Abstract
The incidence of pregnancy-associated breast cancer (PABC) increases as more women choose to delay childbearing and the population-based incidence of breast cancer rises. Reliably and safely staging PABC is necessary to choose between starting with local or systemic therapy. With regard to local therapy, both lumpectomy and mastectomy can be considered depending on gestational age and the stage at diagnosis. By mirroring nonpregnant treatment regimens as much as possible, chemotherapy may improve long-term oncologic outcomes while allowing for surgical downstaging during pregnancy. Delaying treatment due to misconceptions regarding risk of local and systemic therapy most certainly worsens oncologic outcomes, and most neonatal morbidity is related to gestational age at delivery and not in utero exposures. Pregnancy itself was once considered an independent risk factor for worse outcome, but the prognosis of these patients is not significantly different than nonpregnant counterparts of a similar age.
Introduction
P
Diagnosis of PABC
The evaluation of breast symptoms during pregnancy and the postpartum period can be challenging due to the hormonally induced changes in breast tissue that may lead to increased firmness and nodularity. Furthermore, the symptoms of postpartum lactational mastitis mimic locally advanced or inflammatory breast cancer. The majority of PABCs are diagnosed after presenting with a palpable mass. However, skin thickening and skin redness can be present up to a quarter of the time. 3,6
Completing a reliable diagnostic workup with imaging to determine the extent of disease is important in treatment decision-making. In a nonpregnant patient, breast imaging can include ultrasound, mammogram, and breast magnetic resonance imaging (MRI). Ultrasound helps discern between cystic and solid masses, and mammography can reveal calcifications that may not be visible by ultrasound alone. 7 Ultrasound is widely used in pregnancy and the safety has been previously established. 8 Mammography confers minimal dose to the fetus with abdominal shielding (0.001–0.01 mGy with two views), far below the minimum threshold of 200 mGy for adverse effects during organogenesis (up to 10 weeks of gestation). 9
Although contrast-enhanced breast MRI can be a useful diagnostic tool in non-PABC, the safety of gadolinium in pregnancy is controversial. Free gadolinium is considered toxic and therefore only administered to humans in chelated form. It crosses the placenta and remains in the amniotic fluid to be swallowed by the fetus and re-enters the fetal circulation. Furthermore, the prone positioning necessary for breast MRI may lead to prolonged pressure on the gravid uterus, disrupting uterine blood flow. Although lactating women are encouraged to pump or breastfeed immediately before MRI to reduce the background enhancement, gadolinium is excreted into breast milk at a rate of 0.0004% of the maternal dose, and therefore, the American College of Radiology guidelines do not require that patients discontinue breastfeeding after contrast-enhanced MRI. 10
In some advanced PABC cases where metastases are suspected, a metastatic workup before delivery may be necessary to guide treatment decisions. Given that lungs, bone, and liver are the most common metastatic sites of breast cancer, a pregnant patient may undergo a chest X-ray with abdominal shielding, liver ultrasound, and noncontrast supine MRI in place of the bone scan to complete the metastatic workup. The fetal dose of PET/CT has been found to be 10–50 mGy and therefore is usually deferred to the postpartum period. 9 See Figure 1 for a diagnosis and treatment algorithm of PABC.

Algorithm for the diagnosis and treatment of pregnancy-associated breast cancer.
Treatment of PABC
Initiation of local versus systemic therapy in PABC
Depending on the clinical scenario, PABC patients and their providers may be confronted with the choice to either proceed with local control or to initiate systemic treatment. For patients diagnosed with early-stage PABC, both breast conservation and mastectomy are considered viable surgical options. However, diagnosis in the first or second trimester may influence the decision to proceed with mastectomy to avoid a potential delay in administration of post-lumpectomy external beam radiation, which is contraindicated during pregnancy. See Table 1 for treatment options of PABC by gestational age.
Although nonobstetric surgery during the first trimester has not been found to be unsafe according to the American College of Obstetricians and Gynecologists. 11
Cyclophosphamide, 5-fluorouracil, and doxorubicin use reported in the literature (taxanes, platinum-based agents are less studied). 21 –25
EFM, electronic fetal monitoring; GA, gestational age.
With regard to specific risks of any surgery during pregnancy, the second trimester is considered by many practitioners to be the safest time for surgery, with the lowest risk of miscarriage and preterm labor. However, the American College of Obstetrics and Gynecology has advised that nonobstetrical surgery during the first trimester is routinely performed and considered to be low risk to the mother and fetus, and none of the contemporary anesthetic agents has been shown to have teratogenic effects when used in standard doses at any gestational age. 11 Therefore, surgery during the first or second trimesters would be appropriate.
When nonobstetric surgery during pregnancy is deemed necessary, a team approach should be used. Obstetricians and anesthesiologists familiar with aspects of maternal physiology and anatomy that could affect intraoperative management should be involved. After 20 weeks of gestation, positioning the patient in left lateral tilt position displaces the uterus away from the inferior vena cava, allowing for optimal venous return and uterine perfusion while supine in the operating room. If the fetus is considered viable (24 weeks), simultaneous electronic fetal and contraction monitoring should be performed before and after the procedure to assess fetal well-being and monitor for preterm contractions. If intraoperative fetal monitoring is used after 24 weeks, a qualified individual should be available to interpret the fetal heart rate pattern and an obstetric care provider with cesarean delivery privileges should be readily available. In a previable fetus, fetal heart rate by Doppler before and after the procedure is sufficient. 11
Breast conservation versus mastectomy
After controlling for age, stage, tumor size, and hormone receptor status, there is no associated survival advantage for pregnant patients undergoing mastectomy versus breast conservation. 12,13 Patients who are pregnant should have the same discussions with their surgeons regarding breast conservation therapy versus mastectomy as their nonpregnant counterparts. Autologous breast reconstruction should be delayed until the breasts return to their baseline architecture to improve cosmesis; however, there is no contraindication to the placement of tissue expanders during pregnancy. Nevertheless, attention should be paid to minimizing prolonged surgeries that may involve reconstruction due to increased risks of venous thromboembolism seen in the hypercoagulable state in pregnancy. 14 Of note, both unfractionated heparin and low-molecular-weight heparin are safe to use in pregnancy and perioperative chemical thromboprophylaxis should be considered in addition to pneumatic compression devices in a pregnant patient with cancer. 15
Sentinel lymph node biopsy in pregnancy
The National Surgical Adjuvant Breast and Bowel Project (NSABP) B-32 trial found that in clinically node-negative patients, sentinel lymph node biopsy has a low false-negative rate and is preferred in nonpregnant patients due to decreased risk of chronic lymphedema. 16 Radioactive tracer (technetium sulfur colloid and technetium-labeled albumin) is not contraindicated in pregnancy. 17 –19 Studies that have estimated the radiation exposure to the fetus have determined that the maximum absorbed dose of radiation using the standard dose of radioactive colloid is well below the absolute minimum of 50 mGy, believed to be the threshold absorbed dose for adverse effects. 20,21 While most practitioners prefer the use of radiocolloid in pregnancy and recommend against the use of blue dyes (such as methylene, lymphazurin, and isosulfan), a recent study suggests methylene blue may also be safely used. 22
Radiation
Patients who choose breast conservation or who require postmastectomy radiation must delay radiation treatment until after delivery to avoid exposure to the fetus. 23 Timely treatment initiation is imperative, as patients undergoing lumpectomy with adjuvant radiation should begin radiotherapy within 8–12 weeks to maintain disease-free survival advantage and to avoid increased risk of local recurrence. 24
Principles of Systemic Treatment
Timing of systemic treatment
Chemotherapy should be deferred until after the first trimester (after 14 weeks) when organogenesis is considered complete (10 weeks). 25 The rate of malformations is up to 14% if chemotherapy is given in the first trimester, and even higher with combination chemotherapy. However, if administered in the second or third trimester, the rate of malformation mirrors the baseline population's 3% risk of congenital malformations. 26 At least 3 weeks between a cycle of chemotherapy during pregnancy and delivery are recommended. 27
Reports of oncologic outcomes after chemotherapy during versus after pregnancy suggest that delaying systemic treatment in PABC patients may worsen prognosis. 27 Nettleton et al. found that by delaying chemotherapy by 3–6 months, the risk of metastases was increased by 5%–10%. 28 In a retrospective study by Beadle et al. of young women with PABC, all patients had a 10-year local recurrence risk of 20%. When these patients were matched by age and stage of diagnosis to non-PABC patients, there was no difference in oncologic outcome. However, there was a trend toward worse overall survival if treatment, which in this study included both medical and surgical, was delayed until the postpartum period. 25
Systemic chemotherapy
Cytotoxic chemotherapy during pregnancy should mimic nonpregnant regimens as closely as possible. Anthracyclines, specifically doxorubicin, have not been found to significantly affect cardiac outcomes of children exposed in utero, and no differences in ejection fraction have been noted in large retrospective groups of patients' offspring who were followed for up to 29 years of age. 29 Cyclophosphamide has also not been shown to increase neonatal morbidity. 30 In a study by Hahn et al., 57 pregnant breast cancer patients were treated on a single-arm protocol with FAC (5-fluorouracil, doxorubicin, cyclophosphamide) in the neoadjuvant or adjuvant setting. Four cycles of this anthracycline- and cyclophosphamide-containing regimen were given through 35 weeks of gestation without an increased incidence of stillbirth, miscarriage, or perinatal death. 31
Taxanes, specifically paclitaxel, have not been found to be teratogenic when administered in the third trimester. Paclitaxel is preferred over docetaxel due to docetaxel's greater transplacental transfer. 32 Although taxanes may play a promising role in the therapeutic strategy of PABC, there are overall less data on the safety of taxanes in pregnancy. 33 Although poor outcomes have not been associated with taxanes, the studies are all smaller case series and should be interpreted with caution.
Platinum derivatives are alkylating agents and may play a role in Her2-overexpressing or triple-negative disease. Highly protein bound, the unbound fraction may cross the placenta. Carboplatin may be associated with derangements in trophoblast invasion at the maternal/placental interface, disrupting placental development that is not complete until 20 weeks of gestation. Although the data regarding the safety of carboplatin in pregnancy are limited, a systematic review of the use of carboplatin and cisplatin in pregnancy found that no malformations or toxicity was reported in seven carboplatin-exposed neonates. 34
Contraindicated chemotherapy agents
Trastuzumab, used in the neoadjuvant and adjuvant treatment of Her2-positive disease, is relatively contraindicated in pregnancy. It has been linked to cases of reversible oligohydramnios (low amniotic fluid) or anhydramnios (no amniotic fluid) in 33% of patients who receive it while pregnant. 2 However, this Her2-targeted treatment may be discussed in special high-risk situations, and inadvertent exposure is not considered an indication for termination. There are not yet published data on the use of pertuzumab in pregnancy, although pertuzumab, along with trastuzumab and ado-trastuzumab emtansine, is currently being studied in the MotHER trial. 35
Selective estrogen receptor modulators such as tamoxifen citrate and raloxifene are generally contraindicated during pregnancy due to risk of ambiguous genitalia. 36 More dramatic congenital anomalies, such as spinal abnormalities, absent ears, or craniofacial abnormalities, and the cardiac malformations seen in Goldenhar's syndrome have also been associated with tamoxifen use. 37
Chemotherapy dosing in pregnancy
Pregnancy leads to a 40%–60% increase in plasma volume starting as early as 6 weeks of gestation. The increase in volume is accompanied by a drop in serum albumin, which can alter the concentration of protein-bound drugs such as taxanes. Furthermore, renal and hepatic clearance is increased, which can further decrease plasma concentrations of chemotherapy drugs, leading to decreased drug exposure. 38
Although the physiologic derangements in pregnancy may result in altered drug pharmacokinetics such as lower maximal concentrations, the current recommendations are to not decrease or increase doses, and to dose according to actual body weight as per the American Society of Clinical Oncology guidelines. Treatment intervals should not be adjusted, and dose-dense regimens may be considered in pregnancy as well. However, the patient may need more frequent monitoring for anemia, the need for transfusion, or febrile neutropenia. 39
Supportive therapy
Pregnant patients may be more vulnerable to the noxious side effects of chemotherapy such as dizziness, nausea, and vomiting and may warrant aggressive supportive therapy. The increase in circulating progesterone leads to decreased gastric motility and contributes to the oftentimes hyperemetic state that pregnancy induces. 40 Ondansetron (Pregnancy Category B) has not been causally linked to an increased risk of adverse outcomes in pregnancy. Although the 5-HT3 antagonist was reportedly correlated to an increased risk of cleft palate in a retrospective epidemiological study, more recent higher quality data have demonstrated that having a history of hyperemesis gravidarum, not the exposure to ondansetron, may be associated with an increased risk of birth defects. 41,42 Metoclopramide (Pregnancy Category B) is commonly used to treat hyperemesis in pregnancy and helps improve gastric motility. 43,44
Steroids are often given along with other antiemetic drugs prophylactically before the administration of chemotherapy. Dexamethasone has been linked to cleft palate in murine models so methylprednisolone is usually preferred. 45,46 Histamine (H1 and H2) blockers such as ranitidine (Pregnancy Category B) are routinely used to treat pregnancy-related gastritis and reflux symptoms, and therefore may also be helpful in the multimodal antiemetic regimens during chemotherapy. 47
Data regarding the use of granulocyte colony stimulating factors (G-CSF) filgrastim and pegfilgrastim to reduce the incidence of febrile neutropenia are limited. 48,49 If the decision is made to utilize dose-dense chemotherapy during pregnancy, G-CSF should not be withheld. 2 A retrospective analysis of 176 women who received the stem cell mobilizer during pregnancy demonstrated no difference in gestational age at delivery, birth weight, incidence of congenital malformations, or leukocyte count at birth when fetuses were exposed in utero. 49
Peripartum Considerations
