Key Abbreviations
Acute lymphoblastic leukemia ALL
Acute myeloid leukemia AML
Acute nonlymphocytic leukemia ANLL
Alpha-fetoprotein AFP
Beta human chorionic gonadotropin β-hCG
Carcinoembryonic antigen CEA
Central nervous system CNS
Chronic myelocytic leukemia CML
Computed tomography CT
Fine-needle aspiration FNA
Food and Drug Administration FDA
Gestational trophoblastic disease GTD
Gray Gy
Hodgkin lymphoma HL
Magnetic resonance imaging MRI
Non-Hodgkin lymphoma NHL
The juxtaposition of life and death can present numerous emotional and ethical conflicts to the patient, her family, and her physicians. The diagnosis of cancer for anyone is understandably frightening. To deal with cancer in the context of pregnancy is particularly burdensome, because the patient may have to balance competing maternal and fetal interests. On occasion, a pregnant woman may be required to make decisions that affect her life or longevity versus the life or the well-being of her unborn child. Cancer in pregnancy complicates the management of both the cancer and the pregnancy. Diagnostic and therapeutic interventions must carefully address the associated risks to both the patient and the fetus. Informed decisions require evaluation of a number of factors, and with counseling, these considerations are the foundation on which treatment decisions are made. An evolution has taken place in the philosophy of care, from one of total disregard of the pregnancy with frequent immediate termination to a more thoughtful approach in which management decisions consider both maternal and fetal outcomes so as to limit risk of death or morbidity to both.
It is estimated that approximately 20% to 30% of malignancies occur in women younger than age 45 years. Although cancer is the second most common cause of death for women in their reproductive years, only about 1 in 1000 pregnancies is complicated by cancer. Because no large prospective studies have addressed cancer treatment in pregnancy, physicians tend to base treatment strategies on small retrospective studies or anecdotal reports that occasionally present conflicting information. A successful outcome is dependent on a cooperative multidisciplinary approach. The management plan must be formulated within a medical, moral, ethical, legal, and religious framework that is acceptable to the patient and guided by communication and educational resources of the health care team.
Delays in diagnosis of cancer during pregnancy are common for various reasons: (1) many of the presenting symptoms of cancer are often attributed to the pregnancy; (2) many of the physiologic and anatomic alterations of pregnancy can compromise physical examination; (3) serum tumor markers such as β-human chorionic gonadotropin (β-hCG), alpha-fetoprotein (AFP), and cancer antigen 125 (CA 125) are increased during pregnancy; and (4) the ability to optimally perform either imaging studies or invasive diagnostic procedures may be altered during pregnancy. Because the gestational age is significant when evaluating the risks of treatments, it is important to determine gestational age accurately. An early ultrasound evaluation may be useful to ensure accurate dating.
The malignancies most commonly encountered in the pregnant patient are breast cancer, cervical cancer, and melanoma—cancers that most commonly occur in this age group—but also ovarian, thyroid, and colorectal cancer as well as leukemia and lymphoma. The frequencies of these diseases complicating pregnancy may increase secondary to the trend to delay childbearing, and age is the most potent predictor of cancer. Before specific malignancies are discussed below, some general principles are reviewed.
Chemotherapy During Pregnancy
Pharmacology of Chemotherapy During Pregnancy
Because pregnancy profoundly changes maternal physiology (see Chapter 3 ), the potential exists for altered pharmacokinetics associated with chemotherapy. Orally administered medications are subject to changes in gastrointestinal (GI) motility. Peak drug concentrations are decreased owing to the 50% expansion in plasma volume, which produces a longer drug half-life unless a concurrent increase occurs in metabolism or excretion. The increase in plasma proteins and fall in albumin may alter drug availability, and amniotic fluid may act as a pharmacologic third space that potentially increases toxicity because of delayed metabolism and excretion. Hepatic oxidation and renal blood flow are both elevated during pregnancy and may influence the metabolism and excretion of most drugs. However, because pharmacologic studies in pregnant women are lacking, we currently assume that initial drug dosages are similar to those given to the nonpregnant woman, and adjustments to dose are based on toxicity on a course-by-course basis. Because most antineoplastic agents can be found in breast milk, breastfeeding is contraindicated (see Chapter 24 ).
