Abstract
The diagnosis of breast cancer during pregnancy, athough uncommon, can be a devastating and confusing diagnosis for many women. Many clinicians are unaware of the treatment options available for the pregnant patient and are unable to counsel their patients. The treatment of pregnancy-associated breast cancer is determined by the age of gestation and the stage of cancer at presentation. A multidisciplinary approach involving a medical oncologist, surgical oncologist, radiation oncologist and high-risk obstetrician will lead to an optimal outcome for the mother and child. This article will discuss the management of a pregnant patient with breast cancer.
Keywords
A diagnosis of breast cancer during pregnancy is a devastating and stressful time for the patient and her family. Misconceptions and false advice could lead the patient to feel as if she needed to choose between herself and her child. When the patient and her caregivers are educated in the multimodality treatment of breast cancer during pregnancy, the focus can change to optimizing the health and safety of both the mother and child. The treatment of breast cancer during pregnancy involves a multidisciplinary team of surgical oncologists, medical oncologists, radiation oncologists and high-risk obstetricians. This article will discuss the current information regarding the diagnosis and treatment of pregnancy-associated breast cancer (PABC).
Incidence
According to the American Cancer Society, there will be an estimated 211,240 new cases of breast cancer diagnosed in women in the USA in the year 2005 [101]. Of these, approximately 0.2–3.8% will coincide with pregnancy or lactation [1]. PABC is defined as breast cancer diagnosed during pregnancy or within the 12 months following delivery. It is estimated that 1 in 3000 to 3 in 10,000 deliveries are to women diagnosed with breast cancer during pregnancy [2–4]. A review of deliveries in California from 1992 to 1997 revealed 0.13 cases of breast cancer per 1000 live births [5]. Although an uncommon occurrence, it has been postulated that the incidence of PABC will increase as women delay childbearing for personal or professional reasons [6].
Diagnosis
Breast cancer in pregnancy most often presents as a painless mass or thickening in the breast [7]. Physiologic changes of the breast during pregnancy, including increased density and nodularity of the breast tissue, can lead to a difficult physical exam and delayed presentation [8]. Previous reports have cited delays in diagnosis of PABC of up to 6 months, but more recent studies report a mean delay of 1 or 2 months [9]. Nettleton and colleagues developed a mathematical model to quantify the risk of delaying treatment of breast cancer during pregnancy [10]. For tumors with a doubling time of 130 days, the daily increased risk of axillary metastases due to treatment delay was 0.028%. A 1-, 3- or 6-month delay increased risk by 0.9, 2.6 and 5.1% respectively. For more aggressive tumors with a doubling time of 65 days, the risks at each time interval doubled.
All breast masses in pregnant patients should be assessed by clinical exam, diagnostic imaging and biopsy. Mammography with abdominal shielding is safe in pregnancy, but may be associated with a high false-negative rate secondary to the increased density of the breast [9]. Ahn and colleagues recently reported abnormal mammograms in 13 of 15 (86.7%) patients with PABC [11]. Eight of the patients had masses, while the rest had other suspicious radiographic abnormalities. Breast ultrasound has not been extensively evaluated in women with PABC. Two studies reported the finding of solid masses in all evaluated cases of PABC (a total of 25 cases) [11,12], while another study reported that the ultrasound findings of two of four patients with PABC appeared benign [13]. Yang and colleagues reported a review of 23 women at MD Anderson Cancer Center, TX, USA, with PABC who underwent imaging prior to surgery with mammography (n = 3), sonography (n = 4), or mammography and sonography (n = 17) [14]. Mammography was positive for malignancy in 18 out of 20 cases (90%), while sonography showed a mass in all 21 cases (100%). Sonography also correctly demonstrated axillary metastasis in 15 women and was utilized to determine response to chemotherapy in 12 women. Studies of breast magnetic resonance imaging (MRI) for PABC are not available at this time. MRI is not recommended for imaging of breast malignancy during pregnancy in the first trimester because its influence on fetal organogenesis is unclear. MRI for other staging purposes appears to be safe during the second and third trimesters. In addition, gadolinium-based contrast media should not be used in pregnancy because it crosses the placenta and has been shown to cause fetal abnormalities in rats [15,16].
