DRD2 expression based on 18F-fallypride PET/MR predicts the dopamine agonist resistance of prolactinomas: a pilot study

The dopamine agonists (DA) have been used widely to treat prolactinomas. However, it is difficult to predict whether the patient will be responsive to DA treatment. We aimed to investigate whether the in vivo expression of DRD2 based on 18F-fallypride PET/MR could predict the therapeutic effect of DA on prolactinomas. Seven patients with prolactinomas completed 18F-fallypride PET/MR. Among them, three patients underwent surgery and further tumor immunohistochemistry. Imaging findings and immunohistochemical staining were compared with treatment outcomes. 18F-fallypride PET/MR was visually positive in 7 of 7 patients, and DRD2 target specificity could be confirmed by immunohistochemical staining. A significantly lower tracer standard uptake value (SUV) could be detected in the resistant patients (n = 3) than in the sensitive patients (n = 4; SUVmean, 4.67 ± 1.32 vs. 13.57 ± 2.42, p < 0.05). DRD2 expression determined by 18F-fallypride PET/MR corresponded with the DA treatment response. 18F-fallypride PET/MR may be a promising technique for predicting DA response in patients with prolactinoma.


Introduction
Prolactinomas are well-differentiated tumors that originate from pituitary lactotroph cells, causing hyperprolactinemia and resulting in gonadal and sexual dysfunction [1,2].
Prolactinomas comprise 50-60% of all pituitary neuroendocrine tumors, and occur most frequently in women aged 20-50 years [1,2]. As the first-line treatment option, drug treatment for prolactinomas relies on the use of dopamine agonists (DA), such as bromocriptine (BRC) and cabergoline (CAB) [2,3]. The majority of patients treated with DA have curative effects in normalizing prolactin (PRL) levels and reducing tumor size [3]. However, some patients may require high doses or be resistant to medical therapy [3].
The dopamine receptor family is divided into five subtypes (dopamine receptor D1-5, DRD1-5) [4,5]. DRD1, DRD2, DRD4 and DRD5 are expressed in normal human pituitary glands or pituitary neuroendocrine tumors, except DRD3 [6,7]. Under normal circumstances, dopamine is implicated in activating DRD2 and inhibiting prolactin secretion by pituitary lactotroph cells [8]. DRD2 is also highly expressed in prolactinoma, which becomes a selective therapy target by DA [4,5]. However, approximately 20-30% patients with prolactinomas are resistant to BRC and around 10-20% to CAB. DA resistance has been defined as failure to achieve normal PRL levels together with a ≥50% reduction in tumor size at maximally tolerated doses. Generally, the resistance of BRC or CAB has been related to a decreased expression in DRD2, although other molecular mechanisms also underpin the DA resistance, such as abnormal expressed miRNAs, PR domain containing 2 with ZNF domain (PRDM2), and proline-rich protein BstNI subfamily 3 (PRB3), etc [9,10]. In other words, DRD2 density in prolactinomas may positively correlate with their clinical responsiveness to DA. However, the preclinical evaluation of DRD2 expression in prolactinomas is still rare.
Positron emission tomography (PET) is an effective imaging technique to study receptor distribution in the brains of live animals and humans. To date, PET studies in humans have been widely used to quantitatively analyze dopamine receptors in the central nervous system in vivo to diagnose and research a variety of neuropsychiatric diseases [11,12]. However, dopaminergic imaging has rarely been studied in prolactinomas. Only 11 C-N-methylspiperone, 11 C-raclopride, 18 F-FDOPA and 18 F-fallypride as PET tracers have been used in prolactinomas [13][14][15]. Due to its convenience and stability, 18 F-fallypride has recently been successfully tested as a PET tracer to study prolactinomas [15]. However, this study was performed using PET/CT which can not show the detailed anatomical characteristics when compared with PET/MR. In addition, 18 F-fallypride PET/MR was not used to study prolactinomas.
In the present study, 18 F-fallypride PET/MR was first used to study prolactinomas in humans. The aim of the study was to demonstrate whether DRD2 expression based on 18 F-fallypride PET/MR could predict DA treatment resistance in prolactinomas.

