The ability to predict the efficacy of molecularly targeted therapies for non–small cell lung cancer (NSCLC) for an individual patient remains problematic. The purpose of this study was to identify, using a refined “coexpression extrapolation (COXEN)” algorithm with a continuous spectrum of drug activity, tumor biomarkers that predict drug sensitivity and therapeutic efficacy in NSCLC to Vorinostat, a histone deacetylase inhibitor, and Velcade, a proteasome inhibitor. Using our refined COXEN algorithm, biomarker prediction models were discovered and trained for Vorinostat and Velcade based on the in vitro drug activity profiles of nine NSCLC cell lines (NCI-9). Independently, a panel of 40 NSCLC cell lines (UVA-40) were treated with Vorinostat or Velcade to obtain 50% growth inhibition values. Genome-wide expression profiles for both the NCI-9 and UVA-40 cell lines were determined using the Affymetrix HG-U133A platform. Modeling generated multigene expression signatures for Vorinostat (45-gene; P = 0.002) and Velcade (15-gene; P = 0.0002), with one overlapping gene (CFLAR). Examination of Vorinostat gene ontogeny revealed a predilection for cellular replication and death, whereas that of Velcade suggested involvement in cellular development and carcinogenesis. Multivariate regression modeling of the refined COXEN scores significantly predicted the activity of combination therapy in NSCLC cells (P = 0.007). Through the refinement of the COXEN algorithm, we provide an in silico method to generate biomarkers that predict tumor sensitivity to molecularly targeted therapies. Use of this refined COXEN method has significant implications for the a priori examination of targeted therapies to more effectively streamline subsequent clinical trial design and cost. Mol Cancer Ther; 9(10); 2834–43. ©2010 AACR.

This article is featured in Highlights of This Issue, p. 2639

The need for new pharmacogenomic approaches to drug discovery and subsequent clinical validation in the treatment of non–small cell lung cancer (NSCLC) is based on two important observations. First, approximately 70% of NSCLC patients present with stage III/IV disease, for whom the standard of care consists of a platinum-based doublet chemotherapeutic regimen with or without local therapies such as surgery or radiation (1). Unfortunately, the results of these platinum-based doublet therapies in the treatment of advanced stage NSCLC patients are poor (2). Second, several recent phase III clinical trials have shown only an 8% to 10% improvement in 5-year survival in patients following surgery who receive adjuvant chemotherapy for node-positive NSCLC (36). Conversely, 90% of patients who receive that same adjuvant chemotherapy receive no benefit, but incur all the risks of subsequent toxicities and the cost of such an all-inclusive approach. Therefore, an unmet need exists for drug discovery platforms to identify new, clinically effective therapeutic agents.

Research is currently focused on identifying novel molecularly targeted therapies that exploit the underlying mechanisms of tumorigenesis and/or tumor cell signal transduction pathways involved in chemoresistance (79). In addition, the inability to predict accurately the efficacy of these molecularly targeted agents for an individual patient remains problematic. Advances in gene expression profiling have begun to dissect the problems outlined above by using a signature-based therapeutic approach for drug discovery and also for predicting chemosensitivity profiles for individual patients (1014).

We have investigated the utility of the combination of histone deacetylase (HDAC) and proteasome inhibitors as a molecularly targeted treatment strategy in NSCLC (8, 15, 16). Isolated HDAC therapy has little effect on NSCLC cell viability, in large part due to secondary activation of the antiapoptotic transcription factor NF-κB through Akt-mediated enhancement of p300 acetyltransferase activity, which promotes acetylation of RelA/p65, the transcriptionally active subunit of NF-κB (16). However, when HDAC inhibitors are combined with a proteasome inhibitor, there was a robust, dose-dependent increase in NSCLC cell apoptosis (8). Whereas this increased NSCLC cell death was encouraging, it was apparent that there were varying degrees of combination-drug sensitivity in selected NSCLC cell lines, regardless of tumor p53, K-Ras, or p16 mutational profiles. Thus, whereas in vitro and in vivo studies (8, 15, 16) suggest that combined HDAC and proteasome inhibition has promise in the treatment of NSCLC, selection of which patient could benefit from such therapy is uncertain. Furthermore, traditional “bench to market” methods (i.e., phase I–III clinical trials) used to assess drug therapy efficacy are lengthy (17), costly (18, 19), and often fail to yield “positive” results (20); these very real concerns all demand a different approach to the problem of which patient will benefit from which drug(s).

In this report, we use a refined version of the “coexpression extrapolation (COXEN)” algorithm (21, 22) and apply it to 40 NSCLC cell lines to identify tumor biomarkers that predict drug sensitivity to Vorinostat (Supplementary Fig. S1A; Merck, Inc.), a HDAC inhibitor, and Velcade (Supplementary Fig. S1B; Millennium Pharmaceuticals), a proteasome inhibitor. We show the high prediction capability of this refined COXEN method through which we develop a formula that predicts the probable efficacy of combined Vorinostat and Velcade therapy in NSCLC. Based on this, we provide an in silico method through which in vitro assessment of compounds such as Vorinostat and Velcade, in isolation or in combination, can be used to generate biomarkers that are highly predictive of tumor sensitivity.

Cell culture, cell lines, reagents, and RNA isolation

Forty human NSCLC cell lines (UVA-40; Supplementary Table S1) were obtained from the American Type Culture Collection and the laboratory of John D. Minna, M.D. (University of Texas Southwestern Medical Center, Dallas, TX) from October 2008 until April 2009 and were grown as previously described (23). All cell lines were used within 6 months of receipt and have been DNA fingerprinted for provenance using the Promega GenePrint PowerPlex 1.2 kit and confirmed to be the same as the DNA fingerprint library maintained by the American Type Culture Collection or the Minna/Gazdar laboratory. Vorinostat was purchased from Sigma-Aldrich. Velcade was provided through a materials transfer agreement with Millennium Pharmaceuticals. Total RNA isolation from these cell lines was done as previously described (23).

