Endocrine Abstracts

Introduction: The variable tumor behavior in patients with gastro-entero-pancreatic neuroendocrine tumors (GEPNETs) is challenging. Current general biomarkers are insufficient to predict the disease course. An emerging biomarker is the NETest, a blood-based gene signature that can predict disease status based on the expression of genes involved in tumor biology. While promising, the accuracy and reproducibility of results in daily practice during years of follow up has never been assessed. Evaluation of serial NETest measurements in an individual is needed to determine its place in the clinical armamentarium.

Aims: To evaluate if serial NETest measurements can predict treatment response and reflect disease evolution during years of follow up.

Methods: Serial NETest scores were compared with RECIST1.1 defined disease status in 132 GEP-NET patients over 46 (6-71) months of follow-up. A median of 4 samples was collected in patients on a watch-and-wait strategy or undergoing systemic treatment. Pre- and post-treatment scores (<6 months) were compared with progression-free survival (PFS).

Results: Fluctuating scores [0-100%] were seen in patients with no evidence of disease (NED) and stable disease (SD). None of the 30 patients with NED and 1 of the 28 (4%) patients with SD had all outcomes within the low range. In patients with progressive disease (PD) and not receiving any treatment (n=16), ongoing tumor progression was confirmed in consecutive samples in 82%. Patients responding to treatment (PFS > 12 months) had higher pre-treatment NETest scores (76.5; n=22) compared to non-responders (33; n=12;P=0.001). Patients with low pre-treatment scores had a 21 months shorter PFS after treatment (10 vs 31 months;P=0.008). The accuracy for treatment response prediction was 0.73 (P=0.009). Post-treatment scores had no discriminative value.

Conclusion: Low NETest scores are associated with an indolent tumor behavior in the follow up of individuals, but scores fluctuate over time in patients with NED and SD. Elevated NETest scores in patients on watch-and-wait strategy had limited predictive value while elevated scores – measured before treatment initiation – predicted treatment response and might therefore be used for individualizing decisions on starting systemic therapy.