Prognostic test is standard of care in eye cancer
One of the primary challenges for doctors in managing uveal melanoma, as well as all cancers, is predicting the potential for the disease to spread. Fortunately, patients today have access to the DecisionDx-UM gene expression profile (GEP) test, which predicts a patient’s risk based upon their tumor’s own unique biology.
Developed by ocular oncologist Dr. J. William Harbour while at Washington University (now at Bascom Palmer Eye Institute) and exclusively licensed to Castle Biosciences Inc., DecisionDx-UM is the most widely used prognostic test in uveal melanoma. It is used to predict statistical rates of survival over five years, the period for which we currently have scientific data.
The GEP test is extremely valuable in guiding a patient’s care following treatment of the original eye tumor. It measures the activity or “expression” of certain genes within the tumor to determine its risk profile, or Class:
- Class 1A: Very low risk, with a 2% chance of the eye cancer spreading over five years;
- Class 1B: Low risk, with a 21% chance of metastasis over five years; and
- Class 2: High risk, with 72% odds of metastasis over five years.
Historically, doctors relied on physical characteristics like tumor size, location, and color patterns to determine metastatic risk. These factors still play a role, but alone, are not accurate predictors. In the 1990s, this method was improved upon by tests that analyze a chromosome mutation. Called Monosomy 3, these tests identify tumors with a loss of one copy of chromosome 3, which is associated with a higher risk of metastasis than those with both copies intact. In a head-to-head study however, Monosomy 3 testing proved to be far less accurate than DecisionDx-UM.
For these reasons as well as the scientific rigor that went into its development, the GEP test has been adopted by the majority of the country’s leading ocular oncologists (100 of the estimated 110 specialists) as standard of care in the management of eye cancer.
DecisionDx-UM: Rigorously Validated for Accuracy, Reliability
The DecisionDx-UM test is the only test that has undergone extensive validation in multiple studies, including an independent, prospective study conducted by the Collaborative Ocular Oncology Group (COOG). In 494 patients at 12 centers in the United States and Canada, the researchers found the test could successfully classify tumors more than 97 percent of the time. The GEP test was found to be superior in predictive accuracy and reliability to all other prognostic methods. The COOG study results were published in Ophthalmology in August 2012.
Formal comparisons (using a method called net reclassification improvement) at the three-year mark of the study showed the DecisionDx-UM test to have a 43% improvement over physical characteristics like tumor size and shape, and a 38% improvement over Monosomy 3 testing.
In other studies, Monosomy 3 testing has been documented to technically fail to return a result in up to 50% of specimens tested (Young, 2007; Desjardins, 2010). Getting a result the first time is extremely important in uveal melanoma, since timing of the biopsy is critical.
BAP1 Testing for Inherited Risk
Mutations of BAP1, a gene located on chromosome 3, can either be inherited or developed by the uveal melanoma tumor. Drs. Harbour and Anne M. Bowcock at Washington University discovered a link between metastasis and BAP1 mutations in 2010, and published data regarding the correlation between BAP1 mutations, metastasis, and the DecisionDx-UM test. Castle Biosciences’ own analysis confirmed the researchers’ finding that DecisionDx-UM was superior to BAP1 testing in predicting metastatic risk in uveal melanoma.
Castle is completing validation for BAP1 testing, and based on current data, believes initial clinical use will be used to assist in identifying inherited BAP1 mutations and will not be a replacement for the more accurate DecisionDx-UM test. If you are interested in learning when BAP1 testing becomes available, please email firstname.lastname@example.org.