Precision Oncology: The Road Ahead

Therapeutic biomarkers in metastatic castration-resistant prostate cancer: does the state matter?

The treatment of metastatic castration-resistant prostate cancer (mCRPC) has been fundamentally transformed by our greater understanding of its complex biological mechanisms and its entrance into the era of precision oncology. A broad aim is to use the extreme heterogeneity of mCRPC by matching already approved or new targeted therapies to the correct tumor genotype. To achieve this, tumor DNA must be obtained, sequenced, and correctly interpreted, with individual aberrations explored for their druggability, taking into account the hierarchy of driving molecular pathways. Although tumor tissue sequencing is the gold standard, tumor tissue can be challenging to obtain, and a biopsy from one metastatic site or primary tumor may not provide an accurate representation of the current genetic underpinning. Sequencing of circulating tumor DNA (ctDNA) might catalyze precision oncology in mCRPC, as it enables real-time observation of genomic changes in tumors and allows for monitoring of treatment response and identification of resistance mechanisms. Moreover, ctDNA can be used to identify mutations that may not be detected in solitary metastatic lesions and can provide a more in-depth understanding of inter- and intra-tumor heterogeneity. Finally, ctDNA abundance can serve as a prognostic biomarker in patients with mCRPC.The androgen receptor (AR)-axis is a well-established therapeutical target for prostate cancer, and through ctDNA sequencing, insights have been obtained in (temporal) resistance mechanisms that develop through castration resistance. New third-generation AR-axis inhibitors are being developed to overcome some of these resistance mechanisms. The druggability of defects in the DNA damage repair machinery has impacted the treatment landscape of mCRPC in recent years. For patients with deleterious gene aberrations in genes linked to homologous recombination, particularly BRCA1 or BRCA2, PARP inhibitors have shown efficacy compared to the standard of care armamentarium, but platinum-based chemotherapy may be equally effective. A hierarchy exists in genes associated with homologous recombination, where, besides the canonical genes in this pathway, not every other gene aberration predicts the same likelihood of response. Moreover, evidence is emerging on cross-resistance between therapies such as PARP inhibitors, platinum-based chemotherapy and even radioligand therapy that target this genotype. Mismatch repair-deficient patients can experience a beneficial response to immune checkpoint inhibitors. Activation of other cellular signaling pathways such as PI3K, cell cycle, and MAPK have shown limited success with monotherapy, but there is potential in co-targeting these pathways with combination therapy, either already witnessed or anticipated. This review outlines precision medicine in mCRPC, zooming in on the role of ctDNA, to identify genomic biomarkers that may be used to tailor molecularly targeted therapies. The most common druggable pathways and outcomes of therapies matched to these pathways are discussed.

PERCEPTION predicts patient response and resistance to treatment using single-cell transcriptomics of their tumors

Tailoring optimal treatment for individual cancer patients remains a significant challenge. To address this issue, we developed PERCEPTION (PERsonalized Single-Cell Expression-Based Planning for Treatments In ONcology), a precision oncology computational pipeline. Our approach uses publicly available matched bulk and single-cell (sc) expression profiles from large-scale cell-line drug screens. These profiles help build treatment response models based on patients' sc-tumor transcriptomics. PERCEPTION demonstrates success in predicting responses to targeted therapies in cultured and patient-tumor-derived primary cells, as well as in two clinical trials for multiple myeloma and breast cancer. It also captures the resistance development in patients with lung cancer treated with tyrosine kinase inhibitors. PERCEPTION outperforms published state-of-the-art sc-based and bulk-based predictors in all clinical cohorts. PERCEPTION is accessible at https://github.com/ruppinlab/PERCEPTION . Our work, showcasing patient stratification using sc-expression profiles of their tumors, will encourage the adoption of sc-omics profiling in clinical settings, enhancing precision oncology tools based on sc-omics.

Community cohesion looseness in gene networks reveals individualized drug targets and resistance

Community cohesion plays a critical role in the determination of an individual's health in social science. Intriguingly, a community structure of gene networks indicates that the concept of community cohesion could be applied between the genes as well to overcome the limitations of single gene-based biomarkers for precision oncology. Here, we develop community cohesion scores which precisely quantify the community ability to retain the interactions between the genes and their cellular functions in each individualized gene network. Using breast cancer as a proof-of-concept study, we measure the community cohesion score profiles of 950 case samples and predict the individualized therapeutic targets in 2-fold. First, we prioritize them by finding druggable genes present in the community with the most and relatively decreased scores in each individual. Then, we pinpoint more individualized therapeutic targets by discovering the genes which greatly contribute to the community cohesion looseness in each individualized gene network. Compared with the previous approaches, the community cohesion scores show at least four times higher performance in predicting effective individualized chemotherapy targets based on drug sensitivity data. Furthermore, the community cohesion scores successfully discover the known breast cancer subtypes and we suggest new targeted therapy targets for triple negative breast cancer (e.g. KIT and GABRP). Lastly, we demonstrate that the community cohesion scores can predict tamoxifen responses in ER+ breast cancer and suggest potential combination therapies (e.g. NAMPT and RXRA inhibitors) to reduce endocrine therapy resistance based on individualized characteristics. Our method opens new perspectives for the biomarker development in precision oncology.

Culture and expansion of murine proximal airway basal stem cells

BACKGROUND

The stem cell characteristic makes basal cells desirable for ex vivo modeling of airway diseases. However, to date, approaches allowing them extensively in vitro serial expansion and maintaining bona fide stem cell property are still awaiting to be established. This study aims to develop a feeder-free culture system of mouse airway basal stem cells (ABSCs) that sustain their stem cell potential in vitro, providing an experimental basis for further in-depth research and mechanism exploration.