Drug activity assay for cell line growth inhibition

Human NSCLC cell lines were plated in 96-well culture plates (Costar) at a density of 1,000 cells per well in 50 μL of complete medium, as previously described (23), with experimental and control plates, incubated at 37°C. After 6 hours of incubation, the control plates were treated with Alamar Blue (Invitrogen Corporation) to assess a baseline value for each cell line. After 24 hours of incubation, the experimental plates were treated with Vorinostat, Velcade, or a combination of the two drugs. Each drug dose was plated in eight repeats. The doses for Vorinostat were 0, 0.1, 0.5, 1, 5, 10, and 20 μmol/L. The doses for Velcade were 0, 5, 10, 25, 50, 100, and 200 nmol/L. The doses for the Vorinostat (μmol/L)/Velcade (nmol/L) combination therapy were 0/0, 0.1/5, 0.5/10, 1/25, 5/50, 10/100, and 20/200. The experimental plates were then incubated for 72 hours (at 37°C) and subsequently treated with Alamar Blue to assess growth inhibition. Drug doses were determined based on previous growth inhibition assays (8).

Microarray expression analysis

The gene profile data of 40 NSCLC cell lines were collected using the Affymetrix HG-U133A platform and are available at Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/). Four of these 40 NSCLC cell lines (NCI-H125, NCI-H226, NCI-H292, and NCI-H596) were profiled using the Affymetrix HG-U133A GeneChips at the University of Virginia. Samples of RNA were first assessed for quality using the Agilent 2100 Bioanalyzer (Agilent Technologies), performing electrophoretic separations that allowed the inspection for two ribosomal peaks, suggesting that no degradation had occurred. Samples were then prepared for analysis using the protocol outlined in the Affymetrix GeneChip Expression Analysis Technical Manual (http://www.affymetrix.com/support/downloads/manuals/expression_analysis_technical_manual.pdf).

Statistical analysis

Estimation of GI50 values.

A nonparametric spline regression technique with the constraint that each drug's higher dose concentration provides at least equal or higher drug efficacy (inhibition) than its lower concentration was applied for estimating the drug activities across each drug's experimental range of dose concentration. The smoothness parameter of spline was tuned objectively using the generalized cross-validation method (24). The software for generalized additive models in the mgcv package (25) for R was used for the dose-effect curve estimation, and a combination of golden section search and successive parabolic interpolation for one-dimensional optimization (26), implemented with the nlminb routine of R, was used to obtain the final GI50 estimates by inverting the dose-effect curves.

Refined COXEN algorithm.

The COXEN algorithm (21) is an in silico method that is used to develop molecular-based prediction models by identifying and using the biomarkers that are concordantly expressed between two independent cancer systems or populations (i.e., the NCI-9 and UVA-40 sets in this study). In brief, COXEN is composed of six distinct components or steps: obtaining relevant drug activity data on the training set (step 1), molecular expression data on both the training and test sets (steps 2 and 3), initial drug sensitivity biomarker discovery on the training set (step 4), subselection of COXEN biomarkers (step 5), and multivariate prediction modeling with these COXEN biomarkers on the training set (step 6). Note that drug activity data on the test set are not needed in these steps and are only prospectively compared with the molecular-based COXEN prediction scores.

In this study, we developed and applied a refined version of the COXEN algorithm. Specifically, principal component regression in lieu of linear discriminate analysis was used for multivariate prediction modeling (step 6). As such, the drug activity data were measured on a continuous scale and no longer divided into either sensitive or resistant group at the expense of potential information loss. The resultant COXEN scores were predicted GI50 values with a continuous spectrum. Applying the refined COXEN algorithm (Fig. 1), we obtained predictive COXEN biomarkers on a continuous spectrum of drug activity for both individual and combination therapies using Vorinostat and Velcade (Supplementary Table S2).

Figure 1.

Schematic plot of the COXEN algorithm and associated data sets. The COXEN algorithm consists of three main steps: identification of profile signature, construction of prediction model, and validation of prediction model. Identification of the profile signature entails using the nine NSCLC cell line subset (NCI-9) of the NCI-60 cancer cell line panel and training them for Vorinostat and Velcade using the Affymetrix HG-U133A GeneChips and estimated GI50 values. The UVA-40 panel of NSCLC cell lines was split into Construction and Validation Subsets. Generated COXEN scores were then used for biomarker selection. The resultant COXEN scores from the single-drug prediction modeling for Vorinostat and Velcade served as the input data to create a prediction model for the Vorinostat and Velcade combination therapy.

Figure 1.

Schematic plot of the COXEN algorithm and associated data sets. The COXEN algorithm consists of three main steps: identification of profile signature, construction of prediction model, and validation of prediction model. Identification of the profile signature entails using the nine NSCLC cell line subset (NCI-9) of the NCI-60 cancer cell line panel and training them for Vorinostat and Velcade using the Affymetrix HG-U133A GeneChips and estimated GI50 values. The UVA-40 panel of NSCLC cell lines was split into Construction and Validation Subsets. Generated COXEN scores were then used for biomarker selection. The resultant COXEN scores from the single-drug prediction modeling for Vorinostat and Velcade served as the input data to create a prediction model for the Vorinostat and Velcade combination therapy.

Close modal

COXEN biomarker selection and prediction modeling for single-drug activity.

