Correlations between leukocyte count and absolute granulocyte count in patients receiving cancer chemotherapy

The absolute granulocyte count (AGC) has been considered the best index for estimating the risk of infection in patients receiving myelosuppressive therapy. However, many investigators and cooperative oncology groups use the leukocyte count and extrapolate concurrent AGC values from an arbitrary conversion scale. Our review of the literature revealed no analysis of the relationship between the leukocyte count and the AGC in patients receiving cancer chemotherapy. It also failed to provide a method for converting toxicity criteria from one scale to the other. To explore the possible relationship of the leukocyte count to the AGC, we have completed a retrospective analysis of leukocyte count and accompanying AGC in patients receiving cancer chemotherapy. The leukocyte count and the AGC are shown to be linearly related over the entire population, enabling predictable cross‐indexing from leukocyte count to AGC by the use of the formula: AGC = −0.7 + 0.8 (leukocyte count). This provides a rational basis for the development of guidelines for drug dosing and toxicity. In the patient group with leukocyte count ⩽ 4500, however, the magnitude of random variability decreased predictive ability. Numerous patients in this group received differing toxicity scale scores when classified according to the Eastern Cooperative Oncology Group (ECOG) scales for AGC and leukocyte count. In some cases, as much as 46% disagreement occurred. New toxicity scales for AGC and leukocyte count, which were developed based upon the linear relationship above, reduced this disagreement substantially. These scales result in a greater agreement of toxicity ratings, and may provide a more accurate method of classifying toxicity and regulating dosages of chemotherapeutic agents.