Malignancy and Pregnancy

Epidemiology

Ovarian cancer in pregnancy is a very rare event, with an estimated occurrence rate of 1 in 18,000–55,000 pregnancies. ^{, ,} Though adnexal masses are detected in ∼2% of pregnancies, they persist in only a small fraction. ^, It is estimated that 1 in 1000 pregnant women undergo exploratory surgery to evaluate an adnexal mass and approximately 1%–5% of these are malignant. ^, Most ovarian malignancies diagnosed in pregnancy are germ cell tumors; however, the incidence of epithelial ovarian malignancies identified during pregnancy may increase as women defer childbearing later into their reproductive years. The diagnosis of sex cord stromal tumor in pregnancy is rare. ^{, ,}

Evaluation and Management

With extensive use of obstetrical ultrasound (US), the number of adnexal masses diagnosed in pregnancy has increased with time. Transvaginal or abdominal ultrasound is the preferred imaging technique to evaluate adnexal masses as it is safe, widely accessible, and has a relatively high sensitivity and specificity. ^, Magnetic resonance imaging (MRI) may add value when ultrasound is not conclusive. For example, MRI can better depict the characteristic findings of an exophytic leiomyoma, degenerating leiomyoma, endometrioma, decidualized endometrioma, and ovarian edema. MRI can be used safely throughout pregnancy, though use of the contrast agent gadolinium is discouraged as it crosses the placenta and is excreted by the fetal kidneys into the amniotic fluid. Exposure to ionizing radiation should be avoided in pregnancy where possible; when unable, the lowest reasonably achievable dose should be utilized. Though computed tomography (CT) is not absolutely contraindicated in pregnancy, it should be considered only if maternal benefits outweigh risk of fetal radiation exposure.

Approximately 75% of adnexal masses complicating pregnancy are simple cysts measuring less than 5 cm in diameter, and the remaining 25% are simple or complex masses that exceed 5 cm in diameter. During pregnancy, ∼70% of ovarian masses spontaneously resolve by the early second trimester, becoming undetectable by 14 to 15 weeks’ gestational age. ^, The likelihood of spontaneous resolution decreases with increasing size and complexity. Adnexal masses larger than 8 cm in diameter are more often complicated by pain, torsion, rupture, or internal hemorrhage, compared with smaller lesions. Preterm labor, preterm premature rupture of membranes, obstruction of labor, and fetal death have rarely been observed. ^{, ,} Ovarian torsion occurs most commonly in the late first or early second trimester, when the uterus is growing out of the true pelvis, and in the puerperium, when the uterus undergoes rapid involution. If clinical signs or symptoms consistent with torsion, rupture, or hemorrhage occur, emergent surgery is indicated independent of gestational age. In the case of an ovarian tumor larger than 8 cm in diameter in a woman who does not undergo surgery, the clinician and patient must be aware of the increased risk for complications and their symptoms.

In the nonemergent setting, radiographic features and symptom assessment dictate need for surgical exploration. In one study of more than 550 pregnant patients, complex masses (loculations, septations, solid features, papillary projections, or poorly defined borders) were found to be malignant in 9% of cases, whereas only 1% of simple cysts were malignant. Surgical exploration should therefore be considered for complex adnexal masses that persist, increase in size, or have radiographic features concerning for malignancy, whereas observation is reasonable for simple and asymptomatic masses in pregnancy. ^{, ,} There is no consensus on the management of adnexal masses based on size alone. The American College of Obstetricians and Gynecologists (ACOG) identifies adnexal masses >10 cm as concerning for malignancy; however, simple and asymptomatic masses remain low risk and observation can be considered.