For any suspicious breast masses, a definitive biopsy should be performed. This can be done through fine needle aspiration (FNA), core biopsy, incisional biopsy or excisional biopsy. FNA will provide a sample for cytology and can be accurate in the diagnosis of breast cancer [17,18]. Accurate diagnosis requires a pathologist familiar with the typical cytologic changes of pregnancy and a lactating breast. Core biopsies provide tissue for histologic study, allow the diagnosis of invasive cancer, and allow the assessment of prognostic markers including hormone receptor and human epidermal growth factor receptor 2 (HER-2/neu) status. There is a single case report of a milk fistula after core biopsy [19]. An excisional or incisional biopsy can be performed relatively safely during any trimester of pregnancy [20].
Staging
Pregnant women with breast cancer should have adequate staging in order to determine the appropriate treatment modalities and counsel the patient regarding the risks and benefits of treatment. Because of the delay in presentation or diagnosis of breast cancer, many patients present with more advanced disease. A case–control study from France evaluated 154 patients with PABC and 308 matched controls [21]. The patients in the PABC group had a larger tumor size (4 vs 3 cm), more lymph node involvement, but no increased metastases at diagnosis. Middleton and colleagues evaluated 39 PABC patients from MD Anderson Cancer Center and found that 79% of patients presented with lymph node metastases [22].
Clinical staging should involve a thorough physical exam, including the nodal basins. For any clinically suspicious lymph nodes, an ultrasound with FNA should be performed. The potential for metastatic disease to spread to other sites, especially lung, liver and bone, needs to be assessed. Radiographic staging should be modified in pregnancy to limit the amount of radiation exposure to the fetus. According to the experience of pregnant women exposed to radiation following the atomic bombs in Hiroshima and Nagasaki, a level of 1–9 centiGrays (cGy) can lead to increased fetal abnormalities, depending on the time of exposure [23]. A total radiation dose exposure of less than or equal to 5 cGy is considered acceptable with low likelihood of adverse outcome [24,25]. A chest x-ray with abdominal shielding exposes the fetus to 0.02–0.07 miliGrays (mGy) and is considered safe in pregnancy [26]. Liver ultrasound can safely screen for liver metastases, but may be less accurate if the patient has developed fatty infiltration of the liver during pregnancy. Computed tomography (CT) scans should not be routinely used. MRI is more commonly utilized for screening for liver metastases, but again its safety in pregnancy has not been fully evaluated. Radionucleotide bone scans can be safely performed during pregnancy with fetal exposure to 1.94 mGy of radiation. However, they require adequate hydration and placement of an indwelling bladder catheter for 8 h to prevent retention of radioisotope in the bladder [27]. MRI screening, noncontrast of the thoracic/lumbar spine may be useful in patients with signs or symptoms of bone disease [9].
Pathologic features
Breast cancer in patients with PABC is histologically similar to that in nonpregnant patients. Ductal carcinomas occur in 75–90% of patients [9]. Previous studies reported a decreased incidence of estrogen receptor (ER) and progesterone receptor positivity in tumors of patients with PABC. Bonnier and colleagues compared the tumors of 75 patients with PABC with those of 182 nonpregnant patients with breast cancer [21]. The group with PABC had 42% ER-negative cancers while the control group had 21% ER-negative disease. These results may be related to the ligand-binding assay methodology used to determine hormone receptor status [28], which could lead to false-negative results in those with higher levels of circulating estrogen and progesterone downregulating receptors. Elledge and colleagues compared the ligand-binding assay technique with immunohistochemistry staining (IHC) in the breast cancers of pregnant patients [28]. They did not find a difference in the numbers of hormone receptor-positive patients with PABC when matched with controls. In Bonnier's paper, 75% of the tumors were evaluated by ligand-binding assay [21]. When IHC was the method utilized for determination of hormone receptor status, there was no difference between the patients with PABC and their controls. Middleton and colleagues reported 28% ER-receptor positivity by IHC [22]. Further studies using IHC in larger numbers of patients would be helpful in clarifying the hormone receptor status for this group. Middleton and colleagues found that 28% of the tumors of patients with PABC stained positive for HER-2/neu [22], while Elledge and colleagues found that 58% stained positive for HER-2/neu [28]. Further studies are needed.
