Cancer in Pregnancy

Malaika Amneus and Christine H. Holschneider

The diagnosis of cancer complicates approximately 1 in 1000 pregnancies. Once diagnosed, emotional, ethical, diagnostic, and treatment dilemmas confront both the patient and the treating physicians, posing unique challenges. Questions regarding whether or not to terminate the pregnancy, potential maternal risk of delays in cancer treatment, fetal risks of early delivery, and maternal and fetal effects of cancer treatments during gestation are complex and competing factors that make decision making both medically and emotionally challenging. Limited data on the treatment of malignancies during pregnancy and absence of randomized controlled studies in this population contribute to the lack of generalized treatment algorithms. Individualization of treatment planning with a multidisciplinary team is essential. Considerations not only include the risk/benefit assessment of treatment modalities such as chemotherapy, radiation therapy, and surgery during pregnancy, but also include the potential maternal and fetal consequences of diagnostic procedures.

The most common malignancies during pregnancy include those that are most commonly found in women of reproductive age and include cervical and ovarian cancer. The most common nongynecologic malignancies are breast cancer, malignant melanoma, thyroid cancer, and hematologic malignancies.^1–3 Given that the incidence of malignancies increases with increasing age, as more women choose to delay childbearing, it is expected that the incidence of cancer during pregnancy will increase.

RADIATION IN PREGNANCY

Key Points

1. The developmental effects of radiation exposure on pregnancy is related both to the dose of radiation as well as the gestational age.

2. Computed tomography and magnetic resonance imaging, including scans of the pelvis, are associated with negligible fetal risk.

Ionizing radiation is used routinely during imaging for cancer staging or disease surveillance, and radiation therapy is a common component of the treatment of many cancers in the nonpregnant patient. In pregnancy, consideration has to be given not only to the radiation exposure of the mother, but also that of the developing fetus. Much of what we know about the effects of radiation on pregnancy is based on animal studies, accidental or incidental human exposures to diagnostic and therapeutic radiation, and data gathered from victims of radiation exposure after the atomic bombings of Hiroshima and Nagasaki. There are many confounding factors that limit our interpretation of these data, including species differences, potential differential effects of various types of ionizing radiation, lack of certainty regarding the doses of radiation received, potential differential effects of single versus multiple exposures, and the baseline rate of human malformations and other negative outcomes. Given the lack of controlled studies on the issue, patient counseling regarding radiation exposure in pregnancy may be challenging, even as it relates to imaging procedures. The growing importance of this issue is evidenced by a recent review that noted a 107% increase over the past decade in the use of imaging studies using ionizing radiation during pregnancy.⁴

The developmental effects of radiation exposure on a pregnancy are related not only to the dose of radiation, but also to the gestational age of the pregnancy. Gestation can be generally divided into 3 periods: preimplantation (0-2 weeks after conception), organogenesis (2-8 weeks after conception), and the fetal period (> 8 weeks after conception). Radiation exposure has different potential developmental effects during each of these periods (Table 16-1).^5,6 In addition, radiation exposure can lead to genetic cell injury and an increased malignancy risk at any gestational age.

Table 16-1 Developmental Effects of Ionizing Radiation at Different Gestational Ages^7,8

During the preimplantation period, radiation exposure is thought to have an “all or none” effect; that is, the embryo dies, or there is no consequence.⁷ This may be explained by the fact that at this early stage of development, the embryo is composed of totipotent cells. If the radiation causes the death of a sufficient number of these cells, the embryo will not survive. Otherwise, the remaining cells will continue to divide and the insult will be overcome without further sequelae. Exposure to 10 cGy (10 rads) or more is generally thought to result in pregnancy loss during the preimplantation period.⁸

During the period of organogenesis, the fetus is most susceptible to the teratogenic effects of radiation exposure. Notable effects include microcephaly, microphthalmia, eye anomalies, mental retardation, genital malformations, and growth restriction.^6,9 It is the current consensus that exposure to < 5 cGy (< 5 rads) of radiation is not related to an increased risk of fetal malformation,⁸ and it is important to note that no single common diagnostic imaging procedure in use today results in exposures at this level (Table 16-2). However, such levels can be quickly reached with multiple imaging studies, therapeutic radiation, and fluoroscopic procedures. Although concerns about exposure in the range of 5 to 10 cGy (5-10 rads) have been raised, significantly increased developmental risk to the embryo and fetus is not known until the absorbed dose reaches at least 10 cGy (10 rads).

Table 16-2 Estimated Fetal Radiation Exposure From Common Diagnostic Radiologic Procedures^10,11

During the fetal period, the effects of radiation exposure seem to be limited to growth retardation, microcephaly, and central nervous system (CNS) defects (decreased IQ and mental retardation).^5,6 The period from the 8th to 15th weeks appears to be the period of greatest vulnerability of the CNS, and it is estimated that the probability of mental retardation is approximately 40% for every 100 cGy (100 rads) exposure above 10 cGy (10 rads).¹² The sensitivity of the CNS to radiation is less during the 16th to 25th weeks, and studies have estimated a 13- to 21-point reduction in IQ for every 100 cGy (100 rads) of radiation exposure above 10 cGy (10 rads).⁵ The radiation sensitivity of the CNS is even further reduced after 25 weeks.