Tumor markers may be useful when assessing risk of malignancy in patients with adnexal masses; however, these markers must be interpreted with caution in pregnancy. As levels of cancer antigen 125 (CA 125) are usually elevated during the beginning of pregnancy and can persist throughout pregnancy, serum testing is unreliable for evaluating the risk for epithelial malignancy. The serum level of maternal lactate dehydrogenase (LDH), the tumor marker for dysgerminomas, is not altered by pregnancy, and it can serve as a marker for the disease during pregnancy. ^, Serum alpha fetoprotein (AFP), the tumor marker for endodermal sinus tumors, is produced by the yolk sac, fetal liver, and gastrointestinal tract. Additionally, it can be elevated in the setting of neural tube defects. Thus AFP is nonspecific, though if severely elevated, a malignant germ cell tumor should be considered. Human chorionic gonadotropin (hCG), the tumor marker for choriocarcinoma, is unreliable in pregnancy. Inhibin B and antimullerian hormone (AMH) can be elevated in women with granulosa cell tumors. These levels are not elevated in normal pregnancy and thus can be useful for diagnostic evaluation.

Surgical exploration for an ovarian mass is optimally undertaken in the second trimester, ideally between 14 and 20 weeks’ gestation, though it can be performed at any time in gestation provided specific measures are taken. This is the optimal window for surgical intervention as it allows time for spontaneous regression of benign masses, limits disruption of corpus luteum, and avoids use of anesthetic agents during organogenesis. Additionally, the risk of spontaneous abortion declines dramatically in the second trimester. When possible, minimally invasive techniques are preferred. ^, If an adnexal mass is discovered late in pregnancy, evaluation, management, and surgical exploration may be deferred until or after delivery in some circumstances. If abdominal delivery is required, diagnosis and potential staging can be undertaken at the time of cesarean section; if delivery is vaginal, surgery may be postponed, depending on the nature and appearance of the mass.

If a malignancy is diagnosed at the time of surgery, next steps are dictated by the extent of disease, tumor histology, and patient preferences with respect to the patient’s current pregnancy and future fertility. Ideally, thorough counseling about the risks and benefits of staging and/or tumor reductive surgery is done prior to proceeding to the operating room. This conversation must be had in the context of both her current pregnancy and oncologic outcomes and is ideally conducted using a multidisciplinary approach. Once the diagnosis of malignancy is made, appropriate therapy should not be withheld, although it may need to be modified or delayed in order to most optimally care for the patient and her fetus.

Malignant ovarian germ cell tumors are the most common ovarian malignancy diagnosed in pregnancy and the majority are diagnosed at an early stage. ^, Dysgerminoma is found most frequently, followed by yolk sac tumors; these account for 38% and 30.4% of malignant germ cell tumors diagnosed in pregnancy, respectively. Surgical management of suspected malignant germ cell tumor in pregnancy consists of unilateral salpingo-oophorectomy with strong consideration for comprehensive fertility-sparing staging ( Box 56.1 ). As 15% of dysgerminomas are bilateral, inspection of the contralateral ovary with biopsy or removal of abnormalities should be performed for therapeutic and prognostic reasons. Wedge resection or biopsy of a normal-appearing contralateral ovary is unnecessary because of the risk for increased adhesion formation and possible infertility. Patients with stage I dysgerminoma and stage I, grade 1 immature teratoma can be appropriately managed by observation following surgical staging per the National Comprehensive Cancer Network (NCCN) guidelines. For patients with other histologies (endodermal sinus, embryonal, choriocarcinoma, polyembryonal, grade 2–3 immature teratoma) and/or advanced disease, adjuvant chemotherapy is warranted. Chemotherapy typically includes bleomycin, etoposide, and cisplatin (BEP). Malignant germ cell tumors have a propensity for rapid growth and often recur when adjuvant therapy is withheld; thus timely initiation is recommended.

The prognosis for women with early-stage dysgerminoma is excellent, and they have a very low likelihood of recurrence. In women with advanced disease, approximately 10% of tumors will recur, but most are treatable. Endodermal sinus tumors (i.e., yolk sac tumors) of the ovary are rare, aggressive tumors that confer a poor prognosis. These tumors are marked by increased serum levels of maternal AFP, which may be elevated in normal and abnormal pregnancies. Nevertheless, an extremely elevated AFP level in an apparently normal pregnancy may be associated with an endodermal sinus tumor. Because of their aggressive nature, surgery is indicated for the diagnosis of an endodermal sinus tumor, and adjuvant chemotherapy is administered to all women with this diagnosis.