Drug Effects on the Fetus
All drugs undergo animal teratogenicity testing, and based on these results, the drugs are assigned risk categories ( Table 50-1 ) by the U.S. Food and Drug Administration (FDA). Based on this system, most chemotherapeutic agents are rated as C, D, or X. However, animal teratogenicity testing cannot always be reliably extrapolated to humans. For example, a drug (e.g., aspirin) may show teratogenic effects in animals but will not affect humans. The opposite is also true and, as such, has serious potential to do harm; for example, a drug may demonstrate no animal teratogenicity (e.g., thalidomide), but it can cause serious human anomalies (see Chapter 8 ). Because detailed ultrasonography may fail to identify subtle anatomic but serious functional abnormalities before 20 weeks’ gestation, patients should be appropriately counseled and may want to consider the option of pregnancy termination if first-trimester chemotherapy is planned or administered. The risk of teratogenicity during the second and third trimesters is significantly reduced and is likely no different from that for pregnant women who are not exposed to chemotherapy.
| CATEGORY | DEFINITION |
|---|---|
| A | Controlled studies have demonstrated no risk in the first trimester, and the possibility of fetal harm appears remote. |
| B | Either animal studies have failed to identify a risk but no controlled studies have been done in women or animal studies have shown an unconfirmed adverse effect. |
| C | Either animal studies have revealed adverse effects and no controlled studies have been done in women or studies in women and animals are not available. Use the drugs only if the potential benefit justifies the potential risks. |
| D | Evidence of fetal risk exists, but the benefits may be acceptable despite the risk in either a life-threatening situation or a serious disease for which safer drugs are ineffective. |
| X | Studies in animals or humans have demonstrated fetal abnormalities, and the risk of the drug clearly outweighs any possible benefit. |
Although the literature that addresses chemotherapy administration during pregnancy is somewhat limited and dated, literature reviews provide us with some information regarding the frequency of affected offspring. Because antineoplastic agents target the rapidly dividing malignant cell, it could be expected that the exposed fetus would be particularly susceptible to serious toxicity. Clear documentation of such is not the case. Excluding the intentional use of abortifacients, it is difficult to clearly demonstrate that the use of chemotherapy results in an increase in the clinically recognized spontaneous abortion rate over the expected 15% to 20%. However, if continuation of the pregnancy is desired, chemotherapy is not advised in the first trimester because of an increased rate of major malformations that ranges from 10% to 17% with single-agent therapy and up to 25% with combination chemotherapy. Administration in the second and third trimester has been associated with intrauterine growth restriction (IUGR), stillbirth, and low birthweight. Maternal effects such as chemotherapy-induced nausea and vomiting may also affect fetal growth and birthweight. Because early induction of labor or surgical delivery is often a component of the overall treatment plan, it has been difficult to identify premature birth as a specific result of the chemotherapy. Preterm delivery has been reported to occur in more than half of pregnancies associated with malignancies, with a majority being iatrogenically induced and associated with increased neonatal morbidity. Therefore the timing of delivery and consideration of delay of therapy in the case of early-stage disease should be part of the treatment planning.
When chemotherapy is administered during the latter half of gestation, fetal organ toxicity has not been reported as a major problem, although neonatal myelosuppression and hearing loss have been reported. Although second malignancies, impaired growth and development, intellectual impairment, and infertility have been reported after chemotherapy administration to children, the delayed effects of in utero exposure are less well documented. Although few data are available, concerns regarding cognitive development have been evaluated, and the outcomes are comparable to those of individuals not exposed to chemotherapy in utero. Growth and fertility in these children have also been shown to be similar to those not exposed.
Classification of Chemotherapy Agents
Antimetabolites
Historically, the antimetabolite aminopterin was used as an abortifacient, and in cases of failed abortion, the risk of fetal malformation was about 50%. Methotrexate has replaced aminopterin for chemotherapeutic purposes, and although similar types of anomalies occur, a lower overall frequency (<10%) has been reported. In the first trimester, methotrexate has been associated with skeletal and central nervous system (CNS) defects. Although no anomalies were reported in the second and third trimesters, methotrexate was associated with low birthweight and neonatal myelosuppresion. The use of low-dose methotrexate for systemic diseases (e.g., rheumatic disease and psoriasis) does not appear to produce teratogenicity. In a prospective study, 5-fluorouracil, in combination with other agents, was well tolerated and resulted in no perinatal deaths.
Alkylating Agents
Alkylating agents such as cyclophosphamide and chlorambucil are commonly used for the management of malignancies. Unfortunately, most of these agents have demonstrated some teratogenic potential when administered in the first trimester, with defects that include renal agenesis, ocular abnormalities, and cleft palate. In the second and third trimesters, alkylating agents have been shown to be acceptably safe.
Antitumor Antibiotics
Even when administered in early pregnancy, antitumor antibiotics—such as doxorubicin, idarubicin, bleomycin, and daunorubicin—appear to demonstrate a low risk of teratogenicity. Doxorubicin has been associated with one case of multiple anomalies; however, the patient was receiving combination chemotherapy. One potential reason for the low rate of anthracycline-associated teratogenicity is its questionable ability to cross the placenta.
Vinca Alkaloids
Although potent teratogens in animals, vincristine and vinblastine do not appear to be as teratogenic in humans. In limited studies, vinca alkaloids were well tolerated in both early and later stages of pregnancy.
