Aripiprazole use as a cause of dopamine agonist failure in the treatment of prolactinomas

Introduction

Prolactinomas refer to prolactin-secreting pituitary tumours of lactotroph lineage. With a prevalence of 50 per 100,000 persons, they account for approximately 50% of all functional pituitary adenomas in clinical practice (Chanson & Maiter 2019). Affected women typically present with microprolactinomas (<1 cm tumour diameter), at least partly due to the high burden of endocrine symptoms such as menstrual disturbance and galactorrhoea (Lamba et al. 2019). In contrast, approximately 80% of men present with symptoms of tumour mass effect (Nishioka et al. 2003), with macroprolactinomas in men inherently considered an aggressive pituitary tumour subtype (De Sousa & McCormack 2022). Dopamine agonist (DA) therapy is a highly effective first-line therapy to improve endocrine function and reduce tumour mass (Melmed et al. 2011), with cabergoline having superior efficacy to bromocriptine and quinagolide (Petersenn et al. 2023). Whilst cabergoline is effective in >80% of prolactinomas, this and other DA agents carry a risk of drug-induced psychosis as dopamine is an established central mediator of psychosis, with most existing antipsychotics acting via dopamine antagonism (De Sousa 2022). DA use is also associated with increases in impulse control disorders, anxiety and depression (De Sousa et al. 2019).

The antipsychotic agent, aripiprazole, has been proposed as a treatment for both hyperprolactinaemia and psychiatric disease as it is a partial agonist of dopamine receptors (Hoffer et al. 2009). The recent Pituitary Society consensus statement on prolactinomas suggested that aripiprazole use in individuals with hyperprolactinaemia and an underlying psychiatric disorder could enable dose reduction or even cessation of concurrent DA therapy in patients with a prolactinoma (Petersenn et al. 2023). We present a cautionary case report of profoundly reduced DA efficacy in an aggressive prolactinoma immediately after commencing aripiprazole, as well as a systematic review of aripiprazole use in prolactinomas.

Case report

A 41-year-old male was incidentally found to have a pituitary mass after presenting with palatal myoclonus. He had a past medical history of nasal polyps, asthma, coeliac disease, gastro-oesophageal reflux disease and depression/anxiety, with his only medications being escitalopram 10 mg daily and esomeprazole 20 mg daily. His depression/anxiety was very well controlled on the low-dose escitalopram, not requiring any psychology or psychiatry input.

A large 44 × 40 × 37 mm rounded destructive soft tissue mass centred on the pituitary fossa was seen on a magnetic resonance imaging (MRI) scan of the brain, abutting and elevating the optic chiasm, expanding and destroying bone, eroding into the sphenoid sinus, the clivus and down towards the hypoglossal canal. The patient reported low mood and fatigue for 2–3 years prior, along with erectile dysfunction, low libido and difficulty losing weight. On examination, he had bilateral tender gynaecomastia but no other features of pituitary hormone dysfunction. There were no visual field deficits.

View Full Size

Baseline brain MRI on presentation (A, B) and postoperative brain MRI (C, D).

Citation: Endocrine Oncology 5, 1; 10.1530/EO-24-0065

• Download Figure
• Download figure as PowerPoint slide

A pituitary panel revealed an elevated prolactin level of 53,516 mU/L (73–407 mU/L), a low testosterone level of 3.79 nmol/L (8.0–30.0 nmol/L), an inappropriately low-normal luteinising hormone level of 2.6 U/L (0.6–12.0 U/L) and a follicle-stimulating hormone level of 2.5 U/L (1.0–12.0 U/L), in keeping with a prolactinoma and associated secondary hypogonadism. Macroprolactin was excluded using polyethylene glycogen precipitation. Insulin-like growth factor 1 (IGF-1), thyroid stimulating hormone (TSH), free thyroxine (fT4), morning cortisol and adrenocorticotropic hormone (ACTH) were normal. Visual field perimetry confirmed intact vision. Cabergoline was commenced as the first-line therapy, with potential adverse effects of impulse control disorder, mood effects and cerebrospinal fluid rhinorrhoea discussed. Prolactin peaked at 73,155 mU/L just before the commencement of cabergoline 250 μg twice weekly which was well tolerated symptomatically, escalating to 500 μg twice weekly after 4 weeks.

