Molecular epidemiological studies have demonstrated that correlation of sexual activity with cervical carcinoma is related to transmission of the epithelial trophic and oncogenic HPV.4–4 Most HPV infections are transient, resulting in no changes or low-grade intraepithelial lesions (cervical intraepithelial neoplasia [CIN] 1) that will be spontaneously cleared in most young women.^5,6 Development of high-grade intraepithelial lesions will occur in a small minority of women, usually within 24 months.⁷ High-grade lesions (CIN 2/3) may progress to invasive cervical cancer if not treated.^8,9

Cases of CIN 2/3 that progress to invasive cancer will do so over a period of 8 to 12 years, which has been referred to as the detectable preclinical phase.10–13 Thus the opportunities for early detection and intervention are abundant. However, an estimated one half of the invasive cervical cancers diagnosed in the United States are found in women who have never been screened, and an additional 10% are diagnosed in women who have not been screened within the preceding 5 years.¹⁴

Based on a comprehensive review of available evidence, the American Cancer Society released new screening recommendations for the prevention and early detection of cervical cancer on March 12, 2012. Table 87-1 is a synopsis of their recommendations.¹⁵ The biggest change from the past is that Pap tests are no longer recommended every year and there are two types of test that are used for cervical cancer screening. The Pap test can find early cell changes and treat them before they become cancer. The HPV test finds certain infections that can lead to cell changes and cancer. The HPV test may be used along with a Pap test or to help doctors decide how to treat women who have an abnormal Pap test.

Table 87-1

American Cancer Society Guidelines for Screening by Cytology for the Early Detection of Cervical Neoplasia and Cancer

• Cervical cancer screening should begin approximately 3 years after the onset of vaginal intercourse and no later than age 21 years.*

• Women between the ages of 21 and 29 years should have a Pap test every 3 years. They should not be tested for HPV unless it is needed after an abnormal Pap test result.

• Women between the ages of 30 and 65 years should have both a Pap test and an HPV test every 5 years. This is the preferred approach, but it is also OK to have a Pap test alone every 3 years.

• Women older than age 65 years who have had regular screenings with normal results should not be screened for cervical cancer. Women who have been diagnosed with cervical precancer should continue to be screened.

• Women who have had their uterus and cervix removed in a hysterectomy and have no history of cervical cancer or precancer should not be screened.

• Women who have had the HPV vaccine should still follow the screening recommendations for their age group.

• Women who are at high risk for cervical cancer may need to be screened more often. Women who are at high risk might include those with human immunodeficiency virus (HIV) infection, organ transplantation, or exposure to the drug diethylstilbestrol (DES). They should talk with their doctor or nurse.

Human Papillomavirus Biology

HPV is a double-stranded DNA virus in the Papovaviridae family (Fig. 87-1). This virus, made up of approximately 8000 nucleotides, encodes seven early genes and two late genes, as well as having a small, untranslated region. Approximately 100 different serotypes with limited DNA homology have been identified. Although the identification of high-risk subtypes of HPV has been important in defining potential therapeutic targets for the prevention of cervical carcinoma, HPV infection has not adequately explained all of the biological features of preinvasive disease and progression to invasive disease. Studies in sexually active college women demonstrated that infection with HPV is extremely common, occurring in up to 50% of women who become sexually active between the ages of 16 and 21 years, with the first abnormal Pap smear appearing in only a subset of women 1 year after infection. These infections typically are associated with low-grade dysplastic lesions and usually are transient. Persistent infection with associated high-grade dysplastic lesions is seen in only a small proportion of infected women (perhaps 1% or 2%). Biological and/or immunologic cofactors that allow the persistence of HPV infection in a small subset of women remain unclear. Epidemiological studies suggest that coinfection with herpes simplex virus type II, long-term oral contraceptive use, cigarette smoking, and high parity may increase the risk of persistent infection, carcinoma in situ (CIS), and invasive disease.^16,17 Although long-term epidemiological studies undertaken since appreciation of the role of HPV remain incomplete, it is known that in the large majority of women, the period between HPV infection, dysplasia, and then invasive carcinoma typically is years to decades, offering the potential for screening and early intervention to change the natural history and morbidity associated with this disease.^18–22

Figure 87-1 Human papillomavirus (HPV) genome. The HPV virus is a double-stranded DNA virus in the Papovaviridae family. This virus, constituted of approximately 8000 nucleotides, encodes seven early genes and two late genes, and has a small, untranslated region. Approximately 100 different serotypes with limited DNA homology have been identified. (From Parker MF, Sausville EA, Birrer MJ. Basic biology and biochemistry of gynecologic cancer. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia: Lippincott-Raven; 1997. Fig. 3.6.)

Molecular epidemiological studies have divided HPV serotypes into high-, intermediate-, and low-risk subtypes for the development of cervical neoplasia.²³ Low-risk subtypes are associated with venereal warts (condylomata acuminata), whereas intermediate- and high-risk subtypes are associated with cervical dysplasia and invasive carcinoma. Recent worldwide review of HPV typing demonstrated that 87% of squamous cell carcinomas had an identifiable HPV genome associated with the tumor as compared with 76.4% of adenocarcinomas. HPV-16 was the predominant type, associated with between 46% and 63% of the squamous carcinomas, whereas HPV-18 was associated with 10% to 14% of squamous cell carcinomas. Sixteen other HPV types were associated with the remaining 25% of cases, including HPV-45, -31, and -33. Most epidemiological studies have demonstrated a higher incidence of HPV-18 (37% to 41%), followed by HPV-16 (26% to 36%) in women with adenocarcinoma of the cervix. Several authors have shown a positive association between the presence of certain HPV subtypes and prognosis. Barnes and colleagues²⁴ showed that, among invasive carcinomas, HPV-18 is associated with poorly differentiated histology and higher incidence of nodal metastases. Similarly, Walker²⁵ reported that HPV-18–associated cancers were more likely to recur than were HPV-16–associated cancers. By contrast, HPV-16 was associated with large cell-keratinizing tumors, and these tumors were less likely to recur.²⁶ Lombard²⁷ demonstrated HPV-18–associated tumors to have a relative risk of death 2.4 times greater than that observed for patients with HPV-16–associated tumors and 4.4 times greater than that for patients with a tumor associated with another HPV type.

