In This Issue

See article by Catenacci et al., p. 573.

Monoclonal antibodies targeting receptor tyrosine kinases (RTK) represent an attractive therapeutic approach for cancers addicted to oncogenic RTK signaling pathways. Catenacci and colleagues report a case of a patient with heavily pretreated, progressive, metastatic gastric cancer who was enrolled in a phase I study testing a monoclonal antibody against the RTK MET (MetMAb). The patient was the only one in the trial to respond to therapy, with a complete response (CR) after 4 MetMAb doses lasting 2 years, at which point the cancer recurred. After 3 additional doses of MetMAb therapy upon recurrence, widespread peritoneal progression had developed, despite a partial response in the 2 initial recurrent lesions. To better understand why this particular patient had such a dramatic, durable initial response, the authors analyzed tumor MET gene copy, MET and MET ligand [hepatocyte growth factor (HGF)] expression, and serum HGF levels. They found that prior to MetMAb treatment, the patient had one of the highest baseline plasma HGF levels, and the primary tumor exhibited MET polysomy (≥ 4 copies/cell). Furthermore, MET and HGF expression was high in the primary tumor and increased with histologic progression. Immediately following initial MetMAb treatment, HGF serum levels dropped precipitously and remained low. Upon recurrence, MET expression increased slightly without a concomitant increase in MET gene copy number or serum HGF, suggesting the tumor had become HGF independent and resistant to MetMAb therapy. Together, these results demonstrate the promise of MetMAb in MET-addicted tumors, and identify predictors of response to treatment.

See article by Pecot et al., p. 580.

Most current methods to capture circulating tumor cells (CTC) detect epithelial markers such as EpCAM or cytokeratin (CK) and therefore miss tumor cells that are mesenchymally derived or have undergone an epithelial–mesenchymal transition (EMT). Pecot and colleagues have developed a platform for CTC detection that aims to increase capture efficiency by enriching for more cell populations. A cocktail of antibodies against an expanded panel of epithelial, mesenchymal, and stem cell antigens is first added directly to blood to facilitate recovery of multiple CTC phenotypes. The microfluidic-based system then captures rare antibody-bound cells from blood and allows subsequent molecular analysis of captured cells directly within a microchannel without requiring further cell manipulation. This approach improved recovery of CTCs in patients with advanced lung, breast, colorectal, and prostate cancer and was more sensitive than the only U.S. Food and Drug Administration–approved technique for CTC detection. Furthermore, FISH analysis demonstrated that CK⁻ CTCs exhibited genetic characteristics of primary tumors in human patients and that the total number of CK⁻ cells correlated with the aggregate tumor burden in a mouse model of metastasis. Additional studies using TGF-β to induce EMT indicated that CTCs can lose CK staining and thus have a partial EMT phenotype. This study demonstrates the inefficiency of CK as a marker to identify all potential CTC populations and presents a versatile platform that can potentially be adapted to other antibodies and single-cell microscopic analyses.

See article by Taylor et al., p. 587.

Dedifferentiated liposarcoma (DLPS) is an aggressive subtype of liposarcoma, the most common adult soft-tissue sarcoma. Because DLPS has a poor prognosis and is resistant to conventional cytotoxic therapies, Taylor and colleagues sought to better understand DLPS etiology and identify potential therapeutic targets. Genomic sequencing identified an average of ∼450 chromosomal rearrangements per tumor (particularly involving chromosome 12q) largely caused by genomic amplification at breakpoints. Further, exome sequencing revealed mutations in histone deacetylase 1 (HDAC1), which were ultimately identified in 8.3% of DLPS samples. Consistent with this finding implicating epigenetic abnormalities in the development of a subset of DLPS, the authors identified differentially methylated regions in DLPS compared to normal adipose tissue correlating with significant changes in gene expression. Notably, multiple adipocyte differentiation genes and several microRNAs (miRNA) were specifically hypermethylated and downregulated in DLPS. To determine whether epigenetic therapies could reverse the suppression of these genes and block tumor growth, the authors treated DLPS cells with either the DNA methyltransferase inhibitor 5-Aza-2′-deoxycytidine (5-aza), the HDAC inhibitor suberoylanilide hydroxamic acid (SAHA), or both. SAHA and 5-aza acted synergistically to restore expression of adipocyte differentiation markers, suppress cellular proliferation, induce apoptosis, and inhibit growth of DLPS xenografts. Taken together, these findings implicate epigenetic deregulation of adipocyte differentiation in the development of DLPS and provide a rationale for the use of epigenetic therapies in treatment of this aggressive liposarcoma subtype.

See article by Juergens et al., p. 598.

DNA methyltransferase and histone deacetylase (HDAC) inhibitors approved for use in several hematologic malignancies have so far each resulted in toxicity and minimal efficacy in solid tumors. Juergens and colleagues report the results of a phase I/II clinical trial of combined low-dose azacitidine (a DNA methyltransferase inhibitor) and entinostat (an HDAC inhibitor) therapy in patients with refractory metastatic non–small cell lung cancer (NSCLC). Adverse events were medically manageable, and of the patients who completed at least 2 cycles of therapy, 12 of 34 (35%) had either an objective response or sustained disease stabilization. One patient had a complete response lasting 14 months before fatal progression of a second, molecularly distinct primary NSCLC, and another experienced complete resolution of liver metastases and stable disease approximately 2 years after completing epigenetic therapy. The authors observed that patients with a decreased methylation signature in response to azacitidine and entinostat were more likely to have stable disease or objective responses and had significantly longer overall and progression-free survival, consistent with an on-target epigenetic mechanism of action. Additionally, of the patients who progressed while on epigenetic therapy, 21% had objective responses to subsequent therapy, including 2 long-term survivors (44 and 52 months), suggesting epigenetic therapy may also prime some cancers to respond to cytotoxic therapies. These results indicate combination therapy may be effective in reversing epigenetic mechanisms driving a subset of NSCLCs.

See article by Cheung et al., p. 608.

Focal amplification of chromosome 22q11.21 has been identified in 3% of non–small cell lung cancers (NSCLC), but the genes in this region responsible for driving oncogenic transformation have not been identified. Cheung and colleagues now demonstrate that overexpression of a gene in this amplicon, CRKL [v-crk sarcoma virus CT10 oncogene homolog (avian)-like], is required for the survival and tumorgenicity of NSCLC cells with 22q11.21 amplification. CRKL overexpression also induces anchorage-independent growth in human airway epithelial cells, indicating that CRKL plays a role in oncogenic transformation in the lung. Because CRKL is an adaptor protein with many interacting partners and wide-ranging roles in signal transduction, the authors sought to characterize the cellular consequences of CRKL amplification. Analyses of tyrosine kinase phosphorylation and interactions with known CRKL-binding proteins demonstrated that CRKL overexpression leads to concurrent activation of the SRC-C3G-RAP1, SOS1-RAS-RAF, and PI3K-AKT signaling pathways, which are required for the growth of CRKL-amplified NSCLC cells. Persistent CRKL-induced signaling also contributes to NSCLC resistance to EGFR inhibitor therapy, as CRKL overexpression promoted the proliferation and survival of NSCLC cells in response to gefitinib or dasatinib. FISH analysis in EGFR inhibitor-resistant NSCLC tumors identified a de novo CRKL amplification in one patient that was not present prior to EGFR inhibitor treatment, implicating CRKL amplification in acquired EGFR inhibitor resistance. Together, these results suggest that CRKL may represent a therapeutic target in a subset of NSCLCs.

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