FBXW7 is a biologically validated cancer driver gene for CLL

F-box and WD40 repeat domain containing-7 (FBXW7) is one of the statistically defined candidate cancer driver genes for chronic lymphocytic leukemia (CLL). In this issue of Blood, Close et al describe functional genomic experiments that have verified the biological consequences of mutations in FBXW7, which can now be considered a validated driver gene for CLL.¹ 

The mere presence of somatic mutations is insufficient to implicate a gene in cancer. For example, when analytical methods are applied to whole-exome sequencing data from a given cancer, almost a quarter of the somatically mutated genes encode proteins that are irrelevant for tumor initiation or progression, such as olfactory receptors, cardiac ryanodine receptors (RyR2, RyR3), cytoskeletal dyneins (DNAH5, DNAH11), or neuronal synaptic vesicle proteins (PCLO).² 

Cancer geneticists and bioinformaticians use the term “passenger” to describe such mutations that were likely being randomly acquired to distinguish them from mutations targeting candidate cancer driver genes that were implicated in the tumor biology. In the absence of biological data, a statistical definition of candidate cancer driver genes was created and then used by cancer geneticists and bioinformaticians. Any given gene was labeled as a candidate cancer driver if it harbored somatic point mutations (that is, substitutions and small insertion or deletions) at a statistically significant rate or pattern in cancer tumor samples. Three independent statistical signals are used to separate candidate cancer driver genes from other genes: (1) they have a high mutational burden relative to background expectation, (2) there is a clustering of mutations within the gene, and (3) mutations are enriched in the evolutionarily conserved amino acids of the corresponding proteins.³ 

In CLL, more than 40 genes fulfill the statistical definition of candidate cancer driver, but only a few of them are biologically validated as oncogenes or tumor-suppressor genes (ie, SF3B1, NOTCH1, TP53, ATM).⁴ 

FBXW7 encodes an E3 ubiquitin ligase whose WD40 domain interacts with proteins that are subsequently subject to proteasomal degradation (see figure panel A). Among FBXW7 targets for proteasomal degradation, there are a number of known proto-oncogenes, such as cleaved NOTCH1 (NOTCH1-NOTCH1 intracellular domain [NICD]), MYC, and NF-κB2.⁵  In CLL, FBXW7 mutations occur at a frequency of 2% to 6% and commonly affect the WD40 substrate binding domain, including hotspot residues that are required for substrate recognition. Close et al identify cleaved NOTCH1 (NICD) as the target of FBXW7 proteasomal degradation activity in CLL cells and prove that FBXW7 mutations in CLL correlate with increased NICD protein and increased NOTCH1 target gene expression (see figure panel B).

Most patients with CLL express NICD, a transcriptional coactivator of oncogenic programs that are generated from the NOTCH1 receptor upon its activation.⁶  In CLL, NICD accumulates in the nucleus as a consequence of frameshift or nonsense mutations targeting exon 34, and it is selected to disrupt the proline, glutamic acid, serine, and threonine (PEST) domain of the protein.⁷  Under physiologic conditions, the PEST domain of NICD is required to limit the intensity and duration of NOTCH1 signaling activation by tailoring NICD for proteasomal degradation. A proportion of CLL patients show NICD accumulation despite the lack of NOTCH1 exon 34 mutations.⁶  A fraction of such patients with stable NICD expression can be distinguished by the occurrence of mutations in the noncoding 3′-untranslated region (3′UTR) of NOTCH1.⁸  Noncoding mutations in the 3′-UTR of NOTCH1 lead to a novel splicing event between a cryptic donor site located in the coding region of NOTCH1 and a newly created acceptor site in the 3′-UTR, which results in a deletion that removes the PEST domain and increases protein stability, as in the previously described NOTCH1 mutation affecting exon 34.⁸ 

The article by Close et al helps expand our understanding of the genetic mechanisms that lead to upregulated NICD in CLL. By disrupting the function of the FBXW7 ubiquitin ligase, FBXW7 mutations prevent NICD proteasomal degradation, thus mimicking the consequence of NOTCH1 mutations in stabilizing NICD. Because both NOTCH1 and FBXW7 mutations point to the same biological consequences, namely stabilization of NICD, they are generally mutually exclusive in CLL. Besides mimicking the biological effect of NOTCH1 mutations, FBXW7 mutations also share the same clinical consequences, including co-occurrence with trisomy 12 and inferior survival, a resemblance that further underlines their functional similarity.^6-8 

Among CLL patients who harbor NOTCH1 mutations, treatment with chemotherapy regimens containing rituximab does not result in the expected increase in progression-free survival compared with treatment with chemotherapy regimens that do not contain rituximab; this points to NOTCH1 mutations as a biomarker of resistance to the anti-CD20 antibody rituximab.⁹  Given the biological and clinical similarities between NOTCH1 and FBXW7 mutations, the findings of Close et al may also have clinical translations, which will prompt the assessment of FBXW7 mutations as another biomarker of anti-CD20 resistance in CLL.