Bisecreting Pituitary Neuroendocrine Tumors: About 9 Cases at the Abass Ndao Hospital Center in Dakar ()
1. Introduction
The term pituitary adenoma is increasingly being replaced by the term pituitary neuroendocrine tumor (PitNET) [1]. Multi-secreting forms are tumors capable of producing two or more different hormones [2]-[4]. Two theories are currently being put forward [5]. The “multiple hit” theory favors the proliferation of different mutated cell types. The other is the transdifferentiation theory, in which cells already present in a pituitary adenoma may differentiate into another cell type. The prevalence of multisecreting PitNETs has been reported to range from 0.01% to 9% and 0.37% to 2.6% respectively in autopsy studies [6] and surgical studies [7].
Functional multisecreting PitNETs are manifested by clinical signs related to hypersecretion [8]. The two most common forms are acromegaly [8] and Cushing’s disease [5]. Asymptomatic neuroendocrine tumors (PitNETs) known as non-functional tumors have also been described. They are immunohistochemically positive for one or more hormones or transcription factors. These non-functional PitNETs represent 15% to 30% [9] of all PitNETs and are mainly represented by gonadotrophic PitNETs, i.e. 70% to 75% [10] [11].
The sensitivity of magnetic resonance imaging is 85% for the detection of microadenomas [12]. The challenge is then to detect contiguous tumors that appear to be unique, or a macroadenoma dwarfing a microadenoma. To refine the result, 3T MRI or MRI coupled with positron emission tomography (PET-MET/3T MRI) is still decisive [13].
Treatment will depend on the symptoms associated with hypersecreting hormones and any mass effects. When surgery is indicated, transphenoidal resection is the most common way. However, corticotropic and somatotropic PitNETs are more prone to recurrence [14]. The best indicator is Ki67, which correlates with recurrence [15]. In view of these advances in the management of these multisecreting pituitary neuroendocrine tumors and the lack of published data in Africa, and in Senegal particularly, a survey was carried out with the aim of describing the epidemiological, clinical and therapeutic aspects of multisecreting PitNETs.
2. Patients and Method
This was a multicenter, retrospective, descriptive and analytical study. The survey was conducted on files collected from 1st January 2008 to 31 December 2022 (i.e. over a period of 14 years) in the neurosurgery departments of the Fann and Principal hospitals in Dakar and the endocrinology department of the Abass Ndao hospital in Dakar. The diagnosis of pituitary neuroendocrine tumor was based on clinical, hormonal, and morphological criteria. Immunohistochemistry, the reference technique, is not yet available in Senegal. Any incomplete file was excluded from the study. The data required for the study were:
Epidemiological data: age, sex, terrain and antecedents, consultation time.
Clinical data: hypersecretory syndromes (Cushing, dysmorphic, thyrotropic, amenorrhoea-galactorrhoea), pituitary tumor syndrome (headache and visual acuity or visual field disorders). The visual field was assessed using the Goldmann.
Pituitary hormone data: baseline pituitary work-up including prolactinemia (norms: 3.46 - 19.4 ng/ml), ACTH (morning norms 5 - 60 ng/L; evening norms: <30 ng/L), midnight cortisolemia (norms: 37 - 194 ng/ml), 24-hour urinary free cortisol (norms: 28 - 138 nmol/24h), plasma GH and IGF1 (age-dependent norms), FSH/LH, testosterone (male norms: 8.7 - 34.7 nmol/l), estrogen (follicular phase: 21 - 251 pg/ml, ovulatory peak: 38 - 649 pg/ml, luteal phase: 21 - 312 pg/ml, menopause: <28 pg/ml), TSHus (norms: 0.35 - 4.94 µIU/ml) and free T4 (norms: 9 - 19 pmol/l). The dynamic tests consisted of a minute braking test with dexamethasone and an oral glucose tolerance test (OGTT) with GH.
Morphological data had enabled the PitNET to be identified using magnetic resonance imaging.
Management: this involved medical treatment and its modalities (frenative or substitutive therapy). Surgery was performed by transphenoidal way, guided by endoscopy in the neurosurgery units of Senegal.
Evolution was assessed based on the secretory syndrome, the pituitary tumor syndrome, the size of the tumor on the follow-up MRI, and the search for surgical complications: breach (leakage of cerebrospinal fluid), meningitis, transient diabetes insipidus, death, etc.
The data collected was entered into Excel 2016. The analysis was carried out using Excel 2016 and R version 4.1.3. For the descriptive analysis, the qualitative variables were described using frequency tables and bar charts. Quantitative variables were described by their position (mean, median) and dispersion (standard deviation, extremes) parameters. For bivariate analysis, the Student’s t-test was used for comparisons of means, and the Chi-square or Fisher’s t-test for comparisons of proportions. A link or difference is noted if the p is less than 0.05.
3. Results
3.1. Socio-Demographic Data
Of the 242 patients treated for PitNET, 09 were presented with bihormonal PitNET, i.e. a proportion of 3.71%. This proportion was 7.76% of the 116 cases of secretory PitNET. Two types of association were found: 08 tumors with GH + prolactin secretion and 01 PitNET with ACTH + prolactin secretion. The mean age of our patients was 41.7 ± 11 years with extremes of 25 and 59 years. The median age was 45 years. Our study included 5 (55.6%) males and 4 (44.4%) females. The average age of the women was over 45, and most of the men were under 45 (80%). The associated pathologies were diabetes mellitus and high blood pressure in 22.2% each.
3.2. Clinical Data
The mean time to consultation was 24.2 ± 12 months, with extremes of 6 and 36 months. The median was 24 months. The consultation time was longer in women, with an average of 25.5 months, whereas in men the average was 23.2 months. Clinically, gonadotropic insufficiency was found in all patients (100%). Dysmorphic syndrome was found in 6 patients (66.7%) and tumor syndrome in 7 patients (77.8%). Ophthalmological evaluation revealed a decrease in visual acuity in 66.6% of patients and left-sided blindness in one patient. We found no cases of quadranopsia or hemianopsia. Table 1 summarizes the epidemiological and clinical characteristics of patients on admission.
Table 1. Initial clinical characteristics of patients with multisecreting tumors.
Parameters |
Results |
Epidemiological data |
|
Population size |
9 (100%) |
|
Mean age |
41.7 ± 11 years |
|
Average consultation delay |
24.2 ± 12 months |
|
Sex ratio |
1.25 |
|
Diabetes mellitus |
22.2% |
|
Hypertension |
22.2% |
Clinical data |
|
Tumor syndrome |
7 (77.8%) |
|
Amenorrhea-galactorrhea syndrome |
9 (100%) |
|
Dysmorphic syndrome |
6 (66.7%) |
|
Cushing’s syndrome |
1 (11.1%) |
|
Diplopia |
1 (11.1%) |
|
Reduced visual acuity |
6 (66.6%) |
|
Left-sided blindness |
1 (11.1%) |
3.3. Biological Data
Biological analysis revealed hyperprolactinaemia > 500 ng/ml in all patients, unrelated to pituitary disconnection. This hyperprolactinemia was associated with ACTH-dependent hypercortisolism in one patient and with elevated GH and IGF1 (Insulin Growth Factor) levels in 8 patients (88.9%).
3.4. Morphological Data
Morphological exploration was performed using magnetic resonance imaging (MRI) in all patients. The results were in favor of pituitary macroadenoma in all patients, with invasion noted in 02 patients. Table 2 summarizes the various paraclinical characteristics of patients on admission.
Table 2. Initial paraclinical characteristics of patients.
Parameters |
Results |
Hormone data |
9 (100%) |
|
Mixed secretion (ACTH + Prolactin) |
1 (11.1%) |
|
Mixed secretion (GH + Prolactin) |
8 (88.9%) |
|
Prolactin levels (average and extremes) (n = 9) |
2438 ng/ml (510 - 3843) |
|
GH levels (average and extremes) (n = 8) |
112 ng/ml (34 - 150) |
|
IGF1 levels (average and extremes) (n = 8) |
618.5 ng/ml (579 - 730) |
|
ACTH levels (n = 1) |
97.3 ng/l |
Morphological data (MRI) |
9 (100%) |
|
Macroadenoma |
9 (100%) |
|
Tumor invasion |
2(22.2%) |
3.5. Therapeutic and Evolutionary Data
Of the 9 patients identified, 8 had received treatment with cabergoline (Dostinex®) 0.5 mg/week, i.e. a proportion of 88.9%. These were patients with mixed hypersecretion of prolactinemia and GH. The only patient with hyperprolactinaemia associated with Cushing’s disease did not receive drug treatment for the hormonal disorder.
Transsphenoidal surgery was performed in all patients. Complications were noted after surgery. These were mainly transient diabetes insipidus, followed by cerebrospinal fluid leaks. We noted one case of death following a post-operative haematoma in a patient with mixed secretion of GH and prolactin. Post-operatively, hormonal monitoring in 5 patients showed remission. Imaging revealed incomplete excision in 2 of the 5 patients who were able to undergo MRI. Ophthalmological symptoms improved after surgery in 7 patients. Table 3 summarizes the evolutionary data after surgical management.
3.6. Analytical Study (Table 4)
We compared 3 different types of PitNETs in their epidemiological, clinical, morphological and surgical aspects. Of all 242 patients followed up for PitNET, we had more non-functioning PitNETs (52.1%), (compared with 44.2% of monosecreting
Table 3. Post-operative evolutionary data.
Evolution data |
Population size (frequency) |
Post-operative complications |
|
|
Transient diabetes insipidus (n = 9) |
3 (33.3%) |
|
Cerebrospinal fluid (CSF) leak (n = 9) |
1 (11.1%) |
|
Transient diabetes insipidus + CSF leakage (n = 9) |
3 (33.3%) |
|
Deaths (n = 9) |
1 (11.1%) |
Post-operative paraclinical evaluation |
|
|
Biological remission (n = 5) |
5 (100%) |
|
Complete excision (n = 5) |
3 (60%) |
|
Improved visual condition (n = 8) |
7 (87.5%) |
Table 4. Comparative Table of the different functional forms of PitNETs.
Parameters |
Bisecreting PitNETs |
Monosecreting PitNETs |
Nonfunctional PitNETs |
P-value |
Epidemiological data |
Population size (n = 242) |
9 |
107 |
126 |
|
Prevalence |
3.7 % |
44.2 % |
52.1 % |
|
Female |
4 (44.4%) |
79 (73.8%) |
44 (34.9%) |
<0.001 |
Average age (years) |
41.8 ± 11 |
36.3 ± 11 |
48.2 ± 14 |
<0.001 |
Clinical data |
Average consultation time (months) |
24.2 ± 12 |
27.0 ± 29 |
14.0 ± 17 |
<0.001 |
Tumor syndrome |
7 (77.8%) |
61 (57.0%) |
108 (85.7%) |
<0.001 |
Gonadal syndrome |
9 (100%) |
81 (75.7%) |
13 (10.3%) |
<0.001 |
Dysmorphic syndrome |
6 (66.7%) |
13 (12.1%) |
0 (0.00%) |
<0.001 |
Cushing’s syndrome |
1 (11.1%) |
14 (13.1%) |
0 (0.00%) |
<0.001 |
Morphological data |
Macroadenoma |
9 (100%) |
63 (58.9%) |
125 (99.2%) |
<0.001 |
Tumor invasion |
1 (11.1%) |
13 (12.1%) |
8 (6.35%) |
<0.001 |
Surgical and evolution data |
Pituitary surgery |
9 (100%) |
81 (75.7%) |
83 (67.5%) |
|
Cerebrospinal fluid leak |
4 (50.0%) |
4 (6.56%) |
10 (21.3%) |
0.003 |
Transient diabetes insipidus |
6 (75.0%) |
46 (75.4%) |
30 (63.8%) |
0.427 |
Deaths |
1 (12.5%) |
9 (14.8%) |
8 (17.0%) |
0.919 |
Meningitis |
0 (0.00%) |
0 (0.00%) |
1 (2.13%) |
0.474 |
Remission |
1 (20.0%) |
13 (15.9%) |
10 (13.0%) |
0.646 |
PitNETs and 3.7% of bisecreting PitNETs). Unfortunately, as mentioned above, we did not have immunohistochemistry to better type these PitNETs. We found more women and a younger population in the group of monosecreting PitNETs, with a mean age of 36.3 years. Consultation was earlier in non-functioning PitNETs, with an average delay of 14 months. Among the reasons for consultation, tumor syndrome was significantly more frequent in non-functioning PitNETs than in monosecreting PitNETs. However, when the PitNET was secretive, hypogonadism syndrome was more frequent and was more common in bisecreting PitNETs (100%). Morphologically, there was a lower proportion of macroadenomas in monosecreting PitNETs, but they were significantly more invasive comparing to other types. We did not find any statistically significant difference concerning the morphological data other than a greater frequency of cerebrospinal fluid leak when the operated PitNETs was bisecreting. We also noted that whatever the type of PitNETs, diabetes insipidus was the most frequent complication.
4. Discussion
4.1. Epidemiological Data
In our PitNETs management database, we found a slightly higher proportion of bisecreting PitNETs than in previous studies. [16]. The prevalence was 0.37% in the series by Kontogeorgeos et al. [16] with a cohort of three thousand patients. In the study by Magri et al. [17] the prevalence was 2.6% in 117 patients operated on for pituitary neuroendocrine tumors. In our series, the prevalence was underestimated due to a lack of immunohistochemical evaluation. Autopsy studies have reported a much higher prevalence, up to 34%. [18]. While some studies have reported cases of triple secreting PitNETs [8] [16], this was not the case in our study. The lack of immunohistochemistry could explain this incomplete diagnostic. There was no clear gender trend [7] [19]-[21]. Our patients were in their forties with a mean age of 41.7 ± 11 years. In the literature, the mean age was 46.5 years in the series by Ogando-Rivas et al. [21] 45.7 years in the series by Ho et al. [19].