The mechanism by which these epithelial trophic viruses cause neoplasia has been defined through an elaborate collection of experiments that have focused on the role of the E6 and E7 early genes. Virally encoded E6 binds to the p53 protein, leading to targeting of this cellular checkpoint protein for proteasomal destruction. The E6 protein product also enters the nucleus and interacts with the other important nuclear proteins. The second virally encoded protein, E7, sequesters the retinoblastoma gene product (Rb), releasing the Rb-associated E2F transcription factor, which results in the initiation of a number of nuclear events important in driving the cell through the cell cycle. Recent studies also have suggested that p21 is sequestered and inactivated by the E7 protein. Elimination of the p53 and Rb checkpoint results in dysregulated cell growth and dysplasia. In addition, the E7/E6 proteins both independently lead to mitotic instability, with abnormality in chromosomal segregation during mitosis and increased chromosome instability. This may explain, in part, the long latency period between HPV infection and carcinoma. Loss-of-heterozygosity studies and other molecular events have defined chromosomal changes and changes in other oncogenic proteins that are neither primary nor secondary effects of HPV infection, but are instead caused by coinfection or other non-HPV factors. Although data suggest that HPV infection may be critical in initiating the neoplastic transformation, it is not clear that therapy that targets the HPV genome or HPV-related protein products will be useful in reversing the fully transformed phenotype.

The observation that, in a large portion of HPV-infected women, invasive carcinoma never develops, suggests that the preinvasive neoplastic cells can be cleared by the immune system. Recent studies identified T cells that specifically identify peptide epitopes of the E6 and E7 proteins. Certain human leukocyte antigen DR2 and TAP genes have been associated with invasive cervical carcinoma. Understanding the biology of cervical cancer initiation and propagation requires an understanding of HPV biology and its interaction with its epithelial cell target, as well as an understanding of cervical immunity.

The HPV virus is epithelial trophic. Unlike other oncogenic viruses (such as the hepatitis viruses), systemic exposure to the virus is not required for infection of the target organ. Analysis has demonstrated that E6-specific T cells in the majority of healthy women are responsible (at least in part) for eradication of infection. Tetramer analysis in patients with cervical CIS or invasive cervical cancer has in general demonstrated few HPV-16 E7-reactive T cells, typically representing less than 0.1% of the CD8+ T cells in the systemic circulation. Recent attempts to generate productive and/or long-term HPV-16 infection have been accomplished by using an HPV-16 virus like particle vaccine.

In a recent randomized, placebo-controlled study, the delivery of three vaccinations at day 0, month 2, and month 6 reduced the risk of persistent new HPV-16 infection from 3.8 cases per 100 women-years at risk to 0 per 100 woman-years in the group that were vaccinated, for a 100% efficacy rate. In this study, involving 2392 young women between the ages of 16 and 23 years, nine cases of HPV-16–related CIN were noted, all in the placebo recipients.²⁸ The HPV vaccine with just HPV-16 and another double-variant vaccine are now commercially available and are recommended for both girls and boys from the ages of 9 through 31 years. The use of the vaccine worldwide should help eliminate cervical and head and neck cancers caused by HPV-16.

Maiman, Fruchter, and associates 29–32 investigated the disease characteristics, recurrence risks, and survival rates of human immunodeficiency virus (HIV)-seropositive patients with CIN and invasive cervical cancer. HIV-infected women had significantly higher rates of recurrence of CIN after standard therapies than did seronegative women. HIV-infected women with cervical cancer had significantly more advanced disease than did those who were not infected. Only 3 (19%) of 16 HIV-seropositive patients were first seen with an early-stage disease (defined as stage Ia or nonbulky Ib) compared with 35 (52%) of 68 in the HIV-seronegative group. When upstaged based on surgicopathological findings, only one (6%) HIV-infected patient had early-stage disease compared with 40% of uninfected patients. The response to therapy and prognosis was poorer among HIV-seropositive women, with higher recurrence and death rates. The majority of seropositive women had lymph node metastases and high-grade tumors. They generally were asymptomatic with respect to their HIV disease, but died of cervical cancer. The significant impact of immune status on disease progression was made evident by prolonged disease-free follow-up in seropositive patients with CD4 counts greater than 500/mm³, in contrast to those with CD4 counts less than 500/mm³. The observed marked increased risk of invasive cervical carcinoma in women infected with the HIV virus and the demonstration that highly effective antiretroviral therapy is capable of doubling the CIN regression rates in women infected with HIV, as compared with those women not receiving highly effective antiretroviral therapy, provides suggestive, indirect clinical evidence supporting the critical importance of the intact immune system in limiting the progression of HPV infection to invasive cancer in healthier populations.

Pathology

The epithelium of the cervix is composed of squamous epithelium that covers the exocervix and glands and columnar epithelial cells that line the endocervix. The border between the squamous and columnar epithelium is called the squamocolumnar junction, the site of ongoing squamous metaplasia believed to be most vulnerable to viral neoplastic transformation. With increasing age, the squamocolumnar junction migrates from the exocervix into the distal endocervical canal (Fig. 87-2), with the region between the original and subsequent locations termed the transformation zone. The transformation zone is the most common location for detection of early cervical cancers (Fig. 87-3). Tumors arising on the ectocervix typically are squamous cell carcinomas, whereas adenocarcinomas are more likely to have their epicenter in the endocervix.

Figure 87-2 The transformation zone and squamocolumnar junction. NSJC, New squamocolumnar junction; OSJC, original squamocolumnar junction; T-zone, region between original and new squamocolumnar junctions. (From Wright TC Jr, Richart RM. Pathogenesis and diagnosis of preinvasive lesions of the lower genital tract. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia: Lippincott; 1997. p. 678; and Berek JS, Hacker NF. Practical gynecologic oncology. 2nd ed. Baltimore: Williams & Wilkins; 1994.)

Figure 87-3 Invasive squamous cell carcinoma. A mass is present on the external os. (From Clement PB, Young RH. Atlas of gynecologic surgical pathology. Philadelphia: Saunders; 2000. p. 103.)

