Dopamine agonists (DAs) are chemical compounds that, by directly binding to the dopamine receptors (DRs), induce biological actions resembling that of endogenous dopamine (1). In the context of pituitary neuroendocrine tumors (PitNET), DAs lead to the inhibition of hormonal secretion and tumor shrinkage in different pituitary tumor histotypes by mainly binding to the dopamine receptor type 2 (D2DR), largely expressed in lactotroph pituitary cells but also localized in other pituitary cell sub-populations including somatotroph and corticotroph cells (2).

DAs can be classified in ergot derivatives, including bromocriptine (BRC), cabergoline (CAB), pergolide and lisuride, and in non-ergot derivatives such as quinagolide. The most commonly used DAs as a medical treatment for PitNET are currently BRC and CAB (3).

The main localization of D2DR is on normal and tumoral animal and human lactotroph pituitary cells (2, 4, 5), therefore DAs represent the treatment of choice for PRL-secreting PitNET, for whose management, unlike other PitNET, the consensus guidelines have recommended the medical therapy as first-line treatment of choice rather than surgery (6). In prolactinomas, medical therapy relies on the preferential use of CAB compared to BRC because of the higher affinity of CAB to D2DR than BRC, its better tolerability profile, and its higher and long-lasting efficacy in normalizing prolactin (PRL) levels and in inducing tumor shrinkage (6). Indeed, in prolactinomas receiving treatment with CAB biochemical control and relevant tumor shrinkage are reportedly recorded in the vast majority of patients up to 100% and 96% of cases, respectively (7).

The D2DR localization in normal and tumoral animal and human somatotroph (4, 8–10) and normal and tumoral human corticotroph pituitary cells (11, 12) has provided the basis for DAs application also in the therapeutic algorithm of acromegaly and, as off-label use, in Cushing’s Disease (CD). However, in patients harbouring GH-secreting PitNET and ACTH-secreting PitNET, the biochemical and tumoral efficacy of DAs is lower than in prolactinomas. Indeed, DAs, mainly CAB, are indicated as monotherapy in patients with GH-secreting PitNET with mild disease (13), defined as mild signs and symptoms of GH excess and modest elevation in serum IGF-I levels (lower than 2 times the upper limit of normal) (14). In such patients, CAB administration has resulted in the achievement of biochemical control in approximately one-third of patients (15). Alternatively, CAB can be used in combination with somatostatin receptor ligands (SRLs) in patients displaying partial responsiveness to SRLs monotherapy (14), defined as incomplete or inadequate biochemical and tumoral control (16), or in addition to the GH receptor antagonist pegvisomant (17). In such patients, CAB addition to SRLs has resulted in the achievement of biochemical control in 52% of patients (15). When combined with pegvisomant, CAB administration has been demonstrated to induce biochemical control in 68% of patients (17). In CD, DAs are listed among the pituitary-directed drugs for patients with ACTH-secreting PitNET, including those who have experienced a noncurative surgery or a postoperative recurrence and are not candidates for additional pituitary surgery (18). In such patients, the use of CAB has resulted in the achievement of biochemical control in up to 40% of cases (19). Added to other compounds, CAB has been shown to induce biochemical control in 56-79% when used in combination with steroidogenesis inhibitors (20–22) and up to 88% when administered in addition to pasireotide and ketoconazole (23).

Despite the proven biochemical and tumoral efficacy of DAs in prolactinomas, a minority of patients, accounting for 10% of patients with macroprolactinoma and less than 20% of those with macroprolactinoma (24, 25), fail to achieve the biochemical control of PRL excess and/or the reduction in tumor mass during treatment with DAs (6, 26).

In patients with persistent/recurrent clinically nonfunctioning PitNET, for which no medications have been approved, the efficacy of DAs, including CAB and BRC, in reducing the tumor size has been tested based on the D2DR expression established in most of the nonfunctioning PitNET tissues in several preclinical studies (27–29). DA therapy evaluated in some case reports and in small series provided variable results with a reduction in tumor size in 30% of cases and induction of stable disease in 58% of cases (30). However, due to insufficient and variable evidence, DA therapy is not routinely administrated in patients with nonfunctioning PitNET and DAs use is still a matter of debate (31).

The molecular mechanisms of DAs resistance in PitNET are not fully understood. This mini-review focuses on the molecular mechanisms underlying the resistance to DAs in PitNET.

