Survival and Prognostic Factors in Patients with Carcinoma of Cervical Stump

Purpose: To evaluate patients with carcinoma of cervical stump (CCS) and analyse different clinico-pathologic factors affect prognosis. Patients and Methods: This study was carried out through review of clinical records of patients. Recorded data included information on age, tumor stage, presenting symptoms, size of tumor, histopathology, grade, type, cause of subtotal hysterectomy (STH), treatment and follow-up results. Staging according to International Federation of Gynecology and Obstetrics (FIGO) staging system was done through: PHYSICAL examination, pelvic examination under anaesthesia, chest X-ray, magnetic resonance imaging (MRI) of the abdomen and pelvis, cystoscopy, rectosigmoidoscopy and intravenous pyelography. Prognostic factors as age, size of tumor, stage, lymph node (LN) involvement, pathological type, grade and type of CCS either true or coincidental were analysed through multivariate analysis. Results: 62% of patients are above 50 years with stage II in 48.7%. Squamous cell carcinoma was more common but 54% are of GIII. 89% were true CCS. Positive lymph nodes were reported in 27%. The predominant reason for STH was abnormal bleeding (73%). In about 95% of cases, women seeked medical attention because of symptoms and the most common presenting symptom was bleeding (54%). According to the stage and performance status of patients, treatment consisted of radiotherapy either external or interstitial, chemotherapy and chemoradiotherapy. Through multivariate analysis, the following was found to have adverse impact on survival: Coincidental type (P = 0.04), high grade (P = 0.03), advanced stage (P = 0.01), larger tumor size (P = 0.02), lymph node involvement (P = 0.029) and older age (P = 0.035). While pathological type was not (P = 0.52). After median follow-up of 52 months; 5-year overall survival was 65%. Conclusion: CCS has a low morbidity. Adverse survival outcomes can be anticipated in those patients with: high grade lesions, advanced stages, large tumor size, coincidental type, older age and positive lymph node involvement.


Introduction
Sparing the cervix at time of hysterectomy which called subtotal hysterectomy (STH) was carried out in the past in order to reduce surgical complications as blood loss, vaginal vault prolapse, enterocele, ureteral injuries and vaginal cuff abscess [1].However, three randomized controlled trials comparing perioperative or post-operative complications of total hysterectomy versus STH, concluded that there is no clinically significant difference between blood loss and surgical time in both approaches despite that STH may be associated with less blood loss and shorter surgical time [2]- [4].Also, it was suggested that the retaining cervix reduce the adverse effect on psychosexual behavior of the patients.Roovers and colleagues [5] found that sexual activity did not change after surgery whether it was total or subtotal hysterectomy, this finding was also reported by El-Touky et al. [6].But, the retaining cervix carries the risk of developing cervical stump symptoms as vaginal bleeding, pelvic pain and cervical malignancy.Carcinoma of cervical stump (CCS) has been divided into: 1) Coincidental cases, when detected within 2 years after hysterectomy suggesting presence of pre-existing disease that not recognized at time of surgery; 2) True cases, when detected later than 2 years and considered to arise de novo [1].The aim of this study is to retrospectively review patients with CCS as regard age, clinical stage, histopathology, treatment and survival, also to determine clinico-pathologic factors which are prognostic for patients with CCS.

Patients and Methods
A retrospective review of the records of patients with CCS attended to Clinical Oncology and Nuclear Medicine Department, Mansoura University at the period from January 2000 to December 2013.Recorded data included information on age, tumor stage, presenting symptoms, size of tumor, histopathology, type, cause of STH, treatment and follow-up results, the institutional review board approved the study.Staging according to FIGO staging system [7] was carried out through: Physical examination, pelvic examination under anaesthesia, chest xray, magnetic resonance imaging (MRI) of the abdomen and pelvis, cystoscopy, rectosigmoidoscopy and intravenous pyelography.After treatment, all patients were followed-up every 3 months for the first year then every 4 -6 months for 3 years then annually.At each visit, full physical examination was performed.MRI of abdomen and pelvis was done every six months.
Prognostic factors as age, size of tumor, stage, lymph node (LN) involvement, pathological type, grade and type of CCS either true or coincidental were analysed through multivariate analysis.

Statistical Methods
Statistical Package for Social Sciences (SPSS) version 15.0 (Chicago, IL, USA) was used for statistical analysis, Number and percentage for demographic data.The categorical variables were compared using chi-square and Fisher's Exact test.P-value of ˂0.05 indicate statistical significance.Kaplan-Meier test was used for survival function.

Results
This study included 37 patients Table 1 showed their demographic data; mean age was 54 years, 62% of patients are above 50 years with stage II in 48.7%.Squamous cell carcinoma was more common but 54% are of GIII.89% were true CCS.Positive lymph nodes were reported in 27%.
The predominant reason for STH was abnormal bleeding (73%).In about 95% of cases, women seeked medical attention because of symptoms and the most common presenting symptom was bleeding (54%).
According to the stage and performance status of patients, treatment consisted of radiotherapy either external or interstitial, chemotherapy and chemoradiotherapy.Seventeen cases were treated by surgery followed by radiotherapy in the form of external pelvic irradiation at dose of 45 -50 Gy followed by brachytherapy.Five cases received neoadjuvant radiotherapy followed by surgery, 11 patients were treated by chemoradiotherapy and  4 received chemotherapy alone.Through multivariate analysis; the following was found to have adverse impact on survival: Coincidental type (P = 0.04), high grade (P = 0.03), advanced stage (P = 0.01), larger tumor size (P = 0.02), lymph node involvement (P = 0.029) and older age (P = 0.035).While pathological type was not (P = 0.52).

Discussion
The main disadvantage of STH is the risk of developing CCS and the requirement for regular cervical screening following surgery [8].So STH should be avoided whenever possible.CCS is found in 1% -3% of patients with STH and accounts for 3% -9% of all cervical cancer [9] [10].In our series; mean age was 54 years that was less than found by Hellström A.C., et al. [11].Igboeli P., et al. [12] reported higher incidence of true CCS similar to that found in our patients.The majority of our cases were diagnosed as stage II (48.7%) and most of them with SCC (92%), similar finding were reported by Hannoun-Levi J.M., et al. [13] and Petersen L.K. et al. [14].
Only 5% of cases were incidentally discovered.This can be explained by lack of screening after STH or some patients may not have received adequate information about the extent of the hysterectomy and therefore may have erroneously concluded that they no longer need cancer screening.Patients with coincidental CCS does worse than true one that was also found by Gibbons SK, Keys HM [15].This can be explained by presence of disease which not recognized at time of STH.
It was found that AC cases had worse prognosis compares with the squamous cases [16].But there was no significant impact of pathological type on survival in our patients; this result is not conclusive because of small number of AC patients (8%).
Our results support the suggestion that high-grade, large tumor size (4 cm) and positive lymph node involvement adversely affect survival [17]- [19].
Calais G., et al. [20] mentioned that stage and nodal status had impact on prognosis; similar to our finding.Most authors have reported that experienced institutions can achieve survival results in patients with CCS comparable to those that are achieved in patients with carcinoma of intact cervix [13] [16] [21].However, Igboeli P., et al. [12] has reported superior results with CCS cases when compared to those with an intact uterine cervix.In our patients, 5-year overall survival rate was comparable to that reported by Calais G., et al. [20] and Chen L., et al. [22].

Conclusion
CCS has a low morbidity.Adverse survival outcomes can be anticipated in those patients with: high grade lesions, advanced stages, large tumor size, coincidental type, older age and positive lymph node involvement.