Participants
For this pilot observation, we recruited 7 patients with prolactinomas underwent molecular imaging with 18 F-fallypride PET/MR. Inclusion criteria were patients diagnosed with prolactinoma and who had signed an informed consent form. Exclusion criteria included 1) declining to participate, 2) concomitantly taking psychotropic drugs or other drugs causing elevated prolactin, 3) being diagnosed with Parkinson's disease or taking dopaminergic agents, 4) having received gamma knife treatment, 5) being pregnant or lactating, or preparing for pregnancy, 6) having claustrophobia, and 7) poor compliance. After diagnosis, patients received the following medication: less than 15 mg per day of bromocriptine or 3 mg per week of cabergoline for at least 3 months. Regular blood tests and sellar enhancement MRI were performed on patients every 3 months. Parameters for outcome assessment included prolactin level, tumor size, visual field defects, and other symptoms. Failure to achieve normalization of prolactin levels and failure to achieve at least 50% tumor shrinkage after at least 3-month either bromocriptine (Because cabergoline is not available in China) or cabergoline treatment were regarded as resistance to DA in our study (normal prolactin levels were defined as less than 25 ng/ml for females, and 20 ng/ml for males) [16]. Demographic information and follow-up data were collected in the database for clinical reference and research purposes. The local ethics committee approved the study protocol, and all participants signed informed consent forms. The pilot study has applied for a clinical trial (ClinicalTrials.gov Identifier: NCT03717454).

PET/MR imaging and analysis
Simultaneous PET/MR images were acquired using a Biograph mMR scanner (Siemens Healthcare, Erlangen, Germany), and the MR field strength is 3.0 T. PET/MR studies were performed and reviewed by two board-certified specialists in radiology and nuclear medicine (X. L. and X. H.), were blinded to the clinical data. Standard uptake value (SUV) was used for semi-quantitative PET analysis, which refers to the radioactive activity of the PET tracers in tumor and normal tissues. The maximum of SUV (SUV max ) and mean of SUV (SUV mean ) of the tumor and the SUV mean of the reference regions including the cerebellum, hypothalamus and substantia nigra were documented. Details are described in the Supplementary Materials.

Immunohistochemistry
Immunohistochemistry staining of PIT-1, prolactin, CAM5.2, DRD2, and DRD3 were performed on 10% formalin-fixed, paraffin-embedded prolactinoma surgery samples from 3 of 7 patients. Details are described in the Supplementary Materials. As for tumors with apoplexy, the parenchyma, but not the cystic samples were harvested for the DRD2 immunohistochemistry.

DRD2/3 expression in normal tissues and PitNETs
The NIH Genotype-Tissue Expression (GTEx) project was created to establish a sample and data resource for studies on the relationship between genetic variation and gene expression in multiple human tissues (http://genome.ucsc. edu). DRD2 and DRD3 expression data in 54 tissues from GTEx RNA-seq of 17382 samples and 948 donors (V8, Aug 2019) were downloaded. In our previous study, a total of 180 pituitary tumors were collected to perform transcriptome analysis [17]. Data are available athttp://www. biosino.org/node/project/detail/OEP001353. DRD2 and DRD3 expression profile was generated.

Statistical analysis
Statistical analyses were performed using Prism software (version 6.0; GraphPad Software, Inc.). The results are shown as the mean ± SEM as indicated. Differences between two groups were compared with a non-parametric test. All statistical tests were 2-sided, and a p value of less than 0.05 was considered to indicate statistical significance.

