The next horizon in precision oncology: Proteogenomics to inform cancer diagnosis and treatment.

When it comes to precision oncology, proteogenomics may provide better prospects to the clinical characterization of tumors, help make a more accurate diagnosis of cancer, and improve treatment for patients with cancer. This perspective describes the significant contributions of The Cancer Genome Atlas and the Clinical Proteomic Tumor Analysis Consortium to precision oncology and makes the case that proteogenomics needs to be fully integrated into clinical trials and patient care in order for precision oncology to deliver the right cancer treatment to the right patient at the right dose and at the right time.

On the Design of Combination Cancer Therapy.

Combination therapy programs are the hallmark of the successful treatment of all forms of human malignancies. In this issue of Cell, Palmer and Sorger present data suggesting that cell culture results indicative of synergistic anticancer drug interactions rarely translate clinically and that the results of combination therapies in mouse models or human clinical trials, even if successful, are best explained by the independent activities of the individually administered drugs.

Atezolizumab for Advanced Alveolar Soft Part Sarcoma.

BACKGROUND: Alveolar soft part sarcoma (ASPS) is a rare soft-tissue sarcoma with a poor prognosis and no established therapy. Recently, encouraging responses to immune checkpoint inhibitors have been reported. METHODS: We conducted an investigator-initiated, multicenter, single-group, phase 2 study of the anti-programmed death ligand 1 (PD-L1) agent atezolizumab in adult and pediatric patients with advanced ASPS. Atezolizumab was administered intravenously at a dose of 1200 mg (in patients ≥18 years of age) or 15 mg per kilogram of body weight with a 1200-mg cap (in patients <18 years of age) once every 21 days. Study end points included objective response, duration of response, and progression-free survival according to Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1, as well as pharmacodynamic biomarkers of multistep drug action. RESULTS: A total of 52 patients were evaluated. An objective response was observed in 19 of 52 patients (37%), with 1 complete response and 18 partial responses. The median time to response was 3.6 months (range, 2.1 to 19.1), the median duration of response was 24.7 months (range, 4.1 to 55.8), and the median progression-free survival was 20.8 months. Seven patients took a treatment break after 2 years of treatment, and their responses were maintained through the data-cutoff date. No treatment-related grade 4 or 5 adverse events were recorded. Responses were noted despite variable baseline expression of programmed death 1 and PD-L1. CONCLUSIONS: Atezolizumab was effective at inducing sustained responses in approximately one third of patients with advanced ASPS. (Funded by the National Cancer Institute and others; ClinicalTrials.gov number, NCT03141684.).

Nox4 regulates InsP3 receptor-dependent Ca2+ release into mitochondria to promote cell survival.

Cells subjected to environmental stresses undergo regulated cell death (RCD) when homeostatic programs fail to maintain viability. A major mechanism of RCD is the excessive calcium loading of mitochondria and consequent triggering of the mitochondrial permeability transition (mPT), which is especially important in post-mitotic cells such as cardiomyocytes and neurons. Here, we show that stress-induced upregulation of the ROS-generating protein Nox4 at the ER-mitochondria contact sites (MAMs) is a pro-survival mechanism that inhibits calcium transfer through InsP3 receptors (InsP3 R). Nox4 mediates redox signaling at the MAM of stressed cells to augment Akt-dependent phosphorylation of InsP3 R, thereby inhibiting calcium flux and mPT-dependent necrosis. In hearts subjected to ischemia-reperfusion, Nox4 limits infarct size through this mechanism. These results uncover a hitherto unrecognized stress pathway, whereby a ROS-generating protein mediates pro-survival effects through spatially confined signaling at the MAM to regulate ER to mitochondria calcium flux and triggering of the mPT.

A cross-study analysis of drug response prediction in cancer cell lines.

To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross-validation within a single study to assess model accuracy. While an essential first step, cross-validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics of Drug Sensitivity in Cancer, Cancer Cell Line Encyclopedia and Genentech Cell Line Screening Initiative (gCSI). Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.

Early drug development in solid tumours: analysis of National Cancer Institute-sponsored phase 1 trials.

