Management of malignant gliomas during pregnancy

A case series


  • Deborah T. Blumenthal MD,

    Corresponding author
    1. Department of Oncology, Neuro-oncology Service, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
    2. Department of Neurosurgery, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
    3. Department of Medicine, Neuro-oncology Service, Huntsman Cancer Institute at the University of Utah, Salt Lake City, Utah
    • Neuro-oncology Service, Tel Aviv Medical Center, 6 Weizmann Street, Tel Aviv, Israel 64239
    Search for more papers by this author
    • The first and second authors contributed equally to this article.

    • Fax: (011) 972 36974789

  • Mary Grace H. Parreño MD,

    1. Department of Neurosurgery, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
    Search for more papers by this author
    • The first and second authors contributed equally to this article.

  • Julia Batten ANP,

    1. Department of Medicine, Neuro-oncology Service, Huntsman Cancer Institute at the University of Utah, Salt Lake City, Utah
    Search for more papers by this author
  • Marc C. Chamberlain MD

    1. Department of Neurology and Neurological Surgery, Division of Neuro-oncology, University of Washington, Seattle, Washington
    Search for more papers by this author



Limited data are available on the management of glioma in pregnant women. Therefore, the aim of the current article was to describe the outcome of women with malignant gliomas who were exposed to chemotherapy early in the gestation period of their pregnancies.


The authors presented a case series of 6 women with malignant gliomas who during glioma-directed treatment were discovered to have an unplanned pregnancy. All patients elected to discontinue chemotherapy and carry their pregnancy to term.


All women had uneventful pregnancies with no glioma-related complications. All women delivered healthy newborns without evidence of congenital malformations despite exposure to cytotoxic chemotherapy and anticonvulsant medications.


Management of malignant glioma during pregnancy is challenging; however, normal delivery and healthy live birth is possible. Cancer 2008. © 2008 American Cancer Society.

The subject of pregnancy in patients with malignant gliomas is multifaceted and presents several clinical challenges.1‒4 Because of the relative paucity of reported cases of gliomas diagnosed and treated during pregnancy, the literature comprises only case reports and small series.4‒9 The majority of reports address diagnosis and surgical management of glioma but not adjuvant medical treatment.1, 10, 11 Consequently, there is limited literature on radiation treatment of malignant gliomas during pregnancy, and there are no reports to our knowledge of pregnant glioma patients treated with chemotherapy.11, 12 Furthermore, prospective studies of adjuvant glioma therapies in pregnant patients are not feasible because of ethical considerations. Therefore, there are no guidelines for the adjuvant administration of chemotherapy or radiotherapy or the ideal timing of neurosurgical or obstetrical intervention in pregnant women. The difficulty in establishing guidelines for therapeutic strategies is further complicated by a variety of clinical and social factors including the estimated gestational age (EGA) of the fetus, medical condition of the mother, risks of teratogenicity from antiepileptic drugs and chemotherapy, religious and philosophical beliefs of the patient, the stance of her family and medical providers concerning her option for pregnancy termination, and the natural history of malignant glioma.13

Given the paucity of information in the literature, we describe our experience with and the outcomes of a series of women with malignant glioma who were exposed to chemotherapy in early pregnancy. This case series comprises 6 patients with previously diagnosed malignant gliomas who, while undergoing postradiation chemotherapy, discovered that they were pregnant. Despite the unintentional chemotherapeutic exposure, all patients elected to proceed with their pregnancies.


Six women previously diagnosed with malignant glioma were receiving chemotherapy treatment and discovered during treatment that they were pregnant (Table 1). Data for these patients was collected from 5 centers (Huntsman Cancer Institute at the University of Utah; University of California, San Diego; Kaiser Permanente Medical Center, Baldwin Park, Calif; University of Southern California, Los Angeles, Calif; and the Moffitt Cancer Center, University of South Florida, Tampa, Fla) contributed by 2 neuro-oncologists (DTB and MCC).

