Carcinoma of the uterine cervix is the most common female malignancy worldwide, although in the United States, it ranks only eighth (Section 42.2) 135. According to the Centers for Disease Control and Prevention, this remarkable 70 per cent reduction in deaths from cervical cancer is the result of effective Papanicolaou (Pap) smear screening programmes which identify the neoplasia before it becomes invasive. However, the American Cancer Society estimates that during 1993 in the United States there will be 13 500 new cases of invasive cervical cancer and 4400 deaths from the disease.

The incidence of cervical cancer correlates with sexual intercourse at an early age, a history of multiple sexual partners, and a high number of pregnancies. Some time ago herpes simplex virus Type 2 (HSV2) was also implicated as a likely cause of cervical cancer. Miller reported a high prevalence of HSV2 antibodies associated with the disease. Several recent studies, however, have not shown the same relationship. In retrospect, it may be that HSV2 antibodies are a ‘marker’ for the known epidemiological risk factor, namely multiple sexual partners, through which serious cancer cofactors are encountered.

Nonetheless, research continues to implicate viruses, along with cigarette smoking, as major cofactors for cervical cancer. In one study, the DNA of human papilloma virus (HPV) types 16 and 18 was found in 62 per cent of patients with the malignancy but in only 32 per cent of controls. Other series have shown even stronger associations between certain HPV subtypes and cervical cancer. Most recently a correlation between the AIDS virus (HIV), intravenous drug use, and cervical cancer has been reported. Indeed, by definition, the presence of cervical cancer in a patient hitherto reported as HIV-positive upgrades her disease status to that of having active AIDS.

The long recognized, clear relationship, between squamous cell cancer of the cervix and sexual/social behaviour permits the malignancy to be categorized as a sexually/socially transmitted disease and thus theoretically preventable.

Over 90 per cent of invasive cervical cancers are histologically defined as squamous cell in origin; the other 10 per cent comprise various adenocarcinomas. Although a less important distinction, 70 per cent of squamous cell cancers are of the large cell non-keratinizing subtype. Adenocarcinoma arises primarily from the columnar epithelium of the endocervix, rather than from the better known squamous epithelium covering the exocervix and the site of squamous neoplasia.

There are many less common cervical cancer variants, each with unique biological behaviours and prognoses. Adenosquamous and adenoid cystic carcinomas generally carry a poor prognosis, whereas adenoid basal and verrucous carcinomas of the cervix have good prognoses. The last tumour, characterized by pushing borders, is locally aggressive and is treatable by local resection. The infrequent small cell cancer of neuroendocrine origin is an extremely virulent cervical neoplasm that can present with paraendocrine symptoms. Attempts to treat small cell cancers of the cervix have been modelled upon therapies used against the synonymous lesion occurring in the lung. But even when surgery, radiation, and chemotherapy are combined, the prognosis for this cervical malignancy is poor. Very occasionally lymphomas, embryomas, embryonal rhabdomyosarcomas, leiomyosarcomas, malignant mixed muellerian tumours, melanomas, carcinoids, and primary sarcomas have all been reported arising from the uterine cervix.

Cancers found on the cervix are not always primary tumours from that site. Malignancies from the corpus of the uterus, breast, colon, and kidney are known to metastasize to the cervix.

Given the dominating prevalence of the squamous cell histology among the cervical malignancies, most of this Section will address the clinical questions related to squamous cell cancer of the cervix.

