Predicting Ovarian Cancer Patients' Clinical Response to Platinum-Based Chemotherapy by Their Tumor Proteomic Signatures

Histopathology and proteomics are synergistic for high-grade serous ovarian cancer platinum response prediction

Patients with High-Grade Serous Ovarian Cancer (HGSOC) exhibit varied responses to treatment, with 20-30% showing de novo resistance to platinum-based chemotherapy. While hematoxylin-eosin (H&E)-stained pathological slides are used for routine diagnosis of cancer type, they may also contain diagnostically useful information about treatment response. Our study demonstrates that combining H&E-stained whole slide images (WSIs) with proteomic signatures using a multimodal deep learning framework significantly improves the prediction of platinum response in both discovery and validation cohorts. This method outperforms the Homologous Recombination Deficiency (HRD) score in predicting platinum response and overall patient survival. Our study suggests that histology and proteomics contain complementary information about biological processes determining response to first line platinum treatment in HGSOC. This integrative approach has the potential to improve personalized treatment and provide insights into the therapeutic vulnerabilities of HGSOC.

Development and Validation of a Predictive Model for Resistance to Platinum-Based Chemotherapy in Patients with Ovarian Cancer through Proteomic Analysis

Platinum resistance in ovarian cancer poses a significant challenge, substantially impacting patient outcomes. Developing an accurate predictive model is crucial for improving clinical decision-making and guiding treatment strategies. Proteomic data from 217 high-grade serous ovarian cancer (HGSOC) biospecimens obtained from JHU, PNNL, and PTRC were used to construct a prediction model for identifying individuals who are resistant to platinum-based chemotherapy. A total of 6437 common proteins were detected across all data sets, with 26 proteins overlapping between the development cohorts JHU and PNNL. Using LASSO and logistic regression analysis, a six-protein model (P31323_PRKAR2B, Q13309_SKP2, Q14997_PSME4, Q6ZRP7_QSOX2, Q7LGA3_HS2ST1, and Q7Z2Z2_EFL1) was developed, which accurately predicted platinum resistance, with an AUC of 0.964 (95% CI, 0.929-0.999). Internal validation by resampling resulted in a C-index of 0.972 (95% CI 0.894-0.988). External validation performed on the PTRC cohort achieved an AUC of 0.855 (95% CI 0.748-0.963). Calibration curves showed good consistency, and DCA indicated superior clinical utility. The model also performed well in predicting PFS and OS at various time points. Based on these proteins, our predictive model can precisely predict platinum response and survival outcomes in HGSOC patients, which can assist clinicians in promptly identifying potentially platinum-resistant individuals.

Identification of a Proteomic Signature for Predicting Immunotherapy Response in Patients With Metastatic Non-Small Cell Lung Cancer

Immunotherapy has improved survival rates in patients with cancer, but identifying those who will respond to treatment remains a challenge. Advances in proteomic technologies have enabled the identification and quantification of nearly all expressed proteins in a single experiment. Integrating mass spectrometry with high-throughput technologies has facilitated comprehensive analysis of the plasma proteome in cancer, facilitating early diagnosis and personalized treatment. In this context, our study aimed to investigate the predictive and prognostic value of plasma proteome analysis using the SWATH-MS (Sequential Window Acquisition of All Theoretical Mass Spectra) strategy in newly diagnosed patients with non-small cell lung cancer (NSCLC) receiving pembrolizumab therapy. We enrolled 64 newly diagnosed patients with advanced NSCLC treated with pembrolizumab. Blood samples were collected from all patients before and during therapy. A total of 171 blood samples were analyzed using the SWATH-MS strategy. Plasma protein expression in metastatic NSCLC patients prior to receiving pembrolizumab was analyzed. A first cohort (discovery cohort) was employed to identify a proteomic signature predicting immunotherapy response. Thus, 324 differentially expressed proteins between responder and non-responder patients were identified. In addition, we developed a predictive model and found a combination of seven proteins, including ATG9A, DCDC2, HPS5, FIL1L, LZTL1, PGTA, and SPTN2, with stronger predictive value than PD-L1 expression alone. Additionally, survival analyses showed an association between the levels of ATG9A, DCDC2, SPTN2 and HPS5 with progression-free survival (PFS) and/or overall survival (OS). Our findings highlight the potential of proteomic technologies to detect predictive biomarkers in blood samples from NSCLC patients, emphasizing the correlation between immunotherapy response and the idenfied protein set.

Histopathology and proteomics are synergistic for High-Grade Serous Ovarian Cancer platinum response prediction

Patients with High-Grade Serous Ovarian Cancer (HGSOC) exhibit varied responses to treatment, with 20-30% showing de novo resistance to platinum-based chemotherapy. While hematoxylin-eosin (H&E) pathological slides are used for routine diagnosis of cancer type, they may also contain diagnostically useful information about treatment response. Our study demonstrates that combining H&E-stained Whole Slide Images (WSIs) with proteomic signatures using a multimodal deep learning framework significantly improves the prediction of platinum response in both discovery and validation cohorts. This method outperforms the Homologous Recombination Deficiency (HRD) score in predicting platinum response and overall patient survival. The study sets new performance benchmarks and explores the intersection of histology and proteomics, highlighting phenotypes related to treatment response pathways, including homologous recombination, DNA damage response, nucleotide synthesis, apoptosis, and ER stress. This integrative approach has the potential to improve personalized treatment and provide insights into the therapeutic vulnerabilities of HGSOC.