METHODS

We used ROCK inhibitor Y-27632-containing 3T3-CM, MEF-CM, and RbEF-CM to determine the proper feeder-free culture system that could maintain in vitro stem cell morphology of mouse ABSCs. Immunocytofluorescence was used to identify the basal cell markers of obtained cells. Serial propagation was carried out to observe whether the stem cell morphology and basal cell markers could be preserved in this cultivation system. Next, we examined the in vitro expansion and self-renewal ability by evaluating population doubling time and colony-forming efficiency. Moreover, the differentiation potential was detected by an in vitro differentiation culture and a 3D tracheosphere assay.

RESULTS

When the mouse ABSCs were cultured using 3T3-CM containing ROCK inhibitor Y-27632 in combination with Matrigel-coated culture dishes, they could stably expand and maintain stem cell-like clones. We confirmed that the obtained clones comprised p63/Krt5 double-positive ABSCs. In continuous passage and maintenance culture, we found that it could be subculture to at least 15 passages in vitro, stably maintaining its stem cell morphology, basal cell markers, and in vitro expansion and self-renewal capabilities. Meanwhile, through in vitro differentiation culture and 3D tracheosphere culture, we found that in addition to maintaining self-renewal, mouse ABSCs could differentiate into other airway epithelial cells such as acetylated tubulin (Act-Tub) + ciliated and MUC5AC + mucus-secreting cells. However, they failed to differentiate into alveoli epithelial cells, including alveolar type I and alveolar type II.

CONCLUSION

We established an in vitro feeder-free culture system that allows mouse ABSCs to maintain their stem cell characteristics, including self-renewal and airway epithelium differentiation potential, while keeping up in vitro expansion stability.

Preclinical evaluation of the CDK4/6 inhibitor palbociclib in combination with a PI3K or MEK inhibitor in colorectal cancer

BACKGROUND

Studies have demonstrated the efficacy of Palbociclib (CDK 4/6 inhibitor), Gedatolisib (PI3K/mTOR dual inhibitor) and PD0325901 (MEK1/2 inhibitor) in colorectal cancer (CRC), however single agent therapeutics are often limited by the development of resistance.

METHODS

We compared the anti-proliferative effects of the combination of Gedatolisib and Palbociclib and Gedatolisib and PD0325901 in five CRC cell lines with varying mutational background and tested their combinations on total and phosphoprotein levels of signaling pathway proteins.

RESULTS

The combination of Palbociclib and Gedatolisib was superior to the combination of Palbociclib and PD0325901. The combination of Palbociclib and Gedatolisib had synergistic anti-proliferative effects in all cell lines tested [CI range: 0.11-0.69] and resulted in the suppression of S6rp (S240/244), without AKT reactivation. The combination of Palbociclib and Gedatolisib increased BAX and Bcl-2 levels in PIK3CA mutated cell lines. The combination of Palbociclib and Gedatolisib caused MAPK/ERK reactivation, as seen by an increase in expression of total EGFR, regardless of the mutational status of the cells.

CONCLUSION

This study shows that the combination of Palbociclib and Gedatolisib has synergistic anti-proliferative effects in both wild-type and mutated CRC cell lines. Separately, the phosphorylation of S6rp may be a promising biomarker of responsiveness to this combination.

Revolutionizing Biological Science: The Synergy of Genomics in Health, Bioinformatics, Agriculture, and Artificial Intelligence

With climate emergency, COVID-19, and the rise of planetary health scholarship, the binary of human and ecosystem health has been deeply challenged. The interdependence of human and nonhuman animal health is increasingly acknowledged and paving the way for new frontiers in integrative biology. The convergence of genomics in health, bioinformatics, agriculture, and artificial intelligence (AI) has ushered in a new era of possibilities and applications. However, the sheer volume of genomic/multiomics big data generated also presents formidable sociotechnical challenges in extracting meaningful biological, planetary health and ecological insights. Over the past few years, AI-guided bioinformatics has emerged as a powerful tool for managing, analyzing, and interpreting complex biological datasets. The advances in AI, particularly in machine learning and deep learning, have been transforming the fields of genomics, planetary health, and agriculture. This article aims to unpack and explore the formidable range of possibilities and challenges that result from such transdisciplinary integration, and emphasizes its radically transformative potential for human and ecosystem health. The integration of these disciplines is also driving significant advancements in precision medicine and personalized health care. This presents an unprecedented opportunity to deepen our understanding of complex biological systems and advance the well-being of all life in planetary ecosystems. Notwithstanding in mind its sociotechnical, ethical, and critical policy challenges, the integration of genomics, multiomics, planetary health, and agriculture with AI-guided bioinformatics opens up vast opportunities for transnational collaborative efforts, data sharing, analysis, valorization, and interdisciplinary innovations in life sciences and integrative biology.

Spherical nucleic acids-based nanoplatforms for tumor precision medicine and immunotherapy

Precise diagnosis and treatment of tumors currently still face considerable challenges due to the development of highly degreed heterogeneity in the dynamic evolution of tumors. With the rapid development of genomics, personalized diagnosis and treatment using specific genes may be a robust strategy to break through the bottleneck of traditional tumor treatment. Nevertheless, efficient in vivo gene delivery has been frequently hampered by the inherent defects of vectors and various biological barriers. Encouragingly, spherical nucleic acids (SNAs) with good modularity and programmability are excellent candidates capable of addressing traditional gene transfer-associated issues, which enables SNAs a precision nanoplatform with great potential for diverse biomedical applications. In this regard, there have been detailed reviews of SNA in drug delivery, gene regulation, and dermatology treatment. Still, to the best of our knowledge, there is no published systematic review summarizing the use of SNAs in oncology precision medicine and immunotherapy, which are considered new guidelines for oncology treatment. To this end, we summarized the notable advances in SNAs-based precision therapy and immunotherapy for tumors following a classification standard of different types of precise spatiotemporal control on active species by SNAs. Specifically, we focus on the structural diversity and programmability of SNAs. Finally, the challenges and possible solutions were discussed in the concluding remarks. This review will promote the rational design and development of SNAs for tumor-precise medicine and immunotherapy.

