Utilization of Next Generation Sequencing Identifies Potentially Actionable Mutations in Chronic Lymphocytic Leukaemia
Jie Wang,Jennifer D. Morrissette,David Lieberman,Colleen Timlin,Stephen J. Schuster,Anthony R. Mato
DOI: https://doi.org/10.1111/bjh.14315
2016-01-01
British Journal of Haematology
Abstract:Chronic lymphocytic leukaemia (CLL) is characterized by extensive genetic heterogeneity. Mutations in TP53 are associated with refractoriness to chemoimmunotherapy. Although new therapies have improved clinical outcomes in this group of patients, many patients eventually relapse. Several studies using high throughput sequencing have identified novel genetic mutations with prognostic significance in CLL (Wang et al, 2011). However, there has not been a systematic effort to identify high priority, targetable mutations for which pharmaceuticals already exist for other indications, or for which pharmaceuticals are already in development. We performed a retrospective analysis of CLL patients who underwent next generation sequencing (NGS) of tumour samples obtained from peripheral blood and bone marrow sources. We describe our findings in which we identify potentially targetable somatic mutations that may inform future clinical trial design. In particular, our results highlight the occurrence of mutations in BRAF as a potential target in CLL management. A total of 122 sequential patients with CLL who underwent NGS were identified from our clinical database. NGS was performed on an Illumina MiSeq (Illumina, Inc., San Diego, CA, USA) using a custom 68-gene amplicon-based panel developed at our centre with a detection limit of 5% allele frequency and a minimum depth of coverage of 250×. We used a custom bioinformatics pipeline, combining open source tools and custom algorithms for analysis (Daber et al, 2013; Sehgal et al, 2015). All mutational events in the coding regions were classified as pathogenic, benign polymorphisms or variants of uncertain significance. We defined pathogenic mutations as those that alter protein function in a way that drives tumourigenesis with functional evidence showing that the mutation is associated with disease. Benign polymorphisms are defined as those that have an allele frequency >0.1% in the general population as reported in the Exome Variant Server (EVS) and Exome Aggregation Consortium (ExAC) databases (http://evs.gs.washington.edu/EVS/ http://exac.broadinstitute.org/ ). Variants of uncertain significance were defined as those mutational events that did not meet the definition of pathogenic or benign polymorphism. Seventy-two patients (59%) had at least one pathogenic mutation (range of 1–8 mutations). Pathogenic mutations were found in 27 genes, with the top five (TP53, SF3B1, ATM, NOTCH1 and BRAF) affecting 62 patients and these represented 64% of the pathogenic mutational events. Sixty-two patients were found to have pathogenic mutations in genes for which targeted therapeutic agents exist, i.e. those having shown efficacy, either in preclinical in vivo models or in the clinical setting. Pathogenic mutations in NOTCH1, ATM, BRAF and XPO1 were the most frequently implicated targetable mutations in our cohort (Fig 1). Among the most common mutations found in our cohort, mutations in BRAF are the most readily targeted given currently available pharmaceuticals. In contrast to the limited experience in targeting malignancies with NOTCH1, ATM and XPO1 mutations, BRAF inhibition has not only transformed the treatment of BRAF-mutated melanoma, but has also been successfully employed in other malignancies. Both the frequency of its occurrence in CLL, and efficacy in different malignancies supports the rationale for clinical trial design in BRAF inhibition in CLL. Among the seven patients affected, four different amino acid changes were identified in BRAF: V600E, K601E, D594G and D594N. The BRAF V600E mutation is well documented to have oncogenic activity in papillary thyroid cancers, melanomas and hairy cell leukaemia (Li et al, 2009; Wellbrock & Hurlstone, 2010), and it results in the constitutive activation of BRAF kinase activity. BRAF V600E is found in nearly all cases of