Abstract
Background: Data on female fertility, pregnancy, and outcome of offspring after childhood-onset craniopharyngioma (CP) are rare. Study Design: Observational study on pregnancy rate and offspring outcome in female CP patients recruited in KRANIOPHARYNGEOM 2000/2007 since 2000. Results: A total of 451 CP patients (223 female) have been recruited, and 269 (133 female) were postpubertal at study. Six of 133 female CP patients (4.5%) with a median age of 14.9 years at CP diagnosis had 9 pregnancies, giving birth to 10 newborns. Three patients achieved complete surgical resections. No patient underwent postoperative irradiation. Five natural pregnancies occurred in 3 CP patients without pituitary deficiencies. Four pregnancies in 3 CP patients with hypopituitarism were achieved under assisted reproductive techniques (ART) (median 4.5 cycles, range: 3–6 cycles). Median maternal age at pregnancy was 30 years (range: 22–41 years). Six babies (60%) were delivered by caesarean section. Median gestational age at delivery was 38 weeks (range: 34–43 weeks); median birth weight was 2,920 g (range: 2,270–3,520 g), the rate of preterm delivery was 33%. Enlargements of CP cysts occurred in 2 women during pregnancy. Other complications during pregnancy, delivery, and postnatal period were not observed. Conclusions: Pregnancies after CP are rare and were only achieved after ART in patients with hypopituitarism. Close monitoring by an experienced reproductive physician is necessary. Due to a potentially increased risk for cystic enlargement, clinical, ophthalmological, and MRI monitoring are recommended in patients at risk. Severe perinatal complications, birth defects, and postnatal morbidity of mothers and offspring were not observed.
Introduction
Craniopharyngioma (CP) is a rare benign embryonal malformation derived from ectoblastic remnants of Rathke’s pouch [1‒4]. Treatment options for CP include surgical resection with or without consecutive radiation therapy. Surgical interventions should be performed with the goal of preserving the function of vital neurovascular, optic, and hypothalamic structures [5]. Long-term morbidity of CP patients is caused by visual, neuroendocrine, and neurobehavioral sequelae. Social relations and partnerships are frequently impaired by neurobehavioral disturbances due to hypothalamic dysregulation [6‒8]. Female CP patients with hypopituitarism frequently suffer from amenorrhea and infertility. Hypopituitarism is associated with pregnancy complications, such as abortion, anemia, pregnancy-induced hypertension, placental abruption, premature birth, postpartum hemorrhage [9], transverse line, and small-for-gestational-age (SGA) newborns [10]. Pregnancies in childhood-onset CP patients are uncommon and rarely reported in the literature.
Patients and Methods
This observational study included 223 female patients diagnosed with childhood-onset, adamantinomatous CP (aCP) who have been recruited in the multinational trials KRANIOPHARYNGEOM 2000/2007 since the year 2000. One hundred and thirty-three of 223 female CP patients were postpubertal at the time of study. We contacted all 133 female postpubertal patients recruited in KRANIOPHARYNGEOM 2000/2007 and asked by questionnaire and/or telephone interview about previous pregnancies including abortions. The pregnancy rate, ovulation induction, and obstetric and neonatal outcomes were investigated in this study.
The histological diagnosis of aCP was confirmed by a neuropathological reference panel in all cases. Pre- and postoperative imaging as well as follow-up imaging were reference-assessed by a neuroradiologist, blinded to clinical data. Neuroradiological assessment of CP location, degree of surgical resection, preoperative hypothalamic involvement (HI), and surgical hypothalamic lesions (HL) was performed on pre- and postoperative axial, coronal, and sagittal magnetic resonance imaging (MRI) as previously described [11, 12].
Statistical Methods
Descriptive data are provided. Patients’ data are displayed as median (range) or frequency (percent).
Results
From the year 2000 onward, 451 patients (223 female/228 male) with childhood-onset aCP were recruited in the German trials KRANIOPHARYNGEOM 2000/2007 with a high grade of recruiting completeness. A total of 269 of these CP patients (133 female/136 male) were postpubertal at the time of study. The median age of 133 postpubertal females with aCP at the time of study was 24 years, ranging from 18 to 38 years (1st quartile: 21 years; 3rd quartile: 27 years). Six out of these 133 adult female CP patients (4.5%) reported 9 pregnancies resulting in the birth of 10 healthy newborns.
Pregnancy Cases
Case 1
An 8-year-old girl presented with precocious puberty leading to the diagnosis of aCP (tumor size: 2.04 cm2; solid tumor without cystic compartments), which extended to the suprasellar region with HI (grade 1). She underwent subtotal CP resection without postoperative irradiation. Subsequently, she developed isolated central diabetes insipidus (DI). As a result, she was treated with desmopressin as a treatment for central DI and a gonadotrophin-releasing hormone (GnRH) agonist for the treatment of precocious puberty. The desmopressin and GnRH agonist medication were finished at the age of 15 years without the necessity of hormonal substitution therapy later on. At the age of 22 years, she became naturally pregnant. A healthy 38-week male newborn was delivered by normal delivery weighing 2,880 g (25th centile [13, 14]). The mother breastfed her baby for 9 months.
Case 2
A 15-year-old girl presented with a reduced growth rate leading to the diagnosis of aCP (tumor size: 7.2 cm2), which could be completely resected. After surgery, she developed central hypothyroidism, adrenal insufficiency, and central DI. She received replacement therapy with L-thyroxine, hydrocortisone, and desmopressin. As she had the desire to have a child, she underwent six and three cycles of ovulation induction with human chorionic gonadotrophin (HCG) for her first and second pregnancy, consecutively. She conceived through the ovulation induction, and 2 healthy offspring were delivered by normal delivery at the gestational age of 38 and 38+6 weeks, respectively, weighing 2,660 g (15th centile [14]) and 2,960 g (25th centile [14]), respectively. The mother successfully breastfed both babies.
Case 3
A 21-year-old female presented with headaches. Subsequently, she was diagnosed with aCP, and her tumor was completely resected. She developed central DI, central hypothyroidism, central adrenal insufficiency, and gonadotrophin deficiency. Consequently, she required desmopressin, L-thyroxine, and hydrocortisone for hormonal substitution. She was diagnosed with type 2 diabetes mellitus when she was 24 years old. At the age of 37 years, she underwent fertility treatment and conceived after four cycles of ovulation induction with human menopausal gonadotrophin (HMG). Her body mass index at pregnancy was 41.4 kg/m2. Following a cesarean section, she delivered a female newborn weighing 3,520 g (91st centile [14]) at the gestational age of 37+2 weeks. The mother was unable to breastfeed.
Case 4
A 15-year-old female was diagnosed with aCP due to a history of headaches. MRI of the brain revealed a cystic lesion at diagnosis. Her pituitary function was preserved after subtotal tumor resection without postoperative irradiation. Therefore, she did not require hormonal replacement. She became pregnant naturally two times, when she was 30 and 36 years old, respectively. Her pregnancies were otherwise uncomplicated apart from a significant enlargement of a pituitary cystic lesion during her first pregnancy (Fig. 1a). The cystic CP was completely resected during the first pregnancy (Fig. 1b). After complete resection, no cystic enlargement/relapse occurred during the second pregnancy. The patient underwent delivery by means of cesarean section for both pregnancies at the gestational ages of 40 and 41 weeks, respectively. The birth weight of the newborns was 2,270 g (50th centile [14]) and 2,350 g (50th centile [14]), consecutively. Both infants were bottle fed.
Case 5
A 10-year-old female presented with headaches. MRI revealed a solid CP without cystic compartments. An incomplete tumor resection was achieved without postoperative irradiation. Thereafter, she developed central adrenal insufficiency, central hypothyroidism, central DI, prolactin deficiency, and hypogonadotrophic hypogonadism. She was on replacement therapy with hydrocortisone, L-thyroxine, desmopressin, and sex steroids. At the age of 34 years, she conceived dizygotic twins after five courses of HCG therapy. At 34 weeks of gestation, she developed preterm labor. Following a cesarean section, she delivered twins, a female infant weighing 2,270 g (50th centile [14]) and male infant weighing 2,350 g (50th centile [14]). The female newborn was diagnosed with a choroid plexus cyst. Artificial milk feeding was required due to lactation failure.
Case 6
A 26-year-old female complained of headaches and fatigue for 1 year. She had a normal past medical history and denied having any symptoms of endocrinopathy. The first MRI of the brain was conducted during her first pregnancy and demonstrated a partly cystic, sellar mass with typical imaging characteristics of CP. Her endocrine evaluation was normal. Thus, she did not receive any hormonal replacement therapy. No surgical intervention was performed. Apart from bleeding during early pregnancy, her first pregnancy was uneventful. At 43 weeks of gestation, a healthy male newborn was born with a birth weight of 3,800 g (25–50th centile [14]). Two years later, during her second natural pregnancy, it was discovered that her cystic lesion had significantly increased in size. Therefore, she underwent total tumor removal during pregnancy. The histopathological analysis showed the tumor to be aCP. She subsequently developed panhypopituitarism and required hormonal supplementation. She gave birth to a healthy late preterm baby weighing 2,800 g (50–75th centile [14]) by cesarean section.