Invasive squamous cell cancer is frequently associated or coexistent with squamous intraepithelial lesions (dysplasias, cervical intraepithelial neoplasias (CIN)). This finding reinforces the concept that cervical neoplasia is a ‘continuum’ of disease ranging from mild dysplasia (CIN 1) through to invasive cancer. Traditionally, there has been a slow, orderly biological progression across the continuum (over 2 to 20 years) from preinvasive neoplasia to frankly invasive squamous cancer. Peterson followed 127 patients with untreated carcinoma- in-situ of the uterine cervix. Thirty per cent of the patients developed invasive cervical cancer by the tenth year of observation. Today, however, the caveat ‘all bets are off’ more accurately describes the less predictable clinical behaviour of cervical intraepithelial neoplasia. On the one hand, many cases appear to resolve spontaneously. On the other, rapid, unpredictable, malignant transformation can occur, possibly in conjunction with high-risk cofactors, such as HPV subtypes 16 and 18, HIV, and smoking. The occasionally virulent new biology of cervical neoplasia challenges the wisdom of recommendations for screening that suggest increasing the interval between Pap smears.

The next level of neoplasia after carcinoma- in-situ (CIN 3, high grade SIL) is microinvasion. ‘Microinvasion’ is a concept suggesting that an invasive cancer can exist, but be so minimal at the moment that it is still a local phenomena (easily resectable for cure). Fundamental to the concept is a total absence of metastases.

Expert opinions differ in the definition of ‘microinvasion’. Pragmatically, the Society of Gynecologic Oncologists defined the lesion as a malignancy with a depth of invasion of up to 3 mm (from the base of the epithelium), without vascular or lymphatic space involvement, and without multiple foci of invasion. More recently the International Federation of Obstetrics and Gynecology (FIGO) modified its staging of cervical cancer, overall, in which a new, more complex, possibly less practical definition of microinvasion is given (Table 1) 413. The higher definition of microinvasion (Stage 1a²) by FIGO allows, with conditions, a depth of cervical stromal invasion up to 5 mm.

The deeper penetration permitted by FIGO Stage 1a² has been associated with a greater risk of pelvic lymph node metastasis. In one study the risk of pelvic nodal spread was 6.8 per cent for FIGO Stage 1a² patients and 0.21 per cent for patients where the Society of Gynecologic Oncologists' definition of less than 3 mm invasion was used. Similarly, other studies found that with stromal invasion of less than 3 mm the risk of lymph node metastasis was very low, 0–1 per cent, but when invasion deepened to between 3.1 and 5 mm the risk of node metastasis increased to 4 to 7.3 per cent. Moreover, in a disturbing study 85 FIGO Stage 1a² patients were followed 5 years after conservative treatment (non-radical surgery), and three patients died from recurrent disease.

Considering that ‘unambiguous curability’ is the keystone of microinvasion, the above results are incongruous and conceptually unacceptable. A safer clinical axiom is that ‘microinvasion’ exists, but must be carefully defined, and FIGO designation 1a² is only a technical stage and not a guideline for choosing treatment.

After microinvasion, at the worst end of the neoplastic spectrum, is frankly invasive squamous cell cancer. Inherent in this ultimate form of neoplasia is the property of angiogenesis, uncontrolled growth, and metastasis.

Recently a spectrum of ‘adenoneoplasia’ of the cervix has been described, which appears similar to that established for squamous cells. However, unlike squamous lesions in which a clear-cut demarcation (the basement membrane) defining invasion can be readily determined microscopically, no equivalent absolute boundaries exist with glandular neoplasia. This problem makes diagnostic distinctions of degree more difficult. Clinical wariness is justified when formulating treatment plans for ‘adenocarcinoma in-situ’, or ‘microinvasive adenocarcinoma’, since the diagnoses are, by necessity, based on less precise criteria.

Beginning on and expanding in the cervix, the squamous lesion, over time, usually extends laterally into the paracervical and parametrial soft tissues ultimately involving the pelvic sidewall and obstructing the ureters. Simultaneously, the lesion often spreads inferiorly into the vagina and superiorly into the lower uterine segment. Direct, early extension into the bladder or rectum can occur but is uncommon.

Once the malignancy has invaded the underlying cervical stroma, it may enter the paracervical and parametrial lymphatic/vessel spaces. As a consequence, the obturator, internal iliac, and external iliac lymph nodes become sites of metastasis. As the malignancy ascends lymphatically, the para-aortic nodal chain up through the supraclavicular group becomes involved.