Sex cord–stromal tumors (mainly granulosa cell tumors and Sertoli-Leydig cell tumors) are rarely diagnosed in pregnancy. These tend to present with early-stage disease and have a slow, indolent course. Independent of pregnancy, these tumors manifest with evidence of hormone excess (i.e., virilization or hyperestrogenism). During pregnancy, they more commonly manifest with hemorrhagic rupture leading to hemoperitoneum. Management includes unilateral oophorectomy and surgical staging. Adjuvant therapy is reserved for patients with advanced or recurrent disease. Based on the results of a large prospective clinical trial, carboplatin and paclitaxel is the preferred regimen, though BEP is a reasonable alternative.

Epithelial ovarian cancer is extremely rare in pregnancy. A systematic review identified 105 cases of epithelial ovarian cancer in pregnancy between 1955 and 2013. Serous carcinoma was the most common histologic subtype, accounting for 47.6% of cases, followed by mucinous (27.6%) and endometrioid (10.5%). Once the diagnosis is made, appropriate therapy should be initiated. A multidisciplinary approach should be taken and should include gynecologic oncology, maternal-fetal medicine, neonatology, and anesthesiology. For women with advanced disease, the decision to proceed with primary cytoreductive surgery versus neoadjuvant chemotherapy depends on disease distribution, gestational age at diagnosis, patient’s wishes with respect to pregnancy continuation, and maternal condition.

For patients with ovarian cancer diagnosed during pregnancy, oncologic outcomes appear to be similar to those for patients who are not pregnant.

Cervical Intraepithelial Neoplasia

Cervical cytology, HPV testing, and physical examination are the principal forms of cervical cancer screening during pregnancy. Endocervical curettage should be avoided to prevent direct or indirect injury to the pregnancy, but an endocervical brush should be employed to increase the adequacy of the test. This can increase the incidence of spotting after collection, but it appears to have no effect on the risk for serious adverse outcomes related to the pregnancy. Approximately 2%–7% of pregnancies may be complicated by an abnormal Pap test or HPV result. During pregnancy, the goal of evaluation of an abnormal cervical screening result is determination of the extent of neoplasia. The aim is to rule out invasive cancer and allow therapy for preinvasive disease to be deferred until after delivery. In an immunocompetent woman, cervical dysplasia rarely progresses in pregnancy, whereas regression is common.

The Bethesda system is the standard classification system for cervical neoplasia and is used to guide management for patients with abnormal cervical cytology. Initial management of abnormal cervical cytology is similar to that in the nonpregnant patient. For women with atypical squamous cells of uncertain significance (ASC-US), high-risk (HR) HPV testing should be used to determine need for colposcopy in women >25 years old. If HR HPV testing is positive, colposcopy is recommended; if negative, repeat co-testing in 3 years is preferred. Cases in which a high-grade lesion cannot be excluded (ASC-H) are typically managed with colposcopy. Women who are found to have low- or high-grade squamous intraepithelial lesions (LSILs or HSILs) should undergo colposcopy to evaluate the extent and severity of neoplasia. In a large, multicenter retrospective study of >1000 patients with abnormal cytology in pregnancy, cervical biopsy revealed CIN 2–3 in 20% of patients when the colposcopic impression was normal/LSIL and in 55% when the colposcopic impression was similarly high grade.

Pap tests revealing atypical glandular cells (AGCs), although rare in pregnancy, also warrant colposcopic examination. Pregnancy complicates the cytologic interpretation of AGC, because sloughed decidual cells, endocervical gland hyperplasia, and cells demonstrating an Arias-Stella reaction, all of which are benign, can occur in a normal pregnancy. Compared with nonpregnant patients, for whom AGC is associated with malignancy in as many as 25% of cases, AGC found in pregnancy is less likely to indicate malignancy. However, it should be noted that the inability to perform an endocervical curettage and endometrial sampling limits the evaluation of AGC during pregnancy.