Integration of proteomics in the molecular tumor board

Cancer remains one of the most complex and challenging diseases in mankind. To address the need for a personalized treatment approach for particularly complex tumor cases, molecular tumor boards (MTBs) have been initiated. MTBs are interdisciplinary teams that perform in-depth molecular diagnostics to cooperatively and interdisciplinarily advise on the best therapeutic strategy. Current molecular diagnostics are routinely performed on the transcriptomic and genomic levels, aiming to identify tumor-driving mutations. However, these approaches can only partially capture the actual phenotype and the molecular key players of tumor growth and progression. Thus, direct investigation of the expressed proteins and activated signaling pathways provide valuable complementary information on the tumor-driving molecular characteristics of the tissue. Technological advancements in mass spectrometry-based proteomics enable the robust, rapid, and sensitive detection of thousands of proteins in minimal sample amounts, paving the way for clinical proteomics and the probing of oncogenic signaling activity. Therefore, proteomics is currently being integrated into molecular diagnostics within MTBs and holds promising potential in aiding tumor classification and identifying personalized treatment strategies. This review introduces MTBs and describes current clinical proteomics, its potential in precision oncology, and highlights the benefits of multi-omic data integration.

Evaluating Ovarian Cancer Chemotherapy Response Using Gene Expression Data and Machine Learning

Background

Ovarian cancer (OC) is the most lethal gynecological cancer in the United States. Among the different types of OC, serous ovarian cancer (SOC) stands out as the most prevalent. Transcriptomics techniques generate extensive gene expression data, yet only a few of these genes are relevant to clinical diagnosis.

Methods

Methods for feature selection (FS) address the challenges of high dimensionality in extensive datasets. This study proposes a computational framework that applies FS techniques to identify genes highly associated with platinum-based chemotherapy response on SOC patients. Using SOC datasets from the Gene Expression Omnibus (GEO) database, LASSO and varSelRF FS methods were employed. Machine learning classification algorithms such as random forest (RF) and support vector machine (SVM) were also used to evaluate the performance of the models.

Results

The proposed framework has identified biomarkers panels with 9 and 10 genes that are highly correlated with platinum-paclitaxel and platinum-only response in SOC patients, respectively. The predictive models have been trained using the identified gene signatures and accuracy of above 90% was achieved.

Conclusions

In this study, we propose that applying multiple feature selection methods not only effectively reduces the number of identified biomarkers, enhancing their biological relevance, but also corroborates the efficacy of drug response prediction models in cancer treatment.

Histopathologic image-based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

Platinum-based chemotherapy is the cornerstone treatment for female high-grade serous ovarian carcinoma (HGSOC), but choosing an appropriate treatment for patients hinges on their responsiveness to it. Currently, no available biomarkers can promptly predict responses to platinum-based treatment. Therefore, we developed the Pathologic Risk Classifier for HGSOC (PathoRiCH), a histopathologic image-based classifier. PathoRiCH was trained on an in-house cohort (n = 394) and validated on two independent external cohorts (n = 284 and n = 136). The PathoRiCH-predicted favorable and poor response groups show significantly different platinum-free intervals in all three cohorts. Combining PathoRiCH with molecular biomarkers provides an even more powerful tool for the risk stratification of patients. The decisions of PathoRiCH are explained through visualization and a transcriptomic analysis, which bolster the reliability of our model's decisions. PathoRiCH exhibits better predictive performance than current molecular biomarkers. PathoRiCH will provide a solid foundation for developing an innovative tool to transform the current diagnostic pipeline for HGSOC.

Automated detection of nine infantile fundus diseases and conditions in retinal images using a deep learning system

Purpose

We developed an Infant Retinal Intelligent Diagnosis System (IRIDS), an automated system to aid early diagnosis and monitoring of infantile fundus diseases and health conditions to satisfy urgent needs of ophthalmologists.

Methods

We developed IRIDS by combining convolutional neural networks and transformer structures, using a dataset of 7697 retinal images (1089 infants) from four hospitals. It identifies nine fundus diseases and conditions, namely, retinopathy of prematurity (ROP) (mild ROP, moderate ROP, and severe ROP), retinoblastoma (RB), retinitis pigmentosa (RP), Coats disease, coloboma of the choroid, congenital retinal fold (CRF), and normal. IRIDS also includes depth attention modules, ResNet-18 (Res-18), and Multi-Axis Vision Transformer (MaxViT). Performance was compared to that of ophthalmologists using 450 retinal images. The IRIDS employed a five-fold cross-validation approach to generate the classification results.

Results

Several baseline models achieved the following metrics: accuracy, precision, recall, F1-score (F1), kappa, and area under the receiver operating characteristic curve (AUC) with best values of 94.62% (95% CI, 94.34%-94.90%), 94.07% (95% CI, 93.32%-94.82%), 90.56% (95% CI, 88.64%-92.48%), 92.34% (95% CI, 91.87%-92.81%), 91.15% (95% CI, 90.37%-91.93%), and 99.08% (95% CI, 99.07%-99.09%), respectively. In comparison, IRIDS showed promising results compared to ophthalmologists, demonstrating an average accuracy, precision, recall, F1, kappa, and AUC of 96.45% (95% CI, 96.37%-96.53%), 95.86% (95% CI, 94.56%-97.16%), 94.37% (95% CI, 93.95%-94.79%), 95.03% (95% CI, 94.45%-95.61%), 94.43% (95% CI, 93.96%-94.90%), and 99.51% (95% CI, 99.51%-99.51%), respectively, in multi-label classification on the test dataset, utilizing the Res-18 and MaxViT models. These results suggest that, particularly in terms of AUC, IRIDS achieved performance that warrants further investigation for the detection of retinal abnormalities.