BACKGROUND: The low expectation of clinical benefit from phase 1 cancer therapeutics trials might negatively affect patient and physician participation, study reimbursement, and slow the progress of oncology research. Advances in cancer drug development, meanwhile, might have favourably improved treatment responses; however, little comprehensive data exist describing the response and toxicity associated with phase 1 trials across solid tumours. The aim of the study is to evaluate the trend of toxicity and response in phase 1 trials for solid tumours over time. METHODS: We analysed patient-level data from the Cancer Therapy Evaluation Program of the National Cancer Institute-sponsored investigator-initiated phase 1 trials for solid tumours, from Jan 1, 2000, to May 31, 2019. We assessed risks of treatment-related death (grade 5 toxicity ratings possibly, probably, or definitely attributable to treatment), all on-treatment deaths (deaths during protocol treatment regardless of attribution), grade 3-4 toxicity, and proportion of overall response (complete response and partial response) and complete response rate in the study periods of 2000-05, 2006-12, and 2013-2019, and evaluated their trends over time. We also analysed cancer type-specific and investigational agent-specific response, and analysed the trend of response in each cancer type over time. Univariate associations of overall response rates with patients' baseline characteristics (age, sex, performance status, BMI, albumin concentration, and haemoglobin concentration), enrolment period, investigational agents, and trial design were assessed using risk ratio based on the modified Poisson regression model. FINDINGS: We analysed 465 protocols that enrolled 13 847 patients using 261 agents. 144 (31%) trials used a monotherapy and 321 (69%) used combination therapies. The overall treatment-related death rate was 0·7% (95% CI 0·5-0·8) across all periods. Risks of treatment-related deaths did not change over time (p=0·52). All on-treatment death risk during the study period was 8·0% (95% CI 7·6-8·5). The most common grade 3-4 adverse events were haematological; grade 3-4 neutropenia occurred in 2336 (16·9%) of 13 847 patients, lymphopenia in 1230 (8·9%), anaemia in 894 (6·5%), and thrombocytopenia in 979 (7·1%). The overall response rate for all trials during the study period was 12·2% (95% CI 11·5-12·8; 1133 of 9325 patients) and complete response rate was 2·7% (2·4-3·0; 249 of 9325). Overall response increased from 9·6% (95% CI 8·7-10·6) in 2000-05 to 18·0% (15·7-20·5) in 2013-19, and complete response rates from 2·5% (2·0-3·0) to 4·3% (3·2-5·7). Overall response rates for combination therapy were substantially higher than for monotherapy (15·8% [15·0-16·8] vs 3·5% [2·8-4·2]). The overall response by class of agents differed across diseases. Anti-angiogenesis agents were associated with higher overall response rate for bladder, colon, kidney and ovarian cancer. DNA repair inhibitors were associated with higher overall response rate in ovarian and pancreatic cancer. The rates of overall response over time differed markedly by disease; there were notable improvements in bladder, breast, and kidney cancer and melanoma, but no change in the low response of pancreatic and colon cancer. INTERPRETATION: During the past 20 years, the response rate in phase 1 trials nearly doubled without an increase in the treatment-related death rate. However, there is significant heterogeneity in overall response by various factors such as cancer type, investigational agent, and trial design. Therefore, informed decision making is crucial for patients before participating in phase 1 trials. This study provides updated encouraging outcomes of modern phase 1 trials in solid tumours. FUNDING: National Cancer Institute.

TPWshiny: an interactive R/Shiny app to explore cell line transcriptional responses to anti-cancer drugs.

SUMMARY: The NCI Transcriptional Pharmacodynamics Workbench (NCI TPW) is an extensive compilation of directly measured transcriptional responses to anti-cancer agents across the well-characterized NCI-60 cancer cell lines. The NCI TPW data are publicly available through a web interface that allows limited user interaction with the data. We developed 'TPWshiny' as a standalone, easy to install, R application to facilitate more interactive data exploration.With no programming skills required, TPWshiny provides an intuitive and comprehensive graphical interface to help researchers understand the response of tumor cell lines to 15 therapeutic agents. The data are presented in interactive scatter plots, heatmaps, time series and Venn diagrams. Data can be queried by drug concentration, time point, gene and tissue type. Researchers can download the data for further analysis. AVAILABILITY AND IMPLEMENTATION: Users can download the ready-to-use, self-extracting package for Windows or macOS, and R source code from the project website (https://brb.nci.nih.gov/TPWshiny/). TPWshiny documentation and additional information can be found on the project website.

3'-[18F]fluoro-3'-deoxythymidine ([18F]FLT) Positron Emission Tomography as an In Vivo Biomarker of inhibition of CDK 4/6-Rb pathway by Palbociclib in a patient derived bladder tumor.

BACKGROUND: Several new generation CDK4/6 inhibitors have been developed and approved for breast cancer therapy in combination with endocrine therapeutics. Application of these inhibitors either alone or in combination in other solid tumors has been proposed, but no imaging biomarkers of response have been reported in non-breast cancer animal models. The purpose of this study was to evaluate 3'-[18F]fluoro-3'-deoxythymidine ([18F]FLT) Positron Emission Tomography (PET) as in vivo biomarker of response to palbociclib in a non-breast cancer model. METHODS: Twenty-four NSG mice bearing patient derived xenografts (PDX) of a well-characterized bladder tumor were randomized into 4 treatment groups: vehicle (n = 6); palbociclib (n = 6); temozolomide (n = 6); and palbociclib plus temozolomide (n = 6) and treated with two cycles of therapy or vehicle. Tumor uptake of [18F]FLT was determined by micro-PET/CT at baseline, 3 days, and 9 days post initiation of therapy. Following the second cycle of therapy, the mice were maintained until their tumors reached a size requiring humane termination. RESULTS: [18F]FLT uptake decreased significantly in the palbociclib and combination arms (p = 0.0423 and 0.0106 respectively at day 3 and 0.0012 and 0.0031 at day 9) with stable tumor volume. In the temozolomide arm [18F]FLT uptake increased with day 9 uptake significantly different than baseline (p = 0.0418) and progressive tumor growth was observed during the treatment phase. All groups exhibited progressive disease after day 22, 10 days following cessation of therapy. CONCLUSION: Significant decreases in [18F]FLT uptake as early as three days post initiation of therapy with palbociclib, alone or in combination with temozolomide, in this bladder cancer model correlates with an absence of tumor growth during therapy that persists until day 18 for the palbociclib group and day 22 for the combination group (6 days and 10 days) following cessation of therapy. These results support early modulation of [18F]FLT as an in vivo biomarker predictive of palbociclib therapy response in a non-breast cancer model.

Learning curves for drug response prediction in cancer cell lines.

BACKGROUND: Motivated by the size and availability of cell line drug sensitivity data, researchers have been developing machine learning (ML) models for predicting drug response to advance cancer treatment. As drug sensitivity studies continue generating drug response data, a common question is whether the generalization performance of existing prediction models can be further improved with more training data. METHODS: We utilize empirical learning curves for evaluating and comparing the data scaling properties of two neural networks (NNs) and two gradient boosting decision tree (GBDT) models trained on four cell line drug screening datasets. The learning curves are accurately fitted to a power law model, providing a framework for assessing the data scaling behavior of these models. RESULTS: The curves demonstrate that no single model dominates in terms of prediction performance across all datasets and training sizes, thus suggesting that the actual shape of these curves depends on the unique pair of an ML model and a dataset. The multi-input NN (mNN), in which gene expressions of cancer cells and molecular drug descriptors are input into separate subnetworks, outperforms a single-input NN (sNN), where the cell and drug features are concatenated for the input layer. In contrast, a GBDT with hyperparameter tuning exhibits superior performance as compared with both NNs at the lower range of training set sizes for two of the tested datasets, whereas the mNN consistently performs better at the higher range of training sizes. Moreover, the trajectory of the curves suggests that increasing the sample size is expected to further improve prediction scores of both NNs. These observations demonstrate the benefit of using learning curves to evaluate prediction models, providing a broader perspective on the overall data scaling characteristics. CONCLUSIONS: A fitted power law learning curve provides a forward-looking metric for analyzing prediction performance and can serve as a co-design tool to guide experimental biologists and computational scientists in the design of future experiments in prospective research studies.

