Hyperprolactinemia caused by extra-pituitary prolactin secretion: a systematic review

Abstract

This study aims to summarize the clinical and pathological characteristics, as well as the treatment and therapeutic outcomes, of hyperprolactinemia caused by ectopic prolactin secretion. Case reports of patients with hyperprolactinemia caused by extra-pituitary prolactin secretion were collected by conducting searches in three databases using the terms (ectopic prolactin secretion) OR (ectopic hyperprolactinemia) OR (ectopic prolactinoma). Fifty-two cases were included (age: 45.5 years (34–55.25), baseline serum prolactin level: 218 ng/mL (110.3–680.5)). Extra-pituitary prolactin-secreting sites include ectopic pituitary adenomas (age: 55 years (47–65), baseline prolactin level: 382 ng/mL (200–1,598)) and non-pituitary-derived extracranial lesions (age: 38 years (30–45.5), baseline prolactin level: 148 ng/mL (75.25–246)). The most common symptoms of the two types of patients are, respectively, intracranial mass effect and galactorrhea or amenorrhea. 42.3% of cases received dopamine agonists as initial treatment, and among them, all patients with non-pituitary-derived lesions failed to achieve normalization in prolactin levels by receiving medication alone. 38.5% of cases received surgery as initial treatment, and 70% achieved an immediate decrease in prolactin level. In conclusion, hyperprolactinemia caused by ectopic prolactin secretion is rare, but it should still be considered in patients with hyperprolactinemia of unclear cause. Age, baseline prolactin levels, major symptoms, histology, pathology, and therapeutic outcomes varied between patients with prolactin-secreting ectopic pituitary adenomas and non-pituitary-derived lesions. Hyperprolactinemia caused by the latter had a female predominance. Dopamine agonists were effective for most ectopic pituitary adenomas, while patients with non-pituitary-derived lesions tended to resist dopamine agonists and responded well to surgery.

Introduction

Prolactin (PRL) is a hormone that primarily stimulates lactation in the mammary glands. It is mainly secreted by the anterior pituitary gland lactotrophs, but can also be secreted by extra-pituitary tissues, either in normal or pathological states. Extra-pituitary PRL-secreting sites in non-pathological states include decidua, mammary tissue, ovary, male reproductive organs, endothelial cells, immune system, brain, hair follicles, skin, and adipose tissue (1). Pathological ectopic PRL secretion has been described in ectopic pituitary adenomas (EPAs), which are defined as pituitary-derived adenomas without any connection to the pituitary gland and stalk (2, 3, 4). EPAs can be located in the peri-pituitary region, including sphenoid sinus, clivus, suprasellar or parasellar locations, and nasopharynx (5, 6). It is accepted that PRL-secreting tumors located in peri-pituitary regions originate from remnants of Rathke’s pouch (7). Furthermore, pathological ectopic secretion of PRL was also found in extracranial non-pituitary-derived neoplasms and hematologic malignancies (8, 9, 10). The causes of PRL secretion from other ectopic lesions were not fully studied.

Clinically, hyperprolactinemia caused by extra-pituitary PRL secretion is difficult to diagnose, bringing difficulties to its efficient treatment. On the one hand, intracranial EPAs are not easy to distinguish from other intracranial tumors, including chordomas, meningiomas, craniopharyngiomas, germ cell tumors, chondrosarcomas, astrocytomas, melanomas, and metastatic carcinomas (11). On the other hand, for patients with extracranial lesions, it is often challenging to associate hyperprolactinemia with their extracranial lesions during the diagnostic process.

Here, we provide a systematic review of hyperprolactinemia caused by extra-pituitary PRL-secreting lesions and compare the clinical and pathological characteristics, as well as treatment and therapeutic outcomes, between patients with EPAs and extracranial non-pituitary-derived lesions.

