Page 99 - Haematologica3

Prediction of primary resistant AML
combination with patients with RD (non-responder; see Figure 1A) was analyzed separately in the validation set.
Overall survival was defined as time from study entry until death from any cause. Patients alive were censored at the time of their last follow up. The prognostic impact of the classifier was evaluated by the Kaplan-Meier method and the log-rank test. Because resistant AML can only be cured by stem cell transplanta- tion (SCT), all survival analyses were censored for SCT if not oth- erwise indicated. The χ2-test was applied for categorical variables and the Wilcoxon test for continuous variables for statistical com- parisons. For three-way comparison of numerical variables, we used the Kruskal-Wallis Test. Multivariable logistic models for resistance to induction treatment and multivariable Cox models for OS were used to adjust for potential confounders. P≤0.05 was considered statistically significant.
Results
Patients
A flow chart of the study is given in Figure 1B. A total of 1106 patients were included in the analysis. Patients' characteristics are shown in Table 1. Patients in training
set 2 were significantly younger and had a higher response rate than patients in the other two sets. Due to the addi- tion of resistant patients, the validation set included signif- icantly fewer patients with favorable cytogenetics. The median follow up was more than eight years in the train- ing sets and 4.2 years in the validation set.
Validation of the final predictor
The final score (Predictive Score 29 MRC, PS29MRC) consists of 29 gene expression markers and the cytogenet- ic risk groups defined according to the MRC classification,22 and is calculated as a weighted linear sum of the individual predictors (Figure 2).
The score in the validation set ranged from -2.75 to +3.72. In univariate analysis, PS29MRC as a continuous variable (PS29MRCcont) was a highly significant predictor of RD in the validation set with an odds ratio (OR) of 2.39 (95%CI: 1.80, 3.26; P=8.63·10-9, AUC=0.76) (Figure 3A). Similar results were seen with PS29MRC as a dichoto- mous variable (PS29MRCdic) applying the pre-defined cut off (Online Supplementary Appendix) defined in training set 1 (OR 8.03, 95%CI: 4.07, 16.46; P=4.29·10-9). We subse- quently designed multivariable models including all vari-
Table 1. Patients’ characteristics. Number of patients
de novo AML, n (%)
Age, median (range)
Male sex, n (%)
MRC favorable, n (%)
MRC intermediate, n (%)
MRC unfavorable, n (%)
White-cell count (x 10 9/L), median (range) Hemoglobin (g/dL), median (range) Platelet count (x 10 9/L), median (range) LDH (U/L), median (range)
Bone marrow blasts, % median (range) ECOG performance status > 1, n (%) CR/CRi, n (%)
AML with resistant disease, n (%) Cumulative relapse, n (%)
5-year survival in %, [n of deaths] Median follow up (years)
Training Set 1§
407
354 (87)
57 (18-85)
199 (49)
52 (13)
253 (63)
94 (24)
21.2 (0.4-666)
9.0 (3.5-15.4)
53.0 (1-1760)
452 (76-4613)
80 (10-100)
112 (28)
300 (74)
107 (26)
188 (71)
32 [283]
8.6
Training Set 2
449
NA
46 (15-77) 227 (51) 70 (16) 298 (66) 81 (18) NA
NA
NA
NA
NA
NA 372 (83) NA* 189 (59) 38 [289] 8.6
Validation Set$
250
208 (83)
58 (18-74)
120 (48)
14 (6)
188 (75)
48 (19)
17.0 (0.5-406)
9.0 (4.5-16.0)
61.5 (1-997)
408 (107-6601)
73 (6-100)
40 (16)
164 (66)
71 (28)
67 (56)
36 [150]
4.2
P
0.22 <0.001 0.79 <0.001
0.14 0.57 0.063 0.13 0.0016 <0.001 <0.001 0.33 0.0031 0.096
§The data set was restricted to patients with definitive induction results (CR/CRi or resistant disease). $ The data set includes additional 40 patients with resistant disease from the AMLCG-1999 trial. * In training set 2 only the information of responder (CR/CRi) and non-responder (n=77, 17%) was available. n: number; AML: acute myeloid leukemia; LDH: lactate dehydrogenase; ECOG: Eastern Cooperative Oncology Group; CR: complete remission; CRi: incomplete hematologic regeneration.
Table 2. Univariate and multivariable analysis of the prediction of resistant disease in the validation set.
Variable
PS29MRCdic
Age continuous
NPM1mut
RUNX1mut
TP53mut
Multivariable analysis, n=235 OR [95%-CI]
4.44 [2.00; 10.16]
1.06 [1.03; 1.10]
0.48 [0.19; 1.142]
1.05 [0.50; 2.44]
7.16 [1.76; 38.61]
P
0.00030 0.00012 0.094 0.90 0.010
Univariate analysis OR [95%-CI]
8.03 [4.07; 16.46] 1.07 [1.04; 1.10] 0.23 [0.11; 0.46] 2.13 [1.07; 4.21] 12.03 [3.72; 53.84]
P
4.29·10-9 3.87·10-6 6.62·10-5 0.029 0.00016
OR: Odds Ratio; CI: Confidence Interval.
haematologica | 2018; 103(3)
459