Treatment of pregnant breast cancer patients and outcomes of children exposed to chemotherapy in utero




As women in the US delay childbearing, it has been hypothesized that the incidence of breast cancer diagnosed during pregnancy will increase. There are very little prospective data on the treatment of pregnant women with breast cancer with chemotherapy and even less data on the outcomes of their children who were exposed to chemotherapy in utero.


Fifty-seven pregnant breast cancer patients were treated on a single-arm, multidisciplinary, institutional review board-approved protocol with FAC (5-fluorouracil, doxorubicin, cyclophosphamide) in the adjuvant (n = 32) or neoadjuvant (n = 25) setting. Parents/guardians were surveyed by mail or telephone regarding outcomes of children exposed to chemotherapy in utero.


Of the 57 women, 40 are alive and disease-free, 3 have recurrent breast cancer, 12 died from breast cancer, 1 died from other causes, and 1 was lost to follow-up. Of the 25 patients who received neoadjuvant FAC, 6 had a pathologic complete response, whereas 4 had no tumor response to chemotherapy and eventually died from their disease. All women who delivered had live births. One child has Down syndrome and 2 have congenital anomalies (club foot; congenital bilateral ureteral reflux). The children are healthy and those in school are doing well, although 2 have special educational needs.


Breast cancer can be treated with FAC chemotherapy during the second and third trimesters without significant short-term complications for the majority of children exposed to chemotherapy in utero. Longer follow-up of the children is needed to evaluate possible late side effects such as impaired cardiac function and fertility. Cancer 2006. © 2006 American Cancer Society.

Although breast cancer is one of the most common malignancies diagnosed during pregnancy, it is uncommon with an estimated 1 in 3000 to 3 in 10,000 deliveries being to pregnant breast cancer patients.1–3 Because the incidence of breast cancer increases with age, it has been hypothesized that the incidence of breast cancer diagnosed during pregnancy will increase as more women delay childbearing.4, 5 Breast cancer during pregnancy is often perceived as a situation that puts the life of the mother in conflict with that of her fetus. However, limited data suggest that pregnancy termination does not improve the outcome for pregnant women with breast cancer.6–9

Case-control studies have shown that pregnant women with breast cancer are diagnosed with more advanced-stage disease when compared with nonpregnant women.10–14 Diagnosis delays secondary to patient- and physician-related factors have been documented for pregnant women with breast cancer.11, 12, 15–18 However, the prognosis for pregnant patients does not appear to be worse than that for age- and stage-matched nonpregnant controls.19, 20

The majority of information on the treatment of breast cancer in pregnant women is based on retrospective case series and case control studies. However, at the University of Texas M. D. Anderson Cancer Center (MDACC), we have treated breast cancer in pregnant women with systemic chemotherapy (FAC: 5-fluorouracil, doxorubicin, cyclophosphamide) on a prospective, single-arm, multidisciplinary protocol for more than 15 years. The results of treatment of the first 24 pregnant women with breast cancer treated at MDACC were published previously.21 We present herein a more detailed follow-up report on 57 women with nonmetastatic, newly diagnosed breast cancer who were treated under our supervision with FAC chemotherapy in the second and/or third trimesters of pregnancy. We also present the outcomes of the children who were exposed to chemotherapy in utero in this study.


Since 1992, pregnant women with nonmetastatic breast cancer at MDACC have been given systemic chemotherapy (FAC) on a single-arm, multidisciplinary, Institutional Review Board-approved protocol. Before this protocol the treatment of breast cancer in pregnant women was administered based on a consensus decision by a multidisciplinary panel that included breast surgeons and radiation and medical oncologists (1989–1992). The protocol instituted in 1992 was modified in 2001 to comply with the Health Insurance Portability and Accountability Act and to include a survey of the parents and guardians of the children exposed to chemotherapy in utero to determine the childrens' health status. The modified protocol also allowed for the treatment of the patients with additional types of chemotherapy, such as taxanes, after delivery.

Of 64 pregnant women with invasive breast cancer who provided written consent for treatment with FAC, 57 met the inclusion criteria for this analysis (pregnancy plus localized, invasive breast cancer not previously treated with systemic therapy). Seven women were excluded from this analysis: 2 received cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) before presentation at MDACC, 2 had metastatic disease at presentation, 1 had recurrent breast cancer, 1 underwent diagnosis and treatment after delivery of her child, and 1, who was an international patient and registered on this protocol, was never seen and evaluated at our institution.

Treatment Protocol

Patients with intact tumors or residual disease in the nodal basins and/or breast (after incisional or excisional biopsies) at the time of presentation to MDACC underwent bilateral mammography with abdominal shielding and sonography of the affected breast and nodal basins. Core biopsy of the breast lesion and/or fine-needle aspirations (FNA) of suspicious regional lymph nodes were performed to confirm a diagnosis of invasive breast cancer. Excisional biopsy was performed if needed for definitive diagnosis. The metastatic work-up included a 2-view chest X-ray with abdominal shielding, sonography of the liver, and screening noncontrast magnetic resonance imaging of the thoracic and lumbosacral spine if the patient complained of back pain or had advanced-stage disease at diagnosis. A complete blood count and renal and liver function tests were performed.

Pathology was reviewed at MDACC for histology, nuclear grade, estrogen and progesterone receptor status, and, later, HER-2/neu with the use of immunohistochemistry and/or fluorescence in situ hybridization (FISH). A standardized review of pathology obtained from 38 pregnant women with breast cancer seen, but not all treated while pregnant, at MDACC was published previously.22 Not all patients in our prospective cohort, particularly those whose tumor blocks were at outside institutions, had pathology available for standardized pathology review. The date of pathologic diagnosis of invasive breast cancer was defined as the date of diagnosis.

Once a diagnosis of invasive breast cancer was confirmed, the patient was referred to a maternal-fetal medicine specialist for accurate assessment of gestational age using ultrasound. Patients were counseled regarding the potential effects of chemotherapy on the fetus. The limited data regarding the potential risks and benefits of continuing the pregnancy with or without systemic therapy versus termination of the pregnancy were reviewed with the patient.

Surgical consultation was obtained for those patients with operable disease at diagnosis. Neoadjuvant chemotherapy was offered to those with locally advanced breast cancer at diagnosis and/or those who could have been candidates for breast conservation surgery if their disease was adequately down-staged. A patient was only offered breast conservation surgery if radiation therapy could be delivered within a reasonable time frame after delivery of her child and completion of systemic chemotherapy. Radiation therapy, if indicated, was delayed until after delivery.

Pregnant women in the second and third trimester who met the criteria for systemic chemotherapy were given outpatient combination chemotherapy with cyclophosphamide (500 mg/m2 intravenously on Day 1), doxorubicin (50 mg/m2 by continuous intravenous infusion over 72 hours), and 2 bolus doses of 5-fluorouracil (500 mg/m2 intravenously on Days 1 and 4) via a central venous catheter every 21 to 28 days through gestational Week 35.23 Complete blood counts and renal and liver function tests were monitored as indicated clinically. If a patient had febrile neutropenia, granulocyte colony-stimulating factor (filgrastim) was administered first to treat it and then to prevent further episodes. After delivery, patients were offered other systemic therapies, such as taxane-based chemotherapy for node-positive breast cancer and tamoxifen for those with hormone-sensitive breast cancer. We would then offer trastuzumab after delivery to those with HER-2/neu-positive breast cancer

High-risk obstetrical care allowed for intensive fetal monitoring. Fetal growth sonography was performed every 3 to 4 weeks (preferably before each cycle of FAC chemotherapy) or if indicated clinically. Amniocentesis was offered if there were appropriate obstetric indications. Preterm deliveries were performed if indicated obstetrically, and the mode of delivery (vaginal vs. cesarean section) was determined by the obstetrician.

Clinical Assessments

For patients who received neoadjuvant chemotherapy, clinical stage and pathologic stage was recorded using the American Joint Committee on Cancer staging system (6th ed.) with those diagnosed before the 6th edition being restaged based on available data.24 The type of surgical outcome (no surgery, breast conservation, or mastectomy) was also documented for those who received neoadjuvant chemotherapy. Data on additional therapies received, such as taxanes, tamoxifen, and radiation therapy, were collected for all patients. Maternal status as of April 24, 2005 (alive and disease-free, alive with recurrent breast cancer, dead of breast cancer, dead of other causes, lost to follow-up, and unable to verify vital status and/or disease recurrence) was determined by reviewing a variety of sources including the MDACC medical record, the MDACC Tumor Registry, as well as discussion with their primary medical oncologist. Length of follow-up was defined as the date of their first visit to MDACC for invasive breast cancer to the date of last contact with MDACC or death.

For the children exposed to chemotherapy in utero, the gestational age when they were exposed to FAC, number of cycles of FAC to which they were exposed, and total dosages of FAC to which they were exposed were recorded.

To determine the delivery outcomes, neonatal complications, and subsequent health, development, and educational performance, of the children exposed to chemotherapy in utero, we conducted a survey of the parents and guardians of the children in 2003 (n = 30; mothers treated up to 2001, when the revised protocol was introduced) and in 2005 (n = 13; mothers treated from 2001–2005). This survey, developed in conjunction with our colleagues in the Department of Pediatrics at MDACC, asked the parent or guardian to assess the child's health, development, and performance in school if the child was of school age. This survey was mailed to parents and guardians for whom we could verify the address and who could read and speak English. The parents and guardians were given the choice to opt out of the survey, complete the survey and return it by mail, or complete the survey by telephone.

Statistical Methods

Descriptive statistical analysis was performed using the SPSS software program for Windows (v. 10.0; SPSS, Chicago, IL).


Maternal Outcomes

The demographic characteristics of our patient population are listed in Table 1. The majority of patients were white, with 14% African American, and 19% Hispanic (according to self-reported race and/or ethnicity). Despite their young age at diagnosis, the majority of our patients did not have a strong family history of breast or ovarian cancer.

Table 1. Demographic Characteristics of Pregnant Breast Cancer Patients Treated With FAC During Pregnancy*
Patient Characteristics
  • FAC indicates 5-fluorouracil, doxorubicin, and cyclophosphamide.

  • *

    n = 57 except where indicated.

Mean age at diagnosis33.5 y (range, 24.0–45.0 y)
Mean gestational age at diagnosis17.4 wks (range, 2.0–33.0 wks)
Primagravid20% (of 56 patients)
Race/ethnicity (self-reported)
 African American14%
Family history of breast or ovarian cancer
 1st-degree relative: breast cancer17.5%
 2nd-degree relative: breast cancer1.8%
 1st-degree relative: ovarian cancer3.5%

As shown in Table 2, the majority of the patients presented with tumors that were invasive ductal carcinoma (85%), poorly differentiated (82%), HER-2/neu-negative (71%), as well as estrogen and progesterone receptor-negative (69% and 83%, respectively).

Table 2. Characteristics of the Breast Cancers of Patients Treated With FAC Chemotherapy While Pregnant
Pathology CharacteristicPercent
  • FAC indicates 5-fluorouracil, doxorubicin, and cyclophosphamide.

  • *

    Determined by immunohistochemistry and/or fluorescence in situ hybridization.

Histology (n = 53)
 Invasive ductal85
Modified Black nuclear grade (n = 55)
 Grade 2 (moderately differentiated)16
 Grade 3 (poorly differentiated)82
HER-2/neu (n = 35)*
Hormone receptor status
 Estrogen receptor-negative (n = 36)69
 Progesterone receptor-negative (n = 35)83

Patients who underwent definitive surgery before receiving systemic therapy presented with advanced-stage disease, with almost 70% having positive lymph nodes at diagnosis. Of the 32 patients who received adjuvant chemotherapy, approximately 53% had Stage II disease and almost 38% had Stage III disease at diagnosis. Possibly reflecting their more advanced stage at diagnosis, the majority of these patients (81%) had a mastectomy for local control of their disease.

