KRAS Mutations in Ovarian Cancer -- Maybe Not

KRAS Mutations in Ovarian Cancer -- Maybe Not
Maurie Markman, MD

Authors and Disclosures

Posted: 04/15/2011

Learn more
The Role of KRAS rs61764370 in Invasive Epithelial Ovarian Cancer: Implications for Clinical Testing

Pharoah PD, Palmieri RT, Ramus SJ, et al
Clin Cancer Res. 2011 Mar 9. [Epub ahead of print]

Summary

In a previously published report,[1] investigators at Yale University noted a higher incidence of an uncommon KRAS oncogene variant (rs61764370) in approximately 500 women with ovarian cancer vs a noncancer control population. Particularly striking was the observation that 61% (19 of 31) of ovarian cancer patients in their series who had a family history of this malignancy but who did not have a documented BRCA1 or BRCA2 mutation were found to possess this KRAS variant.

In the current study, a multinational group of investigators examined data from patients in studies participating in the Ovarian Cancer Association Consortium and the Consortium of Investigators of Modifiers of BRCA1/2. In these much larger ovarian cancer patient (~ 9500) and control (~ 12,000) populations, they were unable to find any association between the presence of the rs61764370 KRAS variant and the development of ovarian cancer. In addition, there was no association between this genetic abnormality and familial ovarian cancer. Finally, there was also no evidence that this variant influenced either progression-free survival or overall survival. These investigators concluded that the use of this single nucleotide polymorphism in evaluating risk for the development of ovarian cancer "appears unwarranted."

Viewpoint
The association between a documented BRCA1 or BRCA2 abnormality and an increased lifetime risk for the development of epithelial ovarian cancer is well established.[2] However, there remains a number of families in whom ovarian cancer has been documented to occur in a pattern suggestive of a germline mutation but where no such genetic abnormality has been identified. As a result, the report that 61% of a small group of patients in this setting possessed a KRAS variant was potentially of great interest. Indeed, based on this report, a commercial test was fairly rapidly developed for ovarian cancer risk assessment.

Unfortunately, the new report involving a far larger sample size rather convincingly fails to confirm the earlier findings. As a result, there appears to be no current role (outside the setting of a well-designed research study) for examining any patient population in the assessment of ovarian cancer risk, or to use this biological marker as a strategy to evaluate prognosis.

This experience emphasizes once again the critical importance of validation of any reported association between a molecular or genetic abnormality and the risk for or prognosis of cancer. This is particularly relevant, as in the current discussion, if the original observation is made in a very small patient population where the potential for a spurious "statistically significant" outcome is quite real.

We are again reminded that while there may be an understandable rush to commercialize a novel concept that may be of genuine clinical value, it is absolutely critical that such ideas be rigorously and objectively evaluated before they are introduced into the marketplace.

KRAS Mutations in Ovarian Cancer -- Maybe Not