A total of 127 women with prolactinoma (102 with micro- and 25 with macroprolactinoma) were identified. Data on the female study population at the time of diagnosis of prolactinoma is reported in Table 1. At initial diagnosis, no significant differences in RBC parameters were identified between women with microprolactinoma and macroprolactinoma. Moreover, considering the entire female population, no significant correlation between prolactin levels and any RBC parameters was observed.

Hypogonadism was reported in 61 (60%) patients with microprolactinoma and 19 (76%) patients with macroprolactinoma, whereas deficiency of the other pituitary hormones was found only in a subset of patients with macroprolactinoma (Table 1). Serum iron, ferritin, transferrin, folate and vitamin B12 levels were available in 9% of the patients and are reported in Supplementary Table 1. The number of patients under hormone replacement therapy at the time of diagnosis is reported in Supplementary Table 2.

Data on 65 (51%) patients (48 micro- and 17 macroprolactinomas at initial diagnosis) with prolactin normalization under dopamine agonists (cabergoline, n = 57; bromocriptine, n = 6; quinagolide, n = 2) were available. Median time from diagnosis of prolactinoma to RBC parameter analysis after PRL normalization was 24 months (IQR 24). No difference in any RBC parameters was observed in patients with micro- and macroprolactinoma between the time of diagnosis and the time of prolactin normalization (Table 2). In line of normalization of PRL levels, hypogonadism significantly improved in 23 out 28 patients with microprolactinoma and in 8 out 11 patients with macroprolactinoma (Supplementary Table 3). In patients with macroprolactinoma, ACTH and TSH deficiency improved in 2 out 3 and 2 out 5 cases, respectively, at time of PRL normalization, whereas GH was still present in 2 patients (Supplementary Table 3).

To assess the impact of overt hypogonadism, overt hypothyroidism (primary or secondary), prolactin change from diagnosis, age, smoking, and hormone replacement therapies (glucocorticoids, thyroid hormones, estrogen) on RBC, HCT and Hb at PRL normalization, a multivariable regression was performed. Under dopamine agonists, untreated hypogonadism was observed in 5 (8%) women, and untreated/inadequately treated hypothyroidism in 6 patients (10%). The mean prolactin change from baseline was − 3.23 ± 8.28 µg/L, and the mean age was 40 ± 17 years. Smoking was reported by 5 (8%) women. One patient was on glucocorticoid replacement, 12 on thyroid hormone, and 3 on estrogen; none were on GH replacement. As shown in Table 3, age minimally but significantly positively influenced RBC and Hb increase.

The male population comprised 78 patients (23 with micro- and 55 with macroprolactinoma). Hormonal and hematological data of these patients at baseline is reported in Table 1.

Serum iron, ferritin, transferrin, folate and vitamin B12 levels were available in 13% of the patients and are reported in Supplementary Table 1. At initial diagnosis, patients with microprolactinoma showed higher HCT (median 42.8 [2.7] vs. 40.2 [4.0] %, p = 0.0122) and Hb (14.7 [1.2] g/dl vs. 14.0 [2.0], p = 0.0337) than those with macroprolactinoma (Table 1). Patients with microprolactinoma had higher testosterone levels (median 1.7 [1.3] vs. 1.2 [0.8] µg/l, p = 0.006) than patients with macroprolactinoma. Of note, hypogonadism was observed in 61% of patients with microprolactinoma and 78% with macroprolactinoma (p = 0.161). In patients with macroprolactinoma, ACTH and TSH deficiency was found in approximately one third of cases, while GH deficiency was confirmed in 11% of the affected patients (Table 1). The number of patients under hormone replacement therapy at the time of diagnosis is reported in Supplementary Table 2.

Data on 54 patients (69%) with normalized prolactin levels under dopamine agonists (cabergoline, n = 47; bromocriptine, n = 6; quinagolide, n = 1) were available, including 13 micro- and 41 macroprolactinomas. Median time from diagnosis of prolactinoma to RBC parameter analysis after PRL normalization was 12 (IQR 24) months. After PRL normalization, HCT and Hb levels significantly improved in men with microprolactinoma (median HCT 42.3 [1.8] vs. 44.0 [1.6] % and median Hb 14.5 [0.6] vs. 15.1 [0.7] g/dl, both p < 0.05) as well as in those with macroprolactinoma (median HCT 40.2 [3.7] vs. 43.9 [2.2] % and median Hb 13.9 [2.0] vs. 15.1 [1.5] g/dl, both p < 0.001) (Table 4). Furthermore, in men with macroprolactinoma a higher RBC count (4.7 [0.4] vs. 5.0 [0.6] n*106/µl, p < 0.0001) was identified after PRL normalization than at the time of diagnosis. Of note, 16 out 24 patients with macroprolactinoma recovered from anemia under dopamine agonists (Table 4). Hypogonadism improved after PRL normalization in 5 out 6 patients with microprolactinoma and 28 out 33 patients with macroprolactinoma (Supplementary Table 4). Furthermore, in patients with macroprolactinoma, TSH deficiency improved in 9 out 15 cases (p = 0.0228) and ACTH deficiency improved

in 7 out 16 cases after PRL normalization. No improvement in terms of GH deficiency was reported under dopamine agonists in patients with available data (Supplementary Table 4).

