Therapeutic strategies and challenges in the management of craniospinal tumours in pregnancy: a 10-year retrospective tertiary centre study, systematic review, and proposal of treatment algorithms

The study aims to evaluate therapeutic strategies in the management of craniospinal tumours in pregnant patients and the factors that inuence the management along with their inuence on maternal & foetal outcomes. A retrospective single-centre cohort study was performed at a tertiary neurosurgical referral centre. Pregnant patients referred to neuro-oncology multidisciplinary meeting (MDM) with craniospinal tumour were included. Ten-year patient data were collected from hospital records and neuro-oncology MDM outcomes. A systematic review was performed on the available literature in PubMed as per PRISMA guidelines. Lateral/park Systematic review identied 26 eligible articles. Treatment algorithms are proposed addressing the therapeutic strategy for management of cranio-spinal tumours during pregnancy and the challenges for maternal and foetal outcomes were


Introduction
Craniospinal tumours presenting in pregnancy are rare despite primary intracranial tumours being the 5 th leading cause of cancer-related death in women of age 20-39 years [1]. Primary malignant brain tumours in pregnant women occur in 26-150/million; meningioma comprises 13% of all CNS neoplasms identi ed during pregnancy with the incidence of around 7/million in the females in the 15-44 years age group; and vestibular schwannoma has an incidence of 17/10,000,000 in females aged 15-44 years [2,3,4]. Management of craniospinal tumours in pregnancy pose a considerable challenge to the neurosurgical community worldwide and requires a multidisciplinary team including neurosurgeons, neurologists, obstetrician, neuroradiologists and foetal medicine experts. Robust evidence-based guidelines are the current unmet need. This study aims to evaluate therapeutic strategies in the management of craniospinal tumours in pregnant patients and evaluate the factors that in uence the management along with their consequence on maternal & foetal outcomes.

10-year Retrospective single centre Study
A retrospective single-centre cohort study was performed at a tertiary neurosurgical referral centre. Pregnant patients referred to our neuro-oncology multidisciplinary meeting (MDM) with craniospinal tumours were included. Ten-year patient data were collected from hospital records and neuro-oncology MDM outcomes between 2010-2020. All patients were discussed in neuro-oncology MDM. The inclusion criteria were pregnant patients with cranial and spinal tumours. Patients with craniospinal tumours operated prior to conception/pregnancy and craniospinal tumours diagnosed after delivery were excluded from the study.

Consent and Ethics Approval
All data collection were approved by the department of neurosurgery at the hospital. All data were collected retrospectively from patient records and thus did not require ethics committee approval. Patient consent for data collection were not required as it was retrospective, and all the images were anonymised.

Systematic Review
Systematic review with PRISMA guidelines was performed on data collected from PubMed ( Figure 1). The systematic review has been registered in PROSPERO register with registration number [BLINDED FOR REVIEW]. Articles from 1950-2021 were searched on PubMed. The search strategy included MeSH items: ((Pregnancy OR Pregnant) AND (Glioma OR brain Tumour OR Meningioma OR Astrocytoma OR spinal tumour)). Two reviewers individually screened published papers based on inclusion and exclusion criteria and screened the title and abstract. Discrepancies was resolved by mutual consent obtained from Patients who required treatment during pregnancy but with pre-term delivery 6/33 patients underwent surgical management during pregnancy. There was post-neurosurgical pre-term delivery of the foetus in 5/33 patients through c-section managed by the obstetric/ neonatal team. No infant deaths were reported within the rst 30 days of delivery. In this group, 3/6 patients had high grade lesions (Glioblastoma WHO Grade IV.). 1/6 patient underwent endoscopic third ventriculostomy for obstructive hydrocephalus secondary to posterior fossa pilocytic astrocytoma Grade I and the tumour was later resected post-delivery. 1/6 patient had sellar apoplexy secondary to prolactinoma requiring transnasal transsphenoidal approach for decompression of the optic apparatus. 1/6 patient (metastatic carcinoma from the lung) required termination of pregnancy prior to craniotomy and proceeded to adjuvant treatment. Decision was made with MDT discussion with a patient centred approach.

Systematic Review
Systematic review identi ed 26 eligible articles which were ltered through the inclusion and exclusion criteria as per PRISMA guidelines. Among these, 4 studies contained algorithms for trimester-based management of pregnant patients with brain tumour (general), glioma, meningioma, and skull base tumours but no article was identi ed that demonstrated an algorithm for management of spinal tumours.
Glioma algorithm proposed by Peteers S et al was the only algorithm in the literature to address management of glioma diagnosed during pregnancy, antiepileptics management, post-delivery management and management of glioma diagnosed prior to pregnancy. They strati ed their strategy based on the de nition of stable/low grade glioma and unstable/progressive/high-grade glioma. The relevant articles proposing algorithms were summarised on to a tabulated form (Table 2). Articles discussing management of craniospinal tumours with their conclusion and recommendations were tabulated in Table 3.

Algorithms
We propose two algorithms addressing the therapeutic strategy for management of cranial ( Figure 3) and spinal tumours (Figure 4) in the rst, second and third trimesters of pregnancy. These proposed algorithms address important factors in the management of these complex scenarios, which have not been described in the literature in a single ow diagram before, including multidisciplinary discussion, surgical management, adjuvant treatment/radiotherapy, antiepileptic and steroids management, counselling regarding termination of pregnancy, progression of pregnancy with monitoring, decision making for planned caesarean section or supervised vaginal delivery.

Challenges
The challenges and factors that in uenced the maternal and foetal outcomes during management of these craniospinal tumours in pregnancy were identi ed from our data and the systematic literature review. These have been tabulated in Table 4.

Discussion
This single-centre retrospective study provided an overview of the strategies available for the management of different subtypes of craniospinal tumours presenting during pregnancy, depending on their presentation, clinical and imaging ndings. The range of presentations varied from incidental ndings on imaging such as arachnoid cyst/pineal cyst/meningioma to extreme emergencies such as a patient in coma with the imaging suggestive of a contrast enhancing high grade tumour such as a glioblastoma. All patients underwent referral to the neuro-oncology MDT and consensus was obtained for management which was then recommended to the referring physician/oncologist. Patients considered as candidates for surgical management, other than those requiring emergency admission, were advised on AEDs, steroids and were seen in the pre-assessment clinic where imaging ndings and surgical management options were discussed with the patients and family. This was then followed by the patient being offered a date for surgery and anaesthetic assessment with routine pre-operative investigations including haematological pro le, lower limb dopplers and further MRI imaging as deemed necessary. Noncontrast imaging was obtained in all patients requiring surgery whilst being pregnant.
In the absence of prospective large data sets, our experience combined with the information gathered from the literature, constitute useful guidelines in the management of these patients.

Cranial Tumours in Pregnancy
Studies have demonstrated that tailored individualised case-based protocol for the treatment of brain tumours during pregnancy can allow term delivery with even the possibility of safe vaginal delivery of the foetus [2,5,6,7,8].

Glioma and Pregnancy
Pregnancy does not seem to in uence the clinical course of glioma patients nor impact the delivered children's health [9]. However, a population-based case control study demonstrated a decreased glioma risk associated with ever-pregnancy compared with never-pregnancy and that breast feeding among parous women increased glioma risk, concluding that reproductive hormones may in uence the occurrence of glioma [5]. Findings from observational studies show that there is no known effect of pregnancy on survival in low-grade glioma patients although pregnancy can provoke clinical deterioration and tumour growth on MRI. In symptomatically stable women at term, there is no bene t of caesarean section over vaginal delivery with respect to adverse events in mother or child. Unanswered questions in literature include when pregnancy should be discouraged, what the best monitoring schedule is for both mother and the foetus, and if/ how chemo-and radiation therapy can be safely administered during pregnancy [4]. Ronning et al demonstrated that pregnancy did not seem to in uence the overall survival in patients with low grade glioma who became pregnant [7]. However, the recommendation was that the prospective mothers must also be thoroughly informed about the reduced expected overall survival associated with low grade glioma, regardless of the decision to become pregnant [7].