Platinum Agents
Platinum agents have been used with relatively acceptable risks, and normal outcomes have been reported along with cases of IUGR. In addition, case studies have reported hearing loss and ventriculomegaly in 2.7% of patients exposed to cisplatin. One case of bilateral hearing loss was noted in the offspring of a woman treated with carboplatin in the second trimester of pregnancy. However, the cause cannot be fully attributed to the platinum agent because other causes of hearing loss were also present, including prematurity and administration of gentamycin in the neonatal period.
Miscellaneous
Taxane chemotherapy is used in the treatment of malignant tumors of multiple sites, and experience with this class of drugs remains limited, although a few reports to date are favorable.
Targeted Therapies
Despite the emerging role of targeted agents in cancer treatment, they have been used only inadvertently in pregnancy. Trastuzumab has been linked to oligohydramnios, and its use is not recommended in pregnancy. Rituximab has been associated with transient neonatal lymphopenia, and further studies are necessary to determine an accurate safety profile. Imatinib has been associated with low birthweight and premature delivery, whereas erlotinib showed no adverse effects in one case. The antiangiogenic agents (bevacizumab, sunitinib, and sorafenib) are not recommended for use in pregnant women.
Radiation Therapy
Ionizing radiation is a known teratogen, and the developing embryo is particularly sensitive to its effects. Doses higher than 0.20 Gray (Gy) are considered teratogenic. Effects during early pregnancy are typically lethal or result in congenital malformations. During late gestation, radiation may cause specific organ damage in addition to mental retardation, skeletal anomalies, and ophthalmologic abnormalities. Reports have indicated that radiation exposure in utero increases the risk of malignancies such as leukemia and other childhood tumors, particularly when exposure occurred in the first trimester. It is important to remember that these data are conflicting and are extrapolated from data derived from atomic catastrosphes. Regardless, recommendations for radiation exposure in pregnancy should be less than 0.05 Gy. Radiation therapy should be postponed until the postpartum period, and consideration should be given to suitable alternative therapies such as surgery or chemotherapy. If delay of therapy is not possible, termination of the pregnancy may be required.
Surgery and Anesthesia
Aspects of surgery are addressed later in the sections on specific cancers, but some general principles should be considered first. Although complications of surgery can threaten the fetus, extraperitoneal surgery is not related to spontaneous abortion or preterm labor. Surgery can be performed safely in all three trimesters. If flexibility in timing exists, abdominal or pelvic surgery is best performed in the second trimester to limit the risk of first-trimester spontaneous abortion or preterm labor. In the first trimester, progesterone therapy is indicated (weeks 7 to 12) following resection of the corpus luteum. When laparoscopy is selected for abdominal surgery, the open technique is recommended to avoid injury to the uterus. Perioperative cautions include attention to the relative safety of all drugs administered. When appropriate, antibiotic prophylaxis can be safely administered. Fever secondary to either infection or atelectasis should be treated promptly because it may be associated with fetal abnormalities.
No evidence suggests that significant risks of anesthesia exist independent of coexisting disease (see Chapter 16 ). Preoxygenation is critical due to the increase in oxygen consumption, which can lead to desaturation. In the second and third trimesters, careful attention to positioning is required so as to avoid compression of the vena cava from the enlarged uterus. Use of local or regional anesthesia should be considered. Finally, continuous or intermittent fetal monitoring should be used when extrauterine survival is possible.
Pregnancy Following Cancer Treatment
With improved survival rates for many childhood and adolescent malignancies, the clinician must be prepared to offer prenatal counseling to the young woman who presents with a cancer history. Issues worthy of review and in need of clarification for the obstetrician and the patient are listed in Box 50-1 .
What is the risk of recurrence of the malignancy?
If a recurrence were to be diagnosed, depending on the most likely sites, what would be the nature of the probable treatment?
How would such treatment compromise both the patient and the fetus?
Will prior treatments—such as pelvic surgery, radiation to pelvis or abdomen, or chemotherapy—affect fertility or reproductive outcome?
Will the hormonal milieu of pregnancy adversely affect an estrogen receptor–positive tumor?
Previous abdominal irradiation for a Wilms tumor appears to adversely affect the risk of pregnancy complications, including increased perinatal mortality, low birthweight, and abnormal pregnancy. In contrast, a review of pregnancies following treatment for Hodgkin lymphoma revealed no increase in poor pregnancy outcome. However, the rate of ovarian failure following the multiple drug combinations is greater than 50% in some reports. Also, a combination of pelvic irradiation and chemotherapy for Hodgkin disease results in an even higher rate of ovarian failure. Over recent years, a number of women with early cervical cancer have received fertility preservation surgery with radical trachelectomy and regional lymphadenectomy. The preliminary fertility and pregnancy outcomes have been favorable.