Overt hypogonadism and hypothyroidism (primary or secondary) were observed in 4 (7%) male patients after PRL normalization. The mean change in PRL levels from baseline was − 16.44 ± 26.29 µg/L. The mean age of the patients was 48 ± 15 years. Smoking was reported by 7 (13%) men. 9 patients were on glucocorticoid replacement, 3 on adequate thyroid hormone replacement and 2 on testosterone replacement therapy; none were on GH replacement. As shown in Table 5, presence of hypogonadism significantly negatively impacted RBC count, HCT and Hb increase from levels at diagnosis.

In total, 101 SF-36 QoL questionnaires were completed by 51 women and 34 men.

Among women, the SF-36 questionnaire was completed at the time of the initial diagnosis and/or after PRL normalization during medical therapy by 21 and 40 patients, respectively. As shown in Fig. 1, similar SF-36 scores were observed at initial diagnosis and after remission. Of note, 10 women completed the SF-36 questionnaire both at initial diagnosis and after dopamine-induced PRL normalization, with results indicating no significant differences in QoL scores between the two time points (Fig. 2). In these 10 patients, there were also no significant differences in terms of HCT and Hb at initial diagnosis and after PRL normalization (median HCT 39.6 [2.8] vs. 38.5 [2.9] %, p = 0.92; median Hb 13.6 [1.2] vs. 13.3 [0.8] g/dl, p = 0.57).

SF-36 scores at initial diagnosis and after prolactin normalization under dopamine agonists. Questionnaires were completed either at the time of the initial diagnosis (women, n = 21; man, n = 10) or after prolactin normalization (women, n = 40; man, n = 30). Results are reported as mean and 95%-confidence interval. No significant difference in any of the scores was found

Abbreviation: PRL, prolactin

SF-36 quality of life scores of 10 women who completed the questionnaire both at initial diagnosis and after prolactin normalization under dopamine agonists. No significant difference in any of the scores was observed

Abbreviation: PRL, prolactin

To assess the impact of overt hypogonadism and hypothyroidism (primary or secondary), PRL change from diagnosis, age, HCT and Hb changes from diagnosis on QoL scores after PRL normalization, a multivariable regression analysis was performed. Age was negatively impacting physical function (b= -0.493 [-0.907; -0.078], p = 0.021), emotional wellbeing (b= -0.389, [-0.737; -0.041], p = 0.029), (energy b= -0.420 [-0.833; -0.007] p = 0.047), general health (b= -0.767 [-1.197; -0.338], p < 0.0001). Of note, both HCT and Hb changes did not impact any QoL scores (Supplementary Table 5).

Regarding the male study population, each one SF-36 questionnaire was completed either at initial diagnosis and/or after prolactin normalization under dopamine agonists by 10 and 30 patients, respectively. Mirroring the findings in women, there were no significant differences in the SF-36 scores between men at initial diagnosis and after PRL normalization (Fig. 1). Six men completed the questionnaire both at initial diagnosis and after PRL normalization during medical therapy. In these 6 patients, a significant improvement in energy scores between diagnosis and remission was observed (50 vs. 60, p = 0.022) (Fig. 3). Importantly, in these subgroup of patients, an increase of median testosterone levels from 1.1 (0.6) µg/l at the time of diagnosis to 2.8 (3.3) µg/l after PRL normalization (p = 0.031), as well as an increase in HCT (from 40.9 [3.3] to 44.0 [2.0] %) and Hb (from 14.1 [1.4] to 15.2 [0.9] g/dl) was detected (both p < 0.05).

SF-36 quality of life scores of 6 men who completed the questionnaire both at initial diagnosis and after prolactin normalization under dopamine agonists. Over the time of treatment, significant differences in the energy score was observed

Abbreviation: PRL, prolactin

Persistency of hypogonadism was significantly impairing all the QoL scores, while difference in PRL from diagnosis (b = 0.508 [0.159; 0.858], p = 0.007) and age (b= -1.166, [-2.238; -0.94], p = 0.035) were impacting respectively physical function and physical health (Supplementary Table 6).