Treatment
Management of patients with PABC should involve a multidisciplinary team of medical oncologists, surgical oncologists, radiation oncologists and obstetricians. They should work in close conjunction to ensure the optimum outcome for the patient and the child. The patient needs to have obstetric visits with routine high-risk obstetric care, including serial fetal ultrasounds as clinically indicated by the obstetrician [29]. Timing of delivery is also important. Chemotherapy administration should be avoided in the 3 weeks prior to anticipated delivery to avoid myelosuppression and risk of infection [30]. Abortion has not been shown to improve survival [31] and should not be considered standard practice for pregnant patients diagnosed with breast cancer.
Local treatment
Breast surgery can be performed at any time during pregnancy and is considered to pose little or no harm to the fetus [32]. The surgical management of PABC is dictated primarily by the gestational age of the pregnancy at time of diagnosis. Other considerations, such as patient preference and the size and location of the tumor, influence the overall management of the patient. Modified radical mastectomy with axillary lymph node dissection has been the standard treatment for patients with PABC [9]. In many situations it circumvents the need for radiation therapy, which is contraindicated in pregnancy. The typical dose of radiation for breast radiotherapy is 5000 cGy, which would expose the fetus to between 10 cGy early in pregnancy and 200 cGy or more later in pregnancy [33]. This would pose a risk of fetal toxicity.
For this reason, pregnancy was considered to be a contraindication for breast conservation therapy (lumpectomy and axillary lymph node dissection followed by radiation therapy). However, patients who are diagnosed late in pregnancy can undergo lumpectomy with radiation following delivery. Women who are diagnosed at an advanced stage might benefit from the use of neoadjuvant chemotherapy, which would allow for lumpectomy later in pregnancy followed by radiation postpartum. Kuerer and colleagues reported four cases of women with PABC who underwent breast-conserving therapy and chemotherapy [34]. Three of the women received neoadjuvant chemotherapy during pregnancy, followed by surgery and radiation postpartum. They were all without evidence of disease at follow-up of 44 months.
Sentinel lymph node biopsy is routinely performed in nonpregnant patients with early stage breast cancer. Its safety in pregnant patients has not been established, and therefore it is not recommended for patients with PABC [35]. The two methods for sentinel node mapping include 99mTc radioactive isotope or isosulfan blue dye. Gentilini and colleagues recently evaluated the potential risk to the fetus during lymphoscintigraphy using 99mTc radioactive isotope by measuring the radioactivity in 26 premenopausal nonpregnant patients with breast cancer undergoing lymphoscintigraphy [36]. They found the radioisotope localizing in the area of injection. Sites along the surface of the abdomen exhibited doses lower than the sensitivity of the dosimeter in 23 of 26 patients, with the other three patients with levels less than 320 μGy. Further studies will augment these data and may allow a change in the current recommendations.
Chemotherapy
The indication for chemotherapy in patients with PABC is the same as for nonpregnant patients. As many patients with PABC present at a more advanced disease stage, the decision of chemotherapy during pregnancy must be carefully discussed with the patient. There are no randomized clinical trials addressing the use of cytotoxic agents in PABC [31]. Most of the information is from case reports and retrospective case-control studies. Berry and colleagues published the results of an ongoing prospective standardized protocol from the MD Anderson Cancer Center [29]. A recent review of chemotherapy during pregnancy analyzed 72 cases of PABC [37]. While this information is limited, an increasing number of women have been exposed to chemotherapy during pregnancy for a variety of cancer treatments [37].
The risk of cytotoxic agents is greatest during the first trimester, which is the period of organogenesis [37]. Fetuses exposed to cytotoxic agents in the first 12 weeks have a higher risk of spontaneous abortion, fetal death and malformations. A review of 217 cases of pregnant women treated with chemotherapeutic agents for a variety of diseases reported the majority of fetal abnormalities occurring in pregnancies exposed during the first trimester [38]. A review of 139 pregnant patients treated with cytotoxic therapy during the first trimester reported a 17% incidence of fetal malformations [39]. In the same review, patients treated in the second and third trimesters had a 1.3% incidence of fetal malformations. Therefore, chemotherapy administration for breast cancer treatment during the first trimester is contraindicated [37]. The use of methotrexate, a component of the regimen of a combination of three chemotherapy drugs: cyclophosphamide, methotrexate and 5-fluorouracil (CMF), is contraindicated throughout pregnancy, as it is an abortifacient [9].
Chemotherapeutic regimens containing doxorubicin and cyclophosphamide with or without 5-fluorouracil are commonly used in the treatment of breast cancer. Berry and colleagues reported 24 patients treated with the regimen of 5-fluorouracil, doxorubicin and cyclophosphamide (FAC), all after 12 weeks gestation [29]. There were no adverse antepartum maternal complications directly attributable to treatment. One patient was hospitalized for diarrhea and suspected pyelonephritis and another was hospitalized for intravenous heparin therapy for deep venous thrombosis with a history of the same. Three patients experienced preterm labor and another experienced severe pre-eclampsia. The median gestational age at delivery was 38 weeks and the neonates experienced no unusual complications. One neonate delivered 2 days after chemotherapy administration and developed transient leukopenia that resolved without medical intervention. Only one neonate graphed below the tenth percentile for birthweight. Other retrospective studies support the safety of FAC or doxorubicin with cyclophosphamide (AC) in pregnancy during the second and third trimesters [37]. Germann and colleagues reviewed the literature involving 160 patients treated with anthracyclines during pregnancy between 1976 and 2001 [40]. A total of 72% of patients had a diagnosis of a hematologic malignancy while 21% had breast cancer. Doxorubicin or daunorubicin were given to the majority of patients (93%) and 90% of patients received anthracyclines in combination with other chemotherapeutic agents. Administration of anthracyclines was found to be more toxic when given during the first trimester and in doses greater than 70 mg/m2/cycle. In patients treated for solid tumors, two severe events out of five patients were associated with treatment during the first trimester when compared with one event out of 39 patients treated during the second and third trimesters. A total of five malformations were reported, three of which occurred after treatment during the first trimester.
Less information is available for epirubicin and idarubicin. In a recent review by Cardonick and Iacobucci [37], four fetal deaths and two cases of transient cardiomyopathy occurred among 17 cases evaluated. A recent letter by Peccatori and colleagues disputed the information regarding epirubicin [41]. They reported giving nine patients with PABC epirubicin after 15 weeks of gestation. Six patients were given a once-weekly dose of epirubicin while the other three were given epirubicin once every 3 weeks. Although no data are available, they report no severe maternal or fetal complications. One newborn was treated for congenital vescicoureteral reflux. More information regarding the treatment of PABC with epirubicin is required.
Taxanes are widely used in nonpregnant patients with node-positive breast cancer. There are case reports of their use in pregnancy without adverse fetal outcomes [42,43], but further information needs to be collected. The routine use of taxanes in the treatment of PABC cannot be recommended given the limited data available.
The US Food and Drug Administration labels trastuzumab pregnancy class B because it was tested in pregnant animals without significant risk to the fetuses [38]. Only two case reports of the use of trastuzumab during pregnancy are available in the literature [44,45]. Watson published a case report of a woman who had been taking trastuzumab for 5 months when she was found to be 23 weeks pregnant [44]. By ultrasound the fetus appeared normal except for a lack of amniotic fluid. Trastuzumab was discontinued and the amniotic fluid index slowly increased. The patient delivered at 37 weeks and the baby was doing well at 6-month follow up with no evidence of pulmonary hypoplasia. Fanale and colleagues reported a patient who presented with metastatic breast cancer at 27 weeks gestation [45]. She had previously received treatment with paclitaxel followed by 5-fluorouracil, epirubicin and cyclophosphamide with progression of disease. She then underwent modified radical mastectomy and radiation therapy to the chest wall. Fourteen months after her initial diagnosis, and at 27 weeks gestation, the patient presented with right upper quadrant abdominal pain. She was diagnosed with liver metastases and began treatment with trastuzumab and vinorelbine. Amniotic fluids indexes were noted to be low and the patient was diagnosed with oligohydramnios. Additional intravenous fluids were given to the patient with each administration of trastuzumab and vinorelbine. However, at 34–37 weeks gestation the patient underwent induction of labor for decreased fetal movement. No neonatal complications were noted and the infant is reported to be healthy, with normal development. Of note, both patients developed oligohydramnios during administration of trastuzumab. Further study is warranted to determine the safety of trastuzumab during pregnancy and the need for additional intravenous fluids.
Based on prospective data, the current recommendation for the treatment of PABC is to use a regimen containing AC or FAC after the first trimester. There are no guidelines or standardized treatments for PABC. A central database for information regarding treatment of PABC would enhance data available for further recommendations. Future prospective studies would also be beneficial.
Endocrine therapy
Use of tamoxifen citrate during pregnancy is not recommended owing to potential teratogenic effects [46]. This recommendation is based on animal models, as there are no clinical trial data available. One fetus received tamoxifen throughout all three trimesters and was born with Goldenhar syndrome (oculoauriculovertebral dysplasia) at 26 weeks [9]. A direct causal link cannot be established; however, as the fetus might have also had other exposures. Another female fetus exposed to tamoxifen throughout the first 20 weeks was born with ambiguous genitalia. A third report discussed a patient given tamoxifen as treatment for metastatic disease who did not want to change her regimen [46]. Her son was delivered at 31 weeks and was meeting all developmental milestones at follow up of 2 years. Female fetuses of rats and mice given tamoxifen during pregnancy have demonstrated changes in the epithelium of the uterus and reproductive tract [37]. Since there is no definitive human data and the majority of PABC tumors are receptor-negative, at this time tamoxifen use is not recommended during pregnancy.
Outcomes
Initial reports of the prognosis of patients with PABC were grim [9]. However, many studies have shown that when matched for stage and age, women with PABC have the same outcome as nonpregnant patients. Petrek and colleagues compared 56 patients with PABC to age- and stage-matched controls [47]. They found no statistically significant difference in survival. In patients with PABC 5-year survival rates were 82% for node-negative disease and 47% in patients with node-positive disease. It is notable that more patients with PABC presented with node-positive disease, 62% compared with 39% in the nonpregnant group. Other studies corroborate these results [35].
Children born to mothers receiving chemotherapy typically have normal growth and development. In a prospective study by Berry and colleagues, there were no developmental abnormalities in the 24 children born to mothers who received FAC chemotherapy for PABC [29]. The median age at follow-up was 4.5 years. Other retrospective studies corroborate these findings [48]. A central registry is under development by the Motherisk Program in Canada. The hope is to document the long-term effects or complications arising in children exposed to chemotherapy in utero [102].
Future pregnancies
Patients with PABC are typically young and may be interested in having more children in the future. The question arises as to the safety of future pregnancies. Velentgas and colleagues evaluated the outcomes of 53 patients who became pregnant after a diagnosis of breast cancer [49]. They first compared the rate of miscarriage in these patients to matched controls who did not have a history of breast cancer. A total of 68% of the women who became pregnant after breast cancer gave birth to at least one child. The miscarriage rate was 24% in the breast cancer group versus 18% in the control group. No explanation for this difference is known at this time. They then compared overall survival in these patients with matched controls who previously had a diagnosis of breast cancer but did not have subsequent pregnancy. The relative risk for mortality in women with subsequent pregnancy was 0.8. These results, although in one small study, may even suggest that pregnancy after breast cancer has a beneficial effect on survival. Mueller and colleagues compared 438 women with breast cancer with births after diagnosis with 2775 matched women with breast cancer without births after diagnosis [25]. Women with live births 10 months or more after a diagnosis of breast cancer had a decreased risk of death when compared with women without subsequent births (relative risk 0.54). Women with live births within 10 months after a diagnosis of breast cancer do not have a statistically significant difference in survival when compared with controls. Other studies by von Schoultz and colleagues and Gelber and colleagues agree with these findings [51,52], but a prospective study is needed for further clarification. Many oncologists recommend waiting at least 2 years after completing therapy to become pregnant, but this is not based on clinical evidence [9].
Conclusion
Treatment of breast cancer during pregnancy involves a well co-ordinated multidisciplinary approach. Management needs to be determined for each individual patient according to their age of gestation at diagnosis and stage of cancer at presentation (Table 1). Surgery can be performed safely at any time during pregnancy. Patients diagnosed in the first trimester are recommended to have modified radical mastectomy for local treatment and then delay further chemotherapy until the second or third trimester. This will circumvent the need for radiation therapy, which is contraindicated in pregnancy. For patients diagnosed in the second or third trimester, they have more flexibility in timing of surgery. Women qualifying for breast conservation are able to have either neoadjuvant or adjuvant chemotherapy, with radiation following delivery.