Colposcopy during pregnancy has high diagnostic accuracy and a very low complication rate of 0.6% that includes bleeding, infection, spontaneous abortion, and preterm labor. ^{, ,} Colposcopic evaluation in pregnancy is facilitated by eversion of the transformation zone. Punctations, mosaicism, atypical vessels, or friable lesions should raise suspicion for malignancy. Biopsies should be performed for any suspicious lesions seen at the time of colposcopic evaluation. However, taking many samples at one examination should be avoided as this may increase risk of complications. The most common complication associated with colposcopically directed biopsy is hemorrhage resulting from increased vascularity of the cervix during pregnancy. Bleeding can be stopped by direct pressure to the site, with application of Monsel (ferric subsulfate) solution, silver nitrate, vaginal packing, and, rarely, suture ligation of vessels. Colposcopy and biopsy do not jeopardize the pregnancy as long as endocervical curettage is avoided.

Observation without therapy (with repeat cytology or colposcopy at intervals no more frequent than every 12 weeks) is appropriate for pregnant women with HSIL (CIN 2–3) if invasive cervical cancer has been excluded by colposcopy (with or without biopsies). ^, This approach is supported by the American Society of Colposcopy and Cervical Pathology (ASCCP). Repeat biopsy is recommended only if the appearance of the lesion or cytology suggests invasive cancer, and deferring reevaluation until at least 6 weeks postpartum is acceptable. Inadequate colposcopic evaluation is an indication for further evaluation with a cone biopsy in the nonpregnant patient. However, pregnant women with an unsatisfactory initial colposcopic result may undergo repeat colposcopy in 6 to 12 weeks. Because the transformation zone undergoes further eversion as pregnancy progresses, a later examination may be satisfactory.

Treatment for cervical dysplasia (or further evaluation of dysplasia), such as laser therapy, cryotherapy, loop electrosurgical excision procedure (LEEP), and cone biopsy, is typically deferred to the postpartum period. At 6 months after delivery, almost 70% of CIN 2 and CIN 3 lesions have resolved, which is higher than the rate for the nonpregnant population. ^, For this reason, delay of definitive therapy until after delivery is appropriate in a patient without evidence of invasive cervical cancer; however, the maximum interval from delivery until postpartum evaluation and management of cervical dysplasia has not been determined. Cervical cytology, colposcopy, directed biopsies, and endocervical curettage usually are performed after the 6-week postpartum evaluation.

In pregnant patients, excisional biopsies should be reserved for the exclusion of invasive disease. Risks associated with these procedures in pregnancy include cramping, bleeding, infection, preterm premature rupture of membranes, spontaneous abortion, preterm labor, and pregnancy loss. Complication rates are similar for LEEP and cone biopsy. If an excisional biopsy is indicated to rule out malignancy, the safest time to perform the procedure seems to be in the first or early second trimester. Though considered safe in the first trimester, most providers avoid conization prior to 12 weeks’ gestational age as this is the highest risk time period for spontaneous abortion independent of the procedure. In an effort to minimize complications related to the procedure, a coin-shaped (shallow cone) or even wedge excision of a lesion can be performed. For patients in whom a more traditional conization is indicated (suspected endocervical lesion), some physicians have advocated concurrent cerclage at the time of cone biopsy, although efficacy is uncertain. In population-based studies, the risk for adverse pregnancy outcomes following treatment for CIN seems to be higher at earlier gestational ages and with a larger volume of cervix removed. Although there were no complications in the largest study of this strategy, only 17 patients were evaluated.

For patients with cervical dysplasia diagnosed during pregnancy without any clinical evidence of invasive cervical cancer, the route of delivery should not be affected by the dysplasia. Some physicians have documented an increased rate of spontaneous remission of cervical dysplasia after vaginal delivery compared with cesarean delivery, but others have not found this to be the case.

Cervical Carcinoma

The occurrence of cervical carcinoma in pregnancy is rare, comprising only 1% of all cervical cancers diagnosed annually. However, because cervical cytology and examination are often performed early in prenatal care, cervical cancer is one of the cancers more commonly diagnosed during pregnancy. When detected in pregnancy, cervical cancer is often early stage owing to increased surveillance. Manifestation of cervical cancer during pregnancy is similar to that outside pregnancy, and most women present without symptoms. The most common sign of cervical cancer in pregnancy is bleeding, especially postcoital bleeding. It is imperative for clinicians caring for pregnant women to recognize that vaginal bleeding is not necessarily related to the pregnancy and warrants thorough evaluation.