In collected series, patients with Stage I (Table 1) 413 cervical cancer have a 15 per cent incidence of pelvic lymph node metastasis, Stage II, 27 per cent, and Stage III, 46 per cent. By comparison, the para-aortic nodes have metastases of 2, 15, and 30 per cent respectively. It is rare for the para-aortic lymph nodes to be affected if the pelvic nodes are not.

The most common distant sites of metastases include the lungs, mediastinum, bone, and liver. Haematogenous dissemination occurs in less than 10 per cent of patients.

The first clinical manifestation of invasive cervical cancer is usually abnormal vaginal bleeding. Among symptomatic patients, there are complaints of intermenstrual bleeding, postcoital (the ‘classic’ but less common clinical history), and postmenopausal bleeding, occurring in friable, frequently ulcerated, cervical epithelium. When cervical cancer is necrotic, ‘vaginal discharge’ is a common feature. Pain (pelvic, lumbosacral, gluteal, or in the back), is a symptom that often heralds advanced disease. For example, abdominal and low back pain may be caused by paraaortic lymph node metastases. Occasionally, at presentation, there may be haematuria or rectal bleeding suggesting involvement of the bladder or rectum. In countries where Pap smear screening exists, however, most cervical cancer patients are actually asymptomatic at the time of their diagnosis.

The diagnosis of cervical cancer is established by biopsy and histological evaluation. All abnormal lesions on the cervix should be examined by biopsy. In a clinically normal appearing cervix, cytology by Papanicolaou smear is obtained. However, even though the correlation between cytology (pap smear) and histology (biopsy) from the cervix is 60 to 85 per cent, cytology itself is insufficient for diagnosis of any degree of cervical neoplasia, including malignancy. Instead, Pap smears are ‘red flags’, warning of possible neoplasia and demanding that further investigation be made to make a histological diagnosis.

The follow-up test done to evaluate an abnormal Pap smear is colposcopy. An office technique, colposcopy effectively determines where on the cervix to take a biopsy sample in pursuit of dysplastic/malignant tissue. Because colposcopy is a simple, accurate procedure, the more costly and morbid surgical cervical conization under anaesthesia is rarely needed. However, cone biopsy is still indicated when colposcopically-directed cervical biopsies fail to determine the degree or depth of the lesion, or when non-visualized neoplasia exists in the cervical canal as revealed by an endocervical curettage.

FIGO staging of cervical cancer is based on its usual malignant progression (Table 1) 413. In contrast to ovarian and endometrial malignancies, which are staged surgically, cervical cancer is staged clinically. As such, the single most important staging parameter with cervical cancer is the clinical/pelvic examination. Chest radiograph and a radiographic evaluation of the ureters are also performed routinely. Though intravenous pyelography is still the standard means of evaluating the genitourinary system in patients with cervical cancer, the CT scan, is quickly replacing it because of the more comprehensive information provided. Cystoscopy and proctoscopy can be important staging tests, depending on location and extent of disease, but generally they are neither needed, nor cost effective, in staging of early disease.

Other tests not formally part of FIGO staging, but helpful in planning treatment for certain cases, are abdominal-pelvic CT scan, ultrasound, and magnetic resonance imaging (MRI). Heller studied 320 patients with Stage IIB, III, and IV disease with lymphangiogram, CT scan, and ultrasound. When evaluating pelvic lymph nodes, the false negative rates were 14, 25, and 30 per cent respectively. In another study, CT scan was the most valuable test in assessing para-aortic nodes (100 per cent specificity and 67 per cent sensitivity). MRI may be helpful for depicting lower uterine segment tumour involvement and extracervical extension.

