Mass Spectrometry-Based Proteogenomics: New Therapeutic Opportunities for Precision Medicine

Proteogenomics refers to the integration of comprehensive genomic, transcriptomic, and proteomic measurements from the same samples with the goal of fully understanding the regulatory processes converting genotypes to phenotypes, often with an emphasis on gaining a deeper understanding of disease processes. Although specific genetic mutations have long been known to drive the development of multiple cancers, gene mutations alone do not always predict prognosis or response to targeted therapy. The benefit of proteogenomics research is that information obtained from proteins and their corresponding pathways provides insight into therapeutic targets that can complement genomic information by providing an additional dimension regarding the underlying mechanisms and pathophysiology of tumors. This review describes the novel insights into tumor biology and drug resistance derived from proteogenomic analysis while highlighting the clinical potential of proteogenomic observations and advances in technique and analysis tools.

Mapping the proteogenomic landscape enables prediction of drug response in acute myeloid leukemia

Acute myeloid leukemia is a poor-prognosis cancer commonly stratified by genetic aberrations, but these mutations are often heterogeneous and fail to consistently predict therapeutic response. Here, we combine transcriptomic, proteomic, and phosphoproteomic datasets with ex vivo drug sensitivity data to help understand the underlying pathophysiology of AML beyond mutations. We measure the proteome and phosphoproteome of 210 patients and combine them with genomic and transcriptomic measurements to identify four proteogenomic subtypes that complement existing genetic subtypes. We build a predictor to classify samples into subtypes and map them to a "landscape" that identifies specific drug response patterns. We then build a drug response prediction model to identify drugs that target distinct subtypes and validate our findings on cell lines representing various stages of quizartinib resistance. Our results show how multiomics data together with drug sensitivity data can inform therapy stratification and drug combinations in AML.

Deep learning integrates histopathology and proteogenomics at a pan-cancer level

We introduce a pioneering approach that integrates pathology imaging with transcriptomics and proteomics to identify predictive histology features associated with critical clinical outcomes in cancer. We utilize 2,755 H&E-stained histopathological slides from 657 patients across 6 cancer types from CPTAC. Our models effectively recapitulate distinctions readily made by human pathologists: tumor vs. normal (AUROC = 0.995) and tissue-of-origin (AUROC = 0.979). We further investigate predictive power on tasks not normally performed from H&E alone, including TP53 prediction and pathologic stage. Importantly, we describe predictive morphologies not previously utilized in a clinical setting. The incorporation of transcriptomics and proteomics identifies pathway-level signatures and cellular processes driving predictive histology features. Model generalizability and interpretability is confirmed using TCGA. We propose a classification system for these tasks, and suggest potential clinical applications for this integrated human and machine learning approach. A publicly available web-based platform implements these models.

Proteogenomic insights suggest druggable pathways in endometrial carcinoma

We characterized a prospective endometrial carcinoma (EC) cohort containing 138 tumors and 20 enriched normal tissues using 10 different omics platforms. Targeted quantitation of two peptides can predict antigen processing and presentation machinery activity, and may inform patient selection for immunotherapy. Association analysis between MYC activity and metformin treatment in both patients and cell lines suggests a potential role for metformin treatment in non-diabetic patients with elevated MYC activity. PIK3R1 in-frame indels are associated with elevated AKT phosphorylation and increased sensitivity to AKT inhibitors. CTNNB1 hotspot mutations are concentrated near phosphorylation sites mediating pS45-induced degradation of β-catenin, which may render Wnt-FZD antagonists ineffective. Deep learning accurately predicts EC subtypes and mutations from histopathology images, which may be useful for rapid diagnosis. Overall, this study identified molecular and imaging markers that can be further investigated to guide patient stratification for more precise treatment of EC.

Integrative multi-omic cancer profiling reveals DNA methylation patterns associated with therapeutic vulnerability and cell-of-origin

DNA methylation plays a critical role in establishing and maintaining cellular identity. However, it is frequently dysregulated during tumor development and is closely intertwined with other genetic alterations. Here, we leveraged multi-omic profiling of 687 tumors and matched non-involved adjacent tissues from the kidney, brain, pancreas, lung, head and neck, and endometrium to identify aberrant methylation associated with RNA and protein abundance changes and build a Pan-Cancer catalog. We uncovered lineage-specific epigenetic drivers including hypomethylated FGFR2 in endometrial cancer. We showed that hypermethylated STAT5A is associated with pervasive regulon downregulation and immune cell depletion, suggesting that epigenetic regulation of STAT5A expression constitutes a molecular switch for immunosuppression in squamous tumors. We further demonstrated that methylation subtype-enrichment information can explain cell-of-origin, intra-tumor heterogeneity, and tumor phenotypes. Overall, we identified cis-acting DNA methylation events that drive transcriptional and translational changes, shedding light on the tumor's epigenetic landscape and the role of its cell-of-origin.

Validation of a Proteomic Signature of Lung Cancer Risk from Bronchial Specimens of Risk-Stratified Individuals

A major challenge in lung cancer prevention and cure hinges on identifying the at-risk population that ultimately develops lung cancer. Previously, we reported proteomic alterations in the cytologically normal bronchial epithelial cells collected from the bronchial brushings of individuals at risk for lung cancer. The purpose of this study is to validate, in an independent cohort, a selected list of 55 candidate proteins associated with risk for lung cancer with sensitive targeted proteomics using selected reaction monitoring (SRM). Bronchial brushings collected from individuals at low and high risk for developing lung cancer as well as patients with lung cancer, from both a subset of the original cohort (batch 1: n = 10 per group) and an independent cohort of 149 individuals (batch 2: low risk (n = 32), high risk (n = 34), and lung cancer (n = 83)), were analyzed using multiplexed SRM assays. ALDH3A1 and AKR1B10 were found to be consistently overexpressed in the high-risk group in both batch 1 and batch 2 brushing specimens as well as in the biopsies of batch 1. Validation of highly discriminatory proteins and metabolic enzymes by SRM in a larger independent cohort supported their use to identify patients at high risk for developing lung cancer.

Proteome Mapping of the Human Pancreatic Islet Microenvironment Reveals Endocrine-Exocrine Signaling Sphere of Influence

The need for a clinically accessible method with the ability to match protein activity within heterogeneous tissues is currently unmet by existing technologies. Our proteomics sample preparation platform, named microPOTS (Microdroplet Processing in One pot for Trace Samples), can be used to measure relative protein abundance in micron-scale samples alongside the spatial location of each measurement, thereby tying biologically interesting proteins and pathways to distinct regions. However, given the smaller pixel/voxel number and amount of tissue measured, standard mass spectrometric analysis pipelines have proven inadequate. Here we describe how existing computational approaches can be adapted to focus on the specific biological questions asked in spatial proteomics experiments. We apply this approach to present an unbiased characterization of the human islet microenvironment comprising the entire complex array of cell types involved while maintaining spatial information and the degree of the islet's sphere of influence. We identify specific functional activity unique to the pancreatic islet cells and demonstrate how far their signature can be detected in the adjacent tissue. Our results show that we can distinguish pancreatic islet cells from the neighboring exocrine tissue environment, recapitulate known biological functions of islet cells, and identify a spatial gradient in the expression of RNA processing proteins within the islet microenvironment.

Abstract 955: Proteogenomic insights suggest druggable pathways in endometrial cancer

Endometrial carcinoma (EC) is the most common gynecologic cancer diagnosed in developed countries, with incidence on the rise globally. Over the past decade, EC outcomes have worsened in the United States, despite the fact that early stage, well-differentiated cancers are often cured by simple hysterectomy. Work by the Cancer Genome Atlas Consortium (TCGA) and others has stratified EC patients by four genomically-defined subtypes: POLE ultramutated, microsatellite instability high (MSI-H) hypermutated, copy-number low (CNV-L), and copy-number high (CNV-H). These subtypes not only correlate with clinical outcome but also, in specific instances, response to targeted agents such as immune checkpoint inhibitors (ICI). By integrating genomic and proteomic data, our recent study with the Clinical Proteomic Tumor Analysis Consortium (CPTAC) built upon findings from the TCGA, specifically identifying immunologic subtypes which differentiate tumors based on their overall mutation burdens and antigen processing machinery (APM) profiles. Despite this progress, effective treatments for advanced stage, metastatic and/or recurrent disease are often lacking. Thus, a better understanding of how EC subtypes can be efficiently identified and targeted to improve their outcomes is urgently needed. Here, we present the results of a comprehensive proteogenomic analysis of a new prospectively curated set of 138 EC tumors and 20 specimens enriched for normal endometrium. We also evaluated the utility of a novel machine learning algorithm for parsing clinically-relevant genomic features based solely on images from Hematoxylin and Eosin (H&E) stained cross-sections. Our integrative multi-omic analyses examined the diverse dimensions of molecular pathways to better describe diagnostic and therapeutic targets. From our targeted proteomic assay, we describe a model that uses two peptides to accurately predict tumor APM status. Using gene expression and clinical data, we observed an association between lower Myc activity and Metformin usage in our patients with diabetes. Linking genomics, proteomics, and phospho-proteomics, we uncovered an upregulation of AKT-pT308 in tumors with PIK3R1 in-frame indels. Multi-omic clustering revealed a subset of CNV-L tumors enriched for ß-catenin hotspot mutations, which have elevated Wnt signaling due to impaired ß-catenin degradation signals. Lastly, we observe a subset of MSI-H tumors with a gain of chromosome 1q that show lowered immune infiltration. Analysis of this independent cohort, incorporating published EC tumor and cell line cohorts, has not only confirmed findings from our recent exploratory study but also provided new biological insights relevant to potential therapeutic strategies. These findings can be further investigated to guide patient stratification for improved precision treatment of EC.