Table 1. Data From Patient Cases
Case No123456
Mother's age at exposure, y332826242928
Gravida and parityG3 P3 A0G1 P1 A0G2 P2 A0G3 P2 A1G2 P2 A0G1 P0 A0
Pathology of tumor/date of diagnosis/extent of resectionAnaplastic astrocytomaAnaplastic oligodendrogliomaAnaplastic astrocytomaOligodendrogliomaOligoastrocytomaGlioblastoma
Dx 5/03/STRDx 9/87/STRDx 4/90/STRDx 3/93/BiopsyDx 4/97/STRDx/STR8/04
Adjuvant treatmentRTRT + PCVRT + PCVRTRTRT + TMZ
Indication of treatmentRecurrentAdjuvantRecurrentRecurrentRecurrentAdjuvant
Date of delivery12/039/8911/931/992/039/05
Estimated gestational age at discovery, wk244236
Chemotherapeutic agent and dose regimen5th Cycle of TMZ2nd Cycle of PCV4th Cycle of PCV3rd Cycle of PCV9th Cycle of TMZ2nd Cycle of ZTM
Fertility difficultyNoNoNoNoNoNo
Period of exposure to the drugFull cycle (5/28 days)Full cycle (6 weeks)Full cycle 6 wkFull cycle 6 wkFull cycle 5/28 dFull cycle 5/28 d
Fetal weight at delivery, kg3.
Method of deliveryNormal spontaneous vaginal deliveryNormal spontaneous vaginal deliveryRepeat elective Caesarean sectionNormal spontaneous vaginal deliveryRepeat elective Caesarean sectionNormal spontaneous vaginal delivery
Hours of labor31na5na4
Still birth/low birth weight/immediate complications000000
Breast feeding statusEarly cessation of milk production which did not occur in 2 previous pregnanciesDid not breast feedBreast-fed without complicationsBreast-fed without complicationsBreast-fed without complicationsBreast-fed without complications
Antiepileptic medicationNone- (Phenytoin was stopped 10/03)PhenytoinNoneLevetiracetamLevetiracetamLevetiracetam
Duration of follow-up4 y and alive, disease free2 y2 y1 y& 6 mo1 y & 6 mo1 y 1&6 mo
Date of recurrence (after parturition)na11/9010/941/008/0410/06
Salvage therapy after parturitionnaResection and cyclophosphamide X 6Resection and cyclophosphamide X 8CPT-11 X 4CPT-11 X 6TMZ X 2; bevacizumab x5
Time of deathna9/9111/957/002/056/07


(See Table 1 for results.) Pathological diagnoses included 2 anaplastic astrocytomas and 1 each of anaplastic oligodendroglioma, glioblastoma, oligodendroglioma, and oligoastrocytoma. Five patients underwent subtotal resection (STR) at diagnosis (before pregnancy). One patient with oligodendroglioma underwent biopsy only at diagnosis. The mean age of the mothers at the time of pregnancy was 28 years (range, 24 years to 33 years) with parity ranging from 1 to 3 (1: n = 2; 2: n = 3; 3: n = 1). Four mothers were receiving chemotherapy for first tumor recurrence, and 2 mothers were receiving chemotherapy as part of first-line therapy.

All cases of chemotherapeutic drug exposure occurred during the first trimester. Two of the mothers were exposed during the second week of estimated gestational age (EGA), 1 during the third week, and another 3 during the fourth week. Three patients were receiving PCV (procarbazine, CCNU [lomustine], and vincristine) during their second, third, and fourth cycles of chemotherapy, and 3 patients were being treated with temozolomide (TMZ) during their second, fifth, and ninth cycles, respectively.

Four patients underwent normal spontaneous vaginal delivery (1, 3, 4 and 5 hours of labor). Two underwent repeat elective Caesarean sections. The neonatal outcomes included 6 healthy full-term infants with birth weights of 3.3‒3.7 kg (mean = 3.5), all of whom had normal Apgar scores and uneventful courses in the neonatal care unit.