Will pregnancy increase the risk of recurrence or accelerate recurrence? Even in women with estrogen receptor–positive breast cancer, there is no evidence that subsequent pregnancy adversely affects survival. In addition to the altered hormonal milieu, concern is also directed toward the potential for accelerated tumor activity associated with alterations in the pregnant patient’s immune system. No available data support this concern; however, it has been recommended that after a cancer diagnosis, pregnancy should be delayed for at least 2 years, during which recurrence risk is highest.
Cancer During Pregnancy
Breast Cancer
The estimated number of breast cancer cases in women in the United States exceeds 230,000 cases with approximately 40,000 deaths each year. Although the lifetime risk of developing breast cancer is 1 in 8, the risk is 1 in 206 for women 40 years or younger. Each year, this results in approximately 7% to 15% of premenopausal breast cancers being diagnosed in women who are pregnant or lactating or within 1 year after delivery. Although an increase in the frequency of breast cancer complicating pregnancy has been predicted because of delayed childbearing, recent publications are consistent with a frequency of breast cancer concurrent with pregnancy of 1 per 3000 to 10,000 live births.
In general, the risk of breast cancer is directly related to the duration of ovarian function. Therefore both early menarche and late menopause appear to increase the likelihood of developing breast cancer. However, interruption of the normal cyclic ovarian function by pregnancy appears to be protective. This apparent protective effect may be secondary to the normal hormonal milieu of pregnancy that produces epithelial proliferation, followed by marked differentiation and mitotic rest. Multiparous women, particularly those who breastfeed, have a lower risk of developing breast cancer than do nulliparous women. However, based on one study of almost 90,000 women, breastfeeding may not be an independent protective factor. Paradoxically, carriers of BRCA1 and BRCA2 mutations may have an increased risk of developing breast cancer by having children.
Diagnosis and Staging
Breast abnormalities should be evaluated in the same manner as in a nonpregnant patient. The most common presentation of breast cancer in pregnancy is a painless lump discovered by the patient. Because the breast changes become more pronounced in later pregnancy, it is important to perform a thorough breast examination at the initial visit (see Chapter 24 ). Despite the striking physiologic breast changes of pregnancy, including nipple enlargement and increases in glandular tissue that result in engorgement and tenderness, a newly found or persistent breast mass should be evaluated promptly. Diagnostic delays may increase the risk of nodal involvement and are often attributed to physician reluctance to evaluate breast complaints or abnormal findings in pregnancy. The lengths of delays in diagnosis of breast cancer in pregnancy are commonly 3 to 7 months or longer and may result in more advanced stages of diagnosis compared with the general population.
Although bilateral serosanguinous discharge may be normal in late pregnancy, less common presentations such as bloody nipple discharge should be evaluated with mammography and ultrasound. In cases of mastitis or breast abscesses or for evaluation of breast edema or inflammation, a skin biopsy should be considered to evaluate for inflammatory breast cancer.
Mammography in pregnancy is controversial. Although the radiation exposure to the fetus is negligible, the hyperproliferative changes in the breast during pregnancy are characterized by increased tissue density, which makes interpretation more difficult. Breast ultrasound has a high sensitivity and specificity and can distinguish between solid and cystic masses, which makes it the preferred imaging modality for pregnant women. Magnetic resonance imaging (MRI) without gadolinium has also been used; however, data are limited in regard to its accuracy in this population. Percutaneous biopsy is recommended for any lesion that does not meet criteria for a simple cyst. Fine-needle aspiration (FNA) of a mass for cytologic study may be used, but it is often misleading because of changes in the breast during pregnancy. For breast masses, a core needle biopsy is the preferred method for histologic diagnosis.
Before proceeding with treatment, staging should be undertaken. All draining lymph nodes should be evaluated. Recent studies have evaluated the role of sentinel lymph node dissection in pregnant women with breast cancer. Although blue dye is associated with a risk of anaphylaxis and should not be used, several reports have demonstrated the safety of sentinel lymph node biopsy using a low-dose lymphoscintigraphy with 99m-Technetium–labeled sulfur colloid. This technique may be considered in select cases at an experienced center. The contralateral breast must be carefully assessed. Laboratory tests should include baseline liver function tests and serum tumor markers, carcinoembryonic antigen (CEA), and also cancer antigen 15-3 (CA 15-3), which appears to be a useful tumor marker for monitoring breast cancer in pregnancy. A chest radiograph is indicated, and if the liver function tests are abnormal, the liver can be evaluated by ultrasound. If bone metastases are suspected, a bone scan can be performed in pregnancy. In a symptomatic patient, radiographs of the specific symptomatic bones are advised. However, in an asymptomatic patient with normal blood tests, because the yield is low, bone scanning is usually not performed.
