Genetic Profiling of Cancers of Unknown Primary Improves Tx

Roxanne Nelson

September 29, 2013

AMSTERDAM ― If the primary tumor site is unknown, identifying the molecular profile of the metastatic tumor is essential to choosing appropriate therapy, and this is becoming increasingly possible, a US researcher reported here at European Cancer Congress 2013 (ECCO-ECMO-ESTRO).

"We are absolutely sure that just finding the tissue of origin and the histological subtype is truly insufficient to guide therapy," Zoran Gatalica, MD, DSc, executive medical director, Caris Life Sciences, Phoenix, Arizona, commented at a press briefing. "There is definitely a need for finding genetic alterations that could prove useful in selecting targeted therapies."

He was reporting a study in which his team identified actionable biomarkers in a high proportion (77%) of more than 1350 cases of cancers with an unknown primary site. The actionable biomarkers include targeted protein overexpression, such as steroid receptors and MET, protein loss, including PTEN, and activating mutations (EGRF, BRAF, PiK3CA,) and gene copy number variations (HER2, TOP2A and MET amplification).

This is an extremely important study, commented Cora Sternberg, MD, chief of the Department of Medical Oncology and chairman of the Division of Medical Oncology at the San Camillo and Forlanini Hospitals in Rome, Italy. "This is the largest group of patients with unknown primaries ever to be studied and characterized."

"At least 10% of our patients have cancers with unknown primaries, and it is very difficult to discuss this with these patients and their families," said Dr. Sternberg, who moderated the press briefing. "These patients have had an adequate workup, but in some, we just cannot find the primary tumor site."

Improving Therapy Selection

Traditionally, when the site of the primary tumor is unknown, the choice of treatment has largely depended on identifying biomarkers that are indicative of a potential site of tumor origin, in order to better tailor therapy, Dr. Gatalica explained.

However, this approach can only provide statistical odds of one site being preferred over another. For the most part, this strategy does not address nor provide information on the actionable biomarkers that can be targeted for effective treatment, he said.

Dr. Zoran Gatalica

In their study, they reviewed profiling data on patients with cancers of unknown primary site from a large referral laboratory in order to look for information on biomarkers associated with potential for pharmacologic responses.

They employed multiple methods, including mutation analysis, in-situ hybrization, immunohistochemistry, and fragment analysis.

"We found that cancer of unknown primary was more common in women," he said, "And the average age was 61 years. Only a handful of patients were under the age of 20 years."

One example of successful profiling, he cited, was a woman with an unknown primary who had extensive bone disease on her left side. She was profiled at Dr. Gatalica's institution and was found to be carrying an activating mutation in epidermal growth factor receptor (EGFR). On the basis of that finding, she was placed on treatment with the EGFR targeted therapy erlotonib (Tarceva, Osi Pharmaceuticals).

This is not the type of therapy that would usually be considered in such a case, he commented.

"Four months into that therapy, she had a remarkable, near complete response," Dr. Gatalica said. "And that was a drug that would normally not be considered in a case of cancer with an unknown primary."

In other cases, profiling allowed for the reclassification of disease. For example, identifying the somatic GNA11 mutation resulted in the reclassification of metastatic melanoma of unknown primary to primary leptomeningeal melanocytoma.

"When we took all of our cases together, we were pleasantly surprised that we were able to identify biomarkers predicting responses in almost 80% of cases," he said. "Only about 18% of cancers, after extensive profiling, still remained without any identifying target."

More importantly, Dr. Gatalica noted, they were able to find conventional chemotherapy targets as well as biological targets. This provided "an option for drugs that may not be as expensive as one may think," he pointed out.

"A CUP [cancer of unknown primary] diagnosis is challenging for patient and clinician, and a search for the primary site, to define treatment options, is often distressing and fruitless," said John Symons, director of the CUP Foundation, in a statement. "Using predictive biomarker information directly from the tumor offers doctors powerful insights into the best treatment options for CUP patients."

European Cancer Congress 2013 (ECCO-ECMO-ESTRO). Abstract LBA 39. Presented September 29, 2013.


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.
Post as: