Practice Guidelines: Cervical Cancer

Invasive cancer of the uterine cervix is either the leading or second leading cause of death from cancer among women worldwide and is the leading cause of death from cancer among women in developing countries. In some developing countries in the age group of 35 to 45, it is the leading cause of death from any cause. There are approximately 450,000 new cases of cervical cancer annually worldwide, compared to 13,500 new cases in the United States and 4,500 annual deaths.

It is widely recognized that cervical intraepithelial neoplasia (CIN) is a precursor to invasive squamous cell cancers and that adenocarcinoma in-situ is a precursor to invasive endocervical adenocarcinomas, but there are few reliable data on the number of new cases of these precursor lesions that are detected in most populations.

Screening

In the US, screening is performed almost exclusively using exfoliative cytology. In general, it is recommended that screening be initiated with onset of sexual activity and that, after three consecutive annual negative Papanicolaou tests, the subsequent screening interval be chosen by the patient and physician. For all practical purposes, this has translated into an annual recommendation.

The Papanicolaou smear should always include both an endocervical sample taken using a specialized brush device and a scraping sample from the immature transformation zone, including the present squamocolumnar junctions. Recent economic pressures have led to combining these samples on a single slide. There is no apparent loss in the detection efficiency when a one-slide technique is used. The specimen must be fixed immediately in order to avoid air-drying artifact.

TABLE 1

Bethesda Classification of Cervical Cytology

The terminology for reporting the results of the Papanicolaou (Pap) smear interpretation generally follows the Bethesda System (TBS) nomenclature (Table 1). A statement of adequacy is important, and smears are classified as within normal limits if no abnormalities are detected. If abnormal cells are seen and they are characteristic of an human papillomavirus (HPV)-related lesion of the squamous epithelium, they are generally referred to as being consistent with the presence of a squamous intraepithelial lesion (SIL) and further stratified as being low-grade or high-grade. This is in keeping with the molecular evidence that squamous cell cancer precursors are best classified in a two-tiered system. Smears that contain cells that are abnormal but not readily classified are generally diagnosed as having atypical squamous cells of undetermined significance (ASCUS) and further subclassified as to the most likely lesion to be present. Invasive cancers, metastatic disease, and other neoplasms are classified appropriately.

Although there have been problems relating to the introduction of TBS in the United States, it is the dominant system of the nomenclature and brings a uniform language to a previously fragmented nomenclature, although variations are becoming apparent. False-negative Pap smears are inevitable and occur even in the most highly capable laboratories with extensive quality-control systems in place. A negative smear never rules out the presence of neoplasia, and patients with signs or symptoms that could be attributable to cervical neoplasia should be examined and neoplasia excluded.

Screening for cervical neoplasia with the Pap smear is the most cost-effective cancer reduction program yet devised. In all populations in which it has been studied adequately, it has been reported that there is a direct relationship between the proportion of the population screened and a decline in the incidence of cervical cancer and deaths from cervical cancer. It is clear that screening should be used in all at-risk populations if the resources are available to undertake such a program.

A number of potentially useful alternate or complementary screening techniques are being developed, including computer-analyzed Pap smears and HPV DNA typing. These new techniques have the potential to lower the false-negative rate, lower the cost of screening, or increase the screening interval. More data are needed before these adjuncts gain wider acceptance.

Diagnosis

The diagnosis of invasive cervical carcinoma requires microscopic examination of tissue. Cytology and colposcopy are not diagnostic, although each is extremely important in the diagnostic process. Once cytology has indicated the presence of cells suspicious for invasive disease and no lesion is visible, colposcopy is invaluable in finding the area to be biopsied for proof of invasion. Endocervical and endometrial sampling are important in the absence of colposcopic lesions.

The initial work-up of invasive cervical cancer patients includes a history and physical examination, chest radiography, intravenous pyelogram (IVP) or computed tomography (CT) scan, cystoscopy/proctosigmoidoscopy, and HIV testing (especially for the younger, at-risk patient). In microinvasive disease, IVP or CT scan is done selectively, and some centers do not routinely employ IVP/CT or cystoscopy/proctosigmoidoscopy for early stage I disease (less than 2 to 3 cm) because of relatively low yield. During the last decade, however, CT scan has become increasingly popular in the work-up and management of cervical cancer. Several authors have suggested its use in clinical staging; however, there are important limitations with its use. It has been found that CT scans are unreliable in detecting subclinical parametrial disease and nodal metastasis less than 2 cm in diameter. In patients who are not candidates for surgical staging, CT scan can be helpful in assessing nodal disease. All enlarged lymph nodes should be studied histologically/cytologically by either surgical excision or fine-needle aspiration because of the 5% to 10% false-positive rate of CT.

