2011 SGO News: MicroRNA May Help Predict Ovarian Cancer Response

Changes in expression of a microRNA showed potential for predicting response to the angiogenesis inhibitor bevacizumab (Avastin) in recurrent serous ovarian cancer, according to results of a small study reported here.

Women with low-level expression of miR-378 after bevacizumab and chemotherapy had a median progression-free survival (PFS) more than twice as long as that of patients who had normal miR-378 expression.

Moreover, 60% of patients with low miR-378 expression had a PFS of at least six months compared with none of the patients with normal expression, researchers reported at the Society of Gynecologic Oncology meeting.

The association between miR-378 expression and response was specific to bevacizumab and PFS, said former OCRF grantee John K. Chan, MD, of the University of California San Francisco. Separate analyses showed that changes in the microRNA’s expression did not predict response to chemotherapy or overall survival.

“Differential microRNA expression may have clinical utility in determining response to novel therapeutics,” said Chan.

MicroRNA is small, non-coding, single-stranded RNA that regulates gene expression, has been implicated in survival, and has been shown to predict response to chemotherapy in several types of cancer. However, no studies have produced information about the ability of microRNA to predict response to biologic therapy, said Chan.

In an effort to address the lack of information, investigators analyzed data provided by the Cancer Genome Atlas. Beginning with clinical and genetic information on 435 patients with ovarian cancer, they excluded patients who did not have recurrent disease or who did not receive bevacizumab plus chemotherapy at recurrence.  The exclusions left 34 patients for the analysis.

The investigators evaluated microRNA expression in a test group of three patients who had good responses to bevacizumab and chemotherapy.

On the basis of the findings in the test group, Chan and colleagues searched for microRNAs associated with a PFS of at least nine months (responders) and that exhibited at least a twofold difference in expression in cancer tissue compared with normal tissue.

Of 800 microRNAs in the Cancer Genome Atlas database, 33 satisfied the inclusion criteria.

The investigators examined the impact of microRNA expression on the likelihood that a patient would achieve a PFS of at least six months.

A preliminary analysis identified five microRNAs that had significant associations with six-month PFS. After correction for multiple comparisons, only miR-378 remained a significant predictor (P=0.005).

Chan and colleagues also examined various clinical factors associated with response to bevacizumab and chemotherapy and subsequent PFS. They found that the treatment-free interval after the last therapy before bevacizumab plus chemotherapy predicted PFS after therapy.

Specifically, patients with a treatment-free interval of at least one month had a significantly better PFS compared with patients who had treatment-free intervals of less than one month (P=0.02).

In a multivariate analysis, both the treatment-free interval and miR-378 expression remained significantly associated with PFS.

By means of gene target scans, Chan and colleagues identified three genes that appear to be targeted by miR-378. All are involved in activities that are associated with promotion or inhibition of angiogenesis.

Future work will be directed toward elucidating mechanistic explanations for the association between miR-378 expression and response to bevacizumab.