In This Issue

See article, p. 42.

KRAS- and BRAF-mutant colorectal cancers (CRC) are refractory to treatments targeting the EGF receptor due to persistent MEK activation, underscoring the need to identify alternative or combinatorial therapeutic strategies for this subset of patients. Using data from a high-throughput drug screen, Faber and colleagues found that silencing of the antiapoptotic protein MCL-1 sensitized KRAS- and BRAF-mutant but not wild-type CRC cells to the BCL-2/BCL-XL inhibitor ABT-263. Consistent with this finding, treatment with AZD8055, an inhibitor of mTOR complexes 1 and 2 (mTORC1/2), which are required for capdependent translation of MCL-1 in some cancers, suppressed MCL-1 protein expression and cooperated with ABT-263 to induce both cell-cycle arrest and apoptosis specifically in KRAS- and BRAF-mutant CRC cell lines. This genotype-selective sensitivity was independent of MEK signaling and mediated by disruption of BIM–MCL-1 complexes following AZD8055-induced MCL-1 downregulation, thereby facilitating apoptosis in KRAS-mutant cells. Furthermore, dual treatment with ABT-263 and AZD8055 preferentially impaired tumor growth and induced tumor regression in KRAS-mutant CRC xenograft and genetically engineered mouse models. Intriguingly, although KRAS-mutant non–small cell lung cancer (NSCLC) cells were similarly sensitive to combined inhibition of BCL-2, BCL-XL, and MCL-1, AZD8055 did not diminish MCL-1 expression in NSCLC cells, suggesting that MCL-1 is broadly required for KRAS-mutant cancer cell survival but is regulated by distinct mechanisms in NSCLC. These results support further clinical development of this therapeutic combination specifically for patients with KRAS- and BRAF-mutant CRC.

See article, p. 53.

Mutation or deletion of the von Hippel Lindau (VHL) tumor suppressor gene results in stabilization of the hypoxia-inducible factor (HIF) transcription factors in clear cell renal cell carcinoma (ccRCC). Although HIF1α has been implicated as a tumor suppressor in ccRCC, it is coexpressed with the oncoprotein HIF2α in a large subset of ccRCCs, and the molecular mechanisms that overcome the inhibitory effect of HIF1α in these tumors remain unclear. Mathew and colleagues found that the locus encoding the microRNAs (miRNA, miR) miR-30c-2-3p and miR-30a-3p was selectively repressed in ccRCCs expressing both HIF1α and HIF2α compared with ccRCCs expressing only HIF2α. Repression of these miRNAs was pVHL-dependent but HIF-independent and resulted in increased HIF2α levels in HIF1α/HIF2α–positive ccRCC cell lines and human tumor samples, as miR-30c-2-3p and miR-30a-3p directly bound and inhibited the expression of HIF2A (also known as EPAS1) transcripts. miR-30c-2-3p– and miR-30a-3p–mediated suppression of HIF2α impaired xenograft tumor formation, whereas inhibition of miR-30a-3p induced HIF2α expression and augmented tumor growth and angiogenesis. Moreover, decreased expression of miR-30c-2-3p and miR-30a-3p was correlated with poor prognosis in patients with ccRCC, further supporting the importance of HIF2α in tumorigenesis. These findings suggest that downregulation of miR-30c-2-3p and miR-30a-3p in ccRCC enhances HIF2α expression as a means to compensate for HIF1α-mediated tumor suppression and promote tumor progression.

See article, p. 61.

