GEP tests show poor prognostic ability in recurrent stage I melanoma

Two gene expression profile tests for cutaneous melanoma often failed to accurately identify recurrent stage I disease, according to a study.

“The performance of prognostic gene expression profile (GEP) tests for cutaneous melanoma is poorly characterized,” Michael A. Marchetti, MD, of the dermatology service in the department of medicine at Memorial Sloan Kettering Cancer Center and the department of dermatology at Weill Medical College of Cornell University, and colleagues wrote.

In the systematic review and meta-analysis of PubMed/MEDLINE, Embase and Web of Science databases, the researchers aimed to evaluate the performance of GEP tests in patients with American Joint Committee on Cancer (AJCC) stage I or stage II localized melanoma.

Eligible studies included those that were in humans and in the English language and did not have date restrictions. The final analysis included 1,450 participants from seven studies. Five of the studies assessed DecisionDx-Melanoma (Castle Biosciences) and two assessed MelaGenix (NeraCare).

The ratio of patients with or without melanoma recurrence who were correctly classified by the GEP test as being high risk or low risk served as the primary endpoint.

Results showed that the performance of both DecisionDx-Melanoma and MelaGenix varied based on AJCC stage.

The researchers reported findings for 623 patients with stage I disease and 212 patients with stage II disease tested with DecisionDx-Melanoma. Of the stage I patients, six had true positive results, while there were 15 false negatives, 61 false positives and 541 true negatives. For the stage II group, there were 59 true positives, 13 false negatives, 78 false positives and 62 true negatives.

DecisionDx-Melanoma correctly classified recurrence in 29% of stage I patients and 82% of stage II patients.

DecisionDx-Melanoma correctly classified non-recurrent disease in 90% of patients with stage I disease and 44% of those with stage II disease.

For participants tested with the MelaGenix test, 88 had stage I melanoma and 245 had stage II. In the stage I group, MelaGenix yielded seven true positives, 15 false negatives, 15 false positives and 51 true negative results. For stage II, this test yielded 59 true positive results, 19 false negatives, 95 false positives and 72 true negatives.

MelaGenix correctly classified recurrence in 32% of patients with stage I melanoma and 76% with stage II disease. For non-recurrence, MelaGenix was correct in 77% of patients with stage I disease and 43% of those with stage II group.

“The prognostic ability of GEP tests among patients with localized melanoma varied by AJCC stage and appeared to be poor at correctly identifying recurrence in patients with stage I disease, suggesting limited potential for clinical utility in these patients,” the researchers wrote.