CHAPTER 30: Cervical Cancer

Cervical cancer is the most common gynecologic cancer in women worldwide. Most of these cancers stem from infection with the human papillomavirus (HPV), although other host factors affect neoplastic progression following initial infection. Compared with other gynecologic malignancies, cervical cancer develops in a younger population. Thus, screening for this neoplasia typically begins in young adulthood.

Most early cancers are asymptomatic. Thus, diagnosis usually follows histologic evaluation of biopsies taken during colposcopic examination or from a grossly abnormal cervix. This cancer is staged clinically, and this in turn directs treatment. In general, early-stage disease is effectively eradicated surgically. For advanced disease, chemoradiation is primarily selected. As expected, disease prognosis differs with tumor stage, and stage is the most important indicator of long-term survival.

Prevention lies mainly in identifying and treating women with high-grade dysplasia, and in HPV vaccination. Accordingly, as detailed in Chapter 29, regular screening is recommended and HPV vaccination is encouraged to lower rates of cervical cancer in the future.

Worldwide, cervical cancer is common, and it ranks fourth among all malignancies for women (World Health Organization, 2012). In general, higher incidences are found in developing countries, and these countries contribute 85 percent of reported cases annually. Mortality rates are similarly higher in these populations (Torre, 2015). This incidence and survival disparity highlights successes achieved by long-term cervical cancer screening programs.

In the United States, cervical cancer is the third most common gynecologic cancer and the 11th most common solid malignant neoplasm among women. Women have a 1 in 132 lifetime risk of developing this cancer. In 2015, the American Cancer Society estimated 12,900 new cases and 4100 deaths from this malignancy (Siegel, 2015). Of U.S. women, black women and those in lower socioeconomic groups have the highest age-adjusted cervical cancer death rates, and Hispanic women have the highest incidence rates (Table 30-1). This trend is thought to stem mainly from financial and cultural characteristics affecting access to screening and treatment. The age at which cervical cancer develops is in general earlier than that of other gynecologic malignancies, and the median age at diagnosis is 49 years (Howlader, 2014).

In addition to demographic differences, other risks may alter HPV acquisition or action. Notably, the greatest risk is the lack of regular cervical cancer screening (Abed, 2006; Leyden, 2005). Most communities adopting such screening have documented decreased incidences of this cancer (Jemal, 2006).

HPV is the primary etiologic infectious agent associated with cervical cancer (Ley, 1991; Schiffman, 1993). Although other sexually transmitted factors, including herpes simplex virus 2, may play a concurrent causative role, 99.7 percent of cervical cancers are associated with an oncogenic HPV subtype (Walboomers, 1999). In one study, 57 percent of invasive cervical cancer cases were attributable to HPV serotype 16. Serotype 18 was associated with 16 percent of invasive disease cases (Li, 2010). Each of these serotypes can lead to either squamous cell carcinoma or adenocarcinoma of the cervix. However, HPV 16 is more commonly associated with squamous cell carcinoma of the cervix, whereas HPV 18 is a risk factor for cervical adenocarcinoma (Bulk, 2006).

Of other risks, lower educational attainment, older age, obesity, smoking, and neighborhood poverty are independently related to lower rates of cervical cancer screening. Specifically, those living in impoverished neighborhoods have limited access to testing and may benefit from screening outreach programs (Datta, 2006).

Cigarette smoking, both active and passive, increases cervical cancer risk. Among HPV-infected women, current and former smokers have a two- to threefold increased incidence of high-grade squamous intraepithelial lesion (HSIL) or invasive cancer. Passive smoking is also associated with increased risk, but to a lesser extent (Trimble, 2005). The mechanism underlying the association between smoking and this cancer is unclear, but smoking may alter HPV infection in those who smoke. For example, “ever smoking” has been associated with reduced clearance of high-risk HPV (Koshiol, 2006; Plummer, 2003). Tobacco smoke may also alter viral oncoprotein expression in cells in which the HPV is not integrated into the host genome (Wei, 2014).

Parity has a significant association with cervical cancer. Specifically, women with seven prior full-term pregnancies have an approximately fourfold risk, and those with one or two have a twofold risk compared with nulliparas (Muñoz, 2002).

Long-term combination oral contraceptive (COC) use is another risk. In women who are positive for cervical HPV DNA and who use COCs, cervical carcinoma rates increase by up to fourfold compared with women who are HPV-positive and never users (Moreno, 2002). Additionally, current COC users and women who are within 9 years of use have a significantly higher risk of developing both squamous cell and adenocarcinoma of the cervix (International Collaboration of Epidemiological Studies of Cervical Cancer, 2006). Encouragingly, the relative risk to COC users appears to decline after cessation. Data from 24 epidemiologic studies showed that by 10 or more years following COC cessation, cervical cancer risk returns to that of never users (International Collaboration of Epidemiological Studies of Cervical Cancer, 2007).

Sexual activity logically has an association as HPV is sexually transmitted. Having more than six lifetime sexual partners elevates the relative risk of cervical cancer. Similarly, an early age of first intercourse before age 20 confers an increased risk of developing this malignancy. Intercourse after age 21 only shows a trend toward an elevated risk. Moreover, abstinence from sexual activity and barrier protection during sexual intercourse decrease cervical cancer incidence (International Collaboration of Epidemiological Studies of Cervical Cancer, 2006).

Immunosuppressed women have an increased risk of developing cervical cancer. Cervical cancer is an acquired immune deficiency syndrome (AIDS)-defining illness. The standardized incidence ratio (SIR) of developing this cancer in HIV-infected women is 5.82. For transplant recipients the SIR is 2.013 for this malignancy (Grulich, 2007). Women with autoimmune diseases who use immunosuppressants do not appear to have an increased cervical cancer risk, except for azathioprine users (Dugue, 2015).

Tumorigenesis

Most women readily clear HPV, but those with persistent infection may develop preinvasive dysplastic cervical lesions. From such lesions, squamous cell carcinoma of the cervix typically arises at the squamocolumnar junction (Bosch, 2002). In general, progression from dysplasia to invasive cancer requires several years, although times can vary widely. The molecular alterations involved with cervical carcinogenesis are complex and not fully understood. Carcinogenesis currently is suspected to result from the interactive effects among environmental insults, host immunity, and somatic-cell genomic variations (Helt, 2002; Jones, 1997, 2006; Wentzensen, 2004).

Increasing evidence suggests that HPV oncoproteins may be a critical component of continued cancer cell proliferation (Mantovani, 1999; Munger, 2001). Unlike low-risk serotypes, oncogenic HPV serotypes can integrate into human DNA (Fig. 30-1). As a result, with infection, oncogenic HPV’s early replication proteins E1 and E2 enable the virus to replicate within cervical cells. These proteins are expressed at high levels early in HPV infection. They can lead to cytologic changes detected as low-grade squamous intraepithelial (LSIL) cytologic findings on Pap smears.

Amplification of viral replication and subsequent transformation of normal cells into tumor cells may follow (Mantovani, 1999). Specifically, the viral gene products E6 and E7 oncoproteins are implicated in this transformation (Fig. 30-2). E7 protein binds to the retinoblastoma (Rb) tumor suppressor protein, whereas E6 binds to the p53 tumor suppressor protein. In both instances, binding leads to degradation of these suppressor proteins. The E6 effect of p53 degradation is well studied and linked with the proliferation and immortalization of cervical cells (Jones, 1997, 2006; Mantovani, 1999; Munger, 2001).

Tumor Spread

Following tumorigenesis, the pattern of local growth may be exophytic if a cancer arises from the ectocervix, or may be endophytic if it arises from the endocervical canal (Fig. 30-3). Lesions lower in the canal and on the ectocervix are more likely to be clinically visible during physical examination. Alternatively, growth may be infiltrative, and in these cases, ulcerated lesions are common if necrosis accompanies this growth. As primary lesions enlarge and lymphatic involvement progresses, local invasion increases and will eventually become extensive.

Lymphatic Spread

Lymph Node Groups

The pattern of tumor spread typically follows cervical lymphatic drainage. Thus, familiarity with this drainage aids understanding the surgical steps of radical hysterectomy performed for this cancer (Section 46-1). The cervix has a rich network of lymphatics, which follow the course of the uterine artery (Fig. 30-4). These channels drain principally into the paracervical and parametrial lymph nodes. These lymph nodes are clinically important and thus are removed as part of parametrial resection during radical hysterectomy. From the parametrial and paracervical nodes, lymph subsequently flows into the obturator lymph nodes and into the internal, external, common iliac lymph nodes, and ultimately the paraaortic lymph nodes. Accordingly, pelvic and common iliac lymph nodes are also traditionally removed concurrently with radical hysterectomy. In contrast, lymphatic channels from the posterior cervix course through the rectal pillars and the uterosacral ligaments to the rectal lymph nodes. These nodes are also encountered and are removed during the extended resection of the uterosacral ligaments that is characteristic of radical hysterectomy.

Lymphovascular Space Involvement

As tumor invades deeper into the stroma, it enters blood capillaries and lymphatic channels (Fig. 30-5). Termed lymphovascular space involvement (LVSI), this type of invasive growth is not included in the clinical staging of cervical cancer. However, its presence is regarded as a poor prognostic indicator, especially in early-stage cervical cancers. Thus, the presence of LVSI often requires tailoring the planned surgical procedure and adjuvant radiation treatment.

Local and Distant Tumor Extension

With extension through the parametria to the pelvic sidewall, ureteral blockage frequently develops, resulting in hydronephrosis (Fig. 30-6). Additionally, the bladder may be invaded by direct tumor extension through the vesicouterine ligaments (bladder pillars). The rectum is invaded less often because it is anatomically separated from cervix by the posterior cul-de-sac. Distant metastasis results from hematogenous dissemination, and the lungs, ovaries, liver, and bone are the most frequently affected.

Squamous Cell Carcinoma

The two most common histologic subtypes of cervical cancer are squamous cell and adenocarcinoma (Table 30-2). Of these, squamous cell tumors predominate, comprise about 70 percent of all cervical cancers, and arise from the ectocervix. Over the past 30 years, the incidence of squamous cell cancers has declined, whereas that of cervical adenocarcinoma has risen. These changes may be attributed to an improved method of screening for early squamous lesions of the cervix and an increase in HPV prevalence (Vizcaino, 2000). Squamous cell carcinomas can be subdivided into keratinizing and nonkeratinizing carcinomas (Fig. 30-7).

Adenocarcinomas

Adenocarcinomas are a group of cervical cancers composed of the subtypes listed in Table 30-2. In contrast to squamous cell cervical carcinoma, adenocarcinomas make up 25 percent of cervical cancers and arise from the endocervical mucus-producing columnar cells. Because of this origin within the endocervix, adenocarcinomas are often occult and may be advanced before becoming clinically evident. They often give the cervix a palpable barrel shape during pelvic examination.

Adenocarcinomas exhibit various histologic patterns composed of diverse cell types. Of these, mucinous adenocarcinomas are the most common and are subdivided as shown in Table 30-2. The mucinous endocervical type retains resemblance to normal endocervical tissue (Fig. 30-8). The intestinal type resembles intestinal cells and may include goblet cells. Minimal deviation adenocarcinoma, also known as adenoma malignum, is characterized by cytologically bland glands that are abnormal in size and shape. These tumors contain an increased number of glands positioned at a deeper level than normal endocervical glands. Women with Peutz-Jeghers syndrome are at increased risk of developing adenoma malignum. Villoglandular adenocarcinomas are made up of surface papillae. Endometrioid adenocarcinomas are the second most frequently identified and display glands resembling those of the endometrium. Serous carcinoma is identical to serous carcinomas of the ovaries or uterus and is rare. Clear cell adenocarcinoma accounts for less than 5 percent of cervical adenocarcinomas and is named for its clear cytoplasm (Jaworski, 2009). Rarely, adenocarcinomas arise in mesonephric remnants in the cervix and are termed mesonephric adenocarcinomas.

Prognosis Comparison

Evidence describing the prognosis of squamous cell carcinoma compared with that of adenocarcinoma is contradictory. For stage IB and IIA cervical cancer, one study showed a statistically significant lower overall survival rate in those with adenocarcinoma compared with women with squamous cell carcinoma (Landoni, 1997). However, the Gynecologic Oncology Group (GOG) found in a subsequent study that overall survival rates in women with stage IB squamous and adenocarcinomas of the cervix are similar (Look, 1996). For advanced-stage (stage IIB to IVA) cancers, evidence suggests that cervical adenocarcinomas may portend a poorer overall survival rate compared with similarly staged squamous cell carcinomas (Eifel, 1990; Lea, 2002). The 2006 International Federation of Obstetricians and Gynecologists (FIGO) annual report, which reported on more than 11,000 squamous carcinomas and 1613 adenocarcinomas, demonstrated that women with adenocarcinomas have worse overall survival rates at every stage compared with those with squamous cell carcinoma (Quinn, 2006). In sum, evidence suggests that adenocarcinoma of the cervix is a high-risk cell type.

Other Tumor Types

Mixed cervical carcinomas are rare. Of these, adenosquamous carcinomas do not differ grossly from adenocarcinomas of the cervix. The squamous component is poorly differentiated and shows little keratinization. Glassy cell carcinoma describes a form of poorly differentiated adenosquamous carcinoma in which cells display cytoplasm with a ground-glass appearance.

Neuroendocrine tumors of the cervix include large cell and small cell tumors. These rare malignancies are highly aggressive, and even early-stage cancers have a relatively low disease-free survival rate despite treatment with radical hysterectomy and adjuvant chemotherapy (Albores-Saavedra, 1997; Viswanathan, 2004). Often, neuroendocrine markers, including chromogranin, synaptophysin, and CD56, are used to confirm the diagnosis. Uncommonly, endocrine and paraendocrine tumors are associated with these neuroendocrine tumors.

Of other rare types, the cervix may be the site of sarcomas, malignant lymphomas, and melanomas. Cervical leiomyosarcomas and cervical stromal sarcomas have a poor prognosis, similar to that for uterine sarcomas. Melanomas often present as ulcerated blue or black nodules and also have a poor prognosis.