RAF and MEK inhibitors have each shown activity in patients with BRAF^V600E-mutant melanoma, but resistance often arises through MAPK pathway reactivation. Compared with monotherapy, combined RAF and MEK inhibition has improved clinical outcomes, but resistance still develops. To gain insight into mechanisms of acquired resistance to combined RAF/MEK inhibition, Wagle and colleagues performed whole-exome and whole-transcriptome sequencing of BRAF-mutant melanomas before treatment with dabrafenib (a RAF inhibitor) plus trametinib (a MEK inhibitor) and after disease progression. Although combination therapy was expected to enhance MAPK suppression and thus prevent resistance caused by MAPK pathway reactivation, two resistant tumors acquired MAPK pathway alterations (BRAF alternative splicing and BRAF amplification) previously associated with resistance to RAF or MEK inhibitor monotherapy. A third tumor acquired a previously uncharacterized MEK2^Q60P mutation that conferred resistance to singleagent or combined dabrafenib and trametinib when expressed in a melanoma cell line. Of note, MEK2^Q60P-expressing cells retained sensitivity to downstream ERK inhibition, suggesting that ERK inhibitors may be useful in overcoming resistance to combined RAF/MEK inhibition. Cross-referencing of acquired mutations with in vitro RAF/MEK inhibitor resistance data also implicated mutations in ETS2 and SAMD4B as potential resistance mechanisms, but no acquired mutations in the remaining 2 patients affected known resistance-associated genes. Together, these findings establish that mutations arising in the context of single-agent targeted therapy also confer resistance to combination therapy and underscore the need for serial biopsies and molecular profiling to identify unanticipated resistance mechanisms.

See article, p. 69.

Genetic alterations such as those that reactivate the MAPK pathway have been shown to promote acquired resistance to BRAF inhibitors in patients with BRAF-mutant melanoma. In addition, adaptive tumor responses to BRAF inhibition likely also contribute to resistance; however, the identity of these signals and their role in the selection of late, acquired resistance mutations remain unknown. Shi and colleagues found that AKT activation was frequently increased in melanoma biopsies and BRAF-mutant melanoma cell lines as an early response to treatment with MAPK inhibitors. This rebound in AKT phosphorylation was mediated by upregulation of receptor tyrosine kinases (RTK), including platelet-derived growth factor β (PDGFRβ), accumulation of phosphatidylinositol (3,4,5)-trisphosphate (PIP₃), and enhanced pleckstrin homology domain (PHD) recruitment to the membrane, and was suppressed by PTEN expression. In contrast, the gain-of-function AKT1^Q79K PHD mutant, which exhibited increased cell-surface recruitment, rescued BRAF inhibitor–induced AKT phosphorylation and robustly induced vemurafenib resistance in PTEN-expressing cells, indicating that AKT1^Q79K amplifies BRAF inhibitor–driven adaptive AKT signaling in a cell context–dependent manner. Furthermore, PTEN-deficient, BRAF-mutant cells were intrinsically sensitive to AKT inhibition alone, and dual blockade of BRAF and AKT decreased the growth and survival of PTEN-expressing melanoma cells, supporting the use of this therapeutic combination to overcome MAPK inhibitor resistance. Together, these results suggest that BRAF inhibitor–induced early, adaptive AKT signaling shapes the subsequent selection of AKT-upregulating alterations that promote late, acquired resistance.

See article, p. 80.

The clinical efficacy of BRAF inhibitors in patients with BRAF^V600E/K-mutant melanoma is limited by acquisition of drug-resistance mechanisms that facilitate disease progression. Reactivation of MAPK signaling has been implicated as one such mechanism of acquired resistance. However, the relative contributions of MAPK-dependent and -independent mechanisms and the role of BRAF inhibitor–driven genomic hetero geneity in acquired resistance remain poorly understood. Using whole-exome sequencing, Shi and colleagues detected MAPK-reactivating alterations, including mutations in RAS, MEK1, and CDKN2A and amplification or alternative splicing of mutant BRAF, in 70% of BRAF inhibitor–resistant melanoma samples, consistent with a primary role for this pathway in acquired resistance. In addition, mutations that activated phosphoi-nositide 3-kinase (PI3K)–AKT signaling were detected in 22% of disease-progressive tumors, identifying dysregulation of this pathway as a second core mechanism of acquired resistance. Structure modeling predicted that these genetic lesions included gain-of-function mutations in AKT1, AKT3, and PIK3CA and loss-of-function mutations in PIK3R2 and PTEN, and overexpression of PI3K–AKT-upregulating mutations was sufficient to confer vemurafenib resistance. Multiple resistance mechanisms and concomitant mutations in both core pathways were frequently observed within the same patient or the same tumor. Furthermore, tumor heterogeneity among progressive tumors was associated with branched clonal evolution, increased tumor-cell fitness, and altered mutation spectra distinct from UV-induced mutagenesis. These findings define critical mechanisms underlying acquired resistance in BRAF-mutant melanoma and suggest that early targeting of both core pathways may improve treatment outcomes.