Recommended treatment for pregnancy-associated breast cancer (PABC).
AC: Doxorubicin and cyclophosphamide therapy; FAC: 5-fluorouracil, doxorubicin and cyclophosphamide therapy.
Women with PABC typically present at a more advanced stage than their nonpregnant counterparts. However, when matched by stage, the prognosis is not worsened by pregnancy. Based on prospective trials, FAC chemotherapy is safe in pregnancy. Epirubicin, taxanes and trastuzumab have all been reported in pregnancy with good outcomes, but no prospective trial is available. Most children born after exposure to chemotherapy reach developmental milestones and do not have abnormalities. Future pregnancies are optional for women after treatment of their breast cancer, and should be planned in conjunction with follow up.
Future perspective
The multimodality treatment of breast cancer needs to be adopted as the universal standard of care for patients with PABC. As more patients are diagnosed and more clinicians are exposed to the care of patients with PABC, the level of confidence in treatment will increase. Central databases are essential in gathering more information regarding treatment of patients in this population. As new treatments become available, they need to be specifically tested in pregnant animal models and then offered to pregnant patients if they are found to be potentially safe. Prospective clinical trials would allow for the standardization of therapies.
Executive summary
Pregnancy-associated breast cancer (PABC) is breast cancer diagnosed during pregnancy or within the first 12 months following delivery.
1 in 3000 to 3 in 10,000 deliveries are associated with PABC.
Approximately 0.2–3.8% of breast cancer cases each year.
Incidence is rising as women delay childbirth.
Painless mass or thickening.
Delayed presentation usually secondary to changes during pregnancy and lactation.
All breast masses should be assessed by clinical exam, diagnostic imaging and biopsy.
Mammography with abdominal shielding, ultrasound or magnetic resonance imaging (MRI) should be used for evaluation of breast mass.
Biopsy including fine needle aspiration (FNA), core biopsy, excisional or incisional biopsy should be carried out for definitive diagnosis.
Physical exam.
Lymph node evaluation with ultrasound and FNA.
Chest x-ray with abdominal shielding.
Liver evaluation by ultrasound or MRI.
Bone scan is safe during pregnancy, but MRI may be used to evaluate symptomatic lesions.
Ductal carcinomas in 75–90% of cases.
Incidence of estrogen and progesterone receptor-negative tumors is controversial at this time. It is thought to be similar to nonpregnant patients when measured using immunohistochemistry.
Breast surgery can be performed at any time during pregnancy.
Radiation therapy is contraindicated in pregnancy.
Patients diagnosed in the first trimester are recommended to have modified radical mastectomy.
Patients diagnosed in the second and third trimester may qualify for breast conservation surgery with radiation deferred until after delivery.
The use of sentinel node biopsy during pregnancy is controversial and requires more studies.
Chemotherapy is contraindicated during the first trimester of pregnancy secondary to a higher rate of fetal malformations (17% in the first trimester vs 1.3% in the second and third trimester).
Multiple retrospective studies have concluded the safety of chemotherapy after the first trimester.
Only one prospective study regarding chemotherapy in pregnancy has been published. Berry and colleagues published a study of 24 patients treated with 5-fluorouracil, doxorubicin and cyclophosphamide (FAC) after the first trimester and there were no adverse antepartum or neonatal complications directly related to therapy.
FAC (or doxorubicin and cyclophosphamide [AC]) chemotherapy is safe during pregnancy and is the recommended chemotherapy for pregnant patients.
Other chemotherapeutic agents have been used during pregnancy (taxanes, epirubicin, trastuzumab) but no prospective trials are available.
Tamoxifen use during pregnancy is not recommended secondary to a lack of data.
When matched for stage and age, women with PABC have the same outcome as nonpregnant patients.
Children born to mothers receiving chemotherapy typically have normal growth and development.
Future pregnancies are safe and do not have an adverse effect on the health of the mother. Many oncologists recommend waiting at least 2 years after completing therapy to become pregnant.