Materials and methods

This systematic literature review has followed the PRISMA guidelines and has been registered at PROSPERO (ID: CRD42024613890). The literature search was conducted in PubMed, Embase, and Web of Science with the following search terms: (ectopic prolactin secretion) OR (ectopic hyperprolactinemia) OR (ectopic prolactinoma). All case reports and case series published in English up to 1 June 2024 were considered, with no restriction on the start date. References of relevant review articles which meet the inclusion and exclusion criteria were also included. The inclusion criteria were as follows:

• Patients with hyperprolactinemia (>20 ng/mL in males and >30 ng/mL in females);

• Evidence excluding the presence of pituitary tumor was provided;

• Immunohistology of ectopic lesions was positive for PRL or shrinkage of the ectopic lesion was accompanied by a decrease in PRL level.

Patients with physiologic (pregnancy or lactation) or pharmacologic (taking dopamine D2 antagonists and selective serotonin reuptake inhibitors) factors that may cause hyperprolactinemia, patients with histories of pituitary diseases, and patients with certain complications (including hypothalamic-pituitary stalk damage, pituitary mass, chronic renal failure, and primary hypothyroidism) were excluded.

The analysis between the two groups was carried out using the unpaired Mann–Whitney test. P value was adjusted using the Bonferroni correction method. A P value of <0.05 was considered significant. Statistical analysis was done using SPSS Statistics, version 27 (IBM, USA). Data are expressed as median (interquartile range); proportions are expressed as numbers (percentage).

Results

The search results and screening process are summarized in a flowchart (Fig. 1). Our research identified 52 cases reported by 50 case reports published between 1971 and 2024 (Supplementary Table 1 (see section on Supplementary materials given at the end of the article)) (5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56), including 13 men (25%) and 39 women (75%). Baseline characteristics are presented in Table 1. The median age at presentation was 45.5 years (34–55.25), and the median baseline serum PRL level was 218 ng/mL (110.3–680.5). There was no significant difference in both age and baseline PRL levels between males (53 years (41–60), 269 ng/mL (159–881)) and females (41 years (34–49.5), 198 ng/mL (101–508)) (P = 0.16, P = 0.39).

Figure 1.

Flowchart of the literature search and screening. A total of 522 articles were initially identified through the literature search, and 77 articles were deemed relevant to the topic. Following the exclusion of studies that did not meet the inclusion and exclusion criteria, and the addition of eligible references, 50 case reports comprising 52 individual cases were included.

Table 1.

Baseline characteristics (n = 52) (5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56).

	Total	Men	Women
Cases	52	13	39
Age at presentation (years)	45.5 (34–55.25)	53 (41–60)	41 (34–49.5)
Serum PRL level (ng/mL)	218 (110.3–680.5)	269 (159–881)	198 (101–508)
Peri- pituitary tumors	25 (48.1%)	11 (84.6%)	14 (35.9%)
Sphenoid sinus	10 (40.0%)	2 (18.2%)	8 (57.1%)
Clivus	10 (40.0%)	6 (54.5%)	4 (28.6%)
Suprasellar region	4 (16.0%)	2 (18.2%)	2 (14.3%)
Nasopharynx	1 (4.0%)	1 (9.1%)	0
Extracranial solid tumors	23 (44.2%)	2 (15.4%)	21 (53.8%)
Uterus	11 (47.8%)	– –	11 (52.4%)
Ovary	5 (21.7%)	– –	5 (23.8%)
Abdomen	5 (21.7%)	1 (50.0%)	4 (19.0%)
Chest	1 (4.3%)	1 (50.0%)	0
Neck	1 (4.3%)	0	1 (4.8%)
Leukemia	4 (7.7%)	0	4 (10.3%)

Data are expressed as median (interquartile range); proportions are expressed as numbers (percentages).

Lesions in 25 cases (48.1%) were EPAs in the peri-pituitary region; the sphenoid sinus and clivus were the most easily affected parts. Lesions in the other 27 cases (52%) were non-pituitary-derived neoplasms, including leukemia (four cases) and solid tumors of the uterus, ovary, abdomen, chest, and neck (23 cases). 92.6% of the non-pituitary-derived lesions were found in females (25 cases). Baseline PRL level was significantly higher in patients with peri-pituitary EPAs (382 ng/mL (200–1,598)) than in patients with non-pituitary-derived lesions (149 ng/mL (74.93–257.5)) (P = 0.002). However, the maximum diameters of extracranial non-pituitary-derived neoplasms were significantly larger than those of EPAs (data from 25 cases provided). EPAs were primarily located in the sphenoid sinus and clivus, and many showed invasiveness, tending to invade surrounding bony tissues and cavernous sinus.