In addition to the potential developmental effects of radiation exposure, there is also concern for an increased risk of childhood cancers (both leukemias as well as solid tumors) in children exposed to radiation in utero. The relative risk for childhood cancer associated with in utero radiation exposure, as found in the large Oxford Survey of Childhood Cancers as well as in a 1993 meta-analysis, is approximately 1.4.¹³ The excess childhood cancer risk attributable to in utero radiation exposure is estimated to be 6% per 100 cGy, with an increase in risk starting to become appreciable at doses as low as 1 cGy.¹³ Although it may sound alarming that a standard pelvic CT scan may potentially double the risk of fatal childhood cancer, this must be viewed in context of the low baseline risk (approximately 1 in 2000)⁹ and the potential benefit to the management of mother and pregnancy of the information gained by the imaging study. One significant difficulty in interpreting this literature is due to the fact that much of the data stem from case-control studies collected over many decades, during and after which radiation exposure with diagnostic imaging has been greatly reduced. In fact, recent population-based data from Ontario using 1.8 million maternal-child pairs from 1991 to 2008 found no increase in childhood cancers in the offspring of 5590 mothers exposed to major radiodiagnostic testing in pregnancy (crude hazard ratio of 0.69; 95% confidence interval, 0.26-1.82).¹⁴

Computed Tomography in Pregnancy

It is generally agreed that imaging modalities that do not involve ionizing radiation, such as ultrasound and magnetic resonance imaging (MRI), are preferred in pregnancy as long as they adequately provide the desired diagnostic information. However, as demonstrated in Table 16-1, the radiation dose to the fetus from a computed tomography (CT) scan, even of the pelvis, is below a dose expected to be associated with considerable fetal risk. It is thus unacceptable to delay diagnostic work-up of a pregnant patient with suspected cancer if the information gained from the imaging study is expected to affect management. In addition, such exposure per se should not alter the management of the pregnancy. Despite this, a survey of physicians published in 2004 indicated that up to 6% would recommend pregnancy termination after an abdominal CT scan in early pregnancy,¹⁵ highlighting the need for provider education. In addition to the potential effects of exposure to ionizing radiation during a CT scan discussed previously, there may also be concerns about the safety of intravenous (IV) contrast administration during pregnancy. Iodinated contrast media do not appear to be teratogenic in animals. Human data are lacking, and thus they are considered US Food and Drug Administration (FDA) category B drugs. The use of iodinated contrast agents is considered acceptable in human pregnancy when it is necessary for appropriate diagnosis and after informed consent.¹⁶ There is a theoretical risk of depression of neonatal thyroid function with in utero exposure to iodinated agents. However, there are no clear data that demonstrate any considerable increased risk associated with contrast exposure for CT scan during pregnancy. The thyroid function of neonates is routinely assessed in the United States, regardless of any in utero exposure to iodinated agents.

Magnetic Resonance Imaging in Pregnancy

MRI during pregnancy has been used for both maternal and fetal indications, and there is no evidence that human fetal exposure to MRI with 1.5 Tesla magnets has resulted in any negative fetal effects. The 2007 American College of Radiology Guidance Document for Safe MR Practices states that it is acceptable to perform magnetic resonance scans at any time point during pregnancy when the radiologist and referring physician believe that the risk-benefit ratio warrants performance of the study.¹⁷ Some theoretical concerns involve potential teratogenicity and acoustic damage. However, thus far, studies have failed to substantiate these concerns. A study of 35 children who were exposed to 1.5-Tesla MRI during the third trimester of pregnancy found no harmful effects attributable to MRI, in particular no negative effects on vision or hearing.¹⁸ Normal pediatric assessment and developmental outcomes were found in a study of 20 9-month-old infants who were exposed in utero to 4 series of echoplanar imaging MRI between 20 weeks and term.¹⁹

The use of gadolinium-based contrast is controversial. Animal studies have shown potential teratogenic effects at high doses.⁶ Studies show that the contrast agents cross the placenta, are filtered by fetal kidneys, and appear in fetal urine. Because they are excreted into the amniotic fluid, there are concerns regarding potential long-term exposure of the fetus due to recirculation and delayed elimination. The American College of Radiology recommends that before administration of gadolinium-based contrast agents in pregnancy, “a well-documented and thoughtful risk-benefit analysis…” be performed that substantiates an “overwhelming potential benefit to the patient or fetus…”,.¹⁷ However, the US FDA classifies gadolinium as a class C drug, and the European Society of Radiology states that based on available evidence, the use of gadolinium in pregnancy appears to be safe.²⁰ Thus, even though current radiology practices and recommendations in the United States discourage the use of gadolinium-based contrast agents during pregnancy because their safety for the fetus has not yet been proven, gadolinium use should be considered when the diagnostic study is important for the health of the mother.

Therapeutic Radiation in Pregnancy

Although the dose of fetal radiation for most diagnostic procedures is low, administration of therapeutic radiation during pregnancy can result in significant fetal doses. Therapeutic radiation to sites remote from the uterus may be indicated during pregnancy for some cancers and can be administered to a well-informed patient. The dose of radiation administered to the fetus depends on several factors, including the target dose of radiation, the size of the radiation field, the type of teletherapy machine used, the distance of the fetus from the edge of the radiation field, and the use of wedges, blocks, and other objects that cause scatter.²¹ In addition to the scatter from the teletherapy machine and from beam modifiers, internal scatter within the patient also affects the dose received by the fetus. The use of proper shielding of the uterus can reduce the fetal dose by up to a factor of 4.^22,23 Cobalt-60 irradiation is associated with a higher fetal dose compared with high-energy photons. Additionally, the use of high-energy photon beams > 10 MV produce a photoneutron contribution to the radiation dose, and it is generally recommended that photons < 10 MV be used whenever possible.