Recently, magnetic resonance imaging has emerged as a radiologic modality capable of detecting early parametrial and nodal disease. Further experience is needed to establish its role and cost-effectiveness in the work-up and management of cervical carcinoma. Because of infrequent colon involvement, the use of barium enema should be restricted to symptomatic patients. Routine bone scan is unproductive unless the patient complains of bone pain.

Staging

Clinical

TABLE 2

1995 FIGO Staging (Montreal) for Carcinoma of the Cervix Uteri

The official International Federation of Gynecology and Obstetrics (FIGO) staging system for cervical carcinoma (Table 2) is based on clinical assessment (staging table from SGO Staging Handbook). To stage cervical cancer adequately, the following studies may be required: cervical biopsy, cystoscopy, proctosigmoidoscopy, chest radiography, and IVP or CT scan to rule out hydronephrosis. Optimal staging may be performed more easily under general anesthesia if cystoscopy and proctosigmoidoscopy are carried out. Because both the surgeon and radiation oncologist often are closely involved in the management of cervical cancer, a joint effort in staging should be made by both specialists.

Surgical

Accumulating data continue to show the deficiency of clinical staging. The FIGO staging system is based on the belief that cervical cancer is primarily a local disease in the pelvis, and that a surgical staging system cannot be widely employed worldwide, especially in the developing countries.

Surgical staging of cervical cancer, which includes lymphadenectomy, results in important intraoperative information for the gynecologic oncologist planning radical hysterectomy, as well as the radiation oncologist who may be consulted postoperatively for administration of radiotherapy. Furthermore, consistent surgical and pathologic data from staging laparotomies are important for analysis of survival and prognostic risk factors. Future staging systems may incorporate surgical and histopathologic variables to further refine the utility of staging in treatment decision making.

Treatment

Stage IA

The rationale for dividing cervical cancers into stage IA and stage IB is to identify a group of early invasive lesions that have low risk of extracervical spread. These lesions can then be treated by conservative surgery, thereby avoiding the complications associated with radical surgery or radiation therapy. In this regard, the definition of microinvasive (stage IA) cervical cancer proposed in 1973 by the Society of Gynecologic Oncologists (SGO) is optimal. According to the SGO, microinvasion is defined as squamous cell carcinoma invading the cervical stoma to a depth of 3 mm or less and in which there is no evidence of lymphatic or vascular space invasion and no confluence or invasive “tongues.” When the SGO definition is used, patients with microinvasive cervical cancer may be treated by extrafascial hysterectomy. Patients desiring to preserve fertility may be treated by cervical conization alone, provided that all conization margins are free of disease.

In the new FIGO staging classification for cervical cancer originating from the 1994 meeting in Montreal (Table 2), stage IA is redefined to include the SGO criteria of 3 mm for defining microinvasion. Stage IA1 now represents stromal invasion £ 3 mm. Stage IA2 is still defined by an upper limit of stromal invasion of 5.0 mm or less and horizontal spread of 7.0 mm or less. Lymph-vascular space invasion is not taken into consideration in this definition. About 7% of all patients with cervical cancer invading the stroma to a depth of 3 to 5 mm have lymph node metastasis. Patients whose tumors invade the stroma more than 3 mm or have any lymph-vascular space involvement (LVSI) should be treated with radical hysterectomy and lymphadenec- tomy or with radiation therapy.

Stage IB

The new FIGO staging system also subdivides stage IB into IB1 (£ 4 cm) and IB2 (more than 4 cm). Patients with stage IB1 cervical cancer can be treated effectively by either radical hysterectomy with pelvic and para-aortic lymphadenectomy or radiation therapy. Although both surgery and radiation therapy produce similar survival rates, in early-stage cervical cancer radical hysterectomy is considered by many to be the treatment of choice, especially in young, healthy patients with stage IB1 cervical cancers. Excellent survival rates can be achieved in these patients, along with preservation of ovarian function. Radiation therapy consisting of a combination of external and intracavitary radiation is indicated in those patients who have medical contraindications to radical surgery.

Stage IB2 or barrel-shaped cervical cancers can be treated by radiation therapy alone or by radical hysterectomy and pelvic and para-aortic node dissection. However, many of these tumors extend anatomically beyond the curative isodose curve of radiation, and contain central hypoxic areas that are resistant to ionizing radiation. As a result, central recurrence rates for patients with stage IB2 tumors treated with radiation therapy alone have been significant. Many centers now prefer to treat these patients with preoperative radiation therapy followed by extrafascial hysterectomy. Since pelvic lymph node metastases are present in approximately 25% of patients with stage IB2 cervical cancers, para-aortic lymph node dissection should also be performed at the time of extrafascial hysterectomy if the para-aortic nodes were not included in the preoperative radiation fields.

