Skip to main content

Table 2 Chronic lymphocytic leukemia: REMARK-like profile of the analysis performed on the dataset

From: Added predictive value of omics data: specific issues related to validation illustrated by two case studies

a) Patients, treatment and variables

Study and marker

Remarks

Marker

OS = 8-probe-set gene-expression signature

Further variables

v1 = age, v2 = sex, v3 = FISH, v4 = IGVH

Reference

Herold et al. (2011)

Source of the data

GEO (reference: GSE22762)

Patients

n

Remarks

 
 

Assessed for eligibility

151

Disease: chronic lymphocytic leukemia

   

Patient source: Department of Internal Medicine III, University of Munich (2001 - 2005)

Training set

Excluded

0

  
 

Included

151

Criteria: sample availability

   

Gene expression profiling: 44 Affymetrix HG-U133 A&B microarrays, 107 Affymetrix HG-U133 plus 2.0 microarrays

 

With outcome events

41

Overall survival

 

Assessed for eligibility

149

Disease: chronic lymphocytic leukemia

   

Patient source: Department of Internal Medicine III, University of Munich (2005 - 2007)

Validation set

Excluded

18

Due to missing clinical information

 

Included

131

Criteria: sample availability

   

Gene expression profiling: 149 qRT-PCR (only selected genes)

 

With outcome events

40

Overall survival

Relevant differences between training and validation sets

Data source

Same institution, different time (see above)

 

Measurement of gene expressions

Affymetrix HG-U133 vs. TaqMan LDA (see text)

 

Survival rate

Lower in the validation set (see Figure 4)

 

b) Statistical analyses of survival outcomes

Analysis

n

e

Variables considered

Results/remarks

F: preliminary analysis (separately on training and validation sets)

F1: univariate

151

41

v1 to v4

Kaplan-Meier curves (Figure 3)

131

40

G: Multivariate testing of the omics score in the validation data (only validation set involved)

G1: significance

131

40

OS, v1 to v4

Multivariate Cox model (Table 5)

H: Comparison of the predictive accuracy of clinical and combined models through cross-validation in the validation data (only validation set involved)

H1: Overall prediction

131

40

OS, v1 to v4

Prediction error curves based on cross-validation (Figure 11)

Integrated Brier score based on cross-validation (text)