Conclusions

IRIDS identifies nine infantile fundus diseases and conditions accurately. It may aid non-ophthalmologist personnel in underserved areas in infantile fundus disease screening. Thus, preventing severe complications. The IRIDS serves as an example of artificial intelligence integration into ophthalmology to achieve better outcomes in predictive, preventive, and personalized medicine (PPPM / 3PM) in the treatment of infantile fundus diseases.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13167-024-00350-y.

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

BACKGROUND

Machine learning is a potentially effective method for predicting the response to platinum-based treatment for ovarian cancer. However, the predictive performance of various machine learning methods and variables is still a matter of controversy and debate.

OBJECTIVE

This study aims to systematically review relevant literature on the predictive value of machine learning for platinum-based chemotherapy responses in patients with ovarian cancer.

METHODS

Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we systematically searched the PubMed, Embase, Web of Science, and Cochrane databases for relevant studies on predictive models for platinum-based therapies for the treatment of ovarian cancer published before April 26, 2023. The Prediction Model Risk of Bias Assessment tool was used to evaluate the risk of bias in the included articles. Concordance index (C-index), sensitivity, and specificity were used to evaluate the performance of the prediction models to investigate the predictive value of machine learning for platinum chemotherapy responses in patients with ovarian cancer.

RESULTS

A total of 1749 articles were examined, and 19 of them involving 39 models were eligible for this study. The most commonly used modeling methods were logistic regression (16/39, 41%), Extreme Gradient Boosting (4/39, 10%), and support vector machine (4/39, 10%). The training cohort reported C-index in 39 predictive models, with a pooled value of 0.806; the validation cohort reported C-index in 12 predictive models, with a pooled value of 0.831. Support vector machine performed well in both the training and validation cohorts, with a C-index of 0.942 and 0.879, respectively. The pooled sensitivity was 0.890, and the pooled specificity was 0.790 in the training cohort.

CONCLUSIONS

Machine learning can effectively predict how patients with ovarian cancer respond to platinum-based chemotherapy and may provide a reference for the development or updating of subsequent scoring systems.

Mass spectrometry-based high-throughput proteomics and its role in biomedical studies and systems biology

There are multiple reasons why the next generation of biological and medical studies require increasing numbers of samples. Biological systems are dynamic, and the effect of a perturbation depends on the genetic background and environment. As a consequence, many conditions need to be considered to reach generalizable conclusions. Moreover, human population and clinical studies only reach sufficient statistical power if conducted at scale and with precise measurement methods. Finally, many proteins remain without sufficient functional annotations, because they have not been systematically studied under a broad range of conditions. In this review, we discuss the latest technical developments in mass spectrometry (MS)-based proteomics that facilitate large-scale studies by fast and efficient chromatography, fast scanning mass spectrometers, data-independent acquisition (DIA), and new software. We further highlight recent studies which demonstrate how high-throughput (HT) proteomics can be applied to capture biological diversity, to annotate gene functions or to generate predictive and prognostic models for human diseases.

Clinical applications of mass spectrometry-based proteomics in cancer: where are we?

Tumor tissue processing methodologies in combination with data-independent acquisition mass spectrometry (DIA-MS) have emerged that can comprehensively analyze the proteome of multiple tumor samples accurately and reproducibly. Increasing recognition and adoption of these technologies has resulted in a tranche of studies providing novel insights into cancer classification systems, functional tumor biology, cancer biomarkers, treatment response and drug targets. Despite this, with some limited exceptions, MS-based proteomics has not yet been implemented in routine cancer clinical practice. Here, we summarize the use of DIA-MS in studies that may pave the way for future clinical cancer applications, and highlight the role of alternative MS technologies and multi-omic strategies. We discuss limitations and challenges of studies in this field to date and propose steps for integrating proteomic data into the cancer clinic. This article is protected by copyright. All rights reserved.

Effectiveness of Artificial Intelligence for Personalized Medicine in Neoplasms: A Systematic Review

Purpose

Artificial intelligence (AI) techniques are used in precision medicine to explore novel genotypes and phenotypes data. The main aims of precision medicine include early diagnosis, screening, and personalized treatment regime for a patient based on genetic-oriented features and characteristics. The main objective of this study was to review AI techniques and their effectiveness in neoplasm precision medicine.

Materials and Methods

A comprehensive search was performed in Medline (through PubMed), Scopus, ISI Web of Science, IEEE Xplore, Embase, and Cochrane databases from inception to December 29, 2021, in order to identify the studies that used AI methods for cancer precision medicine and evaluate outcomes of the models.

Results

Sixty-three studies were included in this systematic review. The main AI approaches in 17 papers (26.9%) were linear and nonlinear categories (random forest or decision trees), and in 21 citations, rule-based systems and deep learning models were used. Notably, 62% of the articles were done in the United States and China. R package was the most frequent software, and breast and lung cancer were the most selected neoplasms in the papers. Out of 63 papers, in 34 articles, genomic data like gene expression, somatic mutation data, phenotype data, and proteomics with drug-response which is functional data was used as input in AI methods; in 16 papers' (25.3%) drug response, functional data was utilized in personalization of treatment. The maximum values of the assessment indicators such as accuracy, sensitivity, specificity, precision, recall, and area under the curve (AUC) in included studies were 0.99, 1.00, 0.96, 0.98, 0.99, and 0.9929, respectively.

Conclusion

The findings showed that in many cases, the use of artificial intelligence methods had effective application in personalized medicine.

The role of patient-derived ovarian cancer organoids in the study of PARP inhibitors sensitivity and resistance: from genomic analysis to functional testing

Epithelial ovarian cancer (EOC) harbors distinct genetic features such as homologous recombination repair (HRR) deficiency, and therefore may respond to poly ADP-ribose polymerase inhibitors (PARPi). Over the past few years, PARPi have been added to the standard of care for EOC patients in both front-line and recurrent settings. Next-generation sequencing (NGS) genomic analysis provides key information, allowing for the prediction of PARPi response in patients who are PARPi naïve. However, there are indeed some limitations in NGS analyses. A subset of patients can benefit from PARPi, despite the failed detection of the predictive biomarkers such as BRCA1/2 mutations or HRR deficiency. Moreover, in the recurrent setting, the sequencing of initial tumor does not allow for the detection of reversions or secondary mutations restoring proficient HRR and thus leading to PARPi resistance. Therefore, it becomes crucial to better screen patients who will likely benefit from PARPi treatment, especially those with prior receipt of maintenance PARPi therapy. Recently, patient-derived organoids (PDOs) have been regarded as a reliable preclinical platform with clonal heterogeneity and genetic features of original tumors. PDOs are found feasible for functional testing and interrogation of biomarkers for predicting response to PARPi in EOC. Hence, we review the strengths and limitations of various predictive biomarkers and highlight the role of patient-derived ovarian cancer organoids as functional assays in the study of PARPi response. It was found that a combination of NGS and functional assays using PDOs could enhance the efficient screening of EOC patients suitable for PARPi, thus prolonging their survival time.

CT-Based Sarcopenic Nomogram for Predicting Progressive Disease in Advanced Non-Small-Cell Lung Cancer

Background

It is prudent to identify the risk for progressive disease (PD) in patients with non-small-cell lung cancer (NSCLC) who undergo platinum-based chemotherapy. The present study aimed to develop a CT imaging-based sarcopenic nomogram for predicting the risk of PD prior to chemotherapy treatment.

Methods

We retrospectively enrolled patients with NSCLC who underwent platinum-based chemotherapy. Imaging-based body composition parameters such as skeletal muscle index (SMI) for assessment of sarcopenia were obtained from pre-chemotherapy chest CT images at the level of the eleventh thoracic vertebral body (T11). Sarcopenic nomogram was constructed using multivariate logistic regression and performance of the nomogram was evaluated by discrimination, calibration curve, and decision curve.

Results

Sixty (14.7%) of the 408 patients in the study cohort developed PD during chemotherapy. The prediction nomogram for developing PD achieved a moderate efficiency with an area under the curve (AUC) of 0.75 (95% CI: 0.69-0.80) for the training cohort, and 0.76 (95%CI: 0.68-0.84) for the validation cohort, as well as a good performance of consistence (bootstrap for training cohort: 0.75 ± 0.02; validation cohort: 0.74 ± 0.06). Favorable clinical application was observed in the decision curve analysis.

Conclusion

Our CT-based sarcopenic nomogram showed the potential for an individualized prediction of progression for patients with NSCLC receiving platinum-based chemotherapy.

Clinicopathologic predictors of early relapse in advanced epithelial ovarian cancer: development of prediction models using nationwide data

OBJECTIVE

To identify clinicopathologic factors predictive of early relapse (platinum-free interval (PFI) of ≤6 months) in advanced epithelial ovarian cancer (EOC) in first-line treatment, and to develop and internally validate risk prediction models for early relapse.

METHODS

All consecutive patients diagnosed with advanced stage EOC between 01-01-2008 and 31-12-2015 were identified from the Netherlands Cancer Registry. Patients who underwent cytoreductive surgery and platinum-based chemotherapy as initial EOC treatment were selected. Two prediction models, i.e. pretreatment and postoperative, were developed. Candidate predictors of early relapse were fitted into multivariable logistic regression models. Model performance was assessed on calibration and discrimination. Internal validation was performed through bootstrapping to correct for model optimism.

RESULTS

A total of 4,557 advanced EOC patients were identified, including 1,302 early relapsers and 3,171 late or non-relapsers. Early relapsers were more likely to have FIGO stage IV, mucinous or clear cell type EOC, ascites, >1 cm residual disease, and to have undergone NACT-ICS. The final pretreatment model demonstrated subpar model performance (AUC = 0.64 [95 %-CI 0.62-0.66]). The final postoperative model based on age, FIGO stage, pretreatment CA-125 level, histologic subtype, presence of ascites, treatment approach, and residual disease after debulking, demonstrated adequate model performance (AUC = 0.72 [95 %-CI 0.71-0.74]). Bootstrap validation revealed minimal optimism of the final postoperative model.

CONCLUSION

A (postoperative) discriminative model has been developed and presented online that predicts the risk of early relapse in advanced EOC patients. Although external validation is still required, this prediction model can support patient counselling in daily clinical practice.

Radioproteomics in patients with ovarian cancer

Radioproteomics is the integration of proteomics, the systematic study of the protein expression of an organism, with radiomics, the extraction and analysis of large numbers of quantitative features from medical images. This article examines this developing field, and it's application in high grade serous ovarian carcinoma. Seminal proteomic studies in the area of ovarian cancer, such as the PROVAR and CPTA studies are discussed, along side recent research, such as that highlighting the central role of methyltransferase nicotinamide N-methyltransferase as the metabolic regulation of cancer progression in the tumour stroma. Finally, this article considers a novel, hypothesis generating approach to integrate CT-based qualitative and radiomic features with proteomic analysis, and the future direction of the field. Combined advances in radiomic, proteomic and genomic analysis has the potential to signal the age of true precision medicine, where treatment is centered specifically on the molecular profile of the tumour, rather than based on empirical knowledge, thus altering the course of a disease that has the highest mortality of all cancers of the female reproductive system.