Stage IIA

The new FIGO staging system for cervical cancer remains unchanged for stages above IB2. The optimal treatment for most patients with stage IIA cervical cancer is radiation therapy consisting of a combination of external and intracavitary therapy. However, selected patients can be treated effectively by radical hysterectomy with pelvic and para-aortic lymphadenectomy and upper vaginectomy.

Stages IIB, IIIA, and IIIB

The treatment of choice for these stages in the United States is radiation therapy, including brachytherapy and external radiation. In stage IIIB, an extension of the external treatment field may be used to encompass the para-aortic field unless a surgical staging laparotomy with para-aortic lymphadenectomy ruled out para-aortic involvement.

Stage IVA disease is extremely rare, particularly with extension to the rectal mucosa. Treatment aimed at pelvic control usually consists of radical irradiation. In the rare patient with central pelvic disease only extending into the bladder mucosa, exenterative surgery may be chosen instead of primary irradiation.

Special Clinical Problems

Cervical Stump Carcinoma

Carcinoma arising in a cervical stump has decreased in incidence over the previous 20 years as the incidence and indications for subtotal hysterectomy have decreased. Staging of these lesions is identical to patients with an intact uterus.

Patients with stage IB or early stage IIA lesions who are candidates for radical surgery can be treated with a radical trachelectomy and bilateral pelvic lymph node dissection, or with radiotherapy. Patients with more advanced disease or those who are not surgical candidates are treated with primary radiation therapy. This usually involves standard external radiation therapy ports. The brachytherapy requires more individualization based on the length of the cervical canal and extent of parametrial disease compared to patients with an intact uterus.

Incidental Invasive Cervical Carcinoma Diagnosed in the Hysterectomy Specimen

This group of patients includes those who underwent a total hysterectomy with the incidental finding of a frankly invasive carcinoma in the surgical specimen. Fortunately, many of these patients have early disease and can have an excellent prognosis with appropriate treatment.

Patients with microinvasive disease may not require additional therapy. Patients who are candidates for radical surgery and who have no gross residual disease can be treated with radical parametrectomy, upper vaginectomy, and bilateral pelvic and para-aortic lymphadenectomy. This procedure is especially valuable in young patients in whom preservation of normal ovarian and vaginal function is most impor-tant. Radiation therapy is otherwise suitable alternative to radical parametrectomy.

In women whose hysterectomy specimen reveals that cancer has been “cut-through,” the prognosis is poor and radical parametrectomy, upper vaginectomy, and node dissection are inadvisable. These and other patients who are not candidates for radical pelvic surgery or who have more extensive disease can be treated with radiation therapy.

Invasive Cervical Carcinoma During Pregnancy

Diagnosis of invasive cervical carcinoma with a coexisting pregnancy occurs in about 3% of cases. The treatment and timing of treatment are dependent on stage of the disease, duration of the pregnancy, and the patient’s wishes.

Patients with carcinoma in situ of the cervix diagnosed by cytology and colposcopic-directed biopsies can be followed throughout the pregnancy and definitive treatment can be delayed until after reevaluation of the cervix 6 weeks postpartum. When there is suspicion of microinvasive or invasive carcinoma, a cone biopsy should be performed for diagnosis. Cone biopsy in pregnancy is a formidable procedure, especially for the inexperienced operator. Blood loss may be excessive and pregnancy loss may be increased. Microinvasive carcinomas (diagnosed by cone biopsy with free margins) can be followed throughout the pregnancy. All treatment delays are associated with some increased risk to the mother.

The treatment for stage IB or stage IIA cervical carcinoma in pregnancy is usually radical hysterectomy and bilateral pelvic and para-aortic lymphadenectomy. This can be performed in any trimester of pregnancy. If the fetus is viable, a classical cesarean section is carried out, followed by radical hysterectomy and bilateral pelvic and para-aortic node dissection. With nonviable pregnancies, a hysterotomy can be performed if necessary to improve surgical exposure.

Patients with more advanced disease are treated with radiation therapy. Prior to fetal viability, whole-pelvic radiation is initiated and the patient usually undergoes a spontaneous abortion 2 to 3 weeks after initiation of therapy. If spontaneous abortion has not occurred by the time the external radiation is completed, a dilatation and curettage (or rarely a hysterotomy) is required to empty the uterus. A delay of several weeks is required for involution of the uterus to occur prior to brachytherapy.