Beyond the Frontline: A Triple-Line Approach of Thoracic Surgeons in Lung Cancer Management-State of the Art

Non-small cell lung cancer (NSCLC) is now described as an extremely heterogeneous disease in its clinical presentation, histology, molecular characteristics, and patient conditions. Over the past 20 years, the management of lung cancer has evolved with positive results. Immune checkpoint inhibitors have revolutionized the treatment landscape for NSCLC in both metastatic and locally advanced stages. The identification of molecular alterations in NSCLC has also allowed the development of targeted therapies, which provide better outcomes than chemotherapy in selected patients. However, patients usually develop acquired resistance to these treatments. On the other hand, thoracic surgery has progressed thanks to minimally invasive procedures, pre-habilitation and enhanced recovery after surgery. Moreover, within thoracic surgery, precision surgery considers the patient and his/her disease in their entirety to offer the best oncologic strategy. Surgeons support patients from pre-operative rehabilitation to surgery and beyond. They are involved in post-treatment follow-up and lung cancer recurrence. When conventional therapies are no longer effective, salvage surgery can be performed on selected patients.

Patient-derived glioblastoma cell lines with conserved genome profiles of the original tissue

Glioblastoma (GBM) is the most lethal intracranial tumor. Sequencing technologies have supported personalized therapy for precise diagnosis and optimal treatment of GBM by revealing clinically actionable molecular characteristics. Although accumulating sequence data from brain tumors and matched normal tissues have facilitated a comprehensive understanding of genomic features of GBM, these in silico evaluations could gain more biological credibility when they are verified with in vitro and in vivo models. From this perspective, GBM cell lines with whole exome sequencing (WES) datasets of matched tumor tissues and normal blood are suitable biological platforms to not only investigate molecular markers of GBM but also validate the applicability of druggable targets. Here, we provide a complete WES dataset of 26 GBM patient-derived cell lines along with their matched tumor tissues and blood to demonstrate that cell lines can mostly recapitulate genomic profiles of original tumors such as mutational signatures and copy number alterations.

Drug screening at single-organoid resolution via bioprinting and interferometry

High throughput drug screening is an established approach to investigate tumor biology and identify therapeutic leads. Traditional platforms use two-dimensional cultures which do not accurately reflect the biology of human tumors. More clinically relevant model systems such as three-dimensional tumor organoids can be difficult to scale and screen. Manually seeded organoids coupled to destructive endpoint assays allow for the characterization of treatment response, but do not capture transitory changes and intra-sample heterogeneity underlying clinically observed resistance to therapy. We present a pipeline to generate bioprinted tumor organoids linked to label-free, time-resolved imaging via high-speed live cell interferometry (HSLCI) and machine learning-based quantitation of individual organoids. Bioprinting cells gives rise to 3D structures with unaltered tumor histology and gene expression profiles. HSLCI imaging in tandem with machine learning-based segmentation and classification tools enables accurate, label-free parallel mass measurements for thousands of organoids. We demonstrate that this strategy identifies organoids transiently or persistently sensitive or resistant to specific therapies, information that could be used to guide rapid therapy selection.

Suspension state and shear stress enhance breast tumor cells EMT through YAP by microRNA-29b

Except for biochemical effects, suspension state (Sus) is proved to induce epithelial-mesenchymal transition (EMT) of circulating tumor cells (CTCs) mechanically. However, the difference between the effects of the mechanical microenvironment in capillaries (simplified as shear stress (SS) and Sus) and single Sus on EMT is unclear, nor the underlying mechanism. Here, breast tumor cells (BTCs) were loaded with Sus and SS to mimic the situation of CTCs stimulated by these two kinds of mechanics. It was demonstrated that the EMT of BTCs was enhanced by Sus and SS and the mechanotransductor yes-associated protein (YAP) was partially cytoplasmic stored with microRNA (miR)-29b decreased, which was detected by miR sequencing. Though it couldn't possess a feedback regulation, YAP promoted miR-29b expression and posttranscriptionally regulated BTCs EMT through miR-29b, where transforming growth factor β involved. Analysis of clinical database showed that high miR-29b expression was beneficial to high survival rate stabilizing its role of tumor suppressor. This study discovers the mechanism that Sus and SS promote BTCs EMT by YAP through miR-29b posttranscriptionally and highlight the potential of YAP and miR-29b in tumor therapy. The combination of suspension state and shear stress promotes transforming growth factor β involved epithelial-mesenchymal transition by yes-associated protein through microRNA-29b.

Systematic assessment of prognostic molecular features across cancers

Precision oncology promises accurate prediction of disease trajectories by utilizing molecular features of tumors. We present a systematic analysis of the prognostic potential of diverse molecular features across large cancer cohorts. We find that the mRNA expression of biologically coherent sets of genes (modules) is substantially more predictive of patient survival than single-locus genomic and transcriptomic aberrations. Extending our analysis beyond existing curated gene modules, we find a large novel class of highly prognostic DNA/RNA cis-regulatory modules associated with dynamic gene expression within cancers. Remarkably, in more than 82% of cancers, modules substantially improve survival stratification compared with conventional clinical factors and prominent genomic aberrations. The prognostic potential of cancer modules generalizes to external cohorts better than conventionally used single-gene features. Finally, a machine-learning framework demonstrates the combined predictive power of multiple modules, yielding prognostic models that perform substantially better than existing histopathological and clinical factors in common use.

