Can manipulation of a pelvic tumour influence the serum level of cancer antigen 125 and cancer-associated serum antigen?

Cancer antigen 125 (CA 125) and cancer-associated serum antigen (CASA) were measured in 24 women with pelvic masses before and after a gynaecological examination and ultrasonography. CA 125 decreased median 16% after manipulation (P < 0.0001) and CASA decreased median 8% (P = 0.0077). The decline was found in patients with benign tumours as well as in patients with malignant tumours.

Tumour markers are widely used in the diagnosis, monitoring and follow-up of ovarian cancer and serum is often drawn in relation to a gynaecological examination and ultrasonography. In order to evaluate the influence of these procedures on the tumour marker level we measured cancer antigen 125 (CA 125) and cancerassociated serum antigen (CASA) in patients with benign and malignant pelvic masses before and after a gynaecological examination combined with ultrasonography.

MATERIAL AND METHODS
We investigated 24 women with pelvic masses who underwent laparotomy at the Department of Obstetrics and Gynaecology, Odense University Hospital. The first blood sample was drawn 6-8 h before a preoperative gynaecological examination and abdominal and transvaginal sonography. The second blood sample was drawn 16-18 hours after the examinations and before the operation. Serum was stored at -80°C within I h after sampling. CA 125 and CASA were measured in duplicate by enzyme immunoassays as described previously (Kierkegaard et al, 1995). The specimens from each patient were measured simultaneously with the same kit. The detection limit of CA 125 was 0.5 U ml-' and the overall imprecision of the assay was 5.0-8.2 U ml-' (Mogensen and Mogensen, 1995). The results were statistically evaluated by Wilcoxon's matched-pairs test. When the CASA level was outside the detection limit of 2.0-64.0 U ml the values were considered unchanged.
Only nine patients had CASA levels within the detection limit and the median decrease was 8% (range 4-50%, P = 0.0077).
No patients classified as true positive turned out to be false negative after manipulation using the most common thresholds of CA 125 (35, 20 or 15 U ml-') and CASA (8, 6 or 4 U ml-'). With a CA 125 threshold of 20 or 15 U ml-' the patient with a fibroma changed from false positive to true negative (Table 1). Similarly, a patient with a simple cyst changed from false positive to true negative using the CASA threshold of 4 U ml'.

DISCUSSION
In the present paper we have evaluated the effect of a gynaecological examination and ultrasonography on serum levels of CA 125 and CASA. An increased marker leveltrue or falsemay involve extensive investigations and certainly worries the patient. On the other hand, a false-negative result may delay treatment and may result in decreased quality of life and survival. Therefore, it is important to look for conditions other than the malignant disease that may influence the tumour marker level in serum.
Abdominal surgery may cause an increase in CA 125 irrespective of whether the diagnosis of the patient was ovarian cancer, cervical cancer or atherosclerotic disease with aortic surgery; CA 125 reaches its maximum during the second postoperative week (Van der Zee et Mogensen et al, 1993). This non-specific increase may be explained by a traumatically induced inflammation of the peritoneum. Healthy women may have moderately increased CA 125 values during menstruation (Lehtovirta et al, 1990), which may be due to antigen leakage from endometrial glands into the circulation. Increased marker levels have also been found during the first trimester of pregnancy (Kobayashi et al, 1989) and this finding may be caused by CA 125 leakage from the decidual cells and the amniotic fluid into the circulation.
Tumour markers and manipulation of pelvic tumour 2013 From a theoretical point of view we expected a release of CA 125 and CASA from the tumours as a result of the manipulation. However, we observed a small but statistically significant decrease in CA 125 and CASA levels after manipulation of the pelvic masses.
The results are not likely to be explained by analytical procedures or within-day variations. The serum samples were analysed blindly in pairs and the two samples were drawn at an interval of approximately 24 h. The patients fasted and had had no intravenous fluids before the second sample was drawn, therefore haemodilution did not occur.
In conclusion, the present study demonstrated a decrease in the serum levels of CA 125 and CASA after a gynaecological examination combined with sonography. So far, this decline cannot be explained, but the time of sampling should be taken into consideration to avoid misclassification of the patient.