Page 101 - 2019_08-Haematologica-web

TKI discontinuation in CML clinical practice
No progressions occurred. Nine deaths were reported but none of them was disease related.
The patients who resumed therapy (Table 2) were given imatinib (n=77), nilotinib (n=22), dasatinib (n=9), bosu- tinib (n=3), ponatinib (n=1), or IFN (n=2). Most of the patients who stopped imatinib restarted imatinib after relapse, and patients who were on second generation TKI mainly stayed with second generation TKI. Ninety-four percent of the patients who were retreated achieved at least another MMR, and 82% of them achieved another DMR, fitting the criteria for a second attempt at discontin- uation.31
In 20 patients who had lost MR4, and in four patients who had lost MMR, the treatment was not resumed fol-
lowing a shared decision with the doctor. Interestingly, they were still on the same response after a median time of 12 months (IQR 1-32).
Prognostic factors
Univariate analysis – Univariate analysis was used to assess age (considered as continuous variable), sex (female vs. male), Sokal score (intermediate vs. low; high vs. low), type of therapy (second generation TKI vs. ima- tinib), line of therapy at stop (imatinib vs. imatinib post IFN; first-line second generation vs. second generation in second or further lines), type of transcript (b2a2 vs. others), duration of therapy with the last TKI and any TKI (contin- uous variables), duration of total treatment (continuous
Table 2. Type of retreatment after failure of discontinuation.
Overall 2nd generation TKI
Imatinib (n=88)
75 (85) 4 (5) 5 (6) 2 (2) 0 (0) 2(2)
P
0.41
0.19
0.06 0.31
0.77 0.04 0.97 0.84 0.73 0.27 0.07
0.1 0.14 0.18
0.03
0.18
0.22
0.22
Type of retreatment (%) Imatinib
(n=114) (n=26)
77 (67) 2 (8) 22 (19) 18 (69) 9 (8) 4 (15) 3 (3) 1 (4) 1 (1) 1 (4)
Nilotinib
Dasatinib
Bosutinib
Ponatinib
IFNα 2(2) 0(0)
Table 3. Hazard Ratios (HRs) computed at univariate analysis.
Sex
female vs. male
Sokal score
Intermediate vs. low
high vs. low Type of therapy
2nd generation vs. imatinib Type of transcript
b2a2 vs. others
Age at discontinuation (older vs. younger; diff. of 22 ys) Duration of DMR* (diff. of 43 mos)
Time to DMR before stop* (32 mos increase)
Duration of therapy with last TKI* (57 mos increase) Duration of treatment with any TKIs* (58 mos increase) Duration of total treatment* (68 mos increase)
Depth of MR at stop
MR4.5 vs.MR4
MR5 vs. MR4 Undetectable vs. MR4
HR 95%CI
1.17 0.81 1.69
0.74 0.47 1.17
1.66 0.98 2.81
0.8 0.52 1.23
0.93 0.58 1.49 0.76 0.58 0.98 1.01 0.77 1.31 0.97 0.75 1.27 1.04 0.80 1.37 0.85 0.64 1.13 0.79 0.62 1.02
0.67 0.42 1.07 0.68 0.4 1.14 0.7 0.4 1.19
1.53 1.04 2.24
1.57 0.81 3.05
1.40 0.82 2.38
0.73 0.43 1.21
Line of therapy at stop
1st line vs. ≥ 2nd line
Reason for discontinuation
Pregnancy vs. shared decision with MD
ISAV vs. shared decision with MD
Toxicity vs.shared decision with MD
*For each variable the difference of months between groups of patients considered for computing HR corresponds to the Interquartile Range (IQR); ys: years; DMR: deep molecular response; mos: months; MD: medical doctor.
haematologica | 2019; 104(8)
1593