After 3 months of cabergoline therapy, there was a good biochemical response with a 93% reduction in serum prolactin to 5,047 mU/L, and modest tumour shrinkage from 44 × 40 × 37 to 43 × 32 × 32 mm on repeat MRI. However, the patient had developed insomnia, emotional lability, low mood, lethargy and intermittent suicidality whilst on 500 μg twice weekly of cabergoline. Noting the absence of other discernible factors accounting for his mood change, and the clear temporal relationship with cabergoline initiation, his mood symptoms were attributed to DA therapy. Escitalopram was increased to 20 mg daily, and an urgent psychiatry review was arranged. Aripiprazole 5 mg daily was then added to augment the antidepressant effect of escitalopram. The case was discussed at a pituitary multi-disciplinary team meeting where surgery was considered, with the final consensus to continue cabergoline as long as side effects were tolerable without suicidality, given the tumour was responding well biochemically and radiologically.

Whilst aripiprazole quickly abated his suicidal ideation, there was a concerning excursion of his prolactin level from 5,047 to 9,618 mU/L one week after initiation of aripiprazole (190% rise), and then to 17,612 mU/L a month later (349% rise from the pre-aripiprazole level) despite halving the aripiprazole dose due to the side effect of akathisia. No accompanying imaging was performed at this time. Cabergoline was maintained at 500 μg twice weekly throughout this period; hence, the rise in prolactin was attributable to aripiprazole. There was consideration for intrinsic DA resistance, sometimes seen in aggressive macroprolactinomas. However, given that prolactin levels were monitored just before and one week after commencement of aripiprazole, the extremely temporal relationship of such a dramatic rise was most in keeping with aripiprazole effect rather than coincident new-onset DA resistance of the prolactinoma. Notably, serum prolactin did not rise at any time during DA therapy until aripiprazole was commenced. Overall, the trajectory indicated that aripiprazole directly and significantly interfered with the efficacy of cabergoline. Considering the mixed agonistic and antagonistic effects that aripiprazole can have on dopamine receptors, this was assumed to be due to either the partial DA effect of aripiprazole competing at the D₂ dopamine receptor and evidently having greater binding affinity than cabergoline (a full DA) or aripiprazole-induced antagonism of dopamine receptors.

After extensive discussions with the patient, a decision was made to pursue surgical resection due to the adverse effects associated with cabergoline (suicidality) and aripiprazole (akathisia and apparent DA failure). Both medications were ceased, and the patient underwent transsphenoidal resection without complications. Residual tumour was evident postoperatively within the right cavernous sinus (Fig. 1), and the prolactin level improved but remained elevated at 5,833 mU/L, as expected, given the degree of tumour invasion. Histopathology confirmed a prolactinoma with positive immunostaining for PIT-1 and prolactin and an elevated Ki-67 proliferative index of 8.2%. With the aim of reducing the risk of residual disease progression, the patient accepted adjuvant radiotherapy (54 Gy/30 fractions) and recommencement of the maximal tolerated dose of cabergoline without associated psychological adverse effects at 250 μg twice weekly whilst awaiting radiotherapy effect. He has remained off aripiprazole. Prolactin levels have dramatically improved, most recently at a nadir level of 671 mU/L (Fig. 2) two months after transsphenoidal resection with a completed course of external beam radiotherapy.

View Full Size

Prolactin trend with the addition of aripiprazole to stable cabergoline dosing.