Woo et al²⁴ reported on 16 patients with Hodgkin lymphoma who were treated with radiation during pregnancy. Eleven patients received mantle radiation, 3 received radiation to the neck and mediastinum, and in 2 patients, radiation was limited to the neck. The dose to the mid-fetus was estimated in 9 cases and ranged from 1.4 to 5.5 cGy with 6-MV photons and from 10 to 13.6 cGy for cobalt-60. All patients went on to delivery healthy infants. Antypas et al²⁵ performed in vivo as well as phantom measurements of the fetal dose of radiation in a patient irradiated for breast cancer from the second to sixth week of gestation. There was no shielding used. The total tumor dose was 46 Gy, and the fetal dose was estimated at 3.9 cGy. The importance of gestational age and uterine size for the fetal dose of radiation is exemplified when these measurements are compared with those obtained by Ngu et al²⁶; breast irradiation to 50 Gy in late pregnancy resulted in an estimated 21-cGy unshielded dose, which reduced to 14 to 18 cGy with shielding.

If a patient presents with an unplanned pregnancy during radiation therapy for cancer, the risk of poor fetal outcome after fetal radiation exposure is dependent on the fetal dose received. The radiation oncologist and physicist involved in the patient’s radiotherapy should be asked to estimate fetal radiation exposure. The results of such evaluation should be discussed with the patient and her family to allow her to make an informed decision regarding her pregnancy in the context of her underlying disease and other concurrent therapy received or planned.

CHEMOTHERAPY IN PREGNANCY

Key Points

1. Chemotherapy exposure in the first trimester is associated with a 20% incidence of fetal malformations, which exceeds the 3% to 4% rate of fetal malformations in the general population.

2. In the second and third trimester, the most significant potential negative effects of chemotherapy during pregnancy are growth restriction and pre-term birth.

To patients, the prospect of undergoing chemotherapy is often accompanied by apprehension and fear, both in relation to the anticipated physical and emotional effects of treatment, as well as in regard to whether or not the treatment will be effective. When a pregnant patient faces chemotherapy, these concerns are compounded by concerns about the effect of the treatment on the pregnancy and the developing fetus. Several factors, including the chemotherapeutic agent, placental transfer, timing of treatment, dose, and frequency of exposure, influence the effects of chemotherapy on the fetus. Again, there is a lack of controlled data in humans, and thus our knowledge of the effects of chemotherapy during pregnancy are limited to animal studies and human case series and case reports. Depending on the type of malignancy, the stage of disease, and the gestational age of the pregnancy, there are times when appropriate therapy can be delayed until after delivery, or other modalities of treatment, such as surgery, can be used during the pregnancy with adjuvant therapy delayed until after delivery. However, when chemotherapy is likely to improve maternal outcome, it can be administered during pregnancy to an appropriately informed patient.

The concepts regarding the importance of gestational age on the effects of chemotherapy on a developing pregnancy are similar to those in radiation. Chemotherapy within the first 2 weeks after conception appears to have an “all or none” effect, resulting in either spontaneous abortion or no effect. The most sensitive time for potential teratogenesis is during organogenesis. The organs that continue to develop throughout gestation, such as the nervous system, eyes, and bone marrow, remain at risk even after this window. In general, chemotherapy exposure in the first trimester is associated with a 20% incidence of fetal malformations, which exceeds the 3% to 4% rate of fetal malformations in the general population.

In the second and third trimester, the most significant potential negative effects of chemotherapy during pregnancy are growth restriction and preterm birth. A review of published cases of in utero exposure to chemotherapy identified a less than 4% incidence of malformations, a 7% incidence of intrauterine growth restriction, and a 5% incidence of spontaneous preterm delivery. Of the 11 cases of malformation, 9 were associated with first trimester exposure.²⁷ The most recent large report comes from the Cancer and Pregnancy Registry and includes data from 231 women diagnosed with a variety of cancers during pregnancy who were voluntarily reported to the registry. Of these patients, 13 terminated their pregnancy and 157 received chemotherapy during pregnancy. Of the infants exposed to chemotherapy in utero, 3.8% had a malformation. Fewer than 8% of infants had a birthweight less than the 10th percentile, and 6% spontaneously delivered prematurely (iatrogenic preterm delivery excluded).²⁸ The rates of malformation, intrauterine growth restriction and spontaneous preterm delivery in these studies do not exceed those seen in the general population. Although these data are encouraging and can be used for the counseling of patients, some uncertainty continues as a result of the lack of long-term follow-up. The incidence of clear cell carcinoma of the cervix and vagina in women exposed to diethylstilbestrol in utero is a reminder of the importance of continued long-term follow-up.

Although the timing of delivery is not always predictable for obstetrical reasons, it is generally recommended to avoid delivery during the peak of the hematologic toxicity of chemotherapy treatment in the mother. An additional consideration is the potential for neonatal myelosuppression. For example, in a study on the treatment of leukemia during pregnancy, one-third of the infants exposed to chemotherapy within 1 month before delivery were cytopenic at birth.²⁹ Due to this potential complication and the resultant potential risk for neonatal sepsis or bleeding, it is typically recommended that administration of chemotherapy be avoided within 3 weeks of anticipated delivery. This also allows for fetal drug excretion via the placenta, as drug metabolism in the neonate may be impaired, leading to potential increased toxicity.