Diagnostic value of liquid biopsy in the era of precision medicine: 10 years of clinical evidence in cancer

Liquid biopsy is a diagnostic repeatable test, which in last years has emerged as a powerful tool for profiling cancer genomes in real-time with minimal invasiveness and tailoring oncological decision-making. It analyzes different blood-circulating biomarkers and circulating tumor DNA (ctDNA) is the preferred one. Nevertheless, tissue biopsy remains the gold standard for molecular evaluation of solid tumors whereas liquid biopsy is a complementary tool in many different clinical settings, such as treatment selection, monitoring treatment response, cancer clonal evolution, prognostic evaluation, as well as the detection of early disease and minimal residual disease (MRD). A wide number of technologies have been developed with the aim of increasing their sensitivity and specificity with acceptable costs. Moreover, several preclinical and clinical studies have been conducted to better understand liquid biopsy clinical utility. Anyway, several issues are still a limitation of its use such as false positive and negative results, results interpretation, and standardization of the panel tests. Although there has been rapid development of the research in these fields and recent advances in the clinical setting, many clinical trials and studies are still needed to make liquid biopsy an instrument of clinical routine. This review provides an overview of the current and future clinical applications and opening questions of liquid biopsy in different oncological settings, with particular attention to ctDNA liquid biopsy.

Construction of IL-1 signalling pathway correlation model in lung adenocarcinoma and association with immune microenvironment prognosis and immunotherapy: Multi-data validation

Numerous studies have confirmed the inextricable link between inflammation and malignancy, which is also involved in developing lung adenocarcinoma, where IL-1 signalling is crucial. However, the predictive role of single gene biomarkers is insufficient, and more accurate prognostic models are needed. We downloaded data related to lung adenocarcinoma patients from the GDC, GEO, TISCH2 and TCGA databases for data analysis, model construction and differential gene expression analysis. The genes of IL-1 signalling-related factors were screened from published papers for subgroup typing and predictive correlation analysis. Five prognostic genes associated with IL-1 signalling were finally identified to construct prognostic prediction models. The K-M curves indicated that the prognostic models had significant predictive efficacy. Further immune infiltration scores showed that IL-1 signalling was mainly associated with enhanced immune cells, drug sensitivity of model genes was analysed using the GDSC database, and correlation of critical memories with cell subpopulation components was observed using single-cell analysis. In conclusion, we propose a predictive model based on IL-1 signalling-related factors, a non-invasive predictive approach for genomic characterisation, in predicting patients' survival outcomes. The therapeutic response has shown satisfactory and effective performance. More interdisciplinary areas combining medicine and electronics will be explored in the future.

Crosstalk between protein kinases AKT and ERK1/2 in human lung tumor-derived cell models

There is no doubt that cell signaling manipulation is a key strategy for anticancer therapy. Furthermore, cell state determines drug response. Thus, establishing the relationship between cell state and therapeutic sensitivity is essential for the development of cancer therapies. In the era of personalized medicine, the use of patient-derived ex vivo cell models is a promising approach in the translation of key research findings into clinics. Here, we were focused on the non-oncogene dependencies of cell resistance to anticancer treatments. Signaling-related mechanisms of response to inhibitors of MEK/ERK and PI3K/AKT pathways (regulators of key cellular functions) were investigated using a panel of patients' lung tumor-derived cell lines with various stemness- and EMT-related markers, varying degrees of ERK1/2 and AKT phosphorylation, and response to anticancer treatment. The study of interactions between kinases was the goal of our research. Although MEK/ERK and PI3K/AKT interactions are thought to be cell line-specific, where oncogenic mutations have a decisive role, we demonstrated negative feedback loops between MEK/ERK and PI3K/AKT signaling pathways in all cell lines studied, regardless of genotype and phenotype differences. Our work showed that various and distinct inhibitors of ERK signaling - selumetinib, trametinib, and SCH772984 - increased AKT phosphorylation, and conversely, inhibitors of AKT - capivasertib, idelalisib, and AKT inhibitor VIII - increased ERK phosphorylation in both control and cisplatin-treated cells. Interaction between kinases, however, was dependent on cellular state. The feedback between ERK and AKT was attenuated by the focal adhesion kinase inhibitor PF573228, and in cells grown in suspension, showing the possible role of extracellular contacts in the regulation of crosstalk between kinases. Moreover, studies have shown that the interplay between MEK/ERK and PI3K/AKT signaling pathways may be dependent on the strength of the chemotherapeutic stimulus. The study highlights the importance of spatial location of the cells and the strength of the treatment during anticancer therapy.

Phosphoproteomics After Guanidinium Thiocyanate Extraction of Tissue Biopsies

Proteogenomic analysis is emerging as an advantageous tool to assist personalized therapy decisions in clinical health care and integrates complementary information from the genome, transcriptome, and (phospho)proteome. A prerequisite for such analysis is a workflow for the simultaneous isolation of DNA, RNA, and protein from a single sample that does not compromise the different biological molecules and their examination. Focusing on the phosphoproteomic aspect of this workflow, we here provide detailed information on our protocol, which is based on commonly used acid guanidinium thiocyanate-phenol-chloroform (AGPC) extraction with RNA-Bee. We describe the necessary steps for biopsy collection, cryoprocessing, and protein extraction. We further share our practice on protein digestion and cleanup of small samples (200 μg protein) and describe settings for automated IMAC-based phosphopeptide enrichment with the AssayMAP Bravo platform.

