Alternative discriminating function for determining hormone dependency of breast cancer.

Abstract

It is common practice to consider performing hypophysectomy or adrenalectomy combined with oophorectomy on women suffering from metastatic breast cancer. Satisfactory, albeit temporary, improvement can be expected in only about one half of those operated on, so that it would be advantageous if it were possible to predict whether a useful response could be anticipated. Accurate prediction would prevent patients being subjected to a major operation from which no benefit could be expected. By studying the urinary steroid metabolites of a number of these patients before one or other of these operations were per formed, and relating the results obtained to the clinical response after operation, Bulbrook, Greenwood, and Hayward (1960) were able to derive a function which discriminated between the responsive and non-responsive groups. This discriminant function was calculated as 80-80 (17-OHCS (mg./24 hours)) + aetiocholanolone (/xg./24 hours). 17-OHCS (17-hydroxy corticosteroids or total 17-oxogenic steroids) were measured by the method of Appleby, Gibson, Norymberski, and Stubbs (1955), and aetiocholanolone (3a-hydroxy-5/?-androstan-17-one) by the method of Kellie and Wade (1957). A positive numerical result was obtained in the responsive patients and a negative result in the non-responsive group. When this function was determined in a series of normal women and women suffering from non-metastatic breast cancer it was shown that, whereas it was unusual to find a negative discriminant In normal women under 65, over half of the early breast cancer cases gave negative discriminants (Bul brook, Hayward, Spicer, and Thomas, 1962b). The British Medical Journal (1965) recognized the possible clinical significance of this discriminant function even though its physiological meaning was unknown. There has been little published by other laboratories in evaluation of this work, and this apparent lack of interest may be due to the technical difficulties of estimating urinary aetio cholanolone and to the amount of working time needed to do so.