TY - JOUR
T1 - Prognostic nomogram and index for overall survival in previously untreated patients with chronic lymphocytic leukemia
AU - Wierda, William G.
AU - O'Brien, Susan
AU - Wang, Xuemei
AU - Faderl, Stefan
AU - Ferrajoli, Alessandra
AU - Do, Kim Anh
AU - Cortes, Jorge
AU - Thomas, Deborah
AU - Garcia-Manero, Guillermo
AU - Koller, Charles
AU - Beran, Miloslav
AU - Giles, Francis
AU - Ravandi, Farhad
AU - Lerner, Susan
AU - Kantarjian, Hagop
AU - Keating, Michael
PY - 2007/6/1
Y1 - 2007/6/1
N2 - The clinical course for patients with chronic lymphocytic leukemia is extremely heterogeneous. The Rai and Binet staging systems have been used to risk-stratify patients; most patients present with early-stage disease. We evaluated a group of previously untreated patients with chronic lymphocytic leukemia (CLL) at initial presentation to University of Texas M. D. Anderson Cancer Center to identify independent characteristics that predict for overall survival. Clinical and routine laboratory characteristics for 1674 previously untreated patients who presented for evaluation of CLL from 1981 to 2004 were included. Univariate and multivariate analyses identified several patient characteristics at presentation that predicted for overall survival in previously untreated patients with CLL. A multivariate Cox proportional hazards model was developed, including the following independent characteristics: age, β-2 microglobulin, absolute lymphocyte count, sex, Rai stage, and number of involved lymph node groups. Inclusion of patients from a single institution and the proportion of patients younger than 65 years may limit this model. A weighted prognostic model, or nomogram, predictive for overall survival was constructed using these 6 characteristics for 5- and 10-year survival probability and estimated median survival time. This prognostic model may help patients and clinicians in clinical decision making as well as in clinical research and clinical trial design.
AB - The clinical course for patients with chronic lymphocytic leukemia is extremely heterogeneous. The Rai and Binet staging systems have been used to risk-stratify patients; most patients present with early-stage disease. We evaluated a group of previously untreated patients with chronic lymphocytic leukemia (CLL) at initial presentation to University of Texas M. D. Anderson Cancer Center to identify independent characteristics that predict for overall survival. Clinical and routine laboratory characteristics for 1674 previously untreated patients who presented for evaluation of CLL from 1981 to 2004 were included. Univariate and multivariate analyses identified several patient characteristics at presentation that predicted for overall survival in previously untreated patients with CLL. A multivariate Cox proportional hazards model was developed, including the following independent characteristics: age, β-2 microglobulin, absolute lymphocyte count, sex, Rai stage, and number of involved lymph node groups. Inclusion of patients from a single institution and the proportion of patients younger than 65 years may limit this model. A weighted prognostic model, or nomogram, predictive for overall survival was constructed using these 6 characteristics for 5- and 10-year survival probability and estimated median survival time. This prognostic model may help patients and clinicians in clinical decision making as well as in clinical research and clinical trial design.
UR - http://www.scopus.com/inward/record.url?scp=34249733155&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34249733155&partnerID=8YFLogxK
U2 - 10.1182/blood-2005-12-051458
DO - 10.1182/blood-2005-12-051458
M3 - Article
C2 - 17299097
AN - SCOPUS:34249733155
SN - 0006-4971
VL - 109
SP - 4679
EP - 4685
JO - Blood
JF - Blood
IS - 11
ER -