Of the 25 pregnant breast cancer patients who received neoadjuvant chemotherapy, 6 patients did not have definitive surgery: 4 did not respond to chemotherapy, 1 died of a pulmonary embolism after undergoing a cesarean section, and 1 was still receiving therapy as of April 25, 2005. Compared with patients receiving adjuvant chemotherapy, patients receiving neoadjuvant chemotherapy were more likely to have Stage III disease at diagnosis (56% vs. 38%). Of the 19 patients who received neoadjuvant chemotherapy and went on to have surgery, almost 32% had a pathologic complete response (no evidence of tumor in the breast or axillary lymph nodes) at surgery. Compared with the women receiving adjuvant chemotherapy, patients receiving neoadjuvant chemotherapy and who underwent definitive surgery were more likely to have breast-conservation therapy (37% vs. 19%).

For the women with complete data on the use of additional systemic therapies after delivery and completion of FAC (4–6 cycles; n = 52): 16 did not receive any additional systemic therapy, 31% received taxanes, and 12% received tamoxifen. Two patients underwent high-dose chemotherapy with autologous bone marrow transplantation.

At a median follow-up duration of 38.5 months (range, 1.0–189.0 months), the majority of the patients were alive and free of disease as of April 24, 2005 (n = 40). Three patients had recurrent breast cancer, and 1 patient who had metastatic disease at multiple sites, including the brain, left the US more than 1 year before the writing of this article and her status is unknown. Of the 13 patients confirmed to have died, all died of metastatic breast cancer except 1, who died from a pulmonary embolism after a cesarean section. Of the 12 patients who have died from metastatic breast cancer, 10 presented with Stage III disease and 2 presented with Stage II (node-positive) disease.

Outcomes of the Children Exposed to Chemotherapy In Utero

Most of the patients underwent definitive surgery in their second trimester (35.1%), although many underwent surgery after delivery (33.3%). The median gestational age at definitive surgery was approximately 16.0 weeks (range, 4.29–29 weeks). There were no fetal complications noted with any of the surgeries.

Most of the children were exposed to chemotherapy in the second trimester (Table 3). None of the children were exposed to other systemic chemotherapeutic, hormonal, or biologic therapies in utero. One patient received brachytherapy to her breast mass after failure of FAC and refusing further systemic therapy while pregnant.

Table 3. Chemotherapy Exposure In Utero
  1. FAC indicates 5-fluorouracil, doxorubicin, and cyclophosphamide.

Gestational age at chemotherapy initiation (n = 51)23 wks11–34 wks
Number of cycles of FAC given in utero (n = 52)41–6
Chemotherapy doses given to pregnant breast cancer patients
 5-Fluorouracil (n = 35)6931 mg1640–13,600 mg
 Doxorubicin (n = 36)362 mg82–680 mg
 Cyclophosphamide (n = 36)3639 mg820–6800 mg

No stillbirths, miscarriages, or perinatal deaths occurred in our cohort of patients who received FAC chemotherapy during their second and/or third trimester. Table 4 shows the delivery outcomes of the patients who delivered compared with reported norms in the general population.25, 26 The majority of patients delivered at a gestational age of at least 34 weeks; only 3 patients delivered before this gestational age. Only the child born at 29 weeks secondary to the mother's development of preeclampsia weighed less than 2000 g at birth; overall, only 6 of the children weighed less than 2500 g at birth (birth weight range, 1389–2495 g; gestational age range, 29–40 2/7 weeks, with 5 of the 6 being born at less than 38 weeks gestation).

Table 4. Delivery Outcomes in Children With Chemotherapy Exposure In Utero vs. the General Population
OutcomePercentReported norm of general population
  • *

    Perinatal mortality rate includes both late fetal (at least 28 weeks of gestation) and early neonatal (<7 days) deaths.

