Gene tests predicts melanoma staging

The gene expression profile DecisionDx-Melanoma test (Castle Biosciences) is as effective as the more invasive sentinel lymph node biopsy at identifying which stage 1 and 2 tumors would turn deadly in five years. This is according to results from a new multicenter performance study of cutaneous melanoma tumors from 334 patients, which Castle Biosciences presented in early June at the American Society of Clinical Oncology (ASCO) annual meeting in Chicago.

The investigators also reported that physicians providing follow up care for patients diagnosed with class 1 or high risk class 2 melanoma are overwhelmingly changing care based on the patient’s relative risk for future metastasis as defined by the gene test, along with other measures such as tumor thickness and whether a sentinel lymph node biopsy is positive or negative.

Findings

While Breslow’s thickness, ulceration, mitotic rate, sentinel lymph node (SLN) status and results of the gene expression profile test were significant predictors of distant metastasis risk, only the gene expression profile test followed by SLN were independent predictors of the risk. 

Accuracy of distant metastasis risk prediction by the gene expression profile test showed 76% sensitivity, 55% specificity, 42% positive predictive value and 85% negative predictive value, compared to 72% sensitivity, 64% specificity, 45% positive predictive value and 84% negative predictive value for SLN status.

NEXT: Further independent findings

Independent findings

The investigators also reported on an independent, prospective study in cutaneous melanoma involving 174 consecutively diagnosed melanoma patients undergoing sentinel lymph node biopsy (SLNB) and testing with the DecisionDx-Melanoma test of their primary tumors. Investigators reported 159 of those patients were analyzed with the genetic test. At a median 12-month follow-up, SLN was positive for 20 patients, or 13 percent, and negative for 139 patients, or 87 percent. Seven, or 35 percent, of 20 SLN positive patients and seven, or five percent, of 139 SLN negative recurred within two years. Among the gene expression profile test results: 42 were high-risk class 2 and 117 patients were low-risk class 1. Thirteen, or 31 percent, of the 42 Class 2 patients, and one of the 117 class 1 patients recurred within two years. Of 10 patients with both a Class 2 gene test status and positive SLN, seven, or 70 percent, recurred.

In a third abstract, investigators reported on a multicenter, decision impact study on patients tested with DecisionDx-Melanoma at six medical practices from May 2013 to September 2015. They reported that results showed that the test outcome changed clinical management for 52 percent of the patients tested. Of the 80 cases with observed changes in management, 95 percent were adjusted in a manner consistent with the class prediction (lower intensity for class 1 patients and higher intensity for class 2 patients).

The DecisionDx-Melanoma test’s accuracy and utility has been studied in more than 900 melanoma patients’ tumors and studies have confirmed the test’s accuracy in predicting metastatic risk, regardless of the size of the initially non-metastatic tumor. The test uses tumor samples from biopsy or tumor excision, according to company press materials. 

Desiree Ratner, M.D., director, Comprehensive Skin Cancer Program, Mount Sinai Beth Israel and professor of dermatology at the Icahn School of Medicine at Mount Sinai, says the results of the multicenter performance study looking at the impact of gene profiling on clinical decision making in patients with cutaneous melanoma are very exciting. 

“Patients with a negative sentinel lymph node biopsy are normally classified as having a low risk of nodal or distant metastasis,” says Dr. Ratner, who is not a study author. “However, patients with high-risk class 2 test results, despite a negative lymph node biopsy, have been shown to have a higher risk of melanoma progression. The [gene expression profile] test outcomes actually changed clinical management of this high-risk group. These high-risk patients can now be identified earlier, and followed more frequently, so that if they develop metastatic disease, appropriate intervention can be initiated earlier in their disease course.”