Adenocarcinoma of the uterine cervix prognosis and patterns of failure in 367 cases

Abstract

Between 1965 and 1985, 367 patients received initial treatment for adenocarcinoma of the uterine cervix at the M. D. Anderson Cancer Center (MDACC). Of the 334 patients treated with curative intent, 223 had International Federation of Gynecology and Obstetrics (FIGO) Stage I, 60 had Stage II, and 51 had Stage III/IV disease. The 5-year and 10-year relapse-free survival (RFS) rates for all patients treated for Stage I disease were 73% and 70%, respectively. RFS was strongly correlated with initial bulk of disease (P = 0.002), although locoregional control (LRC) was good in all groups: 91 patients with a normal-sized cervix (tumor < 3 cm) had a 5-year RFS rate of 88% and an actuarial LRC rate of 94%; 102 patients with lesions 3 to 5.9 cm in diameter had an RFS rate of 64% and an LRC rate of 82%; and 22 patients with bulky lesions greater than 6 cm in diameter had a comparable LRC rate of 81%, but an RFS rate of only 45%. Decreased RFS also was strongly correlated with positive lymphangiogram (LAG) results (P = 0.02) and poorly differentiated lesions (P = 0.0014). When initial primary tumor size was taken into account, there was no significant difference in RFS or LRC between patients treated with radiation (RT) alone or RT plus extrafascial hysterectomy (R + S). The 5-year and 10-year RFS rates of 60 patients who received curative therapy for Stage II disease were 32% and 25%, respectively, with an LRC rate of 62% at 5 years. Patients with bulky Stage II disease did particularly poorly, with a 5-year RFS rate of 15%. Decreased RFS was correlated with positive LAG results and poorly differentiated tumors. Most Stage II patients whose disease relapsed died with distant metastases (73%). Forty-eight patients with Stage III/IV disease treated with curative intent had a 5-year survival rate of 31% and a 5-year pelvic disease control rate of 52%. In summary, patients with small volume Stage IB lesions have excellent LRC and survival with RT alone. RT achieves good LRC of bulkier Stage I lesions, but survival decreases with increasing primary tumor size. R + S holds no apparent advantage over RT alone. Patients with more advanced disease have a high rate of relapse with frequent distant metastasis. In particular, the survival of patients with FIGO Stage II disease is much lower than what we have observed after treatment of comparable stage squamous carcinoma. However, the late recurrences and prolonged natural history experienced by a few patients suggest a broad spectrum of biologic aggressiveness among cervical adenocarcinomas.