LCM-RNAseq Highlights Intratumor Heterogeneity and a lncRNA Signature from Archival Tissues of GH-Secreting PitNETs

Abstract

Background: This study explores the potential for hidden variations within seemingly uniform regions of growth hormone-secreting pituitary neuroendocrine tumors (GH-PitNETs). We employed archived tissue samples using Laser Capture Microdissection Sequencing (LCM-RNAseq) to probe the molecular landscape of these tumors at a deeper level.

Methods: A customized protocol was developed to extract, process, and sequence small amounts of RNA from formalin-fixed, paraffin-embedded (FFPE) tissues derived from five patients with GH-secreting PitNETs and long-term follow-up (≥10 years). This approach ensured precise isolation of starting material of enough quality for subsequent sequencing.

Results: The LCM-RNAseq analysis revealed a surprising level of diversity within seemingly homogeneous tumor regions. Interestingly, the 30 most highly expressed genes included the well-known long noncoding RNA (lncRNA) MALAT1. We further validated the levels of MALAT1 and of other tumor-associated lncRNAs using digital droplet PCR.

Conclusions: This study demonstrates the potential of LCM-RNAseq to unlock hidden molecular diversity within archived pituitary tumor samples. By focusing on specific cell populations, we identified lncRNAs expressed at different levels within the tumors, potentially offering new insights into the complex biology of GH-secreting PitNETs. This evidence prompts further research into the role of lncRNAs in pituitary neuroendocrine tumor aggressiveness and personalized treatment strategies.

Keywords: GH-secreting pituitary neuroendocrine tumor; RNA sequencing; biomarkers; laser capture microdissection; precision medicine.

Figure 1

Transcriptome analysis by Laser-Capture Microdissection (LCM)-RNAseq of Growth Hormone (GH)-secreting PitNET. (A,B) Venn diagram of the genes expressed in common by the Recurrent (left) and Non-Recurrent (right) GH-secreting PitNET, respectively. (C) Heatmap of the genes with higher (red) or lower (blue) expression. Each sample is identified by its name and grouped according to the outcome and the patient from which it was derived. (D) Heatmap of the sample-to-sample distance: compares sample-sample pairs, displaying the greater or lesser difference between them on a blue scale, from lightest to darkest. (E) PCA of the samples, according to the groups of interest, “Outcome” (Recurrent/Non-Recurrent) and “Replicates” (A = Pit4 and Pit3, B = Pit11, C = Pit34, Pit35 and Pit36, D = Pit1, E = Pit18 and Pit29) on the normalized data of the whole dds (DESeqDataSet).