Patients with advanced disease (stage IIB or greater) with viable fetuses are delivered by classical cesarean section and external radiation therapy is initiated. Standard brachytherapy is added at the completion of external radiation.

Recurrent Cervical Carcinoma

Less than 5% of patients who develop recurrent carcinoma of the cervix are alive 5 years later. Patients who are candidates for curative treatment would include those with an isolated late pulmonary metastasis, central pelvic recurrences after primary treatment with radiation therapy, or those who develop central pelvic recurrences after radical hysterectomy.

Patients with isolated late lung metastasis from cervical carcinoma have a 25% 5-year survival with surgical resection. Likewise, patients who develop a pelvic recurrence after a radical hysterectomy have up to a 40% 5-year survival when treated with radiation therapy.

Patients with central pelvic recurrences after primary treatment with radiation therapy may be salvaged with surgery. Patients with small recurrences limited to the cervix or upper vagina can occasionally be treated with radical hysterectomy and partial vaginectomy with excellent results. Patients with larger central recurrences or who received previous very high-dose radiation therapy frequently have marked fibrosis between the bladder and/or rectum and the uterus, which requires a pelvic exenteration for surgical salvage.

One-third of patients with negative pelvic lymph nodes and free surgical margins survive 5 years after pelvic exenterations.

Recently, more attention has been directed to reconstructive procedures performed at the time of pelvic exenteration to improve quality of life. These include performing continent urinary conduits, primary colon reanastomosis, and vaginal reconstruction.

Follow-up After Primary Therapy

As the majority of treated patients who develop recurrences do so in the first 2 years following their therapy, physical examination, including nodal assessment (especially supraclavicular), rectovaginal examination, and Papanicolaou smears, should be performed at 3- to 4-month intervals during this time. Thereafter, semiannual examinations are appropriate and, beyond 5 years, annual examinations. Symptoms of pain, vaginal bleeding, and gastrointestinal or genitourinary dysfunction must be promptly investigated.

Interval chest films and abdominal-pelvic CT scans should be considered in those patients with high risk of recurrence, especially in the first 2 years. CT scan or IVPs post-treatment may also diagnose ureteral obstruction (pathologic or treatment-related) at potentially early stages.

The information in the Society of Gynecologic Oncologists clinical practice guidelines should not be viewed as a body of rigid rules. The guidelines are general and are intended to be adapted to many different situations, taking into account the needs and resources particular to the locality, the institution, or the type of practice. Variations and innovations that improve the quality of patient care are to be encouraged rather than restricted. The purpose of these guidelines will be well served if they provide a firm basis on which local norms may be built.

These guidelines are copyrighted by the Society of Gynecologic Oncologists (SGO). All rights reserved. These guidelines may not be reproduced in any form without the express written permission of the SGO. Requests for reprints should be sent to: Ms. Karen Carlson, SGO Publications, Society of Gynecologic Oncologists, 401 North Michigan Avenue, Chicago, IL 60611.

References:

Averette HE, Donato DM, Lovecchio JL, et al: Surgical staging of gynecologic malignancies. Cancer 60:2010-2021, 1987.

Averette HE, Nasser N, Yankow SL, et al: Cervical conization in pregnancy: Analysis of 180 operations. Am J Obstet Gynecol 106:543-549, 1970.

Brenner DE, Whitley NO: Computed tomography in invasive carcinoma of the cevix: An appraisal. Obstet Gynecol 62: 218-224, 1983.

Boronow RC: Should whole pelvic radiation therapy become past history? A case for the routine use of extended field therapy and multimodality therapy. Gynecol Oncol 43:71-76, 1991

National Cancer Institute Workshop: The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. JAMA 262(7):931-934, 1989.

Nelson JH Jr, Macaset MA, Lu T, et al: The incidence and significance of para-aortic Iymph node matastases in late invasive carcinoma of the cervix. Am J Obstet Gynecol 118:749, 1974.

Rotman M, Pajak TF, Choi K, et al: Prophylactic extended-field irradiation of para-aortic lymph nodes in stages IIB and bulky Ib and IIA cervical carcinomas. JAMA 274(5): 387-394, 1995

Stehman F, Bundy BN: Carcinoma of the cervix treated with chemotherapy and radiation therapy: Cooperative studies of the Gynecologic Oncology Group. Cancer 71(4; suppl):1697-1701, 1993

Van Nagell JR Jr, Donaldson ES, Wood EG, et al: The significance of vascular invasion and lymphatic infiltration in invasive cervical cancer. Cancer 41:228,1978b.

Nevin J, Soeters R, Dehaeck K, et al: Cervical carcinoma associated with pregnancy. Obstet Gynecol Surv 50(3):228-39, 1995.