Aims
Aggressive and metastatic pituitary neuroendocrine tumours (PitNETs) are associated with significant morbidity and mortality and progress despite treatment with standard therapies. This study aimed to investigate the spatially defined gene expression of aggressive and metastatic PitNETs to identify prognostic and therapeutic biomarkers.
Methods
Aggressive and non-aggressive Pit-1, SF-1, T-Pit, and co-expressing Pit-1 and SF-1 pituitary neuroendocrine tumours were analysed using Nanostring GeoMx Digital Spatial Profiling. Tissue sections were stained with Pan-cytokeratin (PanCK), CD45, CD68 and the nuclear stain SYTO13. Regions of interest (ROIs) were selected to represent tumour areas with and without immune cell infiltrate. Gene expression was assessed using the GeoMx Human Whole Transcriptome Atlas. Samples were sequenced on a NextSeq 2000 and data processed with the GeoMx DSP software. Two normal pituitary samples were included and 33 pituitary tumours samples including aggressive Pit-1 (n=6), non-aggressive Pit-1 (n=6), aggressive SF1 (n=1), non-aggressive SF1 (n=5), non-aggressive SF1/Pit1 (n=6), aggressive T-Pit (n=3) and non-aggressive T-Pit (n=6).
Results
Aggressive PitNETs had reduced expression of tumours suppressor genes (DIRAS, DAPK1, KNDC1, COX7A1, NEBL and KANK1), and increased expression of oncogenes and immune-related genes (SLC38A2, ANXA1, TGFBR2, HLADQA1 and STAT6) compared to non-aggressive PitNETs. The differential gene expression of aggressive versus non-aggressive PitNETs varied within tumours in ROIs with increased compared to reduced immune cell infiltrate. Whilst a subset of genes were consistently upregulated (GCH1, KPNA2 and RNF157I) and downregulated (SEZ6L2, DZIP3, PODXL2, CTNND1) in both aggressive lactotroph Pit1 and T-Pit PitNETs, the majority of differentially expressed genes were unique to each PitNET type. This suggests that although some molecular mechanisms are shared, the biology of aggressive behaviour differs between PitNET types.
Conclusions
PitNETs exhibit molecular intratumoural heterogeneity and distinct molecular profiles between tumour types. Differential gene expression between aggressive and non-aggressive PitNETs may present novel prognostic biomarkers and therapeutic targets.