Pregnancy was once believed to alter the course of cervical cancer compared with nonpregnant patients, but there is no difference in survival or disease characteristics when matched cohorts are studied. When compared with nonpregnant counterparts, pregnant women with cervical cancer are more likely to have early-stage disease. Because of the physiologic and anatomic changes of pregnancy, induration or nodularity at the inferior cardinal ligament is less prominent, leading to underestimation of the stage and degree of tumor involvement. Nonetheless, pregnancy does not affect the survival rate for cervical cancer. The overall survival rate is 80%, compared with 82% in nonpregnant patients.

Clinical staging of cervical cancer has historically included examination under anesthesia, intravenous pyelography, chest radiography, cystoscopy, and sigmoidoscopy. However, the most recent International Federation of Gynecology and Obstetrics (FIGO) staging update allows for the incorporation of findings from advanced imaging studies and/or surgical pathology. Intravenous contrast–enhanced computed tomography (CT) or positron emission tomography (PET) has often been used for staging and treatment planning. However, the use and timing of these staging studies during pregnancy must be considered carefully because of the ionizing radiation exposure ( Table 56.2 ). PET uses intravenous radioactively labeled fluorodeoxyglucose (18F-FDG). The fetal effects of this radioisotope are relatively unknown, though the estimated doses to the fetus are beneath the threshold for any deterministic effects. Because of the relative lack of safety data in human pregnancy, alternative imaging modalities (MRI, US, and/or CT) are most often chosen. Chest radiography is acceptable during pregnancy with appropriate abdominal and pelvic shielding. In some case reports, MRI, which has been shown to be safe during pregnancy, has been used to define the extent of the cervical tumor and/or extracervical tumor spread.

Management of cervical cancer in pregnancy must be individualized. A multidisciplinary team, which may include a perinatologist, neonatologist, radiation oncologist, and gynecologic oncologist, should be recruited to counsel the patient regarding treatment options related to the stage of disease, fetal status, and gestational age and to determine her desire to continue the pregnancy. In certain circumstances, treatment of the cancer is recommended despite the potential lethal or otherwise adverse effect of the therapy on the pregnancy. In others, treatment can be delayed until after delivery or until a gestational age at which delivery will not produce significant neonatal morbidity. Intentional delays in treatment have been reported from 6 to 32 weeks for women with stage I–II disease without significant compromise in outcome. ^,

When microinvasive cervical cancer is suggested by a cervical biopsy or invasive disease is suspected at colposcopy but not confirmed on biopsy, cervical conization is required. The procedure is optimally performed in the operating room, with a knife, after the period of organogenesis has passed and after appropriate counseling about the risks of fetal loss and transfusion. As discussed above, cervical conization in pregnancy carries additional risks over that performed outside of pregnancy. The ideal timing of cone biopsy during pregnancy is controversial, with some physicians reporting fetal loss rates approximating 25% when conization is performed during the first trimester, whereas others have described the relative safety of early conization. ^, Risk of pregnancy loss decreases with increasing gestational age. However, blood loss associated with cone biopsy increases with advancing gestational age.

After conization with negative margins in a woman with microinvasive carcinoma, follow-up colposcopy can be used to monitor disease progression during pregnancy, with no alteration in the intrapartum management. Definitive management is typically deferred to the postpartum period. Surgical treatment for stage IA1 cervical cancer may include cervical conization (with negative margins) or extrafascial hysterectomy, depending on the desire for future fertility. If conization with negative margins occurred during pregnancy and the patient desires future fertility, surveillance alone is appropriate.

Although the approach is controversial, patients with cervical adenocarcinoma in situ or stage IA1 cervical adenocarcinoma with negative cone biopsy margins probably can be managed conservatively during their pregnancy, because most studies demonstrate a low risk for parametrial and lymphatic metastasis with early invasive cervical adenocarcinoma. Women with positive margins after conization during pregnancy must be counseled regarding the risk for current invasive disease, with management based on the risks and benefits of observation, repeat conization, or definitive therapy for possible invasive cervical cancer during pregnancy.

