IKZF3 p.L162R Is a Recurrent Hotspot Mutation in Chronic Lymphocytic Leukemia (CLL)

Recent advances in massively parallel DNA sequencing has enabled identification of various somatic driver mutations in chronic lymphocytic leukemia (CLL). However, recent study has shown that the search of recurrent driver mutations are far from complete, and sequencing of extended cohort continues to reveal rare but recurrent driver mutations.

To search for rare but biologically relevant driver mutations, we conducted whole exome sequencing (WES) on 54 bone marrow samples from patients with CLL, 35 (60%) of whom were untreated and 19 (40%) relapsed after chemoimmunotherapy such as FCR or BR regimens. WES achieved mean targeted sequencing coverage of 124x. Somatic single nucleotide variant (SNV) and small insertions/deletions were called using Mutect and Pindel algorithm, respectively, against matched normal control.

Table 1 shows the comparison of clinical characteristics between untreated and treated patients. As expected, patients previously treated with chemo-immunotherapy had more advanced clinical features compared to untreated patients, such as older age, more advanced Rai stage, higher incidence of 17p deletion, higher levels of serum beta 2 microglobulin, and worse thrombocytopenia.

The median number of somatic SNV was 29 (range: 14-59) per case. There was no significant difference in the numbers of somatic SNV between untreated and treated patients (median number of SNV: 27 [range: 12-59] in untreated vs. 32 [range: 14-59] in treated, P = 0.913). There was no association between the number of SNV and other clinical features such as Rai stage, IGHV mutation status, and 17p deletion status, although analyses are limited by patient number. Mutation signature was also similar between untreated and treated patients. In addition to previously reported recurrent mutations in CLL, 3 patients (6%) had somatic hot-spot mutation in IKZF3, which was a heterozygous missense mutation (T>G) resulted in p.L162R. Two of the 3 patients with IKZF3 mutation were previously treated and the mutation was clonal by ABSOLUTE analysis (cancer cell fraction > 0.9). Both patients had high risk disease features such as deletion 17p and ZAP70 positivity. The other treatment-naive patient had subclonal IKZF3 mutation. One patient with clonal IKZF3 mutation responded well to IPI-145 (PI3K delta gamma inhibitor).

IKZF3 is a transcription factor, which plays an important role in B-cell differentiation, proliferation and maturation. It has three zinc finger domains, the L162R mutation resides in the second zinc finger domain. The mutated residue was highly conserved among all species. In the COSMIC database, IKZF3 p.L162R somatic mutation was also found in diffuse large B cell lymphoma and mantle cell lymphoma, suggesting shared genetic mechanisms among other B-cell ,malignancies. In a recent study, IKZF3 p.L162R was shown to inhibit ubiquitination of IKZF3 protein by lenalidomide therapy in 293T cell (Kršnke et al. Science 2014). Therefore, we tested expression of IKZF3 protein in CLL bone marrow by immunohistochemistry. Very strong nuclear expression of IKZF3 was observed in the IKZF3 mutated cases whereas IKZF3 expression was weak in wild type cases, suggesting that p.L162R mutation leads to longer-lived IKZF3 protein. We are currently sequencing bone marrow samples from CLL patients treated with lenalidomide to evaluate association between IKZF3 mutation and response to lenalidomide therapy.

Table 1. Characteristics of 54 CLL patients who underwent bone marrow WES.

Characteristic	Untreated	Treated	P
	n = 35	n = 19
Median age (range), y	61 (37-84)	69 (56-81)	0.009
Female
Rai Stage (%)			0.131
0	12 (34)	2 (11)
I	9 (26)	4 (21)
II	2 (6)	1 (5)
III	5 (14)	2 (11)
IV	7 (20)	10 (53)
Rai Stage 0-II (%)	23 (66)	7 (37)	0.041
Rai Stage III-IV (%)	12 (34)	12 (63)
Deletion 17p (%)	4 (11)	7 (37)	0.027
Complex karyotype (%)	4 (12)	4 (24)	0.243
IGHV mutation status (%)
Mutated	15 (43)	7 (37)	0.546
Unmutated	10 (19)	7 (37)
Unknown	10 (19)	5 (26)
ZAP70 status (%)
Negative	16 (46)	6 (32)	0.313
Positive	19 (54)	13 (68)
Median WBC count (range), x 109/L	25.2 (4.3-140.8)	12.5 (2.5-161.7)	0.163
Median HGB count (range), g/dL	13.6 (7.4-16.5)	12.3 (8.5-16.1)	0.101
Median BM CLL cells by flow (range), %	71 (22-91)	76 (0.8-96)	0.825
Serum B2MG (range), mg/dL	1.69 (0.9-7.0)	4.4 (2.2-12.4)	0.001