Symptoms

Some women diagnosed with this cancer are asymptomatic. In others, early-stage cervical cancer may create a watery, blood-tinged vaginal discharge. Intermittent vaginal bleeding that follows coitus or douching can also be noted. As a malignancy enlarges, bleeding typically intensifies, and occasionally, a woman presents with uncontrolled hemorrhage from a tumor bed. In such cases, bleeding can often be controlled with a combination of Monsel paste (ferric subsulfate) and vaginal packing. Topical acetone can also be used to obtain hemostasis, especially in cases refractory to Monsel paste (Patsner, 1993). Burning from acetone makes it a less desirable choice than Monsel paste. With vaginal packing, the patient ideally remains at bed rest, and a Foley catheter is concurrently inserted to drain the bladder. The pack may interfere with normal voiding, and the catheter also allows for accurate monitoring of urine output during volume repletion. If bleeding continues, emergent radiation can be delivered. Alternatively, internal iliac artery embolization or ligation can be performed in cases of refractory hemorrhage. However, caution is used for these latter two options, as tumor oxygenation is decreased if these blood supplies are occluded. Radiation therapy is a primary component of advanced-stage cancer, and as noted in Chapter 28, radiation’s effects are diminished in low-oxygen environments and can translate into worsening disease-specific survival rates (Kapp, 2005). In those with significant bleeding, hemodynamic support of the patient follows that described in Chapter 40.

Physical Examination

Most women with this cancer have normal general physical examination findings. In those with suspected cervical cancer, a thorough external genital and vaginal examination is performed. Because HPV is a shared risk factor for cervical, vaginal, vulvar, and anal cancers, concomitant lesions are sought. With speculum examination, the cervix may appear grossly normal if cancer is microinvasive. Visible disease displays varied appearances. Lesions may be an exophytic or endophytic growth; a polypoid mass, papillary tissue, or barrel-shaped cervix; a cervical ulceration or granular mass; or necrotic tissue. A watery, purulent, or bloody discharge can also be seen. For this reason, cervical cancer may mirror the appearance of a cervical leiomyoma, cervical polyp, vaginitis, cervical eversion, cervicitis, threatened abortion, placenta previa, cervical pregnancy, condyloma acuminata, herpetic ulcer, chancre, or a prolapsing uterine leiomyoma, polyp, or sarcoma.

During bimanual examination, a clinician may palpate an enlarged uterus resulting from tumor invasion and growth. Alternatively, hematometra or pyometra may expand the endometrial cavity following obstruction of fluid egress by a primary cervical cancer. In this case, the uterus may feel enlarged and boggy. Advanced cervical cancer cases can extend into the vagina, and disease extent can be appreciated during anterior vaginal wall palpation or during rectovaginal examination. With posterior spread, palpation of the rectovaginal septum between the index and middle finger of an examiner’s hand reveals a thick, hard, irregular septum. The proximal posterior vaginal wall is most commonly invaded. In addition, during digital rectal examination, parametrial, uterosacral, and pelvic sidewall involvement can be appreciated. Either one or both parametria may be invaded, and involved tissues feel thick, irregular, firm, and less mobile. A fixed mass indicates that tumor has probably extended to the pelvic sidewalls. However, a central lesion can become as large as 8 to 10 cm in diameter before reaching the sidewall.

With advancing disease, enlarged supraclavicular or inguinal lymphadenopathy suggest lymphatic tumor spread. Lower extremity edema and low back pain, often radiating down the posterior leg, may reflect compression of the sciatic nerve root, lymphatics, veins, or ureter by an expanding tumor. With ureteral obstruction, hydronephrosis and uremia can follow and may occasionally be the initial presenting finding. In these cases, ureteral stenting or percutaneous nephrostomy tube insertion are usually required. Kidney function is ideally preserved for chemotherapy. Additionally, with tumor invasion into the bladder or rectum, hematuria and/or vesicovaginal or rectovaginal fistula may be found.

Papanicolaou Test and Cervical Biopsy

Histologic evaluation of cervical biopsy is the primary tool to diagnose cervical cancer. Although Papanicolaou (Pap) tests are performed extensively to screen for this cancer, this test does not always detect cervical cancer. Specifically, Pap testing has only a 53- to 80-percent sensitivity for detecting high-grade lesions on any given single test (Agorastos, 2015; Benoit, 1984; Soost, 1991). Thus, the preventive power of Pap testing lies in regular serial screening (Fig. 30-9). Moreover, in women who have stage I cervical cancer, only 30 to 50 percent of single cytologic smears obtained are read as positive for cancer (Benoit, 1984). Hence, Pap testing alone for evaluation of a suspicious lesion is discouraged. Instead, these lesions are directly biopsied with Tischler biopsy forceps or a Kevorkian curette (Fig. 29-16). When possible, biopsies are taken from the tumor periphery, as central portions often contain only necrotic tissue, which will fail to yield a diagnosis. Moreover, biopsies ideally include underlying stroma, so that invasion, if present, can be assessed.

If abnormal Pap test findings are noted, colposcopy is often performed, and adequate cervical and endocervical biopsies are obtained. In some cases, cold knife conization is needed for this. Indications for colposcopy and conization are outlined in Chapter 29, and cervical punch biopsies or conization specimens are the most accurate for allowing assessment of cervical cancer invasion. Both sample types typically contain underlying stroma and enable differentiation between invasive and in situ carcinomas. For this determination, conization specimens provide a larger tissue sample and are most helpful.

Cervical cancers are staged clinically. Allowable components of staging include cold-knife conization, pelvic examination under anesthesia, cystoscopy, proctoscopy, chest radiograph, and intravenous pyelogram (or this portion of the computed tomography [CT] scan can be used). Table 30-3 lists these and also contains radiologic and laboratory tools that are not included in formal staging but may contribute additional information. Bullous edema is not sufficient for the diagnosis of bladder involvement, and this involvement must be biopsy proven. Lymph node metastasis does not change the clinical stage. The staging system widely used for cervical cancer is that developed by the FIGO in collaboration with the World Health Organization (WHO) and the International Union Against Cancer (UICC). This staging was updated in 2009 and is detailed in Table 30-4 and Figure 30-10. In this chapter, early-stage disease refers to FIGO stages I through IIA. The term advanced-stage disease describes stages IIB and higher.

Within each oncology chapter of this text, staging of each cancer type (cervix, vulva, vagina, uterus, ovary) will be presented. FIGO classification is used for gynecologic cancers. In contrast, the American Joint Committee on Cancer (AJCC) developed the TNM Staging System, which is based on the extent of the tumor (T), the extent of spread to the lymph nodes (N), and the presence of metastasis (M). Breast cancer is staged with this latter system, as shown in Table 12-5.

As discussed, cervical cancer is staged clinically, and accurate evaluation is critical to appropriate treatment planning. For example, early-stage tumors may be treated surgically, whereas more advanced tumors require radiation and/or chemotherapy. Although imaging does not affect assignment of stage (except lung metastases seen on chest radiograph and hydronephrosis seen on CT scan), imaging results can tailor treatment for an individual. In addition, lymph node metastases, although not included in the FIGO system, worsen patient prognosis and may be identified with imaging. Thus, radiologic tools such as CT scanning, magnetic resonance (MR) imaging, or positron emission tomography (PET) scanning are commonly used as adjuncts in initial cervical cancer evaluation. However, no uniform approach to the use of these has been developed.

Magnetic Resonance Imaging

For defining anatomy, this high-resolution imaging tool offers superior contrast resolution at soft-tissue interfaces. Thus, MR imaging effectively measures tumor size, delineates cervical tumor boundaries, and identifies surrounding bladder, rectal, or parametrial invasion. Unfortunately, MR imaging is less accurate for diagnosing microscopic or deep cervical stromal invasion or identifying minimal parametrial extension (Mitchell, 2006). These particular distinctions are important as both stromal and parametrial invasion may affect suitability for a given planned treatment. In addition, false-negative findings occur with small volumes of disease and with tissue foci in which cancer cannot be differentiated from scar or necrosis. In these cases, PET scanning identifies metabolic rather than anatomic changes and can be a complementary tool.