PRL-secreting EPAs

The 25 PRL-secreting EPAs included 11 men and 14 women. The median age at presentation and baseline serum PRL levels were 55 years (47–65) and 382 ng/mL (200–1,598), respectively. There were no significant differences in age and baseline PRL level between men (49 years (39–59), 269 ng/mL (191.7–940.7)) and women (58 years (48.5–65.8), 508 ng/mL (216.8–1922)) (P = 0.24, P = 0.54). Aside from three cases that did not undergo biopsy, pathological and immunohistochemical examinations of the others all showed PRL-secreting EPAs. The maximum diameters of seven tumors were provided, all of which exceeded 10 mm, and the mean value was 23.57 mm. Imaging findings in 19 cases were described; among them, tumors in 16 cases had invaded surrounding bony tissue or cavernous sinus.

Presenting symptoms

The primary symptoms at presentation were intracranial mass effects, the most common being headaches (44%) and visual impairment (20%). Nasal obstruction and epistaxis were also caused by compression by the tumor located in the nasopharynx and sphenoid sinus. Only eight cases presented symptoms related to hyperprolactinemia, including irregular menstrual cycle (16%), galactorrhea (8%) and gynecomastia (8%) (Table 2).

Table 2.

Symptoms of patients with PRL-secreting EPAs (n = 25) (5, 6, 12, 13, 17, 19, 21, 23, 24, 26, 28, 29, 30, 31, 33, 34, 36, 37, 38, 44, 50, 51, 52, 54, 55).

	Total (n = 25)	Sphenoid sinus (n = 10)	Clivus (n = 10)	Suprasellar region (n = 4)	Nasopharynx (n = 1)
Hormone effects
Irregular menstrual cycle	4 (16%)	3		1
Galactorrhea	2 (8%)	1	1
Gynecomastia	2 (8%)		1		1
Decreased libido	2 (8%)			1	1
Sexual dysfunction	1 (4%)	1
Decreased testicular size	1 (4%)				1
Mass effects
Headaches	11 (44%)	5	6
Visual impairment	5 (20%)		2	3
Epistaxis	3 (12%)	2			1
Vomiting	1 (4%)			1
Nasal obstruction	1 (4%)				1
Rhinorrhea	1 (4%)	1
Tinnitus	1 (4%)	1
Hearing loss	1 (4%)		1
Retroorbital pain	1 (4%)		1
Diplopia	1 (4%)		1

Proportions are expressed as numbers (percentages).

Treatment

Treatments of cases with PRL-secreting EPAs were summarized in a flowchart (Fig. 2). Eleven cases were treated with dopamine agonists (DAs) initially, and six of them achieved normalization of serum PRL level directly after medication. Among the 11 cases, the medication histories of ten cases were described. Cabergoline (five of ten cases) and bromocriptine (four of ten cases) were both frequently used, and one case underwent a switch from bromocriptine to cabergoline due to poor remission but remained hyperprolactinemic after cabergoline treatment (12). On the other hand, 12 cases received surgery as initial treatment. Postoperative PRL levels in four cases reached normal and were only slightly elevated in two cases (34 ng/mL and 26 ng/mL, respectively), who thus did not receive further treatment. Five cases received DA therapy after surgery due to partial or poor remission, and all of them achieved normalization of PRL levels after medication. Among them, three cases took bromocriptine, and the other two cases received cabergoline. One case received DAs and surgery simultaneously and achieved full remission (29). One case received radiotherapy but died of intracranial infection 1 month after discharge. The serum PRL level after radiotherapy was not reported (34).

Figure 2.

Treatment of cases with PRL-secreting EPAs (n = 25). H: hyperprolactinemia. N/A: not available. N: normalized PRL levels. D: death. DA: dopamine agonist. Other: chemotherapy, radiotherapy, or immunotherapy.

