50 %
50 %
Information about AcuteLymphoblasticLeukemiaALL

Published on January 12, 2009

Author: aSGuest10171


Acute Lymphoblastic Leukemia (ALL) : 1 Acute Lymphoblastic Leukemia (ALL) Presented by: Payam Refaeilzadeh What is ALL? : 2 What is ALL? Most commonly diagnosed cancer in children. Constitutes nearly 1/3 of all pediatric cancers. In ALL Lymphoid ancestor cells become genetically altered and undergo abnormal reproduction. ALL cells exhibit altered expression levels for genes associated with the development of B-Cells and T-Cells B-Cells: play a role in the immune system by secreting antibodies T-Cells: play a role in the immune system by directly attacking pathogens Study 1: Predicting PK parameters from limited samples : 3 Study 1: Predicting PK parameters from limited samples Etoposide is a drug used in ALL treatment Pharmacokinetic parameters for this drug vary across patients Problem Statement: How can we best predict PK parameters such as clearance using only a limited number of concentration samples as input? Two Compartment PK Model : 4 Two Compartment PK Model Catechol is metabolite product of etoposide A two compartment first-order PK model is assumed: dx/dt = -kx Experimental Method : 5 Experimental Method Clinical data contains measured values for the PK parameters: Vc, Kvp-cat, Kcp, etc Assume multivariate Gaussian distribution for PK parameters Generate 1000 synthetic time-course datasets of concentration values using Monte Carlo methods. Use multiple linear regression and Bayesian estimation to estimate Clearance using only two time-points and compare the estimate to the actual clearance (computed from all time points) Multiple Linear Regression : 6 Multiple Linear Regression Search for the two time-points that give the best correlation coefficient. Two variants were also tested: ln(Clearance) and ln(Concentration) Bayesian Estimation : 7 Bayesian Estimation In Bayesian estimation, the target variable is predicted as a PDF instead of a single value. Use maximum a posteriori (MAP) to build a probability model for clearance as function of the PDFs for PK parameters. Pick the two time-points as to minimize variance Several variants were tested each only taking a subset of the PK parameters as input Results : 8 Results Bayesian estimation yielded lower error overall The best model was Bayesian estimation using only the V and Ke PK parameters for etoposide. The best linear model was achieved by log-scaling concentration time-points Study 2: Pharmacology in ALL patients who develop AML : 9 Study 2: Pharmacology in ALL patients who develop AML Etoposide, a drug used in treating ALL causes secondary AML (Acute Myeloid Leukemia) in some patients. Goal of Study: determine whether patients who developed AML, displayed different PK or other pharmacologic characteristics as compared to those who did not. Data : 10 Data Plasma samples were taken in regular time intervals after Etoposide treatment began Measurements of Etoposide and Catechol metabolite concentrations were made. A two compartment first-order PK model (same as previous study) was assumed. Values for PK parameters were estimated using a Bayesian estimation algorithm. Parameters were compared between AML and control groups using Wilcoxon and Conditional Logistic Regression test Statistical Tests: Wilcoxon Test : 11 Statistical Tests: Wilcoxon Test A non-parametric test for determining whether a significant difference exists between two populations Calculation: Taking each observation from one of the two samples, count the number of observations in the other sample that are smaller than it. U = total of these counts Significance: U must be looked up against a table of Mann-Whitney U distribution to obtain a critical value Generally a p value < 0.05 means significant difference Conditional Logistic Regression : 12 Conditional Logistic Regression Simple Logistic Regression In our case probability of developing AML would be conditioned on all input variables Conditional Logistic Regression: Marginalize or condition-out all parameters except for the parameter of interest Results : 13 Results Onset of AML vs. Various PK parameters : 14 Onset of AML vs. Various PK parameters Study 3: Genes Associated with Chemotherapy Resistance : 15 Study 3: Genes Associated with Chemotherapy Resistance Chemotherapy is an effective treatment for 75% of ALL patients. The cause of treatment failure in the remaining 25% is not known Problem Statement: Are there Genes that are differentially expressed between cells sensitive / resistant to chemoteraphy? Experimental Method : 16 Experimental Method ALL Cells were taken from 441 patients In-Vitro experiments were performed to measure the lethal concentration (LC50) of four chemo-drugs on these cells. PCA was performed on the original data First PC accounted for 48% in variance with nearly equal weights for the four drugs Second PC accounted for 24% variance with nearly equal weights for VCR and ASP but in opposite directions Identifying Sensitive and Resistant Cells : 17 Identifying Sensitive and Resistant Cells Weights from first PC were used to calculate a Cross-resistance score (CR) Weights from second PC were used to calculate a VCR-ASP score The top and bottom quartiles were used to define patients with cross-sensitive and cross-resistant ALL (CR) and two groups of patients exhibiting discordant resistance to VCR and ASP Identifying genes : 18 Identifying genes Correlation between gene expression values and resistance scores (CR, VCR-ASP) was calculated. 45 discriminating genes were identified for CR and 139 for VCR-ASP Identified genes were validated by performing clustering on the gene expression data References : 19 References JC Panetta, et al. Limited and optimal sampling strategies for etoposide and etoposide catechol in children with leukemia. Journal of Pharmacokinetics and Pharmacodynamics. 2002. MV Relling, et al. Etoposide and antimetabolite pharmacology in patients who develop secondary acute myeloid leukemia. Leukemia. 1998. S. Lugthart et. al, Identification of genes associated with chemotherapy crossresistance and treatment response in childhood acute lymphoblastic leukemia, Cancer Cell, Volume 7, Issue 4, , April 2005, Pages 375-386.

Add a comment

Related presentations

Related pages

Acute lymphoblastic leukemia - Wikipedia, the free ...

acute lymphoblastic leukemia (ALL), peripheral blood of a child, Pappenheim stain, magnification x100. ... Acute lymphoblastic leukemia; Myeloid:
Read more

Acute Lymphoblastic Leukemia: Survival, Treatments, and More

WebMD explains acute lymphoblastic leukemia (ALL) including its symptoms, risk factors, and treatments.
Read more

Childhood Acute Lymphoblastic Leukemia Treatment (PDQ ...

Childhood acute lymphoblastic leukemia (ALL) is a type of cancer in which the bone marrow makes too many immature lymphocytes (a type of white blood cell).
Read more

What is acute lymphocytic leukemia? | American Cancer Society

Acute lymphocytic leukemia (ALL), also called acute lymphoblastic leukemia, is a cancer that starts from white blood cells called lymphocytes in ...
Read more

Acute lymphocytic leukemia - Mayo Clinic

Acute lymphocytic leukemia is also known as acute lymphoblastic leukemia. Acute lymphocytic leukemia is the most common type of cancer in children, ...
Read more

Acute lymphoblastic leukemia (ALL): MedlinePlus Medical ...

Acute lymphoblastic leukemia (ALL) is a fast-growing cancer of a type of white blood cell called a lymphoblast. ALL occurs when the bone marrow ...
Read more

Acute lymphoblastic leukemia (ALL) -

Acute lymphoblastic leukemia (ALL), also called acute lymphocytic leukemia or acute lymphoid leukemia, is a fast-growing cancer of a type of white blood ...
Read more