AMSTERDAM—A molecularly directed approach to making treatment decisions with patients who have metastatic cancers of unknown primary (CUP) was being urged here at the European Cancer Congress. This was conclusion of a study in which biomarkers predictive of drug response were identified in nearly 80 percent of patients with CUP by molecular profiling of cancer tissues (Abstract 39, accessible via http://eccamsterdam2013.ecco-org.eu/Scientific-Programme/Abstract-search.aspx#).
The traditional approach of using such biomarkers in tissue specimens to refine best estimates about the site of origin—and then tailor therapy according to a presumed primary site—should be abandoned, said Zoran Gatalica, MD, DSc, Executive Medical Director of Caris Life Sciences in Phoenix, Arizona, and Adjunct Professor of Pathology at Creighton University School of Medicine. Having CUP should not deny a patient individualized and targeted therapy, he said.
Up to five percent of all metastatic cancers have no identifiable primary site, he said, noting that guesswork about the primary tumor type can now be superseded by rational decision-making by spotting targets in the cancer capable of responding to specific drugs. “We have profiled about 1,400 cancers of unknown primary and have been able to identify prognostic and predictive biomarkers expressed in these tumors that can help fine-tune therapy to these patients,” he said in an interview.
Many useful actionable biomarkers are already well-known and drugs targeting them are already available, he continued. Instead of confining these drugs to cancers where the marker is believed to have been present on a presumed primary tumor, the drugs can be more successful when targets have definitely been identified in metastatic disease, he said. “For instance, topoisomerase expression can be targeted with drugs that inhibit topoisomerase—regardless of the primary site of the tumor.”
In the study, 77 percent of patients turned out to have biomarkers that could be targeted by known drugs. These included protein overexpression, protein loss, activating mutations, and gene copy number variations. The researchers also found “disease-defining gene mutations” and in one case they were able to adjust a diagnosis of “metastatic melanoma of unknown primary” to instead be primary leptomeningeal melanocytoma.
Gatalica also listed metastatic tumors that overexpress HER2—originally targeted in breast cancer but now proving useful in colorectal and gastro-esophageal junction cancers—among the candidates for rational treatment planning following molecular profiling of CUPs: “Finding an overexpression of HER2 in a carcinoma of unknown primary site might help decide on trastuzumab therapy in those cases,” he noted.
He also suggested that finding an activating mutation in the epidermal growth factor receptor could direct the oncologist's attention, for example, to specific tyrosine kinase inhibitors (TKIs) for some patients.
‘Classical Chemotherapy’ Rationally Used
Gatalica said that quite apart from the new molecular treatments, optimum classical chemotherapy—with or without a targeted treatment—can also be recommended on a rational basis if the predictive biomarkers in a CUP are identified: “Cancers of unknown primary should be profiled for the biomarkers of the known drug therapies. Regardless of their site they are still solid tumors, and mechanistically, they will probably respond the same way as tumors of a known primary site,” he noted.
In an interview, Cora Sternberg MD, FACP, Professor and Chief of Medical Oncology at the San Camillo and Forlanini Hospitals in Rome, who chaired a news conference at which these data were discussed, called the research a major step forward.
“In the typical such situation, looked for all of their tumors, we've done gastroscopy, colonoscopy, CT scans, mammography, and after we've done all of this we still don't know what the primary is!” She said she approved of the fresh approach Gatalica suggested—i.e., to just forget about the primary and instead check the molecular profiles of biopsy specimens to find targets for individual patients.
“What we've been doing in the past is telling patients: We don't know where your tumor comes from, that we're going to use our best guess and give you a chemotherapy that might work for any tumor,” she remarked.
She said she was concerned about the “psychological toxicity” of such a vague approach on patients and their families—not to mention the cost of giving useless chemotherapy and enduring side effects needlessly. “This is a major breakthrough and this is the largest series ever of patients of unknown primaries,” she added.
Sternberg emphasized that these patients need personalized therapy: “In the past we said this tumor comes from the colon, or this tumor comes from the ovary, and the chemotherapies would be completely different. So it was really difficult as an oncologist trying to guess what would work for both of these tumors.”
She said that despite the heterogeneity of tumors, she strongly supported the idea of checking to see what could be driving any patient's particular tumor: “To have an idea of what might work for that patient is extremely important and is the way we need to go in the future.”
ECCO President Cornelis van de Velde, MD, PhD, Professor of Surgery at Leiden University, commented, “These encouraging results show that even in cases where the primary site is not known it is increasingly possible to select effective targeted treatment and hence improve survival times.”
Zoran Gatalica, MD, DSc, discusses more about his study of 1,400 patients with cancers of unknown primary site in whom 80 percent were found to have actionable molecular targets in a video interview at the Congress with Sarah Maxwell.
If you are not yet receiving our iPad issues, download the free Oncology Times app from the App Store today! Visit http://bit.ly/OT-iPadApp, search in the App Store, or follow the link on oncology-times.com.