Page 113 - 2020_11-Haematologica-web

Prognostic impact of CLL stereotyped subsets
Table 2. Baseline characteristics of the advanced stage chronic lymphocytic leukemia (CLL) cases according to subset classification.
Target analysis population, n (%)
Age at study entry (years) median (range) Gender, n (%) Male
ECOG performance status median (range) Binet stage, n (%)
A B C
CLL-IPI risk group, n (%)
Low Intermediate High
Very high
Subset #1
39 (2.09%)
66 (43-84) 24 (61.5) 0.5 (0-2) 39
6 (15.4) 20 (51.3) 13 (33.3) 38
1 (2.6) 10 (26.3) 26 (68.4) 1 (2.6) 104.7 (13.7-375.0) 26.6 (0.5-330.6) 3.4 (0.6-8.1) 1 (2.6) 17 (43.6) 4 (10.3) 23 (59.0) 1 (3.8)
4 (16.0)
0 (0.0)
1 (2.6)
Subset #2
61 (3.28%)
66 (48-88) 36 (59.0) 1 (0-2) 61
7 (11.5) 33 (54.1) 21 (34.4) 57
18 (31.6) 19 (33.3) 18 (31.6) 2 (3.5) 72.3 (3.1-492.0) 20.2 (0.0-163.0) 3.0 (0.0-9.8) 0 (0.0) 18 (31.0) 2 (3.4) 49 (84.5) 2 (4.8)
2 (4.8) 19 (45.2)
2 (3.3)
Subset #4
11 (0.59%)
52 (42-74) 6 (54.5) 1 (0-1) 11
1 (9.1) 6 (54.5) 4 (36.4) 11
7 (63.6) 3 (27.3) 1 (9.1) 0 (0.0) 73.5
Subset #8
16 (0.86%)
67 (35-84) 8 (50.0) 0.5 (0-2) 16
2 (12.5) 11 (68.8) 3 (18.8) 16
0 (0.0) 4 (25.0) 12 (75.0) 0 (0.0) 56.4
u-IGHV
1088 (58.46%)
65 (30-90) 775 (71.2) 1 (0-3) 1087 180 (16.6) 568 (52.3) 339 (31.2) 1029
0 (0.0) 388 (37.7) 534 (51.9) 107 (10.4) 85.3
m-IGHV
646 (34.71%)
65 (36-89) 437 (67.6) 0 (0-3) 646 112 (17.3) 243 (37.6) 291 (45.0) 593 252 (42.5) 228 (38.4) 105 (17.7) 8 (1.3) 65.9
All
1,861
65 (30-90)
1286 (69.1)
1 (0-3)
1860
308 (16.6)
881 (47.4)
671 (36.1)
1744
278 (15.9)
652 (37.4)
696 (39.9)
118 (6.8)
79.1
(0.2-867.0)
16.5
(0.0-970)
3.0
(0.0-17.8)
101 (5.6)
378 (21.0)
222 (12.3)
648 (36.0)
136 (10.6)
185 (14.6)
202 (15.9)
53 (2.8)
Leukocyte count (x 109/L) median (range)
Serum thymidine kinase c(U/L) median (range)
(19.2-144.1) (12.9-118.0) (0.2-867.0) (3.1-741.9)
Serum β2-microglobulin (MG/L) median (range)
Deletion 17P by FISH, n (%) Deletion 11Q by FISH, n (%) Trisomy 12 by FISH, n (%) Deletion 13Q by FISH, n (%) TP53 mutated, n (%)
10.2
(4.1-36.3) 2.2 (0.5-5.1) 0 (0.0) 0 (0.0) 1 (9.1) 6 (54.5) 0 (0.0) 0 (0.0) 0 (0.0)
0
55.9
(0.5-165.3) 3.8 (0.3-8.1) 0 (0.0)
6 (37.5) 11 (68.8) 2 (12.5) 0 (0.0)
5 (38.5) 0 (0.0)
1 (6.3)
19.2
(0.0-848.0) 3.0 (0.0-17.8) 79 (7.5) 316 (29.8) 163 (15.4) 477 (45.1) 103 (13.6) 154 (20.5) 131 (17.4) 35 (3.2)
11.3
(0.0-970.0) 2.8 (0.0-12.2) 21 (3.4) 37 (6.0) 80 (13.0) 409 (66.3) 30 (6.9) 20 (4.7) 52 (12.1) 14 (2.2)
NOTCH1 mutated, n (%)
SF3B1 mutated, n (%)
Richter syndrome status, n (%)
n: number of cases; u: unmutated; m: mutated; IGHV: non-subset immunoglobulin heavy variable; ECOG: Eastern Cooperative Oncology Group; IPI: International Prognostic Index;FISH: fluorescence in situ hybridization.