A continuum appears to exist from CIN to frankly invasive squamous cell carcinoma (Fig. 87-4). The mean age of women with CIN is 15.6 years younger than that of women with invasive cancer, suggesting slow progression of CIN to invasive carcinoma.¹⁸ The natural history of HPV infection and CIN in part reflects the host immune system response to the virus. Seventy-five percent of CIN 1 lesions will spontaneously regress or persist as CIN 1, without progression to invasive carcinoma.^18–21

Figure 87-4 A to F, Histologic portraits of normal stratified squamous epithelium of the cervix and cervical intraepithelial neoplasia. (From Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia: JB Lippincott; 1997. p. 678; and Crum CP. Papillomavirus-related changes and premalignant and malignant squamous lesions of the uterine cervix. In: Clement PB, Young RH, editors. Tumors and tumorlike lesions of the uterine corpus and cervix. Contemporary issues in surgical pathology. Vol 22. New York: Churchill Livingstone; 1993. p. 91.)

Miller ²² reported, in a 13-year observational study, that only 14% of CIN 3 lesions had progressed, whereas 61% persisted, and the remainder disappeared. Patients taking corticosteroids or other immunosuppressive drugs and patients with HIV infection are at higher risk of progressing to invasive cancer and may have a shorter transit time for this progression.

Neuroendocrine Tumors of the Cervix

Neuroendocrine tumors in the cervix include typical and atypical carcinoid tumors, and large cell, small cell neuroendocrine carcinomas, and undifferentiated small cell carcinoma. Both large cell and small cell neuroendocrine tumors resemble similar carcinomas arising in the lung and other aerodigestive sites.39,40 Undifferentiated small cell carcinoma is a very poorly differentiated carcinoma with neuroendocrine features, histologically similar to anaplastic small cell carcinoma of the lung.⁴¹ Neuroendocrine carcinomas tend to behave very aggressively, with frequent widespread metastasis to multiple sites, including bone, liver, and skin. Brain metastases may occur when disease is advanced but usually are preceded by lung metastases.⁴³ Efforts to treat these tumors by using approaches typically used for small cell carcinomas of the lung have had mixed results.⁴²

Clinical Presentation

Preinvasive neoplastic disorders of the cervix usually are asymptomatic, hence the need for regular screening by cytologic evaluation of cells collected from the exocervix and endocervix. Early invasive cancers also may be asymptomatic, although some women will notice postcoital, intermenstrual, or postmenopausal spotting. Other symptoms may include malodorous vaginal discharge, dyspareunia, or cramping pelvic pain from uterine contractions caused by the accumulation of blood and uterine deciduas in menstruating patients with occlusion of the endocervical canal. Chronic blood loss may result in symptomatic anemia in some patients. Major hemorrhage is unusual except in locally advanced disease. Pelvic pain, lower-extremity swelling (from occlusion of pelvic lymphatics or thrombosis of the external iliac vein), or problems with micturition or defecation indicate advanced regional disease and portend an ominous prognosis. Metastatic disease involving supraclavicular nodes, bones, or lungs can be the cause of presenting symptoms, but rarely in the absence of pelvic symptoms. Constitutional symptoms, including anorexia, dysgeusia, and weight loss, are seen most often in patients with very advanced disease.

Screening

Screening for cervical cancer and its precursors with the Pap smear and pelvic examination has resulted in dramatic reductions in cervical cancer mortality in every country where this has been widely used, and is arguably the most effective screening program in effect for any neoplastic disease, in either gender. Table 87-1 outlines the current screening guidelines of the American Cancer Society. The false-negative rate of the Pap smear is approximately 10% to 15% in women with invasive cancer, but the sensitivity, as defined by the detection of biopsy-proven CIN, is 51%.^45–45 The sensitivity of the test may be improved by ensuring adequate sampling of the squamocolumnar junction and the endocervical canal. Smears without endocervical or metaplastic cells may be inadequate and possibly should be repeated.⁴⁶

Factors contributing to abnormal cytologic smears include the presence of hemorrhage, necrosis, and intense inflammation. Thus gross symptomatic lesions should be sampled with biopsy rather than assessed with exfoliative cytology.

In the 1980s, cytology laboratories began reporting an increasing number of smears with changes of “squamous atypia” in response to concerns from clinicians over an unacceptably high false-negative cytology rate and increased recognition by cytopathologists of the cytologic changes associated with HPV infection. The use of multiple classification systems with inconsistently defined numeric grading conventions added further imprecision. In an attempt to eliminate confusion among clinicians and cytopathologists, a uniform system for reporting epithelial cell abnormalities was established in 1988 at a National Cancer Institute (NCI) workshop for reporting cervical and vaginal cytologic diagnoses.⁴⁷

The Bethesda System has since been revised, in 1991 and again in 2001 (Table 87-2),⁴⁸ to reflect laboratory and clinical experience gained since the original implementation, as well as the increased use of new technologies and results from interval research studies. An important contribution of the Bethesda System was the creation of a standardized format and nomenclature for cytology laboratory reports that includes both a descriptive diagnosis and an evaluation of specimen adequacy.

Table 87-2

2001 Bethesda System

SPECIMEN ADEQUACY

Satisfactory for evaluation (note presence of endocervical/transformation zone component)

Unsatisfactory for evaluation (specify reason)

Specimen rejected/not processed (specify reason)

Specimen processed and examined, but unsatisfactory for evaluation of epithelial abnormality because of (specify reason)

GENERAL CATEGORIZATION (OPTIONAL)

Negative for intraepithelial lesion or malignancy

Epithelial cell abnormality

Other

INTERPRETATION/RESULT

Negative for intraepithelial lesion or malignancy

Organisms

• Trichomonas vaginalis

• Fungal organisms consistent with Candida sp.

• Shift in flora suggestive of bacterial vaginosis

• Bacteria morphologically consistent with Actinomyces sp.

• Cellular changes consistent with herpes simplex virus

Other nonneoplastic findings (Optional. List not comprehensive) Reactive cellular changes associated with inflammation (includes typical repair)

• Radiation

• Intrauterine contraceptive device

Glandular cells after hysterectomy Atrophy Epithelial cell abnormalities Squamous cell

• Atypical squamous cells (ASC) of undetermined significance (ASC-US), cannot exclude high-grade squamous intraepithelial lesion (HSIL [ASC-H])

• Low-grade squamous intraepithelial lesion (LSIL) encompassing human papillomavirus, mild dysplasia, cervical intraepithelial neoplasia (CIN) 1

• HSIL encompassing moderate and severe dysplasia, carcinoma in situ; CIN 2 and CIN 3

• Squamous cell carcinoma

Glandular cell

• Atypical glandular cells (AGCs) (specify endocervical, endometrial, or not otherwise specified)

• Atypical glandular cells, favor neoplastic (specify endocervical or not otherwise specified)

• Endocervical adenocarcinoma in situ (AIS)

• Adenocarcinoma

Other (list not comprehensive)

• Endometrial cells in a woman 40 years or older

AUTOMATED REVIEW AND ANCILLARY TESTING (Include as appropriate) EDUCATIONAL NOTES AND SUGGESTIONS (Optional)

Adapted from Soloman D, Davey D, Kurman R, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA 2002;287:2114.

