples, the team determined that most functionally relevant driver mutations correspond from one metastatic tumor to the next in a given patient — information that is expected to be relevant when selecting targeted treatments in individuals with metastatic tumors.
"Because driver gene mutations increasingly inform clinical treatment decision, undetected driver heterogeneity among metastases poses a barrier to the success of this precision medicine approach," the authors wrote. In contrast, they noted, the driver mutation homogeneity found in the cancer types analyzed so far "suggests that a single biopsy accurately represents the driver gene mutations of a patient's metastases."
Moreover, the researchers noted that the driver mutations detected appear to be relatively common across genomically profiled cancer types, based on comparisons with sequence data for more than 25,500 samples in the "Catalogue of Somatic Mutations in Cancer" (COSMIC). The findings appeared online today in Science, along with a "mathematical model of tumor evolution and metastasis formation" that the authors developed to further understand the driver gene mutation homogeneity that turned up in their analysis.
Building from the current results, Reiter noted that the team is interested in finding out whether "the idea of common functional drivers holds up when dealing with 20 to 30 cancer types and hundreds of untreated samples."
For their current analysis, the researchers profiled three metastatic tumor samples for each patient, on average, using insights from the Cancer Genome Atlas project to determine whether nearly 1,800 non-synonymous mutations found in the available genome or exome sequence data fell in a suspected cancer driver genes and, if so, whether these mutations occurred at the base or in subsequent branches of a tumor phylogenetic tree.
After removing data for two patients with hyper-mutated tumors, the team profiled phylogenetic patterns in tumors from the remaining 18 cancer cases, incorporating COSMIC data to help distinguish driver gene changes with functional consequences.
The authors cautioned that they did not consider recurrent non-coding mutations, copy number changes, epigenetic profiles, or mutations in metastatic lesions too small to detect clinically. In addition, they wrote, "we cannot exclude the possibility that mutations in yet-undiscovered driver genes of metastases are heterogeneous."