Citation: Endocrine Oncology 5, 1; 10.1530/EO-24-0065

• Download Figure
• Download figure as PowerPoint slide

Systematic review

We subsequently conducted a comprehensive search of the MEDLINE database, using relevant MeSH terms, i.e. ‘aripiprazole’, ‘prolactinoma’, ‘microprolactinoma’, ‘macroprolactinoma’, ‘adenoma’, ‘microadenoma’ and ‘macroadenoma’ (Fig. 3). Records were screened to eliminate irrelevant articles. Articles were then retained only if they included original case descriptions of individuals with biochemically and radiologically confirmed prolactinomas treated with aripiprazole, with reporting of prolactin levels correlating to the chronology of DA and aripiprazole therapy. Overall, 18 articles were reviewed as full texts, with 12 deemed suitable for analysis (Tables 1, 2, 3, 4). Of the cases sourced, all patients fell into one of the four following categories: those who had their prolactinoma treated with aripiprazole monotherapy and no DA, those whose DA therapy ceased when aripiprazole commenced, those who had DA and aripiprazole commenced simultaneously and those who had aripiprazole followed by concomitant DA therapy.

View Full Size

Identification of studies via databases and registers.

Citation: Endocrine Oncology 5, 1; 10.1530/EO-24-0065

• Download Figure
• Download figure as PowerPoint slide

There were three reported cases of aripiprazole monotherapy (Table 1). A primary psychiatric disorder was present in all cases, along with features of hyperprolactinaemia prompting a workup for a prolactinoma. All three cases had mild prolactin elevation at the time of diagnosis, ranging from two- to four-fold the upper limit of normal, with maximal dimension of the adenoma between 8 and 10 mm. High-dose aripiprazole was utilised in all three cases at a dose of 30 mg per day. Prolactin levels normalised in cases 1 and 2, with mild elevation remaining in case 3 at 1.5 times the upper limit of normal. The only case to document any follow-up imaging was case 3 with ‘no further increase in size’ reported (Steinhagen 2007, Shao-Tsu & Yi-Lin 2010, Jayaram et al. 2011).

In four cases, aripiprazole was commenced concurrently with DA cessation (Table 2). All had visible microprolactinomas (maximal tumour diameter 2–8 mm) with intolerable side effects from DA therapy and the emergence of psychotic symptoms felt to be related to DAs. Case 4 had trialled cabergoline, quinagolide and bromocriptine consecutively before cessation of all DA agents, whilst Cases 5, 6 and 7 switched directly from cabergoline to aripiprazole. Aripiprazole doses varied between 3.75 and 30 mg daily. Pre-aripiprazole prolactin levels whilst still on DA therapy were normal in cases 5, 6 and 7 and elevated in case 4 at 3,000 mU/L (approximately three-fold the upper limit of normal). Prolactin levels were within the normal range on follow-up in all four cases, and repeat imaging showed stable or reduced size of the adenomas (Freeman et al. 2007, Burback 2015, Bakker et al. 2016, Springer et al. 2023).

