The question of whose DNA or RNA was used to develop a complex genetic test, such as those that use AI to identify patterns of genetic variants or gene expression levels, is fundamental to its applicability—the people most like those who were used to develop the test are the most likely to benefit.
In the U.S., it’s no secret that genomics is heavily skewed toward White males of European descent, also known as non-African-American males (NAAMs). So, it is highly likely that these genetic tests are best suited for White American males, who make up the bulk of the genomes and transcriptomes stored in genetic databases and continue to dominate clinical studies.
Yet, for some indications, White American males are not necessarily the population most likely likely to be affected by and die from the condition the genetic test is being developed for other populations. That’s the case in prostate cancer, where Black men are 1.7 times more likely to be diagnosed with and 2.1 times more likely to die from prostate cancer than White men; Black men are also slightly more likely than White men to be diagnosed with advanced disease.
That’s why Veracyte is testing their Decipher Prostate Genomic Classifier, an AI-based transcriptome genetic test that was developed using primarily White American males, on Black African American males to see if it offers clinicians an improvement over clinical factors alone in guiding prostate cancer treatment decisions.
“Only about a third of men with prostate cancer are getting any kind of molecular diagnostic based on our calculations as of the beginning of this year,” Veracyte CEO Marc Stapley told Inside Precision Medicine. “[Black American men] are 70% more likely to be diagnosed with prostate cancer and twice as likely to die from the disease, and there are lots of reasons for that. Their care quality is a major factor.”
Today, at the annual meeting of the American Society for Radiation Oncology (ASTRO), in Washington, D.C., Veracyte announced results from their prospective multi-center clinical trial, called the Validation Study on the Impact of Decipher Testing in African American Men (VANDAAM). The study found that Decipher accurately predicts aggressive prostate cancer among African American men with early-stage disease, confirming that the market-leading genomic test offers clinicians an improvement over clinical factors alone in guiding prostate cancer treatment decisions for White and Black American men.
“We must be able to give access to Decipher into those patient populations because the data can then be used to encourage more of those patients to be treated,” said Stapley.
Correcting equity gaps with prospective studies
The Decipher Prostate Genomic Classifier is a 22-gene test that uses RNA whole-transcriptome analysis and machine learning to guide treatment decisions for prostate cancer patients. The test accurately predicts the likelihood of metastasis development with conventional treatment and is performed on samples taken during a biopsy or after surgery. With this data, doctors can tailor their patients’ treatment plans to their specific needs, suggesting less drastic measures for those with a lower risk of metastasis and earlier, more drastic measures for those with a higher risk.
Even though the Decipher score was developed with a predominantly White cohort, its utility in retrospective studies has been comparable between African American and non-African American men, even demonstrating similar accuracy in predicting metastasis between African American men and non-African American men. To that end, a prospective evaluation of the Decipher score may help identify aggressive tumor subtypes in African American men more accurately, which in turn could lead to more effective targeted treatments.
“We’re trying to correct gaps created over many years,” said Stapley. “There’s a couple of ways to come at that, including running prospective studies to represent populations.”
1-to-1 population matching
The VANDAAM trial, led by Kosj Yamoah, MD, PhD, of the H. Lee Moffitt Cancer Center and Research Institute, included men with low- or intermediate-risk prostate cancer, as defined by the National Comprehensive Cancer Network (NCCN) Guidelines for Prostate Cancer. Yamoah’s group used the Decipher Prostate classifier to create Decipher risk scores after analyzing data from a clinically balanced cohort of 226 men, with 113 African Americans and 113 non-African Americans.
“What I like about what [Yamoah and his team] did with VANDAAM was that they made a very concerted effort to do a one-to-one matching to make sure that they were truly representative and equitable and equal in many respects to the population that they were comparing,” said Stapley. “That matters because African Americans are underserved and not getting the test as much as European men. So, one of the things you have to answer is if the test is going to serve those patients in the same way that it does the patients who are getting treated.”
Decipher serves both White and Black American males
High Decipher test scores were associated with an eight-fold increased risk of biochemical recurrence (BCR; i.e., an increase in blood levels of prostate-specific antigen, a surrogate for aggressive disease) within two years after standard treatment compared to lower genomic test scores. Even after controlling for factors like age, race, and pre-treatment prostate-specific antigen levels (PSA), which predicts the response of prostate cancer to local therapy, patients with high Decipher test scores still had a very high chance of developing rapid-onset BCR. The results showed no significant difference between African American and non-African-American men.
“[VANDAAM] showed that… Decipher does not differentiate between those populations—it serves them both equally,” said Stapley. “A high-risk Decipher score is as impactful for African-American men as it is for non-African-American men. Whatever the descent, decipher high has a significantly higher risk of recurrence. The bottom line is that [Decipher] should be used in those populations.”
Closing the disparity gap in genomics equity
Stapley emphasized the importance of accelerating the testing of African American men with Decipher, expanding access to the test, and creating treatment plans that consider all relevant factors to mitigate the devastating outcomes. Stapley said that he wants to get Veracyte to the point where they can provide a molecular test for the entire population of prostate cancer in the U.S., which sees 300,000 new incidences per year. To do so, Veracyte is testing Decipher in patients in different buckets for prostate cancer risk, undergoing prostatectomy, and experiencing metastasis.
“Veracyte will continue to do this and make sure that when we do studies, we always keep underrepresented populations in mind and do not leave them out in the cold,” said Stapley. “You’re going to see us do that across the board in all our studies in the future. We’ve got over a dozen studies on Decipher in disparate and underrepresented populations, which is very important to close that equity disparity gap. That is something that the whole industry needs to embrace in whatever indication, whether prostate, bladder, lung, or colon cancer. We need to ensure that new studies don’t start with those gaps and that we have a good representation.”
