Originally published Sept. 9.
NEW YORK (GenomeWeb) – Myriad Genetics this week presented data showing that its multi-gene myPath Melanoma test helped doctors change their treatment decisions, often lowering the intensity of the treatment strategy, for 35 percent of nearly 700 cases of pigmented skin lesions.
Myriad presented the data, from a second clinical utility study for myPath Melanoma, at the College of American Pathologists' annual meeting this week. The results of the prospective analysis aligned with findings from a retrospective clinical utility study presented earlier this year, which showed that doctors changed their decisions for 33 percent of cases using myPath Melanoma results.
Myriad is marketing the 23-gene panel test, list priced at $1,500, as a tool that doctors can use to determine if a patient's skin lesion is malignant melanoma or benign. Late last year the company provided early access to the test to dermatopathologists in the US through "The melEval Program." The test, according to Myriad, is one that pathologists can use to arrive at a more definitive diagnosis for suspicious, difficult-to-assess skin lesions. To date, approximately 120 dermatopathologists have submitted samples for testing with myPath through the early access program, according to the firm.
In the latest prospective clinical utility study, 42 dermatopathologists answered questions about their treatment decisions for 687 cases of pigmented skin lesions, before and after performing the myPath Melanoma test. Myriad researchers asked these doctors to document their level of confidence about their diagnosis (is the skin lesion benign, malignant, or indeterminate) and their treatment recommendations before and after the test was performed.
"The most encouraging results we've seen … is that for the cases that are indeterminate, and probably would be managed as melanoma by default because we can't be certain if they are benign or malignant, we saw a 76 percent reduction" after doctors considered myPath results, Loren Clarke, VP of medical affairs at Myriad Genetic Laboratories and lead researcher of the clinical utility study, told PGx Reporter.
Before receiving the test results, the doctors deemed 373 lesions as benign, 212 as malignant, and 102 as indeterminate. After they received the test results, the doctors recorded 427 cases as benign, 217 as malignant, and 43 as indeterminate. Of the initial 102 cases that the doctors diagnosed as indeterminate before myPath was performed, they changed 60 cases to benign and 18 to malignant, while 24 cases remained indeterminate.
"Interestingly, the majority of those cases went from indeterminate to a diagnosis of benign," Clarke noted. "That dovetails pretty well with what we know the problem is in dermatopathology, which is that there a tendency to overdiagnose melanoma for fear of missing melanoma and mismanaging the patient."
Doctors diagnose approximately 76,000 new melanoma cases annually and more than 9,000 people die from the disease, the American Cancer Society estimates. Although melanoma is considered the deadliest form of skin cancer, if caught early it is often curable. Due to improved diagnostic strategies, melanoma survival rates have increased from 49 percent between 1950 and 1954 to 92 percent between 1996 and 2003.
The effort to catch malignant skin lesions earlier and reduce mortality rates, however, has saddled healthcare systems with costs associated with unnecessary biopsies and overdiagnosis of otherwise benign lesions. Experts say that current diagnostic strategies aren't specific enough to discern which skin lesions truly need aggressive intervention and which don't because they are benign. Each year, pathologists analyze 1.5 million skin biopsies suspecting melanoma, and deem 14 percent of them as indeterminate.
In the clinical study presented by Myriad, doctors deemed 68 percent of samples as benign or malignant with high confidence. The number of highly confident determinate cases increased to 87 percent after doctors factored in myPath results. However, although myPath Melanoma reduced the number of indeterminate cases by 76 percent, there were still approximately 6.3 percent of biopsies that doctors remained uncertain about classifying, and 6.7 percent of samples for which doctors had a benign or malignant diagnosis but at a low level of confidence even with the help of myPath.
"When you have something that is difficult enough that you label it indeterminate, I think dermatopathologists are still extremely conservative and are still reluctant and are going to be reluctant for some time to label something very definitively," Clarke said. "Seeing a 76 percent reduction in indeterminate cases … is actually higher than I would expect knowing the level of caution that we exercise in finding these things out."
Additionally, doctors should use myPath alongside patients' clinical information and other tools to help them make treatment decisions. As such, even with the help of the molecular diagnostic, there might be cases that the doctor remains uncertain about. "The score tells them something, but there may also be something about the histopathology [of the skin lesion] that concerns them enough," Clarke said. Also, "these may be the [doctors] that don't have enough experience with the test to really understand the score entirely and the power of the assay."
Among early adopters, Clarke noted, it takes some time for doctors to warm up to the test and use the results to make more definitive diagnoses. "With any new test, what we're seeing is a buildup of confidence of these people," he noted, adding that Myriad doesn't expect this test to be used by doctors for every skin biopsy. "It's going to be used by dermatopathologists who know when they need it and when they don't, when it's indeterminate."
At the American Society of Clinical Oncology's annual meeting in June, Myriad presented data from the first clinical validation study involving the test. In that study, researchers from Myriad and elsewhere used myPath Melanoma to test 437 pigmented lesions in patient samples from four academic medical centers, and gauged how closely the test results matched diagnoses from expert dermatopathologists. The test showed a sensitivity of 90 percent and a specificity of 91 percent in differentiating between malignant melanoma and benign skin lesions.
There is substantial variability between how different pathologists might diagnose a patient after assessing the tissue from the patients' skin lesion. In the validation study, for example, there was between 30 percent and 40 percent discordance among experts about the diagnosis of a particular sample. After performing myPath, there was much more insight into the difficult-to-treat cases, and some lesions that experts thought were false negatives were reclassified as true negatives.
Myriad has five more clinical validation and utility studies underway involving the myPath test. The next clinical validation study currently ongoing will involve 1,000 skin lesions prospectively submitted to Myriad by early-access doctors. Researchers will send those blinded cases to three expert dermatopathologists, and compare the myPath results for the cases on which there is consensus among doctors.
"The only true gold standard [in melanoma] … has really been a pathologist looking at a tissue biopsy. But that's far from a hundred percent," Clarke said. A diagnosis of malignant melanoma or benign lesion is much more accurate when more than one doctor looks at the biopsy and they come to an agreement about the case.
"A truly 100 percent sensitive marker of malignancy is when a patient has metastatic melanoma, unfortunately," Clarke added. "So, ultimately we want to compare the [myPath] score to outcomes of patients who have known metastatic melanoma." Myriad is planning to do a study of this type, but prospectively following patients to see if they actually develop metastatic disease can take decades. "There are melanomas that metastasize quickly and those that remain dormant for more than 10 years sometimes," he observed.
It's more practical, Clarke said, to perform a retrospective study to gauge if the myPath results accurately differentiate between patients who are known to have the disease and those who don't. "We're in the process of trying to acquire cohorts that would have that type of outcome data," Clarke said. "It's difficult because [those cohorts] are few and far between and the tissue is old, so you have to count on there being some attrition in samples that have degraded RNA."
With these completed and ongoing studies, Myriad's aim of course, is to grow adoption of myPath and garner reimbursement for it. "We're planning on submitting our dossier for private and public payor reimbursement in the first or second quarter of 2015," Myriad spokesperson Ron Rogers said.