Inefficacy of chromogranin a assays as neuroendocrine tumor diagnostic tools compared to the NETest.

Abstract

4612 Background: Chromogranin A (CgA) remains a commonly used diagnostic and monitoring tool for neuroendocrine tumor disease despite NCCN guidelines identifying it as a category 3 (major concerns about utility) biomarker. Several commercial assays have been developed to measure this protein (or its fragments) and are available both at CLIA-certified laboratories (USA) as well as in NET Centres of Excellence (CoEs - Europe). CgA is typically reimbursed by insurance companies and appears in several guidelines (e.g., ENETS). We sought to directly evaluate the accuracy of detecting NET disease using two different CgA assays, one in the USA (NEOLISA, EuroDiagnostica, IBL-America, CLIA-certified laboratory) and one in an ENETS CoE (CgA ELISA, Demeditec Diagnostics, Germany). We compared the results to the NETest, a circulating mRNA assay, recently validated as an IVD for NETs. Methods: Patients: NETs ( n=258) including lung: n=43; duodenum n=9; gastric: n=44; pancreas: n=67; small bowel: n=40; appendix: n=10; rectum: n=45. No image-evidence of disease ( n=122) (IND) and image-positive disease (IPD) ( n=136). CgA assays (plasma): NEOLISA, ULN >108ng/ml, DD: ULN>99ng/ml. Data mean±SEM. NETest (whole blood): qRT-PCR - multianalyte algorithmic analyses, CLIA-laboratory. All samples de-identified and assessed blinded. Statistics: Mann-Whitney U-test, Pearson correlation & McNemar-test. Results: In the entire group ( n=258), NEOLISA assay CgA levels were significantly ( p<0.0001) higher (216±91ng/ml) vs. the DD-assay (76±8ng/ml). The assays exhibited a high concordance in output (Pearson r=0.81, p<0.0001), but there were 10.9% ( n=31) discordant results. This reflected the NEOLISA assay detecting more CgA-positive samples. IPD group: CgA-positives were detected in 48/136 (35%, NEOLISA) vs. 28 (21%, DD-assay). McNemar’s Chi2=15.04, p<0.001 OR: 11.0, indicating the NEOLISA was significantly better than the DD-assay. The NETest, in contrast, was positive in 135/136 (99%; OR: 87-106, p<0.0001). IND group: CgA-positives were detected in 12/122 (10%, NEOLISA) vs. 9 (7%, DD-assay; p=NS). The majority (75%) of positives were associated with gastric NETs. The NETest was positive in 7 (6%); 4 were gastric NETs and 3 exhibited elevated CgA. Conclusions: Two standard CgA assays used for NET management (one accepted by Medicare in the USA, the second used at a CoE in Europe) only detect NET disease in 21-35% of cases. In contrast, a circulating mRNA fingerprint, the NETest, is ~99% accurate for detecting NET disease.