DRs are five subtypes of G protein-coupled receptors (GPCRs) divided into two subgroups based on their structural, functional and pharmacological properties: D1-like family, including D1DR and D5DR, and D2-like family, comprising of D2DR, D3DR, and D4DR (2). The presence of introns within the coding regions of D2-like receptors give rise to the potential formation of isoforms due to the occurrence of alternative splicing in their mRNAs (2). Indeed, D2 receptor exists in two different isoforms, denominated D2 short (D2S) and D2 long (D2L) isoforms, differing for the presence or absence of 29 amino acids in the third intracellular loop, and displaying distinct physiological and pharmacological properties (2).

In lactotroph cells, the dopamine pathway, besides the well-known regulation of hormone secretion, controls the complex physiological mechanism of cell homeostasis. The lactotroph cell homeostasis represents a condition of biological and molecular steady-state to prevent the triggering of pathological conditions. Indeed, in lactotroph cells, a correct function of the dopamine signalling pathway is needful to limit PRL synthesis and secretion, and cell growth and proliferation, to minimize the development of lactotroph hyperplasia, or lactotroph differentiation and expansion, with a consequent formation of PRL-secreting tumors leading to a state of pathological hyperprolactinemia (32–37).

DRs are involved in human and animal hormone secretion inducing the cyclic AMP (cAMP) pathway. Precisely, D1-like receptors, being coupled to stimulatory G proteins (Gα_s), activate cAMP pathway, inducing adenylyl cyclase (AC) activity, cAMP accumulation, protein kinase A (PKA) activation and calcium (Ca²⁺) release from the intracellular compartment. Conversely, D2-like receptors, being coupled to inhibitory G proteins (Gα_i/o), inhibit cAMP pathway, suppressing AC activity, cAMP accumulation, PKA activation and Ca²⁺ release from the intracellular compartment (2). Interestingly, studies in rodents have demonstrated that D2S and D2L isoforms induce two cAMP-independent pathways in the regulation of lactotroph cell homeostasis: the mitogen-activated protein kinase (MAPK) and the phosphatidylinositol 3-kinase (PI3K)/protein-kinase B (PKB) pathways.

Specifically, studies on mouse lactotroph cell models demonstrate that D2S induces ERK1/2 of the MAPK pathway and PKB of PI3K pathway, whereas D2L inhibits ERK1/2 and PKB. In physiological conditions, the expression of both isoforms is sufficient to maintain PRL synthesis and secretion and cell proliferation, while the simultaneous knockdown of both D2DR isoforms induces the inhibition of ERK1/2 and the induction of PKB activity resulting in uncontrolled cell proliferation and consequent pituitary hyperplasia and hyperprolactinemia (35). Conversely, in pathological states, the expression of only one D2DR isoform is no longer sufficient to control lactotroph homeostasis (35). These data suggest that the control of D2L/D2S ratio and therefore the balance between MAPK and PI3K pathway are a prerequisite to maintaining lactotroph homeostasis and preventing pathological lactotroph development.

More recently, a new mechanism of DRs modulation has been demonstrated to be involved in the regulation of pituitary tumor volume shrinkage. Indeed, in in vitro rodent prolactinoma and somatotropinoma cell lines (MMQ and GH3) and in vivo rodent pituitary tumors it has been demonstrated that the activation of D5DR mediates cell growth suppression by autophagic pathway stimulation through PKB/mTOR signalling (38–40). Deeply, D5DR activation inhibits mTOR signalling, decreasing p70S6K and 4eBP1 phosphorylations. Consequently, D5DR activation induces autophagic cell death, enhancing the protein expression of the classical hallmark of autophagy activation, the microtubule-associated protein light chain 3-II (LC3-II), but also regulating the levels of several molecules known as hallmarks of apoptosis but also involved in the autophagic process (38, 39, 41). Indeed, D5DR activation increases reactive oxygen species (ROS) levels, rises the cleavage of poly-(ADP-ribose) polymerase (PARP) and the Caspase-3, and decreases the superoxide dismutases (SOD) protein expression (38, 39, 41). Interestingly, BRC and CAB have been found to prompt cell death via different pathways; BRC induces prolactinoma cell death mainly through the apoptosis pathway, while CAB induces prolactinoma cell death mainly via the autophagic cell death pathway (40).

Interestingly, D2DR has been also demonstrated to be involved in the growth and invasiveness as well as in the growth of stem-like cells, the main source of drug resistance, of nonfunctioning PitNET. Indeed, the activation of D2DR significantly induces antiproliferative effects by activating ERK1/2, p38 MAPK and caspase-3 in primary cultures of nonfunctioning PitNET (42) and induces antiproliferative effects on primary stem-like cell spheres of nonfunctioning PitNET (43). Moreover, the activation of D2DR significantly reduces the migration and invasion of a human cell model and of primary cultures of nonfunctioning PitNET, through the Rho-associated protein kinase (ROCK)-dependent LIMK (LIM kinase, an actin-binding kinase) activation. D2DR activation of ROCK/LIMK pathway determinates the inactivation of the protein cofilin, an actin-binding protein that regulates filament dynamics, with a consequent loss of its ability to bind the actin and, thus, promoting cell migration (44).