Outside of pregnancy, stage IA2 or IB1–2 cervical cancer can be definitively treated with radical surgery (radical hysterectomy versus trachelectomy) with lymphadenectomy or radiation therapy (often with concurrent radiosensitizing chemotherapy). The advantage of primary surgical management includes the ability to preserve ovarian function and to avoid the potential negative effect on sexual function imparted by radiation therapy. However, 15%–20% of patients will have features on final pathology that will impart an increased risk of recurrence for which adjuvant radiation therapy (with or without concurrent radiosensitizing chemotherapy) will be recommended. Diagnosis during pregnancy does not change the recommended treatment plan, though it may alter the timing of interventions or require consideration of neoadjuvant chemotherapy in order to allow delay in definitive therapy for fetal maturation.

For patients who elect for termination of the pregnancy, immediate therapy concordant with NCCN guidelines is appropriate. For stage IA2–IB2 cervical cancer, gravid radical hysterectomy with lymphadenectomy is feasible in the first or second trimester. Small reports detailing institutional experience with this approach have demonstrated the safety and feasibility of this procedure. ^{, ,} For patients requiring or electing to proceed with immediate radiation therapy, termination of the pregnancy prior to the initiation of therapy is preferred. In cases where this is not feasible (i.e., bulky cervical tumor, patient refusal), radiation therapy can be delivered with the fetus in utero, which typically results in spontaneous abortion within approximately 3 weeks, and the uterus can be evacuated subsequently. ^,

For women with early-stage disease who elect to continue their pregnancy, case reports and small case series suggest that a moderate delay in definitive therapy is associated with oncologic outcomes similar to those in women treated promptly. Sood and colleagues described 11 women whose definitive treatment for stage IA and IB cervical cancer was delayed by an average of 16 weeks (range, 3 to 32 weeks). All women remained without evidence of disease and without apparent negative consequences related to the delay in definitive therapy during pregnancy. This study and other small series suggest that a moderate delay of definitive therapy does not incur excessive risk. ^, However, the risk is difficult to quantify, and it is unclear how the length of delay may impact risk because of the small number of patients for whom this management strategy has been reported. Careful counseling and documentation are imperative.

Radical trachelectomy with lymphadenectomy has been proposed as a pregnancy-sparing but immediate definitive surgical approach to early-stage cervical cancer in pregnancy. This approach is generally not recommended due to increased risk of pregnancy loss (upward of 20% based on the available data). However, it may be considered in select circumstances. Neoadjuvant chemotherapy can be used for women electing to continue pregnancy who may be deemed high risk for adverse outcomes with delay in therapy. ^{, ,} This may include women with early-stage disease that progresses during pregnancy as well as those with larger tumors or more advanced disease as determined by imaging and/or surgical staging (lymphadenectomy). For all women with cervical cancer who choose to delay definitive therapy (with or without the use of neoadjuvant chemotherapy), close observation during pregnancy is required with thorough physical examination at least every 4 weeks.

For women with locally advanced cervical cancer, definitive therapy with chemoradiation is recommended. For women who present in the second half of pregnancy, a short delay in therapy to allow for fetal maturity may be considered, though this carries a small but unquantifiable risk for adverse cancer outcome. ^, Cesarean delivery (to avoid delivery through a cervical tumor) followed by radiation therapy with concurrent chemotherapy is most often the preferred approach. Diagnosis in the first trimester and early second trimester, however, would result in prolonged delays in therapy that may significantly increase the risk for a poor oncologic outcome. Termination of pregnancy followed by immediate initiation of therapy is therefore advised. Evacuation of the uterus can be performed before irradiation or after radiation therapy if spontaneous miscarriage does not ensue. The anatomic distortion that occurs in pregnancy must be considered when planning radiotherapy to ensure the appropriate treatment fields. Patients who refuse immediate therapy for advanced cervical cancer during the first half of a pregnancy must be counseled regarding the potential effect on tumor growth and spread and the worsened prognosis. In selected patients, neoadjuvant chemotherapy (to decrease the risk for cancer progression during pregnancy) can be considered, with definitive chemoradiotherapy initiated after delivery. However, there are limited data to support this management strategy. ^,