TPM, FPKM, or Normalized Counts? A Comparative Study of Quantification Measures for the Analysis of RNA-seq Data from the NCI Patient-Derived Models Repository.

BACKGROUND: In order to correctly decode phenotypic information from RNA-sequencing (RNA-seq) data, careful selection of the RNA-seq quantification measure is critical for inter-sample comparisons and for downstream analyses, such as differential gene expression between two or more conditions. Several methods have been proposed and continue to be used. However, a consensus has not been reached regarding the best gene expression quantification method for RNA-seq data analysis. METHODS: In the present study, we used replicate samples from each of 20 patient-derived xenograft (PDX) models spanning 15 tumor types, for a total of 61 human tumor xenograft samples available through the NCI patient-derived model repository (PDMR). We compared the reproducibility across replicate samples based on TPM (transcripts per million), FPKM (fragments per kilobase of transcript per million fragments mapped), and normalized counts using coefficient of variation, intraclass correlation coefficient, and cluster analysis. RESULTS: Our results revealed that hierarchical clustering on normalized count data tended to group replicate samples from the same PDX model together more accurately than TPM and FPKM data. Furthermore, normalized count data were observed to have the lowest median coefficient of variation (CV), and highest intraclass correlation (ICC) values across all replicate samples from the same model and for the same gene across all PDX models compared to TPM and FPKM data. CONCLUSION: We provided compelling evidence for a preferred quantification measure to conduct downstream analyses of PDX RNA-seq data. To our knowledge, this is the first comparative study of RNA-seq data quantification measures conducted on PDX models, which are known to be inherently more variable than cell line models. Our findings are consistent with what others have shown for human tumors and cell lines and add further support to the thesis that normalized counts are the best choice for the analysis of RNA-seq data across samples.

Combination therapy with pazopanib and tivantinib modulates VEGF and c-MET levels in refractory advanced solid tumors.

The vascular endothelial growth factor (VEGF)/VEGFR and hepatocyte growth factor (HGF)/c-MET signaling pathways act synergistically to promote angiogenesis. Studies indicate VEGF inhibition leads to increased levels of phosphorylated c-MET, bypassing VEGF-mediated angiogenesis and leading to chemoresistance. We conducted a phase 1 clinical trial with 32 patients with refractory solid tumors to evaluate the safety, pharmacokinetics, and pharmacodynamics of combinations of VEGF-targeting pazopanib and the putative c-MET inhibitor ARQ197 (tivantinib) at 5 dose levels (DLs). Patients either took pazopanib and tivantinib from treatment initiation (escalation phase) or pazopanib alone for 7 days, with paired tumor sampling, prior to starting combination treatment (expansion phase). Hypertension was the most common adverse event. No more than 1 dose limiting toxicity (DLT) occurred at any DL, so the maximum tolerated dose (MTD) was not determined; DL5 (800 mg pazopanib daily and 360 mg tivantinib BID) was used during the expansion phase. Twenty of 31 evaluable patients achieved stable disease lasting up to 22 cycles. Circulating VEGF, VEGFR2, HGF, and c-MET levels were assessed, and only VEGF levels increased. Tumor c-MET levels (total and phosphorylated) were determined in paired biopsies before and after 7 days of pazopanib treatment. Total intact c-MET decreased in 6 of 7 biopsy pairs, in contrast to previously reported c-MET elevation in response to VEGF inhibition. These results are discussed in the context of our previously reported analysis of epithelial-mesenchymal transition in these tumors.

IL-4 and IL-17A Cooperatively Promote Hydrogen Peroxide Production, Oxidative DNA Damage, and Upregulation of Dual Oxidase 2 in Human Colon and Pancreatic Cancer Cells.