In addition to the concerns regarding potential consequences of chemotherapy exposure for the fetus, careful consideration must be given to the physiologic changes that accompany the pregnant state and how those may affect the dosing and toxicity of chemotherapy. Such physiologic changes include an increased plasma volume, an increased glomerular filtration rate, decreased serum albumin, delayed gastric emptying and reduced intestinal motility, alterations in hepatic metabolism, and the creation of a physiologic third space (amniotic fluid). These changes may significantly affect the pharmacokinetics of drugs due to increased volume of distribution, altered protein binding, increased renal clearance, altered drug absorption, enterohepatic circulation, and hepatic clearance, with subsequent effects on peak drug concentration, drug half-life, and area under the curve (Table 16-3). Due to the paucity of data, chemotherapy during pregnancy is usually administered without dose modification compared with non-pregnant patients. Toxicities, response to treatment, and fetal well-being must be carefully followed throughout treatment and adjustments to treatment regimens and supportive care made when clinically indicated.

Table 16-3 Pharmacokinetic Consideration Regarding Chemotherapeutic Drugs in Pregnancy^30,31

Chemotherapeutic Agents in Pregnancy

Information regarding the use of specific chemo-therapeutic agents during pregnancy is limited largely to case reports, small case series, and some registry reports of single-agent or combination therapy use. In addition to the FDA classification of drugs in pregnancy (Table 16-4), several resources can provide information to guide clinicians in the administration of chemotherapy in this setting. These include textbooks such as Drugs in Pregnancy and Lactation: A Reference Guide to Fetal and Neonatal Risk (Briggs, Freeman, and Yaffe, now in the 9th edition), as well as internet-based resources such as Reprotox (www.reprotox.org), TERIS (depts.washington.edu/terisweb), and the Cancer and Pregnancy Registry (www.cooperhealth.org/content/pregnancyandcancer.htm).

Table 16-4 United States Food and Drug Administration (FDA) Classification of Fetal Risks due to Pharmaceuticals³⁶

Antimetabolites

Antimetabolites, especially folic acid antagonists, are the agents most commonly associated with fetal malformations. First-trimester exposure to aminopterin is associated with a series of abnormalities that includes cranial dysostosis, hypertelorism, abnormalities of the external ear, and micrognathia, as well as potential limb deformities and neurologic abnormalities. Methotrexate exposure in the first trimester has resulted in similar anomalies, and methotrexate is considered one of the most teratogenic medications and is classified as a class X drug by the FDA. These findings have been termed aminopterin-methotrexate syndrome. Of 6 first-trimester exposures to cytarabine (alone or in combination with other chemotherapeutic agents), there were 2 congenital abnormalities.³² Mercaptopurine appears to be associated with a low rate of congenital malformations, with 1 abnormality noted of 34 cases of first-trimester exposure.³³ A recent series of breast cancer patients treated with chemotherapy during pregnancy included 9 patients who received combination therapy with cyclophosphamide, methotrexate, and fluorouracil (none in the first trimester), with no adverse outcomes noted.³⁴

Alkylating Agents

A review of use of alkylating agents during pregnancy found a 14% incidence of malformation with administration in the first trimester, but when treatment was limited to the second and third trimester, the rate of malformations was not above that of the general population.³⁵ Reported malformations after administration of cyclophosphamide during the first trimester of pregnancy have included absent or hypoplastic digits, eye abnormalities, cleft palate, flat nasal bridge, and many other abnormalities.^27,33 An interesting case report involves a diamniotic-dichorionic twin pregnancy treated with cyclophosphamide throughout pregnancy for acute lymphocytic leukemia. One twin (female) was without abnormalities, whereas the other (male) had multiple abnormalities including but not limited to esophageal atresia, an upper extremity deformity, and a renal abnormality. He also went on to develop thyroid cancer at 11 years of age and neuroblastoma at 14 years of age.³⁷ Chlorambucil administration during the first trimester is associated with renal agenesis in both animals and humans.³³ Dacarbazine is used in the treatment of Hodgkin lymphoma and has been administered during pregnancy as part of the doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) regimen. In one review, no malformations were noted in 15 pregnancies (some included first-trimester use).³⁸

Antibiotics

Anthracyclines incompletely cross the placenta, and their use in pregnancy has not been associated with an increase in fetal malformations. A recent review of the literature analyzed outcomes in 160 pregnancies exposed to anthracyclines (90% in combination regimens). This included 31 patients exposed in the first trimester. They noted a malformation in 3% and fetal death in 9%. Eighty percent of the malformations were associated with first-trimester exposure. Fetal death was in conjunction with maternal death in 40% of the cases. Of the fetal deaths, 87% were in patients with acute leukemia, and 73% were associated with daunorubicin exposure, leading many to now avoid this agent in pregnancy in favor of doxorubicin.³⁹ An additional concern with the use of anthracyclines in pregnancy is the potential for cardiac toxicity in children exposed to the agents in utero. A recent report with long-term outcomes (mean, 17 years) of cardiac function of such children was encouraging, finding no evidence of late cardiac toxicity.⁴⁰ Bleomycin use in pregnancy is somewhat more controversial. At least 9 cases of bleomycin, etoposide, and cisplatin (BEP) use in pregnancy have been reported; in 2 of those poor neonatal outcomes that were noted, 1 included ventriculomegaly and cerebral atrophy.⁴¹

