New Test Identifies Early-Stage Melanoma With Metastatic Risk

Alexander M. Castellino, PhD

January 15, 2015

Because two-thirds of melanoma patients who die or experience metastatic disease are initially diagnosed with early-stage disease, it is important to determine risk for metastasis in patients with early-stage melanoma. About 75% of patients with melanoma have early disease (stage I or II) at diagnosis.

A prognostic 28-gene signature (DecisionDx-Melanoma, Castle Biosciences) could predict which patients with stage I or II melanoma are at high risk for metastasis, and could alter the clinical management of these patients, according to a confirmatory study published in the January 1 issue of Clinical Cancer Research.

The 28-gene signature predicts a patient's risk for metastasis, independent of American Joint Committee on Cancer (AJCC) melanoma staging, Breslow thickness, ulceration, mitotic rate, and age.

"The behavior of melanoma tumors is highly variable, and often cannot be accurately predicted using traditional staging methods," said senior author Pedram Gerami, MD, professor of dermatology and director of melanoma research at the Northwestern Skin Cancer Institute at Northwestern University in Chicago, in a press release from the manufacturer.

"These results demonstrate that a gene signature can accurately predict metastasis, particularly in tumors that were assumed to be lower risk due to stage, size, and other characteristics. This test can have a significant impact on the management and potentially long-term outcomes of these melanoma patients," he added.

The dataset supporting the 28-gene signature is small, but the data are compelling, said Jeffrey J. Sussman, MD, professor of surgery and chief of the division of surgical oncology at the University of Cincinnati School of Medicine, who was not involved in the study.

In his clinical practice, he orders the test for patients with node-negative disease with higher risk factors (stage IIa/IIb) in an effort to stratify risk for metastasis, he told Medscape Medical News.

Study Details

In the study, gene-expression patterns (GEP) were analyzed in formalin-fixed paraffin-embedded specimens obtained from biopsies or wide excision procedures of primary melanomas.

RNA from the samples was converted to complementary DNA, amplified, and loaded on to a chip containing "28 discriminating gene targets" and three control genes.

The 28 genes were selected from public databases of primary melanomas. The company performed analyses to identify genes that might detect aggressive disease (high likelihood of metastasis), and found that some of these genes are upregulated and some are downregulated in aggressive melanoma.

GEP was used to classify patients with stage I and II melanoma as being at low-risk (class 1) or high risk (class 2).

The 28-gene signature was established in developmental and training cohorts, and was validated in an independent cohort of 104 patients, 35 of whom developed metastatic disease. Follow-up time for patients who did not develop metastatic disease was 7.3 years.

In the validation cohort, 61 patients were classified as having a low-risk GEP and 43 were classified as having a high-risk GEP.

Five-year disease-free survival was significantly better for low-risk patients in the validation cohort than for high-risk patients (97% vs 31%; P < .0001). This compared well with the 5-year rates seen in the developmental cohort (100% vs 38%).

When the developmental and validation cohorts were combined (n = 220), GEP accurately identified 90% of the patients who did not show any documented evidence of metastasis and correctly identified 80% of the patients with documented metastasis.

The 28-gene signature has been independently validated in a cohort of 217 patients who had a negative sentinel lymph node biopsy, Derek Maetzold, president and CEO of Castle Biosciences, told Medscape Medical News. The study involving this validation cohort is currently in press, he reported.

Changes Clinical Management of Patients

The 28-gene test has shown a significant ability to accurately predict and identify patients at high risk for metastasis and, therefore, mortality, Dr. Gerami told Medscape Medical News.

There are limitations to AJCC staging. Two-thirds of patients who die are initially diagnosed with stage I or II disease and are not classed as high risk on the basis of sentinel lymph node biopsy, he explained.

The 28-gene signature is expected to change the clinical management of patients with early-stage melanoma. Stage I and II patients are typically followed in dermatology or primary care clinics, where they undergo skin examinations but not the rigorous imaging protocol that stage III patients do, he explained. However, now that there are new treatment options for melanoma, identifying patients at high risk for metastases is important, Dr. Gerami said.

Clinical guidelines recommend that patients with stage III melanoma be monitored more aggressively and entered into clinical trials. It makes sense that patients with stage I or II disease who are at high risk for metastasis be offered management and monitoring plans similar to those offered to patients with stage III disease, Maetzold told Medscape Medical News.

"When patients show a high-risk signature, they are monitored more frequently and go through additional testing, such as CT and PET/CT, to identify metastasis early," said Alexander Miller, MD, a surgical oncologist in breast cancer and melanoma from South Texas Oncology and Hematology in San Antonio, who uses the test.

Although no imaging is indicated for patients with stage IIa melanoma in the National Comprehensive Cancer Network (NCCN) guidelines, Dr. Sussman said that for stage IIa patients with a high-risk signature, he orders additional imaging tests and treats them much as he would his patients with stage III melanoma.

However, treating oncologists likely will not provide adjuvant therapy until the test is better validated and/or therapies are less toxic, Dr. Sussman told Medscape Medical News.

Commercialization of the 28-Gene Test

Castle Biosciences started taking orders for this test in 2013, after the completion of the first two clinical validation studies, Maetzold told Medscape Medical News.

"The test is primarily ordered by surgeons and dermatologists to improve management decisions in patients diagnosed with early-stage disease," he explained. Surgical oncologists primarily use the test to help appropriately monitor patients with a negative sentinel lymph node biopsy, he added.

Maetzold noted that clinical study data demonstrate that this test is not indicated for patients with node-positive or stage III disease.

He said he is hopeful that the first two validation studies will provide sufficient evidence for the NCCN committee to consider including the 28-gene test in future clinical practice guidelines for the monitoring of patients who are undermanaged on the basis of traditional staging criteria.

Medscape Medical News reached out to two NCCN committee members for comment on the test, but received no response.

Maetzold said he thinks this test should be used in all patients with stage I or II melanoma.

Dr. Miller, however, said he does not order the test for all patients, but only for patients with stage II node-negative disease with a melanoma thickness of 2.5 to 4.0 mm and for those with stage II disease in whom he is concerned about recurrence.

He reported that has used the test in about 40 patients with the high-risk signature. To date, he has not observed any events in these patients, but he noted that it is too early to expect any events to have occurred.

In the future, Dr. Miller said, he might consider using the test in all stage I and II patients. However, he noted, there have been previous tests that were supposed to predict risk for recurrent disease that did not pan out.

Dr. Sussman said he often discusses the test with his patients who might benefit from it. Because patients are not liable for any payment, most agree to be tested.

To date, no patient has been denied the test because of insurance or financial hardship, Maetzold told Medscape Medical News.

Several study authors report being consultants, being employees, and/or have ownership interests in Castle Biosciences. Dr. Miller and Dr. Sussman have disclosed no relevant financial relationships.

Clin Cancer Res. 2015;21(1):175-183. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: