Calibrating Observational Health Record Data Against a Randomized Trial.

Importance: The conditions required for health record data sources to accurately assess treatment effectiveness remain unclear. Emulation of randomized clinical trials (RCTs) with health record data and subsequent calibration of the results can help elucidate this. Objective: To pilot an emulation of the KEYNOTE-189 RCT using a commercially available electronic health record (EHR) data source. Design, Setting, and Participants: This retrospective cohort study used an EHR database spanning from April 2007 to February 2023. Follow-up began on treatment initiation and proceeded until an outcome event, loss to follow-up, end of data, or end of study period (640 days). The population-based cohort was ascertained from EHRs provided by 52 health systems across the US. Eligibility criteria were defined as closely as possible to the benchmark RCT. Patients with non-small cell lung cancer initiating first-line treatment for metastatic disease were included. Patients with evidence of squamous non-small cell lung cancer, primary nonlung malignant neoplasms, or identified EGFR/ALK variations were excluded. Data were analyzed from June to October 2023. Exposures: Initiation of first-line pembrolizumab and chemotherapy and chemotherapy alone. Chemotherapy in both groups was defined as a combination of pemetrexed and platinum-based (carboplatin or cisplatin) therapy. Main Outcomes and Measures: Outcomes of interest were 12-month survival probability and mortality hazard ratio (HR). Results: A total of 1854 patients (mean [SD] age, 63.7 [9.6] years; 971 [52.4%] men) were eligible, including 589 patients who initiated pembrolizumab and chemotherapy and 1265 patients who initiated chemotherapy only. The cohort included 364 Black patients (19.6%) and 1445 White patients (77.9%). The 12-month survival probabilities were 0.60 (95% CI, 0.54-0.65) in the pembrolizumab group and 0.58 (95% CI, 0.55-0.62) in the chemotherapy-only group, compared with 0.69 (95% CI, 0.64-0.74) in the KEYNOTE-189 pembrolizumab group and 0.49 (95% CI, 0.42-0.56) in the KEYNOTE-189 chemotherapy-only group. The mortality HR was 0.95 (95% CI, 0.78-1.16), compared with 0.49 (95% CI, 0.38-0.64) in the KEYNOTE-189 RCT. Conclusions and Relevance: In this cohort study piloting an RCT emulation, results were incongruous with the benchmark trial. Differences in patient treatment and data capture between the RCT and EHR populations, confounding by indication, treatment crossover, and accuracy of captured diagnoses may explain these findings. Future feasibility assessments will require data sources to have important oncology-specific measures curated.

Development and deployment of a histopathology-based deep learning algorithm for patient prescreening in a clinical trial.

Accurate identification of genetic alterations in tumors, such as Fibroblast Growth Factor Receptor, is crucial for treating with targeted therapies; however, molecular testing can delay patient care due to the time and tissue required. Successful development, validation, and deployment of an AI-based, biomarker-detection algorithm could reduce screening cost and accelerate patient recruitment. Here, we develop a deep-learning algorithm using >3000 H&E-stained whole slide images from patients with advanced urothelial cancers, optimized for high sensitivity to avoid ruling out trial-eligible patients. The algorithm is validated on a dataset of 350 patients, achieving an area under the curve of 0.75, specificity of 31.8% at 88.7% sensitivity, and projected 28.7% reduction in molecular testing. We successfully deploy the system in a non-interventional study comprising 89 global study clinical sites and demonstrate its potential to prioritize/deprioritize molecular testing resources and provide substantial cost savings in the drug development and clinical settings.

A Phase Ib Trial of Personalized Neoantigen Therapy Plus Anti-PD-1 in Patients with Advanced Melanoma, Non-small Cell Lung Cancer, or Bladder Cancer.

Neoantigens arise from mutations in cancer cells and are important targets of T cell-mediated anti-tumor immunity. Here, we report the first open-label, phase Ib clinical trial of a personalized neoantigen-based vaccine, NEO-PV-01, in combination with PD-1 blockade in patients with advanced melanoma, non-small cell lung cancer, or bladder cancer. This analysis of 82 patients demonstrated that the regimen was safe, with no treatment-related serious adverse events observed. De novo neoantigen-specific CD4+ and CD8+ T cell responses were observed post-vaccination in all of the patients. The vaccine-induced T cells had a cytotoxic phenotype and were capable of trafficking to the tumor and mediating cell killing. In addition, epitope spread to neoantigens not included in the vaccine was detected post-vaccination. These data support the safety and immunogenicity of this regimen in patients with advanced solid tumors (Clinicaltrials.gov: NCT02897765).

Harnessing Real-World Evidence for the Development of Novel Cancer Therapies.

Resourcing real-world evidence (RWE) is becoming an increasingly important asset in developing novel therapies for cancer. In this article, an overview of the benefits and challenges of using these data is provided. Through several case examples we highlight future applications and potential.

Defining HLA-II Ligand Processing and Binding Rules with Mass Spectrometry Enhances Cancer Epitope Prediction.

Increasing evidence indicates CD4+ T cells can recognize cancer-specific antigens and control tumor growth. However, it remains difficult to predict the antigens that will be presented by human leukocyte antigen class II molecules (HLA-II), hindering efforts to optimally target them therapeutically. Obstacles include inaccurate peptide-binding prediction and unsolved complexities of the HLA-II pathway. To address these challenges, we developed an improved technology for discovering HLA-II binding motifs and conducted a comprehensive analysis of tumor ligandomes to learn processing rules relevant in the tumor microenvironment. We profiled >40 HLA-II alleles and showed that binding motifs were highly sensitive to HLA-DM, a peptide-loading chaperone. We also revealed that intratumoral HLA-II presentation was dominated by professional antigen-presenting cells (APCs) rather than cancer cells. Integrating these observations, we developed algorithms that accurately predicted APC ligandomes, including peptides from phagocytosed cancer cells. These tools and biological insights will enable improved HLA-II-directed cancer therapies.

High-Order Drug Combinations Are Required to Effectively Kill Colorectal Cancer Cells.

Like classical chemotherapy regimens used to treat cancer, targeted therapies will also rely upon polypharmacology, but tools are still lacking to predict which combinations of molecularly targeted drugs may be most efficacious. In this study, we used image-based proliferation and apoptosis assays in colorectal cancer cell lines to systematically investigate the efficacy of combinations of two to six drugs that target critical oncogenic pathways. Drug pairs targeting key signaling pathways resulted in synergies across a broad spectrum of genetic backgrounds but often yielded only cytostatic responses. Enhanced cytotoxicity was observed when additional processes including apoptosis and cell cycle were targeted as part of the combination. In some cases, where cell lines were resistant to paired and tripled drugs, increased expression of antiapoptotic proteins was observed, requiring a fourth-order combination to induce cytotoxicity. Our results illustrate how high-order drug combinations are needed to kill drug-resistant cancer cells, and they also show how systematic drug combination screening together with a molecular understanding of drug responses may help define optimal cocktails to overcome aggressive cancers. Cancer Res; 76(23); 6950-63. ©2016 AACR.

Targeted Exome Sequencing of the Cancer Genome in Patients with Very High-risk Bladder Cancer.

We completed targeted exome sequencing of the tumors of 50 patients with pTis-pT4b bladder cancer. Mutations were categorized by type, stratified against previously identified cancer loci in the Catalogue of Somatic Mutations in Cancer and The Cancer Genome Atlas databases, and evaluated in pathway analysis and comutation plots. We analyzed mutation associations with receipt of neoadjuvant chemotherapy, nodal involvement, metastatic disease development, and survival. Compared with The Cancer Genome Atlas, we found higher mutation rates in genes encoding products involved in epigenetic regulation and cell cycle regulation. Of the pathways examined, PI3K/mTOR and Cell Cycle/DNA Repair exhibited the greatest frequencies of mutation. RB1 and TP53, as well as NF1 and PIK3CA were frequently comutated. We identified no association between mutations in specific genes and key clinical outcomes of interest when corrected for multiple testing. Discovery phase analysis of the somatic mutations in 50 high-risk bladder cancer patients revealed novel mutations and mutational patterns, which may be useful for developing targeted therapy regimens or new biomarkers for patients at very high risk of disease metastasis and death. PATIENT SUMMARY: In this report we found known, as well as previously unreported, genetic mutations in the tumors of patients with high-risk bladder cancer. These mutations, if validated, may serve as actionable targets for new trials.

Activation of MET via diverse exon 14 splicing alterations occurs in multiple tumor types and confers clinical sensitivity to MET inhibitors.

UNLABELLED: Focal amplification and activating point mutation of the MET gene are well-characterized oncogenic drivers that confer susceptibility to targeted MET inhibitors. Recurrent somatic splice site alterations at MET exon 14 (METex14) that result in exon skipping and MET activation have been characterized, but their full diversity and prevalence across tumor types are unknown. Here, we report analysis of tumor genomic profiles from 38,028 patients to identify 221 cases with METex14 mutations (0.6%), including 126 distinct sequence variants. METex14 mutations are detected most frequently in lung adenocarcinoma (3%), but also frequently in other lung neoplasms (2.3%), brain glioma (0.4%), and tumors of unknown primary origin (0.4%). Further in vitro studies demonstrate sensitivity to MET inhibitors in cells harboring METex14 alterations. We also report three new patient cases with METex14 alterations in lung or histiocytic sarcoma tumors that showed durable response to two different MET-targeted therapies. The diversity of METex14 mutations indicates that diagnostic testing via comprehensive genomic profiling is necessary for detection in a clinical setting. SIGNIFICANCE: Here we report the identification of diverse exon 14 splice site alterations in MET that result in constitutive activity of this receptor and oncogenic transformation in vitro. Patients whose tumors harbored these alterations derived meaningful clinical benefit from MET inhibitors. Collectively, these data support the role of METex14 alterations as drivers of tumorigenesis, and identify a unique subset of patients likely to derive benefit from MET inhibitors.

A functional genomic approach identifies FAL1 as an oncogenic long noncoding RNA that associates with BMI1 and represses p21 expression in cancer.

In a genome-wide survey on somatic copy-number alterations (SCNAs) of long noncoding RNA (lncRNA) in 2,394 tumor specimens from 12 cancer types, we found that about 21.8% of lncRNA genes were located in regions with focal SCNAs. By integrating bioinformatics analyses of lncRNA SCNAs and expression with functional screening assays, we identified an oncogene, focally amplified lncRNA on chromosome 1 (FAL1), whose copy number and expression are correlated with outcomes in ovarian cancer. FAL1 associates with the epigenetic repressor BMI1 and regulates its stability in order to modulate the transcription of a number of genes including CDKN1A. The oncogenic activity of FAL1 is partially attributable to its repression of p21. FAL1-specific siRNAs significantly inhibit tumor growth in vivo.

Quinoline 3-sulfonamides inhibit lactate dehydrogenase A and reverse aerobic glycolysis in cancer cells.

BACKGROUND: Most normal cells in the presence of oxygen utilize glucose for mitochondrial oxidative phosphorylation. In contrast, many cancer cells rapidly convert glucose to lactate in the cytosol, a process termed aerobic glycolysis. This glycolytic phenotype is enabled by lactate dehydrogenase (LDH), which catalyzes the inter-conversion of pyruvate and lactate. The purpose of this study was to identify and characterize potent and selective inhibitors of LDHA. METHODS: High throughput screening and lead optimization were used to generate inhibitors of LDHA enzymatic activity. Effects of these inhibitors on metabolism were evaluated using cell-based lactate production, oxygen consumption, and 13C NMR spectroscopy assays. Changes in comprehensive metabolic profile, cell proliferation, and apoptosis were assessed upon compound treatment. RESULTS: 3-((3-carbamoyl-7-(3,5-dimethylisoxazol-4-yl)-6-methoxyquinolin-4-yl) amino) benzoic acid was identified as an NADH-competitive LDHA inhibitor. Lead optimization yielded molecules with LDHA inhibitory potencies as low as 2 nM and 10 to 80-fold selectivity over LDHB. Molecules in this family rapidly and profoundly inhibited lactate production rates in multiple cancer cell lines including hepatocellular and breast carcinomas. Consistent with selective inhibition of LDHA, the most sensitive breast cancer cell lines to lactate inhibition in hypoxic conditions were cells with low expression of LDHB. Our inhibitors increased rates of oxygen consumption in hepatocellular carcinoma cells at doses up to 3 microM, while higher concentrations directly inhibited mitochondrial function. Analysis of more than 500 metabolites upon LDHA inhibition in Snu398 cells revealed that intracellular concentrations of glycolysis and citric acid cycle intermediates were increased, consistent with enhanced Krebs cycle activity and blockage of cytosolic glycolysis. Treatment with these compounds also potentiated PKM2 activity and promoted apoptosis in Snu398 cells. CONCLUSIONS: Rapid chemical inhibition of LDHA by these quinoline 3-sulfonamids led to profound metabolic alterations and impaired cell survival in carcinoma cells making it a compelling strategy for treating solid tumors that rely on aerobic glycolysis for survival.

Tumor genetic analyses of patients with metastatic melanoma treated with the BRAF inhibitor dabrafenib (GSK2118436).

PURPOSE: Dabrafenib is a selective inhibitor of V600-mutant BRAF kinase, which recently showed improved progression-free survival (PFS) as compared with dacarbazine, in metastatic melanoma patients. This study examined potential genetic markers associated with response and PFS in the phase I study of dabrafenib. EXPERIMENTAL DESIGN: Baseline (pretreatment or archival) melanoma samples were evaluated in 41 patients using a custom genotyping melanoma-specific assay, sequencing of PTEN, and copy number analysis using multiplex ligation amplification and array-based comparative genomic hybridization. Nine patients had on-treatment and/or progression samples available. RESULTS: All baseline patient samples had BRAF(V600E/K) confirmed. Baseline PTEN loss/mutation was not associated with best overall response to dabrafenib, but it showed a trend for shorter median PFS [18.3 (95% confidence interval, CI, 9.1-24.3) vs. 32.1 weeks (95% CI, 24.1-33), P=0.059]. Higher copy number of CCND1 (P=0.009) and lower copy number of CDKN2A (P=0.012) at baseline were significantly associated with decreased PFS. Although no melanomas had high-level amplification of BRAF, the two patients with progressive disease as their best response had BRAF copy gain in their tumors. CONCLUSIONS: Copy number changes in CDKN2A, CCND1, and mutation/copy number changes in PTEN correlated with the duration of PFS in patients treated with dabrafenib. The results suggest that these markers should be considered in the design and interpretation of future trials with selective BRAF inhibitors in advanced melanoma patients.

Lapatinib inhibits the growth of esophageal squamous cell carcinoma and synergistically interacts with 5-fluorouracil in patient-derived xenograft models.

Lapatinib is a dual tyrosine kinase inhibitor of epidermal growth factor receptor (EGFR) and human EGFR-2 (HER2) tyrosine kinase domains. To explore the potential utility of lapatinib for the treatment of esophageal squamous cell carcinoma (ESCC), we examined the expression profiles of EGFR and HER2 in tumor tissues and in paired adjacent non-neoplastic tissues from patients with ESCC. We evaluated the antitumor effects of lapatinib alone or in combination with oxaliplatin or 5-fluorouracil (5-FU) on a panel of primary ESCC cells in vitro with various levels of EGFR and HER2 expression. The in vivo effect of lapatinib alone or in combination with oxaliplatin or 5-FU was evaluated using a primary ESCC xenograft model. EGFR was overexpressed in 80.9% (76/94) of the ESCC samples, while 24.5% (23/94) of the samples overexpressed HER2. EGFR and HER2 co-overexpression was detected in 22.3% of samples (21/94). In vitro, the primary ESCC cells were more sensitive to lapatinib combined with 5-FU or oxaliplatin than to lapatinib alone. Lapatinib in combination with 5-FU had more potent antitumor effects in the primary ESCC xenograft model, and markedly reduced the phosphorylation of EGFR and HER2, compared with lapatinib alone or in combination with oxaliplatin. These data indicate that lapatinib has activity in EGFR- and/or HER2-expressing ESCC primary cells, and that lapatinib in combination with 5-FU may be a promising treatment strategy for patients with ESCC.

Good practice guidelines for biomarker discovery from array data: a case study for breast cancer prognosis.

BACKGROUND: Biomarker discovery holds the promise for advancing personalized medicine as the biomarkers can help match patients to optimal treatment to improve patient outcomes. However, serious concerns have been raised because very few molecular biomarkers or signatures discovered from high dimensional array data can be successfully validated and applied to clinical use. We propose good practice guidelines as well as a novel tool for biomarker discovery and use breast cancer prognosis as a case study to illustrate the proposed approach. RESULTS: We applied the proposed approach to a publicly available breast cancer prognosis dataset and identified small numbers of predictive markers for patient subpopulations stratified by clinical variables. Results from an independent cross-platform validation set show that our model compares favorably to other gene signature and clinical variable based prognostic tools. About half of the discovered candidate markers can individually achieve very good performance, which further demonstrate the high quality of feature selection. These candidate markers perform extremely well for young patient with estrogen receptor-positive, lymph node-negative early stage breast cancers, suggesting a distinct subset of these patients identified by these markers is actually at high risk of recurrence and may benefit from more aggressive treatment than current practice. CONCLUSION: The results show that by following good practice guidelines, we can identify highly predictive genes in high dimensional breast cancer array data. These predictive genes have been successfully validated using an independent cross-platform dataset.

Loss and reduced expression of PTEN correlate with advanced-stage gastric carcinoma.

Phosphatase and tensin homolog (PTEN) is a tumor suppressor involved in multiple cell processes. To investigate the role of PTEN in the development of gastric carcinoma, we determined the expression pattern of PTEN in primary gastric carcinoma and in paired adjacent non-neoplastic tissue. We also determined the correlation of PTEN expression with clinicopathological characteristics and patient survival. Overall, 159 gastric carcinomas and 151 paired adjacent non-neoplastic tissues were used in the present study. PTEN expression was determined using tissue microarrays and immunohistochemistry. The clinical sensitivity and specificity of PTEN expression were calculated using receiver operator characteristic curves. Results showed that the loss of cytoplasmic PTEN was significantly more frequent in carcinoma tissue compared with adjacent non-neoplastic tissue (62 vs. 5%, respectively; P<0.0001). PTEN expression was markedly downregulated in carcinoma tissues compared with adjacent non-neoplastic tissues. The loss of cytoplasmic PTEN expression was positively correlated with histological stage (P=0.016). The loss of nuclear or total PTEN, and downregulation of total PTEN expression, was significantly different between American Joint Committee on Cancer tumors of stage I and stages II-IV. A low cytoplasmic or total PTEN expression showed high clinical sensitivity and specificity for gastric carcinoma. However, PTEN expression was not significantly associated with overall or 3-year survival rates. The findings of the present study indicated that PTEN expression may be a molecular diagnostic marker for gastric cancer. Thus, the loss or reduced expression of PTEN potentially correlate with advanced stages of gastric carcinoma.

High overexpression of fatty acid synthase is associated with poor survival in Chinese patients with gastric carcinoma.

Fatty acid synthase (FAS) is the key enzyme regulating de novo biosynthesis of fatty acids. FAS overexpression has been found in many types of tumors and is associated with poor survival. However, the expression of FAS and its relationship with prognosis in Chinese patients with gastric carcinoma are still unknown. Therefore, in this study, we examined the expression of FAS using tissue microarrays and determined its correlation with clinicopathological characteristics and prognosis of gastric carcinoma in Chinese patients. FAS overexpression was graded as S (T/A) <1, ≥1 to <2, ≥2 to <3 or ≥3 in 35 (38.9%), 20 (22.2%), 9 (10%) and 26 (28.9%) patients, respectively. High FAS overexpression [S (T/A) ≥3] was significantly correlated with poor prognosis (log-rank test, P= 0.0078) and with decreased 3-year survival rate (χ(2) test, P=0.0023). FAS overexpression was not significantly associated with other clinicopathological characteristics. In conclusion, our results suggest that FAS expression might be a potential prognostic marker for gastric carcinoma in Chinese patients.

Combined phosphatase and tensin homolog (PTEN) loss and fatty acid synthase (FAS) overexpression worsens the prognosis of Chinese patients with hepatocellular carcinoma.

