These tools will no longer be maintained as of December 31, 2024. Archived website can be found here. PubMed4Hh GitHub repository can be found here. Contact NLM Customer Service if you have questions.


Pubmed for Handhelds

PUBMED FOR HANDHELDS

Search MEDLINE/PubMed

  • Title: Impact of clinical and morphological variables in classification and regression tree-based survival (CART) analysis of CML with special emphasis on dynamic features.
    Author: Thiele J, Kvasnicka HM, Zirbes TK, Flucke U, Niederle N, Leder LD, Diehl V, Fischer R.
    Journal: Eur J Haematol; 1998 Jan; 60(1):35-46. PubMed ID: 9451426.
    Abstract:
    To determine parameters of predictive value in CML, a retrospective clinico-pathological study was performed. This included laboratory data and (pretreatment) bone marrow biopsies of 120 patients with a monotherapy by busulfan (BU) and 50 patients with interferon-alpha 2b (IFN) treatment. Median survival in the BU group was 39 months and in the IFN-treated patients 65 months. Morphological features (CD61-positive megakaryocytes, argyrophilic fibres, pseudo-Gaucher cells) were evaluated by morphometry. Additionally, we measured the incidence of apoptosis (in situ end-labelling technique) and the expression of the proliferating cell nuclear antigen (PCNA). The ratio between the proliferative and apoptotic cell fraction was coined leukaemia turnover index (LTI). In order to estimate the impact of clinical and various morphological as well as dynamic features of prognostic significance, a multivariate analysis was carried out using the classification and regression tree approach (CART). Discrimination of single disease parameters revealed that fibrosis remained the most significant variable for survival in both therapeutic groups. Indicators of myeloid metaplasia such as occurrence of erythro-normoblasts and/or splenomegaly were important clinical parameters for prognosis. Inclusion of morphological as well as dynamic disease features in risk classification resulted in a substantial improvement of prognostic efficiency compared to other predictive scores which could be demonstrated by means of ROC-analysis.
    [Abstract] [Full Text] [Related] [New Search]