The Mystery Disease

Cancer is always a diagnostic and therapeutic challenge – but far more so when the origin of the primary tumor is unknown. By definition, carcinoma of unknown primary (CUP) is metastatic, and without knowing where the disease began, oncologists are often at a loss to suggest the most appropriate treatment. Although the current standard of care is a platinum-based combination therapy, the approach is often less effective than treatments for known cancers, buying patients only a few short months of life. In liquid biopsy, though, a research team from the University of California San Diego’s Center for Personalized Cancer Therapy, has found a potential answer – sequencing the DNA shed by the mystery tumors (1).

Why turn to liquid biopsy for CUP?

Patients with CUP generally have a poor prognosis, with a median survival of only six to eight months with standard-of-care approaches. We decided to investigate blood-derived circulating tumor DNA (ctDNA) to better understand the biology of CUP.

When patients are suspected of having CUP, the diagnostic team makes an effort to figure out in which organ the disease originated. Patients may undergo additional evaluations including esophagogastroduodenoscopy, colonoscopy, mammogram and CT imaging to find the primary cancer, as well as more high-tech approaches like a microarray-based assay used to predict the origin of a cancer. However, even with this extensive testing, the putative primary origin is only assigned in about one quarter of patients with CUP.

How does liquid biopsy for CUP work?

Liquid biopsy is used not to diagnose CUP, but to understand the molecular makeup of the cancer – which, in the eyes of many laboratory medicine professionals, is as valuable a “diagnosis” as any other. In our work, we currently employ a commercial assay that analyzes up to 73 genes at a time. The test isolates the tumor DNA shed by the cancer into the bloodstream and then applies next generation sequencing to assess the relevant genomic alterations. The implications of such a test for patients with CUP are clear – but its utility doesn’t end there.

Nowadays, for instance, many clinical trials with targeted therapies require patients to have certain genomic markers, many of which a simple blood test can identify. It’s true, of course, that there are other ways of testing for such markers; we could potentially use archival tissue or even obtain a new biopsy – but tissue-based approaches come with a host of challenges. A single sample cannot reflect tumor heterogeneity at multiple metastatic sites; archival tissue cannot reveal any changes that might have occurred as a result of therapy; new biopsies carry the risks inherent in any invasive medical procedure. Nevertheless, biomarker testing is vital – so blood becomes the new “go-to” method.

Of course, we still believe that testing for genomic markers in tissue is equally important. Blood-derived ctDNA has its own disadvantages – for instance, sensitivity is low, so liquid biopsy may not detect as many genes as tissue sequencing. In our clinic, we consider the results of tissue and blood-derived ctDNA sequencing complementary to each other. At the moment, neither reigns supreme – but nor can either one be omitted entirely.

What are your top tips for liquid biopsy?

Liquid biopsy to assess molecular alterations in blood-derived ctDNA is a valuable tool, but pathologists should take a clear-sighted approach to interpreting the results. Differences between ctDNA and tissue sequencing, for instance, should not be a de facto cause for concern regarding the technical validity of the tests; there are valid biological reasons to expect some variation between ctDNA and tissue-based genomic tests – even when the blood and tissue samples are obtained on the same day. Consider the two assays complementary to one another, as we do, rather than relying on either one in isolation, or panicking when the results are not identical.

At this moment, liquid biopsy with ctDNA can potentially be used to identify actionable genetic targets. In fact, testing ctDNA for EGFR mutations in patients with non-small cell lung cancer is FDA approved (2) – and as the technology improves, we expect ctDNA panels to expand to the point where they can test several hundred genes simultaneously.

ctDNA is an important addition to our armamentarium to better understand the underlying abnormalities driving cancer in our patients. Along with other routine tests, ctDNA is set to become a vital tool to guide treatment and selection of clinical trials for patients.

What’s next for the field?

As our understanding of genomics grows, we will probably learn more about the utility of variations of unknown significance (VUS) in the future. Right now, we do not know their functional significance – and, in general, we give most of our attention to characterized variants when deciding on therapy. We have recently shown, though, that the number of alterations in ctDNA can predict a patient’s response to immunotherapy in a manner similar to that of tumor mutational burden in tissue. In other words, patients with hypermutated ctDNA are more successfully treated with checkpoint inhibitor immunotherapy. When we count alterations in ctDNA to assess the hypermutated state, we include VUS – so even without understanding their effects, we still find value in detecting and enumerating such mutations. To that end, it’s vital for pathologists and laboratory medicine professionals to become experienced and knowledgeable in the genomics field.