Utilization of Next Generation Sequencing Identifies Potentially Actionable Mutations with Prognostic Significance in Chronic Lymphocytic Leukemia

Introduction: CLL is a clinically and biologically heterogeneous disease; cytogenetic evaluation with fluorescence in situ hybridization (FISH) is routinely used to guide therapy.  For example, del17p is associated with chemoimmunotherapy (CIT) refractoriness and decreased survival. The use of kinase inhibitors (KI) has improved clinical outcomes; however, some pts progress on KI and require subsequent therapies.  Next generation sequencing (NGS) can further define genetic alterations that may act in concert to drive malignancy, and identify pathways that can be targeted with novel approaches.  Here we describe the mutational landscape of a cohort of 57 CLL pts treated at the University of Pennsylvania and identify potentially targetable pathways for intervention.

Methods:  We identified 57 pts who underwent analysis of tumor DNA using NGS (2013-2015) and analyzed clinical characteristics, genetic mutations, and progression free survival (PFS). NGS was performed on an Illumina MiSeq using a 33 gene amplicon-based panel developed at our center with detection limit of 5% allele frequency with a minimum depth of coverage of 250x. We used custom bioinformatic pipelines combining open source tools and custom algorithms for analysis. Pathogenic mutations were defined as those that have been reported in studies with functional data.

Results: Of the 57 pts who underwent NGS the median age at NGS was 65.5 yr (range 17.7-90.7), 65% were male, 21% patients received CIT alone, 21% patients received KI alone and 39% pts received both CIT and KI, and 32% received KI in relapse. 74% (42/57) of pts had at least one genetic mutation identified by NGS.  The median number of mutations per pt was 1 (range 0-8).  25% of pts had ≥ unique 3 mutations). Mutations in 24 unique genes (n=94) were identified and were categorized as likely pathogenic (69%), variants of uncertain significance (27%), or likely benign (4%). The most frequently mutated genes were ATM (20%), SF3B1 (12%), NOTCH1 (10%), DNMT3A (7%), and TP53 (7%).  We identified 19 low frequency gene mutations, which in aggregate affect 24.5% of the pt cohort (Table).  The median PFS for CIT pts was 31.4 mo (median f/u 15 mo) and 8 mo for KI pts (median f/u 4 mo). Using Cox regression, the presence of ≥ 1 mutation was associated with an inferior PFS (Figure) following CIT when controlled for del11q status (HR 3.1, p=.05) or complex karyotype (HR 3.4, p=.03) and a PFS trend when controlled for del17p (HR 2.7, p=.08).  Pts with ≥ 4 mutations had a shorter PFS on ibrutinib (Ibr) compared to those with fewer mutations (p=0.0002).  

Conclusion: NGS identifies several mutations that may be targetable using agents which have not been tested in CLL.  The presence of a mutation identified by NGS predicts for inferior PFS on CIT, and the presence of ≥ 4 mutations predict early treatment failure on Ibr. These genetic alterations demonstrate the diversity of pathways that are involved in CLL biology.  These results support a rationale for clinical trial design using a precision medicine approach selecting therapies which are already available in practice based on individual pt genetic profiles.

Table 1: Mutation Events Summary
Putative Pathway	Frequency (%)	Potential Therapy
DNA Damage and Cell Cycle Control	32 (34)
ATM TP53 XPO1 STAG2	19 (20.2) 7 (7.4) 5 (5.3) 1 (1.1)	PARP inhibitors Selective inhibitors of Nuclear Export
RNA Processing	20 (21.3)
SF3B1 XPO1 TBL1XR1 PRPF40B ZRSR2	11 (11.7) 5 (5.3) 2 (2.1) 1 (1.1) 1 (1.1)
Epigenetic modification	11 (11.7)
DNMT3A TET2	7 (7.4) 4 (4.3)	DNA methyltransferase inhibitors
RAS-RAF-MEK-MAPK	10 (10.6)
BRAF KRAS NRAS NF1	5 (5.3) 2 (2.1) 2 (2.1) 1 (1.1)	BRAF inhibitors RAS/MEK inhibitors RAF/MEK inhibitors
Transcriptional regulation activity	10 (10.6)
BCOR PHF6 TBL1XR1 ASXL1	4 (4.3) 2 (2.1) 2 (2.1) 2 (2.1)
Notch Signaling	9 (9.6)
Notch1	9 (9.6)	Notch inhibitors
Inflammatory Pathways	3 (3.2)
MYD88 BIRC3	1 (1.1) 2 (2.1)	B cell receptor signal transduction inhibitors
Cellular metabolism	2 (2.2)
IDH1 IDH2	1 (1.1) 1 (1.1)	IDH inhibitors
Telomere maintenance	2 (2.1)
POT1	2 (2.1)
Chromatin modification	2 (2.1)
ZMYM3	2 (2.1)