Citation Format: Lizabeth Katsnelson, Yongchao Dou, Marina A. Gritsenko, Matthew L. Anderson, Karin D. Rodland, Bing Zhang, Tao Liu, David Fenyö, the Clinical Proteomic Tumor Analysis Consortium. Proteogenomic insights suggest druggable pathways in endometrial cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 955.

A streamlined tandem tip-based workflow for sensitive nanoscale phosphoproteomics

Effective phosphoproteome of nanoscale sample analysis remains a daunting task, primarily due to significant sample loss associated with non-specific surface adsorption during enrichment of low stoichiometric phosphopeptide. We develop a tandem tip phosphoproteomics sample preparation method that is capable of sample cleanup and enrichment without additional sample transfer, and its integration with our recently developed SOP (Surfactant-assisted One-Pot sample preparation) and iBASIL (improved Boosting to Amplify Signal with Isobaric Labeling) approaches provides a streamlined workflow enabling sensitive, high-throughput nanoscale phosphoproteome measurements. This approach significantly reduces both sample loss and processing time, allowing the identification of >3000 (>9500) phosphopeptides from 1 (10) µg of cell lysate using the label-free method without a spectral library. It also enables precise quantification of ~600 phosphopeptides from 100 sorted cells (single-cell level input for the enriched phosphopeptides) and ~700 phosphopeptides from human spleen tissue voxels with a spatial resolution of 200 µm (equivalent to ~100 cells) in a high-throughput manner. The new workflow opens avenues for phosphoproteome profiling of mass-limited samples at the low nanogram level.

Histopathologic and proteogenomic heterogeneity reveals features of clear cell renal cell carcinoma aggressiveness

Clear cell renal cell carcinomas (ccRCCs) represent ∼75% of RCC cases and account for most RCC-associated deaths. Inter- and intratumoral heterogeneity (ITH) results in varying prognosis and treatment outcomes. To obtain the most comprehensive profile of ccRCC, we perform integrative histopathologic, proteogenomic, and metabolomic analyses on 305 ccRCC tumor segments and 166 paired adjacent normal tissues from 213 cases. Combining histologic and molecular profiles reveals ITH in 90% of ccRCCs, with 50% demonstrating immune signature heterogeneity. High tumor grade, along with BAP1 mutation, genome instability, increased hypermethylation, and a specific protein glycosylation signature define a high-risk disease subset, where UCHL1 expression displays prognostic value. Single-nuclei RNA sequencing of the adverse sarcomatoid and rhabdoid phenotypes uncover gene signatures and potential insights into tumor evolution. In vitro cell line studies confirm the potential of inhibiting identified phosphoproteome targets. This study molecularly stratifies aggressive histopathologic subtypes that may inform more effective treatment strategies.

Internal Standard Triggered-Parallel Reaction Monitoring Mass Spectrometry Enables Multiplexed Quantification of Candidate Biomarkers in Plasma

Despite advances in proteomic technologies, clinical translation of plasma biomarkers remains low, partly due to a major bottleneck between the discovery of candidate biomarkers and costly clinical validation studies. Due to a dearth of multiplexable assays, generally only a few candidate biomarkers are tested, and the validation success rate is accordingly low. Previously, mass spectrometry-based approaches have been used to fill this gap but feature poor quantitative performance and were generally limited to hundreds of proteins. Here, we demonstrate the capability of an internal standard triggered-parallel reaction monitoring (IS-PRM) assay to greatly expand the numbers of candidates that can be tested with improved quantitative performance. The assay couples immunodepletion and fractionation with IS-PRM and was developed and implemented in human plasma to quantify 5176 peptides representing 1314 breast cancer biomarker candidates. Characterization of the IS-PRM assay demonstrated the precision (median % CV of 7.7%), linearity (median R² > 0.999 over 4 orders of magnitude), and sensitivity (median LLOQ < 1 fmol, approximately) to enable rank-ordering of candidate biomarkers for validation studies. Using three plasma pools from breast cancer patients and three control pools, 893 proteins were quantified, of which 162 candidate biomarkers were verified in at least one of the cancer pools and 22 were verified in all three cancer pools. The assay greatly expands capabilities for quantification of large numbers of proteins and is well suited for prioritization of viable candidate biomarkers.

New Views of Old Proteins: Clarifying the Enigmatic Proteome

All human diseases involve proteins, yet our current tools to characterize and quantify them are limited. To better elucidate proteins across space, time, and molecular composition, we provide a >10 years of projection for technologies to meet the challenges that protein biology presents. With a broad perspective, we discuss grand opportunities to transition the science of proteomics into a more propulsive enterprise. Extrapolating recent trends, we describe a next generation of approaches to define, quantify, and visualize the multiple dimensions of the proteome, thereby transforming our understanding and interactions with human disease in the coming decade.

Evaluation of Differential Peptide Loading on Tandem Mass Tag-Based Proteomic and Phosphoproteomic Data Quality

Global and phosphoproteome profiling has demonstrated great utility for the analysis of clinical specimens. One barrier to the broad clinical application of proteomic profiling is the large amount of biological material required, particularly for phosphoproteomics─currently on the order of 25 mg wet tissue weight. For hematopoietic cancers such as acute myeloid leukemia (AML), the sample requirement is ≥10 million peripheral blood mononuclear cells (PBMCs). Across large study cohorts, this requirement will exceed what is obtainable for many individual patients/time points. For this reason, we were interested in the impact of differential peptide loading across multiplex channels on proteomic data quality. To achieve this, we tested a range of channel loading amounts (approximately the material obtainable from 5E5, 1E6, 2.5E6, 5E6, and 1E7 AML patient cells) to assess proteome coverage, quantification precision, and peptide/phosphopeptide detection in experiments utilizing isobaric tandem mass tag (TMT) labeling. As expected, fewer missing values were observed in TMT channels with higher peptide loading amounts compared to lower loadings. Moreover, channels with a lower loading have greater quantitative variability than channels with higher loadings. A statistical analysis showed that decreased loading amounts result in an increase in the type I error rate. We then examined the impact of differential loading on the detection of known differences between distinct AML cell lines. Similar patterns of increased data missingness and higher quantitative variability were observed as loading was decreased resulting in fewer statistical differences; however, we found good agreement in features identified as differential, demonstrating the value of this approach.

Facile One-Pot Nanoproteomics for Label-Free Proteome Profiling of 50-1000 Mammalian Cells

Recent advances in sample preparation enable label-free mass spectrometry (MS)-based proteome profiling of small numbers of mammalian cells. However, specific devices are often required to downscale sample processing volume from the standard 50-200 μL to sub-μL for effective nanoproteomics, which greatly impedes the implementation of current nanoproteomics methods by the proteomics research community. Herein, we report a facile one-pot nanoproteomics method termed SOPs-MS (surfactant-assisted one-pot sample processing at the standard volume coupled with MS) for convenient robust proteome profiling of 50-1000 mammalian cells. Building upon our recent development of SOPs-MS for label-free single-cell proteomics at a low μL volume, we have systematically evaluated its processing volume at 10-200 μL using 100 human cells. The processing volume of 50 μL that is in the range of volume for standard proteomics sample preparation has been selected for easy sample handling with a benchtop micropipette. SOPs-MS allows for reliable label-free quantification of ∼1200-2700 protein groups from 50 to 1000 MCF10A cells. When applied to small subpopulations of mouse colon crypt cells, SOPs-MS has revealed protein signatures between distinct subpopulation cells with identification of ∼1500-2500 protein groups for each subpopulation. SOPs-MS may pave the way for routine deep proteome profiling of small numbers of cells and low-input samples.

CD63-mediated cloaking of VEGF in small extracellular vesicles contributes to anti-VEGF therapy resistance

Despite wide use of anti-vascular endothelial growth factor (VEGF) therapy for many solid cancers, most individuals become resistant to this therapy, leading to disease progression. Therefore, new biomarkers and strategies for blocking adaptive resistance of cancer to anti-VEGF therapy are needed. As described here, we demonstrate that cancer-derived small extracellular vesicles package increasing quantities of VEGF and other factors in response to anti-VEGF therapy. The packaging process of VEGF into small extracellular vesicles (EVs) is mediated by the tetraspanin CD63. Furthermore, small EV-VEGF (eVEGF) is not accessible to anti-VEGF antibodies and can trigger intracrine VEGF signaling in endothelial cells. eVEGF promotes angiogenesis and enhances tumor growth despite bevacizumab treatment. These data demonstrate a mechanism where VEGF is partitioned into small EVs and promotes tumor angiogenesis and progression. These findings have clinical implications for biomarkers and therapeutic strategies for ovarian cancer.

The AML microenvironment catalyzes a stepwise evolution to gilteritinib resistance

Our study details the stepwise evolution of gilteritinib resistance in FLT3-mutated acute myeloid leukemia (AML). Early resistance is mediated by the bone marrow microenvironment, which protects residual leukemia cells. Over time, leukemia cells evolve intrinsic mechanisms of resistance, or late resistance. We mechanistically define both early and late resistance by integrating whole-exome sequencing, CRISPR-Cas9, metabolomics, proteomics, and pharmacologic approaches. Early resistant cells undergo metabolic reprogramming, grow more slowly, and are dependent upon Aurora kinase B (AURKB). Late resistant cells are characterized by expansion of pre-existing NRAS mutant subclones and continued metabolic reprogramming. Our model closely mirrors the timing and mutations of AML patients treated with gilteritinib. Pharmacological inhibition of AURKB resensitizes both early resistant cell cultures and primary leukemia cells from gilteritinib-treated AML patients. These findings support a combinatorial strategy to target early resistant AML cells with AURKB inhibitors and gilteritinib before the expansion of pre-existing resistance mutations occurs.

Abstract 2170: Proteogenomic and metabolomic characterization of human glioblastoma

Glioblastoma (GBM) is the most aggressive nervous system cancer, with median survival under 2 years. Understanding its molecular pathogenesis is crucial for improving diagnosis and treatment. We performed an integrated analysis of genomic, proteomic, post-translational modification and metabolomic data on 99 treatment-naive GBMs. We identified key phosphorylation events (e.g., phosphorylated PTPN11 and PLCG1) as potential switches mediating oncogenic pathway activation as well as potential targets for EGFR-, TP53- and RB1-altered tumors. We detected immune subtypes, driven by the presence of distinct immune cell populations using bulk omics, validated by snRNA-seq, and they were correlated with specific expression and histone acetylation patterns. Acetylation of histone H2B in classical-like and immune-low GBM was driven largely by BRDs, CREBBP, and EP300. Integrated metabolomic and proteomic data identified specific lipid distributions across subtypes and distinct global metabolic changes in IDH mutated tumors. By comparing the adult GBM proteomics to the adolescent and young adult (AYA) GBM cohort from the HOPE study, we found downregulated IDH1 expression and up-regulated expression of genes in the NADH dehydrogenase family, including NDUFB1, and NDUFB3 among others, which may be related to high IDH1 mutation frequency in AYA. This work highlights biological relationships which could potentially aid GBM patient stratifications for more effective treatments.

Citation Format: Liang-Bo Wang, Alla Karpova, Marina A. Gritsenko, Jennifer E. Kyle, Song Cao, Yize Li, Dmitry Rykunov, Antonio Colaprico, Joseph Rothstein, Runyu Hong, Vasileios Stathias, MacIntosh Cornwell, Francesca Petralia, Richard D. Smith, Antonio Iavarone, Milan G. Chheda, Jill S. Barnholtz-Sloan, Karin D. Rodland, Tao Liu, Li Ding, Clinical Proteomic Tumor Analysis Consortium. Proteogenomic and metabolomic characterization of human glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2170.

Assessment of TMT Labeling Efficiency in Large-Scale Quantitative Proteomics: The Critical Effect of Sample pH

Isobaric labeling via tandem mass tag (TMT) reagents enables sample multiplexing prior to LC-MS/MS, facilitating high-throughput large-scale quantitative proteomics. Consistent and efficient labeling reactions are essential to achieve robust quantification; therefore, embedded in our clinical proteomic protocol is a quality control (QC) sample that contains a small aliquot from each sample within a TMT set, referred to as "Mixing QC." This Mixing QC enables the detection of TMT labeling issues by LC-MS/MS before combining the full samples to allow for salvaging of poor TMT labeling reactions. While TMT labeling is a valuable tool, factors leading to poor reactions are not fully studied. We observed that relabeling does not necessarily rescue TMT reactions and that peptide samples sometimes remained acidic after resuspending in 50 mM HEPES buffer (pH 8.5), which coincided with low labeling efficiency (LE) and relatively low median reporter ion intensities (MRIIs). To obtain a more resilient TMT labeling procedure, we investigated LE, reporter ion missingness, the ratio of mean TMT set MRII to individual channel MRII, and the distribution of log 2 reporter ion ratios of Mixing QC samples. We discovered that sample pH is a critical factor in LE, and increasing the buffer concentration in poorly labeled samples before relabeling resulted in the successful rescue of TMT labeling reactions. Moreover, resuspending peptides in 500 mM HEPES buffer for TMT labeling resulted in consistently higher LE and lower missing data. By better controlling the sample pH for labeling and implementing multiple methods for assessing labeling quality before combining samples, we demonstrate that robust TMT labeling for large-scale quantitative studies is achievable.

Proteogenomic and metabolomic characterization of human glioblastoma

Glioblastoma (GBM) is the most aggressive nervous system cancer. Understanding its molecular pathogenesis is crucial to improving diagnosis and treatment. Integrated analysis of genomic, proteomic, post-translational modification and metabolomic data on 99 treatment-naive GBMs provides insights to GBM biology. We identify key phosphorylation events (e.g., phosphorylated PTPN11 and PLCG1) as potential switches mediating oncogenic pathway activation, as well as potential targets for EGFR-, TP53-, and RB1-altered tumors. Immune subtypes with distinct immune cell types are discovered using bulk omics methodologies, validated by snRNA-seq, and correlated with specific expression and histone acetylation patterns. Histone H2B acetylation in classical-like and immune-low GBM is driven largely by BRDs, CREBBP, and EP300. Integrated metabolomic and proteomic data identify specific lipid distributions across subtypes and distinct global metabolic changes in IDH-mutated tumors. This work highlights biological relationships that could contribute to stratification of GBM patients for more effective treatment.

Abstract LT022: The AML microenvironment catalyzes a step-wise evolution to gilteritinib resistance

In acute myeloid leukemia (AML), activating mutations in FLT3 are the most common genetic abnormality. Multiple FLT3 inhibitors have been developed, including the FDA-approved inhibitor gilteritinib. However, AML patients only respond to gilteritinib for approximately 6 months due to the emergence of drug resistance. While gilteritinib eliminates blasts in peripheral circulation, residual blasts in the bone marrow microenvironment are protected by cytokines and growth factors. Persistence of these residual cells represents early resistance to treatment. How these cells adapt to survive in the marrow microenvironment remains unclear, but over time resistant subclones resume growth and lead to relapsed disease. At relapse, many patients have intrinsic resistance mutations, what we term late resistance. In this study, we used a stepwise model that charts the temporal evolution of early to late gilteritinib resistance. To recapitulate early resistance, we cultured the FLT3-ITD+ AML cell lines, MOLM-14 and MV4;11, with exogenous microenvironmental ligands that allow cells to become resistant to gilteritinib in a ligand-dependent manner. After 7 weeks, all cultures with ligand resumed growth (early resistance), whereas cells without ligand never resumed growth. Following ligand withdrawal, the cells become transiently sensitive to gilteritinib but resumed growth after 2 months (late resistance). We comprehensively analyzed early and late resistance by integrating whole exome sequencing, CRISPR/Cas9 screening, proteomics, metabolomics, and small-molecule inhibitor screening. Early resistance is characterized by slowly dividing cells and metabolic reprogramming, particularly with respect to lipid metabolism. Early resistant cultures also became uniquely dependent on Aurora kinase B (AURKB) for survival. We then validated these findings in primary AML cells from patients (N=11) treated with gilteritinib and found that early resistant cells demonstrated reduced cell cycle and alterations in lipid metabolism. Gene expression analysis of sequential stromal cell samples from AML patients (N=13) pre- and post gilteritinib treatment showed an increase in lipid metabolism following gilteritinib treatment, indicating that the microenvironment is also dynamic and in crosstalk with neighboring AML cells. Primary early resistant AML cells also became dependent on AURKB signaling, and were exquisitely sensitive to the combination of AURKB inhibitors and gilteritinib. In contrast, late resistance is driven by an expansion of pre-existing NRAS mutant subclones, consistent with the resistance profile of AML patients on gilteritinib. Metabolic reprogramming continued to evolve in late resistance with further dependence upon lipid metabolism. Our study provides mechanistic understanding of how the marrow microenvironment contributes to extrinsic early resistance, which then leads to late intrinsic resistance. We also define a unique vulnerability to AURKB inhibitors in early resistance that may thwart the expansion of late resistant NRAS subclones.

Citation Format: Sunil K. Joshi, Tamilla Nechiporuk, Daniel Bottomly, Paul Piehowski, Julie A. Reisz, Janét Pittsenbarger, Andy Kaempf, Sara J.C. Gosline, Yi-Ting Wang, Tao Liu, Cristina E. Tognon, Angelo D’Alessandro, Jeffrey W. Tyner, Shannon K. McWeeney, Karin D. Rodland, Brian J. Druker, Elie Traer. The AML microenvironment catalyzes a step-wise evolution to gilteritinib resistance [abstract]. In: Proceedings of the AACR Virtual Special Conference on the Evolving Tumor Microenvironment in Cancer Progression: Mechanisms and Emerging Therapeutic Opportunities; in association with the Tumor Microenvironment (TME) Working Group; 2021 Jan 11-12. Philadelphia (PA): AACR; Cancer Res 2021;81(5 Suppl):Abstract nr LT022.