With the formulation of Bethesda 2001, a specimen is designated “satisfactory for evaluation” or “unsatisfactory for evaluation.” The category “satisfactory but limited by …” has been eliminated as confusing. Minimal cellularity requirements for a specimen to qualify as satisfactory depend on specimen type, with an estimated 8000 to 12,000 well-visualized squamous cells for conventional smears and 5000 squamous cells for liquid-based preparations. Comments on partially obscuring inflammation or blood may be added to the “satisfactory” designation, with a specimen considered “partially obscured” when 50% to 75% of the epithelial cells cannot be visualized. When more than 75% of epithelial cells are obscured, a specimen is designated “unsatisfactory.” A notation is made regarding the presence or absence of an endocervical/transformation zone component for specimens with adequate squamous cellularity, with the numeric criterion being at least 10 well-preserved endocervical or squamous metaplastic cells. Cell clusters are not required.

Squamous atypia is a mild cellular abnormality characterized by slight nuclear enlargement and minimal changes of the nuclear chromatin. Cellular changes that show morphologic features of papillomavirus infection or mild CIN (grade 1) are termed low-grade squamous intraepithelial lesions (LSILs). Cellular changes consistent with moderate (CIN 2) or severe (CIN 3) dysplasia are termed high-grade squamous intraepithelial lesions (HSILs). The division of squamous intraepithelial lesions into high- and low-grade strata reflects accumulating evidence that LSIL generally is indicative of transient infection with HPV, particularly in young women, and does not routinely require colposcopy or treatment, whereas HSIL cytology detects precursor lesions that require further evaluation and treatment. Important refinements in the 2001 Bethesda System relate to the reporting of equivocal results. Atypical squamous cells (ASCs) are now further qualified as “of undetermined significance” (ASC-US) or “cannot exclude HSIL” (ASC-H). The qualifier “undetermined significance” was retained, as some cases of ASC-US are associated with underlying CIN 2 or CIN 3. All ASCs are considered to be suggestive of squamous intraepithelial lesions, and thus the category “ASC-US favor reactive” has been eliminated, with a portion of cases previously so designated being downgraded to “negative for intraepithelial lesion or malignancy.” The new term ASC-H includes true HSIL and its mimics, and is thought to have a positive predictive value for CIN 2 or CIN 3 intermediate between ASC-US and HSIL. The term atypical glandular cells of undetermined significance (AGUS) has been eliminated to avoid confusion with ASC-US. Glandular cell abnormalities now are classified as atypical endocervical, endometrial, or glandular cells. The term atypical epithelial cells may be used when a squamous versus glandular origin cannot be ascertained.

Diagnosis

Management of the patient with an abnormal Pap smear with respect to further diagnostic assessments, therapeutic intervention, and subsequent surveillance follow-up is a complex arena in which guidelines continue to evolve. The American Society for Colposcopy and Cervical Pathology (ASCCP) has developed contemporary guidelines for management of cervical cytologic abnormalities and CIN based on current reporting terminology (Bethesda 2001).49,50 Techniques for further evaluation can include colposcopy, endocervical curettage or brushing, and cervical conization by cold knife or loop electrodiathermy excision procedure (LEEP).

Testing for high-risk, oncogenic HPV DNA 51–57 synchronously, or subsequent to cytologic screening, is proving to be a useful complementary tool in triage of patients with ASC-US smears, and in determining the need for colposcopy and the intervals for repeated screening. It may be helpful in the follow-up of younger patients with LSIL smears, and a useful tool in identifying older women who can safely be screened at 3-year intervals instead of annually.

HPV triage of patients with an ASC-US smear is at least as sensitive as immediate colposcopy for ultimately detecting CIN 3 and results in referral of about half as many women to colposcopy. A follow-up strategy that uses repeated cytology is sensitive at an ASC-US referral threshold, but requires two follow-up visits and, ultimately, more colposcopic examinations than does HPV triage.⁵⁸

Cytologic and HPV screening in women older than 30 years (after which most sexually active women have been exposed to HPV and have, or have not, spontaneously cleared the infection) may be a mechanism through which low-risk HPV-negative and cytologically negative patients can be identified for whom screening at 3-year intervals is prudent and appropriate.

Colposcopy

Patients with a gross lesion of the cervix should undergo cervical biopsy. For patients with an abnormal cytologic evaluation, without a gross lesion, a colposcopic examination with directed punch biopsies is required. Colposcopy allows the clinician to identify areas suggestive of dysplasia. A 3% acetic acid solution is applied to the cervix for 30 to 90 seconds, which causes a transient reaction with the envelope proteins of the papillomavirus, in addition to producing an osmotic dehydration of the dysplastic cells, thereby accentuating the optically dense chromatin to produce a whitish area. The skilled colposcopist can further distinguish between grades of dysplasia based on acetowhitening and types of vascular patterns (Fig. 87-7). An additional technique to help visualize abnormal areas is the application of quarter-strength Lugol iodine after the initial inspection for acetowhitening. High-grade lesions turn mustard yellow.

