Aberrant expression of cyclin D1 is associated with poor prognosis in early stage cervical cancer of the uterus.

OBJECTIVE: Many investigators have studied the expression of G1 phase regulatory protein in uterine cervical cancer. However, it is unclear which step of the genetic expression participates in cyclin D1 expression and what its prognostic meaning is. The aims of this study were to evaluate the regulatory level of cyclin D1 expression and the relationship between the expression of cyclin D1 and its inhibitor, p21WAF1/CIP1, and to evaluate their impact on the prognosis of early stage cervical cancer.

METHODS: The presence of cyclin D1 mRNA was studied using Northern blot in 6 normal cervices and 7 invasive cervical cancer specimens. Western blot was used to detect the cyclin D1 protein in 8 normal cervices and 8 invasive cancer specimens. Thirty-two cases of FIGO stage Ib-IIa cervical cancers (28 squamous cell carcinomas, 3 adenocarcinomas, 1 adenosquamous cell carcinoma), 31 cases of cervical intraepithelial neoplasia 3 (CIN 3), and 28 normal cervices were stained for cyclin D1 and p21(WAF1/CIP1) using monoclonal antibody. Statistical analysis was performed to assess the differences in expression and their prognostic significance. RESULTS. Cyclin D1 mRNA was found to be underexpressed in cervical cancer. Western blot also revealed underexpression of cyclin D1 protein in cervical cancer compared to normal controls. Positive immunohistochemical staining of cyclin D1 was noted in 28/28 (100%) of the normal controls, 1/31 (3%) cases of CIN 3, and 9/32 (28%) cases of invasive cancer. The number of positively stained specimens was lower than that of normal controls in CIN 3 and cervical cancer specimens (P = 0.005). Fifteen of 28 (54%) normal controls, 15/31 cases (48%) of CIN 3, and 27/32 cases (84%) of invasive cancer were proved positive for p21WAF1/CIP1 immunohistochemistry. p21WAF1/CIP1 was more highly expressed in cervical cancer than in that of either normal controls or CIN specimens (P = 0.001). Positive immunostaining of cyclin D1 and p21WAF1/CIP1 was not related to high-risk factors (pelvic lymph node metastasis, deep cervical stromal invasion, parametrial invasion, large tumor size, and unusual histologic type) and human papilloma virus infection. Positive cyclin D1 immunostaining was associated with decreased disease-free survival and a lower overall survival (P = 0.0175 and 0.0189, respectively). On multivariate analysis, positive cyclin D1 expression was a significant prognostic variable for recurrence (P = 0.0004).

CONCLUSION: Underexpression of cyclin D1 was regulated at the level of transcription in cervical cancer. Although cyclin D1 was underexpressed in cervical neoplasias, it was more frequently expressed in malignant lesions. p21WAF1/CIP1 was more highly expressed in cervical cancers than in either normal cervices or CIN 3 specimens. Unfavorable prognoses were associated with cyclin D1 expression, and not with the expression of its inhibitor, p21WAF1/CIP1.We conclude that immunohistochemical assessment of cyclin D1 can be a useful molecular marker for predicting prognosis in early stage cervical cancer of the uterus.