Gene Expression Ratios Lead to Accurate and Translatable Predictors of DR5 Agonism across Multiple Tumor Lineages

Death Receptor 5 (DR5) agonists demonstrate anti-tumor activity in preclinical models but have yet to demonstrate robust clinical responses. A key limitation may be the lack of patient selection strategies to identify those most likely to respond to treatment. To overcome this limitation, we screened a DR5 agonist Nanobody across >600 cell lines representing 21 tumor lineages and assessed molecular features associated with response. High expression of DR5 and Casp8 were significantly associated with sensitivity, but their expression thresholds were difficult to translate due to low dynamic ranges. To address the translational challenge of establishing thresholds of gene expression, we developed a classifier based on ratios of genes that predicted response across lineages. The ratio classifier outperformed the DR5+Casp8 classifier, as well as standard approaches for feature selection and classification using genes, instead of ratios. This classifier was independently validated using 11 primary patient-derived pancreatic xenograft models showing perfect predictions as well as a striking linearity between prediction probability and anti-tumor response. A network analysis of the genes in the ratio classifier captured important biological relationships mediating drug response, specifically identifying key positive and negative regulators of DR5 mediated apoptosis, including DR5, CASP8, BID, cFLIP, XIAP and PEA15. Importantly, the ratio classifier shows translatability across gene expression platforms (from Affymetrix microarrays to RNA-seq) and across model systems (in vitro to in vivo). Our approach of using gene expression ratios presents a robust and novel method for constructing translatable biomarkers of compound response, which can also probe the underlying biology of treatment response.

FGFR-Mediated Reactivation of MAPK Signaling Attenuates Antitumor Effects of Imatinib in Gastrointestinal Stromal Tumors

Activating mutations in either KIT or PDGFRA are present in approximately 90% of gastrointestinal stromal tumors (GIST). Although treatment with the KIT and PDGFR inhibitor imatinib can control advanced disease in about 80% of GIST patients, the beneficial effect is not durable. Here, we report that ligands from the FGF family reduced the effectiveness of imatinib in GIST cells, and FGF2 and FGFR1 are highly expressed in all primary GIST samples examined. The combination of KIT and FGFR inhibition showed increased growth inhibition in imatinib-sensitive GIST cell lines and improved efficacy in patient-derived GIST xenografts. In addition, inhibition of MAPK signaling by imatinib was not sustained in GIST cells. An ERK rebound occurred through activation of FGF signaling, and was repressed by FGFR1 inhibition. Downregulation of Sprouty proteins played a role in the imatinib-induced feedback activation of FGF signaling in GIST cells.

SIGNIFICANCE

We here show that FGFR-mediated reactivation of the MAPK pathway attenuates the antiproliferation effects of imatinib in GISTs. The imatinib-induced ERK rebound can be repressed by the FGFR inhibitor BGJ398, and combined KIT and FGFR inhibition leads to increased efficacy in vitro and in patient-derived xenografts.

An F876L mutation in androgen receptor confers genetic and phenotypic resistance to MDV3100 (enzalutamide)

Castration-resistant prostate cancer (CRPC) is the most aggressive, incurable form of prostate cancer. MDV3100 (enzalutamide), an antagonist of the androgen receptor (AR), was approved for clinical use in men with metastatic CRPC. Although this compound showed clinical efficacy, many initial responders later developed resistance. To uncover relevant resistant mechanisms, we developed a model of spontaneous resistance to MDV3100 in LNCaP prostate cancer cells. Detailed characterization revealed that emergence of an F876L mutation in AR correlated with blunted AR response to MDV3100 and sustained proliferation during treatment. Functional studies confirmed that AR(F876L) confers an antagonist-to-agonist switch that drives phenotypic resistance. Finally, treatment with distinct antiandrogens or cyclin-dependent kinase (CDK)4/6 inhibitors effectively antagonized AR(F876L) function. Together, these findings suggest that emergence of F876L may (i) serve as a novel biomarker for prediction of drug sensitivity, (ii) predict a "withdrawal" response to MDV3100, and (iii) be suitably targeted with other antiandrogens or CDK4/6 inhibitors.

SIGNIFICANCE

We uncovered an F876L agonist-switch mutation in AR that confers genetic and phenotypic resistance to the antiandrogen drug MDV3100. On the basis of this fi nding, we propose new therapeutic strategies to treat patients with prostate cancer presenting with this AR mutation.