Guideline for hyperprolactinemia has defined the DAs resistance as a failure to achieve normal PRL levels together with a ≥ 50% reduction in tumor size at maximally tolerated doses (6). Only a small subset of patients with prolactinomas does not respond to DAs treatment (primary resistance), as approximately 20-30% is resistant to BRC and around 10-20% to CAB (24, 45), due to modification of DRs expression or to intracellular molecular mechanisms. Very infrequently patients with prolactinomas experience a delayed resistance becoming refractory to DAs prolonged treatment (secondary resistance) (46–51). This condition is generally considered an expression of a negative prognosis as it might indicate a malignant transformation of the prolactinoma (46).

In GH-secreting PitNET, the efficacy of DAs, particularly CAB, is remarkably lower than prolactinomas, regardless of their administration as monotherapy or in addition to other compounds; similarly, in ACTH-secreting PitNET the therapeutic efficacy of DAs, mainly CAB, is limited by a decreased response over time leading to escape to treatment (12, 52, 53).

Several molecular mechanisms, reviewed in the following sections and summarized in Figure 1 , seem to play a role in DAs resistance.

Generally, the effectiveness of BRC or CAB has been related to a decreased expression in DRs, as the expression of such receptors has been shown to correlate with responsiveness to therapy in lactotroph, somatotroph, corticotroph and in clinically nonfunctioning PitNET (12, 54–58). Particularly, a lower messenger expression of D2S and D2S/D2L ratio has been confirmed in the DAs resistant pituitary tumors (12, 27, 28, 54, 56, 59), although in a Japanese study the resistance to DAs in prolactinomas has been surprisingly correlated to a reduced messenger expression of D2L receptor subtype (57). In contrast to these studies, nonfunctioning PitNET response to DAs treatment is not related to D2DR mRNA and protein expression (60, 61), suggesting that other additional factors may mediate the growth inhibitory effects of DAs in nonfunctioning PitNET.

In conclusion, the pharmacological efficacy of DAs used for the medical treatment of PitNETs is based on several multifaceted mechanisms, mainly investigated in prolactinomas. These mechanisms include the reduction of D2DR expression, regulated by internalization and down-regulation processes orchestrated by cytoskeleton proteins and β-arrestins; changes in the proportion of D2DR subtypes; the regulation of cell signalling pathway of D2DR; and the increase of angiogenic and fibrotic markers. Growing evidence has highlighted the role of miRNAs in DAs resistance, particularly through the regulation of fibrotic pathway, although further studies are required to better elucidate the burden and the role of molecular mechanisms underlying pharmacological resistance to DAs.

CP conceived and wrote the manuscript. RPa, MN, and RPir contributed to the literature search and manuscript preparation. AC and RPiv provided a significant expert contribution to the scientific revision process. RSA supervised the manuscript drafting and critically reviewed the content. All authors contributed to the article and approved the submitted version.

This work was supported by the Ministry of Education, University and Research Grants PRIN 2017S55RXB (to RSA) and by POR FESR CAMPANIA 2014–2020 “Diagnostic and therapeutic innovations for rare neuroendocrine, endocrine tumors and glioblastoma by an integrated technological platform of clinical, molecular, genomic, ICT, pharmacological and pharmaceutical expertise” RARE.PLAT.NET (to AC).

RPiv has been the principal investigator of Clinical and/or Translational Research Studies for Novartis, HRA Pharma, Ipsen, Shire, Corcept Therapeutics, Cortendo AB-Strongbridge Biopharma, Janssen Cilag, Camurus and Pfizer; Co-investigator of Research Studies for Pfizer; received research grants from Novartis, Pfizer, Ipsen, HRA Pharma, Shire, IBSA, Strongbridge Biopharma; has been an occasional consultant for Novartis, Ipsen, Pfizer, Shire, HRA Pharma, Cortendo AB-Strongbridge Biopharma, Ferring, Recordati Rare Disease, Corcept Therapeutics, Crinetics Pharmaceuticals, ARH Healthcare, Biohealth Italia, Damor Farmaceutici; and has received fees and honoraria for presentations from Novartis, Shire, Pfizer and Recordati beyond the confines of this work. CP received research grants from Corcept Therapeutics. AC has been the principal investigator of Research Studies for Novartis, Ipsen, Pfizer, Lilly, Merck and Novo Nordisk; consultant for Novartis, Ipsen, Pfizer, and received honoraria from Novartis, Ipsen and Pfizer beyond the confines of this work.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.