Longitudinal evaluation of plasma miR371 to detect minimal residual disease and early relapse of germ cell tumors.

407 Background: Active surveillance is routinely recommended to manage patients (pts) with clinical stage I (CSI) germ cell testicular tumors (GCT), the most common presentation of newly diagnosed GCT. Circulating plasma miR371a-3p (miR371) has shown high sensitivity and specificity in pts with metastatic non teratoma GCT or in pts with clinically detectable testicular GCT prior to orchiectomy. However, limited data are available about this biomarker accuracy to detect minimal residual disease post-orchiectomy in pts on active surveillance for early stage disease. Methods: CSI GCT pts with available plasma samples after radical orchiectomy enrolled in the British Columbia provincial biobank research program were selected for this study. RT-PCR was used for qualitative miR371 analysis. Sensitivity, specificity, negative and positive predictive values (NPV, PPV) and AUC in predicting tumor recurrence were evaluated for miR371 and compared to the same operating characteristics of current gold standard diagnostic tests. Relapse free survival (RFS) was correlated to post-orchiectomy miR371 (positive or negative) status. Fisher’s exact test was used to evaluate the sensitivity and specificity, unpaired t-test for comparison of miR371 expression. RFS was calculated using the Kaplan-Meier method, and differences between groups were estimated using the log rank test, 2-sided and with 5% significance threshold. Results: With a median follow-up of 41 months, 101 pts with CSI GTCwere included, of whom 35 (34.6%) experienced a disease relapse during the follow-up. miR371 was positive in 22/35 (62.8%) of the relapsed pts. miR371 positivity preceded clinical evident disease by a median of 3 months (range: 1-12 months).The specificity and PPV were 100% (95% CI: 94.5 - 100 for both), sensitivity 62.8% (95% CI: 44.9 - 78.5), NPV 83.5% (95% CI: 76.7 - 88.6) and AUC 0.81 (95% CI: 0.71 - 0.91). No false positive results were observed. The RFS of the pts with positive post-orchiectomy miR371 was significantly shorter (median: 3.5 months vs. not reached; p<0.0001) compared to the pts with a negative post-orchiectomy miR371 (HR: 16.9; 95% CI: 2.1 - 135.7; p<0.0001). miR371 sensitivity correlated with tumor burden, time between tumor relapse and miRNA testing and histology (nonseminoma > seminoma). Conclusions: miR371 has high specificity and PPV in detecting GCT at an early stage and could be used to guide treatment selection after orchiectomy. Further studies, including the SWOG S1823 clinical trial, are ongoing or have been planned in this setting for validation of clinical utility.