Vinca Alkaloids

Vinca alkaloids are embryocidal and teratogenic in animal studies, and scattered reports suggest potential malformations after administration of these drugs.^30,33,38 However, several case reports also suggest the relative safety of these agents in human pregnancy.³⁰ For example, in a review of 6 cases of vinorelbine exposure during pregnancy (none in the first trimester), there were no malformations noted, and short-term outcomes of the infants were positive.⁴² The use of vinblastine and vincristine (including during the first trimester) has also been reported, and there is no clear evidence of an increased risk of fetal malformations.³³

Platinum Agents

Although platinums are the most commonly used chemotherapeutic agents in pregnancy, data regarding the use of these agents during pregnancy is still limited. There are several case reports on the use of cisplatin or carboplatin alone or in combination with other agents after the first trimester without resultant congenital abnormalities.^10,43-46 A recent review by Mir et al⁴⁷ identified 43 patients in the literature who had been treated with cisplatin (36 patients), carboplatin (6 patients), or both (1 patient), with more than 80% of the cases receiving combination therapy. In the 36 cases of cisplatin exposure, they noted 3 cases of intrauterine growth restriction (IUGR), 2 cases of oligohydramnios, 1 case of polyhydramnios, 1 case of microphthalmos, and 1 case of ventriculomegaly. The association of these malformations with cisplatin remains speculative due to concomitant exposure to other potentially teratogenic drugs. Unremarkable neonatal examination and pediatric development were reported in 34 of 36 children after in utero exposure to cisplatin. In the small number of carboplatin exposure cases described, there were no noted malformations or fetal toxicities noted. A recent study of 7 women who received cisplatin monotherapy for cervical cancer in the second and third trimester of pregnancy reports the concentration of cisplatin in the maternal blood, umbilical cord blood, and amniotic fluid measured at the time of delivery. Maternal serum cisplatin concentrations ranged widely, depending on the number of prior cycles and time interval to last treatment. Amniotic fluid cisplatin concentrations were 13% to 42% and umbilical artery levels 31% to 65% of the maternal serum concentration. Eight healthy infants were born with no anomalies and normal short-term development.⁴⁸

Taxanes

The large molecular weight and high degree of protein binding of paclitaxel likely limit transfer of the drug across the placenta.³³ Human data remain very limited. There are case reports of paclitaxel administration along with a platinum agent for the treatment of ovarian cancer during pregnancy^10,49 without evidence of fetal malformation or developmental abnormalities. Additional cases of taxane administration during pregnancy are reported through the Cancer and Pregnancy Registry²⁸ without fetal malformations.

Topoisomerase Inhibitors

Data regarding the use of etoposide, topotecan, and irinotecan during pregnancy are scant to nonexistent. Use of etoposide has been reported during the second and third trimester, and pancytopenia and IUGR have been noted,^50,51 but there are no reports of fetal malformations.

Supportive Medications

Recombinant erythropoietin does not appear to cross the placenta and does not seem to pose a risk to the fetus. Given the increased risk of thromboembolic disease associated with the use of recombinant erythropoietin and darbepoetin in cancer patients⁵² and the already increased risk of thromboembolic disease associated with pregnancy,⁵³ caution may be warranted.

Granulocyte colony-stimulating factor administration is not teratogenic in animals.³³ It does cross the placenta,⁵⁴ but administration during pregnancy appears to be safe. For example, a study from the Severe Chronic Neutropenia International Registry included 20 pregnancies exposed to treatment, with no adverse outcomes noted.⁵⁵

The antiemetics metoclopramide and ondansetron are pregnancy class B and considered safe in pregnancy. Although malformations have been described in children exposed to prochlorperazine in utero, a large study including 2023 total exposures and 877 exposures during the first trimester found no increase in malformations and no other negative impact.⁵⁶

Large studies have noted an association between the use of systemic corticosteroids in the first trimester and orofacial clefts. For example, a case-control study noted an odds ratio (OR) of 4.3 for isolated cleft lip with or without cleft palate and an OR of 5.3 for isolated cleft palate with first trimester exposure.⁵⁷

The antihistamines ranitidine, cimetidine, and diphenhydramine are all class B agents and are considered safe in pregnancy. Due to concerns about antiandrogenic activity of cimetidine, some recommend that ranitidine is preferred over cimetidine.⁵⁸

For pain control, acetaminophen is the first-line choice for mild pain and is not teratogenic. Nonsteroidal anti-inflammatory drugs (NSAIDs) have been associated with an increased rate of spontaneous abortion with first trimester use. Additionally, in the third trimester, NSAIDs have been linked to decreased amniotic fluid volume and constriction of the ductus arteriosus.⁵⁹ Additionally, a large nested case-control study with more than 36,000 pregnant women noted that the OR for congenital anomalies for women who filled prescriptions for NSAIDs during the first trimester was 2.21 (95% confidence interval [CI], 1.72-2.85) and for anomalies related to cardiac septal closure the OR was 3.34 (95% CI, 1.87-5.98).⁶⁰ Opioids are not teratogenic, but do cross the placenta and pose the risk of fetal or neonatal withdrawal. Acute opioid withdrawal should be avoided during pregnancy, as it can be life-threatening to the fetus. Infants born to mothers who chronically take opioids should be carefully monitored for neonatal abstinence syndrome, characterized by apnea, autonomic dysfunction, diarrhea, diaphoresis, lacrimation, irritability, respiratory distress, seizures, tachypnea, and wakefulness, and treated with careful weaning of opioids.

CERVICAL CANCER

Key Points

1. The mechanisms to diagnose cervical cancer during pregnancy are the same as those used in nonpregnant patients and include cytology, colposcopy, cervical biopsy, and, in carefully selected cases, cervical conization.