High grade Glioma
Glioblastoma although rare can present as a neurological emergency requiring surgery during pregnancy.
A case report by Mackenzie et al described a patient undergoing caesarean section for delivery at 36+2 weeks followed by craniotomy for resection of GBM and then radiotherapy and chemotherapy postpartum [10]. A literature review in 2012 discussed the effects of radiation during CT imaging, chemotherapy during second and third trimester, timing of surgery and delivery of pregnancy, postoperative management of pregnancy in neurosurgical patients, anaesthetic considerations, use of anticonvulsants and steroids [11]. The paper suggested that adjuvant therapy (chemo-, radiotherapy) in pregnancy was likely to cause deleterious effects ranging from intrauterine growth retardation, low birth weight, premature delivery and still birth [11]. Detailed discussion with the patient regarding potential adverse outcome and uncertainty was recommended to be carried out in conjunction with the oncologist.
There was an increased risk of miscarriage associated with surgery in the rst trimester and some evidence for increased risk of birth defects [11]. The proposal therefore was that surgery should be delayed until second trimester whenever possible [11]. The study also highlighted the co-ordination within the multidisciplinary team and suggested individualised approach for each patient during the tumour and pregnancy management [11]. An early population-based assessment demonstrated that the observed to expected ratios were substantially reduced for brain malignancy and meningioma and thus, rejected the view that intracranial neoplasms presented more often during pregnancy [2]. This led to the current recommendations on individualised MDT based approach to different types of cranial tumours in pregnancy [2].

Meningioma and Pregnancy
Risk of meningiomas in women <50 years was increased with an increase in number of pregnancies as compared to nulliparous women [5], concluding the in uence on the occurrence of meningiomas by reproductive hormones. Reports of shrinkage or disappearances of meningiomas after pregnancy suggest that repeat imaging should be performed during pre-operative assessment. Acceleration of growth rates during increased levels of serum oestrogen (O) and progesterone (P) in pregnancy have been demonstrated with overexpression of O/P receptors [12]. Large symptomatic meningiomas require surgical resection regardless of the status of pregnancy; preferred timing of intervention is, however, a matter of debate. Surgical resection during pregnancy may be associated with maternal and foetal morbidities/mortalities but overall outcomes are similar to post-partum resections for the mother. Patients with pregnancy >27 weeks are likely to undergo foetal delivery rst followed by planned tumour resection.
Maternal health and well-being are paramount in the decision-making process but aiming for term delivery has been shown to avoid prematurity associated complications in the neonate [13]. A systematic review demonstrated similar incidence of meningiomas in pregnant and non-pregnant women; symptoms are up is likely during pregnancy due to uid retention, engorgement of vessels, and presence of sex hormone receptors in the tumour cells. Foetal monitoring by biophysical pro le (ultrasound, amniotic uid volume, foetal breathing movements) and cardiotocography (CTG) is recommended [8]. Surgery is the preferred choice of treatment for meningiomas, but individualised approach is recommended. Foetal biophysical pro le scoring system, Umbilical arterial Doppler assessment, neuro-anaesthesia, and microsurgical techniques permit safe neurosurgical management of meningiomas and other brain tumours during pregnancy. Emergency surgical intervention should be reserved for patients with fast tumour growth rates, active hydrocephalus requiring shunting, signs of impending brain herniation, or progressive neurological de cits [8].

Other Cranial tumours and Pregnancy
A case series published in 2019 reported sellar region lesions that underwent surgical resection in four patients, diagnosed during pregnancy, presenting with deterioration in vision. All patients continued their pregnancy to term post operatively. They concluded that MDM guided decision-making process, leading to surgical resection of sellar region lesions in pregnant women with severe impairment, was the key to the positive outcome of pregnancy [14]. Priddy et al have suggested an algorithm for management of symptomatic benign skull base tumours and have highlighted red ag symptoms of paresis, complete visual loss, brain herniation, complete III nerve palsy and coma as needing emergency surgery. Trimester based management was highlighted in this series and discussion regarding continuation vs termination of pregnancy was considered for patients diagnosed in the rst trimester of pregnancy. Postponement of surgery as far as safely possible was suggested for patients in second and 3 rd trimester [15].

Spinal tumours and Pregnancy
The These factors will play a crucial role in the decision-making process regarding whether to operate or not to operate and, when to offer surgery, for these high risk, complex patients.

Limitations
Our data was based on a retrospective series with modest numbers. Further large-sample size prospective studies would provide robust evidence regarding management of pregnant patients with craniospinal    Algorithm for therapeutic strategy for spinal tumours in pregnancy