Abstract 1620: Dovitinib has anti-tumor activity in gastrointestinal stromal tumor (GIST) cell lines

Background: Gastrointestinal stromal tumors (GISTs) are the most common gastrointestinal tract sarcoma, with 5000-6000 cases per year diagnosed in the United States. Most GISTs carry activating mutations in KIT (≈ 75% of cases) or PDGFR alpha (10%-15%). Imatinib, a KIT/PDGFR inhibitor, is the frontline treatment for GIST, although secondary mutations commonly lead to acquired resistance. Dovitinib inhibits several kinase targets important in the development and maintenance of GISTs, including KIT, PDGFR, VEGFR, and FGFR. This study was designed to evaluate the single-agent activities of imatinib and dovitinib in a panel of GIST cell lines and an imatinib-sensitive mouse model.

Methods: In vitro sensitivity was examined in both imatinib-sensitive (GIST-T1 and GIST882) and -insensitive (GIST430 and GIST48) cell lines. KIT phosphorylation was measured by Western blot and sandwich ELISA. Imatinib and dovitinib monotherapy were evaluated for tumor growth inhibition and delay in a nu/nu mouse xenograft model using GIST-T1 xenografts. Dovitinib was administered by mouth at either 60 mg/kg daily for 14 days or 30 mg/kg daily for 21 days. Imatinib was given at 100 mg/kg by mouth twice daily. Following treatment, dosing was terminated, and tumors were allowed to regrow off treatment. Kaplan-Meier analysis was used to determine tumor growth delay, reported as median time to reach endpoint (tumor volume ≥ 1000 mm3).

Results: Imatinib and dovitinib inhibit KIT phosphorylation in GIST-T1,

GIST882, GIST430 and GIST48 cells, although imatinib appeared to be more potent. Dovitinib and imatinib both potently inhibited the proliferation, GI50 10-100 nM, of imatinib-sensitive GIST-T1 and GIST882 cell lines; GI50 was > 500 nM in imatinib-insensitive GIST48 and GIST430 cell lines. In vivo, imatinib and dovitinib were well tolerated for 14- and 21-day treatment cycles. In the 14-day treatment model, imatinib and dovitinib led to 52% and 59% tumor regression, respectively, and median time to regrowth was 69 days for both agents compared with 31 days for vehicle. In the 21-day treatment model, imatinib and dovitinib led to 66% and 36% tumor regression, respectively, and median time to regrowth was 81 and 69 days, respectively, compared with 27 days for vehicle.

Conclusions: Dovitinib inhibits KIT signaling and has similar growth-inhibition activity to imatinib in imatinib-sensitive GIST cell lines. In xenograft models, both imatinib and dovitinib are well tolerated, induce tumor regression, and delay tumor regrowth. The additional kinase inhibitory activities of dovitinib offer the possibility for enhanced or differentiated activity to imatinib in certain settings. Experiments are ongoing to explore this possibility. These data suggest that dovitinib may have therapeutic benefit in GIST, with efficacy resulting from a combination of the inhibition of KIT and other kinases.

Citation Format: Joseph D. Growney, Fang Li, Shumei Qiu, Bella Gorbatcheva, Linda Battalagine, Juergen Mestan, Paul Manley, Matthew Squires, Alex Cao, John E. Monahan. Dovitinib has anti-tumor activity in gastrointestinal stromal tumor (GIST) cell lines. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1620. doi:10.1158/1538-7445.AM2013-1620

FGFR genetic alterations predict for sensitivity to NVP-BGJ398, a selective pan-FGFR inhibitor

Patient stratification biomarkers that enable the translation of cancer genetic knowledge into clinical use are essential for the successful and rapid development of emerging targeted anticancer therapeutics. Here, we describe the identification of patient stratification biomarkers for NVP-BGJ398, a novel and selective fibroblast growth factor receptor (FGFR) inhibitor. By intersecting genome-wide gene expression and genomic alteration data with cell line-sensitivity data across an annotated collection of cancer cell lines called the Cancer Cell Line Encyclopedia, we show that genetic alterations for FGFR family members predict for sensitivity to NVP-BGJ398. For the first time, we report oncogenic FGFR1 amplification in osteosarcoma as a potential patient selection biomarker. Furthermore, we show that cancer cell lines harboring FGF19 copy number gain at the 11q13 amplicon are sensitive to NVP-BGJ398 only when concomitant expression of β-klotho occurs. Thus, our findings provide the rationale for the clinical development of FGFR inhibitors in selected patients with cancer harboring tumors with the identified predictors of sensitivity.

SIGNIFICANCE

The success of a personalized medicine approach using targeted therapies ultimately depends on being able to identify the patients who will benefit the most from any given drug. To this end, we have integrated the molecular profiles for more than 500 cancer cell lines with sensitivity data for the novel anticancer drug NVP-BGJ398 and showed that FGFR genetic alterations are the most significant predictors for sensitivity. This work has ultimately endorsed the incorporation of specific patient selection biomakers in the clinical trials for NVP-BGJ398.