FIGO Stage Ia1 (early microinvasion) can be treated by either cone biopsy or simple total hysterectomy (Rutledge Type I; Table 2 414). The same surgical options exist for the Stage Ia2 disease, but there is 5 per cent recurrence rate after cone biopsy compared with a zero rate following radical hysterectomy (Rutledge Type III) and pelvic lymphadenectomy (see Section 30.16 214). Thus, after careful consideration, the increased risks inherent in pursuing a conservative treatment course (i.e. cone biopsy) must be assumed by the well-informed patient. Generally, this difficult choice of the higher risks of recurrence after limited surgery has to be balanced against a strong desire to preserve future fertility.

For Stages Ib to IIa lesions of squamous or typical adenocarcinoma histology, primary treatment options are either a radical hysterectomy (Wertheim–Meigs) with lymphadenectomy or radical radiation therapy (whole pelvis plus brachytherapy). Although comparable 5-year survival rates are reported, apart from ovarian conservation, surgery as primary treatment has additional advantages, for example, less long-term morbidity. Radiation is associated with vaginal fibrosis and chronic bladder and bowel problems, 8 per cent of which require medical and/or surgical intervention. Another major advantage of surgery is the ability to determine the extent of disease more precisely. The better understanding of actual disease distribution in turn, permits a more exacting separation into good and poor prognosis groups.

In 1912 Wertheim introduced the radical hysterectomy as a primary management of cervical cancer, and during his career carried out 500 radical operations with a substantial 30 per cent operative mortality rate and a survival rate of 40 per cent. In 1944, Meigs from the Massachusetts General Hospital enhanced the surgical approach to cervical cancer by integrating pelvic lymphadenectomy with the radical hysterectomy. Despite the greater complexity of the operation, Meigs reported a mortality rate of only 1 per cent. Moreover, the 5-year survival was better (90 per cent for node-negative patients and 42 per cent with those who were node positive). Acute complications of radical hysterectomy and pelvic lymphadenectomy then, as now, include haemorrhage, infection, and deep vein thrombophlebitis. Delayed complications are bladder atony in 3 per cent, genitourinary fistula in 1 to 2 per cent, and lymphoceles (see Section 10.5 73).

The advantages of radical surgery over radical radiation are shorter treatment duration, less expense, fewer long-term complications, and better post-treatment sexual function. These benefits, coupled with modern surgery, anaesthesia, and postoperative care which have lessened operative morbidity still further, explain why surgery has emerged as the primary treatment of choice for early cervical cancer, even for patients once dogmatically declared ‘too old’. If, however, at laparotomy unexpected intraoperative findings are encountered (e.g. unresectable disease or nodal metastases), departure from the intended radical hysterectomy may be appropriate in deference to pursuing an alternative treatment approach (e.g. radiation with or without chemotherapy) more consistent with the new-found scope of disease.

In a prospective study of 940 patients with Stage Ib cervical cancer who had radical hysterectomies and no gross intraoperative findings, the Gynecologic Oncology Group analysed pathological findings predictive of prognosis. Table 3 415 summarizes some of those features. Among them, grade, vascular invasion, parametrial extension, and depth of stromal invasion were correlated with disease of the pelvic nodes. Interestingly, lesion diameter was not a important variable in this study but is generally considered an important risk factor based on other reports. The presence of positive lymph nodes carries a range of prognoses which are poor overall (Table 4) 416.

Based on the information obtained at surgery, patients found to be at high risk for recurrence can undergo adjuvant radiotherapy or chemotherapy in a protocol setting in an effort to improve outcome. Surprisingly, despite years of use as the adjuvant local/regional therapy after radical hysterectomy for high-risk patients, pelvic radiation has not shown survival benefit, and has additive morbidity. The reason for failure of adjuvant radiation therapy may be its inability to sterilize gross nodal disease and also its inability, as a local/regional therapy, to control disseminated disease. A role for adjuvant chemotherapy here, as in breast and ovarian cancer would seem logical, although there are few data. Several small studies have suggested adjuvant chemotherapy can improve outcome for some patients with high-risk disease after surgery. More trials are needed.