Treatment
The treatment of breast carcinoma at any time is often overshadowed by psychological and emotional factors. Because of potential risks to the developing fetus, treatment decisions carry an additional burden. Therapy must be individualized in accordance with present knowledge and with the specific desires of the patient, gestational age, and tumor stage and biology; at all times, maternal treatment should adhere to standard recommendations. At the time of diagnosis, it is important to have a baseline assessment of fetal growth and development by ultrasound as well as routine fetal monitoring throughout the pregnancy and treatment.
Local Therapy
The usual criteria for breast-preserving therapy versus modified radical mastectomy pertain to the patient with breast cancer in stages I to III. Although breast surgery can be safely performed in all trimesters with minimal fetal risks, the use of radiation is complicated by the presence of the pregnancy. Consideration should be given to the delay of irradiation until after delivery. Experimental calculations suggest that when the fetus is within the true pelvis, a tumor dose of 5 Gy will expose the fetus to 0.010 to 0.015 Gy. Later in pregnancy, parts of the fetus may receive as much as 0.200 Gy. Because these levels exceed the upper limit of safety, alternative approaches should be considered, and reports have shown that in the late second or third trimester, delaying treatment without impacting maternal outcome may be reasonable.
Pregnancy Termination
Because early studies suggested more unfavorable outcomes in pregnant women, it was assumed that the hormonal changes of pregnancy contributed to rapid tumor growth; therefore therapeutic abortion was frequently advised. At present, a harmful effect of continuing pregnancy has not been demonstrated in most published series. Although data are limited, studies have shown similar survival outcome if the patient had a spontaneous pregnancy loss, chose to terminate, or continued the pregnancy. However, it is difficult to evaluate potential selection bias toward performing an abortion in patients with advanced disease. Because young women tend to have hormone receptor–negative tumors, it is difficult to make an argument, based on hormonal concerns, for either termination of pregnancy or oophorectomy as an adjunct to therapy.
Chemotherapy
In general, women who present with either metastatic breast carcinoma or rapidly progressive inflammatory carcinoma should avoid any delays in therapy. Immediate initiation of chemotherapy is critical to providing the patient with inflammatory carcinoma with any chance for long-term survival. If the patient has a clinical indication for adjuvant chemotherapy other than inflammatory carcinoma, the delay of instituting chemotherapy and awaiting fetal pulmonary maturity should be considered in select third-trimester situations. Current literature shows that the use of cytotoxic agents for breast cancer in pregnancy is relatively safe; and anthracycline-based regimens, which are the most widely used, are associated with a favorable safety profile. Additional agents such as 5-fluorouracil, cyclophosphamide, and taxanes have also been used, but data on these agents are limited. Others concur with the use of chemotherapy in the second and third trimesters in the management of breast cancer. Pregnancy is a contraindication to the use of tamoxifen because of the risk of adverse fetal outcomes, including craniofacial malformations and ambiguous genitalia. Data regarding the safety of trastuzumab in pregnancy are limited, and currently it is not recommended because of reports of oligohydramnios and neonatal deaths as a result of respiratory and renal failure.
Prognosis
As with any malignant disease, prognosis best correlates with the anatomic extent of disease at the time of diagnosis. Outcomes in stage I and II disease are favorable, and survival rates approach 86% to 100%; however, the presence and extent of nodal involvement is especially predictive of prognosis in both nonpregnant and pregnant patients. Although nodal status is of prognostic significance, the number of positive nodes is also important. In the pregnant patient, the 5-year survival rate is 82% for patients with three or fewer positive nodes and 27% if greater than three nodes contain tumor. Probably because of the associated delays in diagnosis, pregnancy appears to increase the frequency of nodal disease, with 53% to 71% of patients exhibiting nodal involvement at diagnosis.
Reports have demonstrated that up to 60% to 80% of breast cancer cases in the pregnant population are estrogen receptor negative and have a rate of ERBB2 positivity between 25% and 58%. In addition to a higher incidence of lymph node metastases, breast tumors in pregnant women are more frequently reported to be larger and to be of a higher grade that those of their nonpregnant counterparts. However, these rates are histologically identical to the nonpregnant patient of similar age, and when controlled for age and stage, pregnancy does not seem to affect prognosis adversely. Of note, pregnant women with early-stage disease who had a delay in therapy did not experience a negative effect on disease-free survival. However, breast cancer therapy during pregnancy is feasible, and patients should be counseled thoroughly. Although infants exposed to chemotherapy for breast cancer in utero had a lower birthweight and more complications, this was not clinically significant, and most adverse outcomes were associated with preterm delivery; thus full-term delivery should be planned when possible.
Subsequent Pregnancy
Although the consensus is that subsequent pregnancies do not adversely affect survival, recommendations have been made regarding the timing of a subsequent pregnancy. It is generally advised that women with node-negative disease wait 2 to 3 years, and this interval should be extended to 5 years for patients with positive nodes. Others have advised that no delay is indicated for the patient with a good prognosis who does not receive postoperative adjuvant chemotherapy. It has been advised that patients should undergo a complete metastatic workup before a subsequent pregnancy.