Non-pituitary-derived PRL-secreting lesions

Twenty-seven of 52 cases were found to have extracranial non-pituitary-derived lesions with PRL-secreting activity, including two men and 25 women, exhibiting a female predominance. The median age at presentation was 38 years (30–45.5), and the median baseline PRL level was 148 ng/mL (75.25–246). The median maximum diameter of solid tumors was 99 mm (88.5–147.25, 18 of 23 cases provided).

Pathology

Twenty-five of the total 27 cases underwent pathological examinations, and among them, 19 cases underwent immunohistochemical staining for PRL, of which seven cases (36.8%) were positive for PRL (Table 3). Leiomyomas, perivascular epithelioid cell tumors (PEComas), dermoid cysts, uterine tumors resembling ovarian sex cord tumors (UTROSCTs), acute lymphoblastic leukemia (ALL), and acute myeloid leukemia (AML) were found in more than one case. Cases with neoplasms located in the uterus and abdomen had a relatively low positive rate for PRL staining (two of ten cases (20%) and one of four cases (25%), respectively). Ovarian neoplasms had a 100% positive rate in PRL staining. As for leukemia cases, bone marrow specimens were subjected to pathologic examinations, immunohistochemical staining, and RT-PCR (reverse transcription polymerase chain reaction) of PRL mRNA. Although PRL expression was not found in the leukemic blasts of one case with AML, interstitial mononuclear cells in the bone marrow specimen were stained positive for PRL. Furthermore, mRNA of PRL was found in both ALL cases under disease states, and repeated RT-PCR after chemotherapy of both cases revealed disappearance of PRL mRNA (8, 47).

Table 3.

Pathological and immunohistochemical results of non-pituitary-derived lesions (n = 27).

	Cases	IHC positive cases
Total	25/27, N/A: 2	7/19
Uterus	11/11	2/10, N/A: 1
Leiomyoma (9, 18, 25, 43, 45, 46, 48)	7	2/6, N/A: 1
Uterine tumor resembling ovarian sex cord tumor (20, 39)	2	0/2
Mesenchymal tumor (49)	1	0/1
Perivascular epithelioid cell tumor (16)	1	0/1
Ovary	5/5	3/3, N/A: 2
Dermoid cyst (27, 40, 56)	4	2/2, N/A: 2
Primary small cell ovarian cancer of pulmonary type (22)	1	1/1
Abdomen	4/5, N/A: 1	1/4
Perivascular epithelioid cell tumor (10, 41, 42)	3	1/3
Metastatic melanoma (32)	1	0/1
Neck	1/1	N/A: 1
Large cell lymphoma (15)	1	N/A: 1
Chest (53)	0/1, N/A: 1	– –
Leukemia	4/4	1/2, N/A: 2
ALL (8, 47)	2	0/1, N/A: 1
AML (14, 35)	2	1/1, N/A: 1

ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; N/A: not available. Proportions are expressed as included numbers/total numbers (percentages).

Presenting symptoms

The most common symptoms were hormone effects caused by hyperprolactinemia, namely galactorrhea (88.9%), irregular menstrual cycles (59.3%), decreased libido (14.8%), and gynecomastia (3.7%). Other symptoms include systemic symptoms and local mass effects caused by tumor compression (Table 4).

Table 4.

Symptoms of patients with non-pituitary-derived PRL-secreting lesions (n = 27) (8, 9, 10, 14, 15, 16, 18, 20, 22, 25, 27, 32, 35, 39, 40, 41, 42, 43, 45, 46, 47, 48, 49, 53, 56).