Application of Artificial Intelligence Methods to Pharmacy Data for Cancer Surveillance and Epidemiology Research: A Systematic Review

PURPOSE

The implementation and utilization of electronic health records is generating a large volume and variety of data, which are difficult to process using traditional techniques. However, these data could help answer important questions in cancer surveillance and epidemiology research. Artificial intelligence (AI) data processing methods are capable of evaluating large volumes of data, yet current literature on their use in this context of pharmacy informatics is not well characterized.

METHODS

A systematic literature review was conducted to evaluate relevant publications within four domains (cancer, pharmacy, AI methods, population science) across PubMed, EMBASE, Scopus, and the Cochrane Library and included all publications indexed between July 17, 2008, and December 31, 2018. The search returned 3,271 publications, which were evaluated for inclusion.

RESULTS

There were 36 studies that met criteria for full-text abstraction. Of those, only 45% specifically identified the pharmacy data source, and 55% specified drug agents or drug classes. Multiple AI methods were used; 25% used machine learning (ML), 67% used natural language processing (NLP), and 8% combined ML and NLP.

CONCLUSION

This review demonstrates that the application of AI data methods for pharmacy informatics and cancer epidemiology research is expanding. However, the data sources and representations are often missing, challenging study replicability. In addition, there is no consistent format for reporting results, and one of the preferred metrics, F-score, is often missing. There is a resultant need for greater transparency of original data sources and performance of AI methods with pharmacy data to improve the translation of these results into meaningful outcomes.

Comprehending the Proteomic Landscape of Ovarian Cancer: A Road to the Discovery of Disease Biomarkers

Despite recent technological advancements allowing the characterization of cancers at a molecular level along with biomarkers for cancer diagnosis, the management of ovarian cancers (OC) remains challenging. Proteins assume functions encoded by the genome and the complete set of proteins, termed the proteome, reflects the health state. Comprehending the circulatory proteomic profiles for OC subtypes, therefore, has the potential to reveal biomarkers with clinical utility concerning early diagnosis or to predict response to specific therapies. Furthermore, characterization of the proteomic landscape of tumor-derived tissue, cell lines, and PDX models has led to the molecular stratification of patient groups, with implications for personalized therapy and management of drug resistance. Here, we review single and multiple marker panels that have been identified through proteomic investigations of patient sera, effusions, and other biospecimens. We discuss their clinical utility and implementation into clinical practice.

Proteomic Studies on the Management of High-Grade Serous Ovarian Cancer Patients: A Mini-Review

High-grade serous ovarian cancer (HGSC) remains the most common and deadly subtype of ovarian cancer. It is characterized by its late diagnosis and frequent relapse despite standardized treatment with cytoreductive surgery and platinum-based chemotherapy. The past decade has seen significant advances in the clinical management and molecular understanding of HGSC following the publication of the Cancer Genome Atlas (TCGA) researchers and the introduction of targeted therapies with anti-angiogenic drugs and poly(ADP-ribose) polymerase inhibitors in specific subgroups of patients. We provide a comprehensive review of HGSC, focusing on the most important molecular advances aimed at providing a better understanding of the disease and its response to treatment. We emphasize the role that proteomic technologies are now playing in these two aspects of the disease, through the identification of proteins and their post-translational modifications in ovarian cancer tumors. Finally, we highlight how the integration of proteomics with genomics, exemplified by the work performed by the Clinical Proteomic Tumor Analysis Consortium (CPTAC), can guide the development of new biomarkers and therapeutic targets.

Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks

BACKGROUND

Ovarian cancer causes 151,900 deaths per year worldwide. Treatment and prognosis are primarily determined by the histopathologic interpretation in combination with molecular diagnosis. However, the relationship between histopathology patterns and molecular alterations is not fully understood, and it is difficult to predict patients' chemotherapy response using the known clinical and histological variables.

METHODS

We analyzed the whole-slide histopathology images, RNA-Seq, and proteomics data from 587 primary serous ovarian adenocarcinoma patients and developed a systematic algorithm to integrate histopathology and functional omics findings and to predict patients' response to platinum-based chemotherapy.

RESULTS

Our convolutional neural networks identified the cancerous regions with areas under the receiver operating characteristic curve (AUCs) > 0.95 and classified tumor grade with AUCs > 0.80. Functional omics analysis revealed that expression levels of proteins participated in innate immune responses and catabolic pathways are associated with tumor grade. Quantitative histopathology analysis successfully stratified patients with different response to platinum-based chemotherapy (P = 0.003).

CONCLUSIONS

These results indicated the potential clinical utility of quantitative histopathology evaluation in tumor cell detection and chemotherapy response prediction. The developed algorithm is easily extensible to other tumor types and treatment modalities.

