Assessment of macroprolactinemia inpatients with prolactinoma

Purpose: Macroprolactin, the high-molecular mass prolactin isoform, is considered to be an inactive product with extrapituitary origin. Although macroprolactinemia is considered a benign condition, there is evidence of over-lapping clinical features among patients with hyperprolactinemia. Data on the prevalence of macroprolactinemia in prolactinomas is also quite limited. The aim of this study was to assess the prevalence of macroprolactinemia in our patients with prolactinoma. Methods: The study included patients with macroprolactinoma (n = 50) and microprolactinoma (n = 16). Prolactin level was measured with an electrochemiluminescent immunoassay, and macroprolactinemia was defined as the percentage of prolactin recovery < 40% after the polyethylene glycol precipitation. Results: Macroprolactinemia was not detected in our patients with prolactinoma (the percentage of PRL recovery range; 55%–96%). The mean percentage of prolactin recovery was similar in patients with macroprolactinoma and microprolactinoma (67.7% ± 8.0% and 70% ± 9.4%, respectively, p = 0.96). Conclusion: Macroprolactinemia is generally associated with negative findings on pituitary imaging. Although the monomeric prolactin is dominant, rarely macroprolactin may also be present in prolactinomas. We did not detect presence of macroprolactin in any of the patients and there was no statistically significant difference between micro-and macroprolactinomas in terms of prolactin recovery.


Introduction
Human prolactin (PRL) consists of three major circulating forms, termed as monomeric PRL (mPRL), big PRL (bPRL), and "big big" PRL (bbPRL) or macroprolactin (MPRL) [1]. The monomeric form has a molecular weight of 23 kDa and accounts for most of the total PRL immunoreactivity in the serum of both normal subjects and patients with hyperprolactinemia. The bPRL has a molecular weight of 48-56 kDa and accounts for 10%-15% of PRL immunoreactivity. MPRL is defined as PRL with a molecular mass of ≥150 kDa which is a large antigen-antibody complex consisting of monomeric PRL and anti-PRL autoantibodies [1,2]. MPRL molecules may also occur in different sizes and compositions. These include aggregates of PRL monomers with varying degrees of glycosylation and covalent or noncovalent bonding along with additional isoforms [3]. Normally, MPRL represents a negligibly small percentage of the total PRL amount but MPRL may be predominant in some serum specimens [4,5]. This phenomenon is called macroprolactinemia [6]. Current commercial PRL immunoassays may exhibit variable degrees of immunoreactivity with MPRL [7]. The prevelance of macroprolactinemia in general population was reported as 4% while the incidence of macroprolactinaemia in patients with hyperprolactinemia ranges from 10% to 40% when samples from different reference laboratories are assayed [7][8][9][10][11].
Both Current Endocrine Society and Pituitary Society Guidelines recommend MPRL measurement only in patients with asymptomatic hyperprolactinemia [12,13]. Due to low biological activity and slower clearance rate of MPRL, symptoms related with hyperprolactinemia are not present in patients with macroprolactinemia and most authors do not recommend treatment or long-term follow up. However, recently several studies showed the co-occurrence of macroprolactinemia after systematic screening for macroprolactinemia in patients with prolactinoma [14][15][16][17][18]. Therefore, we aimed to assess the prevalence and clinical significance of macroprolactinemia in our patients with prolactinoma by using the polyethylene glycol (PEG) precipitation method.

Materials and methods
This is a retrospective study. The patients who were being followed up with diagnosis of prolactinoma (n = 66) in the pituitary out-patient clinic of the Istanbul Medical Faculty and who were investigated for presence of macroprolactinemia with MPRL measurements were enrolled in the study. The diagnosis of prolactinoma was confirmed according to the typical clinical signs and symptoms and radiographic signs (pituitary adenoma confirmed by magnetic resonance imaging) and laboratory tests (high PRL levels in at least two different blood samples) [12]. Serum PRL levels were measured with an electrochemiluminescent immunoassay (ECLIA; Elecsys, Modular Analytics, Roche Diagnostics) with reference ranges of 4.04-15.2 ng/mL and 4.7-23.3 ng/mL for adult males and females, respectively. Presence of MPRL was assessed with PEG precipitation of the serum samples according to the procedure which has been previously described by Hattori et al. [19] and extensively validated by Olukago and Kane [20]. The procedure was as follows: 200 μL of serum was added to 200 μL of 25 g/dL PEG 6000 solution, after thorough mixing and centrifugation at 3000 rpm for 30 min, the supernatant was removed for analysis, and PRL assay was performed immediately. PEG precipitates PRL molecules with a molecular weight more than 100 kDa, therefore PRL in the supernatant is considered to be free of MPRL. The results of the precipitation test was compared with those obtained from the unprecipitated serum samples. Percentage of MPRL was calculated using the following formula: MPRL% = (PRL serum − PRL supernatant) × 100/PRL serum. Results of the PEG precipitation test were presented as the percentage of PRL recovery (free PRL) = 100% − MPRL%. Macroprolactinemia was considered present when recovery % was <40% [16,[20][21][22].
Statistical analyses were performed using SPSS version 21.0. Categorical variables were defined by frequency and percentage rate, and numeric variables with mean ± standard deviation (SD). In dual independent group comparisons, Student's t-test was used for normally distributed numeric variables and the Mann-Whitney U test was used for non-normally distributed data. Categorical variables were compared using the χ 2 test. Statistically significant results were defined with presence of p value <0.05.