Dual oxidase 2 (DUOX2) generates H2O2 that plays a critical role in both host defense and chronic inflammation. Previously, we demonstrated that the proinflammatory mediators IFN-γ and LPS enhance expression of DUOX2 and its maturation factor DUOXA2 through STAT1- and NF-κB‒mediated signaling in human pancreatic cancer cells. Using a panel of colon and pancreatic cancer cell lines, we now report the induction of DUOX2/DUOXA2 mRNA and protein expression by the TH2 cytokine IL-4. IL-4 activated STAT6 signaling that, when silenced, significantly decreased induction of DUOX2. Furthermore, the TH17 cytokine IL-17A combined synergistically with IL-4 to increase DUOX2 expression in both colon and pancreatic cancer cells mediated, at least in part, by signaling through NF-κB. The upregulation of DUOX2 was associated with a significant increase in the production of extracellular H2O2 and DNA damage-as indicated by the accumulation of 8-oxo-dG and γH2AX-which was suppressed by the NADPH oxidase inhibitor diphenylene iodonium and a DUOX2-specific small interfering RNA. The clinical relevance of these experiments is suggested by immunohistochemical, microarray, and quantitative RT-PCR studies of human colon and pancreatic tumors demonstrating significantly higher DUOX2, IL-4R, and IL-17RA expression in tumors than in adjacent normal tissues; in pancreatic adenocarcinoma, increased DUOX2 expression is adversely associated with overall patient survival. These data suggest a functional association between DUOX2-mediated H2O2 production and induced DNA damage in gastrointestinal malignancies.

A spontaneously metastatic model of bladder cancer: imaging characterization.

BACKGROUND: Spontaneously metastatic xenograft models of cancer are infrequent and the few that exist are resource intensive. In xenografts, caliper measurements can be used to determine primary tumor burden and response to therapy but in metastatic disease models determination of the presence of metastatic disease, metastatic burden, and response to therapy are difficult, often requiring serial necropsy. In this study we characterized the development of visceral metastases in a patient derived xenograft model (PDXM) using in vivo imaging. RESULTS: We identified and characterized the previously unreported development of spontaneous liver and bone metastasis in a known patient derived xenograft, bladder xenograft BL0293F, developed by Jackson Laboratories and the University of California at Davis and available from the National Cancer Institute Patient-Derived Models Repository [1]. Among FDG-PET/CT, contrast-enhanced MRI and non-contrast MRI, non-contrast T2w MRI was the most effective and efficient imaging technique. On non-contrast T2 weighted MRI, hepatic metastases were observed in over 70% of animals at 52 days post tumor implantation without resection of the xenograft and in 100% of animals at day 52 following resection of the xenograft. In a group of animals receiving one cycle of effective chemotherapy, no animals demonstrated metastasis by imaging, confirming the utility of this model for therapy evaluation. There was good agreement between pathologic grade and extent of involvement observed on MRI T2w imaging. CONCLUSION: PDX BL0293F is a reliable visceral organ (liver) metastatic model with high penetrance in both non-aggravated and post excisional situations, providing a reliable window for therapy intervention prior to required excision of the xenograft. The imaging characteristics of this model are highly favorable for non-clinical research studies of metastatic disease when used in conjunction with non-contrast T2 weighted MRI.

Hydrogen peroxide-producing NADPH oxidases and the promotion of migratory phenotypes in cancer.

The cellular microenvironment plays a critical role in cancer initiation and progression. Exposure to oxidative stress, specifically hydrogen peroxide (H2O2), has been linked to aberrant cellular signaling through which the development of cancer may be promoted. Three members of the NADPH oxidase family (NOX4, DUOX1 and DUOX2) explicitly generate this non-radical oxidant in a wide range of tissues, often in support of the inflammatory response. This review summarizes the contributions of each H2O2-producing NOX to the invasive behaviors of tumors and/or the epithelial-mesenchymal transition (EMT) in cancer that plays an essential role in metastasis. Tissue localization in tumorigenesis is also highlighted, with patient-derived TCGA microarray data profiled across 31 cancer cohorts to provide a comprehensive guide to the relevance of NOX4/DUOX1/DUOX2 in cancer studies.