Five cases commenced DA therapy and aripiprazole simultaneously (Table 3). Cases 8 and 9 had pre-existing prolactinomas, which had been ‘lost to follow-up’, returning to medical attention with a psychotic episode, not on any DA therapy at that time. In case 8, the prolactinoma was ‘abutting the optic chiasm’ (no dimensions reported) with a serum prolactin level of 39,160 mU/L, and the patient was commenced on bromocriptine 2.5 mg daily alongside 15 mg daily of aripiprazole. After just 4 weeks of therapy, the prolactin level had normalised; no repeat imaging was documented (Sheldrick & Grunder 2008). Case 9 is the only male identified in the systematic review. He was commenced on aripiprazole 15 mg daily in addition to cabergoline, which was rapidly increased to a high dose at 4 mg twice weekly. He had a macroadenoma (13 × 15 × 16 mm) with prolactin elevation to 20,319 mU/L. After 6 months of dual therapy, his prolactin level had not completely normalised at 1,383 mU/L and his adenoma size had marginally reduced to 11 × 12 × 14 mm (Tsigkaropoulou et al. 2012). Cases 10 and 11 were undergoing treatment for underlying psychotic disorders with risperidone when symptoms consistent with hyperprolactinaemia prompted evaluation for a prolactinoma. Once prolactinoma was diagnosed, risperidone was ceased and both cases 10 and 11 simultaneously commenced aripiprazole and DA therapy. Case 10 commenced cabergoline 1 mg weekly and aripiprazole 5 mg daily. There was a small adenoma present (2 × 3 mm) along with prolactin elevation to 3,553 mU/L (approximately seven-fold the upper limit of normal). Prolactin normalised a month after commencing treatment; no repeat imaging was reported (Liao & Bai 2014). Case 11 underwent primary surgical resection of her prolactinoma due to concerns regarding starting DA therapy with a severe underlying psychotic disorder. Residual disease was present postoperatively on MRI with associated prolactin elevation at 4,468 mU/L (approximately nine-fold the upper limit of normal). She was simultaneously commenced on quinagolide 75 μg daily and aripiprazole 30 mg daily. Serum prolactin fell but was still elevated 5 months later at 2,617 mU/L, and repeat imaging demonstrated a ‘reduction in size’ (Broekhof et al. 2012). Case 12 had limited details available but had a previously known macroprolactinoma of 25 mm in maximal dimension and a suppressed prolactin level (212 mU/L) whilst on DA therapy (and not on any antipsychotics). The patient then presented with a psychotic exacerbation whilst not taking either antipsychotics or DAs and was recommenced with dual cabergoline and aripiprazole. This was followed by moderate prolactin elevation to 2,447 mU/L and a modest reduction in tumour maximal diameter from 25 to 21 mm at the last follow-up. Overall, in these five cases of simultaneous DA and aripiprazole commencement, four had significant reductions in prolactin (normalisation in two) and the largest lesion (case 12) had a significant prolactin rise. The three with reported post-treatment MRIs showed some tumour shrinkage but not resolution.

Case 13 is the only report of a patient with a prolactinoma initially treated with aripiprazole monotherapy, followed by the addition of DA therapy (Table 4). Limited demographic data was available. A giant prolactinoma with maximal dimension of 69 mm and prolactin level of 683,298 mU/L was evident before the commencement of cabergoline. Both parameters improved with the addition of DA therapy but did not lead to remission, with the tumour reducing in size to 30 mm and prolactin level still elevated at 12,765 mU/L. After cessation of aripiprazole and rotation to risperidone (doses not specified for either drug), prolactin normalised to 425 mU/L and the lesion reduced in size to 2 mm despite remaining on a stable dose of 1.5 mg/week of cabergoline (Allard et al. 2020).

When considering all cases, aripiprazole use (either as monotherapy or alongside DA therapy) was associated with prolactin normalisation in 8 of 13 individuals and partial prolactin reduction in 5 of 13 individuals. Tumour shrinkage was evident in 5 of 7 cases with reported imaging and radiologically stable in 2 of 7. In contrast to the present case, in no previous case was aripiprazole added to pre-existing DA therapy, and hence, whether aripiprazole lowered DA efficacy could not be determined.

Discussion

Aripiprazole is a partial DA, having a high intrinsic agonistic effect at dopamine D₂ receptors. This differs from the majority of antipsychotics that express their function through antagonism of postsynaptic D₂ receptors (Mohr et al. 2022). From this effect, it has been demonstrated that aripiprazole can lower prolactin, contrasting with other antipsychotics that frequently cause a prolactin increase (Leucht et al. 2013). It has been suggested that for patients with coexisting psychotic disorders and prolactinomas, there should be consideration of switching antipsychotic treatment to aripiprazole to reduce the dose of cabergoline required (Molitch 2020). Until now, there has been no evidence about how this partial agonistic effect interacts with cabergoline, bromocriptine and quinagolide, which constitute the first-line therapy for prolactinomas through their full D₂ receptor agonistic effect (Chanson & Maiter 2019).