Use of the serum glycan state to predict ovarian cancer patients' clinical response to chemotherapy treatment

Ovarian cancer is the most lethal gynecologic carcinoma; because the tumor often relapses shortly after treatment. Glycosylation plays important roles in cancer drug resistance and could be used as biomarkers to predict the drug response of patients. We used MALDI-QIT-TOF MS to analyze the serum glycomic from patients with different drug responses. Samples were collected before treatment; follow-up visit were performed after 6 months. Forty-eight drug-sensitive patients and 16 drug-resistant patients were enrolled. Compared with drug-sensitive patients, 5 glyco-subclasses and 5 single glycans were significantly altered in drug-resistant patients. Lewis type, α2,3 sialic acid and multibranch glycans were increased, α2,6 sialic acid glycans were decreased. The peak at m/z 2986.44 showed stronger prediction abilities than other single glycans, with an AUC of 0.83. A panel of three increased glycans (m/z 2401.36, H5N4F1S2, a Lewis type biantennary glycan; m/z 2986.44, H6N5S3, a triantennary trisialylated glycan; m/z 3086.39, H6N5F1S3, a Lewis type triantennary glycan) combined with CA125 achieved an AUC value of 0.88, showing a strong discrimination performance. This study provides new insights into N-glycosylation patterns in ovarian cancer patients with different drug response. These altered glycans might serve as biomarkers to reflect patients' drug sensitivity and to guide clinical treatment. SIGNIFICANCE: A large number of ovarian cancer patients experience tumor relapse shortly after initial treatment. Glycosylation plays important roles in cancer drug resistance and could be used as a biomarker to predict the drug response of patients. However, the glycosylation expressed in patients with different drug response have not been elucidated. In the present study, we used MALDI-QIT-TOF MS to analyze the serum glycomic levels of patients with different drug responses. Several glycans were changed significantly between these two groups. A panel of three increased glycans (m/z 2401.36, a Lewis type biantennary glycan, 2986.44, a triantennary trisialylated glycan, and 3086.39, a Lewis type triantennary glycan) combined with CA125 performed better descrimination of these two groups with AUC of 0.88. These altered glycans might serve as biomarkers to reflect patients' drug sensitivity and to guide clinical treatment.

A deep neural network approach to predicting clinical outcomes of neuroblastoma patients

BACKGROUND

The availability of high-throughput omics datasets from large patient cohorts has allowed the development of methods that aim at predicting patient clinical outcomes, such as survival and disease recurrence. Such methods are also important to better understand the biological mechanisms underlying disease etiology and development, as well as treatment responses. Recently, different predictive models, relying on distinct algorithms (including Support Vector Machines and Random Forests) have been investigated. In this context, deep learning strategies are of special interest due to their demonstrated superior performance over a wide range of problems and datasets. One of the main challenges of such strategies is the "small n large p" problem. Indeed, omics datasets typically consist of small numbers of samples and large numbers of features relative to typical deep learning datasets. Neural networks usually tackle this problem through feature selection or by including additional constraints during the learning process.

METHODS

We propose to tackle this problem with a novel strategy that relies on a graph-based method for feature extraction, coupled with a deep neural network for clinical outcome prediction. The omics data are first represented as graphs whose nodes represent patients, and edges represent correlations between the patients' omics profiles. Topological features, such as centralities, are then extracted from these graphs for every node. Lastly, these features are used as input to train and test various classifiers.

RESULTS

We apply this strategy to four neuroblastoma datasets and observe that models based on neural networks are more accurate than state of the art models (DNN: 85%-87%, SVM/RF: 75%-82%). We explore how different parameters and configurations are selected in order to overcome the effects of the small data problem as well as the curse of dimensionality.

CONCLUSIONS

Our results indicate that the deep neural networks capture complex features in the data that help predicting patient clinical outcomes.

Proteomics advances for precision therapy in ovarian cancer

Introduction: Due to the relatively low mutation rate and high frequency of copy number variation, finding actionable genetic drivers of high-grade serous carcinoma (HGSC) is a challenging task. Furthermore, emerging studies show that genetic alterations are frequently poorly represented at the protein level adding a layer of complexity. With improvements in large-scale proteomic technologies, proteomics studies have the potential to provide robust analysis of the pathways driving high HGSC behavior. Areas covered: This review summarizes recent large-scale proteomics findings across adequately sized ovarian cancer sample sets. Key words combined with 'ovarian cancer' including 'proteomics', 'proteogenomic', 'reverse-phase protein array', 'mass spectrometry', and 'adaptive response', were used to search PubMed. Expert opinion: Proteomics analysis of HGSC as well as their adaptive responses to therapy can uncover new therapeutic liabilities, which can reduce the emergence of drug resistance and potentially improve patient outcomes. There is a pressing need to better understand how the genomic and epigenomic heterogeneity intrinsic to ovarian cancer is reflected at the protein level and how this information could be used to improve patient outcomes.

Molecular insights into cancer drug resistance from a proteomics perspective

INTRODUCTION

Resistance to chemotherapy and development of specific and effective molecular targeted therapies are major obstacles facing current cancer treatment. Comparative proteomic approaches have been employed for the discovery of putative biomarkers associated with cancer drug resistance and have yielded a number of candidate proteins, showing great promise for both novel drug target identification and personalized medicine for the treatment of drug-resistant cancer. Areas covered: Herein, we review the recent advances and challenges in proteomics studies on cancer drug resistance with an emphasis on biomarker discovery, as well as understanding the interconnectivity of proteins in disease-related signaling pathways. In addition, we highlight the critical role that post-translational modifications (PTMs) play in the mechanisms of cancer drug resistance. Expert opinion: Revealing changes in proteome profiles and the role of PTMs in drug-resistant cancer is key to deciphering the mechanisms of treatment resistance. With the development of sensitive and specific mass spectrometry (MS)-based proteomics and related technologies, it is now possible to investigate in depth potential biomarkers and the molecular mechanisms of cancer drug resistance, assisting the development of individualized therapeutic strategies for cancer patients.

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.