See article, p. 94.

Intrinsic or acquired resistance to RAF inhibitors such as vemurafenib or dabrafenib remains a challenge in treating patients with BRAF^V600E-mutant metastatic melanoma. Most known RAF inhibitor resistance mechanisms have been identified preclinically before confirmation in a small number of clinical samples, suggesting that the spectrum of mechanisms of clinical resistance to RAF inhibitors has been incompletely characterized. Van Allen and colleagues performed whole-exome sequencing of 45 formalin-fixed, paraffin-embedded BRAF^V600E-mutant metastatic melanomas before vemurafenib or dabrafenib monotherapy and after resistance developed. Resistance-associated genetic alterations predominantly affected MAPK pathway components and included previously identified, highly recurrent alterations, such as NRAS and MEK1 mutations or BRAF amplification, as well as other less common, previously uncharacterized events such as MEK2 mutation and MITF amplification. MEK2 mutations conferred resistance to both RAF and MEK inhibitors and led to sustained MEK and ERK phosphorylation in vitro, though melanoma cells harboring these mutations retained sensitivity to ERK inhibition. However, overexpression of MITF, encoding a MAPK-regulated melanocyte lineage transcription factor, in BRAF^V600E-mutant melanoma cells conferred cross-resistance to RAF, MEK, and ERK inhibitors, suggesting that in rare cases resistance can arise as a result of restoration of transcriptional output downstream of MAPK signaling. In addition to demonstrating the feasibility of using archival tumor material for clinical resistance studies, these results expand our understanding of RAF inhibitor resistance and may guide use of combination or sequential therapies.

See article, p. 110.

Patients with chronic inflammatory or autoimmune diseases have an increased risk of lymphoid malignancies, but the mechanisms that drive oncogenic transformation in these settings are unknown. Immune responses induce modifications of the lymphoid tissue stromal architecture that lead to changes in the number, distribution, and activity of immune cells, raising the possibility that altered stromal remodeling might promote malignant transformation under inflammatory or autoimmune conditions in which lymphoproliferation is increased. To test this hypothesis, Sangaletti and colleagues crossed autoimmunity-prone Fas-mutant mice with mice lacking secreted protein acidic and rich in cysteine (SPARC, also known as osteonectin), a regulator of stromal remodeling that mediates interactions between cellular stromal components and the extracellular matrix. Loss of SPARC led to disrupted splenic architecture in association with exacerbated autoimmunity and malignant proliferation of B cells. In the absence of SPARC, stromal disorganization and decompartmentalization in the spleen and reduced immune inhibitory signals from the extracellular matrix led to unrestricted, noncanonical interactions between B cells and neutrophils, with extrusion of neutrophil extracellular traps (NET) and NETotic cell death stimulating inflammation and B-cell proliferation. In humans, SPARC downregulation was observed in some non-Hodgkin B-cell lymphomas, particularly chronic lymphocytic lymphoma (CLL), and was also accompanied by disrupted lymphoid tissue architecture and altered expression of genes associated with aberrant immune responses. Together, these findings support a link between alterations in stromal matricellular components and lymphomagenesis in the setting of chronic inflammation or autoimmunity.

Note: In This Issue is written by Cancer Discovery Science Writers. Readers are encouraged to consult the original articles for full details.