The development of a prediction tool for chemotherapy-induced anemia in patients with advanced non-small cell lung cancer (NSCLC) palliative receiving chemotherapy

8600 Background: Anemia remains a common complication in patients with advanced NSCLC on palliative chemotherapy. Many oncologists believe that anemia is unpredictable, acting on a low hemoglobin value once the patient is anemic (i.e. reactively). However, patient care could be substantially improved if severe anemia could be accurately predicted by validated mathematical models. The development and validation of our cycle-based prediction model for severe anemia (hemoglobin ≤ 100 g/L) is described. Methods: Data on 536 European patients receiving palliative chemotherapy was prospectively collected as part of the European Cancer Anemia Survey (Ludwig et al, 2004), including information on clinical and biochemistry parameters that could potentially be associated with anemia. The sample was randomly divided into a 2/3 model derivation and 1/3 validation sample. Multivariable logistic regression techniques were applied for the initial model. A risk scoring system based on regression parameters was created, f...