Surfactant-assisted one-pot sample preparation for label-free single-cell proteomics

Large numbers of cells are generally required for quantitative global proteome profiling due to surface adsorption losses associated with sample processing. Such bulk measurement obscures important cell-to-cell variability (cell heterogeneity) and makes proteomic profiling impossible for rare cell populations (e.g., circulating tumor cells (CTCs)). Here we report a surfactant-assisted one-pot sample preparation coupled with mass spectrometry (MS) method termed SOP-MS for label-free global single-cell proteomics. SOP-MS capitalizes on the combination of a MS-compatible nonionic surfactant, n-Dodecyl-β-D-maltoside, and hydrophobic surface-based low-bind tubes or multi-well plates for ‘all-in-one’ one-pot sample preparation. This ‘all-in-one’ method including elimination of all sample transfer steps maximally reduces surface adsorption losses for effective processing of single cells, thus improving detection sensitivity for single-cell proteomics. This method allows convenient label-free quantification of hundreds of proteins from single human cells and ~1200 proteins from small tissue sections (close to ~20 cells). When applied to a patient CTC-derived xenograft (PCDX) model at the single-cell resolution, SOP-MS can reveal distinct protein signatures between primary tumor cells and early metastatic lung cells, which are related to the selection pressure of anti-tumor immunity during breast cancer metastasis. The approach paves the way for routine, precise, quantitative single-cell proteomics.

Tsai, Zhang, Scholten et al. develop a surfactant- assisted one-pot sample preparation coupled with mass spectrometry method (SOP-MS) for label-free global single-cell proteomics. This method allows researchers to measure hundreds of proteins from single human cells, suggesting its utility for quantitative single-cell proteomics.

Integrated Proteogenomic Characterization across Major Histological Types of Pediatric Brain Cancer

We report a comprehensive proteogenomics analysis, including whole-genome sequencing, RNA sequencing, and proteomics and phosphoproteomics profiling, of 218 tumors across 7 histological types of childhood brain cancer: low-grade glioma (n = 93), ependymoma (32), high-grade glioma (25), medulloblastoma (22), ganglioglioma (18), craniopharyngioma (16), and atypical teratoid rhabdoid tumor (12). Proteomics data identify common biological themes that span histological boundaries, suggesting that treatments used for one histological type may be applied effectively to other tumors sharing similar proteomics features. Immune landscape characterization reveals diverse tumor microenvironments across and within diagnoses. Proteomics data further reveal functional effects of somatic mutations and copy number variations (CNVs) not evident in transcriptomics data. Kinase-substrate association and co-expression network analysis identify important biological mechanisms of tumorigenesis. This is the first large-scale proteogenomics analysis across traditional histological boundaries to uncover foundational pediatric brain tumor biology and inform rational treatment selection.

Integrated Proteomic and Glycoproteomic Characterization of Human High-Grade Serous Ovarian Carcinoma

Many gene products exhibit great structural heterogeneity because of an array of modifications. These modifications are not directly encoded in the genomic template but often affect the functionality of proteins. Protein glycosylation plays a vital role in proper protein functions. However, the analysis of glycoproteins has been challenging compared with other protein modifications, such as phosphorylation. Here, we perform an integrated proteomic and glycoproteomic analysis of 83 prospectively collected high-grade serous ovarian carcinoma (HGSC) and 23 non-tumor tissues. Integration of the expression data from global proteomics and glycoproteomics reveals tumor-specific glycosylation, uncovers different glycosylation associated with three tumor clusters, and identifies glycosylation enzymes that were correlated with the altered glycosylation. In addition to providing a valuable resource, these results provide insights into the potential roles of glycosylation in the pathogenesis of HGSC, with the possibility of distinguishing pathological outcomes of ovarian tumors from non-tumors, as well as classifying tumor clusters.

A Review of Imputation Strategies for Isobaric Labeling-Based Shotgun Proteomics

The throughput efficiency and increased depth of coverage provided by isobaric-labeled proteomics measurements have led to increased usage of these techniques. However, the structure of missing data is different than unlabeled studies, which prompts the need for this review to compare the efficacy of nine imputation methods on large isobaric-labeled proteomics data sets to guide researchers on the appropriateness of various imputation methods. Imputation methods were evaluated by accuracy, statistical hypothesis test inference, and run time. In general, expectation maximization and random forest imputation methods yielded the best performance, and constant-based methods consistently performed poorly across all data set sizes and percentages of missing values. For data sets with small sample sizes and higher percentages of missing data, results indicate that statistical inference with no imputation may be preferable. On the basis of the findings in this review, there are core imputation methods that perform better for isobaric-labeled proteomics data, but great care and consideration as to whether imputation is the optimal strategy should be given for data sets comprised of a small number of samples.

Proteomic Analysis of Exosomes for Discovery of Protein Biomarkers for Prostate and Bladder Cancer

Extracellular vesicles (EVs) are released by nearly all cell types as part of normal cell physiology, transporting biological cargo, including nucleic acids and proteins, across the cell membrane. In pathological states such as cancer, EV-derived cargo may mirror the altered state of the cell of origin. Exosomes are the smaller, 50–150 nanometer-sized EVs released from fusion of multivesicular endosomes with the plasma membrane. Exosomes play important roles in cell-cell communication and participate in multiple cancer processes, including invasion and metastasis. Therefore, proteomic analysis of exosomes is a promising approach to discover potential cancer biomarkers, even though it is still at an early stage. Herein, we critically review the advances in exosome isolation methods and their compatibility with mass spectrometry (MS)-based proteomic analysis, as well as studies of exosomes in pathogenesis and progression of prostate and bladder cancer, two common urologic cancers whose incidence rates continue to rise annually. As urological tumors, both urine and blood samples are feasible for noninvasive or minimally invasive analysis. A better understanding of the biological cargo and functions of exosomes via high-throughput proteomics will help provide new insights into complex alterations in cancer and provide potential therapeutic targets and personalized treatment for patients.

Abstract 5125: Evaluation of differential peptide loading on TMT-based proteomic on phosphoproteomic data quality in an AML model

Mass Spectrometry based proteomics profiling has become a powerful tool for broad quantification of proteins and their post-translational modifications for cancer research. Due to extensive efforts in the field to benchmark and standardize complex workflows, particularly those of the Clinical Proteomics Tumor Analysis Consortium (CPTAC), proteomics is now moving into clinical research. Reliable and reproducible quantification of &gt; 10,000 proteins and &gt;30,000 phosphosites is now routinely attainable from mammalian tissue samples. Integration of deep-scale proteomics analysis of human tumors with genomic data has been shown to improve specificity for identifying pathway alterations caused by tumor associated mutations. Furthermore, phosphoproteome measurements provide information on pathway activation not discernible from genetic measurements and thus provides unique insights for potential therapeutic targets.

A substantial challenge in the field of clinical proteomics is obtaining the sufficiently large clinical specimens necessary for deep coverage of the proteome. This challenge is particularly acute in the case of phosphoproteomics, which requires 100-fold more material than global profiling due to the low stoichiometry of the modification. The current gold standard approach for clinical proteomics employs a tandem mass tags (TMT) isobaric labeling approach to achieve deep quantitative proteomic and phosphoproteomic measurements. In this approach, 10 or 11 patient samples are labeled with unique isobaric labels, mixed in equal proportion, and fractionated prior to liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. This creates a second challenge since very often due to external variables, differing quantities of protein are obtainable for patients within the same study. In these cases, the analysts need to decide whether to exclude sample-limited patients, reduce the protein loading per patient for the study, or to include that individual patient at reduced protein loading.

In this study, we set out to interrogate the impact of differential channel loading on global and phosphoproteomics results using an established clinical workflow. First, we determined protein yields from patient samples of decreasing cell count. We then used these results to design a 2 TMT plex experiment to explore a range of peptide loadings that we expect to encounter in executing a clinical proteomics experiment. The results were examined for protein/phosphosite coverage and various aspects of quantitative reproducibility. Our results provide a thorough examination of the impacts of differential channel loading and provide researchers with a resource to make informed decisions concerning their study design.

Citation Format: Paul D. Piehowski, Yang Wang, James A. Sanford, Joshua R. Hansen, Marina A. Gritsenko, Vladislav A. Petyuk, Karl K. Weitz, Cristina Tognon, Wei-Jun Qian, Tao Liu, Brian J. Druker, Karin D. Rodland. Evaluation of differential peptide loading on TMT-based proteomic on phosphoproteomic data quality in an AML model [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5125.

Abstract 5942: An integrated approach reveals novel extrinsic pathways and cytokine signatures associated with drug response in acute myeloid leukemia

Background: Acute myeloid leukemia (AML) is a heterogeneous disease with a dismal 5 year survival rate below 30%. Current AML therapies have provided little improvement in achieving complete remission. This is attributed to the development of drug resistance and extrinsic factors from the microenvironment that promote AML progression. Previously, our lab demonstrated that proinflammatory cytokines from the bone marrow microenvironment such as IL-1, promote AML progression. This emphasizes the rationale to design and implement targeted therapies to block extrinsic pathways coupled with intrinsic pathways conferring drug resistance in AML.

