Novel NEN Diagnostic

About

Neuroendocrine neoplasms (NENs) are clinically diverse tumors that likely arise from neuroendocrine cells scattered throughout the body. The incidence of NEN has risen significantly globally over the last three decades making NEN a growing healthcare concern. Initial diagnosis of NEN is challenging due to heterogeneous clinical presentation with often vague abdominal symptoms. Further complicating this is the high tumor diversity and subtle histologic differences between pathological types. The factors can lead to a delay in diagnosis and studies have found a median delay of almost 54 months from onset of first symptoms. This delay results in later stage diagnosis with significantly poorer outcomes of some NENs compared with non-NENs at the same anatomical site. Current diagnostic testing for NEN detection is a combination of imaging to identify the tumor and pathologic evaluation of biopsy or resection material. The latter technique combines immunohistochemistry and mitotic count figures to predict primary tumor site and grade but has accuracy limitations. With the above limitations, and the high expense of existing diagnostics, there is an urgent need to identify better biomarkers enable faster, cheaper, and more accurate diagnosis of NENs. MicroRNAs (miRNAs) are small (19‑24 nucleotide) regulatory molecules that can be used as biomarkers to classify cancer. miRNAs are highly informative tissue markers due to their abundance and stability, as well as their cell‑type and disease‑stage specificity. miRNAs are also known to have predictable interactions with messenger RNAs, meaning that these molecules can also provide mechanistic information regarding cellular processes. Using barcoded small RNA sequencing and data mining, Queen’s researchers have generated comprehensive microRNA expression profiles for 15 different NEN pathology types. These data have allowed the researchers to create a miRNA‑based hierarchical classifier which is able to first identify whether a tumor is a NEN and subsequently sub‑classify the actual pathological type. Queen’s researchers have identified miR‑375 as a potential universal marker of neuroendocrine cell differentiation. The current leading NEN biomarkers, chromogranin A (CgA) and synaptophysin, have demonstrated inconsistent results depending on the particular NEN subtype. Chromogranin A has shown specificity ranging from 68‑100% and sensitivity ranging from 42‑93% based on tumor primary site, grade, or disease burden1. Similarly, synaptophysin, despite being considered to be less specific than CgA, has been found to be expressed in as low as 41‑75% of certain NEN tumor types. As such, there is a clear need for a universal biomarker that can detect NEN tumors with a high degree of sensitivity and specificity. Their feature selection algorithm also identified 17 miRNAs to discriminate 15 NEN pathological types. Queen’s researchers subsequently constructed a multilayer classifier for discriminating NEN pathological types and successfully identified 217 of 221 samples (98%). This miRNA expression profiling has resulted in the identification of universal and classificatory markers to both identify NEN tumors and determine the NEN pathology type. These improvements in NEN classification can help enable more effective and timely clinical treatment while also allowing for more targeted NEN classification in the context of future clinical trials.