Umeå University's logo

Cervical cancer is the second most common malignancy in women worldwide and one of the leading causes of cancer mortality globally. In patients with invasive cervical cancer prognostic factors are of value for the choice of treatment, monitoring of treatment and follow-up. The most important clinical prognostic factors are stage, tumor volume, parametrial infiltration, vascular invasion, lymph node metastases, and distant metastases. An improved estimation of the prognosis of cervical cancer is desirable, especially in early cancer stages.

The aim of this research was to study possible associations between tumor markers, female sex steroids, smoking, S-phase fraction (SPF), and prognosis in invasive squamous cell cervical cancer (SCC). The study comprised 190 patients with SCC, stages IB-IV, admitted to the Department of Gynecologic Oncology at Norrland University Hospital in Umeå between September 1984 and October1990. Ten year mortality was estimated.

In study I, of a total of 103 patients, it was found that increased tumor growth, measured by the DNA SPF, was associated with elevated serum progesterone and smoking in the premenopasual patients and with aneuploidy in the whole group.

In study II, comprising 128 patients, survival length related to hormone levels and SPF was evaluated in women who died of cervical cancer. In both pre- and postmenopausal women, who died of cervical cancer, SPF at or above 12% was correlated with reduced survival. There was significant positive correlation between a low serum estradiol/progesterone ratio and short survival in those premenopausal women who died of cancer (p=0.02).

In study III, ten-year follow-up results in 128 women were compared with the expression of ten relevant tumor markers, assessed by immunohistochemistry. The overall ten-year survival rate in patients with low COX-2 and high CD4+ expression was 76%, versus 53% in the remaining women. The survival rate with absent p53 and high COX-2 expression in the tumors was 42%, versus 71%, while the corresponding figure for the combination of high COX-2 intensity and expression of c-myc was 27%, versus 62%. None of the single markers correlated significantly with outcome in the final Cox regression analyses, while five combinations did.

Study IV addressed possible associations between selected tumor markers and cofactors in SCC. Ten tumor markers were examined in 128 patients. Smoking habits and previous oral contraceptive use were recorded. Serum estradiol and progesterone levels were evaluated in 80 women. Highly significant associations were found between strong c-myc staining and increased progesterone, low EGFR staining and high serum estradiol, and absence of p53 staining and smoking. There was an association between absence of p53 and high serum progesterone.

In study V, LRIG1 expression was studied in 128 patients and was compared with expression of nine other tumor markers, smoking history, hormone levels, and prognosis. LRIG1 appears to be a significant prognostic predictor in early stage SCC, independent of the other tumor markers that were studied.  Diminished expression in advanced cancer stages and the inverse correlation to serum progesterone and smoking indicate that LRIG1 is a tumor suppressor in squamous cell cervical cancer.

Conclusion: The results of these studies support a role of progesterone as a promoter of cervical cancer and indicate that smoking is associated with tumor progression. A combination of tumor markers might be of help in prognostic prediction. LRIG1 acts as a tumor suppressor. These findings might contribute towards greater understanding of prognostic prediction of squamous cell cervical cancer.