A multidisciplinary team including medical and surgical oncology, along with maternal fetal medicine and social work, should be assembled early in the care of a patient with PABC. Fetuses exposed to treatment for breast cancer will require closer monitoring during pregnancy, which may present further logistical hardship for the patient outside of frequent medical oncology visits. Before initiating therapy, gestational age and normal anatomy should be confirmed. Chemotherapy should be discontinued at 35–37 weeks or 2 weeks before a planned delivery to reduce the risk of maternal or neonatal cytopenia. 2
Similar to patients without cancer, if iatrogenic preterm delivery before 37 weeks is anticipated, then antenatal corticosteroids should be administered 48 hours before delivery to improve fetal lung maturity. 50 With regard to mode of delivery, vaginal delivery is preferred over cesarean section due to the shorter recovery period. The placenta should be sent to pathology after delivery to rule out the rare event of placental metastases. 51,52
Prognosis of PABC
Historically, breast cancer diagnosed during pregnancy, particularly if diagnosed in the first trimester, has been an impetus to recommend termination of the pregnancy to mitigate the potential tumorigenic effects of the pregnancy-induced hormonal milieu and to allow for initiation of treatment. In a study of women diagnosed with breast cancer in the first trimester, there was not a difference in oncologic outcome between those women who terminated and those who did not. 53 The decision to terminate a pregnancy is a personal choice of the well-informed patient, but she may benefit from a discussion regarding the treatment options specific to her tumor and the optimal timing of surgery, radiation, or systemic therapy.
Most data on the prognosis of patients with PABC are based on small cohorts and are limited by the retrospective nature and heterogeneous treatment regimens. However, in a study of more than 300 women with PABC matched to more than 800 nonpregnant controls with breast cancer, poor prognostic factors of high-grade and hormone receptor-negative tumors were more common in those diagnosed during or after pregnancy. However, disease-free survival and overall survival were not significantly different when the patients were matched for age, stage, and tumor characteristics. Therefore, although pregnancy itself was once considered an independent risk factor for worse outcome, the prognosis of these patients is not significantly different than nonpregnant counterparts of a similar age. 1
Diagnoses in the postpartum period, however, may portend a worse prognosis. In a retrospective cohort study of 619 women younger than 46 years diagnosed with breast cancer between 1981 and 2011, cases diagnosed within 5 years postpartum had an almost three times greater risk of distant recurrence and death than nulliparous women, even with adjustments for biologic subtype, stage, and year of diagnosis.
Future Pregnancies and Genetic Testing
Women with a history of PABC should not be discouraged from future pregnancy. However, becoming pregnant within 12 months after delivery may worsen her prognosis. 37 In PABC patients with a history of estrogen receptor-positive tumors, who have been prescribed at least 5 years of adjuvant endocrine therapy, it is not clear if interruption in adjuvant endocrine therapy affects overall prognosis, but disease outcomes are currently being studied in the multicenter prospective POSITIVE trial to address this question. 38
Patients with breast cancer diagnosed during childbearing age (usually less than 40) meet criteria for genetic testing. Those found to be BRCA mutation heterozygotes should be counseled regarding the potential to utilize preimplantation genetic diagnosis (PGD). PGD is an assisted reproductive technology where in vitro embryos without the autosomal dominant mutation are selected for implantation, allowing patients to “eliminate” the deleterious mutation from their future offspring.
Conclusions
Delaying treatment due to misconceptions regarding risk of local and systemic therapy most certainly worsens oncologic outcomes. Treatment timing and delivery planning require the assembly of a multidisciplinary team at diagnosis. By mirroring nonpregnant treatment regimens as much as possible, using neoadjuvant chemotherapy may improve long-term oncologic outcomes while allowing for effective downstaging, positively impacting choice of surgical modality and subsequent management of the axilla. Although breast cancer in the postpartum period may portend a worse prognosis, pregnancy itself is no longer considered an independent risk factor for poor outcome. Historically, uncertainties regarding the safety of diagnostic modalities and treatment of PABC may have led to worse outcomes in this group of younger women with breast cancer. However, more recent data clarify the safety of a complete but adjusted diagnostic workup and broaden the treatment options in this heterogeneous disease process.
Footnotes
Author Disclosure Statement
The authors disclose no financial or personal relationship with other people or organizations that could inappropriately influence their work.