Methods: To identify factors that promote drug sensitivity and resistance in AML, we used a cohort of 350 AML patients with various genetic subtypes. We quantified levels of 41 growth factors in plasma samples using a multiplex luminex assay. The same cohort of AML samples were analyzed for drug response to 130 small molecule inhibitors in in vitro functional drug sensitivity assay, for RNA-seq gene expression, and whole exome sequencing. The data was integrated to identify differential pathways and markers for drug response. To model drug response in vitro we created a series of drug resistant AML cell lines by culturing these cells in specific drugs long-term.

Results: Pathway analysis integrating drug sensitivity, transcriptome, and cytokine expression data identified distinct differentially regulated pathways for various inhibitors. We found that protein synthesis pathways were significantly enriched in sorafenib (FLT-3 inhibitor) resistant samples whereas cytokine and immune pathways (such as TLR and MCP1 signaling) were correlated with trametinib (MEK inhibitor) and venetoclax (Bcl-2 inhibitor) resistance. Specifically, venetoclax resistance correlated with augmented levels of cytokines such as IL-1 and IP-10 and growth factors such as PDGF. Further, trametinib resistance in AML samples is correlated with the upregulation of CCL11, MCP1, and TGFα but a downregulation of CCL22 levels. Accordingly, using AML cell line models, we observed an increase in MCP1 and TGFα at the transcript and secreted protein levels in trametinib resistant cell lines. We demonstrated that MCP1 stimulation activates survival pathways by activating pERK and pJNK signaling, thus reducing apoptosis in AML cells. Targeting MCP1 in combination with trametinib reverses these effects, thus offering a novel therapeutic approach to overcome drug resistance.

Conclusion and perspectives: Our data suggest that distinct extrinsic pathways may regulate the response to specific targeted therapy, thus providing a strong basis to design new treatment regimens. Combination treatment incorporating extrinsic pathways would aid in sensitizing AML cells to therapy. Our study offers an integrated approach and a resource to identify such pathways for a large number of tested drugs, and narrows down functionally important pathways associated with drug response.

Citation Format: Rucha V. Modak, Alisa Damnernsawad, Ted Laderas, Guanming Wu, Jeffery W. Tyner, Karin D. Rodland, Shannon K. McWeeney, Anupriya Agarwal. An integrated approach reveals novel extrinsic pathways and cytokine signatures associated with drug response in acute myeloid leukemia [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5942.

Abstract 6580: Proteogenomic characterization of endometrial carcinoma

Endometrial cancer (EC) is the sixth most common cancer in women globally and is one of the few cancers for which mortality is increasing; to address the need to develop new methods for the diagnosis and treatment of this disease, we performed a comprehensive proteogenomic characterization of 95 endometrial carcinomas.

We prospectively collected 83 endometrioid and 12 serous tumors, along with 49 normal tissue samples, for multi-omic characterization. Each sample underwent whole exome sequencing, whole genome sequencing, total RNA and miRNA sequencing, and DNA methylation array analyses. In addition, we evaluated all samples for protein, phosphorylation, and acetylation levels using an integrated, isobaric tandem mass tag labeling–based quantitative proteomics workflow. This dataset, publicly available through the Genomic Data Commons, CPTAC data portal, and other sources, provides a valuable resource for researchers and clinicians.

Our analyses of these tumors revealed, among other things, a potential role for circRNAs in the epithelial-mesenchymal transition and new information about proteomic markers of clinical and genomic tumor subgroups, including relationships to known druggable pathways. We found possible new consequences of perturbations to the Wnt/β-catenin pathway, particularly CTNNB1 hotspot mutations; moreover, from the genome-wide acetylation survey, we found potential regulatory mechanisms linking Wnt signaling and histone acetylation. Phosphorylation data analysis suggests an elevated DNA damage response in serous tumors, which we hypothesize may be targetable by certain FDA-approved drugs. We found that TP53 hotspot mutations play unique trans-acting roles as compared to the same mutations in colon and ovarian cancers. We also characterized aspects of the tumor immune landscape, including immunogenic alterations, neoantigens, common cancer/testis antigens, and the immune microenvironment, all of which can inform immunotherapy decisions; of particular translational interest, we identified a subset of patients who, despite having high amounts of microsatellite instability, exhibit biomarkers which suggest that immunotherapy would be an ineffective treatment for their disease.

By integrating global measurements of protein abundance, phosphorylation, and acetylation with next-generation genomics and transcriptomics, we have provided novel insights into biological processes underlying the development of endometrial cancers and generated strong hypotheses for new approaches to personalized treatment of individuals with endometrial cancer.

Citation Format: Emily A. Kawaler, Yongchao Dou, Daniel Cui Zhou, Marina A. Gritsenko, Karin D. Rodland, Li Ding, Bing Zhang, Tao Liu, David Fenyö, Clinical Proteomic Tumor Analysis Consortium. Proteogenomic characterization of endometrial carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 6580.

Detection of Head and Neck Cancer Based on Longitudinal Changes in Serum Protein Abundance

BACKGROUND

Approximately 85% of the U.S. military active duty population is male and less than 50 years of age, with elevated levels of known risk factors for oropharyngeal squamous cell carcinoma (OPSCC), including smoking, excessive use of alcohol, and greater numbers of sexual partners, and elevated prevalence of human papilloma virus (HPV). Given the recent rise in incidence of OPSCC related to the HPV, the Department of Defense Serum Repository provides an unparalleled resource for longitudinal studies of OPSCC in the military for the identification of early detection biomarkers.

METHODS

We identified 175 patients diagnosed with OPSCC with 175 matched healthy controls and retrieved a total of 978 serum samples drawn at the time of diagnosis, 2 and 4 years prior to diagnosis, and 2 years after diagnosis. Following immunoaffinity depletion, serum samples were analyzed by targeted proteomics assays for multiplexed quantification of a panel of 146 candidate protein biomarkers from the curated literature.

RESULTS

Using a Random Forest machine learning approach, we derived a 13-protein signature that distinguishes cases versus controls based on longitudinal changes in serum protein concentration. The abundances of each of the 13 proteins remain constant over time in control subjects. The AUC for the derived Random Forest classifier was 0.90.

CONCLUSIONS

This 13-protein classifier is highly promising for detection of OPSCC prior to overt symptoms.

IMPACT

Use of longitudinal samples has significant potential to identify biomarkers for detection and risk stratification.

Proteomic Tissue-Based Classifier for Early Prediction of Prostate Cancer Progression

Although ~40% of screen-detected prostate cancers (PCa) are indolent, advanced-stage PCa is a lethal disease with 5-year survival rates around 29%. Identification of biomarkers for early detection of aggressive disease is a key challenge. Starting with 52 candidate biomarkers, selected from existing PCa genomics datasets and known PCa driver genes, we used targeted mass spectrometry to quantify proteins that significantly differed in primary tumors from PCa patients treated with radical prostatectomy (RP) across three study outcomes: (i) metastasis ≥1-year post-RP, (ii) biochemical recurrence ≥1-year post-RP, and (iii) no progression after ≥10 years post-RP. Sixteen proteins that differed significantly in an initial set of 105 samples were evaluated in the entire cohort (n = 338). A five-protein classifier which combined FOLH1, KLK3, TGFB1, SPARC, and CAMKK2 with existing clinical and pathological standard of care variables demonstrated significant improvement in predicting distant metastasis, achieving an area under the receiver-operating characteristic curve of 0.92 (0.86, 0.99, p = 0.001) and a negative predictive value of 92% in the training/testing analysis. This classifier has the potential to stratify patients based on risk of aggressive, metastatic PCa that will require early intervention compared to low risk patients who could be managed through active surveillance.

An Improved Boosting to Amplify Signal with Isobaric Labeling (iBASIL) Strategy for Precise Quantitative Single-cell Proteomics

Mass spectrometry (MS)-based proteomics has great potential for overcoming the limitations of antibody-based immunoassays for antibody-independent, comprehensive, and quantitative proteomic analysis of single cells. Indeed, recent advances in nanoscale sample preparation have enabled effective processing of single cells. In particular, the concept of using boosting/carrier channels in isobaric labeling to increase the sensitivity in MS detection has also been increasingly used for quantitative proteomic analysis of small-sized samples including single cells. However, the full potential of such boosting/carrier approaches has not been significantly explored, nor has the resulting quantitation quality been carefully evaluated. Herein, we have further evaluated and optimized our recent boosting to amplify signal with isobaric labeling (BASIL) approach, originally developed for quantifying phosphorylation in small number of cells, for highly effective analysis of proteins in single cells. This improved BASIL (iBASIL) approach enables reliable quantitative single-cell proteomics analysis with greater proteome coverage by carefully controlling the boosting-to-sample ratio (e.g. in general <100×) and optimizing MS automatic gain control (AGC) and ion injection time settings in MS/MS analysis (e.g. 5E5 and 300 ms, respectively, which is significantly higher than that used in typical bulk analysis). By coupling with a nanodroplet-based single cell preparation (nanoPOTS) platform, iBASIL enabled identification of ∼2500 proteins and precise quantification of ∼1500 proteins in the analysis of 104 FACS-isolated single cells, with the resulting protein profiles robustly clustering the cells from three different acute myeloid leukemia cell lines. This study highlights the importance of carefully evaluating and optimizing the boosting ratios and MS data acquisition conditions for achieving robust, comprehensive proteomic analysis of single cells.