hairy cell leukaemia (HCL), and the BRAF inhibitor vemurafenib has already been shown to be clinically active in relapsed-refractory HCL (Dietrich et al, 2012; Samuel et al, 2014). The BRAF K601E mutation, like V600E, is also an activating mutation associated with the RAF/MEK/Erk pathway. Preclinical and clinical evidence suggest that BRAF inhibition with vemurafenib and MEK inhibition with trametinib can block downstream signalling induced by the K601E mutation (Dahlman et al, 2012; Kim et al, 2013). The remaining BRAF mutations are both D594 inactivating mutations, against which BRAF inhibitors are not expected to be effective. The mechanism through which BRAF D594 inactivating mutations may contribute to oncogenesis has been elucidated (Heidorn et al, 2010). In the presence of activated RAS, inactivated BRAF can form a complex with CRAF, resulting in the hyperactivation of MEK. Choosing inhibitors of the downstream targets RAS and MEK may be more appropriate. In the absence of concomitant RAS activation, inactivated BRAF may still drive oncogenesis through its interaction with CRAF and may be amenable to treatment with pan-RAF inhibition, such as with sorafenib. In our cohort, 6 of 7 patients with a pathogenic mutation in BRAF had co-occurring mutations. These included mutations in ATM, TP53, XPO1, KRAS, NRAS, NOTCH1, NF1, TET2, BCOR, IDH2, SF3B1 and BIRC3. One patient had concomitant pathogenic mutations in BRAF and NRAS, which is notable because in melanoma, NRAS and BRAF mutations are mutually exclusive, whereas mutations in BRAF and PTEN frequently co-occur (Wellbrock & Hurlstone, 2010). The former is reflective of the fact that NRAS is upstream of BRAF, and the latter illustrates the interaction between the two distinct RAS signalling streams involving the ERK/MAPK and the PI3K/PTEN/AKT pathways. Just as an NRAS mutation activates both signalling streams driving cellular proliferation, concurrent activation of BRAF with loss of PTEN cooperate to effect signalling in both pathways. The co-occurrence of BRAF and NRAS mutations in our patient means that both RAS signalling pathways are affected. This finding highlights the importance of looking at co-occurring mutations prior to selecting targeted pharmaceuticals. Given the genetic heterogeneity of CLL and our characterization of targetable mutations, we believe that mutation-specific clinical trial design in CLL may yield invaluable insight into alternative therapeutic approaches, particularly in patients who have failed chemoimmunotherapy combinations and novel kinase inhibitors. The National Cancer Institute-Molecular Analysis for Therapy Choice (NCI-MATCH) clinical trial (NCT02465060) seeks to assign treatment to patients based on tumour molecular abnormalities rather than specific disease type. Given our findings and the diversity of pathways involved in CLL biology, we believe that CLL is a candidate disease entity for the design of clinical trials using a precision medicine approach that selects therapies based on individual patient genetic profiles. J.W. Performed research, analyzed data, wrote the paper. J.M. Performed NGS, analyzed data, edited the paper. D.B.L. Performed NGS, analyzed data, edited the paper. C.T. Analyzed data, wrote the paper. S.J.S. Analyzed data, wrote the paper. A.R.M. Designed research, performed research, analyzed data, wrote the paper. Wang: No relevant conflicts of interest. Morrissette: No relevant conflicts of interest. Lieberman: No relevant conflicts of interest. Timlin: No relevant conflicts of interest. Schuster: Nordic Nanovector: Membership on an entity's Board of Directors or advisory committees; Janssen: Research Funding; Novartis: Research Funding; Genentech: Consultancy; Gilead: Research Funding; Hoffman-LaRoche: Research Funding; Celgene: Consultancy, Research Funding; Phamacyclics: Consultancy, Research Funding. Mato: Gilead: Consultancy, Research Funding; AbbVie: Consultancy, Research Funding; Pronai Pharmaceuticals: Research Funding; Pharmacyclics: Consultancy, Research Funding; Janssen: Consultancy; Celgene Corporation: Consultancy, TG Therapeutics: Research Funding; Acerta: Research Funding.