The most common symptoms leading to CP diagnosis in these 6 patients were headaches, reduced growth rate, and precocious puberty. Median age at aCP diagnosis was 14.9 years, ranging from 7.8 to 29 years. Two of 6 patients presented with HI at initial CP diagnosis. Complete surgical resections could be achieved in 3 patients; one of them (case 3) presenting with presurgical HI. Surgical HL were observed in 2 patients (case 1: grade 1; case 3: grade 2) [11, 12]. Two of 6 patients presented with mainly cystic lesions at the time of CP diagnosis (cases 4 and 6). Pregnancies could be successfully achieved also in CP patients with initial presurgical HI and surgical HL (cases 1 and 3). No patient underwent irradiation. After surgery, 2 patients (cases 1 and 4) had normal pituitary function, while 4 others had multiple pituitary deficiencies requiring hormonal supplementation. Patient characteristics are shown in Table 1.
Median age at pregnancy in 6 CP patients was 30 years, ranging from 22 to 41 years. Natural pregnancies occurred in 5 pregnancies of 3 CP patients with normal pituitary function. All patients with pituitary insufficiency required assisted reproductive techniques (ART) in order to conceive (after a median of 4.5 cycles of HCG or HMG, range: 3–6 cycles).
During two pregnancies after initial complete CP resection (cases 2 and 3), no relapses were observed. Furthermore, no progression of residual solid CP (without cystic compartments) occurred during pregnancy after initial incomplete CP resection (cases 1 and 5). In the 2 patients (cases 4 and 6) with cystic CP detectable on MRI who became pregnant, cystic enlargement was detectable during pregnancy. In case 4, cystic remnants of the CP after initial incomplete resection progressed. In case 6, a cystic CP was diagnosed by MRI during first pregnancy and progressed during second pregnancy leading to complete resection. Accordingly, progressions were detectable exclusively for cystic CP compartments during pregnancy. Other severe complications during pregnancy, delivery, and postnatal period such as abortion, pregnancy inducing hypertension, placental abruption, and postpartum hemorrhage were not observed.
There were 10 live births with no miscarriages or fetal deaths. Four babies, including one set of twins were born at gestational age less than 38 weeks. Median birth weight was 2,920 g (on 25th centile [14]), ranging from 2,270 to 3,800 g. Only one baby was observed to be SGA with birth weight below the 10th centile [14] (case 4). Six of 10 babies were delivered by caesarean section. There were no reports of congenital abnormalities, although one baby was diagnosed with a choroid plexus cyst. Three patients were unable to breastfeed. Details of the 9 pregnancies, 8 singletons and one set of twins, are shown in Tables 2 and 3.
Discussion
There are two subtypes of CP: aCP and papillary CP (pCP) [15, 16]. The BRAF V600E mutation is characteristic of pCP, while most of aCP harbor a mutation of the β-catenin gene [15, 17]. ACP has a high incidence in children and adolescents [15]. Recurrence in aCP is associated with increased vascular endothelial growth factor (VEGF) and fibroblast growth factor expression [17]. The aim of CP treatment is to maximize disease control while minimizing patient morbidity [18]. Between 2001 and 2019, only 21% of CP patients recruited in KRANIOPHARYNGEOM 2000/2007 underwent gross total resection, while incomplete resection was achieved in 65% of CP patients, followed by irradiation in 40% of CP patients.
Long-term outcome and quality of survival after CP are frequently impaired due to factors which also have direct implications for social and emotional behavior [6], partnership relations and bonding [19‒22], and affectivity and neuropsychological balance. Sterkenburg et al.[20] reported on lower rates of partnership in CP with HI when compared to patients without HI and lower rates of partnerships and pregnancies in CP patients when compared with healthy siblings. This could explain the observed low pregnancy rate, as 38/32% of female CP patients included in our study presented with HI/damage, respectively. However, 2 patients (34%) with successful pregnancy were diagnosed with HI in our cohort (cases 1 and 3).