Perinatal mortality rate* (n = 55)06.9 deaths per 1000 live births and fetal deaths25
Type of delivery (n = 52)
 Vaginal60Cesarean section rate was 27.6% among the U.S. population in 200325
 Cesarean section40
Gestational age at delivery (wks) (n = 52)37 (range, 29–42)47% of live births in the United States. are gestational age 37–39 wks26
Birth weight (g) (n = 47)2890 (range, 1389–3977)3117–3956 g (50th to 95th percentiles at 37 wks of gestation)26

We had an excellent response to the mail/telephone survey of the parents or guardians of the children exposed to chemotherapy in utero (93% [40/43]). At the time of the survey, the children's ages ranged from 2 to 157 months. The majority of the children did not have any significant neonatal complications and seem to be similar to reported norms for the general population27–29 (Table 5). The most common neonatal complication was difficulty breathing, with 10% of the neonates (4/40) requiring ventilation. One child born vaginally at a gestational age of 38 weeks had a subarachnoid hemorrhage (SAH) on Day 2 postpartum. Although this occurred more than 3 weeks after the mother's last course of chemotherapy and the mother's complete blood count was normal, the child had both neutropenia (absolute neutrophil count approximately 0.6 K/UL) and thrombocytopenia (platelet count = 89 K/UL). No etiology for the subarachnoid hemorrhage was found. The child was doing well developmentally at 40 months of follow-up with only residual right leg weakness, for which the child has received physical therapy. One child, born to a woman who was 32 years of age, had Down syndrome.

Table 5. Neonatal Outcomes in Children Exposed to Chemotherapy In Utero vs. General Population
Outcome (n = 40)PercentReported Incidence among the General Population
Neonatal complications
 Breathing difficulties: (requiring ventilation)28 (10)21% of infants at a tertiary hospital had respiratory distress27
 Subarachnoid hemorrhage2.50.8/1000 births in a large Canadian series28
Congenital or chromosomal abnormalities
 Down syndrome2.51:700 births born to mothers 30–34 years old29
 Clubfoot2.51:1000 births29
 Congenital bilateral ureteral reflux2.53–4% of newborns have some abnormality of kidneys/ureters29

All parents and guardians who responded to the survey felt that their child exposed to chemotherapy in utero was healthy. All but the child with Down syndrome were thought to have normal development when compared with siblings or other children at a similar age. Of the school-age children (n = 18), only 2 required special attention in school: 1 had attention deficit disorder, whereas the other was the child with Down syndrome. Most of the children were without significant health problems (Table 6).

Table 6. Postneonatal Outcomes in Children With Chemotherapy Exposure In Utero
  • *

    n = 40 unless otherwise stated.

“Normal development” compared to siblings or other children, per child's family or guardian (n = 39)97
Requires special attention in school (n = 18)11
Reported to have no health problems43
Upper respiratory infections: ear, sinus, bronchiolitis13
Asthma/breathing difficulties10
Attention deficit disorder5
Hypercholesterolemia, with obesity3
“Eye problems,” not otherwise specified3
Heart murmur (resolved by age 1 year) and “lazy eye”3


We have described the presentation, treatment, and outcomes for the largest cohort of pregnant patients with breast cancer treated on a prospective clinical trial. Based on our data, pregnant women with breast cancer can be treated with FAC chemotherapy in the second and/or third trimesters with relative safety for both mother and fetus. In our prospective cohort, the majority of children exposed to chemotherapy in utero were reported to be healthy, with no significant developmental problems except for the child with Down syndrome.

These data are comparable to a cross-sectional study of pregnant women with hematologic malignancies performed by Aviles and Neri,30 who examined the outcomes of 84 children exposed to chemotherapy in utero. In that study, the children's learning and educational performance were normal, and no congenital, neurologic, psychologic, cardiac, or cytogenetic abnormalities or malignancies were noted. Furthermore, the second-generation children were reported to have no significant problems, although their parents refused formal medical or intelligence tests for their children. Despite a number of case series indicating that exposure to chemotherapy in the first trimester can result in pregnancy loss and congenital malformations, a number of patients in the study by Aviles and Neri were exposed to chemotherapy in the first trimester without significant complications.31–33 In our pregnant patients with breast cancer, we did not start chemotherapy until the second trimester, although 1 of our patients started chemotherapy at approximately 11 weeks gestation (by dates). Our follow-up data on the children exposed to chemotherapy in utero is not complete because some of our former pregnant patients have been lost to follow-up or died, and we have been unable to contact the next of kin. However, one would anticipate that the parents and guardians of children exposed to chemotherapy in utero would inform us of potential adverse outcomes, as they are aware of their child's unusual medical history.