Some cases of Stages Ib to IIa cancers merit consideration for special management. The expanded endocervix, ‘barrel cervix’, is an example. Usually an adenocarcinoma arising within the endocervical canal balloons out of the entire cervix drastically (e.g. >6 cm diameter). The now bulky-shaped cervix, because of its configuration, is not geometrically suitable to receiving uniform high-dose brachyradiation for cure. The barrel cervix is so large and distorted that curative tumouricidal isodose curves encompassing the entire ‘barrel’ are not possible, leaving some peripheral and hypo-oxygenated areas of the lesion undertreated. This accounts for a central failure rate of 17 per cent reported after treatment of the bulky cervix with radiation alone. Controversy exists, however, regarding the best clinical management for the problem, but many centres still believe that radical radiation should be followed by extrafascial hysterectomy to enhance local control in the management of barrel/bulky cervix.

Other special cases of cervical cancer are cervical cancer/neoplasia occurring during pregnancy, carcinoma of the cervical stump after supracervical hysterectomy, the unexpected finding of cancer of the cervix after ‘benign hysterectomy’, cervical cancer and a coexisting pelvic mass or pelvic inflammatory disease, cervical cancer and ureteral obstruction, and cervical cancer and AIDS. Each of these complex situations needs appropriate, sometimes unique, interventions (or no intervention) determined within the context of the special circumstances and specific patient.

Stage IIb to IVa cervical cancers, by default, have been managed moderately well by using primary radical radiation therapy, alone. Five-year survival results vary with stage and the institution reporting but are generally between 20 and 50 per cent. Many radiation therapy failures are the result of treating unrealistically large tumour volumes, which are beyond the curative capacities of tolerable radiation doses. Additionally, at least 30 per cent of Stage IIb to IVa patients have metastatic disease in the para-aortic lymph nodes. Extended-field radiation, covering the pelvis and incorporating a para-aortic ‘chimney’, has had minimal impact on widespread nodal disease of this type. The problem is that tumouricidal radiation doses cannot be given because of toxicity, and thus constrained radiation has a limited ability to sterilize malignant lymph nodes, especially those that are large or widely dispersed.

Stage IVb cervical cancer is incurable, and accordingly, treatment is individualized and palliative. For example, high-dose, sometimes even single-fraction, external radiation therapy can be used as an unconventional but creative means to control vaginal bleeding. The advantage of this approach over conventional, 1.8 Gy daily fractions to a total of 45 Gy, is less intrusion into the remaining life of the patient.

The rare Stage IVa lesion (local disease extending directly anterior or posterior into the bladder or rectum) has been a partial exception to the ‘rule’ that primary radiation is the only treatment for advanced cervical cancer. Barring evidence of distant/unresectable disease, Stage IVa cancers may also be amenable to cure with ultraradical surgery (primary pelvic exenteration). Apart from this uncommon situation, surgery has, to date, played little role in the treatment of advanced cervical cancer since it is, by itself, less effective than is radiation in that setting.

Chemotherapy, given as single agents or in combinations, can produce short-term (4–7 months) responses. Drugs that have been used are cisplatin/carboplatin, bleomycin, vincristine, mitomycin C, ifosfamide, 5FU, VP16, methotrexate, both singly and in various groupings. At present, the increased toxicity of combination regimens does not seem justified since cisplatin alone yields equivalent survival prolongation. Cisplatin produces a 20 to 30 per cent response rate with recurrent squamous cancer of the cervix including a 10 per cent complete response rate. Interestingly, no clear dose–response relation has been demonstrated using cisplatin to treat cervical cancer, so a low, less toxic dose (e.g. 50 mg/m²) is appropriate. Unfortunately, responses are limited.