2. There is a high rate of postpartum regression of abnormal cervical cytology and cervical dysplasia diagnosed in pregnancy.

3. Management of invasive cervical cancer diagnosed in pregnancy must include consideration of gestational age, fetal viability, disease stage, and patient preferences regarding pregnancy and treatment.

Epidemiology

The incidence of cervical cancer in pregnancy varies from 0.12 to 1.06 per 1000 pregnancies. It is the most common gynecologic malignancy diagnosed during pregnancy as well as the most common malignancy in pregnancy worldwide. Three percent of cervical cancers are diagnosed during pregnancy. For many women, especially those of lower socioeconomic status, pregnancy may be the only time they receive medical attention. Cervical cancer screening guidelines recommend to start cervical cytology screening at 21 years of age regardless of pregnancy status. The use of Pap tests and gynecologic examination as a component of routine prenatal care likely contributes to the high rate of cervical cancer diagnosis during pregnancy.

Risk factors for the development of cervical cancer in nonpregnant and pregnant populations appear to be similar, with human papilloma virus (HPV) infection, multiple sexual partners, early coitarche, smoking, immunosuppression, human immunodeficiency (HIV) infection, low socioeconomic status, and non-white race associated with higher incidence. Cervical cancers diagnosed during pregnancy tend to be of lower stage, with one study finding a 3.1 relative risk of having stage I disease.⁶¹ The histologic spectrum is similar in pregnant and nonpregnant populations, with squamous cell histology predominating.

Diagnosis

The mechanisms to diagnose cervical cancer during pregnancy are the same as those used in nonpregnant patients and include cytology, colposcopy, cervical biopsy, and, in carefully selected cases, cervical conization. The incidence of abnormal cervical cytology in pregnancy is similar to that seen in nonpregnant cohorts, with rates of approximately 5% to 8%. The performance characteristics of cervical cytology do not appear to differ in pregnant and nonpregnant patients. Pregnancy is associated with increased vascularity and altered appearance of the cervix, leading to exaggerated colposcopic findings, which are more difficult to interpret. Thus, if colposcopy is performed during pregnancy, it should be done by a provider with specific expertise in that area. The reliability of expert colposcopic-directed biopsies during pregnancy appears similar to that of the nonpregnant patient, with 83.7% and 89.4% concordance, respectively, with the final diagnosis.⁶²

Cervical biopsies are safe to perform in pregnancy. Despite the increased vascularity of the pregnant cervix, significant hemorrhage rarely occurs. When bleeding does occur, it can usually be controlled by the application of pressure. If necessary, use of Monsel solution to the biopsy site is another alternative for obtaining hemostasis. Suture ligature is rarely required. Use of an endocervical brush to obtain a cytologic specimen is a safe practice,^63,64 but endocervical curettage is omitted during pregnancy due to concerns for possible disruption of the pregnancy. Theoretical complications, such as rupture of membranes, are of concern, despite a lack of definitive evidence for this concern. As in nonpregnant patients, any suspicious cervical mass in pregnancy should be biopsied to exclude potential malignancy.

Cervical conization during pregnancy carries considerable risks, including bleeding, miscarriage, preterm labor, preterm delivery, and infection. The incidence of hemorrhage (> 500 cc) with cold knife conization of the cervix is correlated with the trimester during which the procedure is performed; there is minimal risk during the first trimester, 5% in the second trimester, and 10% in the third trimester.^65,66 Fetal loss rates are on the order of 4.5%.^65,67 Given these risks, cervical conization during pregnancy should only be used when there is strong suspicion for an invasive malignancy based on cytology, cervical biopsy, or colposcopic appearance, and when the diagnosis of invasive malignancy would significantly alter the management of the pregnancy and the timing or route of delivery. In these situations, consideration should be given to performing a cerclage during the conization procedure in pregnancy, whether performed as cold knife conization or loop electrosurgical excision procedure cone.⁶⁸

Treatment

Abnormal Cervical Cytology and Cervical Dysplasia

Studies indicate a high rate of postpartum regression of abnormal cervical cytology and cervical dysplasia diagnosed in pregnancy. A recent study of 1079 patients who underwent colposcopy during pregnancy, mostly without biopsy (93%), found postpartum regression to normal for 64% of patients referred with atypical squamous cells of undetermined significance (ASCUS) or low-grade squamous intraepithelial lesion (LSIL) Paps, and 53% for those with high-grade squamous intraepithelial lesion (HSIL) Paps.⁶⁹ In a study of 153 cases of biopsy-proven cervical intraepithelial neoplasia (CIN) II and CIN III in pregnancy, no cases of progression to invasive or microinvasive disease were found. This study also noted high resolution rates, with 39% of CIN II and 37% of CIN III regressing to normal postpartum.⁷⁰ In line with these high rates of regression and low rates of progression to invasive disease during pregnancy, the 2006 Bethesda consensus guidelines for the management of abnormal cervical cancer screening tests and CIN favor a conservative approach in the absence of suspected malignancy and aim to avoid treatment during pregnancy.^71,72

The management of abnormal Pap tests in pregnancy is summarized in Figure 16-1. ASCUS cytology in pregnancy can be managed the same as in a nonpregnant patient, but colposcopy may be deferred until at least 6 weeks postpartum. For LSIL cytology in pregnancy, the American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines state a preference for performing colposcopy during pregnancy, but it is acceptable to defer this until postpartum. For HSIL, atypical squamous cells, cannot rule out a high-grade lesion (ASC-H), and atypical glandular cells (AGC) cytology during pregnancy, colposcopy is recommended. During colposcopy, biopsies should be taken of any lesions suspicious for CIN II/III or cancer. For histo-logic abnormalities, the ASCCP guidelines recommend follow-up without treatment in CIN I, with re-evaluation with cytology and colposcopy postpartum. For CIN II and CIN III, repeat colposcopy and cytology at 12-week intervals during pregnancy is reasonable, or repeat cytology and colposcopy may be deferred to the postpartum period, depending on the index of suspicion and the gestational age. Treatment of CIN during pregnancy is not indicated. The only indication for treatment of cervical neoplasia during pregnancy is invasive cancer.