The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity

The systematic translation of cancer genomic data into knowledge of tumour biology and therapeutic possibilities remains challenging. Such efforts should be greatly aided by robust preclinical model systems that reflect the genomic diversity of human cancers and for which detailed genetic and pharmacological annotation is available. Here we describe the Cancer Cell Line Encyclopedia (CCLE): a compilation of gene expression, chromosomal copy number and massively parallel sequencing data from 947 human cancer cell lines. When coupled with pharmacological profiles for 24 anticancer drugs across 479 of the cell lines, this collection allowed identification of genetic, lineage, and gene-expression-based predictors of drug sensitivity. In addition to known predictors, we found that plasma cell lineage correlated with sensitivity to IGF1 receptor inhibitors; AHR expression was associated with MEK inhibitor efficacy in NRAS-mutant lines; and SLFN11 expression predicted sensitivity to topoisomerase inhibitors. Together, our results indicate that large, annotated cell-line collections may help to enable preclinical stratification schemata for anticancer agents. The generation of genetic predictions of drug response in the preclinical setting and their incorporation into cancer clinical trial design could speed the emergence of 'personalized' therapeutic regimens.

The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity

The systematic translation of cancer genomic data into knowledge of tumour biology and therapeutic possibilities remains challenging. Such efforts should be greatly aided by robust preclinical model systems that reflect the genomic diversity of human cancers and for which detailed genetic and pharmacological annotation is available. Here we describe the Cancer Cell Line Encyclopedia (CCLE): a compilation of gene expression, chromosomal copy number and massively parallel sequencing data from 947 human cancer cell lines. When coupled with pharmacological profiles for 24 anticancer drugs across 479 of the cell lines, this collection allowed identification of genetic, lineage, and gene-expression-based predictors of drug sensitivity. In addition to known predictors, we found that plasma cell lineage correlated with sensitivity to IGF1 receptor inhibitors; AHR expression was associated with MEK inhibitor efficacy in NRAS-mutant lines; and SLFN11 expression predicted sensitivity to topoisomerase inhibitors. Together, our results indicate that large, annotated cell-line collections may help to enable preclinical stratification schemata for anticancer agents. The generation of genetic predictions of drug response in the preclinical setting and their incorporation into cancer clinical trial design could speed the emergence of 'personalized' therapeutic regimens.

A drug resistance screen using a selective MET inhibitor reveals a spectrum of mutations that partially overlap with activating mutations found in cancer patients

The emergence of drug resistance is a primary concern in any cancer treatment, including with targeted kinase inhibitors as exemplified by the appearance of Bcr-Abl point mutations in chronic myeloid leukemia (CML) patients treated with imatinib. In vitro approaches to identify resistance mutations in Bcr-Abl have yielded mutation spectra that faithfully recapitulated clinical observations. To predict resistance mutations in the receptor tyrosine kinase MET that could emerge during inhibitor treatment in patients, we conducted a resistance screen in BaF3 TPR-MET cells using the novel selective MET inhibitor NVP-BVU972. The observed spectrum of mutations in resistant cells was dominated by substitutions of tyrosine 1230 but also included other missense mutations and partially overlapped with activating MET mutations that were previously described in cancer patients. Cocrystallization of the MET kinase domain in complex with NVP-BVU972 revealed a key role for Y1230 in binding of NVP-BVU972, as previously reported for multiple other selective MET inhibitors. A second resistance screen in the same format with the MET inhibitor AMG 458 yielded a distinct spectrum of mutations rich in F1200 alterations, which is consistent with a different predicted binding mode. Our findings suggest that amino acid substitutions in the MET kinase domain of cancer patients need to be carefully monitored before and during treatment with MET inhibitors, as resistance may preexist or emerge. Compounds binding in the same manner as NVP-BVU972 might be particularly susceptible to the development of resistance through mutations in Y1230, a condition that may be addressed by MET inhibitors with alternative binding modes.

Interfering with resistance to smoothened antagonists by inhibition of the PI3K pathway in medulloblastoma

The malignant brain cancer medulloblastoma is characterized by mutations in Hedgehog (Hh) signaling pathway genes, which lead to constitutive activation of the G protein (heterotrimeric guanosine triphosphate-binding protein)-coupled receptor Smoothened (Smo). The Smo antagonist NVP-LDE225 inhibits Hh signaling and induces tumor regression in animal models of medulloblastoma. However, evidence of resistance was observed during the course of treatment. Molecular analysis of resistant tumors revealed several resistance mechanisms. We noted chromosomal amplification of Gli2, a downstream effector of Hh signaling, and, more rarely, point mutations in Smo that led to reactivated Hh signaling and restored tumor growth. Analysis of pathway gene expression signatures also, unexpectedly, identified up-regulation of phosphatidylinositol 3-kinase (PI3K) signaling in resistant tumors as another potential mechanism of resistance. Probing the relevance of increased PI3K signaling, we demonstrated that addition of the PI3K inhibitor NVP-BKM120 or the dual PI3K-mTOR (mammalian target of rapamycin) inhibitor NVP-BEZ235 to the initial treatment with the Smo antagonist markedly delayed the development of resistance. Our findings may be useful in informing treatment strategies for medulloblastoma.