Screening of predicted synergistic multi-target therapies in glioblastoma identifies new treatment strategies

Background

IDH-wildtype glioblastoma (GBM) is a highly malignant primary brain tumor with a median survival of 15 months after standard of care, which highlights the need for improved therapy. Personalized combination therapy has shown to be successful in many other tumor types and could be beneficial for GBM patients.

Methods

We performed the largest drug combination screen to date in GBM, using a high-throughput effort where we selected 90 drug combinations for their activity onto 25 patient-derived GBM cultures. 43 drug combinations were selected for interaction analysis based on their monotherapy efficacy and were tested in a short-term (3 days) as well as long-term (18 days) assay. Synergy was assessed using dose-equivalence and multiplicative survival metrics.

Results

We observed a consistent synergistic interaction for 15 out of 43 drug combinations on patient-derived GBM cultures. From these combinations, 11 out of 15 drug combinations showed a longitudinal synergistic effect on GBM cultures. The highest synergies were observed in the drug combinations Lapatinib with Thapsigargin and Lapatinib with Obatoclax Mesylate, both targeting epidermal growth factor receptor and affecting the apoptosis pathway. To further elaborate on the apoptosis cascade, we investigated other, more clinically relevant, apoptosis inducers and observed a strong synergistic effect while combining Venetoclax (BCL targeting) and AZD5991 (MCL1 targeting).

Conclusions

Overall, we have identified via a high-throughput drug screening several new treatment strategies for GBM. Moreover, an exceptionally strong synergistic interaction was discovered between kinase targeting and apoptosis induction which is suitable for further clinical evaluation as multi-targeted combination therapy.

Landscape of somatic alterations in large-scale solid tumors from an Asian population

Extending the benefits of tumor molecular profiling for all cancer patients requires a comprehensive analysis of tumor genomes across distinct patient populations worldwide. In this study, we perform deep next-generation DNA sequencing (NGS) from tumor tissues and matched blood specimens from over 10,000 patients in China by using a 450-gene comprehensive assay, developed and implemented under international clinical regulations. We perform a comprehensive comparison of somatically altered genes, the distribution of tumor mutational burden (TMB), gene fusion patterns, and the spectrum of various somatic alterations between Chinese and American patient populations. Here, we show 64% of cancers from Chinese patients in this study have clinically actionable genomic alterations, which may affect clinical decisions related to targeted therapy or immunotherapy. These findings describe the similarities and differences between tumors from Chinese and American patients, providing valuable information for personalized medicine.

Understanding the mutation landscape of cancer may enable the development of more targeted therapies. Here, the authors sequence a panel of genes in a large Asian cohort and compare to American cohorts and find 64% of the Asian patients have actionable mutations.

Artificial intelligence to guide precision anticancer therapy with multitargeted kinase inhibitors

BACKGROUND

Vast amounts of rapidly accumulating biological data related to cancer and a remarkable progress in the field of artificial intelligence (AI) have paved the way for precision oncology. Our recent contribution to this area of research is CancerOmicsNet, an AI-based system to predict the therapeutic effects of multitargeted kinase inhibitors across various cancers. This approach was previously demonstrated to outperform other deep learning methods, graph kernel models, molecular docking, and drug binding pocket matching.

METHODS

CancerOmicsNet integrates multiple heterogeneous data by utilizing a deep graph learning model with sophisticated attention propagation mechanisms to extract highly predictive features from cancer-specific networks. The AI-based system was devised to provide more accurate and robust predictions than data-driven therapeutic discovery using gene signature reversion.

RESULTS

Selected CancerOmicsNet predictions obtained for "unseen" data are positively validated against the biomedical literature and by live-cell time course inhibition assays performed against breast, pancreatic, and prostate cancer cell lines. Encouragingly, six molecules exhibited dose-dependent antiproliferative activities, with pan-CDK inhibitor JNJ-7706621 and Src inhibitor PP1 being the most potent against the pancreatic cancer cell line Panc 04.03.

CONCLUSIONS

CancerOmicsNet is a promising AI-based platform to help guide the development of new approaches in precision oncology involving a variety of tumor types and therapeutics.

A Multiview Clustering Method With Low-Rank and Sparsity Constraints for Cancer Subtyping

Multiomics data clustering is one of the major challenges in the field of precision medicine. Integration of multiomics data for cancer subtyping can improve the understanding on cancer and reveal systems-level insights. How to integrate multiomics data for accurate cancer subtyping is an interesting and challenging research problem. To capture the global and the local structure of omics data, a novel framework for integrating multiomics data is proposed for cancer subtyping. Multiview clustering with low-rank and sparsity constraints (MVCLRS) can measure the local similarities of samples in each omics data and obtain global consensus structures by integrating the multiomics data. The main insight provided by MVCLRS is that low-rank sparse subspace clustering for the construction of an affinity matrix can best capture the local similarities in omics data. Extensive testing is conducted on 10 real world cancer datasets with multiomics from The Cancer Genome Atlas. Compared with 10 state-of-the-art multiomics clustering algorithms, the MVCLRS performs better in the 10 cancer datasets by providing its clustering results with at least one enriched clinical label in nine of ten cancer subtypes, the most of any method.

Neural Collective Matrix Factorization for integrated analysis of heterogeneous biomedical data

MOTIVATION

In many biomedical studies, there arises the need to integrate data from multiple directly or indirectly related sources. Collective matrix factorization (CMF) and its variants are models designed to collectively learn from arbitrary collections of matrices. The latent factors learnt are rich integrative representations that can be used in downstream tasks, such as clustering or relation prediction with standard machine-learning models. Previous CMF-based methods have numerous modeling limitations. They do not adequately capture complex non-linear interactions and do not explicitly model varying sparsity and noise levels in the inputs, and some cannot model inputs with multiple datatypes. These inadequacies limit their use on many biomedical datasets.