Endocervical Curettage or Endocervical Brush

Study of the endocervical canal is required when no abnormalities are found on colposcopic examination, when the entire squamocolumnar junction cannot be visualized, or when atypical endocervical cells are present on Pap smear. Some experts advocate the use of endocervical curettage (ECC) as part of every colposcopic examination to safeguard against missing occult cancer within the endocervical canal. Others reserve ECC for patients with recurrent cytologic atypia after therapy. A sleeved endocervical brush is an alternative to ECC that is less uncomfortable for many patients. A recent, prospective comparison of ECC specimens with sleeved endocervical brush specimens (both obtained from the same patient sequentially after randomization with respect to the order of obtaining specimens) before cervical conization or hysterectomy revealed a higher rate of inadequate specimens from ECC, comparability of the two techniques with respect to sensitivity and specificity in unmatched analysis, and superior sensitivity of the sleeved endocervical brush in matched analysis. As the sleeved endocervical brush is at least isoeffective and possibly better than ECC but more comfortable for the patient, many clinicians prefer this maneuver, which may serve to increase patient compliance with subsequent surveillance follow-up and repeated assessment.⁵⁹

Excisional Biopsy

Diagnostic cervical conization (cone biopsy) leads to an accurate diagnosis and decreases the incidence of inappropriate therapy in the following situations: (a) the squamocolumnar junction is not visualized on colposcopy; (b) dysplastic epithelium extends into the endocervical canal; (c) cytology is suggestive of high-grade dysplasia or worse; (d) microinvasive carcinoma is found on directed biopsy; (e) sampling of the endocervical canal shows high-grade intraepithelial neoplasia; and (f) cytology is suggestive of adenocarcinoma in situ.

Loop Electrodiathermy Excision Procedure

LEEP uses wire-loop electrodes in conjunction with a radiofrequency alternating current to excise the entire transformation zone and distal canal under local anesthesia (Fig. 87-8). Compared with ablative procedures, LEEP has the major advantage of obtaining tissue for histologic evaluation. At many centers, LEEP has become the preferred treatment for CIN that can be assessed adequately with colposcopy.^60–72

Figure 87-8 Equipment used for loop electrosurgical excision procedures (LEEP). A, LEEP system 6000 (Cooper Surgical). B, Large loop electrodes used for LEEP. (B from Baggish MS. Colposcopy of the cervix, vagina, and vulva: a comprehensive textbook. Philadelphia: Mosby; 2003. p. 96.)

Complications include bleeding, with a reported incidence of 1% to 8%,73,74 cervical stenosis (1%), and, rarely, pelvic cellulitis or adnexal abscess. In some cases, LEEP may not be an adequate alternative to formal excisional conization, such as in those patients in whom microinvasive or invasive cancer is suspected, or in those with adenocarcinoma in situ, because it may treat disease within the cervical canal inadequately and complicate pathological interpretation of the specimen. Appropriate application of this technique will yield tissue suitable for pathological study and reliable diagnosis. Excess heat may result in thermal artifact that can compromise interpretation of margins and the therapeutic adequacy of the LEEP procedure.

Diagnostic or Therapeutic Excisional Conization (Cone Biopsy)

Diagnostic or therapeutic excisional conization (cone biopsy) must be performed under general or regional anesthesia. Complications, which include hemorrhage, sepsis, infertility, stenosis, and cervical incompetence, occur in 2% to 12% of patients, depending on depth and geometry of excision.75–75 Width and depth of the cone should be tailored to the topography of the lesion, to produce the least amount of injury while providing clear surgical margins. Conization may be performed with a cold knife or with the carbon dioxide laser.

Patient Evaluation in Patients with Invasive Disease

A thorough history should be obtained before diagnostic assessments are initiated, because skilled questioning will elicit information that will serve to guide subsequent evaluation. A sexual history should be obtained from all patients, which may alert the physician to screen for HIV and other sexually transmitted diseases. This also will serve to establish a baseline for sexual function and expectations that may be helpful in the posttreatment counseling and sexual rehabilitation of patients after completion of therapy. A complete history and systems review also should assess comorbid conditions (e.g., neurologic, cardiovascular, pulmonary, gastrointestinal, endocrine) that may affect the selection and implementation of treatment strategy.

A social assessment should be routinely obtained and will assist in determining supportive interventions that may be necessary to facilitate patient compliance with complex diagnostic programs and multimodality treatment programs. In the United States, patients with advanced cervical cancer may be uninsured or underinsured, often will have cultural or language barriers to obtaining and complying with care, may have limited educational attainment and medical sophistication, and will have variable degrees of familial or other social support.

A general physical examination should be carried out, with attention to accessible lymph node groups, including the supraclavicular nodes and the inguinal nodes. Physical examination also should assess the heart, lungs, and large vasculature (i.e., the carotid, dorsalis pedis, posterior tibial, and femoral arteries; the aorta; and the veins of the lower extremities). Pelvic examination must include meticulous inspection of all potentially visible sites and palpation of the perineum, the vulva, the full length of the vagina, and the anorectum. Some patients will manifest areas of preinvasive disease at multiple separate sites along the lower reproductive tract or synchronous invasive primary cancers of the lower reproductive tract and the anus, representing “field cancerization” consequent to exposure to a common etiologic agent. Typically, tumors arising from the ectocervix or transformation zone are easily visualized and sampled with biopsy. Occasionally tumors arising in the uterine corpus or lower uterine segment can invade the uterine cervix or protrude through the cervical os, causing confusion in the diagnosis. Often this can be sorted out by histopathology and by diagnostic imaging, with magnetic resonance imaging (MRI) being most useful in this context. Once the diagnosis of cervical carcinoma is established, staging includes careful clinical evaluation of the vagina, cervix, uterus, and parametria. The adnexal areas should be palpated as well, because some patients will harbor adnexal pathology (tuboovarian abscesses or, less commonly, adnexal metastases) that optimally are diagnosed and addressed before initiation of cancer treatment.

Conventionally, pelvic examinations in the setting of invasive cancer were performed under general or regional anesthesia to assist in the most accurate possible physical evaluation of disease extent. Relaxation of the muscles of the pelvic floor and the opportunity to conduct vigorous and prolonged examination without inflicting discomfort or pain may identify disease extension that may be missed in an outpatient clinic examination. For example, examination under anesthesia may be critical in very anxious, young, nulliparous women, as well as in elderly patients who may not be regularly sexually active and in whom the vagina may be stenotic. Examination under anesthesia also affords the opportunity to perform cystoscopy and proctosigmoidoscopy when indicated. However, in an era of cost containment and constrained health care resources, examination under anesthesia may be omitted when a cooperative patient is amenable to satisfactory examination in the clinic or, in rare circumstances, in patients with locally extensive disease requiring rapid initiation of treatment.