The GI50 values and gene expression profiles of the NCI-9 NSCLC cell lines (NCI-9) in the National Cancer Institute public database of 60 cancer cell lines (NCI-60) were used to rank the genes according to their association with the drug activity of each compound. Our application of the refined COXEN algorithm was first rigorously evaluated using a random cross-validation that strictly divided the modeling, training, and test subsets. In brief, the UVA-40 panel of NSCLC cell lines was randomly assigned to two independent subsets. The subset used to filter biomarkers and to construct a prediction model was composed of 19 NSCLC cell lines treated with Vorinostat or Velcade (Construction Subset). The other subset of the UVA-40 panel, which was strictly reserved for validation of the prediction model, was composed of 18 NSCLC cell lines treated with Vorinostat and 19 NSCLC cell lines treated with Velcade (Validation Subset). Note that three NSCLC cell lines treated with Vorinostat and two NSCLC cell lines treated with Velcade were excluded from our prediction modeling due to high experimental variations in drug activities (Supplementary Fig. S2). For a robust statistical inference, we performed this random split 100 times.

With each split, to maintain concordant expression patterns between two independent systems (i.e., the NCI-9 and UVA-40 sets), we triaged the top-ranked 200 biomarkers by filtering the ones that showed inconsistent expression patterns between NCI-9 and the Construction Subset. After the mild filtration, the top-ranked biomarkers served as input variables for constructing multivariate prediction models on the NCI-9 panel. Adding one additional input variable at a time, the candidate prediction models were constructed and the performance of each model was assessed by a rank-based association between its prediction scores and the experimentally measured GI50 values of the Construction Subset along with its 95% bootstrap confidence interval (Supplementary Fig. S3). Note that the sequentially enlarged gene space in high dimension was reduced to a few of major orthogonal directions of the input space with most variations. Among the candidate prediction models, the COXEN prediction model was chosen to minimize the P values of the association tests with the narrowest confidence interval and was independently evaluated by the Validation Subset.

On statistical validation of the refined COXEN algorithm, we obtained the final predictive COXEN biomarkers for Vorinostat or Velcade (Supplementary Table S2) in the UVA-40 panel. The performance of the NCI-9–trained COXEN prediction model was then assessed by measuring the rank-based association between its prediction scores and the experimentally measured GI50 values of the UVA-40.

Multvariate prediction modeling for combination-drug therapy.

The resultant COXEN scores of the single-drug prediction models for Vorinostat and Velcade were used for building a prediction model for combination therapy. A multiple regression form of prediction models with or without interaction terms was considered. Each candidate prediction model for combination therapy was tested using 1,000 runs of a “learning-test split,” where for each run, half of the COXEN scores from the single-drug prediction modeling for Vorinostat and Velcade were jointly sampled with the corresponding observed GI50 values of combination therapy. This subset analysis was used to fit a prediction model having an identical functional form of the final prediction model. Using the other half of the COXEN scores, the performance of the prediction model was measured by an estimate of the rank-based Spearman's correlation coefficient with its respective P value.

Biomarker validation for single-drug activity in NSCLC cells

Initial application of the COXEN algorithm to the Construction Subsets of the UVA-40 panel resulted in COXEN prediction models that selected a 100-gene model for Vorinostat (P < 0.05; Supplementary Fig. S3A) and a 45-gene model for Velcade (P < 0.05; Supplementary Fig. S3B). These models were highly significant, and corresponding 95% bootstrap confidence intervals were simultaneously minimized. COXEN scores were obtained from the constructed prediction model and applied to the Validation Subset of the UVA-40 panel. This subset was strictly reserved for independent prediction where observed GI50 values were plotted against predicted GI50 values for both Vorinostat and Velcade. The resultant plots showed highly significant statistical models with rank-based Spearman's correlations of 0.46 (P = 0.0247) for Vorinostat and 0.53 (P = 0.0093) for Velcade.

Despite these encouraging observations, it is possible that the performance of the developed COXEN prediction model may depend on the specific random split of the UVA-40 panel. To test the robustness of this schema, 100 random splits were done for both Vorinostat (Supplementary Fig. S4A) and Velcade (Supplementary Fig. S4B). The resulting prediction models attained statistical significance (P ≤ 0.1) for both Vorinostat (84%) and Velcade (73%). Therefore, the randomization of the UVA-40 panel cell lines to either the Construction or the Validation Subset does not play a significant role in the resultant prediction model.

With such statistically significant models, we are confident that the COXEN algorithm is capable of developing a prediction model that can be used to predict single-drug sensitivity to Vorinostat and Velcade.

COXEN multigene predictors are predictive of tumor sensitivity to Vorinostat and Velcade in NSCLC cells

With validation of our COXEN approach to prediction modeling, the multigene expression signatures for both Vorinostat and Velcade were developed on the NCI-9 and UVA-40 NSCLC cell lines, resulting in a 45-gene model for Vorinostat (Supplementary Fig. S5A) and a 15-gene model for Velcade (Supplementary Fig. S5B). Plotting of observed versus evaluated GI50 values for Vorinostat (Fig. 2A) and Velcade (Fig. 2B) on UVA-40 NSCLC cell lines showed rank-based Spearman's correlations of 0.46 (P = 0.002) for Vorinostat and 0.55 (P = 0.0002) for Velcade.

Figure 2.

Evaluation of COXEN models for UVA-40 NSCLC cell lines. A, scatter plot of COXEN prediction scores versus experimentally measured GI50 values for Vorinostat resulted in a rank-based Spearman's correlation of 0.46 (P = 0.002). B, scatter plot of COXEN prediction scores versus experimentally measured GI50 values for Velcade resulted in a rank-based Spearman's correlation of 0.55 (P = 0.0002).

Figure 2.