For primary cervical cancer, MR imaging is superior to CT for determining carcinoma size, local tumor extension, and lymph node involvement (Bipat, 2003; Mitchell, 2006; Subak, 1995). MR imaging is often preferred for patients being considered for fertility-sparing radical trachelectomy (Abu-Rustum, 2008; Olawaiye, 2009). Overall, however, both MR imaging and CT perform similarly in cervical cancer (Hricak, 2005).

Computed Tomography

This is the most widely used imaging tool for the assessment of nodal involvement and distant metastatic disease. It offers high-resolution depiction of anatomy, especially when used with contrast. CT scanning is not a component of FIGO staging. However, it is obtained in many women with cervical cancer to evaluate tumor size and bulky extension beyond the cervix. CT can also aid detection of enlarged lymph nodes, ureteral obstruction, or distant metastasis (Follen, 2003).

However, CT has limitations similar to MR imaging. CT is not accurate for assessing subtle parametrial invasion or deep cervical stromal invasion because of its poor soft-tissue contrast resolution. CT is also limited by its inability to detect small-volume metastatic involvement in normal-size lymph nodes. Moreover, internal node architecture is often poorly defined. This makes distinction between reactive node hyperplasia and true metastatic disease difficult.

Positron Emission Tomography

This nuclear medicine imaging technique creates an image of functional processes within the body. With FDG-PET, a radiolabeled analogue of glucose, fluorodeoxyglucose (FDG), is injected intravenously and is taken up by metabolically active cells such as tumor cells. PET provides a poor depiction of detailed anatomy, thus scans are frequently read side-by-side with CT scans. The combination allows correlation of metabolic and anatomic data. As a result, current PET scanners are now commonly integrated with CT scanners, and the two scans can be performed during the same session (Fig. 2-33).

FDG-PET is superior to CT or MR imaging for lymph node metastasis identification (Belhocine, 2002; Havrilesky, 2005; Selman, 2008). However, PET is insensitive for lymphatic metastasis <5 mm. Moreover, its role in early-stage smaller, resectable tumors is limited (Sironi, 2006; Wright, 2005). PET scans can be useful in planning radiation treatment fields and also in identifying those patients who have distant metastatic disease and are candidates for palliative chemotherapy rather than curative-intent chemoradiotherapy.

As noted, cervical cancer is staged clinically and not surgically. However, surgical evaluation of retroperitoneal pelvic and paraaortic lymph nodes offers accurate metastasis detection that is superior to radiologic imaging (Goff, 1999). As a result, lymph node dissection may modify a patient’s primary treatment strategy based on the level of nodal disease. For example, radiation fields may be altered to ensure that patients with negative paraaortic lymph nodes are not overtreated with extended-field radiation and that patients with positive paraaortic lymph nodes are not undertreated. Potential candidates include patients with positive or suspected positive pelvic nodes undergoing chemoradiation treatment. Supporting studies show that if positive pelvic/paraaortic nodes are identified, patients may receive a significant survival benefit from extended chemotherapy and/or extended field radiation therapy (Hacker, 1995; Holcomb, 1999; Leblanc, 2007).

In addition, tumor-laden nodes may be debulked. Several studies report similar disease-free survival rates for patients whose macroscopic nodal disease is resected compared with women with microscopic nodal disease (Cosin, 1998; Downey, 1989; Hacker, 1995). That said, there is virtually no long-term survival for patients with unresectable bulky paraaortic lymph nodes.

During lymphadenectomy, most experts recommend lymph node dissection in the common iliac and paraaortic region and resection of macroscopic lymph nodes (Querleu, 2000). Traditional laparotomy and minimally invasive surgery (MIS) approaches have been compared. Although diagnostically equivalent, laparoscopic approaches offer postoperative MIS advantages. In addition, laparoscopic lymph node dissection has been associated with significantly less radiation morbidity than that with radiation following laparotomic approaches (Vasilev, 1995).

Despite these suggested benefits, some experts argue that the benefits of surgical staging, if any, are minimal. These studies estimate only a 4- to 6-percent survival benefit after aggressive surgical debulking of retroperitoneal lymph nodes (Kupets, 2002; Petereit, 1998).

The significance of tumor burden on survival is well demonstrated, whether measured by FIGO stage, centimeter size, or surgical staging (Stehman, 1991). Of these definers, FIGO stage is the most significant prognostic factor (Table 30-5). However, within each stage distribution, lymph node involvement also becomes an important modifier in determining prognosis. For example, in early-stage cervical cancer (stages I through IIA), nodal metastases are an independent predictor of survival (Delgado, 1990; Tinga, 1990). One GOG study demonstrated a 3-year survival rate of 86 percent for women with early-stage cervical cancer and negative pelvic lymph nodes, compared with a 3-year survival rate of 74 percent in patients who had one or more positive lymph nodes (Delgado, 1990).

In addition, the number of nodal metastases is predictive. Studies demonstrate significantly higher 5-year survival rates in those with one positive lymph node compared with rates in women with multiple involved nodes (Tinga, 1990). In advanced-stage (stage IIB through IV) cervical cancer, lymph node metastases also worsen prognosis. In general, microscopic nodal involvement has a better prognosis than macroscopic nodal disease (Cosin, 1998; Hacker, 1995).

Stage IA

The term microinvasive cervical cancer identifies this subgroup of small tumors. By definition, these tumors are not visible to the naked eye. Specifically, as seen in Table 30-4, criteria for stage IA tumors limit invasion depth to no greater than 5 mm and lateral spread to no wider than 7 mm. Microinvasive cervical cancer carries a minor risk of lymph node involvement and excellent prognosis following treatment. A retrospective study compared tumors with horizontal spread ≤7 mm and those with >7 mm spread. Higher rates of pelvic lymph node metastasis and recurrence rates were noted as tumor spread further than 7 mm (Takeshima, 1999).

Stage IA tumors are further divided into IA1 and IA2. These cancers are subdivided to reflect increasing depth and width of invasion and increasing risks for lymph node involvement.

Stage IA1

These microinvasive tumors invade no deeper than 3 mm, spread no wider than 7 mm, and are associated with the lowest risk for lymph node metastasis. Squamous cervical cancers with stromal invasion less than 1 mm have a 1-percent risk of nodal metastasis, and those with 1 to 3 mm of stromal invasion carry a 1.5-percent risk. Of 4098 women studied with this tumor stage, less than 1 percent died of disease following surgery (Ostor, 1995). Because of the low risk of spread into the parametrial or uterosacral nodes, these lesions may be effectively treated with cervical conization alone (Table 30-6) (Keighley, 1968; Kolstad, 1989; Morris, 1993; Ostor, 1994). However, a total extrafascial hysterectomy (type I hysterectomy) is preferred for women who have completed childbearing. Hysterectomy types are described in Table 30-7.

In stage IA1 microinvasive cancers, the presence of LVSI increases the risk of lymph node metastasis and cancer recurrence to approximately 5 percent. Accordingly, at our institution, these cases are traditionally managed with modified radical hysterectomy (type II hysterectomy) and pelvic lymphadenectomy. Radical trachelectomy with pelvic lymph node dissection can be considered in women desiring fertility preservation (Olawaiye, 2009). This is described on page 670.