Postnatally, 1 mother opted not to breastfeed, and 5 mothers breast fed their infants. All of the newborns are alive and well with no evidence of congenital malformations.

Disease recurrence after parturition occurred in 5 patients at a median time after pregnancy of 13 months (range, 11 months to 18 months). Two patients underwent reresection for recurrent tumor followed by 6 and 8 cycles of cyclophosphamide. Two patients were treated with 4 and 6 cycles, respectively, of CPT-11, and 1 patient received 2 cycles of temozolomide followed by 4 cycles of bevacizumab. Median survival after disease recurrence after delivery was 8 months (range, 6 months to 10 months).

The surviving patient with anaplastic astrocytoma is disease free 4 years after delivery and has not required further treatment after pregnancy.

Five of the 6 mothers died within 30 months after delivery, all from progression of glioma. All mothers with disease recurrence were treated either with reresection2 or salvage chemotherapy.5 One mother with newly diagnosed anaplastic astrocytoma remains alive and disease free more than 4 years after completing treatment.


The administration of chemotherapy during pregnancy, in particular during the first trimester of pregnancy, may induce harmful effects both in the fetus and in the expectant mother. For the fetus, these detrimental effects include congenital malformations, potential carcinogenesis, organ toxicity, prematurity, impaired fetal weight gain, growth retardation, and developmental delay. For the mother, chemotherapy-associated risks include stillbirth, spontaneous abortion, and maternal sterility.14, 15

Procarbazine is a potent teratogen and carcinogen in animal models and, in such models, can cross the placenta in sufficient concentration to cause fetal death or malformation, particularly when the exposure is in early gestation.16 Four case reports have described the use of procarbazine in Hodgkin-disease patients with unplanned pregnancies. All opted to continue with their pregnancy. In 3 cases, chemotherapy was discontinued when the patient was known to be pregnant. In 1 infant, the only detectable abnormality was an erythematous hemangioma on the forearm.17 In the second case, the infant was born with respiratory distress syndrome and an atrial septal defect.18 In the third case, a child was delivered at 24 weeks with orthopedic abnormalities (talipes, hypoplastic tibia) and a cerebral hemorrhage.19 In the fourth case, procarbazine and carmustine were given to the mother for 5 months before conception and throughout the first and second trimesters, and streptozotocin was given throughout the third trimester. She delivered a low birth-weight, male infant who was otherwise normal.20 By contrast in this case series, no malformations were observed among the 6 live-born neonates.

The treatment of malignant gliomas requires a multidisciplinary approach. Current standards of practice for treating newly diagnosed high-grade glioma include maximal safe resection followed by radiotherapy, and, in the instance of newly diagnosed glioblastoma, adjuvant TMZ chemotherapy. The role of adjuvant chemotherapy in other grades of malignant gliomas remains controversial.

In this case series, all patients were exposed to alkylator-based chemotherapy (either PCV or TMZ) during the first trimester of their pregnancy when fetal organogenesis is most active. Three patients received PCV, as, at that time, this drug combination was widely used as a first-line chemotherapy regimen.21, 22 The commonly observed toxicities of PCV are myelosuppressive (thrombocytopenia, leucopenia, and anemia), neurologic (cognitive or mood change and peripheral or autonomic neuropathy), and gastrointestinal (nausea and vomiting); fatigue and allergic reaction also occur. Temozolomide (TMZ) is a novel imidazotetrazine that has become the standard of care for newly diagnosed glioblastoma and recurrent anaplastic gliomas.23 The commonly observed toxicities of TMZ are mild myelosuppression, typically easily controlled gastrointestinal upset (nausea and vomiting), fatigue, and anorexia.24

Although the use of carmustine wafers (Gliadel) in pregnancy has not been specifically reported, studies have shown virtually undetectable amounts of carmustine in the systemic circulation. This modality has been approved by the US Food and Drug Administration for newly diagnosed and recurrent glioblastoma and may be presumed to be relatively safe for use in pregnant patients with malignant glioma who are candidates for surgical resection.1, 25