RESULTS

To address these limitations, we develop Neural Collective Matrix Factorization (NCMF), the first fully neural approach to CMF. We evaluate NCMF on relation prediction tasks of gene-disease association prediction and adverse drug event prediction, using multiple datasets. In each case, data are obtained from heterogeneous publicly available databases and used to learn representations to build predictive models. NCMF is found to outperform previous CMF-based methods and several state-of-the-art graph embedding methods for representation learning in our experiments. Our experiments illustrate the versatility and efficacy of NCMF in representation learning for seamless integration of heterogeneous data.

AVAILABILITY AND IMPLEMENTATION

https://github.com/ajayago/NCMF_bioinformatics.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Computational Biology Predicts the Efficacy of Tumor Immune Checkpoint Blockade

Tumor immunotherapy is considered as one of the most promising methods in cancer treatment in recent years. Immune checkpoint blockade (ICB) can activate immune cells to destroy tumors by relieving the inhibitory pathway of tumor cells to immune cells. In silico prediction of the ICB response is an important step toward achieving effective and personalized cancer immunotherapy. Although immune checkpoint inhibitors have shown exciting clinical effects in the treatment of many types of tumors, there are still some clinical problems in practical application, such as low response rate and large individualized differences. How to predict the efficacy of effective individualized immune checkpoint inhibitors for tumor patients based on specific biomarkers and computational models is one of the key issues in the immunotherapy of this kind of tumor. In our work, from the five levels of genome level, transcription level, epigenetic level, microbial taxonomy level, and the immune cell infiltration profile level, the biomarkers and in silico calculation methods that affect the efficacy of tumor immune checkpoint inhibitors are comprehensively summarized.

Histone methylation modification patterns and relevant M-RiskScore in acute myeloid leukemia

Objective

We tried to identify novel molecular subtypes of acute myeloid leukemia (AML) associated with histone methylation and established a relevant scoring system to predict treatment response and prognosis of AML.

Methods

Gene expression data and clinical characteristics of patients with AML were obtained from The Cancer Genome Atlas (TCGA) database and Gene Expression Omnibus (GEO) database. Molecular subtyping was carried out by consensus clustering analysis, based on the expression of 24 histone methylation modification regulators (HMMRs). The clinical and biological features of each clustered pattern were taken into account. The scoring system was constructed by using differential expression analysis, Cox regression method and lasso regression analysis. Subsequently, the scoring system in the roles of prognostic and chemotherapeutic prediction of AML were explored. Finally, an independent GSE dataset was used for validating the established clustering system.

Results

Two distinct subtypes of AML were identified based on the expression of the 24 HMMRs, which exhibited remarkable differences in several clinical and biological characteristics, including HMMRs expression, AML-M0 distribution, NPM1 mutation, tumor mutation burden, somatic mutations, pathway activation, immune cell infiltration and patient survival. The scoring system, M-RiskScore, was established. Integrated analysis demonstrated that patients with the low M-RiskScore displayed a prominent survival advantage and a good response to decitabine treatment, while patients with high M-RiskScore have resistance to decitabine, but they could benefit from IA regimen therapy.

Conclusion

Detection of HMMRs expression would be a potential strategy for AML subtyping. Meanwhile, targeting histone methylation would be a preferred strategy for either AML-M0 or NPM1 mutant patients. M-RiskScore was a useful prognostic biomarker and a guide for the choice of appropriate chemotherapy strategy.

Pervasive conditional selection of driver mutations and modular epistasis networks in cancer

Cancer driver mutations often display mutual exclusion or co-occurrence, underscoring the key role of epistasis in carcinogenesis. However, estimating the magnitude of epistasis and quantifying its effect on tumor evolution remains a challenge. We develop a method (Coselens) to quantify conditional selection on the excess of nonsynonymous substitutions in cancer genes. Coselens infers the number of drivers per gene in different partitions of a cancer genomics dataset using covariance-based mutation models and determines whether coding mutations in a gene affect selection for drivers in any other gene. Using Coselens, we identify 296 conditionally selected gene pairs across 16 cancer types in the TCGA dataset. Conditional selection affects 25%-50% of driver substitutions in tumors with >2 drivers. Conditionally co-selected genes form modular networks, whose structures challenge the traditional interpretation of within-pathway mutual exclusivity and across-pathway synergy, suggesting a more complex scenario where gene-specific across-pathway epistasis shapes differentiated cancer subtypes.

Risk and benefit for umbrella trials in oncology: a systematic review and meta-analysis

BACKGROUND

Umbrella clinical trials in precision oncology are designed to tailor therapies to the specific genetic changes within a tumor. Little is known about the risk/benefit ratio for umbrella clinical trials. The aim of our systematic review with meta-analysis was to evaluate the efficacy and safety profiles in cancer umbrella trials testing targeted drugs or a combination of targeted therapy with chemotherapy.

METHODS

Our study was prospectively registered in PROSPERO (CRD42020171494). We searched Embase and PubMed for cancer umbrella trials testing targeted agents or a combination of targeted therapies with chemotherapy. We included solid tumor studies published between 1 January 2006 and 7 October 2019. We measured the risk using drug-related grade 3 or higher adverse events (AEs), and the benefit by objective response rate (ORR), progression-free survival (PFS), and overall survival (OS). When possible, data were meta-analyzed.