Adenocarcinomas are typically diagnosed at a more advanced stage than squamous cell cervical cancers. Thus, microinvasive adenocarcinomas present a unique management dilemma, due to sparse data regarding this tumor stage. However, based on evaluation of Surveillance Epidemiology and End Result (SEER) data provided by the National Cancer Institute, the incidence of lymph node involvement is similar to that with squamous cancers (Smith, 2002; Spoozak, 2012). Of microinvasive cervical adenocarcinomas, 59 cases managed with uterine preservation and conization have been reported in the literature (Baalbergen, 2011; Bisseling, 2007; Ceballos, 2006; McHale, 2001; Reynolds, 2010; Schorge, 2000; Yahata, 2010). Of these cases, following conization, no recurrences were identified during surveillance in women without LVSI. According to SEER data, the 5-year overall survival for women with stage IA1 adenocarcinoma treated with conization is 98 percent (Spoozak, 2012).

Stage IA2

These microinvasive cervical lesions have 3 to 5 mm of stromal invasion, have a 7-percent risk of lymph node metastasis, and carry a >4-percent risk of disease recurrence. In this group of women, the safety of conservative therapy is yet to be proven. Thus, for this degree of invasion, radical hysterectomy and pelvic lymphadenectomy is recommended.

For fertility preservation, stage IA2 squamous cervical lesions may be treated with radical trachelectomy and lymphadenectomy. A nonabsorbable cerclage may be placed concurrently with such radical trachelectomy to improve cervical competence during pregnancy. These procedures have high cure rates, and successful pregnancies have been reported. If women are carefully selected for age <45 years, smaller tumor size (<2 cm), and negative nodal involvement, then reported recurrence rates are similar to those of radical hysterectomy (Burnett, 2003; Covens, 1999a,b; Gien, 2010; Olawaiye, 2009). Some experts will offer radical trachelectomy to patients with tumors up to 4 cm (stage IB1). However, prior to surgery, approximately one third of patients with this tumor stage will instead be found to need radical hysterectomy or adjuvant chemoradiation due to intermediate- or high-risk features (Abu-Rustum, 2008; Gien, 2010). Preoperative MR imaging to evaluate the parametria and/or CT scan to evaluate extracervical disease is recommended in these cases. If tumor has extended proximally past the internal cervical os, then trachelectomy is contraindicated. Although this technique is promising, it carries a learning curve, and further studies to validate its efficacy are needed.

In addition to stage IA1 tumors, some centers are evaluating the safety of conization or extrafascial hysterectomy for a broader group of women with early-stage cervical cancer, since parametrial involvement in microinvasive cervical cancer is rare (Hou, 2011). One study, which included 51 women with stage IA1 to stage IB1 cervical cancer, demonstrated no recurrences at a median surveillance of 21 months for women treated with conization or extrafascial hysterectomy and no nodal dissection (Bouchard-Fortier, 2014). Two women received adjuvant chemoradiation based on specimen histologic analysis results. In addition, SEER data that included 3987 women with microinvasive cervical cancer showed similar survival rates for women with adenocarcinoma treated with conization compared with hysterectomy. However, women with squamous cell carcinoma undergoing hysterectomy had improved survival rates compared with women undergoing conization (Spoozak, 2012).

Alternatively, patients with microinvasive carcinoma (stages IA1 and IA2) can be treated with intracavitary brachytherapy alone with excellent results (Grigsby, 1991; Hamberger, 1978). Potential candidates for vaginal brachytherapy include women who are elderly or who are not surgical candidates due to comorbid medical disease.

Hysterectomy

Women with FIGO stage IA2 through IIA cervical cancer, that is, those without obvious parametrial involvement, may be selected for radical hysterectomy with pelvic lymph node dissection and with or without paraaortic lymph node dissection. Surgery is appropriate for those who are physically able to tolerate an aggressive surgical procedure, those who wish to avoid the long-term effects of radiation therapy, and/or those who have contraindications to pelvic radiotherapy. Typical candidates include young patients who desire ovarian preservation and retention of a functional, nonirradiated vagina.

Historically, there are five types of hysterectomy, as described by Piver and colleagues (1974). However, hysterectomy techniques used clinically today vary depending on the degree of surrounding tissue that is resected and are categorized as type I, II, or III (see Table 30-7).

Type I hysterectomy, also known as an extrafascial hysterectomy or simple hysterectomy, removes the uterus and cervix, but does not require excision of the parametrium or paracolpium. It is appropriately selected for benign gynecologic pathology, preinvasive cervical disease, and stage IA1 cervical cancer.

Type II hysterectomy, also known as modified radical hysterectomy, removes the cervix, proximal vagina, and parametrial and paracervical tissue. This hysterectomy is well suited for tumors in patients with stage IA1 cervical cancer who have positive margins following conization and have insufficient cervix to repeat conization. This hysterectomy is also appropriate for patients with stage IA1 cervical cancer with LVSI. Some institutions perform type II hysterectomies in women with stage IA2 tumors and smaller stage IB tumors with good outcomes (Landoni, 2001).

Type III hysterectomy, also known as radical hysterectomy, requires greater resection of the parametria. Its goal is to remove microscopic disease that has extended into the parametrium, paracolpium, and around the uterosacral ligaments. To summarize surgical steps, the uterine arteries are ligated at their origin from the internal iliac arteries near the pelvic sidewall, and all tissue medial to this origin, that is, the parametrium, is resected (Fig. 30-11) (Section 46-1). The ureters are completely dissected from their beds and moved laterally for protection during wide excision of the parametrium and paracolpium. The bladder and rectum are mobilized caudally and off the vagina to permit resection of ≥2 cm of proximal vagina. The uterosacral ligaments are clamped at their midpoint. This procedure is performed for stage IA2, stage IB1, stage IIA1, and for some stage IB2 lesions, and for patients with relative contraindications to radiation. These contraindications include diabetes, pelvic inflammatory disease, hypertension, collagen disease, inflammatory bowel disease, or adnexal masses.

The approach for type I, II, and III hysterectomies can be abdominal, laparoscopic, robot-assisted, or vaginal, depending on patient characteristics and surgeon experience. Advantages of MIS include less blood loss and shorter hospital stay. Intra- and postoperative complications are similar regardless of approach (Ramirez, 2008). Long-term follow-up of patients undergoing laparoscopic radical hysterectomy demonstrates excellent overall survival rates (Lee, 2010).

Radical Trachelectomy

This surgical option can preserve fertility in selected young women with cervical cancer, and the cancer stages appropriate for radical trachelectomy mirror those for radical hysterectomy. Compared with radical hysterectomy, radical trachelectomy is less often performed.

Radical trachelectomy was originally completed vaginally, as described by Dargent (2000), but an abdominal approach is now used more commonly (Abu-Rustum, 2006). The abdominal approach allows for a larger resection of the parametria and is suitable for patients with larger tumors (>2 cm). With radical trachelectomy, steps of radical hysterectomy proceed and thus the uterine vessels are ligated, the parametria is resected, ureterolysis is completed, the bladder and rectum are mobilized, and the upper vagina is resected. To remove the cervix, the uterus is incised at or just below the level of the internal os, with the goal to leave 5 mm of endocervix still attached to the uterus. At this remaining endocervical margin, a thin tissue sample is sharply excised, termed a shave margin, and sent for frozen section. If cancer is absent in this specimen, then reconstruction may proceed. For this, a cerclage using permanent suture is placed, and the knot is tied posteriorly. The uterus is then stitched to the vagina using absorbable sutures. From each side, the corpus ultimately retains blood supply through the uterine branch of the ovarian artery.

