Plasma kisspeptin: a potential biomarker of tumor metastasis in patients with ovarian carcinoma.

More than 15 000 women in the US die from ovarian carcinoma annually (1 ). The most important determinant of survival in ovarian carcinoma cases is tumor stage. Most patients presenting with disease confined to the ovaries (stage 1) can be cured, with a 5-year survival rate of 90% (1 ). Spread of the tumor outside the ovaries (stages 2 to 4) confers a much poorer prognosis, however. Cure is uncommon in patients with tumor spread into the pelvis (stage 2), the abdomen (stage 3), or the liver/ extra-abdominal region (stage 4), with a 5-year survival rate as low as 22% (1 ). It is therefore critically important to develop novel biochemical markers of ovarian carcinoma to stratify patients according to prognosis. Serum cancer antigen 125 (CA125) is a useful marker for screening and monitoring disease response but cannot be reliably used for staging, because 20% of tumors do not secrete CA125 (2 ). There is currently no diagnostic marker of tumor spread in patients with diagnosed ovarian carcinoma. Kisspeptin, a peptide that activates the kisspeptin receptor, is encoded by the KISS1 (KiSS-1 metastasis-suppressor) gene. KISS1 expression negatively correlates with metastasis risk in lung, pancreatic, esophago-gastrointestinal, endometrial, renal cell, and bladder carcinomas (3). In ovarian carcinoma, low KISS1 expression is associated with aggressive disease and a poor prognosis, and KISS1 overexpression inhibits cell migration and reduces metastasis (4). Kisspeptin is therefore a putative metastasis suppressor in ovarian carcinoma. We compared plasma kisspeptin concentrations in patients with stage 1 ovarian carcinoma and patients with disease in stages 2 to 4 to determine its utility as a novel biomarker of tumor metastasis in ovarian carcinoma. After obtaining ethics approval (reference: 04/Q0406/80) and informed consent, we studied 31 patients with ovarian carcinoma and 31 healthy volunteers without known cancer. Blood was collected into a tube coated with lithium heparin and containing 5000 kallikrein inhibitor units (0.2 mL) of aprotinin and centrifuged for 4 min at 1538g. The separated plasma was stored at 20 °C. Plasma kisspeptin immunoreactivity was measured with a previously described RIA (5 ). The limit of detection was 2 pmol/L, and the intraand interassay CVs at 20 pmol/L were 8.3% and 10%, respectively. CA125 was measured with a 2-step ARCHITECT immunoassay (Abbott Diagnostics), with a 1-kU/L analytical sensitivity and intraand interassay CVs between 40 – 170 kU/L of 5%. Data are expressed as the mean (SD) unless stated otherwise. Multiple means were compared with one-way ANOVA and the Tukey multiplecomparison test. Nine patients had stage 1 ovarian carcinoma, with a median plasma kisspeptin concentration of 17.4 pmol/L (range, 10.1–54.2 pmol/L; interquartile range, 12.9 – 36.1 pmol/L). Twenty-three patients had ovarian carcinoma of stages 2 to 4, with a median plasma kisspeptin concentration of 7.8 pmol/L (range, 2– 41.2 pmol/L; interquartile range, 5.7–17.0 pmol/L. The median plasma kisspeptin concentration for the controls was 11.4 pmol/L (range, 2– 29.9; interquartile range, 8.4 –17.9 pmol/L). The plasma kisspeptin concentration appeared dependent on ovarian carcinoma stage (Fig. 1). The mean kisspeptin concentration in the patients with stage 1 ovarian carcinoma was increased significantly, compared with the patients with ovarian carcinoma of stages 2 to 4 and with the controls [25.1 (15.2) pmol/L (stage 1), 11.8 (10.3) pmol/L (stages 2 to 4), and 13.1 (6.92) pmol/L (controls); P 0.05, stage 1 vs stages 2 to 4; P 0.01, stage 1 vs controls]. A plasma kisspeptin concentration 20 pmol/L had a 91.3% sensitivity (95% CI, 72%–99%) and a 44.4% specificity (95% CI, 14%–79%) with respect to patients with ovarian carcinoma in stages 2 to 4. A plasma kisspeptin concentration 10 pmol/L has a lower sensitivity (65.2%; 95% CI, 39%– 80%) but a 100% specificity (95% CI, 66%– 100%) with respect to patients with ovarian carcinoma in stages 2 to 4. By comparison, a serum CA125 concentration 1000 kU/mL had a 26.1% sensitivity (identified 6 of 23 patients with disease in stages 2 to 4) and a 100% specificity (excluded 9 of 9 patients with stage 1 disease) with respect to ovarian carcinoma in stages 2 to 4 (Fig. 1). A ROC curve analysis of plasma kisspeptin in patients with disease in stages 2 to 4 vs patients with stage 1 ovarian carcinoma yielded an area under the curve of 0.80 (95% CI, 0.65– 0.96; P 0.01). Multiple lines of evidence implicate kisspeptin as a metastasis inhibitor in ovarian tumors, but plasma kisspeptin concentrations in patients with ovarian carcinoma have not been reported previously. This small study suggests that patients with stage 1 ovarian carcinoma have significantly increased concentrations of plasma kisspeptin compared with patients with ovarian carcinoma in stages 2 to 4 or compared with healthy controls. Furthermore, plasma kisspeptin may be more sensitive for detecting disease in stages 2 to 4 Clinical Chemistry 58:6 1061–1066 (2012) Letters to the Editor