Estimating risk of ovarian cancer

We have our research paper just been accepted for publication by the Archieve of Obstet and Gynecol (will be on line and printed version in a few months).   Our study included 324 patients undergoing adnexal/ ovarian mass surgery were recruited into the study. All study patients had a preoperative CT scan and serum CA-125 test. CT scan abnormalities included any solid tumor components, ascites, and pelvic or abdominal lymphadenopathy and omental caking.   There were 225 (70%) benign and 99 (30%]) malignant ovarian masses. Using logistic regression with the area under the curve of the receiver operating curve of 82%, the cancer probability was determined by the equation

e^{-3.6372 + 0.0306 * (A) + 0.001 * (B) + 0. 876 * (C)+1.551 * (D) + 1.7377 * (E) + 2.76 * (F)}

 1+e ^{-3.6372 + 0.0306 * (A) + 0.001 * (B) + 0.876 * (C) + 1.551 * (D) + 1.7377 * (E) + 2.76 * (F)}

where A = age, B = CA-125, C = solid adnexal mass is 1 and cystic is 0, D = ascites is 1, E = omental caking is 1 and absence is 0,; F = node size > 1 cm is 1 and < 1cm no is 0 value. The natural logarithm e is a constant [2.718281828]. For example, for a woman of age 60, CA-125 = 50 U/ml, with solid adnexal mass, ascites, omental caking, and lymphadenopathy, the probability is 0.994. Hence, this woman has a 99.4%  probability of having cancer.

I wish to caution that our study is still preliminary.  This paper showed a mathematical formula could be used in combination with CT scan finding to estimate risk for ovarian cancer.

Reference:

Santoso JT, et al.  CT adnexall mass score to estimate ovarian cancer.  Arch Gynecol Obstet. 2014