Following radical trachelectomy, women continue to menstruate, and conception can occur naturally. However, cervical stenosis may develop, and thus intrauterine insemination or in vitro fertilization is often needed. Pregnancies are frequently complicated by second-trimester loss and higher rates of preterm birth (Plante, 2005; Shepherd, 2008). In a review of 485 women for whom a radical abdominal trachelectomy was planned, 47 cases (10 percent) were converted to radical hysterectomy. Another 25 women required adjuvant therapy based on final pathologic specimen findings. Thus, 413 women (85 percent) retained fertility. In this fertile cohort, there were 75 pregnancies, 18 miscarriages, 47 deliveries (19 term, 12 preterm, 16 not stated), and 10 women were pregnant at the time of publication (Pareja, 2013). Cesarean delivery with a classical incision is recommended.

Stage IB to IIA

Stage IB lesions are defined as those extending past the limits of microinvasion yet still confined to the cervix. This stage is subcategorized either as IB1 if tumors measure ≤4 cm or as IB2 if they measure >4 cm (Fig. 30-12).

Stage II cancers extend outside the cervix. They may invade the upper vagina and the parametria but do not reach the pelvic sidewalls. Stage IIA tumors have no parametrial involvement but do extend vaginally as far as the proximal two thirds of the vagina. Stage IIA is further subdivided into stage IIA1 for tumor size ≤4 cm and IIA2 for tumor size >4 cm. Stage IIB cancer may invade the vagina to a similar extent and also invade the parametria.

Treatment

Stage IB to IIA cancers do not extend into the parametria and thus can be managed with either surgery or chemoradiation. In a prospective study of primary therapy, 393 women were randomly assigned to undergo radical hysterectomy and pelvic lymphadenectomy or receive primary radiation therapy. Five-year overall survival and disease-free survival rates were statistically equivalent (83 percent and 74 percent, respectively). Patients who underwent radical surgery followed by radiation had the worst morbidity (Landoni, 1997).

Because chemoradiation and surgery are both viable options, the optimum treatment for each woman ideally assesses clinical factors such as menopausal status, age, concurrent medical illness, tumor histology, and cervical diameter. For stage IB1 and IIA1 cervical cancers, it is left to the physician’s discretion and patient preference as to which treatment modality is preferred. Our general approach to patients with bulky stage IB2 or stage II cervical cancers, that is, those measuring >4 cm, is to manage them primarily with chemoradiation, in a similar fashion to advanced-stage cervical cancers.

In general, radical hysterectomy for stage IB through IIA tumors is usually selected for premenopausal women who wish to preserve ovarian function and for women who have concerns about altered sexual functioning following radiotherapy. Age and weight are not contraindications to surgery. However, in general, older women may have longer hospital stays, and heavier women can have longer operative time, greater blood loss, and higher rates of wound complications. Surgery is contraindicated in patients with severe cardiac or pulmonary disease.

In those electing surgery, oophorectomy may be deferred in younger women. One GOG study evaluated tumor spread to the ovary in those with IB tumors electing radical hysterectomy without adnexectomy. Ovarian metastases were identified in only 0.5 percent of 770 women with stage IB squamous cell cancers and in 2 percent of those with adenocarcinomas (Sutton, 1992). For those electing ovarian preservation, ovarian transposition, accomplished by oophoropexy of the ovary into the upper abdomen, can be performed during radical hysterectomy. This repositioning helps preserve ovarian function, in case postoperative pelvic radiation is indicated. In addition, to reduce complications from radiotherapy that might be needed following radical hysterectomy, a surgeon may perform an omental J-flap. Namely, after surgery, the small bowel may become fixed in the pelvis by adhesions, which renders it vulnerable to radiation damage. The omental J-flap can fill the pelvis to reduce this adhesion risk and is described in Section 46-14.

Systematic lymphadenectomy can lead to complications such as lymphocyst and lymphedema. Therefore, in women with cervical cancer, sentinel lymph node mapping to assess lymphatic spread while avoiding extensive nodal resection is attractive. As a review, the sentinel node is the first node(s) receiving lymphatic drainage from a given tumor. To find this node, either blue dye or a technetium radioactive tracer or both are separately injected preoperatively into the cervix. At surgery, the sentinel node is stained blue and emits radioactivity discernible by Geiger counter. From a metaanalysis including 67 studies, the pooled sentinel node detection rate was 89 percent and sensitivity was 90 percent. Both were highest in women injected with both radiotracer and blue dye. Smaller tumor size (<2 cm) and early-stage diseases were associated with the highest sensitivity and detection rate. At this time, sentinel node for cervical cancer remains experimental (Kadkhodayan, 2015).

Weighing Surgical and Radiotherapy Complications

Complications for early-stage cervical cancer radical surgery include ureteral stricture, ureterovaginal fistula, vesicovaginal fistula, bladder dysfunction, constipation, wound breakdown, lymphocyst, and lymphedema. The risk of venous thromboembolism warrants chemoprophylaxis and/or sequential compression devices as outlined in Table 39-8. If radiotherapy is added as an adjuvant to surgery, the risk of many of these is increased.

On the other hand, radiation therapy also carries long-term complications described in Chapter 28. Of these, altered sexual function secondary to a shortened vagina, dyspareunia, psychological factors, and vaginal stenosis are common. Late urinary and bowel complications such as fistula formation, enteritis, proctitis, and bowel obstruction may also develop following radiotherapy.

Positive Pelvic Lymph Nodes

Approximately 15 percent of patients with stage I through IIA cervical cancers will have positive pelvic nodes. Risk factors for lymph node involvement include those listed in Table 30-8. Of those with involved nodes, 50 percent will have grossly positive pelvic nodes intraoperatively. In most cases involving grossly positive nodes, radical hysterectomy is abandoned. After recovering from surgery, whole-pelvic radiation and brachytherapy with concomitant chemotherapy is administered. The 50 percent of patients with involved nodes not grossly identified intraoperatively are considered to be at high risk of recurrence following their radical hysterectomy. As described subsequently, these women require postoperative adjuvant chemoradiation therapy.

Recurrence Risk

For women who have completed radical surgery for early-stage cervical cancer, the GOG has defined risk factors to help identify women for tumor recurrence. Intermediate risk describes those who on average would have a 30-percent risk of cancer recurrence within 3 years. Factors included in this model are depth of tumor invasion, clinical tumor diameter, and LVSI.

To determine appropriate treatment, patients with these intermediate-risk factors have been studied. In one trial, women were randomly assigned to receive pelvic radiation therapy following radical hysterectomy or to undergo radical hysterectomy and observation. A nearly 50-percent reduced risk of recurrence was found in those who received postoperative adjuvant radiation therapy (Sedlis, 1999). However, this adjuvant radiation does not prolong overall survival. Notably, these patients did not receive chemoradiation. In our practice, these intermediate-risk patients are counseled regarding their recurrence risk and offered the option of adjuvant chemoradiation therapy. A GOG clinical trial (GOG #263) that is assessing chemoradiation in this patient population is ongoing.

A high-risk category of early-stage cervical cancer patients who underwent radical surgery has also been described. High-risk is defined as a 50- to 70-percent risk of recurrence within 5 years. These women have positive lymph nodes, positive surgical margins, or microscopically positive parametria (Peters, 2000). This group is routinely offered adjuvant radiation therapy. Moreover, the GOG demonstrated that the addition of concurrent chemotherapy significantly prolongs disease-free and overall survival rates in this group of women with high-risk early-stage cancer (Peters, 2000).