Evaluation of COXEN models for UVA-40 NSCLC cell lines. A, scatter plot of COXEN prediction scores versus experimentally measured GI50 values for Vorinostat resulted in a rank-based Spearman's correlation of 0.46 (P = 0.002). B, scatter plot of COXEN prediction scores versus experimentally measured GI50 values for Velcade resulted in a rank-based Spearman's correlation of 0.55 (P = 0.0002).

Close modal

To examine the overall expression patterns of the COXEN biomarkers for Vorinostat (Supplementary Table S2A) and Velcade (Supplementary Table S2B), we performed a clustering analysis on the UVA-40 panel (Fig. 3). As shown in Fig. 3A for Vorinostat and in Fig. 3B for Velcade, the UVA-40 cell lines were largely separated based on their drug sensitivity in this unsupervised clustering analysis. To show the relative expression pattern of each gene between sensitive and resistant cell lines, the clustering heatmaps were refined from their original output, where there existed a small number of large intensity values that dominated the heatmap landscape, based on their relative ranks of expression intensities. Although simply compared, these clustering heatmaps could show the predictive potential of these biomarkers.

Figure 3.

Heatmaps of biomarkers highly associated with drug sensitivity to Vorinostat and Velcade. A, clustered image map with two-way unsupervised clustering of expression profile data of the 45 highest-ranked genes for Vorinostat using 37 NSCLC cell lines from the UVA-40 panel of NSCLC cell lines. B, clustered image map with two-way unsupervised clustering of expression profile data of the 15 highest-ranked genes for Velcade using 38 NSCLC cell lines from the UVA-40 panel of NSCLC cell lines. Red, black, and green indicate high, intermediate, and low expression levels, respectively. Orange and blue in the upper bar indicate sensitive and resistant cell types based on mean GI50 values, respectively. Pink and burnt orange in the left bar indicate gene networks. (The top biological functions of each network are summarized in Table 1.)

Figure 3.

Heatmaps of biomarkers highly associated with drug sensitivity to Vorinostat and Velcade. A, clustered image map with two-way unsupervised clustering of expression profile data of the 45 highest-ranked genes for Vorinostat using 37 NSCLC cell lines from the UVA-40 panel of NSCLC cell lines. B, clustered image map with two-way unsupervised clustering of expression profile data of the 15 highest-ranked genes for Velcade using 38 NSCLC cell lines from the UVA-40 panel of NSCLC cell lines. Red, black, and green indicate high, intermediate, and low expression levels, respectively. Orange and blue in the upper bar indicate sensitive and resistant cell types based on mean GI50 values, respectively. Pink and burnt orange in the left bar indicate gene networks. (The top biological functions of each network are summarized in Table 1.)

Close modal

To determine which, if any, specific biomarkers were essential to the highly predictive nature of the models, each set of biomarkers was tested by random sampling. In testing the robustness of the gene signatures, 1,000 random samplings of two thirds of the genes from each drug compound (Vorinostat: 30 genes, Velcade: 10 genes) showed an overall consistent predictability of the models (P < 0.0005; Supplementary Fig. S6). This implies that, within the respective gene models, there were no biomarkers that critically affected the performance of the prediction models.

Examination of specific tumor biomarkers for Vorinostat and Velcade

When evaluating the resultant 45-tumor-biomarker signature for Vorinostat (Fig. 3A), we find that four genes appear multiple times: met proto-oncogene (MET); CD44 molecule (CD44); microtubule associated monoxygenase, calponin and LIM domain containing 2 (MICAL2); and calmodulin 1 (CALM1). The multiple appearances of these genes are a function of how the COXEN algorithm probes for significant biomarkers. In using different probes to scan the genome, genes that span large segments of the genome may be picked up by multiple probes, thereby appearing multiple times. Importantly, the presence of a gene multiple times (i.e., CD44) does not affect the prediction performance of our model (Supplementary Fig. S7) because their contributions were mathematically optimized when multiple probes were included in our multivariate model. Thus, our prediction models were quite robust. Additionally, the gene model output renders one biomarker as having only a gene accession number (AK027225), indicating that this is a novel gene yet to be identified. As such, evaluation of the tumor biomarker set for Vorinostat reveals 34 distinct genes that determine the sensitivity of the tumor to single-drug therapy with Vorinostat.

Evaluating the resultant 15-tumor-biomarker signature for Velcade (Fig. 3B), we find that no genes appear multiple times and all genes are identifiable. Therefore, the COXEN algorithm indicates 15 distinct genes as being critical in determining the sensitivity to single-drug therapy with Velcade.

Examination of the gene ontogeny revealed several gene networks involved in tumor cell replication and death: cellular assembly/organization, cell cycle, and RNA damage/repair gene networks for Vorinostat and cellular development and carcinogenic gene networks for Velcade (Table 1). A review of the selected biomarkers for both drugs revealed no significant overlap, with only one shared biomarker, CASP8 and FADD-like apoptosis regulator (CFLAR), for both Vorinostat and Velcade. Although this suggests independent mechanisms of action for the two drugs, CFLAR is a regulator of apoptosis found in both gene signatures. This finding correlates with the in vitro studies on NSCLC performed by our group (8, 27, 28) and the studies on multiple myeloma cells by Grant and colleagues (29, 30), which showed that treatment with combined Vorinostat and Velcade resulted in caspase-mediated apoptosis.

Table 1.