Today chemotherapy is being employed in new roles in the management of cervical cancer. One example is neoadjuvant chemotherapy. Chemotherapy is used first, to shrink the tumour and second, to perhaps control micrometastases, prior to intense local treatment, with surgery or radiation. Neoadjuvant therapy seems logical, but studies to date are limited and results on efficacy have been mixed.

In another new role, chemotherapy and radiation are being administered concurrently. It is intended that chemotherapy should sensitize or enhance the local ongoing effects of radiation. In the past, similar efforts have been made to boost the therapeutic index of radiation using other innovative concepts, such as hyperbaric oxygen, neutron therapy, and changes in dosing, timing, and fraction size. Therapeutic improvements have not been obvious. It seems that there is relatively little radiobiological difference between tumour cells and normal cells. Thus, all the cells, normal and malignant, are comparably affected by whatever cytotoxic changes are made in treatment. There have been, nevertheless, several small studies suggesting that better local control, and even improved survival, can be achieved with the concurrent use of radiation with either hydroxyurea, 5-fluorouracil, cisplatin, or combinations of these agents. Important prospective trials using these concurrent therapies are ongoing.

All patients at Women and Infants' Hospital/Brown University with gynaecological malignancies are presented to a multidisciplinary, prospective Tumor Board for primary treatment recommendations, and again, as new information or events warrant. Primary treatment protocols generally suggested by the Tumor Board for cervical cancer are as follows.

Stage Ia¹ (less than 1 mm cervical stromal invasion) is managed either conservatively with cone biopsy or, if fertility is not desired, by extended hysterectomy with removal of suspicious lymph nodes. Stage Ia², with up to 3 mm of stromal invasion, can be treated as Stage Ia¹. However, in this subgroup of Ia² (< 3 mm invasion) if fertility is not strongly desired, or for virtually all Stage Ia² patients with deeper or more extensive lesions, a more definitive surgical intervention is recommended, typically a modified radical hysterectomy with pelvic lymph node sampling.

The majority of patients with Stage Ib and early IIa cancers undergo radical hysterectomy, pelvic lymphadenectomy, and additional staging as indicated. An alternative to radical surgery, in specific cases, is primary radical radiation. At laparotomy, the actual surgery may be modified, redirected, or stopped if unexpected circumstances are encountered. After surgery, patients are stratified, based on the surgical–pathological findings, and those deemed at high risk are offered adjuvant treatment on various research protocols (e.g. radiation therapy and/or chemotherapy).

Patients with advanced disease may opt for management (often aggressive) on experimental protocols as an alternative to conventional radiation therapy used alone. For example, some protocols give neoadjuvant (primary) chemotherapy and/or radiotherapy; depending on response, surgery and staging may be added. Other protocols for advanced disease use a primary surgical staging, followed by additional treatments tailored to the surgical findings. Chemotherapy, either as a radiation sensitizing agent, or as a therapeutic modality in its own right, is often combined with radiation after primary surgical staging or after radical pelvic surgery.

The overall 5-year survival rate of patients with cervical cancer is only 50 per cent. Stage of disease is the single most important predictor of survival. Five-year survival rates for patients with Stage Ia disease is 98 per cent, for Stage Ib–IIa, 75 to 85 per cent, Stage IIb, 55 per cent, for Stage III, 10 to 50 per cent, and for Stage IVb essentially none. Discouragingly, despite improved radiotherapy techniques and better delineation of the extent of the disease, survival rates for cervical cancer have not improved much in the last 30 years.

The overall recurrence rate of cervical cancer is 35 per cent. In contrast, only 10 to 20 per cent of Stage Ib patients experience recurrence of disease. This recurrence rate is increased to 45 per cent if affected nodes were found originally. Recurrence generally occurs within the first 2 years following primary treatment, and most patients who have recurrence die from the disease.

Recurrent cervical cancer generally has a poor prognosis and must be treated within the context or constraint of the site of the recurrence and the type of therapy previously given (surgery and/or radiation). Sites of recurrence are categorized as local/central (potentially resectable and curable), regional (in the pelvis, but extending to the sidewall or lymph nodes, unresectable and rarely curable), and distant.