We aimed to investigate the expression pattern of phosphatase and tensin homolog (PTEN), to evaluate the relationship between PTEN expression and clinicopathological characteristics, including fatty acid synthase (FAS) expression, and to determine the correlations of PTEN and FAS expression with survival in Chinese patients with hepatocellular carcinoma (HCC). The expression patterns of PTEN and FAS were determined using tissue microarrays and immunohistochemistry. The expression of PTEN was compared with the clinicopathological characteristics of HCC, including FAS expression. Receiver operator characteristic curves were used to calculate the clinical sensitivity and specificity of PTEN expression. Kaplan-Meier survival curves were constructed to evaluate the correlations of PTEN loss and FAS overexpression with overall survival. We found that the loss of PTEN expression occurred predominantly in the cytoplasm, while FAS was mainly localized to the cytoplasm. Cytoplasmic and total PTEN expression levels were significantly decreased in HCC compared with adjacent non-neoplastic tissue (both, p < 0.0001). Decreased cytoplasmic and total PTEN expression showed significant clinical sensitivity and specificity for HCC (both, p < 0.0001). Downregulation of PTEN in HCC relative to non-neoplastic tissue was significantly correlated with histological grade (p = 0.043 for histological grades I-II versus grade III). Loss of total PTEN was significantly correlated with FAS overexpression (p = 0.014). Loss of PTEN was also associated with poor prognosis of patients with poorly differentiated HCC (p = 0.049). Moreover, loss of PTEN combined with FAS overexpression was associated with significantly worse prognosis compared with other HCC cases (p = 0.011). Our data indicate that PTEN may serve as a potential diagnostic and prognostic marker of HCC. Upregulating PTEN expression and inhibiting FAS expression may offer a novel therapeutic approach for HCC.

Genomic DNA copy-number alterations of the let-7 family in human cancers.

In human cancer, expression of the let-7 family is significantly reduced, and this is associated with shorter survival times in patients. However, the mechanisms leading to let-7 downregulation in cancer are still largely unclear. Since an alteration in copy-number is one of the causes of gene deregulation in cancer, we examined copy number alterations of the let-7 family in 2,969 cancer specimens from a high-resolution SNP array dataset. We found that there was a reduction in the copy number of let-7 genes in a cancer-type specific manner. Importantly, focal deletion of four let-7 family members was found in three cancer types: medulloblastoma (let-7a-2 and let-7e), breast cancer (let-7a-2), and ovarian cancer (let-7a-3/let-7b). For example, the genomic locus harboring let-7a-3/let-7b was deleted in 44% of the specimens from ovarian cancer patients. We also found a positive correlation between the copy number of let-7b and mature let-7b expression in ovarian cancer. Finally, we showed that restoration of let-7b expression dramatically reduced ovarian tumor growth in vitro and in vivo. Our results indicate that copy number deletion is an important mechanism leading to the downregulation of expression of specific let-7 family members in medulloblastoma, breast, and ovarian cancers. Restoration of let-7 expression in tumor cells could provide a novel therapeutic strategy for the treatment of cancer.

Predicting breast cancer chemotherapeutic response using a novel tool for microarray data analysis.

We developed a novel tool for microarray data analysis that can parsimoniously discover highly predictive genes by finding the optimal trade off between fold change and t-test p value through rigorous cross validation. In addition to find a small set of highly predictive genes, the tool also has a procedure that recursively discovers and removes predictive genes from the dataset until no such genes can be found. We applied our tool to a public breast cancer dataset with the goal to discover genes that can predict patient’s response to a preoperative chemotherapy. The results show that estrogen receptor (ER) gene is the most important gene to predict chemotherapeutic response and no gene signatures can add much clinical benefit for the whole patient population. We further identified a clinically homogenous subgroup of patients (ER-negative, PR-negative and HER2-negative) whose response to the chemotherapy can be reasonably predicted. Many of the discovered predictive markers for this subgroup of patients were successfully validated using a blinded validation set.

Systematic analysis of genotype-specific drug responses in cancer.

A systematic understanding of genotype-specific sensitivity or resistance to anticancer agents is required to provide improved patient therapy. The availability of an expansive panel of annotated cancer cell lines enables comparative surveys of associations between genotypes and compounds of various target classes. Thus, one can better predict the optimal treatment for a specific tumor. Here, we present a statistical framework, cell line enrichment analysis (CLEA), to associate the response of anticancer agents with major cancer genotypes. Multilevel omics data, including transcriptome, proteome and phosphatome data, were integrated with drug data based on the genotypic classification of cancer cell lines. The results reproduced known patterns of compound sensitivity associated with particular genotypes. In addition, this approach reveals multiple unexpected associations between compounds and mutational genotypes. The mutational genotypes led to unique protein activation and gene expression signatures, which provided a mechanistic understanding of their functional effects. Furthermore, CLEA maps revealed interconnections between TP53 mutations and other mutations in the context of drug responses. The TP53 mutational status appears to play a dominant role in determining clustering patterns of gene and protein expression profiles for major cancer genotypes. This study provides a framework for the integrative analysis of mutations, drug responses and omics data in cancers.