Promising ‘liquid biopsy’ detects and identifies tumors with a blood test

By Robyn Boyle, RPh, for MDLinx
Published February 12, 2018

Key Takeaways

Researchers have developed a blood test called CancerSEEK that can detect eight common solid tumors, including some highly lethal forms that currently have no screening tests. Results from the NIH-funded investigation team, led by Joshua D. Cohen from Johns Hopkins University School of Medicine in Baltimore, MD, were published in Science.

The test measures levels of circulating proteins and mutations in cell-free DNA, which are non-encapsulated DNA originating from tumor clones.

“The test utilizes combined assays for genetic alterations and protein biomarkers and has the capacity not only to identify the presence of relatively early cancers but also to localize the organ of origin of these cancers,” the authors wrote.

Blood tests for cancer must be very specific; otherwise, too many healthy individuals will receive false positive results, leading to unnecessary procedures as well as anxiety. Liquid biopsies can have high specificity because they are based on driver gene mutations that are usually found only in abnormal clonal proliferations of cells, such as cancers.

Diagnostic sensitivity and identification of the underlying tissue of origin are other important issues. Because the same gene mutations drive multiple tumor types, liquid biopsies based on genomic analysis alone generally cannot identify the anatomical location of the primary tumor.

CancerSEEK has the capacity to identify the presence of relatively early cancers and to localize the organ of these cancers. The study involved 1,005 patients with non-metastatic, clinically detected cancers of the ovary, liver, stomach, pancreas, esophagus, colorectum, lung, or breast. Blood samples were collected before patients received any neo-adjuvant chemotherapy. The median age at diagnosis was 64 years (range 22 to 93).

The control cohort consisted of 812 healthy individuals of median age 55 years (range 17 to 88) with no known history of cancer, high-grade dysplasia, autoimmune disease, or chronic kidney disease.

CancerSEEK evaluates levels of 8 proteins and the presence of mutations in 2,001 genomic positions; each genomic position could be mutated in several ways. The presence of a mutation in an assayed gene or an elevation in the level of any of the proteins would classify a patient as positive. Rigorous statistical methods were used to ensure the accuracy of the test.

The median sensitivity of CancerSEEK among the eight cancer types evaluated was 70%, and ranged from 98% in ovarian cancers to 33% in breast cancers. In the five cancer types (ovary, liver, stomach, pancreas, and esophagus) for which there are no screening tests available for average-risk individuals, sensitivity ranged from 69% to 98%.

At this sensitivity, the specificity was > 99%; 7 of the 812 individuals without known cancers were positive, although it is uncertain if some participants labeled as ‘false-positive’ had a cancer not yet detected.

An important attribute of a screening test is the ability to detect cancers at relatively early stages. The median sensitivity of CancerSEEK was 73% for stage 2, 78% for stage 3 cancers, and 43% for stage 1 cancers. The sensitivity for stage 1 cancers was highest for liver cancer (100%) and lowest for esophageal cancer (20%).

Liquid biopsy is based on the idea that mutant DNA templates in plasma are derived from dying cancer cells and serve as highly specific markers for neoplasia. Tumor tissue was evaluated in 153 patients with identified primary tumor and statistically significant levels circulating tumor DNA (ctDNA).

In 138 of 153 cases (90%), the mutation in the plasma was identical to a mutation found in the primary tumor. This concordance between plasma and primary tumor was evident in all eight cancer types, and ranged from 100% in ovarian and pancreatic cancers to 82% in stomach cancers.

CancerSEEK was also able to localize the source of the cancer to two anatomic sites in a median of 83% of these patients and to localize the source to a single organ in a median of 63% of these patients.

Most localization information was derived from protein markers, as driver gene mutations are usually not tissue-specific. The accuracy of prediction varied with tumor type; it was highest for colorectal cancers (84% and 100% for the top two predictions, respectively) and lowest for lung cancers (44% and 63% for the top two predictions, respectively).

The authors noted that the study had some limitations. First, the patient cohort in the study had known cancers and most were diagnosed on the basis of symptoms. In a routine setting, the sensitivity of detection is likely to be lower than reported in the study. In addition, patients with inflammatory or other diseases could have more false positives than healthy controls in this study. The investigators were not able to use a completely independent set of cases for testing.

Finally, the authors wanted to test as many cancers as possible. With their limited resources, the proportion of cancers of each type in the cohort was not representative of incidence in the US. They estimated the sensitivity of CancerSEEK to be 55% among all eight cancer types, when weighted for actual incidence in the US. However, the high sensitivities of CancerSEEK would not be affected.

“Our study lays the conceptual and practical foundation for a single, multi-analyte blood test for cancers of many types,” the authors concluded. “We estimate the cost of the test to be less than $500, which is comparable or lower than other screening tests for single cancers, such as colonoscopy. The eight cancer types studied here account for 360,000 (60%) of the estimated cancer deaths in the US in 2017, and their earlier detection could conceivably reduce deaths from these diseases.”

They also point out that this test would not be not meant to replace other non-blood based screening tests, but to provide additional information. Furthermore, large prospective studies will be needed to evaluate all cancer types.

To read more about the study, click here

Share with emailShare to FacebookShare to LinkedInShare to Twitter