Staging

The revised 2009 staging criteria (Table 87-3)⁷⁶ of the International Federation of Obstetrics and Gynecology (FIGO) are a mixture of histopathological, clinical, and radiographic assessments that reflect the fact that invasive cervical cancer is most prevalent in less-developed portions of the globe where sophisticated and expensive imaging modalities may not be widely available. Cervical cancer is clinically staged, and staging is based primarily on inspection and palpation of the cervix, vagina, parametrium, and pelvic sidewalls. Only the subclassification of stage I (Ia1, Ia2) requires pathological assessment. The FIGO staging system permits assessment through biopsy, physical examination, cystoscopy, proctoscopy, excretory urography (intravenous pyelography), and plain film radiography of the chest and skeletal system. Results of lymphangiography, computed tomography (CT), MRI, and positron emission tomography (PET) may be of great value in planning treatment, but do not influence assignment of clinical stage in the FIGO formalism. When findings are equivocal, by convention, a patient is assigned to a lower stage. Once clinical stage has been assigned, it cannot be altered by subsequent events or findings. Findings from surgical evaluation (by laparoscopy or by surgical assessment of retroperitoneal lymph nodes via extraperitoneal or transperitoneal node dissection) will not alter assignment of clinical stage. However, these findings may profoundly influence subsequent treatment. Similarly, evidence of nodal or other spread discerned at the time of hysterectomy does not alter clinical stage.

Table 87-3

FIGO 2009 Staging for Cervical Cancer

FIGO STAGES
1	Cervical carcinoma confined to uterus (extension to corpus should be disregarded)
Ia	Invasive carcinoma diagnosed only by microscopy. All macroscopically visible lesions—even with superficial invasion—are T1b/Ib. Stromal invasion with a maximum depth of 5 mm measured from the base of the epithelium and a horizontal spread of 7 mm. Vascular space invasion, venous or lymphatic, does not affect classification
Ia1	Measured stromal invasion ≤3 mm in depth and ≤7 mm in horizontal spread
Ia2	Measured stromal invasion >3 mm and ≤5 mm with a horizontal spread ≤7 mm
Ib	Clinically visible lesion confined to the cervix or microscopic lesion greater than T1a2/Ia2
Ib1	Clinically visible lesion ≤4 cm in greatest dimension
Ib2	Clinically visible lesions >4 cm in greatest dimension
II	Cervical carcinoma invades beyond uterus but not to pelvic wall or to the lower third of vagina
IIa	Tumor without parametrial invasion
IIa1	Clinical visible lesion ≤4 cm in greatest dimension
IIb2	Clinical visible lesion >4 cm in greatest dimension
IIb	Tumor with parametrial invasion
III	Tumor extends to the pelvic wall, and/or involves the lower third of the vagina, and/or causes hydronephrosis or nonfunctioning kidney
IIIa	Tumor involves lower third of the vagina, no extension to pelvic wall
IIIb	Tumor extends to pelvic wall and/or causes hydronephrosis or nonfunctioning kidney
IVa	Tumor invades mucosa of the bladder or rectum, and/or extends beyond true pelvis (bullous edema is not sufficient to classify a tumor as T4)
IVb	Distant metastasis

A staging system based largely on findings from clinical pelvic examination is inherently imprecise and somewhat subjective, with overstaging and understaging of the parametria being the most problematic issue and the one most likely to alter management. However, FIGO stage does correlate with prognosis and does permit cautious interinstitutional and intrainstitutional comparisons of treatment outcomes.

Diagnostic Imaging Evaluation of Cervical Cancer

The goals of clinical staging in patients with invasive disease include determining the appropriateness and extent of initial surgical treatment, as opposed to initial treatment with combined radiation therapy and synchronous chemotherapy. Although not formally part of clinical staging, modern sophisticated diagnostic imaging can enhance clinical assessment of disease volume and extent in women with cervical carcinoma, thus refining both selection of treatment and technical implementation of treatment modalities.

The FIGO staging system is based predominantly on clinical examination under anesthesia, ultrasonography, intravenous urography, cystoscopy, proctoscopy, and chest radiography. Significant inaccuracies occur in this staging because of possible errors in gynecologic examination (24% to 39%).⁷⁷ The various imaging modalities have a complementary role in the accurate staging and complete evaluation of the cancer that have important therapeutic implications.

Today, clinicians most often obtain a contrast-enhanced CT scan of the abdomen and pelvis for patients with disease of stage Ib2 or greater (Fig. 87-9).⁷⁸ Lymph nodes larger than 1.0 to 1.5 cm in diameter are suspicious for tumor involvement and should be biopsied. Unfortunately, microscopic metastases are not readily detected with CT, and the inflammation commonly associated with advanced disease may cause enlargement of nodes that do not contain metastases. The sensitivity of MRI in the detection of regional metastases is similar to that of CT. However, MRI provides better anatomic delineation and accurate estimation of the tumor size, volume, and local extent within the pelvis, which can influence the choice of therapy (Fig. 87-10). Involvement of the vagina, parametrium, pelvic wall muscles, ureter, bladder, and rectum can be better assessed with MRI for accurate staging (Fig. 87-11).⁷⁷ MRI before and after vaginal opacification with contrast medium can be used if imaging evaluation of the vaginal wall or fornices is required.⁷⁹ T₂-weighted images obtained by using phased-array coil, fast spin-echo or conventional spin-echo techniques are accurate in local staging and lymph nodal assessment; the former technique is faster with increased resolution.⁸⁰ Dynamic contrast-enhanced T₁-weighted images are helpful in identifying smaller tumors, fistulous tracts, and invasion into bladder and rectum.⁷⁷ Dynamic contrast-enhanced imaging is useful in differentiating areas composed predominantly of tumor cells (well-enhanced areas) from those of fibrous tissue with scattered cancer cells (poorly enhanced areas). This information can be helpful, as radiation therapy is more effective in well-enhanced tumors.⁸¹

Figure 87-9 Pattern of nodal metastases (*arrows*) on computed tomography. Necrosis is a common feature of squamous cell carcinoma. A, Paraaortic node. B, Retroperitoneal node. C, External iliac node.

Figure 87-10 Cervical cancer on MRI. T2-weighted sagittal image shows a cervical mass (*arrow*) causing stenosis and fluid accumulation in uterus.