Lactation and Breast Reconstruction
Lactation is possible in a small percentage of patients after breast-conserving therapy for early-stage breast cancer. Lumpectomy using a radial incision, rather than the cosmetically preferred circumareolar incision, is less likely to disrupt ductal anatomy (see Chapter 24 ). Disruption of the ductal system may increase the rate of mastitis. Breastfeeding is contraindicated in women receiving chemotherapy, because significant levels of the drug can be found in breast milk.
Breast reconstruction with the use of autologous tissue has increased secondary to questions about the use of silicone-filled implants. The transverse rectus abdominis myocutaneous (TRAM) flap is one popular method of breast reconstruction. Because the donor site is a portion of the anterior abdominal wall, there is potential concern when the patient develops abdominal distension from pregnancy. In one case report and review of the literature, nine women who became pregnant following breast reconstruction experienced no problem with anterior abdominal wall integrity. However, because data are limited, it may be best to address reconstruction in the postpartum period.
Lymphoma
Hodgkin Lymphoma
Approximately 36,000 cases of lymphoma were expected to be diagnosed in women in the United States in the year 2015, and only about 11% of these would have been Hodgkin lymphoma (HL). HL is the second most commonly diagnosed cancer in women aged 15 to 29 years, and it presents at a mean age of 32 years. Lymphomas complicate approximately 1 in 6000 pregnancies, and few large series have evaluated the many issues that affect these women. However, spontaneous abortion, stillbirth rates, preterm births, and adverse pregnancy outcomes do not appear to be increased or affected by the course of the disease, and termination of pregnancy should not be routinely advised. Although some advocate therapeutic abortion for the patient with a first-trimester pregnancy and HL to allow complete staging, others limit therapeutic abortion to those women who require infradiaphragmatic radiation or those with extensive systemic symptoms or visceral disease.
Women with HL frequently present with enlarged cervical or axillary lymph nodes, and the diagnosis is established by biopsy of the suspicious nodes. The presence of systemic symptoms such as night sweats, pruritus, or weight loss suggests more extensive disease. In a series of 17 women diagnosed with HL during pregnancy, the average time of diagnosis was 22 weeks’ gestation, and systemic symptoms were uncommon. Clinical staging for lymphoma requires systemic evaluation by history, laboratory findings, bone marrow, and radiographic imaging. Clinical treatment and staging are individualized. Although pathologic staging for HL in the nonpregnant patient may involve laparotomy and splenectomy, these are not generally performed in pregnancy. The routine evaluation for HL during pregnancy includes a single anterior/posterior view chest radiograph, liver function tests, serum creatinine clearance, complete blood count, erythrocyte sedimentation rate, and lymph node and bone marrow biopsy . Evaluation of the abdomen is compromised by the gravid uterus, and MRI is preferred because it can accurately assess nodes, liver, and spleen. Chest tomography or computed tomographic (CT) scan of the mediastinum may be necessary to evaluate nodal enlargement in the chest. Isotope scans of the liver and bone are best avoided during pregnancy, but ultrasonography is safe and may provide useful information.
Disease stage is the most important factor in treatment planning and prognosis ( Table 50-2 ). The survival rate for early-stage HL exceeds 90%, whereas patients with disseminated nodal disease have a 5-year survival rate of about 50%. As expected, patients with stage IV disease have poor survival rates. Radiation therapy is the mainstay of treatment for early-stage HL, and combination chemotherapy is used for the treatment of advanced-stage disease with organ involvement. Most investigators agree that treatment should not be withheld during pregnancy except in early-stage disease, particularly if the diagnosis is made in late gestation.
| STAGE | DESCRIPTION |
|---|---|
| I | Involvement of a single lymph node region (I) or of a single extralymphatic organ site (I E ) |
| II | Involvement of two or more lymph node regions on the same side of the diaphragm (II) or localized involvement of an extralymphatic organ site and of one or more lymph node regions on the same side of the diaphragm (II E ) |
| III | Involvement of lymph node regions on both sides of the diaphragm (III), which also may be accompanied by localized involvement of an extralymphatic organ or site (III E ), the spleen (III S ), or both (III SE ) |
| IV | Diffuse or disseminated involvement of an extralymphatic organ with or without localized lymph node involvement (liver, bone marrow, lung, skin) |
Radiotherapy to the supradiaphragmatic regions may be performed with abdominal shielding after the first trimester. When this approach is chosen, irradiation should be delayed until after the first trimester, total fetal dose should be limited to 0.1 Gy or less, and the goal is partial therapy with completion therapy after delivery. Spontaneous abortion has been reported with an estimated first-trimester fetal dose of 0.09 Gy secondary to scatter from delivering 44 Gy to the chest of a patient receiving treatment for a recurrence. In general, if the estimated exposure to a first-trimester fetus is expected to exceed 0.1 Gy or if combination chemotherapy is planned for the first trimester, therapeutic abortion should be considered because of an increased risk of fetal malformations. Asymptomatic early-stage disease that presents in the second half of pregnancy may be followed closely while preparations are made for early delivery. The use of corticosteroids and single-agent chemotherapy have been proposed for the patient with systemic symptoms.