Adjuvant Hysterectomy Following Primary Radiation

Treating bulky stage I (IB2) cervical cancers with adjuvant hysterectomy after radiation therapy has been evaluated. Adjuvant hysterectomy reduces locoregional relapse but does not contribute to an overall improvement in survival rates. However, initial lesion size may affect efficacy. In one study, those with tumors measuring <7 cm who underwent postradiation hysterectomy survived longer than did women with equivalent tumors in the radiation-only regimen group. In contrast, those with lesions ≥7 cm who underwent postradiation hysterectomy fared worse than their counterparts receiving only radiotherapy (Keys, 2003).

Early-stage Cervical Adenocarcinoma

These cancers may be more radioresistant than squamous cell cervical carcinomas. Although some prefer radical hysterectomy to radiotherapy, studies suggest equivalent survival rates with either (Eifel, 1991, 1995; Hopkins, 1988; Nakano, 1995). However, larger lesions may not regress if managed by radiation alone (Leveque, 1998; Silver, 1998). The centers of bulky tumors may be less radiosensitive due to relative cellular hypoxia. This effect underscores the advantages of radical hysterectomy for women with stage I cervical adenocarcinoma.

Stages IIB through IVA

Advanced-stage cervical cancers extend past the confines of the cervix and often involve adjacent organs and retroperitoneal lymph nodes. If untreated, these tumors progress rapidly. Treatment for these tumors is individualized, yet most advanced-stage tumors have a poor prognosis and 5-year survival rates are <50 percent.

Radiation Therapy

This modality forms the cornerstone of advanced-stage cervical cancer management. Both external beam pelvic radiation and brachytherapy are typically delivered (Chap. 28). Of these, external beam radiation usually precedes intracavitary radiation, which is one form of brachytherapy. External beam radiation is commonly administered in 25 fractions during 5 weeks (40 to 50 Gy). During evaluation, if paraaortic nodal metastases are found, then extended-field radiation can be added to treat these affected lymph nodes.

During brachytherapy, bowel and bladder are packed away from the intracavitary source using vaginal packing during tandem insertion to limit radiation to these organs. Treatment is often prescribed to point A, that is, a point 2 cm lateral and 2 cm superior to the external cervical os, and to point B, which is a point 3 cm lateral to point A. Side effects during and following radiation therapy are common, and these are discussed in Chapter 28.

Chemoradiation

Current evidence indicates that chemotherapy given concurrently with radiation therapy significantly improves overall and disease-free survival rates of women with cervical cancer. Chemoradiation is also associated with superior survival rates compared with pelvic and extended-field paraaortic region irradiation alone (Morris, 1999). After five trials demonstrated improved survival rates, it is now recommended that cisplatin-based chemotherapy should be considered in women undergoing radiation for cervical cancer (Keys, 1999; Morris, 1999; Peters, 2000; Rose, 1999; Whitney, 1999).

Of chemotherapy agents, cisplatin-containing regimens have been associated with the best survival rates (Rose, 1999; Whitney, 1999). The characteristics of this agent are described in Chapter 27, and Figure 30-13 describes its tumoricidal action. Nonplatinum regimens also have activity but have not been directly compared with cisplatin-containing regimens (Vale, 2008). At our institution, cisplatin is given weekly for 5 weeks. It is administered concurrently with external beam radiation and with brachytherapy. Unfortunately, the recurrence risk remains as high as 40 percent after chemoradiation given for curative intent (Chemoradiotherapy for Cervical Cancer Meta-Analysis Collaboration, 2009). Therefore, a large international randomized Phase III trial is currently evaluating whether the addition of adjuvant chemotherapy after completion of chemoradiation improves overall survival rates for women with stage IB1 with positive nodes and for those with stage IB2 through IVA disease.

Pelvic Exenteration for Primary Disease

This ultraradical surgery encompasses removal of the bladder, rectum, uterus, fallopian tubes and ovaries (if present), vagina, and surrounding tissues (Section 46-4). Primary exenteration may be considered for women with stage IVA cancer, that is, with direct tumor invasion into bladder and/or bowel but without distant spread. Exenteration is rarely performed for this indication, yet for suitable candidates, the survival rate can reach 30 percent (Million, 1972; Upadhyay, 1988).

Stage IVB

Patients with stage IVB disease have a poor prognosis and are treated with a goal of palliation. Pelvic radiation is administered to control vaginal bleeding and pain. Systemic chemotherapy is offered to palliate symptoms and prolong overall survival. The chemotherapy regimens used in this group of women are similar to those used in the setting of recurrent cancer.

Following Radiotherapy

Women who receive radiotherapy are closely monitored to assess their response. Tumors may be expected to regress for up to 3 months after therapy. Pelvic examination and/or radiologic scanning should document progressive shrinkage of the cervical mass. The rectovaginal examination is used to detect nodularity in the ligaments and parametria. If disease progresses locally after this interval, prognosis is poor. Pelvic exenteration may sometimes be indicated for this clinical setting.

In general, patients are seen at 3-month intervals for 2 years, then every 6 months until 5 years have passed from treatment, and then annually. At each visit, in addition to pelvic examination, a thorough manual nodal survey includes neck, supraclavicular, axillary, and inguinal lymph nodes. A cervical or vaginal cuff Pap test is also collected annually for 20 years after treatment completion. Findings of high-grade squamous intraepithelial lesions with screening should prompt colposcopic evaluation and biopsy of suspicious lesions. If recurrent cancer is diagnosed from these biopsies, then CT imaging is performed.

Once radiotherapy is completed, patients are encouraged to use a vaginal dilator or have vaginal intercourse three times per week. This helps keep the vagina patent, aids pelvic examination and Pap testing in the future, and ensures that the patient can remain sexually active if desired. Otherwise, radiation may result in vaginal fibrosis, leading to a shortened, nonfunctional vagina. The use of a water-based lubricant is also recommended.

Following Surgery

After a radical hysterectomy, 80 percent of recurrences are detected within the subsequent 2 years. During patient surveillance, an abnormal pelvic mass or abnormal pelvic examination finding typically prompts CT scanning of the abdomen and pelvis. Findings include cervical or vaginal lesion, rectovaginal nodularity, pain radiating down the posterior thigh, or new-onset lower extremity edema. Pelvic recurrences after radical hysterectomy, if diagnosed early, can be treated with radiation therapy. The same schedule of visits and Pap testing as just outlined for surveillance following radiotherapy is then recommended.

Hormone Therapy

Cervical cancer is not estrogen-dependent and thus hormone therapy is not contraindicated to treat menopausal symptoms, taking into account the risks and benefits discussed in Chapter 22. Moreover, hormone therapy is strongly considered for any premenopausal patient undergoing radiation treatment for cervical cancer until the average age of menopause. This is because radiation doses given for cervical cancer lead to menopause. Ovaries previously transposed out of the pelvic radiation field may be exceptions. Either systemic or vaginal forms are suitable. Estrogen alone is used if the uterus has been surgically removed, whereas combination hormonal therapy is given if the uterus remains.

This is defined as either persistent or recurrent cancer. Persistent disease is cervical cancer that has not completely regressed within 3 months of finishing radiotherapy. Recurrent disease is defined as a new lesion after primary therapy completion and initial regression.

Treatment of persistent or recurrent disease depends on its location and extent. The intent in these cases is usually palliative. However, in certain instances, a woman may qualify for pelvic radiation if she previously had not received this treatment. Alternatively, a woman may be a candidate for curative-intent surgery. Metastatic cervical cancer is not curable. In this setting, the goal of chemotherapy is to maximize existing patient quality of life and prolong survival.