FIGURE 16-1. Management of an abnormal pap test and cervical intraepithelial neoplasia in pregnancy. Based on the 2006 American Society for Colposcopy and Cervical Pathology/Bethesda Consensus Guidelines.^65,66 AGC, atypical glandular cells; ASC-H, atypical squamous cells, cannot rule out a high-grade lesion; ASCUS, atypical squamous cells of undetermined significance; CIN, cervical intraepithelial neoplasia; LSIL, low-grade squamous intraepithelial lesion; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion.

Invasive Cervical Cancer

Once invasive cervical cancer is confirmed by biopsy, a careful staging examination and MRI are indicated to assess for size of disease, sites of disease involvement, and evidence of metastatic disease. If cervical cancer is diagnosed in a pregnant woman at an advanced gestational age with expected fetal lung maturity, expedited delivery and initiation of definitive treatment should be undertaken. A cervical cancer diagnosis made in a pre-viable undesired pregnancy should be managed with immediate initiation of appropriate definitive therapy and resultant termination. For all other patients, decisions regarding the treatment of cervical cancer diagnosed during pregnancy are more challenging and should involve a multidisciplinary approach and careful consideration of disease stage, the gestational age of the pregnancy, the patient’s wishes regarding the pregnancy, and the patient’s preferences for therapy.

Figure 16-2 provides an overview of management options for patients diagnosed with invasive cervical cancer during pregnancy. When the decision for definitive treatment has been made for a patient who does not desire to continue a previable pregnancy, recommendations and treatment options are generally the same as those in nonpregnant patients. For patients who wish to continue a previable pregnancy and those who have a potentially viable but premature pregnancy, careful individualized balancing of maternal treatment needs and the desire to allow for fetal maturation must be done. In general, immediate treatment is appropriate in cases of locally advanced disease, documented lymph node metastasis, progression of disease during pregnancy, and when desired by the patient. In some cases, neoadjuvant chemotherapy may provide an opportunity to treat the mother while allowing for further fetal maturation and deferring definitive therapy to the immediate postpartum period.

FIGURE 16-2. Possible management options of invasive cervical cancer in pregnancy. LVSI, lymphovascular space invasion.

Stage IA1

Patients with stage IA1 disease diagnosed on cone biopsy with negative margins can be followed for the remainder of the pregnancy and anticipate vaginal delivery. This approach is supported by data in nonpregnant patients and reports of excellent outcomes in women treated with conization in pregnancy for both stage IA1 squamous cell and IA1 adenocarcinoma of the cervix. In a study of 8 women with squamous cell carcinoma treated with conization and managed expectantly throughout pregnancy, no invasive disease was found at the time of postpartum hysterectomy.⁷³ Similarly, in a small study of 4 pregnant women with stage IA1 adenocarcinoma, no residual invasive cancer was found in postdelivery treatment specimens.⁷⁴ Because there is no convincing evidence suggesting that vaginal delivery in patients with stage IA1 cervical cancer after conization with negative margins compromises outcomes, cesarean section should be reserved for obstetrical indications. In patients requiring cesarean section who do not desire future fertility, consideration may be given to cesarean hysterectomy, bearing in mind the higher morbidity of hysterectomy at the time of cesarean section.

Extrapolating from data in nonpregnant patients, it can be expected that the risk of residual microinvasive disease after conization for apparent stage IA1 disease with a conization margin positive for CIN III is approximately 22%, and the risk of more than microinvasive disease approximates 10%.⁷⁵ Therefore, it is imperative that pregnant patients with positive cone margins be followed closely during pregnancy and thoroughly evaluated postpartum. Given these risks, we recommend monthly clinical examinations and colposcopy every 3 months during pregnancy. A detailed discussion regarding route of delivery follows. Subsequent treatment for a patient who had been followed through pregnancy after conization for apparent stage IA1 disease with a conization margin positive for CIN III will depend on the patient’s wishes for future fertility and should at a minimum entail repeat conization. If the patient wishes for more definitive therapy, a frozen cone-hysterectomy⁷⁶ postpartum is a reasonable management option.