It is recognized that all alkylator-based chemotherapy crosses the placenta and, therefore, fetal toxicity depends upon gestational age and duration of exposure.26 In most cases, toxic effects of chemotherapy on the fetus have been reported when treatment was given during embryogenesis or in the first trimester and less often when administered during later trimesters. In the instance of fetal PCV exposure, the estimated risk of malformation when exposed during the first trimester is 1 in 3. Considering any chemotherapy exposure during pregnancy, the rate of chemotherapy-associated fetal malformation is 12.7% to 17% and that of low birth weight is 40%.20 By comparison, the rate of severe congenital malformations in the general population is approximately 1% to 3%, and low birth weight is 10% to 15%.10, 27 There are several reports from the breast cancer literature on experience with chemotherapy for the treatment of women with breast cancer during the second and third trimesters of pregnancy. Chemotherapy agents have included anthracyclines, cyclophosphamide, methotrexate, doxorubicin, and fluorouracil. These experiences suggest no significantly increased incidence of adverse short-term fetal or pediatric events when chemotherapy is administered for breast cancer in the second or third trimester.28, 29 The National Comprehensive Cancer Network (NCCN) guidelines advocate the use of chemotherapy for treatment of breast cancer in the second and third, but not the first, trimester of pregnancy, whereas radiation and hormonal therapies for breast cancer are recommended to be avoided throughout the pregnancy. Most available data on teratogenicity of chemotherapy are based on case reports and small series. Long-term studies are needed to evaluate the effects of chemotherapy on the mother, the fetus, and the neonate.

Although all of the patients in our series had uneventful deliveries (4 vaginal and 2 elective repeat Caesarian sections), delivery in a pregnant mother may be complicated by residual intracranial tumor. Increased intracranial pressure can occur from the added effects of pressure from the exertions of labor upon the tumor and associated peritumoral edema. This scenario is more of a concern in an undiagnosed, untreated patient. In the case of a symptomatic tumor, an elective Caesarean section should be encouraged.

It is not known whether TMZ or PCV pass into breast milk.24, 30 As a general admonition, mothers are advised not to take these drugs while breastfeeding because of the possibility of serious harm to the infant. In this case series, all chemotherapy was stopped after discovery of pregnancy and several months before delivery. Five mothers attempted to breast feed, although 1 mother experienced early cessation of the breast milk production, which had not happened in previous pregnancies. One patient chose not to breast feed.

Concerning radiotherapy, no prospective, randomized studies have been undertaken with respect to administration of radiotherapy for newly diagnosed malignant gliomas in pregnancy because of the paucity of cases as well as ethical considerations as mentioned above. There appears to be a correlation between fetal radiation dose and fetal viability with <10 cGy having no major effect and >300 cGy leading to abortion. A dose of >50 cGy poses a risk of damage during all trimesters.11, 14 However, in cases where radiotherapy is required, an estimation of fetal risk could be made by using phantom measurements to assess the likely fetal dose.11 Abdominal shielding devices have been proposed and proven acceptable for protecting the fetus when treating the mother with radiotherapy to the brain. A total fetal dose-exposure of approximately 1 cGy achievable with shielding devices is within the radiation-dose limits suggested by the American Association of Physics in Medicine. Therefore, pregnant patients may receive cranial radiotherapy for newly diagnosed malignant gliomas with abdominal shielding that provides minimal fetal exposure.

