Individual quality assessment of autografting by probability evaluation: a model estimated by analysis of graft-related end points in 204 patients with malignancies

Bone Marrow Transplant. 2003 Mar;31(6):453-8. doi: 10.1038/sj.bmt.1703828.

Abstract

Haematological toxicity is considered a secondary end point important for graft evaluation - in today's practice graft evaluation focuses on the primary impact of health economic end points. This report illustrates the benefit of combining CD34 enumeration and demographic as well as disease-related variables in models for individual quality assessment of autografting following high-dose therapy. A total of 24 centres in Scandinavia enrolled 204 patients younger than 67 years who received high-dose therapy with autologous peripheral blood stem cell transplantation. Using the binary Logistic Regression Analysis, the prognostic value of diagnostic demographic variables, therapy and graft-related factors was entered into a multivariate analysis and the final significant models were used to estimate probabilities for acceptable or unacceptable outcome among different patient scenarios. The model that estimated post-transplant efficacy by selected primary end points (time on antibiotics and use of transfusions) includes six independent variables related to sex, age, disease, conditioning, growth factor administration, and graft CD34+ cell number. The model that estimated transplantation-related toxicity by selected secondary end points (time to blood cell recovery) included four independent variables related to age, disease, growth factor administration and graft CD34+ cell number.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Antigens, CD34 / analysis
  • Breast Neoplasms / therapy
  • Female
  • Hematopoietic Stem Cell Transplantation / standards*
  • Hematopoietic Stem Cell Transplantation / statistics & numerical data*
  • Hodgkin Disease / therapy
  • Humans
  • Leukocytes / chemistry
  • Leukocytes / cytology
  • Logistic Models
  • Lymphoma, Non-Hodgkin / therapy
  • Male
  • Middle Aged
  • Models, Statistical*
  • Multiple Myeloma / therapy
  • Neoplasms / therapy*
  • Probability
  • Prognosis
  • Quality Assurance, Health Care / methods*
  • Testicular Neoplasms / therapy
  • Transplantation, Autologous

Substances

  • Antigens, CD34