4.2. Clinical Data
Hypertension and diabetes mellitus were found in almost a quarter of cases. These two pathologies, which are cardiovascular risk factors, had not been investigated in many other studies [19] [21]. These were some of the circumstances of discovery in GH (acromegaly) and ACTH (Cushing’s disease) secretions. Their secondary nature makes them curable pathologies, depending on how long they have existed. However, the evolution of these cardiovascular risk factors after management of the endocrine pathology was not evaluated in our study. In the literature, the most common endocrine syndromes, in multisecreting PitNETs, were acromegaly, followed by Cushing’s syndrome [7] [19] [21]. In our study, the amenorrhoea-galactorrhoea syndrome was more prevalent. This could explain the earlier consultation in younger subjects due to sexual problems. In the study by von Schöning et al. [20], the authors found multisecreting PitNETs with no clinical manifestations. Neurological signs were more common in our study. In the series by Ho et al. [19], neurological signs were present in 60% of patients. Among these neurological signs, ophthalmological signs such as visual field abnormalities and reduced visual acuity were the most frequent.
4.3. Morphological Data
We only had MRI available pre- and post-operatively. The other morphological means available for morphological exploration were PET-Methionine coupled to MRI intraoperatively and ultrasound intraoperatively in cases where localisation was difficult [7]. In the literature, discordance between imaging results and surgical findings has been reported. In most autopsy and operative studies, many bi-secreting PitNETs were microadenomas with mean sizes of less than 3 mm [5] [13]. We only had pituitary macroadenomas.
4.4. Therapeutic Data
Medical treatment mainly concerned the management of hyperprolactinaemia when associated [7] [8] [19] [21]. In addition to its inhibitory effect on pituitary hormone secretion, this combined treatment helps to reduce the volume of the PitNETs and therefore facilitates its removal. Dopaminergic agonists reduce tumor size by more than 50% in 80% - 90% of patients, and cabergoline is generally more effective and better tolerated than bromocriptine [22]. In acromegaly, in vivo evidence of the antiproliferative effects of somatostatin analogues has been demonstrated by a significantly lower Ki-67 labelling index in somatotrophic tumor tissue obtained from patients previously treated with octreotide [23]. In addition to this effect on the tumor itself, pre-surgical treatment helps to normalise GH levels, control glycaemia, hypertension and left ventricular hypertrophy. Cabergoline has a limited indication in acromegaly. However, these somatostatin analogues are not available in Senegal. In our study, there was no need for drug treatment of Cushing’s disease. Surgery was chosen as first-line treatment [21]. This suggests the low intensity of clinical manifestations in cases with associated Cushing’s disease. Surgery was performed via the transsphenoidal route, as in our study. In some cases, intraoperative imaging was necessary to clarify the location of the lesion and allow a more complete resection [5] [9].
4.5. Evolution Data
Numerous postoperative complications have been reported in previous studies [24]. Resection could be complete when the tumor could be located preoperatively and there was no invasion. Thanks to intraoperative exploration, resection rates could reach 96% [25]. In our study, despite the modest resources available, resection was complete in 60% of cases. We might hope to see an improvement in the success rate with the acquisition of a more advanced technical platform and greater awareness of the need for earlier management. Complications mainly resolved spontaneously, apart from one post-operative haematoma which resulted in death.
Limitations of our study
Our study brought together the largest number of cases in the sub-region. This provides a more accurate picture of the management in our region and within our limited resources. In our study we were unable to distinguish between contiguous bisecreting PitNETs and those that were clearly distinct. Despite the difficulties of accessing certain complementary examinations, the minimum exploration required before and after the operation were performed. An effort still needs to be made to make care available to all. Other points for improvement related to:
The absence of immunohistochemistry for the diagnosis of non-functional components and the demonstration of contiguous bisecreting PitNETs;
Lack of evaluation of multiplication indices such as Ki-67 and the search for metastasis;
The long-term follow-up was not included due to loss of follow-up in some patients.
5. Conclusion
The aim of this study was to describe the epidemiological, clinical and therapeutic aspects of multisecreting PitNETs. Despite the lack of immunohistochemistry, two types of association were found: ACTH-PRL and GH-PRL. All were macroadenomas, the majority with an associated tumor syndrome. The latter was significantly less frequent in bisecreting PitNETs compared with monosecreting and non-functional forms. However, there was no significant difference in the occurrence of post-operative complications between the 3 secretory forms of PitNETs.