	Total (n = 27)	Uterus (n = 11)	Ovary (n = 5)	Abdomen (n = 5)	Chest (n = 1)	Neck (n = 1)	Leukemia (n = 4)
Hormone effects
Galactorrhea	24 (88.9%)	10	5	4		1	4
Irregular menstrual cycle	16 (59.3%)	9	5	2
Decreased libido	4 (14.8%)	2		2
Gynecomastia	1 (3.7%)					1
Other symptoms
Fatigue	4 (14.8%)	1		1			2
Abdominal pain	3 (11.1%)	1	1	1
Headaches	3 (11.1%)	2	1
Hot flashes	3 (11.1%)	1	1	1
Joint pain	3 (11.1%)	1					2
Left flank discomfort	1 (3.7%)			1
Hematuria	1 (3.7%)			1
Fever	1 (3.7%)			1
Cough	1 (3.7%)				1
Chest pain	1 (3.7%)				1
Anemia	1 (3.7%)						1
Swelling	1 (3.7%)						1
Weight loss	1 (3.7%)						1
Menorrhagia	1 (3.7%)	1
Dysmenorrhea	1 (3.7%)	1
Vaginal dryness	1 (3.7%)	1
Abnormal vaginal bleeding	1 (3.7%)	1

Proportions are expressed as numbers (percentages).

Treatment

Treatment and prognosis were summarized in the flowchart (Fig. 3). DAs have poor therapeutic effect in lowering serum PRL levels for patients with non-pituitary-derived neoplasms. All 11 cases who received DA therapy as initial treatment were insensitive to medication. Medication histories of ten cases were available. Among them, eight cases received cabergoline, and two cases received bromocriptine. After medication, they further received surgery (ten cases) or immunotherapy (one case), and all of them have achieved normalization of serum PRL level. However, recurrence of the PRL-secreting uterine tumor occurred in one case 2 years after surgery, leading to the recurrence of hyperprolactinemia. However, surgery, chemotherapy, and immunotherapy generated a good prognosis. Eight cases received surgery as initial treatment, and long-term normalization of PRL level was achieved in six of them after resection of the tumor. Recurrence, which subsequently led to hyperprolactinemia, occurred in two cases, and they underwent secondary surgery combined with chemotherapy, which generated a good prognosis in one case (16, 39). The prognosis of the other case was not provided. Serum PRL level in one case successfully decreased to normal after receiving DA combined with surgery as the initial treatment (22). Chemotherapy or immunotherapy for certain tumors has successfully normalized PRL levels in five cases, including the four cases with leukemia (8, 14, 15, 35, 47). Hyperprolactinemia can be misleading for the diagnosis of extracranial PRL-secreting neoplasms. One case had a dermoid cyst in the right ovary that was misdiagnosed as pituitary adenoma and received surgery of the pituitary at first, but the pathological examination of the biopsy revealed only blood clots and no tumor tissue. She then received DA but had recurring hyperprolactinemia after drug withdrawal. After a right oophorectomy, she finally achieved normalization of the PRL level. The pathological results of this case revealed a PRL-positive dermoid cyst (27).

Figure 3.

Treatment of cases with non-pituitary-derived PRL-secreting lesions (n = 27). H: hyperprolactinemia. R: recurrence. N: normalized PRL levels. N/A: not available. DA: dopamine agonist. Other: chemotherapy, radiotherapy, or immunotherapy.

Discussion

Hyperprolactinemia in pathologic conditions, primarily caused by prolactinoma, whose prevalence ranged from 25 per 100,000 to 63 per 100,000 (57), could also be caused by extra-pituitary PRL secretion. However, this condition is rare, with only 52 eligible cases identified in this systematic review. Two types of ectopic PRL-secreting lesions, namely EPAs and non-pituitary-derived lesions, which include solid tumors and hematologic malignancies, have been reported at present. Differences were found in sex preference, pathology, clinical presentation, and outcomes of the two types of ectopic PRL-secreting lesions and prolactinomas.

Sex preference of PRL-secreting EPAs does not conform to that of prolactinomas, exhibiting a lower female-to-male ratio. Prolactinoma has predominance in females between 20 and 50 years old (58), while we have only identified four female cases at age 20–50 years with PRL-secreting EPAs (13, 19, 24, 26). Higher age at presentation of the cases included in this review (55 years (47–65)) may lead to a similar prevalence in both genders, given that prolactinomas do not show preference in females among people over 50 years old (58).