RESULTS

Of the 6207 records identified, we included 31 sub-trials or arms of nine umbrella trials (N = 1637). The pooled overall ORR was 17.7% (95% confidence interval [CI] 9.5-25.9). The ORR for targeted therapies in the experimental arms was significantly lower than the ORR for a combination of targeted therapy drugs with chemotherapy: 13.3% vs 39.0%; p = 0.005. The median PFS was 2.4 months (95% CI 1.9-2.9), and the median OS was 7.1 months (95% CI 6.1-8.4). The overall drug-related death rate (drug-related grade 5 AEs rate) was 0.8% (95% CI 0.3-1.4), and the average drug-related grade 3/4 AE rate per person was 0.45 (95% CI 0.40-0.50).

CONCLUSIONS

Our findings suggest that, on average, one in five cancer patients in umbrella trials published between 1 January 2006 and 7 October 2019 responded to a given therapy, while one in 125 died due to drug toxicity. Our findings do not support the expectation of increased patient benefit in cancer umbrella trials. Further studies should investigate whether umbrella trial design and the precision oncology approach improve patient outcomes.

Application of non-negative matrix factorization in oncology: one approach for establishing precision medicine

The increase in the expectations of artificial intelligence (AI) technology has led to machine learning technology being actively used in the medical field. Non-negative matrix factorization (NMF) is a machine learning technique used for image analysis, speech recognition, and language processing; recently, it is being applied to medical research. Precision medicine, wherein important information is extracted from large-scale medical data to provide optimal medical care for every individual, is considered important in medical policies globally, and the application of machine learning techniques to this end is being handled in several ways. NMF is also introduced differently because of the characteristics of its algorithms. In this review, the importance of NMF in the field of medicine, with a focus on the field of oncology, is described by explaining the mathematical science of NMF and the characteristics of the algorithm, providing examples of how NMF can be used to establish precision medicine, and presenting the challenges of NMF. Finally, the direction regarding the effective use of NMF in the field of oncology is also discussed.

Differential Gene Expression and Weighted Correlation Network Dynamics in High-Throughput Datasets of Prostate Cancer

Precision oncology is an absolute need today due to the emergence of treatment resistance and heterogeneity among cancerous profiles. Target-propelled cancer therapy is one of the treasures of precision oncology which has come together with substantial medical accomplishment. Prostate cancer is one of the most common cancers in males, with tremendous biological heterogeneity in molecular and clinical behavior. The spectrum of molecular abnormalities and varying clinical patterns in prostate cancer suggest substantial heterogeneity among different profiles. To identify novel therapeutic targets and precise biomarkers implicated with prostate cancer, we performed a state-of-the-art bioinformatics study, beginning with analyzing high-throughput genomic datasets from The Cancer Genome Atlas (TCGA). Weighted gene co-expression network analysis (WGCNA) suggests a set of five dysregulated hub genes (MAF, STAT6, SOX2, FOXO1, and WNT3A) that played crucial roles in biological pathways associated with prostate cancer progression. We found overexpressed STAT6 and SOX2 and proposed them as candidate biomarkers and potential targets in prostate cancer. Furthermore, the alteration frequencies in STAT6 and SOX2 and their impact on the patients' survival were explored through the cBioPortal platform. The Kaplan-Meier survival analysis suggested that the alterations in the candidate genes were linked to the decreased overall survival of the patients. Altogether, the results signify that STAT6 and SOX2 and their genomic alterations can be explored in therapeutic interventions of prostate cancer for precision oncology, utilizing early diagnosis and target-propelled therapy.

N6-Methyladenosine-Related lncRNAs Are Potential Prognostic Biomarkers and Correlated With Tumor Immune Microenvironment in Osteosarcoma

N6-methyladenosine (m6A) and long non-coding RNAs (lncRNAs) play vital roles in the prognostic value and immune microenvironment of malignant tumors. Here, we constructed a m6A-related lncRNA signature in osteosarcoma samples from TCGA dataset and analyzed the association of the signature with tumor immune microenvironment. m6A-related lncRNAs were identified by performing Pearson's correlation analysis and were used to construct a novel m6A-related lncRNA signature in osteosarcoma. Validation in testing and entire cohorts confirmed the satisfactory accuracy of the risk signature. Principal-component analysis verifies the grouping ability of the risk signature. Functional enrichment analyses connected immune with the risk signature based on the six m6A-related lncRNAs. When patients were separated into high- and low-risk group based on their risk scores, we found that patients in the high-risk group had lower stromal scores, immune scores, and ESTIMATE scores, while the tumor purity was higher in the high-risk group than that in the low-risk group. As for immune cell infiltration, the proportion of monocytes was significantly higher in the low-risk group than that in the high-risk group. Of the six lncRNAs, AC004812.2 was a protective factor in osteosarcoma and low expression of AC004812.2 predicted worse overall survival. Overexpression of AC004812.2 inhibited 143B cell proliferation and increased the expression levels of IGF2BP1 and YTHDF1. In all, our m6A-related lncRNA signature was a potential prognostic biomarker and correlated with tumor immune microenvironment and immune cell infiltration, and AC004812.2 might be an important regulator of m6A modification and a promising therapeutic target in osteosarcoma.

Patient-Derived Explants as a Precision Medicine Patient-Proximal Testing Platform Informing Cancer Management

Precision medicine approaches that inform clinical management of individuals with cancer are progressively advancing. Patient-derived explants (PDEs) provide a patient-proximal ex vivo platform that can be used to assess sensitivity to standard of care (SOC) therapies and novel agents. PDEs have several advantages as a patient-proximal model compared to current preclinical models, as they maintain the phenotype and microenvironment of the individual tumor. However, the longevity of PDEs is not compatible with the timeframe required to incorporate candidate therapeutic options identified by whole exome sequencing (WES) of the patient’s tumor. This review investigates how PDE longevity varies across tumor streams and how this is influenced by tissue preparation. Improving longevity of PDEs will enable individualized therapeutics testing, and thus contribute to improving outcomes for people with cancer.