A true local (central) recurrence after surgery can be managed with radiation therapy with survivals reported as high as 25 to 50 per cent. In previously radiated patients with central recurrence of the disease, selective use of pelvic exenteration offers a 30 to 60 per cent 5-year survival if total resection is possible. Because of the high complication rate (Table 5) 417 and the major long-term physical and psychological morbidities that patients can encounter, exenterative surgery is only indicated if cure is possible.

Surgical candidates must be carefully selected and restaged and should have no evidence of disease beyond the, already known, central lesion. The dreaded triad, unilaterally oedema, sciatic pain, and ureteral obstruction, suggest almost certain unresectability, and thus are relative contraindications for exenteration. Unfortunately, even in the face of an asymptomatic patient with ‘normal’ CT scan and MRI, distant disease or other absolute contraindications for exenteration are commonly encountered, but not until laparotomy. When contraindications are found, the operation is curtailed if possible since it cannot provide cure.

Extensive, multidisciplinary, preoperative preparations for exenterative ‘candidates’, and their families, is essential so they understand the indications for surgery, the components of surgery, the morbidities, the sequence of intraoperative events, the possibility of a ‘no go’, the possibility of mortality, and the high possibility of having a recurrence, even if the exenteration is carried out under optimal circumstances. Despite all the negative thinking surrounding exenterative surgery, after consideration, most candidates opt for the aggressive approach as it represents their only chance of cure.

Unfortunately, most cervical recurrences are unresectable or diffuse, and are thus not appropriate for surgical salvage. These patients with recurrence can be considered for experimental treatment protocols using radiation with or without concomitant systemic agents. The occasional temporary response can be quite good. In addition, chemotherapy alone can have a palliative role, although generally responses are brief.

A new, less predictable, more virulent, cervical neoplasia may be emerging associated with high-risk cofactors like HIV, subtypes of human papillomavirus, and cigarette smoking. Future screening and management of cervical neoplasia will need to be adjusted to fit this evolving biology. Traditional treatment of cervical cancer yields good results for early stage disease, but overall treatment achievement is only modest, and our standard approaches seem to have reached their limits. Clinical staging of cervical cancer can be helpful in comparing treatment modalities but does not define the volume and extent of the disease and is inaccurate relative to surgery. Surgery is increasingly preferred as the primary treatment for early cervical cancer because of its effectiveness as therapy, the staging information it provides, its lower cost and lesser long-term morbidity, and the relatively superior quality of life (sexual and otherwise) it affords patients after treatment, regardless of age. Radiation therapy has a marginal capability to sterilize bulky local disease or clinically enlarged lymph nodes, limiting its value as treatment for regional or advanced cervical cancer. Current chemotherapy yields only short-term responses and can be toxic. Single agent low dose cisplatin seems to achieve comparable results to those achieved with more toxic combination regimens. Better, more rational treatment approaches are needed for particularly high-risk local and advanced disease. Trials evaluating neoadjuvant, adjuvant, and sensitizing chemotherapy, blended with surgery and radiation are ongoing and will point the way towards improved management of cervical cancer.

Abayomi O, et al. The value of ‘routine tests’ in the staging evaluation of gynecologic malignancies: a cost effectiveness analysis. Int J Rad Oncol Biol Phys 1982; 8: 241.

Averette H, Nelson JH Jr., Ng AB, Hoskins W, Boyce JG, Ford JH. Diagnosis and management of microinvasive (Stage la) carcinoma of the uterine cervix. Cancer 1976; 38: 414.

Averette HE, Lichtinger M, Sevin BU, Girtanner RE. Pelvic exenteration: a 15-year experience in a general hospital. Am J Obstet Gynecol 1984; 150: 179.

