NEW YORK (GenomeWeb) – Patients with synchronous colorectal cancer — defined as having more than one primary tumor at the time of diagnosis — could be particularly difficult to treat with targeted therapies due to tumor heterogeneity, according to a new study.
Researchers from the National Center for Tumor Diseases in Heidelberg, Germany; the University Hospital Heidelberg; and the Technical University of Munich profiled tumors from 23 synchronous colorectal cancer patients and found that 87 percent displayed genetic heterogeneity among their primary tumors. They reported their results this week in the journal Genes, Chromosomes and Cancer.
"Testing one tumor or a single metastasis may not suffice in the [synchronous colorectal cancer] setting as clinically relevant and tumor-specific genetic information may be left undetected, compromising optimal oncological therapy," the authors wrote.
The researchers sought to study sCRC because the diagnosis and treatment of those patients has proven to be challenging. The molecular underpinnings of sCRC have not been studied extensively and a "comprehensive molecular picture" of sCRC is lacking, the authors wrote.
Although tumor heterogeneity is not a new phenomenon, and has been known to complicate treatment and diagnosis, the large proportion of primary tumors the researchers found to be genetically disparate in this study was "quite surprising," the authors wrote, and is "in contrast" to other studies that "reported a fairly homogeneous genetic make-up of both solitary primary CRC and their distant metastases."
For synchronous colorectal cancer, heterogeneity also means that taking a biopsy from just one primary tumor will not suffice as a way to decide which targeted treatments to use.
To try and understand the genetic makeup of sCRC tumors, the researchers studied 50 primary tumors and five metastases from 23 patients. To characterize the tumors, they analyzed MLH1 methylation status, microsatellite stability, and ran a custom-designed 30-gene hotspot colorectal cancer panel, using Thermo Fisher Scientific's AmpliSeq technology and Ion Torrent PGM.
Of the 23 patients, 19 had two primary tumors and four had three primary tumors. Two patients also had metastases. On average, the panel identified two somatic mutations per tumor. Twenty patients, or 87 percent, showed "striking genetic disparities," while just three patients had primary tumors with the same somatic mutations.
In addition, the genetic heterogeneity was observed in genes that would influence treatment choices. For instance, 10 patients had "considerable differences" in the status of the KRAS gene, with one tumor having a KRAS mutation and the other wild-type KRAS. One of those patients also had four different subclonal KRAS mutations in the same tumor.
As an example of the tumor heterogeneity, one patient harbored a KRAS and TCF7L2 mutation in one primary tumor, but had three different mutations in the second primary tumor, in the APC, CTNNB1, and SYNE1 genes.
Another patient, who had three primary tumors, had mutations in the same genes — for instance, two of the tumors had mutations to PIK3CA and two had mutations to APC — although the specific mutations differed between tumors. The patient also had some vastly different mutations. For instance, one tumor harbored a TP53 mutation, while the other two had wild-type TP53.
Four patients displayed microsatellite instability, which was discordant between the primary tumors for one patient.
Of the three patients who also had metastates, one had primary tumors that were genetically similar, and that patient's metastatis also had a similar genotype.
In a patient whose primary tumors had discordant KRAS status, the metastatis had the KRAS mutation as well as an APC mutation, and therefore, "diagnostic profiling results of this metastatic lesion are potentially misleading," the authors wrote, since the mutations in the metastasis may reflect one of the primary tumors but not he other.
The study shows that genotyping one tumor or a single metastatic lesion "may not be sufficient" and could add a challenge not only for molecular testing but also for treatment.
For instance, tumors with wild-type KRAS may be responsive to EGFR inhibitors, while patients with mutated KRAS will likely not respond.
Aside from KRAS, genetic heterogeneity affected other CRC-related genes that have a well-established causative role, including APC, PIK3CA, and TGFBR2.
The authors noted that additional studies should be done on more patients to fully understand the extent of the heterogeneity as well as the impact on treatment.