Discussion
Macroprolactinemia is generally associated with negative findings on pituitary imaging and clinical presentation or bioactivity of MPRL are still debated [23]. Some studies have determined that serum samples containing high molecular mass PRL may exhibit lower, higher or similar biological activity compared with serum samples containing mPRL and may sometimes be associated with signs and symptoms of hyperprolactinemia [3,24]. In the literature, a small proportion of patients with macroprolactinemia have symptoms of galactorrhea (20%) or oligo/amenorrhea (45%), and MPRL is associated with pituitary adenomas only in 20%-27% [23,[25][26][27]. Furthermore, although the mPRL is dominant, MPRL may also be present in prolactinomas [14][15][16]28]. In the current literature dominance of MPRL was firstly described by Rogol and Rosen [28] in their study with prolactinoma patients. Then, Ohnami et al. [14] showed presence of tumor originated bbPRL in patients with PRL-secreting pituitary adenomas. In subsequent studies on this subject, generally the presence of prolactinoma have been evaluated in patients with macroprolactinemia. In a series of 106 patients with macroprolactinemia studied by Vallette-Kasic et al. [9], prolactinoma has been identified in only four patients and it was proven immunohistochemically.
There are several methods for detection of MPRL such as gel filtration chromatography (GFC), PEG precipitation method, ultrafiltration method, immunoadsorption with protein G-sepharose, protein A-sepharose or anti-hIgG-agarose [11,29,30]. The three forms of PRL are identified by GFC of serum or extracts of normal pituitary gland and pituitary tumors. MPRL identification by GFC is gold standart method but it is difficult, time-consuming, expensive, and not currently used in practice [8,31]. Whereas, PEG-precipitation test is a simple and inexpensive method that can easily be integrated into laboratory practice and this test is proposed by Current Endocrine Society Guideline [12]. Furthermore one current study has reported high sensitivity (up to 100%), but suboptimal specifity levels with this method [31]. PEG precipitation method is also known as the method revealing the most consistent results with GFC measurements [11,29,30]. However, anti-PRL antibody should be present in circulation in order to use this method. Mounier et al. [15] showed that none of the patients with prolactinoma had autoantibodies by the PEG precipitation method, including patients who had predominantly bbPRL in serum analysis performed with the GFC method. Despite the fact that they have demonstrated the existence of bbPRL in tumor exctracts by the GFC method. Similarly, we did not detect macroprolactinemia in any of our patients with the PEG precipitation method. However, recently, using this method, the frequency of macroprolactinemia in prolactinoma patients was reported to be 7.6% [16]. The frequency of macroprolactinemia is higher in patients with microadenoma, but association of macroprolactinemia with macroadenoma is a much rare situation [9,18,26,27]. We did not detect MPRL in both of our patients with micro-and macroprolactinoma and there was no statistically significant difference in terms of PRL recovery. It is also unclear whether pituitary lesions are microprolactinomas or nonfunctional microadenomas with macroprolactinemia. Two recent studies revealed the 10 year follow up data of patients with macroprolactinemia. Abnormal imaging findings were reported to be rare and progression was not detected in the presence of microadenoma, and regression of microadenoma was reported even in the absence of treatment. These results support presence of incidental nonfunctional microadenomas with macroprolactinemia rather than presence of microprolactinoma with macroprolactinemia [32,33]. In contrast to these, Elenkova et al. [16] evaluated 10 patients who had both prolactinoma and macroprolactinemia, four of these patients had macroprolactinoma and one of them had an invasive macroprolactinoma [17]. However, the presence of MPRL could not be clearly defined whether from tumoral or peripheral origin, due to the fact that PEG precipitation method was used only on the serum samples and not on the tumor extracts [16,17].
Consequently, our results suggest that prolactinoma may not be associated with macroprolactinemia. On the contrary to the recent publications implicating coexistence of prolactinoma and macroprolactinemia, we agree with the guidelines that suggest the assesment of MPRL levels merely in the evaluation of asymptomatic hyperprolactinemia. However, the limitations of our study are that the study is under power and further studies with larger patient series is needed in order to make stronger suggestion and also we used the PEG precipitation method in the investigation of macroprolactinemia, whereas the most qualified method in the assessment of macroprolactinemia is defined as the GFC method. Furthermore, especially as it has also been proven to be present immunohistochemically in the tumor exctracts, presence of macroprolactinemia should also be kept in mind in the evaluation of prolactinomas.