Proteogenomic Characterization of Ovarian HGSC Implicates Mitotic Kinases, Replication Stress in Observed Chromosomal Instability

In the absence of a dominant driving mutation other than uniformly present TP53 mutations, deeper understanding of the biology driving ovarian high-grade serous cancer (HGSC) requires analysis at a functional level, including post-translational modifications. Comprehensive proteogenomic and phosphoproteomic characterization of 83 prospectively collected ovarian HGSC and appropriate normal precursor tissue samples (fallopian tube) under strict control of ischemia time reveals pathways that significantly differentiate between HGSC and relevant normal tissues in the context of homologous repair deficiency (HRD) status. In addition to confirming key features of HGSC from previous studies, including a potential survival-associated signature and histone acetylation as a marker of HRD, deep phosphoproteomics provides insights regarding the potential role of proliferation-induced replication stress in promoting the characteristic chromosomal instability of HGSC and suggests potential therapeutic targets for use in precision medicine trials.

Comparison of ovarian cancer and normal precursors identifies key signaling pathways

Mitotic and cyclin-dependent kinases emerge as potential therapeutic targets

Previously identified hallmarks of homologous repair status and survival are confirmed

Replication stress appears to drive increased chromosomal instability

Proteogenomic Characterization of Endometrial Carcinoma

We undertook a comprehensive proteogenomic characterization of 95 prospectively collected endometrial carcinomas, comprising 83 endometrioid and 12 serous tumors. This analysis revealed possible new consequences of perturbations to the p53 and Wnt/β-catenin pathways, identified a potential role for circRNAs in the epithelial-mesenchymal transition, and provided new information about proteomic markers of clinical and genomic tumor subgroups, including relationships to known druggable pathways. An extensive genome-wide acetylation survey yielded insights into regulatory mechanisms linking Wnt signaling and histone acetylation. We also characterized aspects of the tumor immune landscape, including immunogenic alterations, neoantigens, common cancer/testis antigens, and the immune microenvironment, all of which can inform immunotherapy decisions. Collectively, our multi-omic analyses provide a valuable resource for researchers and clinicians, identify new molecular associations of potential mechanistic significance in the development of endometrial cancers, and suggest novel approaches for identifying potential therapeutic targets.

An Integrative Analysis of Tumor Proteomic and Phosphoproteomic Profiles to Examine the Relationships Between Kinase Activity and Phosphorylation

Phosphorylation of proteins is a key way cells regulate function, both at the individual protein level and at the level of signaling pathways. Kinases are responsible for phosphorylation of substrates, generally on serine, threonine, or tyrosine residues. Though particular sequence patterns can be identified that dictate whether a residue will be phosphorylated by a specific kinase, these patterns are not highly predictive of phosphorylation. The availability of large scale proteomic and phosphoproteomic data sets generated using mass-spectrometry-based approaches provides an opportunity to study the important relationship between kinase activity, substrate specificity, and phosphorylation. In this study, we analyze relationships between protein abundance and phosphopeptide abundance across more than 150 tumor samples and show that phosphorylation at specific phosphosites is not well correlated with overall kinase abundance. However, individual kinases show a clear and statistically significant difference in correlation among known phosphosite targets for that kinase and randomly selected phosphosites. We further investigate relationships between phosphorylation of known activating or inhibitory sites on kinases and phosphorylation of their target phosphosites. Combined with motif-based analysis, this approach can predict novel kinase targets and show which subsets of a kinase's target repertoire are specifically active in one condition versus another.

Abstract 1620: Validation of a proteomic signature of lung cancer risk from bronchial brushings

A major challenge in lung cancer prevention and cure hinges on identifying the at-risk population who ultimately develops lung cancer. Previously we reported proteomic alterations in the cytologically normal bronchial epithelial cells collected from the bronchial brushings of individuals at-risk for lung cancer. Proteins were identified by shotgun proteomics using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and selected candidate biomarkers of risk were validated in an independent sample cohort by parallel reaction monitoring. The purpose of the current study is to validate, in independent cohorts from Vanderbilt University Medical Center, a selected list of 54 candidate proteins associated with risk for lung cancer with sensitive targeted proteomics using selected reaction monitoring (SRM) at the Pacific Northwest National Laboratory. We tested proteins from bronchial brushings and bronchial biopsies collected from individuals at low risk (n=10) and high risk (n=10) based on the Tammemagi risk stratification model as well as from patients with lung cancer (n=10). Multiplexed LC-SRM assays were developed for all 54 proteins and the protein concentrations were determined by LC-SRM using heavy isotope-labeled peptides for 44 and 49 proteins in the brushings and biopsy samples, respectively. Pairwise comparisons by Wilcoxon rank sum test demonstrated significant overexpression, in the high-risk group compared to the low-risk group, of ALDH3A1 and AKR1B10 in both brushings and biopsies, and of SFN/YWHAS only in the biopsies. Significant overexpression of ALDH1A1, ALDH3A1, AKR1B10, ANXA1, CTNNB1, G6PD, LGALS7B, ALOX15, PGK1, SFN/YWHAS, and UBA1 were observed in the lung cancer group compared to the low-risk group in the brushings only. LGALS7B was also overexpressed in the brushings of high risk compared to lung cancer group. In conclusion, these SRM results revealed promise of selected candidate proteins to stratify the at-risk population for lung cancer. Next, our validation efforts will test the promising candidates in second, larger independent cohort; methods with increased sensitivity such as PRISM (high-pressure, high-resolution separations with intelligent selection and multiplexing)-SRM for analysis of the previously undetected low abundant protein candidates and carrier-assisted SRM approaches for analysis of very small-sized samples. Ultimately, we hope to deliver a signature of risk that may provide the basis for lung cancer risk assessment and possibly novel future prevention strategies. This work is supported by UO1CA152662.

Citation Format: SM Jamshedur Rahman, Yuqian Gao, Athena A. Schepmoes, Yi-Ting Wang, Tujin Shi, Sheau-Chiann Chen, Heidi Chen, Karin D. Rodland, Tao Liu, Pierre P. Massion. Validation of a proteomic signature of lung cancer risk from bronchial brushings [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1620.

Abstract LB-006: Proteogenomic characterization of human colon cancer reveals new therapeutic opportunities

We prospectively collected matched tumor specimens, adjacent non-tumor tissues, and blood samples from 110 colon cancer patients and analyzed the samples using seven omics platforms, including whole-exome sequencing, copy number arrays, RNA-Seq, miRNA-Seq, label-free global proteomics, isobaric tandem mass tag (TMT) labeling-based global proteomics, and TMT-based phosphoproteomics. Comparative proteomic and phosphoproteomic analysis of paired tumor and adjacent normal samples produced the first comprehensive catalogue of colon cancer-associated proteins and phosphosites, including known and putative new biomarkers and drug targets. These cancer-associated proteins and phosphosites had very little overlap with known cancer genes in the Cancer Gene Census, providing a novel information layer to our knowledge about colon cancer. One notable finding in differential proteome analysis is the identification of several cancer/testis antigens that were recurrently over-expressed in tumors compared to adjacent normal tissue, including IGF2BP3 (51%), SPAG1 (14%), and ATAD2 (8%). Through integrative analysis of the whole-exome sequencing, RNA-Seq, and proteomics data, we further predicted personalized neoantigens for 38% of the patients. In total, we found proteomics-supported neoantigens or cancer/testis antigens for 78% of the tumors in this cohort, demonstrating the potential of proteogenomics in identifying tumor antigens for cancer vaccine development. Proteomics data complemented somatic copy number analysis results and showed that multiple somatic copy number deletion events converge to repress the endocytosis pathway, suggesting its tumor suppressor role in colon cancer. In addition to reinforcing or complementing genomic findings, proteogenomic integration may also contradict genomics data-based inferences and lead to unexpected discoveries and therapeutic opportunities. Proteomics data identified SOX9 as an oncogene in colon cancer, whereas it was predicted to be a tumor suppressor based on somatic mutation data in the TCGA study. Phosphoproteomics data revealed a dual role of Rb phosphorylation in promoting proliferation and repressing apoptosis in colon cancer, clarifying the long-standing puzzle of colon cancer-specific amplification of this tumor suppressor and highlighting a unique opportunity for targeting Rb phosphorylation in colon cancer. Microsatellite instability status has been approved by the FDA as a biomarker for selecting patients for checkpoint inhibitor therapy in colorectal and other solid tumors. However, many MSI-high tumors fail to respond to checkpoint inhibition. Our proteogenomic analysis identified a subtype-specific association between increased glycolysis and decreased CD8 T cell infiltration in MSI-high colon tumors, suggesting glycolysis as a target for overcoming immune evasion in this MSI-H tumors. We make the primary and processed datasets available in publicly accessible data repositories and portals to allow broad use of these datasets for new biological discoveries and therapeutic hypothesis generation.