Pelvic Exenteration for Secondary Disease

When curative-intent surgery is contemplated, local disease should be biopsy proven. Clinically, a patient may be considered for pelvic exenteration if the triad of lower extremity edema, back pain, and hydronephrosis are absent. If present, these suggest disease extension to the pelvic sidewalls, which would contraindicate surgery. In addition, regional and distant metastasis should be excluded by both physical examination and radiologic imaging, which typically is a PET/CT scan.

Pelvic exenteration begins with exploratory laparotomy, biopsies of suspicious lesions, and paraaortic lymph node evaluation. Exenteration is completed only if no disease is found in frozen section specimens sampled at the beginning of the surgery. A complete surgical description of this procedure is found in Section 46-4.

Alternatively, in highly selected patients, radical hysterectomy may be considered as an alternative to pelvic exenteration (Coleman, 1994). In these circumstances, women should have small cervical recurrences measuring less than 2 cm and have disease-free pelvic lymph nodes both prior to and during surgery.

With either operation, intraoperative and postoperative complications can be significant. Reported 5-year survival rates approximate 50 percent. Most recurrences occur in the first 2 years postoperatively (Berek, 2005; Goldberg, 2006).

Radiotherapy or Chemotherapy for Secondary Disease

Patients with central or limited peripheral recurrences who are radiotherapy naïve are candidates for curative-intent chemoradiation treatment. In these groups, survival rates of 30 to 70 percent have been reported (Ijaz, 1998; Ito, 1997; Lanciano, 1996; Potter, 1990).

Antineoplastic drugs are used to palliate both disease and symptoms of advanced, persistent, or recurrent cervical cancer (Table 30-9). Cisplatin is considered the single most active cytotoxic agent in this setting (Thigpen, 1995). Overall, response duration to cisplatin is 4 to 6 months, and survival in such women only approximates 7 months (Vermorken, 1993). A four-arm prospective randomized study demonstrated that the combinations of cisplatin with topotecan, vinorelbine, or gemcitabine are not superior to the combination of cisplatin and paclitaxel (Monk, 2009). Most recently, a randomized study evaluated adding bevacizumab to combination chemotherapy. Bevacizumab (Avastin) is a monoclonal antibody targeting vascular endothelial growth factor (VEGF). This addition increased the median overall survival length by 3.7 months (see Table 30-9) (Tewari, 2014).

Palliative chemotherapy is administered only if this treatment does not significantly lower patient quality of life and is balanced against the benefits of supportive care. Women with persistent nausea and vomiting from tumor-associated ileus may benefit from a gastrostomy tube. Bowel obstruction can be managed surgically, provided a patient is an appropriate surgical candidate. Percutaneous nephrostomy tubes may be placed for urinary fistulas or urinary tract obstruction.

Pain management forms the basis of palliation, and an extensive list of pain medications is found in Table 42-2. Cervical cancer patients can experience significant pain, and this is assessed at each visit. Many will require narcotics. If a patient has been using opioids and is hospitalized for inadequate pain control, then patient-controlled analgesia is considered. The total dose that controls the pain in a 24-hour period is determined. This dose can then be converted an equivalent dose of long-acting opioids. To allow for incomplete cross-tolerance between narcotics, the dose should be decreased by 25 to 50 percent. A supplemental short-acting opioid can be available for breakthrough pain and is typically prescribed at a dose that is 10 to 20 percent of the long-acting total daily dose and given at appropriate intervals. Narcotics can constipate, and patients using these are given a bowel regimen. This can be individualized, and suitable agents are listed in Table 25-6. In particular, a combination of stool softeners (docusate sodium) plus laxative (senna) plus polyethylene glycol is often effective.

We recommend discussion of medical directives if a patient has adequate mental capability. Often, such discussion is conducted over time, giving a woman an opportunity to understand the severity and progression of her disease. Home hospice is an invaluable part of terminal care for most of these women, who require intensive pain management and considerable assistance with daily living activities.

Survival rates between pregnant and nonpregnant women with cervical cancer do not differ when matched by age, stage, and year of diagnosis. Overall survival rates are slightly better for cervical cancer in pregnancy, because an increased proportion of patients have stage I disease.

Diagnosis

A Pap test is recommended for all pregnant patients older than 21 at the initial prenatal visit. Additionally, clinically suspicious lesions are directly biopsied. If Pap test results reveal HSIL, adenocarcinoma in situ (AIS), or suspected malignancy, then colposcopy is performed and biopsies are obtained. However, endocervical curettage is excluded to prevent amnionic sac rupture.

If Pap testing indicates malignant cells and colposcopy-directed biopsy fails to confirm malignancy, then diagnostic conization may be necessary. Many experts recommend delaying conization until the second trimester due to concern about pregnancy loss, however, median blood loss during excisional procedures increases with gestational age, especially in the third trimester. In pregnant patients, a loop electrosurgical excision procedure (LEEP) does not appear to offer an advantage compared with cold-knife conization. Moreover, one study found a surgical complication rate of 25 percent with LEEP in pregnancy, and 47 percent of the women had persistent or recurrent disease (Robinson, 1997).

Stages I and II Cancer in Pregnancy

Women with microinvasive squamous cell cervical carcinoma found during conization that measures ≤3 mm and contains no LVSI (stage IA1) may deliver vaginally and be reevaluated 6 weeks postpartum. Moreover, for those with stage IA or IB disease, studies find no increased maternal risk if treatment is intentionally delayed to optimize fetal maturity regardless of the trimester in which the cancer was diagnosed. Given the outcomes, a planned treatment delay is generally acceptable for women who are 20 or more weeks’ gestational age at diagnosis with stage I disease and who desire to continue their pregnancy. However, a patient may be able to delay from earlier gestational ages if she wishes. For women who wish to continue pregnancy, pelvic lymph node dissection can be performed in pregnancy in the first and second trimester via MIS (Vercellino, 2014). Women with positive nodes may elect to be treated with definitive treatment, rather than delay treatment, or may opt for neoadjuvant chemotherapy during pregnancy or for early delivery. For women who have a previable gestation and who desire definitive treatment of early-stage disease, a radical hysterectomy with the fetus in situ and lymphadenectomy can be performed. For patients with stage IA2 through IIA1, a cesarean section via a classical uterine incision may be performed at term, followed immediately by radical hysterectomy and lymph node dissection. Notably, a classical cesarean incision minimizes the risk of cutting through tumor in the lower uterine segment, which can cause serious blood loss and result in tumor spread.

Advanced Cervical Cancer in Pregnancy

Women with advanced cervical cancer diagnosed prior to fetal viability are offered primary chemoradiation. Spontaneous abortion of the fetus tends to follow whole-pelvis radiation therapy. For women who decline pregnancy termination, systemic chemotherapy can be administered. Cisplatin with vincristine or with paclitaxel can be administered in pregnancy. Congenital anomalies, growth restriction, and preterm delivery do not appear to be increased in fetuses of women who receive chemotherapy after the first trimester (Cardonick, 2010). If cancer is diagnosed after fetal viability is reached and a delay until fetal pulmonary maturity is elected, then a classical cesarean delivery is performed. Chemoradiation is administered after uterine involution. For patients with advanced disease and treatment delay, pregnancy may impair prognosis. A woman who elects to delay treatment, to provide quantifiable benefit to her fetus, will have to accept an undefined risk of disease progression.