A recent study on changes of health-related quality of life during the first 3 years after surgery revealed that HI and surgical damage had a negative impact on social behavior towards friends whereas social behavior in the family was not influenced [22]. These observations might indicate that due to hypothalamic syndrome many CP patients stick to familial bonding, whereas social relations to friends and partnerships are less frequent and intense. Based on reported deficits in psychosocial abilities, neuropsychological sequelae and neuroendocrine secretion of oxytocin after childhood-onset CP [6‒8], low pregnancy rates in these patients are likely to also reflect at least in part deficits in terms of social behavior.
In the literature, pregnancies in CP patients are rarely reported. Seven previous studies describing pregnancy outcomes in CP patients reported on 9 women with 16 pregnancies resulting in 12 live births, as shown in Table 4. Most of the patients (67%) underwent a caesarean delivery at a median gestational age of 38 weeks with no congenital abnormalities reported. Natural pregnancy was reported in 2 pregnancies of one CP patient with normal pituitary function [23]. Similar to the literature, we observed 5 naturally achieved pregnancies in 3 patients with a normal pituitary function. Accordingly, we conclude that treatment strategies preserving pituitary function are best for fertility after CP. However, pituitary integrity is rarely achieved in CP resection due to anatomical location of the tumor. All of the CP patients with hypopituitarism underwent ART in order to conceive. With regard to fetal growth, a higher rate of SGA (36%) was observed in the reviewed literature compared to our study (10%).
Pregnancy is associated with physiological pituitary enlargement reversible within 6 months postpartum [24]. Thus, sellar/parasellar masses such as CP should be monitored carefully during pregnancy [25]. We reviewed the literature concerning cases of CP diagnosed primarily during pregnancy. The first case was reported by Fischer in 1935 [26]. Previous case reports on primary diagnosis of CP and/or CP tumor progression during pregnancy are listed in Table 5. Nine women were diagnosed with primary CP during pregnancy [26‒31]. Among these 9 women, there were 2 who presented with solid CP tumor progression during pregnancy [32, 33]. The CP tumor characteristics comprised 5 solid tumors, 2 cystic tumors, one mixed solid-cystic tumor, and one patient without available data. In these 10 pregnancies, MRI monitoring was performed during 3 pregnancies of 2 patients due to complaints about headaches and visual deterioration [32, 33]. The other 7 pregnancies had no MRI monitoring of CP during pregnancy. In the 3 pregnancies with MRI follow-ups, all MRIs of CP showed solid CP tumor enlargement exclusively, with no cystic progression observed. In our study, we observed 2 cystic CP tumor enlargements during 2 pregnancies. The only patient with solid residual CP after incomplete primary resection (case 5) experienced no solid tumor progression during pregnancy.
There are few molecular studies reporting on an increased expression of estrogen receptors mRNA in CP [34, 35]. Additionally, receptors for progesterone, leptin, insulin-like growth factor-1 (IGF-1), growth hormone (GH)-releasing hormone, and GH were detectable on CP surface [36]. The expression of estrogen and GH receptors positively correlated with tumor size and cell proliferation index [35]. A previous case report revealed an increase in CP size after receiving GH and estrogen substitution in one patient and estrogen substitution alone in 2 patients [37]. The in vitro study of Li et al.[38] indicated that GH and its endogenous mediator IGF-1 promote cell growth, while tamoxifen inhibits CP growth. In clinical trials, GH substitution therapy was not associated with reduced 3-year event-free survival rates in childhood-onset CP patients [39]. Yu and Melmed [40] demonstrated that the overexpression of pituitary tumor transforming gene (PTTG) plays a major role in pituitary oncogenesis. PTTG mRNA expression is regulated by estrogen; a high level of estrogen leads to increased PTTG expression. Increased pituitary PTTG mRNA expression was reported to be associated with an approximately 3.4-fold increase in pituitary fibroblast growth factor and an approximately 2.2-fold increase in VEGF expression, which are linked to CP tumor progression [41]. VEGF, also known as the vascular permeability factor, is a key regulator of increased vascular permeability and neovascularization of a variety of tumors [42]. In an animal model, Senger et al.[43] could show that VEGF promotes the accumulation of ascites fluid. Stockhammer et al.[42] reported on an association between the elevation of VEGF and the progression of brain tumor cysts. Fourteen patients with primary cystic brain tumors of diverse histology, including 3 patients with CP were investigated for VEGF concentrations in cyst fluid. Aspirated fluids of all tumor cysts revealed high VEGF levels. Moreover, several studies observed that cystic formation in cerebellar gliomas and hemangioblastomas was related to VEGF expression [44, 45], being associated with increased microvascular permeability and thus leading to extravasation of plasma proteins and fluid accumulation [46, 47]. We speculate that the physiological increase in VEGF levels during pregnancy could be a factor inducing cystic progression of CP during pregnancy. Due to estrogen normally being formed in the ovaries and also being produced by the placenta at elevated levels during pregnancy to help maintain a healthy pregnancy, we hypothesize that increased estrogen levels during pregnancy might play a role in the enlargement of cystic CP during pregnancy. However, further studies are required to confirm this estrogen effect in CP patients.