Several theories have been advanced to account for the exacerbation of brain tumor-related maternal symptoms that may occur during pregnancy, including steroid hormone-mediated growth (meningiomas) and hemodynamic changes (increased systemic and cerebral blood volume).5, 6 A potential mechanism that could result in clinical deterioration in expectant mothers with malignant gliomas is an increase in intracranial mass effect from increased tumor blood volume caused by hemodynamic changes inherent to pregnancy. However, in the current case series, there did not appear to be an increase of glioma mass effect due to pregnancy, likely a result of prior resection and application of radiotherapy resulting in residual small-volume disease. In addition, it is during the latter half of the second trimester and in the early third trimester, when maternal intravascular volume peaks, that the majority of newly diagnosed brain tumors present in pregnant patients.5 During this time in pregnancy, fluid retention and increased intravascular volume may exacerbate intracranial mass effect, both by increasing peritumoral edema and absolute tumor blood volume resulting in discovery of the underlying glioma. Pregnancy has not convincingly been shown to accelerate glioma growth.

Magnetic resonance imaging (MRI) is the preferable imaging technique for following a patient with a known brain tumor. MRI uses gadolinium-based contrast agents that are paramagnetic and do not contain iodine. These agents have a far better margin of safety compared with computed tomography (CT) contrast agents, as allergic reactions are relatively rare (an occurrence of approximately 1 in 350,000), and there is a minimal chance of nephrotoxicity in patients with normal renal function.31 Gadolinium-containing agents readily cross the placenta but have not been shown to cause birth defects at conventional dosages. Although there are no documented instances of gadolinium teratogenicity, the majority of radiologists prefer MRI without contrast for imaging during pregnancy, as was performed in this case series.

Another problem confronting physicians who are treating pregnant patients for glioma is use of antiepileptic drugs (AEDs). It is recognized that use of AEDs during pregnancy may be teratogenic and may increase the risk of major congenital abnormalities by at least 2-fold. In many women with either established or newly diagnosed gliomas, AEDs cannot be discontinued because of the risk of seizures during pregnancy, which can be harmful to both mother and fetus. The current American Academy of Neurology guidelines state that physicians should optimize treatment before conception, use monotherapy if possible, choose the most effective AED for seizure type and syndrome, use the lowest effective dose, give the mother supplemental folic acid, and treat the child with vitamin K at birth and possibly the mother late in pregnancy if she is using AEDs that interfere with vitamin K absorption.30

At the time of the discovery of unplanned pregnancy, 4 patients in this case series were receiving AEDs (1 phenytoin and 3 levetiracetam). All AEDs in this series were continued on the basis of a history of recent tumor-related epilepsy; however, no apparent side effects were noted either in the mothers or in their newborns albeit with limited follow-up. In addition, all mothers received supplemental folic acid throughout pregnancy. At discovery of pregnancy, all mothers consulted with high-risk obstetricians and were followed with periodic fetal ultrasounds. The long-term outcome of the infants delivered in this case series appears normal; however, the majority of children are not yet school aged, and, therefore, their neurocognitive outcome is uncertain. In addition, it is too early to rule out the possibility of future secondary malignancies in these children.

Five of the 6 patients in this series died within or shortly after 2 years after giving birth. At this time, only 1 patient survives with inactive disease. Due to the natural history of malignant gliomas, life expectancy of the mother is unfortunately short. Consequently, in early childhood, the child is at very high risk of losing the mother and will likely be reared by either the father or relatives. This may have an impact on the child's growth and development that is difficult to quantify. Moreover, while she is providing care for her infant or very young child, the mother may need intensive medical treatment for her malignant glioma. Thus, there will be an additional burden on the immediate family. Upon discovery of malignant glioma in a pregnant woman, physicians should mention these sensitive issues to the patient's family during discussions of glioma treatment and probable outcome.

In conclusion, this case series illustrates that for women with malignant gliomas and unplanned pregnancies complicated by chemotherapy exposure, the outcome most often results in an uncomplicated pregnancy and healthy live birth. Although we do not advocate that pregnant mothers intentionally be exposed to chemotherapy, we report a series of cases that illustrates favorable outcomes. A serious limitation of this series is the short follow-up of the infants exposed to chemotherapy during their fetal development. Consequently, the risk of late effects has not been established. Nonetheless, these cases lend anecdotal support for a normal pregnancy, delivery, and outcome in cases of unintentional early fetal chemotherapy exposure.