Identical Driver Mutations in Metastases as Primary Cancer

Alexander M. Castellino, PhD

September 20, 2018

Metastases drive most cancer-related mortality. So, identifying mutations that are driving cancer growth and that are also present among all metastases of a cancer may be important in disease management. This is exactly what the authors of a genome- and exome-wide analysis of 76 untreated metastases from 20 patients with diverse primary tumors set out to do — and they showed that these so-called driver mutations were present among all metastases.

The findings come from a multi-institutional study published online September 7 in Science.

"Driver gene mutations are homogeneous among all metastases of a primary tumor," Johannes G. Reiter, PhD, computational scientist and instructor from the Canary Center for Cancer Early Detection, Stanford University, California, told Medscape Medical News.

"In the era of targeted therapies, which rely on mutational analysis for diagnosis and treatment, identifying appropriate driver mutations to selectively target may be important in disease management," Reiter suggested.

The researchers analyzed data from 20 cancer patients for whom genome-wide or exome-wide sequencing had been conducted for at least two distinct treatment-naive metastases. Patients included in the analysis had melanoma, as well as pancreatic, endometrial, colorectal, breast, gastric, lung, and prostate cancers.

The researchers studied 115 samples, which included 76 untreated metastases. The mutational analyses identified driver and passenger mutations; the designations were based on 299 putative driver genes from the Cancer Genome Atlas consensus.

In a video commentary on the study findings and their significance, senior author Bert Vogelstein, MD, of the Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, Maryland, noted that their study confirmed tumor mutational heterogeneity, but not in the driver genes. He said: "Our study focused on these driver genes and on patients with metastatic disease that hadn't been treated with any therapy. We found that the driver genes in these patients were generally not heterogeneous.

"The great majority of metastases from the same patient had the identical driver gene mutations," Vogelstein said.

Reiter explained further how they determined this. The team built an evolutionary tree using a computational analysis. Mutations present in all metastases were referred to as MetTrunk, and those present in a subset of metastases were designated as MetBranch.

The team then determined whether heterogeneous mutations in driver genes were likely to be functional by comparing driver genes in MetTrunk with those previously seen in other cancers. A comparison with the Catalogue of Somatic Mutations in Cancer (COSMIC) database showed that 38% of driver mutations identified in the trunk were present in COSMIC and that only 16% of these were present in the branch. The fraction of driver mutations in the branches was similar in frequency to passenger mutations seen in the trunk of branches.

For example, in a patient with colorectal cancer, two mutations in APC and one in TP53 were present in the trunk; AR, CACNA1A, IRF2 mutations were present in the branches; and EPHA3, KMT2C, and PTPRD were present only in some regions of the primary tumor. But of all these, only the mutations in APC and TP53 were present across all cells — cells from the primary and the metastatic sites — and were also predicted to be important drivers of the disease.

Reiter explained that driver mutations are present in the original tumor clone, and as tumor cells grow and are disseminated to sites of metastases, they acquire additional mutations. "That is why identifying the key functional driver mutation(s) is important but challenging," he said. "Most previous studies looked at samples from the primary tumor but lacked the ability to indicate which of the driver mutations are also likely to be present among all metastases and can therefore be targeted by treatment," he explained.

The researchers plan to conduct this type of analysis in larger cohorts of patients and across many other tumors to determine whether the "minimal functional driver gene mutation heterogeneity is a general phenomenon of advanced disease."

Study Implications

Vogelstein explained that the study has significant practical implications. One is critical to the development of therapies that target the drive genes. "If driver gene mutations in different metastatic lesions from the same patient were heterogeneous, then there would be little hope for new therapies to induce clinically important remission or cure," he said.

"These therapies at best would shrink only some of the metastatic lesions, while others would grow unabated. Fortunately, this does not appear to be the case," he said. He suggested that therapies that targeted the driver mutation would provide meaningful clinical outcomes.

In addition, the researchers conclude: "This pan-cancer analysis of untreated metastases suggests that a single biopsy accurately represents the driver gene mutations of a patient's metastases."

When asked how this is possible with a single biopsy sample, Reiter explained that a single biopsy seems to capture all likely key driver mutations.

He further indicated that bioinformatics analyses using available computational methods will provide clinicians information on how to identify the key functional driver mutations and separate them from others that may be in driver genes but that are predicted to not have functional consequences or are in passenger genes.

Vogelstein said: "Results suggest that a single biopsy from a primary tumor generally captures the information needed to determine whether the specific targeted therapy is indicated.

"If numerous biopsies from different parts of the tumors were always required to capture this information, the task for the clinician and the discomfort to the patient would be much more challenging," he added.

The analysis was carried out on biopsy specimens from treatment-naive patients. This is important because studies of treated samples cannot provide the same mechanistic insights of cancer evolution, Reiter noted. Once a patient has been treated, the samples may carry mutations resulting from treatment.

"We want to see if the idea of common functional drivers holds up when dealing with 20 to 30 different cancer types and hundreds of untreated samples," Reiter said in a Stanford news statement.

The study was supported by the National Institutes of Health, the Lustgarten Foundation for Pancreatic Cancer Research, the Sol Goldman Center for Pancreatic Cancer Research, the Virginia and D. K. Ludwig Fund for Cancer Research, an Erwin Schrödinger fellowship, a Landry Cancer Biology fellowship, and the Office of Naval Research. Dr Reiter has disclosed no relevant financial relationships. Dr Vogelstein is the cofounder of Personal Genome Diagnostics and is on the scientific advisory board of Sysmex-Inostics and Exelixs GP. These companies have licensed technologies from Johns Hopkins, and Dr Vogelstein receives royalties from those licenses. Relevant financial relationships of other authors are listed in the original article.

Science. Published online September 7, 2018. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.