Introduction

Cervical cancer is a relatively uncommon cause of neoplastic death in the United States, with declines over the past decade in both incidence and mortality.[1] Estimates indicate that in 1998, 13,700 cases of cervical cancer will be diagnosed in the United States, and 4,900 women will die of this disease.[2] Despite US declines in its incidence and mortality, cervical cancer remains a significant world health problem, and in several developing countries it is the leading cancer.[3]

The majority of cervical malignancies are squamous cell in origin and are thought to progress in an orderly fashion, from mild to severe dysplasia with eventual invasion of local tissues. Aside from invasion of organs in proximity to the cervix, the disease also progresses via the pelvic lymph nodes to the para-aortic and mediastinal lymph nodes. Visceral metastases usually occur late and are uncommon at presentation, even in patients with advanced local disease.[3,4]

Treatment focuses on detection and eradication of early or minimally invasive disease. This is accomplished by regular Papanicolaou smears with colposcopy, directed biopsies, limited excisions of the cervical tissue (eg, cone biopsies), and simple hysterectomy. Lesions that are too invasive or large, however, require more radical therapeutic approaches.

The management of invasive disease depends on the bulk of the lesion and the International Federation of Gynecology and Obstetrics (FIGO) clinical stage (Table 1). Surgery or radiotherapy may be offered to patients with invasive stage I through stage IIA disease. The typical surgical procedure consists of radical hysterectomy and bilateral lymphadenectomy. Radiotherapy generally involves a combination of external-beam and intracavitary techniques. Advanced disease is usually treated with radiotherapy. The outcomes of treatment according to clinical stage are presented in Table 2.

Improving outcomes in patients with stage II disease or greater remains a major problem. The issues are twofold. First, there is a need for strategies that are additive to radiotherapy in order to improve local control of the malignancy. Aside from providing a greater frequency of cure, ideally these strategies would reduce the significant radiation-related morbidity, such as fistulae and pain. A second issue is the eradication of visceral disease. Distant failure is still a component in the majority of advanced disease relapses. Effective systemic therapy could be applied in the neoadjuvant and adjuvant settings.

Current Systemic Therapy

Chemotherapy for refractory cervical cancer has an extensive history. Several conclusions may be drawn from the historical experience. Single-agent studies with positive results have approximately the same response rates of 15% to 30% with rare complete responses (Table 3). Cisplatin (Platinol) is considered to be the most active single drug.

Patients with recurrences in the radiated field seldom respond to chemotherapeutic drugs. This is theoretically attributed to the lower sensitivity of hypoxic tissues to the drugs. Patients in renal failure or those with poor performance status rarely benefit from chemotherapy.

Numerous trials have assessed combination chemotherapy in patients with cervical cancer. High response rates were documented even in patients who had received prior radiotherapy (Table 4). The durability of responses was distressingly short, however, lasting only a few months. Furthermore, when randomized trials were performed comparing single-agent to combination therapy, the results did not favor the combinations (Table 5). In these trials, neoadjuvant combination chemotherapy, usually cisplatin-based regimens, before radiation therapy again yielded high response rates but did not improve overall survival (Table 6).

The use of concurrent radiotherapy and cisplatin-based therapy is being studied, and results of randomized trials will soon be reported.

In summary, current systemic chemotherapy is associated with significant clinical response rates. However, these are seldom complete and are usually of short duration. Unfortunately, there has been no proof of survival prolongation, palliative benefits, or quality-of-life improvements with these therapies. Thus, there is a significant need for newer approaches.

Irinotecan: A Topoisomerase I Inhibitor

Topoisomerase inhibitors are nuclear enzymes with a multiplicity of cellular functions. Topoisomerase I induces single-strand DNA breaks that allow uncoiling and torsion relief in front of the DNA replication fork.[5-7] Camptothecin, an alkaloid from the leaves of the Chinese tree Camptotheca acuminata (Nyssaceae) is the parent compound of irinotecan (CPT-11 [Camptosar]).[8] The latter is a water-soluble derivative of camptothecin. SN-38, the active metabolite of irinotecan, is dependent on the concentration of the closed lactone ring, which is pH dependent.[9]

The pharmacokinetics of irinotecan are linear, with a biphasic or triphasic curve and a mean half-life of 10 hours.[10] The pharmacokinetics and pharmacodynamics of the active metabolite SN-38 are complex and differ from those of irinotecan.[9-11]