tively. TTFT was significantly longer in M-CLL as com- pared to U-CLL (P<0.001). Subset #4 had a longer TTFT compared to subsets #1, #2 and #8 (P=0.002), but similar to M-CLL (P=0.338) (Figure 1A). Therapy was needed only for 1 of 8 patients assigned to subset #4. Statistically, there was no difference in TTFT between subsets #1, 2 and 8, and U-CLL (P=0.204) (Figure 1A). Regarding PFS, a significant difference was observed between M-CLL and U-CLL (P<0.001). Subset #4 had a longer PFS than subsets #1, 2 and 8 (P=0.014) that was similar to M-CLL cases (P=0.249), with only two patients progressing. Subsets #1, 2, 8 and U-CLL had a similar PFS (P=0.198) (Figure 1B). Regarding OS, M-CLL had a significantly longer time to event compared to U-CLL (P<0.001). Subsets #1, 2, 4, and 8 showed no statistically significant differences in terms of OS; however, the OS curves followed a similar pattern to that observed for PFS and TTFT with subsets #1, 2, and 8, showing more events as compared to subset #4 where no incidents of death were observed (Figure 1C).
During a median observation time of 55.1 ms of the advanced stage cohort, there were 1,847, 1,258, 739 and 512 events for TTFT, PFS, TTNT, and OS, respectively. TTFT was longer for M-CLL as compared to U-CLL (P<0.001). Subset #4 had a similar TTFT to that of M-CLL that was longer as compared to subsets #1, 2 and 8
(P=0.009). Subsets #1, 2 and 8 had a similar TTFT as U- CLL (Figure 2A). Regarding PFS, M-CLL had a longer time to event compared to U-CLL (P<0.001). Subset #4 patients had a longer PFS compared to subsets #1, 2, and 8; how- ever; this difference was not statistically significant. Subset #8 had a shorter PFS compared to subsets #1, 2 and U-CLL (P=0.042) (Figure 2C). Regarding TTNT, M-CLL had a longer time to event compared to U-CLL (P<0.001). No subset #4 patient needed subsequent therapy during the observation time and TTNT was longer as compared to subsets #1, 2, and 8 (P=0.027). Subset #4 also had a longer TTNT compared to M-CLL without reaching sta- tistical significance (P=0.058) (Figure 2B). Regarding OS, we observed a significant difference between M-CLL and U-CLL (P<0.001). None of the subset #4 patients died before time of analysis; subset #4 patients had a longer OS than subsets #1, 2 and 8 (P=0.011) (Figure 2D).
Cases expressing IGHV3-21: subset #2 and others
In the cohort of early stage CLL patients, m-subset #2 had a significantly shorter TTFT compared to m-IGHV3- 21 and m-IGHV patients (P<0.001). While m-subset #2 had an outcome similar to that of the u-IGHV group (P=0.794), m-IGHV3-21 was comparable to m-IGHV (Figure 3A). Similarly, PFS was shorter in m-subset #2
haematologica | 2020; 105(11)
2601