DecisionDx^®-Melanoma Prospective Studies

Consistent Results from Prospective Risk of Recurrence Studies Support Routine Use of DecisionDx-Melanoma

Consistent results observed across six prospective clinical studies totaling 1,621 patients, showing high sensitivity and NPV, support routine use of DecisionDx-Melanoma to identify individual risk and inform clinical decisions.

Combined 1,621 patients across prospective studies

Hsueh et al. J Hematol Oncol 2017: First peer-reviewed publication showing performance in a prospective population. Updated with 3-year median follow-up presented in 2020 showing consistent results.

Analysis of 322 patients from 11 dermatologic and surgical centers reported 1.5- and 3-year RFS rates of 97% (at both time points) for Class 1 patients, and 77% and 65%, respectively, for Class 2 patients (p<0.0001).
While the majority (20 of 25) of melanoma recurrences were observed in the Class 2 group, a negative predictive value of 98% for recurrence demonstrated that a Class 1 outcome was associated with a low risk of metastasis.
Cox regression analysis showed that DecisionDx-Melanoma was an independent and the most significant factor predicting risk of recurrence.

Thorpe et al. ASDS 2019: Independent, prospective, multicenter study showing consistent performance in dermatology practice population.

This prospective cohort of 1,124 melanoma patients treated with Mohs surgery at 9 clinical sites demonstrated that Class 2 patients were 16 times more likely to develop metastatic disease than those with Class 1 result.
The NPV of 98% provides confidence in low risk of metastasis and mortality for Class 1 patients and confirms the prognostic value of the DecisionDx-Melanoma test.

Keller et al. Cancer Med 2019: Prospective study of gene expression profiling shows 79% of recurrences in Class 2 patients.

In an independent study performed at a surgical oncology practice, 159 consecutive melanoma patients undergoing sentinel lymph node (SLN) biopsy were tested with DecisionDx-Melanoma.
With a median follow-up time of 44.9 months a total of 29 recurrences were reported, of which 23 (79%) were Class 2 and 6 (21%) were Class 1. In comparison, 10 recurrences were observed in SLN-positive patients and 19 in SLN-negative patients.
Similar to other studies, DecisionDx-Melanoma demonstrated a strong sensitivity (79%) and negative predictive value (98%), and was an independent and significant prognostic factor by multivariate Cox regression analysis (HR=15.0, p<0.0001).

Podlipnik et al. JEAVD 2019: Consistent performance in multicenter study conducted in Spain.

This prospective study of 86 patients at 5 dermatology practices demonstrated DecisionDx-Melanoma accurately predicts risk of recurrence both independently and combined with AJCC staging.
DecisionDx-Melanoma was the only significant predictor of recurrence compared to AJCC staging and age, with a hazard ratio of 18.8%. All recurrences occurred among Class 2 patients.
When using the test with AJCC staging, the study showed that it can identify high-risk patients in a low-risk AJCC group and low-risk patients in a high-risk AJCC group.

Other Prospective Studies

Greenhaw et al. Dermatol Surg 2018: Negative predictive value of 99% for recurrence shown in early-stage melanoma patients.

In a prospective cohort of 256 consecutive Stage I and II patients (mean follow up = 23 months), the test accurately identified 10 of 13 metastatic patients as Class 2 (sensitivity=77%), and 211 of 214 patients without a recurrence of melanoma as Class 1 (NPV=99%).
Overall, Class 2 patients were 22 times more likely to develop metastatic disease, and Kaplan-Meier analysis showed a 5-year metastasis-free survival rate of 93% for Class 1 patients and 69% for Class 2 patients (p < 0.00001).

Renzetti et al. SSO 2017: Combined experience of two tertiary referral centers shows high sensitivity of the test to identify both recurrence and distant metastasis consistent with validation studies.

In this independent study, 95 patients with Stage I-III disease at two tertiary melanoma centers were tested with DecisionDx-Melanoma (median follow up = 11 months).
Only 2 (2.4%) Class 1 patients recurred (only one with distant metastasis). Conversely, 13 (24.1%) Class 2 patients recurred, with 10 of these (18.5%) having distant metastasis.
Class 2 patients had significantly higher rates of recurrence and distant metastasis than Class 1 patients (p=0.02 for both endpoints).
In this cohort, the sensitivity of DecisionDx-Melanoma to identify those with recurrences as Class 2 was 87% and 91% for distant metastasis.