Predictors of in vitro response to somatostatin and dopamine receptor‐targeted treatments are lacking in clinically non‐functioning pituitary tumors

Abstract Clinically non‐functioning pituitary neuroendocrine tumors (NF‐PitNETs) lack effective medical therapies, and the role of somatostatin receptor ligands (SRLs) and dopamine agonists (DAs) in this setting is controversial. We evaluated the in vitro response of NF‐PitNETs to multiple SRL and/or DA treatments, focusing on the role of receptor expression and tumor characteristics in driving tumor response. Forty‐four NF‐PitNET primary cultures were treated (72 h, 10 nM) with octreotide (OCT), pasireotide (PAS), OCT + PAS, BIM‐53097 (D2R agonist), BIM‐53097 + PAS, and BIM‐23B065 (SSTR2/SSTR5/D2R preferential ligand) to evaluate cell proliferation inhibition. Tumors were considered responders to a treatment when a ≥20% reduction in cell proliferation was observed (vs. control); responder tumors to at least one tested condition were classified into the “responder group” for further analysis. Somatostatin (SSTRs) and dopamine type 2 receptor (D2R) expression were evaluated through immunohistochemistry. Data on radiological invasiveness, proliferation indices (Ki67%, mitoses), and p53 immunostaining were collected. Tumor grade was determined using Trouillas and PANOMEN‐3 classifications. Median patient age at surgery was 60.3 years (IQR 49.7–70.7), 29 were males (71%). Overall, no significant in vitro inhibition of cell proliferation was observed [ranging from −4.2% (OCT and PAS) to −7.4% (BIM‐53097 + PAS)]. However, 13 cultures (30%) were included in the “responder group” [mean inhibition ranging from −18.6% (OCT) to −27.7% (BIM‐53097 + PAS)]. D2R was the most expressed receptor (median IRS 8, IQR 6–8), followed by SSTR1 and SSTR2 (median IRS 4, IQR 3–6); SSTR3 and SSTR5 expression was low. In the whole cohort, OCT and PAS efficacy directly correlated with SSTR2 ( p = .013 and p = .027, respectively). D2R and SSTR2 expression was higher in responders vs. non‐responders to BIM‐23B065 ( p = .017 and p = .042, respectively). ROC curve analysis discriminates BIM‐23B065 responders based on D2R and SSTR2 expression with acceptable ability (AUC = 0.784 and AUC = 0.743, respectively). However, the “responder group” did not differ significantly from the “non‐responder group” with respect to patient demographics, tumor clinical–pathological characteristics, receptor expression, and grading. In conclusion, we confirm the limited efficacy of SRL and DA treatment in NF‐PitNETs. D2R and SSTR2 can discriminate tumors responsive to BIM‐23B065. As concerns the other tested compounds, none of the tumor parameters evaluated demonstrated robust predictive value for NF‐PitNET in vitro response.