Pathological examinations indicated that the PRL-secreting EPAs are homologous to prolactinomas, which revealed the embryonic origin of PRL-secreting EPAs. Those located at the sphenoid sinus, clivus, and nasopharynx are thought to arise from the pituitary rests along the developmental path of the anterior pituitary gland (59), while the suprasellar PRL-secreting EPAs were deemed to originate from peri-infundibular pituitary cells (60).

At presentation, 68% of EPA cases had symptoms related to intracranial mass effects, while only 32% of them presented symptoms related to hormone effects. This may be due to a high proportion of postmenopausal females in EPA cases that we have identified (ten of the 14 female cases were older than 50 years), given that specific hormone effects such as amenorrhea and galactorrhea are less noticeable in men and postmenopause women.

Empty sella was the most common complication, which was found in eight cases with EPAs located in the sphenoid sinus and/or clivus. A possible explanation is that the majority of anterior pituitary precursor cells remained in the sphenoid sinus and clivus and grew into EPAs, while only a few precursor cells migrated to the sella turcica during development, leading to an empty sella (12, 30, 61, 62, 63, 64). Dura mater in all eight cases was intact, which was confirmed by either imaging results or surgical exploration. This also ruled out the possibility that the empty sella developed secondary to the invasion and destruction of dura mater by a non-ectopic invasive pituitary adenoma (30).

DA therapy is the first-line treatment for pituitary prolactinomas, effectively reducing PRL levels and adenoma size (65). As with prolactinomas, six of 11 cases with PRL-secreting EPAs responded well to DA treatment (6, 13, 17, 23, 37, 55), and some of them required long-term DA therapy to maintain remission. This clinical responsiveness suggests the potential involvement of dopamine D2 receptors in EPAs, although direct evidence of D2 receptor expression in EPAs remains lacking.

As for non-pituitary-derived ectopic PRL-secreting lesions, they primarily affect women, mainly because ovarian and uterine neoplasms make up a large proportion (59.3% in our research) of these lesions. This may also be due to the fact that the symptoms of hyperprolactinemia, such as amenorrhea and galactorrhea, are more apparent in women, leading to earlier medical attention and diagnosis. Male patients with subclinical hyperprolactinemia caused by non-pituitary-derived PRL-secreting lesions may not be diagnosed as ectopic PRL-secreting cases. 96.3% of patients with non-pituitary-derived neoplasms seek doctors due to symptoms related to elevated PRL levels, which may be caused by the predominance of premenopausal females.

Non-pituitary-derived PRL-secreting neoplasms had various tissue origins and exhibited different histological characteristics. Neoplasms positive for PRL included leiomyomas, ovarian dermoid cysts, primary small cell ovarian cancer of pulmonary type, PEComas, and leukemia.

Leiomyomas may pathologically secrete PRL itself, given that extra-pituitary secretion of PRL has been found not only in decidua (1), but also in both endometrium (66) and myometrium (67). Another explanation for PRL secretion from leiomyomas was that immune reaction against the myoma led to PRL secretion from lymphocytes around the tumor (43, 68), since extra-pituitary PRL secretion has also been reported in lymphocytes (69). Other possible explanations include the secretion of a PRL-stimulating factor from the neoplasms or stress-associated PRL secretion induced by the mass effect of the neoplasm (43). However, normally, hyperprolactinemia is not common in patients with leiomyomas. The PRL-secreting activity of uterine neoplasms might be related to the relatively large size of the tumor and also the location of the tumor (68). PRL secretion from ovarian dermoid cysts could result from secretory lactotrophs in the dermoid. Pituitary components in dermoid cysts were first described by Akhtar in 1975 (70), and the existence of lactotrophs in dermoids was later revealed by McKeel in 1978 (71). Mechanisms of PRL secretion from PEComas were not clear. Extra-pituitary PRL secretion from the adrenal gland, sweat gland, and gut may provide possible explanations (72). Pathologic secretion of PRL in leukemia patients may be from leukemic blasts. In a cohort of 28 cases with AML, four cases were found to have hyperprolactinemia (>30 ng/mL), and the expression of PRL was detected in their leukemic blasts by immunoblotting (73). PRL expression was also found in both myeloid cell lines in vitro and the blasts of one AML patient (74). The expression of PRL in leukemic blasts might be activated by specific factors under pathologic conditions, as shown by Gerlo et al. who demonstrated that the proinflammatory cytokine tumor necrosis factor-alpha (TNFα) could activate an alternative promoter of PRL in myeloid leukemic cells (75). In addition, PRL may also be secreted by stromal cells in the bone marrow. In one AML case, PRL immunoreactive cells from the bone marrow were mainly interstitial cells, not the blasts (35). Bellone et al. have demonstrated PRL secretion from bone marrow stromal cells (76). PRL may act as a paracrine or autocrine factor in bone marrow, regulate immune and hematopoietic cell growth and differentiation, and is associated with various malignancies of the hematopoietic system, including leukemia (77).