Citation Format: Bing Zhang, Suhas Vasaikar, Chen Huang, Xiaojing Wang, Vladislav A. Petyuk, Sara R. Savage, Bo Wen, Yongchao Dou, Yun Zhang, Zhiao Shi, Osama A. Arshad, Marina A. Gritsenko, Lisa J. Zimmerman, Jason E. McDermott, Therese R. Clauss, Ronald J. Moore, Rui Zhao, Matthew E. Monroe, Yi-Ting Wang, Matthew C. Chambers, Robbert J. Slebos, Ken S. Lau, Qianxing Mo, Li Ding, Matthew Ellis, Mathangi Thiagarajan, Christopher R. Kinsinger, Henry Rodriguez, Richard D. Smith, Karin D. Rodland, Daniel C. Liebler, Tao Liu, CPTAC Investigators. Proteogenomic characterization of human colon cancer reveals new therapeutic opportunities [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-006.

Abstract 4538: Global and phosphoproteomic analysis of AML cell line response to phosphatase inhibitor treatment

In the United States, approximately 21,000 people are diagnosed with AML each year and over 10,000 AML related deaths are reported. Genomic analysis has revealed at least 11 genetic classes of AML, and more than 20 subsets can be assigned when considering the differentiation states of Leukaemic blasts. This variability has generated significant interest in the development of targeted therapies for AML. However, complex mutational patterns and a lack of pharmacological agents for most mutational events have challenged the development of effective treatments. Recent results from the Beat AML clinical trial revealed that response to drugs is associated with mutational status, including instances where drug sensitivity is specific to combinatorial mutational events.

To better understand the molecular mechanisms underpinning drug response, we performed global and phosphoproteomic analyses on 4 AML cell lines that carry mutations associated with AML: K562 with a BCR-ABL fusion, MOLM14 with a FLT3 mutation, HL60 with a RAS mutation, and CMK with a JAK3 mutation. These cell lines were each treated with a different kinase inhibitor drug for which they have a known sensitivity. Cell samples were collected prior to treatment and time points of 30 min, 3h, and 16 hrs. Frozen cell pellets were prepared for analysis using the workflow for multiplexed, deep-scale, proteome and phosphoproteome analysis developed by the NCI CPTAC program.

The goal of our Proteogenomic Translational Research Center is to develop drug response signatures allowing the use of baseline proteome and phosphoproteome measurements to predict patient drug response, providing a foundation for rational selection of combination therapies. Preliminary analysis of the data show a dynamic temporal response to drug treatment in both the global and phosphoproteome, Figure 1. Current efforts are focused on combining knowledge-driven and data-driven analysis approaches to identify the pathways associated with drug response.

Citation Format: Paul D. Piehowski, Jason E. McDermott, Joshua R. Hansen, Samantha L. Savage, Cristina E. Tognon, Anupriya Agarwal, Jeffrey W. Tyner, Brian J. Druker, Karin D. Rodland. Global and phosphoproteomic analysis of AML cell line response to phosphatase inhibitor treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4538.

Boosting to Amplify Signal with Isobaric Labeling (BASIL) Strategy for Comprehensive Quantitative Phosphoproteomic Characterization of Small Populations of Cells

Comprehensive phosphoproteomic analysis of small populations of cells remains a daunting task due primarily to the insufficient MS signal intensity from low concentrations of enriched phosphopeptides. Isobaric labeling has a unique multiplexing feature where the "total" peptide signal from all channels (or samples) triggers MS/MS fragmentation for peptide identification, while the reporter ions provide quantitative information. In light of this feature, we tested the concept of using a "boosting" sample (e.g., a biological sample mimicking the study samples but available in a much larger quantity) in multiplexed analysis to enable sensitive and comprehensive quantitative phosphoproteomic measurements with <100 000 cells. This simple boosting to amplify signal with isobaric labeling (BASIL) strategy increased the overall number of quantifiable phosphorylation sites more than 4-fold. Good reproducibility in quantification was demonstrated with a median CV of 15.3% and Pearson correlation coefficient of 0.95 from biological replicates. A proof-of-concept experiment demonstrated the ability of BASIL to distinguish acute myeloid leukemia cells based on the phosphoproteome data. Moreover, in a pilot application, this strategy enabled quantitative analysis of over 20 000 phosphorylation sites from human pancreatic islets treated with interleukin-1β and interferon-γ. Together, this signal boosting strategy provides an attractive solution for comprehensive and quantitative phosphoproteome profiling of relatively small populations of cells where traditional phosphoproteomic workflows lack sufficient sensitivity.

Proteogenomic Analysis of Human Colon Cancer Reveals New Therapeutic Opportunities

We performed the first proteogenomic study on a prospectively collected colon cancer cohort. Comparative proteomic and phosphoproteomic analysis of paired tumor and normal adjacent tissues produced a catalog of colon cancer-associated proteins and phosphosites, including known and putative new biomarkers, drug targets, and cancer/testis antigens. Proteogenomic integration not only prioritized genomically inferred targets, such as copy-number drivers and mutation-derived neoantigens, but also yielded novel findings. Phosphoproteomics data associated Rb phosphorylation with increased proliferation and decreased apoptosis in colon cancer, which explains why this classical tumor suppressor is amplified in colon tumors and suggests a rationale for targeting Rb phosphorylation in colon cancer. Proteomics identified an association between decreased CD8 T cell infiltration and increased glycolysis in microsatellite instability-high (MSI-H) tumors, suggesting glycolysis as a potential target to overcome the resistance of MSI-H tumors to immune checkpoint blockade. Proteogenomics presents new avenues for biological discoveries and therapeutic development.

Clinical potential of mass spectrometry-based proteogenomics

Cancer genomics research aims to advance personalized oncology by finding and targeting specific genetic alterations associated with cancers. In genome-driven oncology, treatments are selected for individual patients on the basis of the findings of tumour genome sequencing. This personalized approach has prolonged the survival of subsets of patients with cancer. However, many patients do not respond to the predicted therapies based on the genomic profiles of their tumours. Furthermore, studies pairing genomic and proteomic analyses of samples from the same tumours have shown that the proteome contains novel information that cannot be discerned through genomic analysis alone. This observation has led to the concept of proteogenomics, in which both types of data are leveraged for a more complete view of tumour biology that might enable patients to be more successfully matched to effective treatments than they would using genomics alone. In this Perspective, we discuss the added value of proteogenomics over the current genome-driven approach to the clinical characterization of cancers and summarize current efforts to incorporate targeted proteomic measurements based on selected/multiple reaction monitoring (SRM/MRM) mass spectrometry into the clinical laboratory to facilitate clinical proteogenomics.

P-Mart: Interactive Analysis of Ion Abundance Global Proteomics Data

The use of mass-spectrometry-based techniques for global protein profiling of biomedical or environmental experiments has become a major focus in research centered on biomarker discovery; however, one of the most important issues recently highlighted in the new era of omics data generation is the ability to perform analyses in a robust and reproducible manner. This has been hypothesized to be one of the issues hindering the ability of clinical proteomics to successfully identify clinical diagnostic and prognostic biomarkers of disease. P-Mart ( https://pmart.labworks.org ) is a new interactive web-based software environment that enables domain scientists to perform quality-control processing, statistics, and exploration of large-complex proteomics data sets without requiring statistical programming. P-Mart is developed in a manner that allows researchers to perform analyses via a series of modules, explore the results using interactive visualization, and finalize the analyses with a collection of output files documenting all stages of the analysis and a report to allow reproduction of the analysis.

Glomerular filtrate proteins in acute cardiorenal syndrome

Acute cardiorenal syndrome (CRS-1) is a morbid complication of acute cardiovascular disease. Heart-to-kidney signals transmitted by "cardiorenal connectors" have been postulated, but investigation into CRS-1 has been limited by technical limitations and a paucity of models. To address these limitations, we developed a translational model of CRS-1, cardiac arrest and cardiopulmonary resuscitation (CA/CPR), and now report findings from nanoscale mass spectrometry proteomic exploration of glomerular filtrate 2 hours after CA/CPR or sham procedure. Filtrate acquisition was confirmed by imaging, molecular weight and charge distribution, and exclusion of protein specific to surrounding cells. Filtration of proteins specific to the heart was detected following CA/CPR and confirmed with mass spectrometry performed using urine collections from mice with deficient tubular endocytosis. Cardiac LIM protein was a CA/CPR-specific filtrate component. Cardiac arrest induced plasma release of cardiac LIM protein in mice and critically ill human cardiac arrest survivors, and administration of recombinant cardiac LIM protein to mice altered renal function. These findings demonstrate that glomerular filtrate is accessible to nanoscale proteomics and elucidate the population of proteins filtered 2 hours after CA/CPR. The identification of cardiac-specific proteins in renal filtrate suggests a novel signaling mechanism in CRS-1. We expect these findings to advance understanding of CRS-1.

