Context: Early and precise delineation of therapeutic responses are key issues in neuroendocrine neoplasm/tumor management. Imaging is currently used but exhibits limitations in sensitivity and specificity. The utility of biomarkers is unclear. objective, setting, and design: This prospective cohort study (11 mo) sought to determine whether measurements of circulating neuroendocrine tumor transcripts (NETest) predict responses to somatostatin analogs (SSAs).
Patients: The test set consisted of 35 SSA-treated gastroenteropancreatic-NETs (RECISTevaluated). The prospective set consisted of 28 SSA-treated Grade 1-Grade 2 GEP-NETs.
Intervention(s): Whole blood for transcript analysis (NETest) and plasma for Chromogranin A (CgA) (baseline), were collected every 4 weeks (prior to SSA injection). Morphologic (multidetector computed tomography/MRI) and functional imaging ((99m)Tc-[HYNIC, Tyr(3)]-Octreotide) was undertaken at entry and 6-month intervals until progression (RECIST 1.0).
Main outcome measure(s): Treatment response.
Results: Test set: NETest (≥80%; scale, 0-100%) differentiated stable (SD) and progressive (PD) disease (P < .0001). Prospective set: 28 patients (26/28 SD) undergoing standard SSA. Grading: 12 G1, 16 G2. SSA Response: progression-free survival: 315 days: 14 (50%) SD, 14 (50%) PD. NETest: Twenty had elevated (≥80%) values; 14 developed PD; six, SD. CgA: Twelve of 28 exhibited elevated baseline values and/or subsequent >25% increase; eight developed PD; four, SD. NETest (P = .002) and grade (P = .054) were the only factors associated with treatment response. Multiple regression analysis established that the NETest could predict disease progression (P = .0002). NETest changes occurred significantly earlier (146 d prior to progression vs 56 d CgA; P < .0001; χ(2) = 19) and in more patients (100 vs 57%; P < .02).
Conclusions: NETest values (80-100%) were more accurate and occurred at a significantly earlier time point than CgA and predicted SSA treatment response.