There were neoplasms stained negative for PRL, though the secretion of PRL from these tumors could still be strongly supported by the immediate decrease in PRL level after resection of the tumor. In addition, in one PRL-negative case, the PRL concentration of peritumoral blood was 91.7 ng/mL, which was higher than that of peripheral blood, indicating direct PRL secretion from the ectopic tumor (46). The negatively stained tumors were large, with a median maximum diameter of 104.5 mm (80–142.5). Thus, presumably, the tumors might consist of a large area without PRL-secreting function and a focal area that secretes PRL, resulting in negative staining for PRL in these cases. Tumors of the two PRL-positive cases both showed focal expression of PRL (43, 45), which further confirmed this explanation.

Non-pituitary-derived PRL-secreting neoplasms tend to be insensitive to DA treatment, which may be due to the fact that these extracranial lesions may lack dopamine D2 receptors. Surprisingly, the PRL level in six cases of the 11 cases who received DAs as initial treatment increased after treatment with DAs. Among them, three cases were diagnosed as leiomyoma. Studies have shown that dopamine can indirectly activate estradiol receptors (78); thus, DAs may have a similar function in promoting leiomyoma growth as estrogen (79). This may contribute to the increase in PRL level after DA therapy (9). For these PRL-secreting neoplasms, resection or traditional chemotherapy generates a good prognosis. However, remissions should be taken care of, especially for malignant tumors.

Our review provides insights into diagnosing hyperprolactinemia with unclear causes (Fig. 4). In hyperprolactinemic patients with no evidence of pituitary masses on MRI, ectopic PRL secretion should be considered as a potential cause, alongside other known etiologies including physiological states (e.g., pregnancy, stress), primary hypothyroidism, impaired PRL clearance due to hepatic or renal dysfunction, and drug-induced hyperprolactinemia. For patients suspected of having ectopic PRL secretion, further diagnosis should be based on whether there are intracranial masses. A mass located at the sphenoid sinus, clivus, suprasellar region, or nasopharynx is likely a PRL-secreting EPA. For those without intracranial abnormalities, non-pituitary-derived PRL-secreting neoplasms should be highly suspected. In premenopausal women, pelvic imaging is recommended to evaluate the uterus and ovaries for potential PRL-secreting tumors. In adolescents, particularly females, the possibility of hematologic malignancies, especially AML and ALL, should be taken into account. In addition, in patients presenting with a pituitary mass but showing no response to DA therapy, there is a possibility that the mass is a non-functioning pituitary adenoma instead of a prolactinoma, and thus, further evaluation is required to assess the presence of an ectopic PRL-secreting neoplasm.

Figure 4.

Diagnostic algorithm for hyperprolactinemia caused by extra-pituitary PRL secretion. When the serum PRL level is elevated and no signs of pituitary adenomas are found in MRI, after excluding physiological causes, pharmacologic causes, primary hypothyroidism, and complications affecting PRL clearance, ectopic secretion of PRL could be the possible cause of hyperprolactinemia. Besides, in DA-resistant patients with suspected prolactinoma, the presence of ectopic PRL secretion should also be considered.

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

This work is supported by the National High Level Hospital Clinical Research Funding, 2022-PUMCH-A-064, and National High Level Hospital Clinical Research Funding, 2022-PUMCH-B-016.