Results
All patient characteristics are presented in Table 1. There were 5 male and 2 female patients with ages ranging from 15 to 56. The patients' serum prolactin levels were all above 200 ng/ml. Six patients had macroadenomas, defined as tumors larger than 1 cm, with maximum diameters ranging from 1.3 to 6.5 cm. In these 7 patients, prolactinomas with cavernous sinus invasion occurred in 4 cases. Three patients were resistant to DA treatment with Knosp grade 3 or 4. Moreover, two patients received operations and one patient decided to continue the medicine treatment. The other four patients were sensitive to DA treatment. However, one patient underwent emergency surgery due to acute apoplexy during the medicine therapy.
DRD2-directed PET/MR imaging with 18 F-fallypride was visually positive in all patients with prolactinomas (Fig. 1). first underwent the PET/MR examination, then those patients received the DA treatment; patient 7 is sensitive to DA treatment with the normal PRL level, but tumor apoplexy was occurred during the medication, PET/MR was examined before the emergency surgery  Table 2). Individual imaging results can be found in Supplemental Table 1. In addition, the SUV ratios of the target/ cerebellum, target/hypothalamus, target/substantia nigra, or putamen/target in resistant and sensitive prolactinomas had no significant differences (p > 0.05, Table 2).
To investigate the DRD2 and DRD3 distribution in the brain and other tissues, we employed the GTEx database. In the pituitary gland, DRD2 had a high level of expression, but DRD3 had almost no expression (Supplemental Figs. 1 and 2). In addition, our RNA sequencing data from a total of 180 cases further revealed that DRD2 was expressed in PitNETs rather than DRD3 (Supplemental Figure 3). Overall, normal human pituitary glands and PitNETs only express DRD2, but not DRD3.
Finally, we verified whether DRD2 expression based on 18 F-fallypride PET/MR was correlated with the expression of the tumor itself. In this pilot study, three patients had their tumor specimens evaluated after the surgery. Immunohistochemistry staining showed that these prolactinomas were positive for PIT-1, prolactin and CAM5.2 expression, and negative for DRD3 expression (Fig. 2). DRD2 in Patient 7 was positive, but it was negative in Patient 2 and Patient 3, which is consistent with the results of 18 F-fallypride PET/MR.

Discussion
In this single-center study, the first case series of patients with prolactinomas completed 18 F-fallypride PET/MR to reveal the expression of DRD2. Previous studies of dopamine receptor distribution in striatal and extrastriatal regions revealed that 18 F-fallypride exhibited high affinities for DRD2 and DRD3 [18][19][20]. The data from GTEx confirmed that DRD3 was expressed in the striatum and nucleus accumbens, but not in the pituitary gland. There is also no sign of DRD3 expression in PitNETs in the literature. In our previous RNA-sequence data from 180 cases of patients with all types of PitNETs, DRD2 generally had a high level of expression, and DRD3 had almost no expression. The three cases of prolactinomas operated on with transsphenoidal surgery also showed no DRD3 expression in immunohistochemistry analysis. These results indicated that 18 F-fallypride only exhibits DRD2 expression in prolactinomas, which might be a sufficiently reliable indicator of the response to DA therapy. Thus, 18 F-fallypride PET/MR may greatly help predict reactivity of dopaminergic treatment and achieve individualized stratified treatment of prolactinomas.
Nuclear medicine methods are clinically validated and have been increasingly integrated into the management of pituitary adenomas [21]. Recently, Damian et al. described that 18 F-fallypride PET/CT coregistered with MRI was used to reveal the expression of DRD2 in prolactinomas and nonfunctioning adenomas [15]. Compared to sequentially acquired PET and MRI, integrated PET/MR systems provide more accurate images, shorten acquisition time, and reduce motion artifacts [22]. Hybrid PET/MR is a unique imaging technology for simultaneous acquisition and analysis of the morphologic and functional data. Given the lower soft tissue resolution and risk of ionizing radiation of PET/CT, PET/MR is also a superior and more favorable imaging modality for tumors [23,24]. In several types of tumors, PET/MR is feasible in terms of technique, reproducibility of results, and comparability of parameters [23][24][25][26]. Our study also indicated that simultaneous PET/ MR is a promising technology for detecting of prolactinomas and predicting of DA therapy. This study has several limitations. Firstly, the number of patients recruited was limited partly because of the expensive imaging cost. Secondly, tumor samples from sensitive patients were not always available, making it difficult to judge the expression of DRD2 in many sensitive tumors. Thirdly, the integrated PET/MR systems can only identify prolactinomas more than 4 mm, which limits the use of this technique to patients whose tumor size <4 mm. Fourthly, our prior study has revealed that DA treatment could mediate internalization of DRD2 proteins [4], which theoretically may influence the SUV values. Further classification of patients underwent 18 F-fallypride PET/MR before or after DA treatment should be made.

Conclusion
Our data are the first to demonstrate that 18 F-fallypride PET/ MR is feasible to visualize and quantify DRD2 density and has the potential to predict dopaminergic treatment sensitivity in prolactinomas. Further evaluation in larger, prospective studies is warranted to further evaluate the predictive value of DA sensitivity for prolactinomas.