A team based at Johns Hopkins hospital in the US have combined new DNA-based techniques with older assays for protein markers in blood to come up with a combination of tests that shows promising results.
The CancerSEEK blood test was used to detect the presence of mutations in just over 2,000 positions within 16 cancer-related genes. Immunoassays were also performed to measure the levels of 39 proteins, but only eight of these protein biomarkers were used in the diagnostic part of the test.
The researchers recruited 1,005 patients with stage I to III cancers of the ovary, liver, oesophagus, pancreas, stomach, colorectum, lung, or breast who had not yet undergone surgery. The most common stage at presentation was stage II (49% of patients). Stage II cancers are larger and may have spread to local lymph nodes. 20% of patients had stage I disease with small cancers that have not spread at all and 31% of patients had stage III disease where the cancer is more advanced. The ideal goal is to detect cancers early at stage I as they are then most likely to be cured by treatment. The study also included 812 healthy individuals (median age, 55 years) who had no history of cancer.
Overall the CancerSEEK test detected 73% of stage II cancers, 78% of stage III cancers, and just 43% for stage I cancers. For stage I cancers, the highest sensitivity was for liver cancer (98%), and the lowest was for oesophageal cancer (20%). The overall ability to find cancer in the population studied was 70% for the eight common cancers, which account for more than 60% of cancer deaths in the United States. The sensitivities ranged from 98% for ovarian and liver cancer to 33% for breast cancer.
In this study the test rarely had false-positive results i.e. it rarely found cancer that wasn't there. Seven of 812 healthy controls, or less than 1%, tested positive. Also by using the genetic and protein markers results from the diagnostic test plus the results from the 31 other protein markers measured, the researchers were able to localise the origin to either of two possible anatomic locations in a median of 83% of patients and to a single organ in a median of 63%.
The researchers believe that the test could be performed at a cost of about USD$500 if used on a large scale. This cost compares favorably with that of other screening methods such as mammography or colonoscopy which can only detect cancers in single locations.
These results are very promising but the test is still a long way from being ready for general use. The test needs to evaluated in studies that are closer to real-world usage. There are significant limitations of the currently reported study. The first of these is that the patients had already been diagnosed with cancer. Many or most of them would have had symptoms and thus would be different from the general population. This means that when the test is applied in screening the general population the sensitivity of the test is likely to be lower than reported here.
The second limitation is the control group used. These were healthy people. In the real world many of the people being screened with the test would have other diseases and it is known that many non-cancerous diseases can cause false-positive results in the protein marker tests that are being used. This means that in real-world usage the test is very likely to have a higher false-positive test result rate.
Further evaluation of the test has already started. The Geisinger Health System in Pennsylvania is collaborating with Johns Hopkins in a 5-year study of up to 50,000 women that is being funded by a private philanthropic group, The Marcus Foundation. The female volunteers between ages 65 and 75 who have never had cancer will be tested. Those who test positive twice will proceed to imaging studies to localise any cancers detected by the CancerSEEK test.
This long-term study will help answer the questions about real-world test sensitivity and specificity and other questions that arise when evaluating screening tests. Are all tumours detected by the test really significant cancers? Would they have gone on to cause harm to the patient if left undetected? Will testing lead to unnecessary treatment causing costs, potential harm and anxiety to patients? Will the testing and follow-up process be cost-effective?