Tamoxifen in residual non-functioning pituitary Neuroendocrine Tumours: an integrated clinical, molecular and computational study

Abstract

Post-operative residual non-functioning pituitary neuroendocrine tumours (NF- PitNETs) have limited medical treatment options to prevent tumour regrowth. This translational proof-of-concept study evaluated the therapeutic potential of tamoxifen in patients with residual NF-PitNETs using an integrated clinical, molecular, and computational framework. Immunohistochemical analysis quantified oestrogen receptor alpha (ER), oestrogen receptor beta (ERB), and proliferative markers in fifty archival tumour tissues. A 12-month randomised pilot trial enrolled adults with magnetic resonance imaging (MRI)-confirmed residual disease and allocated them in a 1:1 ratio to tamoxifen at 20-40 mg daily or active surveillance. Tumour volumes were assessed at baseline, six months, and twelve months using quantitative three- dimensional MRI segmentation. A panel of ten serum biomarkers spanning growth, angiogenic, inflammatory, and Wingless and Integrated-modulating pathways was measured longitudinally. In parallel, a network pharmacology workflow, pathway enrichment analysis, protein-protein interaction mapping, and hub gene ranking was applied to characterise tamoxifen-associated molecular targets and pathways. Ninety- six patients were screened, with twenty completing follow up; each group had comparable baseline clinical, radiological, histological, and biomarker profiles. ERa expression was detected in 18% of tumours and ERß in 68%. Although longitudinal analyses did not show significant overall time or time-by-treatment effects, paired analysis within the tamoxifen group demonstrated a significant reduction in tumour volume from baseline to twelve months (p=0.045). Tumour shrinkage occurred in 40% of tamoxifen-treated patients, with no cases of regrowth observed, whereas 20% of patients under active surveillance demonstrated tumour progression (p = 0.014). In multiple linear regression analysis, post-treatment Dickkopf-related protein 1 independently and inversely predicted final tumour volume (B≈ 0.00, ẞ = −0.78, t = −3.53, p = 0.010). Multivariate modelling identified a dominant growth and angiogenic axis explaining the largest proportion of variance in volumetric response. Computational network analyses implicated key signalling networks, particularly the phosphatidylinositol 3-kinase/protein kinase B and mitogen-activated protein kinase pathways, supporting tamoxifen-associated systems-level modulation. In conclusion, tamoxifen was associated with favourable volumetric trends supported by radiological, biochemical, and computational evidence. While exploratory, these findings provide biologically coherent, hypothesis-generating evidence that tamoxifen may exert disease-modifying effects in selected patients with residual NF-PitNETs and justify larger controlled trials.