Herein, we show that aripiprazole competes effectively with cabergoline and may therefore cause DA treatment failure as exhibited in our case. Although this is a new finding, it is not unexpected, given the higher binding affinity of aripiprazole for D₂ receptors (Ki = 0.34 nM) (Bartolomesis et al. 2015) in comparison with cabergoline (Ki = 0.7 nM) (Odaka et al. 2014). It has also been demonstrated in an ex vivo setting that aripiprazole acts as an antagonist of pre-synaptic D₂ receptors in the presence of high dopaminergic tone (Ma et al. 2015), which is the expected environment with DA use. The clinical impact of reduced DA efficacy will depend on the aggressiveness of the prolactinoma. We especially recommend caution with the use of aripiprazole in DA-treated macroprolactinomas in men, given that these are inherently aggressive pituitary tumours where effective prolactinoma therapy is essential to reduce the risk of tumour growth and subsequent structural effects, such as visual loss and hypopituitarism. This is particularly noteworthy as men appear more susceptible to DA-induced neuropsychological toxicity (De Sousa et al. 2019) and may therefore be considered for add-on aripiprazole therapy more frequently.

From the data summarised in Table 1, aripiprazole can be an effective monotherapy for prolactinomas, although all three cases were women with microprolactinomas and mild hyperprolactinaemia (i.e. less than four-fold the upper reference range) treated with high-dose aripiprazole. A similar pattern is seen in Table 2 with patients switched from DA therapy to aripiprazole, whereby three of the four individuals had a microprolactinoma and disease that was either well controlled or only mildly elevated at the time of rotation. Case 7 had a large macroprolactinoma with normoprolactinaemia achieved through DA therapy, and this was notably maintained when switching to high-dose aripiprazole. These two cohorts reinforce the notion that a ‘mild’ prolactinoma with only slight prolactin elevation could be treated with aripiprazole monotherapy. This should be considered in cases of prolactinoma with comorbid mood disturbance/psychosis that is either endogenous or induced by DA therapy. Case 13 represents an aggressive macroprolactinoma which was not responsive to aripiprazole monotherapy. Normoprolactinaemia was only achieved once aripiprazole was ceased, accompanied by a dramatic reduction in size and almost radiological resolution (Table 4).

Given that aripiprazole appears to be effective for at least mild disease, it raises the question as to whether aripiprazole could be concomitantly used with DAs. There is some data to suggest the efficacy of co-prescription DA therapy and aripiprazole as seen in Table 3, although it is worth noting that in case 9 (only male patient, aggressive disease), there was an incomplete response despite very high dose cabergoline. Case 11 also demonstrated a potentially aggressive prolactinoma, with high levels of prolactin elevation and local invasion, ultimately managed with upfront surgery due to coexistent severe psychosis. Despite standard DA therapy, residual disease remained radiologically and biochemically with prolactin level elevation greater than five-fold the upper reference range. Both cases may be reflective of underlying DA resistance, as seen in aggressive prolactinomas (Chatzellis et al. 2015, Valea et al. 2022), but could alternatively be explained by aripiprazole competing with cabergoline at the D₂ receptor, thereby limiting DA efficacy. By contrast, our case report revealed a clear, sustained aripiprazole-induced rise in serum prolactin compatible with DA treatment failure.

In conclusion, the case presented herein is the first case, to our knowledge, to illustrate the effect of adding aripiprazole to existing DA therapy, finding clear prolactin excursion following aripiprazole commencement despite stable cabergoline dosing. This example of aripiprazole-induced DA treatment failure is an important caveat to the suggested use of aripiprazole in the latest prolactinoma guidelines (Petersenn et al. 2023). In particular, we caution against the addition of aripiprazole to DA therapy where timely prolactinoma treatment is essential – for example, individuals with aggressive prolactinomas or those presenting with visual defects. In addition, all patients on dual aripiprazole and DA therapy for prolactinomas should be closely monitored, especially if there are risk factors for tumour aggressiveness. More clinical and basic research is needed to better understand the interaction between aripiprazole and DAs on D₂ receptors and prolactinomas.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.

Funding

SD is supported by the Royal Adelaide Hospital Mary Overton Early Career Research Fellowship.

Ethical compliance

Written informed consent for publication of clinical details was obtained from the patient, and publication approval was obtained from the Central Adelaide Local Health Network Human Research Ethics Committee.