The Case for Proteomics and Phospho-Proteomics in Personalized Cancer Medicine

The concept of personalized medicine is predominantly been pursued through genomic and transcriptomic technologies, leading to the identification of multiple mutations in a large variety of cancers. However, it has proven challenging to distinguish driver and passenger mutations and to deal with tumor heterogeneity and resistant clonal populations. More generally, these heterogeneous mutation patterns do not in themselves predict the tumor phenotype. Analysis of the expressed proteins in a tumor and their modification states reveals if and how these mutations are translated to the functional level. It is already known that proteomic changes including posttranslational modifications are crucial drivers of oncogenesis, but proteomics technology has only recently become comparable in depth and accuracy to RNAseq. These advances also allow the rapid and highly sensitive analysis of formalin-fixed and paraffin-embedded biobank tissues, on both the proteome and phosphoproteome levels. In this perspective, pioneering mass spectrometry-based proteomic studies are highlighted that pave the way toward clinical implementation. It is argued that proteomics and phosphoproteomics could provide the missing link to make omics analysis actionable in the clinic.

L-Lactate dehydrogenase B may be a predictive marker for sensitivity to anti-EGFR monoclonal antibodies in colorectal cancer cell lines

Recently, proteins derived from cancer cells have been widely investigated as biomarkers for predicting the efficacy of chemotherapy. In this study, to identify a sensitive biomarker for the efficacy of anti-epidermal growth factor receptor monoclonal antibodies (anti-EGFR mAbs), proteins derived from 6 colorectal cancer (CRC) cell lines with different sensitivities to cetuximab, an anti-EGFR mAb, were analyzed. Cytoplasmic and membrane proteins extracted from each CRC cell line were digested using trypsin and analyzed comprehensively using mass spectrometry. As a result, 148 and 146 peaks from cytoplasmic proteins and 363 and 267 peaks from membrane proteins were extracted as specific peaks for cetuximab-resistant and -sensitive CRC cell lines, respectively. By analyzing the proteins identified from the peptide peaks, cytoplasmic L-lactate dehydrogenase B (LDHB) was detected as a marker of cetuximab sensitivity, and it was confirmed that LDHB expression was increased in cetuximab-resistant CRC cell lines. Furthermore, LDHB expression levels were significantly upregulated with the acquisition of resistance to cetuximab in cetuximab-sensitive CRC cell lines. In conclusion, LDHB was identified as an important factor affecting cetuximab sensitivity using comprehensive proteome analysis for the first time.

Precision Medicine: Role of Proteomics in Changing Clinical Management and Care

It is now possible to collect large sums of health-related data which has the potential to transform healthcare. Proteomics, with its central position as downstream of genetics and epigenetic inputs and upstream of biochemical outputs and integrators of environmental signals, is well-positioned to contribute to health discoveries and management. We present our perspective on the role of proteomics and other Omics in precision health and medicine.

Progress on Identifying and Characterizing the Human Proteome: 2018 Metrics from the HUPO Human Proteome Project

The Human Proteome Project (HPP) annually reports on progress throughout the field in credibly identifying and characterizing the human protein parts list and making proteomics an integral part of multiomics studies in medicine and the life sciences. NeXtProt release 2018-01-17, the baseline for this sixth annual HPP special issue of the Journal of Proteome Research, contains 17 470 PE1 proteins, 89% of all neXtProt predicted PE1-4 proteins, up from 17 008 in release 2017-01-23 and 13 975 in release 2012-02-24. Conversely, the number of neXtProt PE2,3,4 missing proteins has been reduced from 2949 to 2579 to 2186 over the past two years. Of the PE1 proteins, 16 092 are based on mass spectrometry results, and 1378 on other kinds of protein studies, notably protein-protein interaction findings. PeptideAtlas has 15 798 canonical proteins, up 625 over the past year, including 269 from SUMOylation studies. The largest reason for missing proteins is low abundance. Meanwhile, the Human Protein Atlas has released its Cell Atlas, Pathology Atlas, and updated Tissue Atlas, and is applying recommendations from the International Working Group on Antibody Validation. Finally, there is progress using the quantitative multiplex organ-specific popular proteins targeted proteomics approach in various disease categories.

Artificial intelligence in healthcare

Artificial intelligence (AI) is gradually changing medical practice. With recent progress in digitized data acquisition, machine learning and computing infrastructure, AI applications are expanding into areas that were previously thought to be only the province of human experts. In this Review Article, we outline recent breakthroughs in AI technologies and their biomedical applications, identify the challenges for further progress in medical AI systems, and summarize the economic, legal and social implications of AI in healthcare.

Mass spectrometry-based proteomics techniques and their application in ovarian cancer research

Ovarian cancer has emerged as one of the leading cause of gynecological malignancies. So far, the measurement of CA125 and HE4 concentrations in blood and transvaginal ultrasound examination are essential ovarian cancer diagnostic methods. However, their sensitivity and specificity are still not sufficient to detect disease at the early stage. Moreover, applied treatment may appear to be ineffective due to drug-resistance. Because of a high mortality rate of ovarian cancer, there is a pressing need to develop innovative strategies leading to a full understanding of complicated molecular pathways related to cancerogenesis. Recent studies have shown the great potential of clinical proteomics in the characterization of many diseases, including ovarian cancer. Therefore, in this review, we summarized achievements of proteomics in ovarian cancer management. Since the development of mass spectrometry has caused a breakthrough in systems biology, we decided to focus on studies based on this technique. According to PubMed engine, in the years 2008-2010 the number of studies concerning OC proteomics was increasing, and since 2010 it has reached a plateau. Proteomics as a rapidly evolving branch of science may be essential in novel biomarkers discovery, therapy decisions, progression predication, monitoring of drug response or resistance. Despite the fact that proteomics has many to offer, we also discussed some limitations occur in ovarian cancer studies. Main difficulties concern both complexity and heterogeneity of ovarian cancer and drawbacks of the mass spectrometry strategies. This review summarizes challenges, capabilities, and promises of the mass spectrometry-based proteomics techniques in ovarian cancer management.

