Page 137 - 2022_01-Haematologica-web

Predicting response to IST plus EPAG in SAA
plus eltrombopag group (Online Supplementary Table S1). Although not statistically significant, the patients with PNH clones showed slightly higher overall response in both IST groups (odds ratio 1.32, P=0.24) (Online Supplementary Table S1), consistent with previous studies, but the favorable tendency disappeared in the multivariate model (odds ratio 1.06, P=0.87), as the presence of PNH clones correlated with higher ALC (P=0.018) and higher ARC (P=0.073). We also noted that the 19 patients who did not have enough ANC to assess PNH clones showed a significantly reduced overall response to IST plus eltrom- bopag at 53% (10 of 19; P=0.0019), which was analogous to the response rate in 19 patients with ANC <0.02×109/L
at baseline.
The response curves revealed that the reduction of
overall response to IST plus eltrombopag was observed only in patients with pre-treatment ARC lower than 10×109/L while overall response to IST alone decreased below an ARC of 30×109/L (Figure 2A). The overall response in subjects with ARC between 10-30×109/L was significantly improved in the group treated with IST plus eltrombopag, to the levels of those of patients with ARC >30×109/L (from 60% to 91%, P<0.0001) (Figure 2B). Addition of eltrombopag did not appear to significantly improve the overall response in subjects with ARC <10×109/L (63% in the IST plus eltrombopag group and 46% in the IST alone group, P=0.062) nor in those with ARC >30×109/L (90% vs. 89%, P=1), but was beneficial in terms of increasing complete response (27% vs. 8% in the groups with ARC <10×109/L, P=0.0036; 42% vs. 21% in
Table 2. Pretreatment variables and responses to immunosuppressive therapy.
NR PR
Therapy N Value N Value N
CR P
Value
OR vs. NR 0.91
0.17
0.70 0.69
0.0009 <0.0001
0.00027
0.0019 0.63
0.079* 0.065* 0.0015 0.095*
0.12 0.28 0.30 0.57
0.82 1 0.27 0.84
0.0037 0.60
Age, years Male sex
ARC, ×109/L ANC, ×109/L ALC, ×109/L
Plt, ×109/L
PNH clone ≥1%
TPO, ng/mL
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG IST alone
IST+EPAG
32 32 (15-58) 91 32 (19-55)
32 17 (53%) 91 55 (60%)
32 9.2 (6.6-22.0) 91 9.6 (3.2-19.1)
32 0.06 (0.01-0.25) 91 0.20 (0.06-0.45)
32 1.00 (0.59-1.5) 91 1.07 (0.77-1.5)
32 5.5 (2.0-11.3) 91 8.0 (5.5-11.0)
23 8 (35%) 90 30 (33%)
26 3000 (2610-3430)
75 31 (20-56) 117 28 (18-50)
75 35 (47%) 117 71 (61%)
75 24.5 (9.6-39.9) 117 22.0 (7.6-36.0)
75 0.36 (0.17-0.64) 117 0.38 (0.19-0.53)
75 1.22 ( 0.98-1.51) 117 1.32 (1.01-1.69)
75 8.0 (5.0-11.0) 117 9.0 (6.0-14.0)
71 28 (39%) 115 49 (43%)
61 2560 (2130-2880)
69 32 (19-53) 32 24 (14-48)
69 35 (51%) 32 15 (47%)
69 18.3 (10.5-35.6) 32 17.8 (10.9-36.6)
69 0.28 (0.09-0.43) 32 0.32 (0.16-0.51)
69 1.33 (1.13-1.77) 32 1.51 (1.17-1.72)
69 8.0 (5.0-13.0) 32 9.5 (5.8-12.3)
63 24 (38%) 31 11 (35%)
53 2530 (2250-3000)
CR vs. no CR 0.88
0.15
0.88 0.18
0.49 0.086
0.60
Data are n (%) or median (interquartile range) unless otherwise noted. NR: no response; PR: partial response; CR: complete response; IST: immunosuppressive therapy; EPAG, eltrombopag; ARC: absolute reticulocyte count; ANC: absolute neutrophil count; ALC: absolute lymphocyte count; Plt: platelet count; PNH: paroxysmal nocturnal hemoglobinuria; TPO: thrombopoietin. *P<0.05 in patients aged ≥10 years.
AB
Figure 2. Pretreatment absolute reticulocyte count and response rates to immunosuppressive therapy. (A) The overall response rates and their 90% confidence interval of individuals with an initial absolute reticulocyte count between X and (X+10) ×109/L in the group treated with immunosuppressive therapy (IST) plus eltrom- bopag (EPAG) and in the group treated with IST alone. (B) The overall response rates and the complete response rates were compared between the IST plus EPAG group and IST alone group according to initial absolute reticulocyte count. Overall response includes complete responses and partial responses. OR: overall response; ARC: absolute reticulocyte count; PR: partial response; CR: complete response.
haematologica | 2022; 107(1)
129