Carrier-Assisted Single-Tube Processing Approach for Targeted Proteomics Analysis of Low Numbers of Mammalian Cells

Heterogeneity in composition is inherent in all cell populations, even those containing a single cell type. Single-cell proteomics characterization of cell heterogeneity is currently achieved by antibody-based technologies, which are limited by the availability of high-quality antibodies. Herein we report a simple, easily implemented, mass spectrometry (MS)-based targeted proteomics approach, termed cLC-SRM (carrier-assisted liquid chromatography coupled to selected reaction monitoring), for reliable multiplexed quantification of proteins in low numbers of mammalian cells. We combine a new single-tube digestion protocol to process low numbers of cells with minimal loss together with sensitive LC-SRM for protein quantification. This single-tube protocol builds upon trifluoroethanol digestion and further minimizes sample losses by tube pretreatment and the addition of carrier proteins. We also optimized the denaturing temperature and trypsin concentration to significantly improve digestion efficiency. cLC-SRM was demonstrated to have sufficient sensitivity for reproducible detection of most epidermal growth factor receptor (EGFR) pathway proteins expressed at levels ≥30 000 and ≥3000 copies per cell for 10 and 100 mammalian cells, respectively. Thus, cLC-SRM enables reliable quantification of low to moderately abundant proteins in less than 100 cells and could be broadly useful for multiplexed quantification of important proteins in small subpopulations of cells or in size-limited clinical samples. Further improvements of this method could eventually enable targeted single-cell proteomics when combined with either SRM or other emerging ultrasensitive MS detection.

Micropuncture of Bowman's Space in Mice Facilitated by 2 Photon Microscopy

Renal micropuncture and renal 2-photon imaging are seminal techniques in renal physiology. However, micropuncture is limited by dependence on conventional microscopy to surface nephron features, and 2-photon studies are limited in that interventions can only be assessed at the organ, rather than the nephron level. In particular, micropuncture studies of the glomeruli of mice have been challenged by the paucity of surface glomeruli in mice. To address this limitation in order to pursue studies of aspirate from Bowman's space in mouse physiologic models, we developed 2-photon glomerular micropuncture. We present a novel surgical preparation that allows lateral access to the kidney while preserving the required vertical imaging column for 2-photon microscopy. Administration of high molecular weight fluorescein isothiocyanate (FITC)-dextran is used to render the renal vasculature and therefore glomeruli visible for 2-photon imaging. A quantum dot-coated pipette is then introduced under stereotactic guidance to a glomerulus selected from the several to many which may be visualized within the imaging window. In this protocol, we provide details of the preparation, materials, and methods necessary to carry out the procedure. This technique facilitates previously-impossible physiologic study of the kidney, including recovery of filtrate from Bowman's space and all segments of the nephron within the imaging depth limit, about 100 µm below the renal capsule. Pressure, charge and flow may all be measured using the introduced pipette. Here, we provide representative data from liquid chromatography/mass spectrometry performed on aspirate from Bowman's space. We expect this technique to have wide applicability in renal physiologic investigation.

Residual tissue repositories as a resource for population-based cancer proteomic studies

BACKGROUND

Mass spectrometry-based proteomics has become a powerful tool for the identification and quantification of proteins from a wide variety of biological specimens. To date, the majority of studies utilizing tissue samples have been carried out on prospectively collected fresh frozen or optimal cutting temperature (OCT) embedded specimens. However, such specimens are often difficult to obtain, in limited in supply, and clinical information and outcomes on patients are inherently delayed as compared to banked samples. Annotated formalin fixed, paraffin embedded (FFPE) tumor tissue specimens are available for research use from a variety of tissue banks, such as from the surveillance, epidemiology and end results (SEER) registries' residual tissue repositories. Given the wealth of outcomes information associated with such samples, the reuse of archived FFPE blocks for deep proteomic characterization with mass spectrometry technologies would provide a valuable resource for population-based cancer studies. Further, due to the widespread availability of FFPE specimens, validation of specimen integrity opens the possibility for thousands of studies that can be conducted worldwide.

METHODS

To examine the suitability of the SEER repository tissues for proteomic and phosphoproteomic analysis, we analyzed 60 SEER patient samples, with time in storage ranging from 7 to 32 years; 60 samples with expression proteomics and 18 with phosphoproteomics, using isobaric labeling. Linear modeling and gene set enrichment analysis was used to evaluate the impacts of collection site and storage time.

RESULTS

All samples, regardless of age, yielded suitable protein mass after extraction for expression analysis and 18 samples yielded sufficient mass for phosphopeptide analysis. Although peptide, protein, and phosphopeptide identifications were reduced by 50, 20 and 76% respectively, from comparable OCT specimens, we found no statistically significant differences in protein quantitation correlating with collection site or specimen age. GSEA analysis of GO-term level measurements of protein abundance differences between FFPE and OCT embedded specimens suggest that the formalin fixation process may alter representation of protein categories in the resulting dataset.

CONCLUSIONS

These studies demonstrate that residual FFPE tissue specimens, of varying age and collection site, are a promising source of protein for proteomic investigations if paired with rigorously verified mass spectrometry workflows.

Abstract 284: Integrated proteogenomic analysis of laser microdissected primary breast tumors define proteome clusters

Introduction: Breast tumors have 4 well-established intrinsic subtypes based on transcriptome profiling. However, clusters defined by proteomics are often in disagreement with those defined by transcriptomics. Here, we report the findings of proteogenomic profiling of 118 laser microdissected (LMD) breast tumors using RNA-Seq and mass-spectrometry (MS)-based proteomic technologies.

Methods: Cases used in this study were drawn from the Clinical Breast Care Project, with patients consented using an IRB-approved protocol. A total of 118 primary breast tumors embedded in OCT were selected and processed by LMD. Total RNA and protein were extracted using the Illustra triplePrep kit. Paired-end RNA sequencing of 118 cases was performed using the Illumina HiSeq platform, and the reads were preprocessed using a PERL-based pipeline involving the preprocessing tool PRINSEQ, splice-aligner GSNAP and HTSeq for quantifying expression. Quantitative global proteomics analyses were performed on 113 cases using isobaric TMT 6-plex labeling with the “universal reference” strategy. MS data were acquired using a Q-Exactive instrument and analyzed using Proteome Discoverer with Byonic node. Sample-to-sample normalization was conducted to remove pipetting errors and ComBat was used to remove batch effect. K-means clustering was done using Bioconductor package Consensusclustering.

Results: The number of preprocessed RNA sequencing reads for the 118 cases ranged from over 43 to 295 million. An average of 83% of reads was mapped, and 24,518 genes with a mean expression of ≥ 10 counts across 118 tumor samples were identified. The PAM50 algorithm was used for intrinsic subtyping, yielding 37 Basal-like, 16 HER2-enriched, 39 Luminal A and 26 Luminal B calls. Unsupervised clustering of 3,000 highly varying genes reflected 4 intrinsic subtypes. In the global proteomics data, 840 proteins were identified across all 113 cases. Unsupervised K-means consensus clustering on all 840 or just using the top 210 highly varying proteins indicated the optimal number of clusters to be 3. These 3 clusters were identified as Basal-enriched, Luminal A-enriched and Luminal B-enriched. HER2-enriched cases were distributed among these clusters.

We did not observe a stromal-enriched cluster in this analysis of LMD-prepared samples that selected against stromal components of the tumor.

Conclusion: Analysis of LMD breast tumors using proteogenomic technologies resulted in 3 clusters for proteome data: basal-enriched, luminal A-enriched and luminal B-enriched. Unlike a recent report on proteomics clustering using bulk processing of tumors, a stromal-enriched cluster was not observed in this analysis which excluded stromal components of the samples.

The views expressed in this abstract are those of the author and do not reflect the official policy of the Department of Army/Navy/Air Force, Department of Defense, or U.S. Government.

Citation Format: Praveen-Kumar Raj-Kumar, Tao Liu, Lori A. Sturtz, Albert J. Kovatich, Marina A. Gritsenko, Vladislav A. Petyuk, Brenda Deyarmin, Viswanadham Sridhara, James Craig, Jason E. McDermott, Anil K. Shukla, Ronald J. Moore, Matthew E. Monroe, Bobbie-Jo M. Webb-Robertson, Jeffrey A. Hooke, J.Leigh Fantacone-Campbell, Leonid Kvecher, Jianfang Liu, Jennifer Kane, Jennifer Melley, Stella Somiari, Stephen C. Benz, Justin Golovato, Shahrooz Rabizadeh, Patrick Soon-Shiong, Richard D. Smith, Richard J. Mural, Karin D. Rodland, Craig D. Shriver, Hai Hu. Integrated proteogenomic analysis of laser microdissected primary breast tumors define proteome clusters [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 284.

Characterization of the Ovarian Tumor Peptidome

Aberrant degradation of proteins is associated with many pathological states, including cancers. Mass spectrometric analysis of the tumor peptidome has the potential to provide biological insights on proteolytic processing in cancer. However, attempts to use the tumors peptidome information in cancer research have been fairly limited to date, largely due to the lack of effective approaches for robust peptidomics identification and quantification, and the prevalence of confounding factors and biases associated with sample handling and processing. To address this need, we have recently developed an effective and robust analytical platform as well as a novel informatics approach for comprehensive analyses of tissue peptidomes. The ability of this new peptidomics pipeline for high-throughput, comprehensive, and quantitative peptidomics analysis, as well as the suitability of clinical ovarian tumor samples with postexcision delay limited to less than 60min before freezing for peptidomics analysis, has been demonstrated. These initial analyses set a stage for further determination of molecular details and functional significance of the peptidomic activities in ovarian cancer.