Subdiaphragmatic or advanced disease requires chemotherapy, and commonly used regimens include doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) and mechlorethamine, vincristine, procarbazine, and prednisone (MOPP). Because many of these chemotherapeutic agents are known teratogens, such treatment is best avoided in the first trimester. A systematic review found 42 pregnancies in which HL was treated with chemotherapy, and 17 were exposed to chemotherapy in the first trimester, which resulted in six congenital anomalies and three spontaneous abortions. Similar treatments should also be approached with caution later in pregnancy, although most case reports have documented only IUGR and neonatal neutropenia as complications. Long-term follow-up toxicity studies are lacking.
Following therapy for HL, it has been suggested that pregnancy planning should take into consideration that about 80% of recurrences manifest within 2 years. Treatments for HL may compromise the reproductive potential of young patients. As reflected in Figure 50-1 , ovarian failure is more likely to occur in older patients even if treated with fewer courses of chemotherapy. Early studies reported normal ovarian function in only 12% of women following therapy for HL, and combined treatment with radiation and chemotherapy have been found to lead to the highest risk of ovarian failure. More recent series have shown the pregnancy rate among reproductive-age HL survivors to be similar to that of the normal female population. Regimen changes that have occurred since the earlier reports include improved pelvic shielding during primary irradiation, ovarian transposition when pelvic radiation is required, and more frequent use of the ABVD regimen instead of MOPP. Depending on their availability, new reproductive technologies that include oocyte donation and embryo cryopreservation can be considered in select situations.
Patients who become pregnant after treatment for HL do not demonstrate increased adverse perinatal outcomes such as fetal wastage, preterm birth, and birth defects when compared with sibling controls. Although fetal anomalies have occurred after treatment for HL, chromosomal abnormalities or a new gene mutation have not been diagnosed. The absence of a repetitive pattern of malformations makes it difficult to imply a causal relationship between any birth defects observed and previous therapy for Hodgkin disease.
General recommendations, therefore, include delaying therapy until at least the second trimester due to the increased first-trimester fetal risk. After the first trimester, therapy should not be delayed for patients with symptomatic, subdiaphragmatic, or progressive HL. Treatment options include supradiaphragmatic radiation therapy for early-stage disease and combination chemotherapy for bulky, subdiaphragmatic, progressive, or advanced disease. Radiation fields should be tailored to limit fetal dose, and a maximal fetal dose calculation should be performed prior to treatment. Pregnant women with HL should be managed by a maternal-fetal medicine specialist, and whether pregnancy termination should be pursued will depend on the clinical situation.
Non-Hodgkin Lymphoma
Non-Hodgkin lymphoma (NHL) occurs at a mean age of 42 years and is therefore observed less frequently than Hodgkin lymphoma with approximately 100 cases reported in pregnancy. In general, NHL is more likely to complicate pregnancy because more patients have an aggressive histology and advanced-stage disease. Indolent lymphomas are less common in younger patients and are rarely described in pregnancy. Up to 70% of NHL in pregnancy is diffuse large B-cell lymphoma. Interestingly, reproductive organs—particularly the breasts and ovaries—are commonly involved in pregnancy-associated lymphoma compared with aged-matched counterparts.
With the exception of aggressive histologic types, the management of lymphoma may be deferred in the first trimester because of the typical indolent nature of the disease. When initiated, treatment often includes chemotherapy and targeted therapy. If required, radiotherapy can be considered in local fields away from the pelvis in late gestational stages; however, it is typically postponed until after delivery.
Treatment regimens evaluated in pregnancy include cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) with or without bleomycin, which did not have an adverse effect on pregnancy outcomes. Non-CHOP regimens were associated with several unfavorable outcomes that included miscarriage, intrauterine fetal demise, and transient neonatal leukopenia. The anti-CD20 antibody rituximab has been evaluated in limited cases with acceptable toxicities; however, further studies are warranted.
Acute Leukemia
Although the incidence of leukemia in pregnancy is not specifically known, it is estimated to occur in fewer than 1 in 75,000 to 100,000 pregnancies. Acute leukemia represents about 90% of leukemias that coexist with pregnancy. Acute myeloid leukemia (AML) accounts for about 60% and acute lymphoblastic leukemia (ALL) for about 30% of cases. More than three fourths of the cases are diagnosed after the first trimester. The prognosis for acute leukemia in pregnancy is guarded. Although no evidence suggests that pregnancy adversely affects the prognosis of acute leukemia, optimal and immediate care of the pregnant patient with acute leukemia necessitates a team effort and is best achieved in a cancer referral center.