The majority of our patients were diagnosed with advanced-stage disease, including regional lymph node metastases. Although a more advanced stage at presentation has been reported in case series and case-control studies of pregnant breast cancer patients, it has also been reported that it is not uncommon for young breast cancer patients to present with lymph node metastases.10–14, 34–38 Consistent with previously published reports of pregnant and nonpregnant young women with breast cancer, our patients were more likely to have carcinomas that had ductal histology and were poorly differentiated.34, 38–45

Despite presenting with more advanced-stage breast cancer, some of our pregnant breast cancer patients were down-staged through the use of neoadjuvant chemotherapy. With careful multidisciplinary planning, 37% of these patients were able to undergo breast conservation. Neoadjuvant chemotherapy and breast conservation can be considered an option for the pregnant breast cancer patient who presents with a larger tumor and/or regional lymph node involvement.

There has been a concern that the hormonal milieu of pregnancy contributes to the development and progression of the breast malignancy. Our data do not support this hypothesis because the majority of our patients had estrogen and/or progesterone receptor-negative tumors. Our percentage of estrogen and progesterone insensitivity is higher than that previously reported for young, nonpregnant breast cancer patients and in smaller case-control and case series of pregnant breast cancer patients.11, 12, 38, 40, 46, 47 Some of the previous studies used the ligand-binding assay for quantification of hormone receptor status and this test may have been inaccurate in the pregnant breast cancer patient.11, 12, 46

Our systemic treatment of choice remains FAC and we postpone the use of antiestrogen therapy such as tamoxifen and radiation therapy until after delivery. Although there are some case reports on the use of taxanes in pregnant women with breast or ovarian cancer, we believe the limited safety data precludes the routine use of taxanes during pregnancy.48–52 We also do not routinely use trastuzumab during pregnancy because there are only a few published case reports in breast cancer. In 2 of the 3 reported cases, trastuzumab may have been associated with oligohydramnios.53–55

To our knowledge, there are no published case reports on the use of dose-dense anthracyclines and/or taxanes in pregnant women. Given the increased risk of bone marrow suppression with the dose-dense regimen as compared with standard dosing, growth factor support with filgrastim or pegfilgrastim and erythropoietin is often part of the treatment regimen to decrease the risk of neutropenia and anemia, respectively.56, 57 The majority of case reports and case series of anemic or neutropenic pregnant women treated with growth factors did not report significant problems for the mother or fetus.58, 59 Hypertension and/or worsening renal failure has been reported in some pregnant women with renal failure treated with erythropoietin.60, 61 However, the majority of patients in these reports were not pregnant women receiving chemotherapy for solid tumor malignancies. It is unclear if such growth factors administered to a pregnant woman undergoing dose-dense chemotherapy for breast cancer would protect the fetus from bone marrow suppression. Given that 1 of our patients gave birth to a child with neutropenia and thrombocytopenia despite the fact that our patient had normal blood counts at delivery, we have been reluctant to pursue dose-dense therapy among our pregnant patients.

In our cohort of children exposed to chemotherapy in utero, further follow-up of the children is required to determine the potential long-term effects of this exposure on a number of parameters, including fertility, cognition, and cardiac function. However, our short-term data are reassuring, in that most children are healthy and the majority of those of school age are reported to be doing well in school. The low rate of complications observed in the children of our pregnant breast cancer patients may be related to the fact that a highly interactive group of specialists familiar with this unique obstetric and oncologic situation work together as a team to provide the best care possible for both mother and child.