Biological functions of tumor biomarkers selected by the modified COXEN algorithm

Network IDNo. of genesTop biological functions
(A) Vorinostat 
    1 14 Tissue development, cellular assembly and organization, cellular function and maintenance 
    2 12 Cell cycle, cellular growth and proliferation, endocrine system development and function 
    3 RNA damage and repair; nucleic acid metabolism; DNA replication, recombination, and repair 
 
(B) Velcade 
    1 12 Cellular development, organ morphology, reproductive system development and function 
    2 Cancer, neurologic disease, infection mechanism 
Network IDNo. of genesTop biological functions
(A) Vorinostat 
    1 14 Tissue development, cellular assembly and organization, cellular function and maintenance 
    2 12 Cell cycle, cellular growth and proliferation, endocrine system development and function 
    3 RNA damage and repair; nucleic acid metabolism; DNA replication, recombination, and repair 
 
(B) Velcade 
    1 12 Cellular development, organ morphology, reproductive system development and function 
    2 Cancer, neurologic disease, infection mechanism 

Multivariate regression modeling of COXEN scores predicts the activity of doublet therapy in NSCLC cell lines

We hypothesized that the efficacy of doublet Vorinostat and Velcade therapy could be predicted using our refined COXEN algorithm (Fig. 1). Although single-drug modeling indicates only one shared tumor biomarker between Vorinostat and Velcade, prior studies on NSCLC (8), multiple myeloma (29, 30), hematologic T-cell leukemia/lymphoma (31), and cutaneous T-cell lymphoma cells (32) confirm that combination therapy results in decreased cell survival compared with single-drug therapy, suggesting that some degree of enhanced activity exists when both drugs are used.

In searching for multivariate prediction models for the combination therapy, models with and without a drug interaction term were considered for Vorinostat and Velcade. To examine what type of function (i.e., linear, parabolic, etc.) would best represent the combined effect of Vorinostat and Velcade therapy, we plotted the GI50 values of combination therapy on the UVA-40 panel against the predicted COXEN scores (GI50 values) for Vorinostat and Velcade (Supplementary Fig. S8), which implied a high linear association with the single-drug effects for Velcade.

Furthermore, with a significant interaction term in the regression modeling, the final fitted multivariate regression model (R2 = 0.9986) included both single-drug Vorinostat (Vor) and Velcade (Vel; P = 0.06) terms and an interaction term (VorVel, P < 0.001). Below is the final regression model for predicting the activity of doublet Vorinostat and Velcade therapy for tumors, where gene signatures have been identified:

f(Vor,Vel)=0.2708Vor+1.5514Vel+0.1106VorVel

Using the fitted model, observed and predicted GI50 values for combination therapy were significantly correlated (P = 0.007, correlation coefficient = 0.41; Fig. 4). The underlying assumptions (i.e., constant variance and Gaussian distributional specification) of the fitted model were verified using studentized residual and normal quantile-quantile plots.

Figure 4.

Evaluation of a prediction model for combination therapy. Bar plot of direct comparison between COXEN prediction scores and experimentally measured GI50 values for combination therapy. The cell lines are ordered based on GI50 values. COXEN scores and GI50 values were standardized by subtracting the overall mean and dividing by the SE across the UVA-40 panel. The statistical significance was assessed by a rank-based Spearman's correlation test (P = 0.007, correlation coefficient = 0.41). The names of the cell lines for which the COXEN scores and GI50 values are matched are given in blue.

Figure 4.

Evaluation of a prediction model for combination therapy. Bar plot of direct comparison between COXEN prediction scores and experimentally measured GI50 values for combination therapy. The cell lines are ordered based on GI50 values. COXEN scores and GI50 values were standardized by subtracting the overall mean and dividing by the SE across the UVA-40 panel. The statistical significance was assessed by a rank-based Spearman's correlation test (P = 0.007, correlation coefficient = 0.41). The names of the cell lines for which the COXEN scores and GI50 values are matched are given in blue.

Close modal

Having created a prediction model for combination therapy, we further statistically validated the performance of the model. Validation of the combination model involved 1,000 runs of learning-test splits, where 46.1% of the predictions attained statistical significance (P ≤ 0.1; Supplementary Fig. S9).

One of the primary obstacles to the successful treatment of NSCLC has been the dismal performance of platinum-based doublet chemotherapeutic regimens (2, 33). More recently, strategies to overcome this impediment have involved the use of genomic signatures to direct the use of primary chemotherapy (12). Multiple studies have focused on in vitro modeling using the NCI-60 panel of cancer cell lines coupled with baseline gene expression profiling to develop signatures to predict sensitivity to various chemotherapeutic regimens (22, 34). The method commonly used is binary in that cell lines are classified as either sensitive or resistant to a given agent. Based on this assessment, the respective gene expression data are then used to generate a biomarker profile for drug sensitivity (22, 35). The inherent problem with this approach is that drug sensitivity is really a continuous spectrum, and as such, it should not be considered a dichotomous variable. Despite the extensive study of genomic signatures, no other studies before this report have developed profiles based on a more natural continuous spectrum of drug activity on novel molecularly targeted therapies.

In this study, we introduce the application of a refined COXEN algorithm to ascertain the efficacy of two novel molecularly targeted agents, Vorinostat, a HDAC inhibitor, and Velcade, a proteasome inhibitor, in the treatment of NSCLC. The rationale for using these novel agents stems from preliminary data that have shown the in vitro utility of combination HDAC and proteasome inhibitors as a molecularly targeted treatment strategy in NSCLC (8, 15, 16), multiple myeloma (29, 30), hematologic T-cell leukemia/lymphoma (31), and cutaneous T-cell lymphoma cells (32). Use of Vorinostat is further justified by preclinical and clinical studies that support its use in combination with other cancer therapies for the treatment of NSCLC (36, 37). Finally, it is known that a putative target for HDAC inhibitors is the cellular mechanism of handling misfolded proteins that are degraded by the proteasome (38).

We have previously used the COXEN algorithm to extrapolate in vitro drug sensitivity results to well-established, conventional chemotherapeutic compounds to predict tumor behavior in patients with bladder, breast, and ovarian cancers, with the results being validated against independent clinical trial data (21, 39). Previous application of the COXEN algorithm involved developing gene expression profiles based on the inherent sensitivity or resistance of a cell line to a chemical compound (21, 22, 39). Although the COXEN method has been extensively validated, we propose a refined version of the COXEN algorithm (Fig. 1), whereby the gene expression profile is developed based on a continuous spectrum of drug activity. This refined COXEN algorithm eliminates arbitrarily assigned sensitivity/resistance valuations and permits the study of novel molecularly targeted compounds whose response levels for the drug have not been clinically established. With the aforementioned reports having studied tumor behavior in bladder, breast, and ovarian cancers, we are introducing the first application of COXEN using novel molecularly targeted agents, Vorinostat and Velcade, in NSCLC. In using the refined COXEN algorithm, we were able to identify tumor biomarkers that highly predict single-drug sensitivity to Vorinostat and Velcade (Fig. 3; Supplementary Table S2).

Another feature of our refined COXEN algorithm is an extension of the method to produce a functional form derived from our single-drug COXEN output scores for combined Vorinostat and Velcade (Fig. 2), which permits the prediction and internal validation of the probable efficacy of combination therapy for these novel agents in individual tumors. Using this method, we developed a model (Fig. 4) that highly predicts the probable efficacy of doublet therapy with Vorinostat and Velcade (Supplementary Fig. S8). With the current reported correlation levels of the predicted scores, we can achieve both a >75% positive predictive value and a negative predictive value for the top 20% most sensitive and 20% most resistant cell lines. That is, with this degree of correlation, a patient guided by our COXEN prediction will have a >75% certainty of either having a clinical benefit or avoiding unnecessary toxicity. Given that the current results of platinum-based doublet therapies for advanced stage NSCLC yield 5-year survival rates of less than 10% (46), the ability to use molecular predictors to guide combined chemotherapeutic therapy, particularly with novel agents, could result in improved survival rates.

Whereas results from previous applications of the COXEN algorithm were validated against independent clinical trial data (22, 39), this method will be nearly impossible to replicate as efforts are made to introduce new, more “personalized” drug therapies clinically. With the costs of clinical trials for novel agents steadily increasing (40) and the significant number of phase III trials yielding “negative” results after years of patient accrual (41, 42), the current clinical trial method is rapidly becoming obsolete. The refined COXEN algorithm provides a framework whereby an in silico method can be applied to screen novel molecularly targeted therapies in vitro in isolation or in combination to generate biomarkers that predict tumor sensitivity. The resulting biomarkers can be used to prospectively evaluate which tumor types would be responsive to the tested agent. This knowledge would then be used to stratify potential responders into a clinical trial and exclude the nonresponders, thereby effectively streamlining clinical trial design and cost (17). Another potential benefit of an a priori knowledge of biomarkers that predict tumor sensitivity is the ability for drug salvage and/or repositioning strategy in another tumor system in the case of an ongoing clinical trial that fails to yield a “positive” result. These secondary clinical trials would be particularly efficient as the pharmacology and toxicity of the compound(s) would have been already well documented (13).

Potential limitations of our analysis include the use of an in vitro analysis of tumorigenic cell lines to determine biomarkers that predict sensitivity to anticancer agents. Cancer cell lines derived from human tissue can exhibit heterogeneity (i.e., stemness; refs. 4345), have differential doubling times, and grow in nonphysiologic conditions—all of which can affect how these cells respond to anticancer agents (46). To mitigate these concerns as much as possible, we performed our experiments using a 72-hour assay that has been shown by previous studies (22, 47) to be an effective strategy to minimize these issues. An additional concern is that in vitro gene expression profiling of cancer cell lines may not reflect the genomic status of the primary tumor. A recent study performed by Sos and colleagues genomically validated 84 NSCLC cell lines, 37 of which make up the UVA-40; they showed through comparative analysis of orthogonal genomic data sets of these cell lines and primary tumors that NSCLC cell lines reflect the genetic and transcriptional landscape of primary NSCLC specimens (48). The validation of both the assay method and the genetic composition of the UVA-40 NSCLC cell lines strengthens the experimental foundation of the refined COXEN algorithm as described in our study. Notably, our study focuses on the prediction of drug sensitivity of lung cancer cell lines to Vorinostat and Velcade. We believe that this strategy may be applicable to other tumor types and/or novel agents, but we by no means claim that our strategy would work for a wide range of applications in cancer therapeutics.

In conclusion, through the application of a refined COXEN algorithm using a continuous spectrum of drug activity on 40 NSCLC cell lines, we are the first to identify tumor biomarkers that highly predict drug sensitivity to Vorinostat, a HDAC inhibitor, and Velcade, a proteasome inhibitor. Additionally, by extension of the refined COXEN algorithm to combination therapy, we show an ability to predict the probable efficacy of this doublet therapy in NSCLC. In silico COXEN models such as these may significantly enhance our ability to predict a priori the efficacy of novel targeted therapeutics such as Vorinostat and Velcade for NSCLC patients and offer important additional data for subsequent clinical trial designs.

David R. Jones is the principal investigator for a phase I clinical trial involving Vorinostat and Velcade at the University of Virginia (UVA) in patients who have surgically resectable NSCLC. Jae K. Lee is the cofounder of Key Genomics, Inc.

This study was aided by the gracious contribution of NSCLC cell lines provided by Michael Peyton Ph.D. from the laboratory of John D. Minna, M.D. (University of Texas Southwestern Medical Center, Dallas, TX).

