Better observed vs. predicted survival rate found for class 2 uveal melanoma tumors
Click Here to Manage Email Alerts
In patients with class 2 uveal melanoma tumors, predicted metastasis-free survival by a gene expression profile test was shown to be worse than that observed in a cohort study and worse than that reported in the literature.
The DecisionDx-UM (Castle Biosciences) is a commercially available prognostication test offered by several institutions to patients with uveal melanoma as part of their management plan. Based on this test, tumors are classified as low risk (class 1A), intermediate risk (class 1B) or high risk (class 2). Predicted survival reported as percent metastasis free for the three classes at 3 years and 5 years was 98% and 98%, 93% and 79%, and 50% and 28%, respectively, according to the study.
The study, conducted by the University of Iowa and the Cleveland Clinic, compared the predicted metastasis-free survival (MFS) obtained by the gene expression profile (GEP) test with the observed MFS in 347 patients with uveal melanoma. The results were also compared with the MFS data obtained through a meta-analysis of eligible published papers.
At baseline, tumors were classified according to the Collaborative Ocular Melanoma Study as small, medium or large and according to the four stages of the American Joint Committee on Cancer staging system. Over a median follow-up of 38 months, 48 patients (five with class 1A tumors, three with class 1B tumors and 40 with class 2 tumors) developed metastasis. The observed 3-year MFS was 93% for patients with class 1 tumors and 67% for patients with class 2 tumors, while the 5-year MFS was 87% for patients with class 1 tumors and 47% for patients with class 2 tumors. Larger tumor size was significantly associated with increased risk for developing metastasis.
The observed MFS data at 3 years and 5 years for class 2 tumors were significantly better as compared with the predicted GEP outcomes (50% and 28%) and were comparable to the data published in the literature (62% and 40%).
“The predicted MFS for metastatic tumors (class 2) may be too pessimistic than that observed by us and reported by others,” the authors wrote. “Incorporation of tumor size in the prediction model may enhance its accuracy.”
Perspective
Back to Top
In this analysis of 347 patients with uveal melanoma who were evaluated over an 8-year period from two ocular oncology centers, the authors studied the accuracy of genetic testing in predicting metastasis-free survival (MFS). They used gene expression profiling and compared their results (observed MFS) to the predicted MFS from the commercially available gene expression profiling kit.
They found that the predicted MFS for high-risk melanoma (class 2) was worse than their observed MFS. Specifically, they found the predicted MFS at 5 years was 28% vs. their observed MFS at 5 years of 47%. This suggests that other factors such as tumor size could be important to improve accuracy of the predicted MFS.
This is vital information as patients with high-risk uveal melanoma (class 2) are often entered into therapeutic trials to reduce their risk for metastasis. The commercial test predictive risk for metastasis should be as accurate as possible and possibly should include other variables such as tumor size, as found in this report. Further studies with larger cohorts should be undertaken to confirm these findings.
Collapse
Read more about
Click Here to Manage Email Alerts