Skip to main content
Premium Trial:

Request an Annual Quote

Study Finds Tumor Heterogeneity May Contribute to Drug Resistance

NEW YORK (GenomeWeb) – Tumor heterogeneity in metastatic colorectal carcinoma patients may contribute to drug resistance, according to a study published in the Annals of Oncology.

Researchers from the Isituto Nazionale Tumori Fondazione Giovanni Pascale in Italy studied tumor profiles of 182 patients enrolled in CAPRI-GOIM, a clinical trial in which patients are assessed for KRAS status and treated with FOLFIRI plus cetuximab as a first line therapy and after progression randomized to receive either FOLFOX or FOLFOX and cetuximab.

In a previous study, the group found significant intra-tumor heterogeneity in patients. However, the goal of the most recent study was to determine the clinical impact of that heterogeneity.

For instance, the researchers wrote that they aimed to see whether there is a threshold of heterogeneity within KRAS mutant tumors to "separate tumors resistant to EGFR monoclonal antibodies from those that might be responsive to these agents." Another goal was to determine whether mutations in other genes contributed to the development of resistance.

They used Thermo Fisher Scientific's Ion AmpliSeq Colon and Lung Cancer panel, which evaluates 22 genes, although they focused their analyses on four: KRAS, NRAS, BRAF, and PIK3CA — evaluating mutational heterogeneity in each.

Mutations in KRAS and NRAS confer resistance to anti-EGFR monoclonal antibodies and are found in approximately 60 percent of all metastatic colorectal cancer tumors. In addition, mutations in BRAF and PIK3CA have been found in 5 to 10 percent and 15 to 20 percent, of metastatic colorectal tumors, respectively, although their role in anti-EGFR monoclonal antibody resistance is still debated.

After running the targeted panel, each sample was given a heterogeneity score for each of the four genes based on the frequency of mutant alleles. A score of 100 meant that all tumor cells had the mutation. A score greater than 100 meant that there were also copy number variants — either gain of the mutant allele or loss of the wild type.

Of the 182 samples, 78 had at least one mutation in the four genes, including 45 with KRAS mutations, 24 with PIK3CA mutations, 15 with BRAF mutations, and 13 with NRAS mutations. KRAS and NRAS mutations were mutually exclusive, while PIK3CA and BRAF mutations were detected in both RAS wild type and mutant tumors.

Next, the researchers determined a heterogeneity score for each sample and each mutated gene. For KRAS, heterogeneity ranged from 12 to 260 with a mean of 87.1 and with 60 percent of tumors having a score greater than 70. For NRAS, 77 percent of cases had a score greater than 70, while the mean score was 102.8. "These data suggest that RAS mutations are likely to occur at early stage of [colorectal cancer] tumorigenesis," the authors wrote.

In contrast, for BRAF, only 26.7 percent of samples had a score greater than 70, and the mean score was 54.8. Similarly, only 33.3 percent of PIK3CA mutants had a score greater than 70, with a mean of 59.5.

The researchers then assessed whether classifying tumors based on heterogeneity could determine whether a sample would be responsive to the drug, reasoning that tumors with less than a third of KRAS mutant cells might benefit from the anti-EGFR therapies.

Of the 10 patients with scores less than 33, seven, or 70 percent responded to the first-line therapy, while only 45.7 percent of patients with a KRAS score greater than 33 responded. Interestingly, though, median progression-free survival was comparable between the two groups — a "peculiar trend," the authors wrote.

The data suggest that "the presence of a low fraction of cells carrying a resistance mutation may not prevent the response to a specific drug, but the duration of the response will be shorter because the resistant clone will expand rapidly and will cause the recurrence of the disease," the authors wrote.

The authors also found differences between the groups of patients with high and low KRAS scores with respect to their mutational landscape in other genes. For instance, the frequency of PIK3CA mutations was higher in low KRAS mutant tumors compared to high KRAS mutant tumors, with PIK3CA mutations in six out of 10 of the low KRAS samples and in eight out of 35 in the high KRAS cases.

These findings further question what is driving the EGFR drug resistance. Further work will need to be done to determine whether resistance is driven "by the low level KRAS mutant; by the mutations in PIK3CA, BRAF, or even other genes not included in our panel; or by a cooperative effect of these pathways, which is likely to be the case," the authors wrote.