We analyzed the rate of pregnancies, gestational complications, and outcome of mothers and newborns in female patients after childhood-onset aCP. Due to the retrospective design and small cohort size, the results of our study are limited. Some observations and our discussion on pathophysiological mechanisms of cystic CP enlargement during pregnancy are speculative at this point. Pregnancies are rare events after CP, which on its own is a rare disease. However, due to the high degree of recruiting completeness in our Craniopharyngioma Registry, we conclude that the observed rate of pregnancies (4.5%) represents a valid and unique finding. However, several factors influencing pregnancy rates, such as male fertility factors and the patients’ wish to become pregnant, were not analyzed in our study.
Our recommendation to monitor cystic CP during pregnancy for early detection of cystic enlargement is based on our observations in few cases and needs to be confirmed by studies on larger cohorts. Beneficial maternal and newborn outcome after CP is an important finding for patients and treating physicians but also needs further analysis in larger groups. The recommendation to resect CP while preserving pituitary gland and stalk integrity for best perspectives on fertility and natural conception is not realistic. Due to the anatomical location, pituitary gland and stalk frequently cannot be separated from the tumor and frequently need to be sacrificed during tumor resection of CP.
We conclude that pregnancies after CP are rare. ART have led to increased pregnancy rates in CP patients with pituitary insufficiency. With regard to the deficiencies of the hypothalamic-pituitary axes, close monitoring and care by an experienced reproductive physician is a vital factor in the achievement of successful pregnancy and neonatal outcomes. The most common presentations in patients who were diagnosed with CP during pregnancy were visual deterioration and headache. Accordingly, we recommend careful history and clinical examinations including ophthalmological assessment for pregnant CP patients. Especially for those CP patients with cystic compartments of their CP, we recommend an MRI without contrast and limited to sellar and parasellar slices [48] at the beginning of the second trimester for baseline assessment. MRI of CP should be performed in all CP patients who experience visual deterioration and headaches during pregnancy respecting the guidelines for MRI during pregnancy [48]. Severe complications during pregnancy, delivery, and the postnatal period as well as any birth defects were not observed.
Acknowledgements
The authors want to thank all participating colleagues for recruiting patients in KRANIOPHARYNGEOM 2007, and the patients and their parents for participating in this study.
Statement of Ethics
All procedures performed in our study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The studies KRANIOPHARYNGEOM 2000 (Clinical trial registration number: NCT00258453) and KRANIOPHARYNGEOM 2007 (Clinical trial registration number: NCT01272622) were approved by the local standing committee on ethical practice of the Medizinische Fakultät, Julius-Maximilians-Universität Würzburg, Germany (140/99; 94/06, respectively), and written parental and/or patient consent was obtained in all cases.
Disclosure Statement
H.L.M. has received reimbursement of participation fees for scientific meetings and continuing medical education events from the following companies: Ferring, Lilly, Pfizer, Sandoz/Hexal, Novo Nordisk, Ipsen, and Merck Serono. He has received reimbursement of travel expenses from Ipsen and lecture honoraria from Pfizer. The other authors declare that they have no conflict of interest.
Funding Sources
This study was funded by a grant (H.L.M.; DKS2014.13) of the German Childhood Cancer Foundation, Bonn, Germany.
Author Contributions
P.S. researched the data and wrote the manuscript. S.B. collected the data by questionnaire or telephone interview, contributed to the analytical plan and discussion and reviewed/edited the manuscript. B.B. did neuroradiological assessment of all imaging. B.B. is the neuroradiologist who performs reference assessment of imaging in all patients recruited in KRANIOPHARYNGEOM 2000/2007. She prepared the imaging data and their presentation and reviewed/edited the manuscript. H.L.M. initiated and conducted the multicenter trials KRANIOPHARYNGEOM 2000 and KRANIOPHARYNGEOM 2007, contributed to the analytical plan and discussion, and reviewed/edited the manuscript.