Early-Stage Disease: Stages IA2 to IIA

In early-stage patients (stages IA2-IIA) who desire termination of a previable pregnancy and immediate definitive therapy, radical hysterectomy and lymph-adenectomy with the fetus in situ is generally recommended. In this typically young patient population, surgical management as opposed to radiation therapy may be preferable because it may allow for preservation of ovarian function and avoidance of radiation complications such as vaginal stricture and long-term gastrointestinal toxicity. Some authors recommend evacuation of the uterus before radical hysterectomy when the procedure will be performed at greater than 20 weeks.⁷⁷ This may be accomplished via hysterotomy during the same procedure. Primary radiation treatment with concomitant chemotherapy may be an alternative option for patients with early-stage disease, especially those who are poor surgical candidates. Consideration should be given to the pretreatment injection of a feticidal agent for second-trimester patients to honor possible patient preferences, avoid the possibility of a live birth during the procedure, and reduce the risk of violating legislation surrounding late previable pregnancy termination.⁷⁸

The operative morbidity and outcomes of radical hysterectomy appear to be similar in pregnant and nonpregnant populations. In a case-control study of 26 patients who underwent radical hysterectomy for the treatment of cervical cancer during pregnancy, there was no difference in operative time, hospital stay, postoperative bladder function, or postoperative complications compared with nonpregnant matched controls. Blood loss was significantly more in the pregnant group, but blood transfusion was no more frequent. There was also no difference in disease status at last contact, with only 1 patient in the pregnant group dead of disease after an average follow-up period of more than 12 years.⁷⁹

For early-stage patients diagnosed near term and those who choose to delay definitive therapy until after a viable delivery, radical cesarean hysterectomy with lymphadenectomy is generally recommended. In this approach, the patient undergoes a classical cesarean section, the hysterotomy is closed, and a radical hysterectomy and staging lymphadenectomy is performed. Radical cesarean hysterectomy has been shown to be associated with a higher blood loss and need for blood transfusion.⁸⁰ However, the rates of other operative and postoperative morbidities are acceptable and on par with radical hysterectomy in nonpregnant patients. The benefits of completing delivery and therapy in a timely manner and in a single procedure appear to outweigh the disadvantage of greater blood loss. Ovarian transposition may be considered if intraoperative findings and tumor characteristics place a young patient at high likelihood for requiring radiation treatment.

Locally Advanced Disease: Stages IIB to IVA

In general, the recommended treatment for patients with stage IIB to IVA disease is radiation with concomitant sensitizing chemotherapy. Very few studies have evaluated the management of cervical cancer in pregnancy with radiotherapy or chemoradiation.^81–83 In early pregnancy, chemoradiation treatment can commence without prior evacuation of the uterus. In the majority of cases, fetal death is expected to occur within 2 to 3 weeks and abortion by 20 to 45 days after the beginning of radiotherapy.^84,85 Data suggest that pregnancy loss during radiation is delayed and occurs less reliably later in gestation. In a series of 14 patients who underwent radiation therapy for cervical cancer during pregnancy, pregnancy loss occurred an average of 33 days after the initiation of treatment in the first trimester and 44 days after initiation of treatment in the second trimester.⁸⁶ Because of this, some recommended uterine evacuation before initiation of radiotherapy in previable gestations greater than 20 weeks. If hysterotomy is planned for uterine evacuation, lymph-adenectomy may be performed during the same surgical procedure. Medical abortion is another alternative for uterine evacuation and has been successfully used to induce uterine evacuation in cases where radiation therapy resulted in fetal death, but not spontaneous abortion.⁸⁷ Injection of a selective feticidal agent should be considered not only before pregnancy evacuation procedures, but also before the initiation of chemoradiation for patients in the second trimester, given the potential psychological impact on patient, family, and providers of performing radiotherapy with a live fetus in utero.

For patients with locally advanced disease with a viable premature or a strongly desired previable pregnancy, management decisions are complex and need to take into consideration the impact of delaying definitive therapy for the mother, the role of staging lymphadenectomy and neoadjuvant chemotherapy during the pregnancy, and the morbidity associated with a premature delivery for the child, all discussed in more detail next.

Fortunately, stage IVB disease is very rare in pregnancy. As in nonpregnant patients, a systemic approach to treatment is usually used, with local therapy focused on palliation of symptoms.

Delaying Treatment

The morbidity and mortality associated with preterm delivery is considerable (Table 16-5). Short delays in delivery may have a significant impact on neonatal survival. For example, in infants admitted to the neonatal intensive care unit without congenital abnormalities, the mortality rate is approximately 30% at 25 to 26 weeks, compared with less than 10% at 28 week and less than 2% at 34 to 35 weeks. Although neonatal survival rates have been improving remarkably at US referral centers, neurodevelopmentally intact survival is still problematic for very prematurely born babies.⁸⁸ Given the risks associated with prematurity, an effort must be made to balance the desire to optimize fetal outcome with the potential risks of delaying treatment for the mother. Data from many case reports, case series, and case-control studies indicate that there is likely minimal maternal risk associated with delaying treatment of cervical cancer when early-stage disease (stages IA-IB1) is diagnosed during the late second or early third trimester of pregnancy. In a review of 129 pregnancies during which cervical cancer therapy was deliberately delayed for 3 to 4 weeks, excellent outcomes were achieved. More than 95% of patients were alive and without evidence of disease at last follow-up (Table 16-6). Given these data, delaying treatment for early-stage cervical cancer until fetal lung maturity is obtained may be a reasonable treatment option for patients who desire to continue their pregnancy. It is unlikely that this carries considerable risk of inferior cancer outcome for the mother. When choosing to delay therapy, close surveillance is imperative. Progression of disease during this period has been observed, portends a poor prognosis, and warrants immediate initiation of treatment. Given these risks, we recommend monthly clinical examinations and colposcopy every 3 months during pregnancy. One of the greatest challenges lies in determining an individual patient’s risk in delaying therapy.

Table 16-5 Neonatal Outcomes of Extremely Preterm Infants⁸⁸

Stay updated, free articles. Join our Telegram channel

Join