Posttherapy Surveillance of Women With Cervical Cancer: An Outcomes Analysis

To characterize and investigate the efficacy of posttherapy surveillance of patients with cervical cancer, the authors reviewed the medical records of all patients who had primary treatment for stage IB cervical cancer at M. D. Anderson Hospital between 1983 and 1993. Initial treatment, pathologic data, site and time of recurrence, and treatment for recurrence were all recorded. A total of 1093 women with stage IB cervical cancer were identified; 993 met the study criteria. Four hundred sixty-one women (46%) underwent primary surgical therapy. The cancer recurred in 40 (9%) of these patients, of whom 33 were symptomatic and 7 were asymptomatic. Among the 532 (54%) women treated initially with radiation, the cancer recurred in 93 (17%). Eighty-one were symptomatic and 12 were asymptomatic. In all, a recurrence developed in 133 (13%) patients, 114 with symptoms and 19 with no symptoms. Twenty-eight percent of recurrences were in the central pelvis, and 16% were in the lungs. These patients lived significantly longer (P < .001) than those whose recurrences were located elsewhere (pelvic wall, 16%; nodes, 16%, and multiple or other locations, 24%). The time to recurrence was similar among symptomatic and asymptomatic patients. When disease recurred in the central pelvis, patients with no symptoms whose recurrence was detected by follow-up pelvic examination had a median survival of 74 months, compared with 14 months among patients who presented with symptoms and were diagnosed by pelvic examination. The seven asymptomatic patients with recurrences in the pelvis were diagnosed by pelvic examination. Two of these woman died of disease; one patient refused treatment and died of unknown causes. Four patients, all treated with radiation therapy for recurrence, were alive without disease at 30, 125, 92, and 138 months after initial treatment. No asymptomatic pelvic recurrence was detected with the Pap smear, and only 2 of the 50 symptomatic patients with recurrence in the pelvis were diagnosed by Pap smear. Chest radiographs detected all asymptomatic recurrences in the lungs (N = 8) or chest nodes (N = 1). Five of the patients with pulmonary recurrences are alive, three with disease and two without. The two disease-free patients were treated with wedge resection and have survived 138 and 127 months. Four women died of disease. The median duration of survival from time of recurrence among patients with pulmonary recurrences was significantly longer for asymptomatic patients than for those with symptoms (3 vs. 1 year, respectively; P < .02). The disease-free interval before recurrence was similar for all patients. However, the median duration of survival from time of recurrence was significantly longer for patients without symptoms than for those with symptoms (42 vs. 11 months, P < .001). Median duration of survival from the time of original diagnosis was also longer for asymptomatic patients than for those with symptoms (83 vs. 31 months, P < .001). After multivariate analysis with other known prognostic factors, symptomatic status was still significant (P < .01). On the basis of these results, the authors proposed changes to the 5-year surveillance schedule for patients with Stage IB cervical cancer at their institution. They recommend a decrease in the number of pelvic and physical examinations performed each year to three for the first 2 years and two each year thereafter. They recommend only one Pap smear a year instead of multiple smears each year. They also recommend an increase in the use of chest radiography, suggesting that radiograph be obtained in the first and fifth years and that two be obtained in each of the middle years. A cost analysis comparing the new schedule with the one used during the study showed the overall costs of each program to be virtually the same.