Preclinical screening demonstrated the activity of irinotecan in several models. Using a subrenal capsule assay, the two tested cervical cell lines showed growth suppressive effects of > 50%.[12] It is of interest that SN-38 augmented the activity of cisplatin, fluorouracil, and etoposide in HST-1, a human squamous cell carcinoma cell line.[13] This may result from inhibition of the removal of cisplatin adducts.[14]

Irinotecan appears to have radiosensitizing properties in small-cell and adenocarcinoma lung cancer cell lines.[15,16] The addition of recombinant tumor necrosis factor (rh-TNF) and irinotecan to several gynecologic cancer cell lines demonstrated synergy.[17] Curiously, caffeine also enhanced the growth inhibition rate of cisplatin plus irinotecan in various gynecologic cell lines.[18]

Single-Agent Activity

Five clinical trials have assessed irinotecan as a single agent in cervical cancer. The first phase II trial in the United States used a schedule of 125 mg/m²/wk for 4 weeks followed by a 2-week rest. A total of 42 patients (median age, 44 years; range, 24 to 59 years) who had not responded to prior chemotherapy were treated with a median of 2 cycles of irinotecan (range, 1 to 14). Irinotecan produced a response rate of 21%, with a median time to response of 6 weeks and a response duration of 12 weeks.

The major dose-limiting side effects were nausea and vomiting (45%), diarrhea (24%), and myelosuppression (36%). Myelosuppression did not decrease with dose reduction, whereas gastrointestinal side effects did. The investigators concluded that irinotecan’s clinical activity was significant and warranted further investigation, but that hematologic and gastrointestinal sequelae were problematic.[19]

The second US trial, conducted by the Gynecologic Oncology Group (GOG), enrolled 54 patients with recurrent or refractory cervical cancer. Most of these patients had received prior radiotherapy, and 12 had also received chemotherapy. Among 45 evaluable patients, there were 6 (13.9%) responses, 1 of which was complete. Gastrointestinal toxicity was grade 3 or 4 in 19 (39%) of patients. The authors concluded that the drug had modest activity with moderate toxicity and should be combined with cisplatin for future study.[20]

The European Organization for Research and Treatment of Cancer (EORTC) conducted a trial of irinotecan as primary chemotherapy in patients with cervical cancer. Patients were stratified according to whether they had measurable disease outside of a previously irradiated area (group A) or within the irradiated area (group B). The dose of irinotecan was 350 mg/m² given every 3 weeks. Five (24%) responses occurred in group A, as compared with none in group B. The overall response rate was 15%, and the duration of response was 6+ months. There were two deaths secondary to myelosuppression, diarrhea, and dehydration, however. Further studies were recommend- ed to better define the gastrointestinal side effects of irinotecan.[21,22]

The Japanese have had extensive experience with irinotecan in gynecologic cancer. One study used a schedule of 100 mg/m² weekly for four doses. Among 24 patients enrolled in this study, 5 (21%) responded. Another 31 patients were treated with a schedule of 150 mg/m² of irinotecan every 2 weeks for three doses. Eight (26%) patients responded. It is notable that the majority of patients in both groups had received prior radiotherapy and chemotherapy.

An analysis of toxicity in the Japanese studies, which included the results of Takeuchi et al, showed that myelosuppression and gastrointestinal side effects were significant and deaths were reported. No recommendation was made regarding further study.[24-26]

Combination Therapy

Irinotecan has not been extensively studied in combination with other agents in the treatment of cervical cancer. Noda et al evaluated the combination of irinotecan and cisplatin. Irinotecan was given on days 1, 8, and 15, and cisplatin was administered on day 1 only. Cycles were repeated every 29 days. The recommended doses were 60 mg/m² of cisplatin and 60 mg/m² of irinotecan. Of 12 patients treated, 6 had major responses. A phase II trial of this combination is under way.[K. Terada, personal communication, August 29, 1997]

Conclusions

The topoisomerase I inhibitors have not been extensively studied in cervical cancer (Table 7). Irinotecan demonstrates definite, but modest, single-agent activity. The drug produces patterns of response similar to those seen in previous studies, ie, relatively low response rates in irradiated areas and in patients with poor performance status. Most investigators recommend that future studies focus on the combination of irinotecan and cisplatin.

Laboratory data show potentially interesting interactions of irinotecan with radiotherapy and cytokines. Such findings need further refinement and the performance of correlative clinical studies.