a) Patients, treatment and variables | ||||
---|---|---|---|---|
Study and marker | Remarks | |||
Marker | OS = 86-probe-set gene-expression signature | |||
Further variables | v1 = age, v2 = sex, v3 = NMP1, v4 = FLT3-ITD | |||
Reference | Metzeler et al. (2008) | |||
Source of the data | GEO (reference: GSE12417) | |||
Patients | n | Remarks | Â | |
Training set | Assessed for eligibility | 163 | Disease: acute myeloid leukemia | |
 |  | Patient source: German AML Cooperative Group 1999-2003 | ||
Excluded | 0 | Â | Â | |
Included | 163 | Treatment: following AMLCG-1999 trial | ||
 |  | Gene expression profiling: Affymetrix HG-U133 A&B microarrays | ||
With outcome events | 105 | Overall survival: death from any cause | ||
Validation set | Assessed for eligibility | 79 | Disease: acute myeloid leukemia | |
 |  | Patient source: German AML Cooperative Group 2004 | ||
Excluded | 0 | Â | Â | |
Included | 79 | Treatment: 62 following AMLCG-1999 trial 17 intensive chemotherapy outside the study | ||
 |  | Gene expression profiling: Affymetrix HG-U133 plus 2.0 microarrays | ||
With outcome events | 33 | Overall survival: death from any cause | ||
Relevant differences between training and validation sets | ||||
Data source | Same research group, different time (see above) | |||
Follow-up time | Much shorter in the validation set (see text) | |||
Survival rate | Higher in the validation set (see Figure 2) | |||
b) Statistical analyses of survival outcomes | ||||
Analysis | n | e | Variables considered | Results/remarks |
A: preliminary analysis (separately on training and validation sets) | ||||
A1: univariate | 163 | 105 | v1 to v4 | Kaplan-Meier curves (Figure 1) |
79 | 33 | |||
B: evaluating clinical model and combined model on validation data (models fitted on training set, evaluated on validation set) | ||||
B1: overall prediction |  |  |  | Prediction error curves (Figure 5) |
 |  |  | Integrated Brier score (text) | |
 | Training |  |  | Comparison of Kaplan-Meier curves for risk groups: |
 | 163 | 105 |  | - Medians as cutpoints (Figure 6), |
B2: discriminative ability | Â | Â | OS, v1 to v4 | - K-mean clustering (data not shown - see text) |
 |  |  | C-index (text) | |
 | Validation |  |  | K-statistic (text) |
B3: calibration | 79 | 33 |  | Kaplan-Meier curve vs average individual survival curves for risk groups (Figure 7) |
 |  |  | Calibration slope (text) | |
C: Multivariate testing of the omics score in the validation data (only validation set involved) | ||||
C1: significance | 79 | 33 | OS, v1 to v4 | Multivariate Cox model (Table 3) |
D: Comparison of the predictive accuracy of clinical and combined models through cross-validation in the validation data (only validation set involved) | ||||
D1: overall prediction | 79 | 33 | OS, v1 to v4 | Prediction error curves based on repeated cross-validation (Figure 8) |
Prediction error curves based on repeated subsampling (data not shown - see text) | ||||
Prediction error curves based on repeated bootstrap resampling (data not shown - see text) | ||||
Integrated Brier score based on cross-validation (text) | ||||
E: Subgroup analysis (E1-E3 based on training and validation sets, E4 and E5 only on validation set; for all, separate analysis for female and male population) | ||||
E1: overall prediction | Female | OS, v1 to v4 | Prediction error curves (Figure 9) | |
E2: discriminative ability | t.: 88 54 | C-index (text) | ||
 | v.: 46 16 | K-statistic (text) | ||
E3: calibration | Male | Calibration slope (text) | ||
E4: significance | t.: 74 51 | Multivariate Cox model (text) | ||
E5: overall prediction | v.: 33 17 | Prediction error curves based on cross-validation (Figure 10) |