Figure 87-11 Cervical cancer on MRI. A, Sagittal T2-weighted image showing a heterogeneous isointense mass in the cervix extending into the posterior wall of the uterine body (*arrow*). B, T2-weighted image sliced coronally through the cervix showing a mass protruding into the uterine cavity (*arrow*). C, T2-weighted image sliced axially across the cervix showing a bulky, eccentric cervical mass on the right side with spread beyond the T2 hypointense ring of the cervical stroma into the parametrium (*arrow*). D, T1-weighted gadolinium-enhanced image revealing clear delineation of the mass relative to the adjacent uterus (*arrow*).

MRI is very useful in the evaluation of tumor volume and enlarged lymph nodes, inherently important prognostic factors as well as determinants of the design of radiation treatment ports and selection of radiation dose. However, MRI cannot identify micrometastases to lymph nodes and differentiate malignant from nonmalignant enlargement. MRI with newer contrast agents like ultrasmall superparamagnetic iron oxide has been useful in distinguishing benign from malignant nodes.⁸² The shape, volume, and direction of the growth of the tumor can be well assessed, and these are crucial for planning brachytherapy and external beam radiotherapy.⁷⁷ MRI also is useful in the follow-up evaluation of the tumor, its response to treatment, and identification of recurrences. However, benign conditions, like edema or inflammation, sometimes cannot be differentiated from tumors. Dynamic contrast-enhanced MRI is useful in differentiating malignant lesions that usually have shorter and stronger enhancement than benign conditions in patients who have abnormalities after treatment for cervical cancer.⁸³

Recent studies have suggested that ¹⁸F-fluorodeoxyglucose PET (FDG-PET) is more sensitive and specific than more standard radiographic studies in the detection of lymph node involvement by cervical cancer.^86–86 Grigsby and colleagues⁸⁷ have reported a strong correlation between abnormal posttherapy FDG uptake and tumor recurrence. PET scans are now Medicare approved and considered standard of care for staging purposes for all newly diagnosed cervical carcinomas.

Laboratory Evaluation

Routine laboratory assessment should include complete blood counts with differential counts and red cell indices. Many patients with advanced disease will be anemic at the time of diagnosis, often reflecting chronic blood loss and iron deficiency. Anemic patients treated with radiation or chemoradiation have poorer outcomes than do patients with near-normal hemoglobin levels.88–91 Anemia at diagnosis (before treatment) correlates significantly with reduced pelvic control and survival with univariate analysis, but not always when data are subjected to multivariate analysis. In contrast, multivariate analysis reveals that hemoglobin level during the course of radiation therapy or chemoradiation is a robust predictor of local outcome and survival.^88–91 Transfusion for anemic patients with maintenance of average weekly nadir hemoglobin levels at or above 11 to 12 g/dL through radiation therapy is associated with improvement in prognosis to that associated with patients with near-normal or normal hemoglobin levels at diagnosis.^88,90 This effect may be partially mediated through better tumor oxygenation and oxygen-enhanced radiation lethality to clonogenic cells, as well as reduced angiogenesis in better-oxygenated tumors. With pelvic or extended-field radiation that will encompass a substantial percentage of adult bone marrow, hemoglobin may decline slowly, even when adequate iron stores are present and patients are supported with hematinics. Concurrent administration of cisplatin further aggravates this problem, and can result in clinically significant reduction in hemoglobin levels over the course of a 6- to 8-week program of chemoradiation, even in patients with normal hemoglobin levels at the time of diagnosis. Frequent monitoring of hemoglobin levels as well as white cell counts and platelets should be performed throughout a course of chemoradiation, and hemoglobin level should be supported, either by transfusion or through the use of recombinant erythropoietin. The minimal hemoglobin level required is uncertain, but it seems prudent to target a minimum of 10 g/dL.

Neutropenia may indicate supportive treatment with granulocyte colony-stimulating factor to avoid prolonged treatment interruptions that are known to adversely affect local tumor control in patients treated with radiation.92,93 Absolute neutrophil counts should be monitored at least weekly in patients undergoing chemoradiation.

An elevated platelet count has been associated with advanced malignancy and is considered a consequence of increased platelet production.⁹⁴ Hernandez and associates⁹⁵ identified this effect in advanced cervical cancer, and Rodriguez and coworkers⁹⁶ identified the preoperative platelet count as an adverse prognostic factor, even for patients with stage Ib cervical carcinoma. The cumulative 5-year survival of women with a platelet count greater than 300,000 (85 women) was 65%, compared with 84% for the group with a normal value (134 women). At issue was the question of whether the value could have been elevated simply because of bleeding from the primary tumor, but no association was found with the preoperative hematocrit. This study of surgically treated patients with early disease also demonstrated that the effect was not a consequence of metastatic disease and did correlate with tumor volume, with nearly half of the patients with platelet counts in excess of 300,000 having “large” lesion size, in contrast to only 28% (32 of 114) patients with normal counts.⁹⁷ In a multivariate analysis, adjusting for age, race, tumor size, and presence of lymph node metastases, high platelet count was still associated with an adverse prognosis.

Additional laboratory tests should include a complete chemistry panel with serum electrolytes and measurements of renal and hepatic function. Electrolytes and renal function should be monitored repeatedly through a course of chemoradiation. Creatinine may be elevated secondary to hydroureter/hydronephrosis that may require ureteral stent placement or nephrostomies before radiation therapy, particularly if cisplatin chemotherapy is anticipated. Serum sodium and potassium may become abnormal during the course of chemoradiation, particularly in elderly patients, consequent to diarrhea with potassium loss, inadequate oral replenishment, and inadequate or inappropriate fluid intake.