Berek JS, et al. Adenocarcinoma of the uterine cervix: histologic variables associated with lymph node metastasis and survival. Obstet Gynecol 1985; 65: 46.

Burghardt E, et al. Microinvasive carcinoma of the uterine cervix (International Federation of Gynecology and Obstetrics Stage la). Cancer 1991; 67: 1037.

Creasman WT, et al. Management of Stage la carcinoma of the cervix. Am J Obstet Gynecol 1985; 153: 164.

Delgado G. Stage Ib squamous cancer of the cervix: the choice of treatment. Obstet Gynecol Surv 1978; 33: 174

Delgado G, et al. A prospective surgical pathological study of stage I squamous carcinoma of the cervix: a Gynecologic Oncology Group Study. Gynecol Oncol 1989; 35: 314.

Diagnosis and management of invasive cervical carcinomas. American College of Obstetricians and Gynecologists Technical Bulletin, 1989; No. 138.

Duguid HLD, Duncan ID. In Singer A, ed. Cancer of the Cervix: Diagnosis and Treatment. Philadelphia: W B Saunders 1985: 77.

Fletcher GH, Wharton JT. Principles of irradiation therapy for gynecologic malignancy. Curr Prob Obstet Gynecol 1978; 2: 2.Graham JB, Sotto LSJ, Paloucek FP. Carcinoma of the cervix. Philadelphia: Saunders, 1962.

Heller PD, Malfetano JH, Bundy BN. Clinical pathologic study of Stages IIb, III, and IVa carcinoma to the cervix: extended diagnostic evaluation for paraaortic node metastasis (a GOG study). Gynecol Oncol 1990; 38: 425.

International Federation of Gynecology and Obstetrics. Annual report on the results of treatment in gynecologic cancer. Int J Gynecol Obstet 1989; 28: 189.

Lawrence WD. Advances in the pathology of the utrine cervix. Hum Pathol 1991; 22: 792.

Marcial VA, Marcial LV. Radiation therapy of cervical cancer: new developments. Cancer 1993; 71: 1438.

Meigs JV. Carcinoma of the cervix—the Wertheim operation. Surg Gynecol Obstet 1944; 78: 195.

Peterson O. Spontaneous course of cervical pre-cancerous conditions. Am J Obstet Gynecol 1956; 72: 1063.

Piver MS, Rutledge FN, Smith PJ. Five classes of extended hysterectomy of women with cervical cancer. Obstet Gynecol 1974; 44: 265.

Raju KS, Bates TD, Taylor RW. Primary adenocarcinoma of the cervix: treatment and results. Br J Obstet Gynaecol 1987; 94: 1212.

Reeves WC, Brinton LA, Garcia M. Human papilloma virus infection and cervical cancer in Latin America. N Engl J Med 1989; 320: 1437.

Richart RM. Causes and management of cervical intraepithelial neoplasia. Cancer 1987; 60: 1951.

Roberts WS. Major morbidity after pelvic exenteration: a 7 year experience. Obstet Gynecol 1987; 69: 617.

Shingleton HM, Gusberg SB. Radical hysterectomy. In: Gusberg SB, Shingleton HM, Deppe G. Female Genital Cancer. New York: Churchill Livingstone, 1988: 534.

Ueki M, Okamura S, Maeda T. Individualization of patients for adjuvant chemotherapy after surgical treatment of cervical cancer. Br J Obstet Gynaecol 1987; 94: 985.

van Nagell JR, Depriest PD, Higgins PV, Powell DE. Surgical therapy for cervical cancer. In: Gershenson DM, DeCherney AH, Curry SL. Operative Gynecology. Philadelphia: W B Saunders Co, 1993: 271.

Vermorken JB. The role of chemotherapy in squamous cell carcinoma of the uterine cervix: a review. Int J Gynecol Cancer 1993; 3: 129.

Wertheim E. The extended abdominal operation for carcinoma uteri. Am J Obstet Gynecol, 1912; 66: 169.