Liver cancer-associated changes to the proteome: what deserves clinical focus?

INTRODUCTION

Hepatocellular carcinoma (HCC) is recognized as the fifth most common neoplasm and currently represents the second leading form of cancer-related death worldwide. Despite great progress has been done in the understanding of its pathogenesis, HCC represents a heavy societal and economic burden as most patients are still diagnosed at advanced stages and the 5-year survival rate remain below 20%. Early detection and revolutionary therapies that rely on the discovery of new molecular biomarkers and therapeutic targets are therefore urgently needed to develop precision medicine strategies for a more efficient management of patients. Areas covered: This review intends to comprehensively analyse the proteomics-based research conducted in the last few years to address some of the principal still open riddles in HCC biology, based on the identification of molecular drivers of tumor progression and metastasis. Expert commentary: The technical advances in mass spectrometry experienced in the last decade have significantly improved the analytical capacity of proteome wide studies. Large-scale protein and protein variant (post-translational modifications) identification and quantification have allowed detailed dissections of molecular mechanisms underlying HCC progression and are already paving the way for the identification of clinically relevant proteins and the development of their use on patient care.

Systematic Protein Prioritization for Targeted Proteomics Studies through Literature Mining

There are more than 3.7 million published articles on the biological functions or disease implications of proteins, constituting an important resource of proteomics knowledge. However, it is difficult to summarize the millions of proteomics findings in the literature manually and quantify their relevance to the biology and diseases of interest. We developed a fully automated bioinformatics framework to identify and prioritize proteins associated with any biological entity. We used the 22 targeted areas of the Biology/Disease-driven (B/D)-Human Proteome Project (HPP) as examples, prioritized the relevant proteins through their Protein Universal Reference Publication-Originated Search Engine (PURPOSE) scores, validated the relevance of the score by comparing the protein prioritization results with a curated database, computed the scores of proteins across the topics of B/D-HPP, and characterized the top proteins in the common model organisms. We further extended the bioinformatics workflow to identify the relevant proteins in all organ systems and human diseases and deployed a cloud-based tool to prioritize proteins related to any custom search terms in real time. Our tool can facilitate the prioritization of proteins for any organ system or disease of interest and can contribute to the development of targeted proteomic studies for precision medicine.

Association of Omics Features with Histopathology Patterns in Lung Adenocarcinoma

Adenocarcinoma accounts for more than 40% of lung malignancy, and microscopic pathology evaluation is indispensable for its diagnosis. However, how histopathology findings relate to molecular abnormalities remains largely unknown. Here, we obtained H&E-stained whole-slide histopathology images, pathology reports, RNA sequencing, and proteomics data of 538 lung adenocarcinoma patients from The Cancer Genome Atlas and used these to identify molecular pathways associated with histopathology patterns. We report cell-cycle regulation and nucleotide binding pathways underpinning tumor cell dedifferentiation, and we predicted histology grade using transcriptomics and proteomics signatures (area under curve >0.80). We built an integrative histopathology-transcriptomics model to generate better prognostic predictions for stage I patients (p = 0.0182 ± 0.0021) compared with gene expression or histopathology studies alone, and the results were replicated in an independent cohort (p = 0.0220 ± 0.0070). These results motivate the integration of histopathology and omics data to investigate molecular mechanisms of pathology findings and enhance clinical prognostic prediction.

Omics AnalySIs System for PRecision Oncology (OASISPRO): a web-based omics analysis tool for clinical phenotype prediction

SUMMARY

Precision oncology is an approach that accounts for individual differences to guide cancer management. Omics signatures have been shown to predict clinical traits for cancer patients. However, the vast amount of omics information poses an informatics challenge in systematically identifying patterns associated with health outcomes, and no general purpose data mining tool exists for physicians, medical researchers and citizen scientists without significant training in programming and bioinformatics. To bridge this gap, we built the Omics AnalySIs System for PRecision Oncology (OASISPRO), a web-based system to mine the quantitative omics information from The Cancer Genome Atlas (TCGA). This system effectively visualizes patients' clinical profiles, executes machine-learning algorithms of choice on the omics data and evaluates the prediction performance using held-out test sets. With this tool, we successfully identified genes strongly associated with tumor stage, and accurately predicted patients' survival outcomes in many cancer types, including adrenocortical carcinoma. By identifying the links between omics and clinical phenotypes, this system will facilitate omics studies on precision cancer medicine and contribute to establishing personalized cancer treatment plans.

AVAILABILITY AND IMPLEMENTATION

This web-based tool is available at http://tinyurl.com/oasispro; source codes are available at http://tinyurl.com/oasisproSourceCode.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Prediction of therapy response in ovarian cancer: Where are we now?

Therapy resistance is a major challenge in the management of ovarian cancer (OC). Advances in detection and new technology validation have led to the emergence of biomarkers that can predict responses to available therapies. It is important to identify predictive biomarkers to select resistant and sensitive patients in order to reduce important toxicities, to reduce costs and to increase survival. The discovery of predictive and prognostic biomarkers for monitoring therapy is a developing field and provides promising perspectives in the era of personalized medicine. This review article will discuss the biology of OC with a focus on targetable pathways; current therapies; mechanisms of resistance; predictive biomarkers for chemotherapy, antiangiogenic and DNA-targeted therapies, and optimal cytoreductive surgery; and the emergence of liquid biopsy using recent studies from the Medline database and ClinicalTrials.gov.