Prognostic Factors

Prognostic factors fall into two groups: tumor-related factors and patient-related factors. Probably the most important tumor-related factor is tumor size, and this is true both for patients treated with hysterectomy 97,98 and for patients treated with radiation therapy.^101–101 In fact, FIGO, in 1995, divided stage Ib into two groups according to size; other stage categories act, in part, as surrogates for tumor size. For patients treated with surgery, histologic evidence of extracervical spread is associated with a poorer prognosis. Parametrial extension is associated with a higher rate of lymph node involvement, local recurrence and death from cancer.^102,103 Uterine body involvement is associated with an increased rate of distant metastases in patients treated with radiation therapy or surgery.^104,105

Lymph node involvement is another important tumor-related factor. After radical hysterectomy, reported-year survival rates usually are about 35% to 40% lower when the pelvic lymph nodes are involved.106,107 However, studies suggest that postoperative chemoradiation improves these results.¹⁰⁸ Several reports suggest that survival decreases with increasing size of the largest involved nodes,^109,110 increasing number of nodes involved,^106,110,111 and the increasing level of regional involvement and with extent of central disease in the cervix. Overall, the survival rates for patients with positive paraaortic nodes are about half those of patients who have similar stages of disease without paraaortic lymph node involvment.^114–114 Lymphovascular space invasion (LVSI) also is correlated with an increased risk of recurrence. This reflects, in part, the strong correlation between LVSI and lymph node involvement; however, in a large number of postoperative studies, LVSI is an independent predictor of prognosis.^{107,115–117}

Investigators have compared the outcome of patients with adenocarcinomas with squamous carcinomas and have reached varying conclusions about the relative prognoses of the two histologic types of cervical cancer.120–120 In a review of 1767 patients with stage Ib disease (229 with adenocarcinoma), Eifel and colleagues¹²¹ found that patients with adenocarcinoma had a significantly higher risk of recurrence and death from disease, independent of age, tumor size, or tumor morphology. The rate of distant metastasis for patients with bulky (>4 cm) adenocarcinomas was almost twice that for patients with squamous carcinoma (37% vs. 21%, P < 0.01). Several investigators have reported high recurrence rates after radical hysterectomy for adenocarcinomas. In a subset analysis of a randomized Gynecologic Oncology Group study, Rotman and associates¹²² reported a high recurrence rate (11 of 25 patients [44%]) after treatment with radical hysterectomy alone for adeno-adenosquamous carcinomas of the cervix; in contrast, only 3 of 31 patients (10%) who received postoperative radiation therapy had recurrences. However, all of these comparisons had a relatively small number of patients treated with cervical adenocarcinomas.

Other tumor factors include correlation between the serum concentration of squamous cell carcinoma antigen and the extent of squamous carcinoma of the cervix.125–125 Increased tumor vascularity has been associated with a relatively poor prognosis,^126,127 and a strong inflammatory response in the cervical stroma tends to predict a good outcome.¹²⁶ Some authors have reported a correlation between HPV subtype and prognosis.²⁷ Several investigators have reported a higher recurrence rate in patients with histologically negative lymph nodes when a polymerase chain reaction assay of the lymph nodes was strongly positive for HPV DNA.^128,129 Other authors have correlated poor prognosis with the presence of HPV messenger RNA¹²⁸ or HPV-related proteins¹³⁰ in the peripheral blood of cervical cancer patients.

Patient-related factors have been discussed elsewhere in this chapter and include age, hemoglobin, platelet counts, race, and smoking.¹³¹ Several studies have demonstrated that in the United States, women of color with invasive cervical cancer present with higher-stage disease^134–134 and lower hemoglobin levels^135,136 than do white women. Differences in socioeconomic status may influence minority patients’ access to care and ability to comply with treatment recommendations.¹³⁶ In a study of 1304 women treated with radiation for cervical cancer, Kucera and colleagues¹³¹ reported that smokers with cervical cancer had a poorer 5-year survival rate than nonsmokers. This difference was statistically significant for patients with stage III disease (5-year survival rate 20.3% vs. 33.9%, P < 0.01). However, the authors did not analyze the possible influence of smoking-related deaths from causes other than cervical cancer.

Treatment Overview

The treatment of patients with cervical cancer is determined primarily by the stage and extent of disease. As a general rule, CIS and microinvasive cervical cancer with no more than 3 mm of stromal invasion and no more than 7 mm breadth (FIGO stage Ia1) are managed with conservative surgery (excisional conization or extrafascial/simple hysterectomy). Microinvasive cervical cancer with 3 to 5 mm of stromal invasion and no more than 7 mm breadth (FIGO stage Ia2) may be managed with modified radical hysterectomy. Stages Ib1 and small-volume IIa may be managed with either radical surgery (generally, radical hysterectomy or modified radical hysterectomy), fertility sparing surgery if less than 2 cm, radiation alone, or chemoradiation. Controversies surround the optimal management of patients with bulky stage Ib2 (>4 cm) cancer, who may be treated with triple-modality therapy in selected instances (surgery, radiation, chemotherapy in a variety of possible sequences).

After initial radical surgical therapy, patients with adverse histopathological factors may be advised to undergo adjuvant pelvic radiation or adjuvant chemoradiation. Risk factors for recurrence include metastatic disease in regional nodes, positive surgical margins, parametrial extension, deep cervical stromal invasion, lymphovascular space invasion, and large tumor size.

Patients with large FIGO stage IIa or with stages IIb through IVa disease usually are managed with chemoradiation. Some patients with an incomplete response to chemoradiation may benefit from a combined approach using adjunctive hysterectomy to clear persistent central disease after chemoradiation. Patients with bulky Ib2 or IIa disease and rare patients with bulky central stage IIb lesions with minimal, medial parametrial invasion may be considered for this approach. For patients in whom disease recurs centrally in the pelvis after maximal chemoradiation, radical exenterative surgery can be performed, provided that no distant disease is present. Optimal candidates have mobile central disease without lymph node involvement.139–139 Intraoperative radiation therapy may be part of salvage surgery when lateral or posterior surgical margins will be predictably inadequate.^140,141 Recurrent disease after initial radical surgery historically has been treated with salvage radiation, but currently should be treated with chemoradiation.¹⁴²

Superficial Ablative Therapy

A major deficiency of all ablative therapies is the absence of full histopathological assessment of the lesion treated, and the hazard of undertreatment of an occult invasive lesion. For this reason, use of ablative techniques is declining. Three superficial ablative techniques used historically are electrocoagulation diathermy, cryosurgery, and CO₂ laser therapy.

Hysterectomy

Hysterectomy involves the removal of the uterus and varying amounts of surrounding tissue (Fig. 87-12). Because the risk of ovarian metastases is low (0.5%), ovarian preservation usually is recommended in premenopausal women, obviating the need for hormone replacement therapy.¹⁴³ An important exception is when nodal metastases from a primary cervical adenocarcinoma are detected intraoperatively, when the risk of ovarian metastasis escalates steeply, perhaps to as high as 25%.¹⁴⁴ Five types of hysterectomy have been described.¹⁴⁵