Coordination of care with a perinatologist, neonatologist, and gynecologic oncologist is critical to determine the appropriate gestational age for delivery that balances the maternal oncologic risks with the fetal risks of prematurity. In a recent study, late preterm deliveries (before 37 weeks) in pregnant patients with cancer were associated with neonatal intensive care stays in more than 50% of subjects. For this reason, delivery at or after 37 weeks is preferred. Cesarean delivery is recommended for most patients with cervical cancer in pregnancy, but this remains controversial. Vaginal delivery is typically reserved for those with preinvasive or microinvasive tumors. Delivery through a bulky and friable cervical cancer leads to the potential risk for hemorrhage. Cervical cancer may also recur at an episiotomy site, typically within 6 months after delivery, and this is associated with a dismal prognosis. ^, For women with early-stage cervical cancer whose primary therapy will be surgical, this may be performed concurrent with cesarean delivery (cesarean hysterectomy or cesarean radical hysterectomy with or without lymphadenectomy) or after a short interval (4–6 weeks postpartum). Alternatively, depending on the clinical extent of disease, delivery followed by definitive pelvic irradiation with or without concurrent cisplatin may be recommended. For those who will need or are likely to need radiation therapy, oophoropexy can be considered at the time of cesarean section in effort to preserve ovarian function.

In selected women with early cervical cancer, fertility-sparing surgery may be considered for definitive therapy with the potential to maintain future fertility. This may include conization, simple trachelectomy, or radical trachelectomy with lymphadenectomy for women with stage I cervical cancer (radicality to be determined by lesion size). Radical trachelectomy is usually performed in nonpregnant patients with small tumors (<2 cm), but it has been reported during pregnancy as discussed above. After radical trachelectomy, oncologic outcomes appear most favorable in women who have small (<2 cm) squamous lesions without lymphovascular invasion where the reported recurrence rates range from 2% to 5%. Conception rates are greater than 50% following radical trachelectomy, the majority of which are spontaneous. Approximately 20% of patients require assisted reproductive technology, which may include cervical dilation with intrauterine insemination (IUI), IUI with or without ovulation induction, or in vitro fertilization (IVF). The live birth rate following fertility-sparing surgery for cervical cancer is approximately 68%, though rates vary by the radicality of the cervical excisional procedure.

Surgical Principles

Fewer than 2% of women undergo nonobstetric surgery during pregnancy (with a small fraction of these undergoing cancer surgery). Among 5405 women undergoing nonobstetric surgical procedures, most performed in the second trimester, the rates of stillbirth and congenital anomalies were not different from those for pregnancies not associated with surgery, but low birth weight and preterm delivery were more common. An increased risk for preterm delivery has been reported after surgery in the first trimester, particularly with longer procedure times. It is challenging, however, to determine whether it is the underlying disease process or the surgical intervention that increases risk.

For nonviable pregnancies, assessment of fetal heart rate prior to and immediately following surgery is recommended. For patients with viable pregnancies, continuous intraoperative fetal heart rate monitoring should be considered if technically feasible, although efficacy is uncertain. This may be helpful to assess fetal well-being in procedures when hemodynamically important maternal blood loss or wide swings in maternal blood pressure may reasonably be anticipated. A sustained fall in fetal heart rate may indicate decreased uterine perfusion that may be restored by repositioning the patient to relieve inferior vena cava compression or by expanding the intravascular volume. Every effort should be made to avoid ancillary procedures and minimize operative time during urgent procedures during pregnancy. In particular, abdominal surgery during pregnancy carries increased fetal risk compared with extraabdominal surgery. It is imperative to have a clear plan in place regarding management and/or plan for delivery should fetal distress be noted intraoperatively. This involves a multidisciplinary discussion between the patient, surgeon, obstetrician, anesthesiologist, and neonatologist.