The diagnosis of acute leukemia is rarely difficult. The signs and symptoms of anemia, granulocytopenia, and thrombocytopenia include fatigue, fever, infection, and easy bleeding or petechiae, which usually prompts a complete blood count. A normal or elevated white blood cell count (WBC) is present in up to 90% of patients with ALL, although a WBC in excess of 50,000 is found in only one fourth of patients. In contrast, patients with acute nonlymphocytic leukemia (ANLL) may present with a markedly elevated WBC, although one third may exhibit leukopenia. The diagnosis of leukemia should be confirmed by bone marrow biopsy and aspirate. The biopsy material is usually hypercellular with leukemic cells. The smear of the aspirate reveals decreased erythrocyte and granulocytic precursors as well as megakaryocytes. Leukemic cells comprise greater than half of the marrow’s cellular elements in most patients. The morphology of the marrow and the peripheral leukemic cells help to distinguish between lymphocytic and nonlymphocytic leukemias. This latter group includes acute myelocytic (granulocytic), promyelocytic, monocytic, and myelomonocytic leukemias and erythroleukemia. Acute myelocytic leukemia is the most common form of ANLL. Patients who develop ANLL as a result of previous chemotherapy have a particularly poor response to treatment.
Acute leukemia requires immediate treatment regardless of gestational age because delay may result in poorer maternal outcomes. When ALL is diagnosed in the first trimester, termination is recommended due to the effect on both the fetus and the mother. This is in part due to the need for intensive treatment that includes stem cell transplantation, which is contraindicated in gestation. However, numerous reports have documented successful pregnancies in patients with acute leukemia who were aggressively treated with leukapheresis and combination chemotherapy in the second and third trimesters. Although reports have shown no serious long-term effects of in utero exposure to chemotherapy, it is important to counsel patients that acute leukemia and its therapy are associated with a high rates of spontaneous abortion, stillbirth, preterm delivery, and fetal growth restriction. If the mother is exposed to cytotoxic drugs within 1 month of delivery, the newborn should be monitored closely for evidence of granulocytopenia or thrombocytopenia.
Chronic Leukemia
Chronic leukemia accounts for approximately 10% of cases of leukemia during pregnancy. A majority of cases are chronic myelocytic leukemia (CML) because the median age is 35 years, compared with chronic lymphocytic leukemia (CLL), which has a median age of 60 years and makes cases during pregnancy rare.
CML is characterized by excessive production of mature myeloid cell elements, with granulocyte counts that average 200,000/dL. Most patients have thrombocytosis and a mild normochromic normocytic anemia. Platelet function is often abnormal, although hemorrhage is usually limited to patients with marked thrombocytopenia. CML tends to be indolent, and normal hematopoiesis is only mildly affected in the early stages of disease. Therefore delay of aggressive treatment is more feasible than with acute leukemia. Unless complications such as severe systemic symptoms, autoimmune hemolytic anemia, recurrent infection, or symptomatic lymphatic enlargement occur, treatment for chronic leukemia should be withheld until after delivery. When necessary, therapy may consist of interferon-alfa, steroids, and chemotherapy agents or more recently, tyrosine kinase inhibitors such as imatinib. However, whereas the data are limited, it does appear that early exposure to imatinib is associated with a high rate of congenital anomalies and miscarriage. Imatinib therapy initiated in the second and third trimesters may not add significant additional risk, but better outcome data are necessary before this can be ensured.
Melanoma
The incidence of malignant melanoma is increasing in the childbearing years, with reported ranges from 1% to 3.3% in pregnant or lactating women; it accounts for 8% of malignancies diagnosed in gestation. A topic of continued debate is whether pregnancy exerts a negative effect on the course of malignant melanoma. Prognostic features of the primary tumor include tumor thickness, ulceration, and location. Reports have suggested that melanoma that arises during pregnancy is associated with an aggressive clinical course; it is more likely to be diagnosed at an advanced stage, to have an increased tumor thickness, to be found in locations associated with a poor prognosis and therefore to have a shorter disease-free interval. A meta-analysis of melanoma suggested that pregnancy-associated melanomas did appear to have poorer outcomes. However, after correcting for tumor thickness, survival rates were similar, and other studies show no increase in poor prognostic location of lesions in the pregnant patient and comparable survival outcomes to nonpregnant cohorts ( Table 50-3 ). A large population-based study found no data to support the notion of advanced stage, thicker tumors, increased rate of lymph node metastases, or worsened survival in women with melanoma diagnosed during pregnancy when compared with nonpregnant women. This series also noted that maternal and neonatal outcomes were equivalent to those of pregnant women without melanoma.