Grant Support: NIH grants R01 CA136705 (D.R. Jones) and R01 HL081690 (J.K. Lee) and a Thoracic Surgery Foundation for Research and Education (TSFRE) Research Fellowship (A.S. Nagji).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Azzoli
CG
,
Baker
S
 Jr.
,
Temin
S
, et al
. 
American Society of Clinical Oncology Clinical Practice Guideline update on chemotherapy for stage IV non-small-cell lung cancer
.
J Clin Oncol
2009
;
27
:
6251
66
.
2
Schiller
JH
,
Harrington
D
,
Belani
CP
, et al
. 
Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer
.
N Engl J Med
2002
;
346
:
92
8
.
3
Douillard
JY
,
Tribodet
H
,
Aubert
D
, et al
. 
Adjuvant cisplatin and vinorelbine for completely resected non-small cell lung cancer: subgroup analysis of the Lung Adjuvant Cisplatin Evaluation
.
J Thorac Oncol
2010
;
5
:
220
8
.
4
Douillard
JY
,
Rosell
R
,
De Lena
M
, et al
. 
Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB-IIIA non-small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA): a randomised controlled trial
.
Lancet Oncol
2006
;
7
:
719
27
.
5
Olaussen
KA
,
Dunant
A
,
Fouret
P
, et al
. 
DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy
.
N Engl J Med
2006
;
355
:
983
91
.
6
Winton
T
,
Livingston
R
,
Johnson
D
, et al
. 
Vinorelbine plus cisplatin vs. observation in resected non-small-cell lung cancer
.
N Engl J Med
2005
;
352
:
2589
97
.
7
Bunn
PA
 Jr
. 
The potential role of proteasome inhibitors in the treatment of lung cancer
.
Clin Cancer Res
2004
;
10
:
4263
5s
.
8
Denlinger
CE
,
Keller
MD
,
Mayo
MW
,
Broad
RM
,
Jones
DR
. 
Combined proteasome and histone deacetylase inhibition in non-small cell lung cancer
.
J Thorac Cardiovasc Surg
2004
;
127
:
1078
86
.
9
Mack
PC
,
Davies
AM
,
Lara
PN
,
Gumerlock
PH
,
Gandara
DR
. 
Integration of the proteasome inhibitor PS-341 (Velcade) into the therapeutic approach to lung cancer
.
Lung Cancer
2003
;
41
Suppl 1
:
S89
96
.
10
Anguiano
A
,
Nevins
JR
,
Potti
A
. 
Toward the individualization of lung cancer therapy
.
Cancer
2008
;
113
:
1760
7
.
11
Potti
A
,
Mukherjee
S
,
Petersen
R
, et al
. 
A genomic strategy to refine prognosis in early-stage non-small-cell lung cancer
.
N Engl J Med
2006
;
355
:
570
80
.
12
Potti
A
,
Nevins
JR
. 
Utilization of genomic signatures to direct use of primary chemotherapy
.
Curr Opin Genet Dev
2008
;
18
:
62
7
.
13
Smith
SC
,
Baras
AS
,
Lee
JK
,
Theodorescu
D
. 
The COXEN principle: translating signatures of in vitro chemosensitivity into tools for clinical outcome prediction and drug discovery in cancer
.
Cancer Res
2010
;
70
:
1753
8
.
14
Smith
SC
,
Theodorescu
D
. 
Learning therapeutic lessons from metastasis suppressor proteins
.
Nat Rev Cancer
2009
;
9
:
253
64
.
15
Denlinger
CE
,
Rundall
BK
,
Jones
DR
. 
Proteasome inhibition sensitizes non-small cell lung cancer to histone deacetylase inhibitor-induced apoptosis through the generation of reactive oxygen species
.
J Thorac Cardiovasc Surg
2004
;
128
:
740
8
.
16
Liu
Y
,
Denlinger
CE
,
Rundall
BK
,
Smith
PW
,
Jones
DR
. 
Suberoylanilide hydroxamic acid induces Akt-mediated phosphorylation of p300, which promotes acetylation and transcriptional activation of RelA/p65
.
J Biol Chem
2006
;
281
:
31359
68
.
17
Ohashi
W
,
Mizushima
H
,
Tanaka
H
. 
Economic advantage of pharmacogenomics—clinical trials with genetic information
.
Stud Health Technol Inform
2008
;
136
:
585
90
.
18
Schmidt
C
. 
Costly cancer drugs trigger proposals to modify clinical trial design
.
J Natl Cancer Inst
2009
;
101
:
1662
4
.
19
Collier
R
. 
Rapidly rising clinical trial costs worry researchers
.
CMAJ
2009
;
180
:
277
8
.
20
Laurence
J
. 
No more boring science, no more waste in clinical trials
.
Transl Res
2009
;
153
:
1
3
.
21
Lee
JK
,
Havaleshko
DM
,
Cho
H
, et al
. 
A strategy for predicting the chemosensitivity of human cancers and its application to drug discovery
.
Proc Natl Acad Sci U S A
2007
;
104
:
13086
91
.
22
Havaleshko
DM
,
Cho
H
,
Conaway
M
, et al
. 
Prediction of drug combination chemosensitivity in human bladder cancer
.
Mol Cancer Ther
2007
;
6
:
578
86
.
23
Liu
Y
,
Smith
PW
,
Jones
DR
. 
Breast cancer metastasis suppressor 1 functions as a corepressor by enhancing histone deacetylase 1-mediated deacetylation of RelA/p65 and promoting apoptosis
.
Mol Cell Biol
2006
;
26
:
8683
96
.
24
Hastie
T
,
Tibshirani
R
.
Generalized additive models
.
London
:
Chapman & Hall
; 
1990
.
25
Wood
S
.
Generalized additive models: an introduction with R
.
Boca Raton (FL)
:
Chapman & Hall/CRC
; 
2006
.
26
Brent
R
.
Algorithms for minimization without derivatives
.
Englewood Cliffs (NJ)
:
Prentice-Hall
; 
1973
.
27
Jones
DR
,
Broad
RM
,
Comeau
LD
,
Parsons
SJ
,
Mayo
MW
. 
Inhibition of nuclear factor κB chemosensitizes non-small cell lung cancer through cytochrome c release and caspase activation
.
J Thorac Cardiovasc Surg
2002
;
123
:
310
7
.
28
Jones
DR
,
Broad
RM
,
Madrid
LV
,
Baldwin
AS
 Jr.
,
Mayo
MW
. 
Inhibition of NF-κB sensitizes non-small cell lung cancer cells to chemotherapy-induced apoptosis
.
Ann Thorac Surg
2000
;
70
:
930
6
;
discussion 6–7
.
29
Pei
XY
,
Dai
Y
,
Grant
S
. 
Synergistic induction of oxidative injury and apoptosis in human multiple myeloma cells by the proteasome inhibitor bortezomib and histone deacetylase inhibitors
.
Clin Cancer Res
2004
;
10
:
3839
52
.
30
Yu
C
,
Rahmani
M
,
Conrad
D
,
Subler
M
,
Dent
P
,
Grant
S
. 
The proteasome inhibitor bortezomib interacts synergistically with histone deacetylase inhibitors to induce apoptosis in Bcr/Abl+ cells sensitive and resistant to STI571
.
Blood
2003
;
102
:
3765
74
.
31
Zhang
QL
,
Wang
L
,
Zhang
YW
, et al
. 
The proteasome inhibitor bortezomib interacts synergistically with the histone deacetylase inhibitor suberoylanilide hydroxamic acid to induce T-leukemia/lymphoma cell apoptosis
.
Leukemia
2009
;
23
:
1507
14
.
32
Heider
U
,
Rademacher
J
,
Lamottke
B
, et al
. 
Synergistic interaction of the histone deacetylase inhibitor SAHA with the proteasome inhibitor bortezomib in cutaneous T cell lymphoma
.
Eur J Haematol
2009
;
82
:
440
9
.
33
Ettinger
DS
. 
Is there a preferred combination chemotherapy regimen for metastastic non-small cell lung cancer?
Oncologist
2002
;
7
:
226
33
.
34
Staunton
JE
,
Slonim
DK
,
Coller
HA
, et al
. 
Chemosensitivity prediction by transcriptional profiling
.
Proc Natl Acad Sci U S A
2001
;
98
:
10787
92
.
35
Potti
A
,
Dressman
HK
,
Bild
A
, et al
. 
Genomic signatures to guide the use of chemotherapeutics
.
Nat Med
2006
;
12
:
1294
300
.
36
Ramalingam
SS
,
Maitland
ML
,
Frankel
P
, et al
. 
Carboplatin and Paclitaxel in combination with either vorinostat or placebo for first-line therapy of advanced non-small-cell lung cancer
.
J Clin Oncol
2010
;
28
:
56
62
.
37
Traynor
AM
,
Dubey
S
,
Eickhoff
JC
, et al
. 
Vorinostat (NSC# 701852) in patients with relapsed non-small cell lung cancer: a Wisconsin Oncology Network phase II study
.
J Thorac Oncol
2009
;
4
:
522
6
.
38
Lane
AA
,
Chabner
BA
. 
Histone deacetylase inhibitors in cancer therapy
.
J Clin Oncol
2009
;
27
:
5459
68
.
39
Williams
PD
,
Cheon
S
,
Havaleshko
DM
, et al
. 
Concordant gene expression signatures predict clinical outcomes of cancer patients undergoing systemic therapy
.
Cancer Res
2009
;
69
:
8302
9
.
40
Frantz
S
. 
Why are clinical costs so high?
Nat Rev Drug Discov
2003
;
2
:
851
2
.
41
Saijo
N
,
Nishio
K
,
Tamura
T
. 
Translational and clinical studies of target-based cancer therapy
.
Int J Clin Oncol
2003
;
8
:
187
92
.
42
Saijo
N
,
Tamura
T
,
Nishio
K
. 
Problems in the development of target-based drugs
.
Cancer Chemother Pharmacol
2000
;
46 Suppl
:
S43
5
.
43
Brock
A
,
Chang
H
,
Huang
S
. 
Non-genetic heterogeneity-a mutation-independent driving force for the somatic evolution of tumours
.
Nat Rev Genet
2009
;
10
:
336
42
.
44
Chen
J
,
Odenike
O
,
Rowley
JD
. 
Leukaemogenesis: more than mutant genes
.
Nat Rev Cancer
2010
;
10
:
23
36
.
45
Irish
JM
,
Kotecha
N
,
Nolan
GP
. 
Mapping normal and cancer cell signalling networks: towards single-cell proteomics
.
Nat Rev Cancer
2006
;
6
:
146
55
.
46
Sharma
SV
,
Haber
DA
,
Settleman
J
. 
Cell line-based platforms to evaluate the therapeutic efficacy of candidate anticancer agents
.
Nat Rev Cancer
10
:
241
53
.
47
McDermott
U
,
Sharma
SV
,
Settleman
J
. 
High-throughput lung cancer cell line screening for genotype-correlated sensitivity to an EGFR kinase inhibitor
.
Methods Enzymol
2008
;
438
:
331
41
.
48
Sos
ML
,
Michel
K
,
Zander
T
, et al
. 
Predicting drug susceptibility of non-small cell lung cancers based on genetic lesions
.
J Clin Invest
2009
;
119
:
1727
40
.