Laparoscopic surgery is feasible and relatively safe during pregnancy and is the preferred surgical approach where feasible. ^, Advancing gestation limits intraperitoneal space and visualization and thus laparoscopy is typically limited to the first and second trimesters. Laparoscopic procedures are associated with less pain in the postoperative period, fewer postoperative contractions, reduced use of analgesics and tocolytics, and overall shorter length of stay. The most common indications for laparoscopic surgery during pregnancy are cholecystectomy, evaluation of an adnexal mass, and appendectomy. Surgical risk inherent to minimally invasive surgery does not appear to be increased in pregnancy, and laparoscopic techniques in pregnant patients should not differ greatly from those in nonpregnant patients. Nasogastric or orogastric decompression can be used to minimize insertion of ports into the stomach, and the patient should be placed in a leftward tilt to minimize inferior vena cava compression. Though somewhat controversial, pneumoperitoneum is most often established through an open technique to minimize the risk for uterine perforation or laceration. Additionally, intraabdominal pressure should be minimized (typically maximum pressure of 10–13 mm Hg) and surgical time should be limited when possible. There is theoretical risk that the developing fetus may be exposed to respiratory acidosis caused by maternal absorption of the carbon dioxide gas, with subsequent hypercarbia and serum conversion to carbonic acid. Modern anesthesia technique during general anesthesia includes continuous monitoring of expired carbon dioxide, and anesthesiologists intentionally hyperventilate sleeping patients under these circumstances to minimize the risk for respiratory acidosis. Studies of fetal sheep have demonstrated that these animals have adequate compensatory response and placental reserve to tolerate insufflation, although one pregnant ewe died during establishment of the pneumoperitoneum. ^, There have been no human reports of fetal loss attributable to acidosis caused by pneumoperitoneum, and this risk may be theoretical.

With any surgical approach during pregnancy (open or minimally invasive), the gravid uterus should be manipulated as little as possible to minimize the risk for spontaneous preterm labor, ruptured membranes, or other complications. Prophylactic tocolytics at the time of surgery in the second trimester do not decrease the risk for preterm labor or preterm delivery, but they may be beneficial in the third trimester. Specific oncologic procedures should be considered individually as to their relative safety during pregnancy. Whereas lumpectomy during breast cancer surgery has been demonstrated to be safe, the use of vital dyes (methylene blue or lymphazurin) or of radionuclide technetium 99 (99Tc) for this purpose has not been extensively evaluated. In a study of patients with breast cancer undergoing lymphoscintigraphy for sentinel lymph node evaluation, the average radiation dose to the uterus was less than 0.0002 cGy, suggesting its safety. ^,

Epidemiology

Breast cancer is one of the malignancies most commonly encountered in pregnant women. The reported incidence ranges from 1.3 to 3.3 cases per 10,000 live births. Although only 0.2% to 3.8% of breast cancers diagnosed in women younger than 50 years of age are pregnancy-associated, almost 10% to 20% of breast cancers diagnosed in women in their 30s are discovered during pregnancy. ^{, ,} This pattern of prevalence is likely to change in the future with an increase in gestational breast cancers as more women delay childbearing. Women with inherited genetic predisposition may be overrepresented among women diagnosed with pregnancy-associated breast cancer.

Clinical Presentation and Diagnosis

The most common clinical presentation of breast cancer in pregnant and nonpregnant women is a painless lump in the breast. Occasionally, refusal by an infant to nurse from a lactating breast may signify an occult carcinoma; this has been described as the milk rejection sign. The physiologic changes in the breast during pregnancy and lactation result in engorgement and increased nodularity, which makes it a challenge for the patient and clinician to identify tumors by palpation. Such challenges result in diagnostic delays of 2 months or longer, which in part is responsible for the advanced stage at diagnosis in pregnant or lactating women. In a mathematical model assessing tumor progression over time, a 1-month delay in treatment of the primary tumor increased the risk for axillary metastases by 0.9% to 1.8%. Clinicians therefore should perform a thorough breast examination at the initial prenatal visit and without hesitation thereafter.

The differential diagnosis of a breast mass in a pregnant or lactating woman is provided in Box 56.2 . Although 80% of breast biopsies obtained from pregnant women are benign, it is important to biopsy any lump that is present for 2 to 4 weeks.