STUDY OF PROGNOSIS IN ACUTE MYELOID LEUKEMIAS (AML) BY CLUSTER-ANALYSIS

Background. Cluster analysis is particularly effective in detecting ho mogeneous subgroups among large series of observations. We applied thi s relatively uncommon approach to the study of prognosis in 137 patien ts affected by acute myeloid leukemia (AML). Methods and Results. Empl oying simple presentation parameters (age, WBC, splenomegaly, hepatome galy) we used cluster analysis to define 3 groups with different overa ll survival (p=0.0019). This classification was obtained following a r escaling of the variables and principal component analysis. Validation was performed through random definition of a control group. With the same variables, univariate analysis demonstrated age was the only prog nostic factor, while Cox's model was not significant. Conclusions. In our series cluster analysis allowed a better definition of prognosis t han Cox's analysis. Since the 3 groups are well identifiable, each pat ient can be rapidly classified and his allocation confirmed by discrim inant functions. For cluster 2 we were able to project a possible myel odysplastic evolution, while cluster 3 was more frequently associated with a monocytic blastic component. We think that cluster analysis des erves consideration as an alternative statistical approach in the anal ysis of large series of data; its usefulness lies in its power to defi ne homogeneous prognostic or biologic subgroups and to elaborate farth er hypotheses for new studies.