A phase I pharmacokinetic study of belinostat in patients with advanced cancers and varying degrees of liver dysfunction.

AIMS: The histone deacetylase inhibitor belinostat has activity in various cancers. Because belinostat is metabolized by the liver, reduced hepatic clearance could lead to excessive drug accumulation and increased toxicity. Safety data in patients with liver dysfunction are needed for this drug to reach its full potential in the clinic. METHODS: We performed a phase 1 trial to determine the safety, maximum tolerated dose (MTD) and pharmacokinetics of belinostat in patients with advanced cancer and varying degrees of liver dysfunction. RESULTS: Seventy-two patients were enrolled and divided into cohorts based on liver function. In patients with mild dysfunction, the MTD was the same as the recommended phase 2 dose (1000 mg/m2 /day). Belinostat was well tolerated in patients with moderate and severe liver dysfunction, although the trial was closed before the MTD in these cohorts could be determined. The mean clearance of belinostat was 661 mL/min/m2 in patients with normal liver function, compared to 542, 505 and 444 mL/min/m2 in patients with mild, moderate and severe hepatic dysfunction. Although this trial was not designed to assess clinical activity, of the 47 patients evaluable for response, 13 patients (28%) experienced stable disease. CONCLUSION: While a statistically significant difference in clearance indicates increased belinostat exposure with worsening liver function, no relationship was observed between belinostat exposure and toxicity. An assessment of belinostat metabolites revealed significant differences in metabolic pathway capability in patients with differing levels of liver dysfunction. Further studies are needed to establish formal dosing guidelines in this patient population.

Criteria for the use of omics-based predictors in clinical trials.

The US National Cancer Institute (NCI), in collaboration with scientists representing multiple areas of expertise relevant to 'omics'-based test development, has developed a checklist of criteria that can be used to determine the readiness of omics-based tests for guiding patient care in clinical trials. The checklist criteria cover issues relating to specimens, assays, mathematical modelling, clinical trial design, and ethical, legal and regulatory aspects. Funding bodies and journals are encouraged to consider the checklist, which they may find useful for assessing study quality and evidence strength. The checklist will be used to evaluate proposals for NCI-sponsored clinical trials in which omics tests will be used to guide therapy.

Predicting tumor cell line response to drug pairs with deep learning.

BACKGROUND: The National Cancer Institute drug pair screening effort against 60 well-characterized human tumor cell lines (NCI-60) presents an unprecedented resource for modeling combinational drug activity. RESULTS: We present a computational model for predicting cell line response to a subset of drug pairs in the NCI-ALMANAC database. Based on residual neural networks for encoding features as well as predicting tumor growth, our model explains 94% of the response variance. While our best result is achieved with a combination of molecular feature types (gene expression, microRNA and proteome), we show that most of the predictive power comes from drug descriptors. To further demonstrate value in detecting anticancer therapy, we rank the drug pairs for each cell line based on model predicted combination effect and recover 80% of the top pairs with enhanced activity. CONCLUSIONS: We present promising results in applying deep learning to predicting combinational drug response. Our feature analysis indicates screening data involving more cell lines are needed for the models to make better use of molecular features.

NADPH oxidase 1 supports proliferation of colon cancer cells by modulating reactive oxygen species-dependent signal transduction.

Reactive oxygen species (ROS) play a critical role in cell signaling and proliferation. NADPH oxidase 1 (NOX1), a membrane-bound flavin dehydrogenase that generates O2̇̄, is highly expressed in colon cancer. To investigate the role that NOX1 plays in colon cancer growth, we used shRNA to decrease NOX1 expression stably in HT-29 human colon cancer cells. The 80-90% decrease in NOX1 expression achieved by RNAi produced a significant decline in ROS production and a G1/S block that translated into a 2-3-fold increase in tumor cell doubling time without increased apoptosis. The block at the G1/S checkpoint was associated with a significant decrease in cyclin D1 expression and profound inhibition of mitogen-activated protein kinase (MAPK) signaling. Decreased steady-state MAPK phosphorylation occurred concomitant with a significant increase in protein phosphatase activity for two colon cancer cell lines in which NOX1 expression was knocked down by RNAi. Diminished NOX1 expression also contributed to decreased growth, blood vessel density, and VEGF and hypoxia-inducible factor 1α (HIF-1α) expression in HT-29 xenografts initiated from NOX1 knockdown cells. Microarray analysis, supplemented by real-time PCR and Western blotting, revealed that the expression of critical regulators of cell proliferation and angiogenesis, including c-MYC, c-MYB, and VEGF, were down-regulated in association with a decline in hypoxic HIF-1α protein expression downstream of silenced NOX1 in both colon cancer cell lines and xenografts. These studies suggest a role for NOX1 in maintaining the proliferative phenotype of some colon cancers and the potential of NOX1 as a therapeutic target in this disease.