Isolated popliteal segment v tibial bypass. Comparison of hemodynamic and clinical results

Clinical and hemodynamic results of isolated popliteal segment, tibial, and sequential bypass grafts were compared in a retrospective review. Results were good with vein grafts to either an isolated segment or infrapopliteal vessel, with five-year patency rates of 71% and 72%, respectively. Prosthetic grafts performed poorly in both groups, and sequential grafts appeared advantageous in such circumstances. Average ankle pressure increased 49 mm Hg following successful isolated segment grafts. Although less than with patent tibial or sequential grafts, improvement was sufficient to relieve rest pain in all instances and heal ischemic lesions or local amputations in all but four patients. If an adequate vein is available and a good tibial vessel exists, distal grafting may be elected, particularly if advanced ischemic lesions demand restoration of pulsatile flow to the foot. If such conditions are not present, isolated segment grafting will give highly satisfactory results.

Isolation and characterization of the gene coding for cytosolic phosphoenolpyruvate carboxykinase (GTP) from the rat

The gene for cytosolic phosphoenolpyruvate carboxykinase (GTP) [GTP:oxaloacetate carboxy-lyase (transphosphorylating), EC 4.1.1.32] from the rat was isolated from a recombinant library containing the rat genome in phage lambda Charon 4A. The isolated clone, lambda PCK1, contains the complete gene for phosphoenolpyruvate carboxykinase and approximately equal to 7 kilobases (kb) of flanking sequence at the 5' end and 1 kb at the 3' terminus. Restriction endonuclease mapping, R-loop mapping, and partial DNA sequence assay indicate that the gene is approximately equal to 6.0 kb in length (coding for a mRNA of 2.8 kb) and contains eight introns. Southern blotting of rat DNA digested with various restriction enzymes shows a pattern predicted from the restriction map of lambda PCK1. A control region at the 5' end of the gene contained in a 1.2-kb restriction fragment was isolated and subcloned into pBR322. This segment of the gene contains the usual transcription start sequences and a 24-base sequence virtually identical to the sequence found in the 5'-flanking region of the human proopiomelonocortin gene, which is known to be regulated by glucocorticoids. The 1.2-kb fragment of the phosphoenolpyruvate carboxykinase gene can be transcribed into a unique RNA fragment of predicted size by an in vitro transcription assay.

Rapid changes in the concentration of phosphoenolpyruvate carboxykinase mRNA in rat liver and kidney. Effects of insulin and cyclic AMP

Starvation and diabetes both caused a marked increase in the concentration of hepatic phosphoenolpyruvate caroboxykinase mRNA while the administration of insulin to diabetic rats or refeeding glucose to starved animals caused a marked reduction in the levels of enzyme mRNA as measured by hybridization using a cDNA probe.l The Administration of dibutyryl cAMP to a starved-refed cat caused an 8-fold induction of phosphoenolpyruvate carboxykinase mRNA in 1 h. Triamcinolone plus acidosis induced the levels of enzyme mRNA in kidney 3-fold within 6 h, however, starvation for 24h had only marginal effects. In all of the above conditions, the levels of phosphoenolpyruvate carboxykinase mRNA measured by hybridization assay agreed well with the relative levels of translatable mRNA for the enzyme. The half-time of phosphoenolpyruvate carboxykinase mRNA, determined after the administration of either alpha-amanitin or cordycepin to starved animals, was approximately 40 min. However, cycloheximide either alone or together with cordycepin, not only prevented the decrease in phosphoenolpyruvate carboxykinase mRNA sequence abundance, but induced it 2-fold. Cycloheximide itself, when injected into 21-day fetal rats in utero caused an induction of enzyme mRNA equal to that noted when dibutyryl cAMP was administered. The mRNA for phosphoenolpyruvate carboxykinase is approximately 2.8 kb in length, but nuclei from the livers of diabetic rats contain a number of putative precursor RNA species for the enzyme, up to 6.5 kb in size, all containing a poly(A) tail. Two hours after refeedng glucose to a starved rat, these nuclear RNA species could no longer be detected by hybridization to our cDNA probe.