Predicting the effectiveness of hydroxyurea in individual sickle cell anemia patients

The study described in this paper was undertaken to develop the ability to predict the response of sickle-cell patients to hydroxyurea (HU) therapy. W e analyzed the effect of HU on the values of 23 parameters of 83 patients. A Student's t-test was used to confirm (Rodgers GP, Dover GJ, Noguchi CT, S chechter AN, Nienhuis AW. Hematologic responses of patients with sickle cel l disease to treatment with hydroxyurea, N Engl J Med 1990;322;1037-44) at the 0.001 level that treatment with HU increases the proportion of fetal he moglobin (HbF), and the average corpuscular volume (MCV) of the red blood c ells. Correlation analysis failed to establish a statistically significant relationship between any of the 23 parameters and the HbF response. Linear regression analysis also failed to predict a patient's response to HU. On t he other hand, artificial neural network (ANN) pattern-recognition analysis of the 23 parameters predicts, with 86.6% accuracy, those patients that re spond positively to HU and those that do not. Furthermore, we have found th at the values of only 10 of the 23 parameters (listed in the body of this p aper) are sufficient to train ANNs to predict which patients will respond t o HU. (C) 2000 Elsevier Science B.V. All rights reserved.