A Phase I Study of Ganetespib and Ziv-Aflibercept in Patients with Advanced Carcinomas and Sarcomas.

LESSONS LEARNED: The combination of the antiangiogenic agent ziv-aflibercept and the heat shock protein 90 inhibitor ganetespib was associated with several serious and unexpected adverse events and was not tolerable on the dosing schedule tested.Studies such as these emphasize the importance of considering overlapping toxicities when designing novel treatment combination regimens. BACKGROUND: Although inhibition of angiogenesis is an effective strategy for cancer treatment, acquired resistance to antiangiogenic therapy is common. Heat shock protein 90 (Hsp90) is a molecular chaperone that regulates various oncogenic signaling pathways involved in acquired resistance and has been shown to play a role in angiogenesis. Combining an antiangiogenic agent with an Hsp90 inhibitor has therefore been proposed as a strategy for preventing resistance and improving antitumor activity. We conducted a single-arm phase I study evaluating the combination of ziv-aflibercept, an antiangiogenic drug, with the Hsp90 inhibitor ganetespib. METHODS: Adult patients were eligible if they had recurrent or metastatic gastrointestinal carcinomas, nonsquamous non-small cell lung carcinomas, urothelial carcinomas, or sarcomas that had progressed after at least one line of standard therapy. Ziv-aflibercept was administered intravenously on days 1 and 15, and ganetespib was administered intravenously on days 1, 8, and 15, of each 28-day cycle. RESULTS: Five patients were treated with the combination. Although three patients achieved stable disease, study treatment was associated with several serious and unexpected adverse events. CONCLUSION: The dose escalation phase of this study was not completed, but the limited data obtained suggest that this combination may be too toxic when administered on this dosing schedule.

First-in-human study of the epichaperome inhibitor PU-H71: clinical results and metabolic profile.

Background Molecular chaperone targeting has shown promise as a therapeutic approach in human cancers of various histologies and genetic backgrounds. The purine-scaffold inhibitor PU-H71 (NSC 750424), selective for Hsp90 in epichaperome networks, has demonstrated antitumor activity in multiple preclinical cancer models. The present study was a first in-human trial of PU-H71 aimed at establishing its safety and tolerability and characterizing its pharmacokinetic (PK) profile on a weekly administration schedule in human subjects with solid tumors refractory to standard treatments. Methods PU-H71 was administered intravenously over 1 h on days 1 and 8 of 21-day cycles in patients with refractory solid tumors. Dose escalation followed a modified accelerated design. Blood and urine were collected during cycles 1 and 2 for pharmacokinetics analysis. Results Seventeen patients were enrolled in this trial. Grade 2 and 3 adverse events were observed but no dose limiting toxicities occurred, thus the human maximum tolerated dose was not determined. The mean terminal half-life (T1/2) was 8.4 ± 3.6 h, with no dependency to dose level. A pathway for the metabolic disposal of PU-H71 in humans was derived from microsome studies. Fourteen patients were also evaluable for clinical response; 6 (35%) achieved a best response of stable disease for >2 cycles, with 2 patients remaining on study for 6 cycles. The study closed prematurely due to discontinuation of drug supply. Conclusions PU-H71 was well tolerated at the doses administered during this study (10 to 470 mg/m2/day), with no dose limiting toxicities.