An Algorithm to Mine Therapeutic Motifs for Cancer from Networks of Genetic Interactions

Study of pairwise genetic interactions, such as mutually exclusive mutations, has led to understanding of underlying mechanisms in cancer. Investigation of various combinatorial motifs within networks of such interactions can lead to deeper insights into its mutational landscape and inform therapy development. One such motif called the Between-Pathway Model (BPM) represents redundant or compensatory pathways that can be therapeutically exploited. Finding such BPM motifs is challenging since most formulations require solving variants of the NP-complete maximum weight bipartite subgraph problem. In this paper we design an algorithm based on Integer Linear Programming (ILP) to solve this problem. In our experiments, our approach outperforms the best previous method to mine BPM motifs. Further, our ILP-based approach allows us to easily model additional application-specific constraints. We illustrate this advantage through a new application of BPM motifs that can potentially aid in finding combination therapies to combat cancer.

Protein scaffolds: Antibody alternative for cancer diagnosis and therapy

Although antibodies are well developed and widely used in cancer therapy and diagnostic fields, some defects remain, such as poor tissue penetration, long in vivo metabolic retention, potential cytotoxicity, patent limitation, and high production cost. These issues have led scientists to explore and develop novel antibody alternatives. Protein scaffolds are small monomeric proteins with stable tertiary structures and mutable residues, which emerged in the 1990s. By combining robust gene engineering and phage display techniques, libraries with sufficient diversity could be established for target binding scaffold selection. Given the properties of small size, high affinity, and excellent specificity and stability, protein scaffolds have been applied in basic research, and preclinical and clinical fields over the past two decades. To date, more than 20 types of protein scaffolds have been developed, with the most frequently used being affibody, adnectin, ANTICALIN®, DARPins, and knottin. In this review, we focus on the protein scaffold applications in cancer therapy and diagnosis in the last 5 years, and discuss the pros and cons, and strategies of optimization and design.

Although antibodies are well developed and widely used in cancer therapy and diagnostic fields, some defects remain, such as poor tissue penetration, long in vivo metabolic retention, potential cytotoxicity, patent limitation, and high production cost.

Predicting chemotherapy response using a variational autoencoder approach

Background

Multiple studies have shown the utility of transcriptome-wide RNA-seq profiles as features for machine learning-based prediction of response to chemotherapy in cancer. While tumor transcriptome profiles are publicly available for thousands of tumors for many cancer types, a relatively modest number of tumor profiles are clinically annotated for response to chemotherapy. The paucity of labeled examples and the high dimension of the feature data limit performance for predicting therapeutic response using fully-supervised classification methods. Recently, multiple studies have established the utility of a deep neural network approach, the variational autoencoder (VAE), for generating meaningful latent features from original data. Here, we report the first study of a semi-supervised approach using VAE-encoded tumor transcriptome features and regularized gradient boosted decision trees (XGBoost) to predict chemotherapy drug response for five cancer types: colon, pancreatic, bladder, breast, and sarcoma.

Results

We found: (1) VAE-encoding of the tumor transcriptome preserves the cancer type identity of the tumor, suggesting preservation of biologically relevant information; and (2) as a feature-set for supervised classification to predict response-to-chemotherapy, the unsupervised VAE encoding of the tumor’s gene expression profile leads to better area under the receiver operating characteristic curve and area under the precision-recall curve classification performance than the original gene expression profile or the PCA principal components or the ICA components of the gene expression profile, in four out of five cancer types that we tested.

Conclusions

Given high-dimensional “omics” data, the VAE is a powerful tool for obtaining a nonlinear low-dimensional embedding; it yields features that retain biological patterns that distinguish between different types of cancer and that enable more accurate tumor transcriptome-based prediction of response to chemotherapy than would be possible using the original data or their principal components.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04339-6.

Interpretable systems biomarkers predict response to immune-checkpoint inhibitors

Cancer cells can leverage several cell-intrinsic and -extrinsic mechanisms to escape immune system recognition. The inherent complexity of the tumor microenvironment, with its multicellular and dynamic nature, poses great challenges for the extraction of biomarkers of immune response and immunotherapy efficacy. Here, we use RNA-sequencing (RNA-seq) data combined with different sources of prior knowledge to derive system-based signatures of the tumor microenvironment, quantifying immune-cell composition and intra- and intercellular communications. We applied multi-task learning to these signatures to predict different hallmarks of immune responses and derive cancer-type-specific models based on interpretable systems biomarkers. By applying our models to independent RNA-seq data from cancer patients treated with PD-1/PD-L1 inhibitors, we demonstrated that our method to Estimate Systems Immune Response (EaSIeR) accurately predicts therapeutic outcome. We anticipate that EaSIeR will be a valuable tool to provide a holistic description of immune responses in complex and dynamic systems such as tumors using available RNA-seq data.

Using machine learning approaches for multi-omics data analysis: A review

With the development of modern high-throughput omic measurement platforms, it has become essential for biomedical studies to undertake an integrative (combined) approach to fully utilise these data to gain insights into biological systems. Data from various omics sources such as genetics, proteomics, and metabolomics can be integrated to unravel the intricate working of systems biology using machine learning-based predictive algorithms. Machine learning methods offer novel techniques to integrate and analyse the various omics data enabling the discovery of new biomarkers. These biomarkers have the potential to help in accurate disease prediction, patient stratification and delivery of precision medicine. This review paper explores different integrative machine learning methods which have been used to provide an in-depth understanding of biological systems during normal physiological functioning and in the presence of a disease. It provides insight and recommendations for interdisciplinary professionals who envisage employing machine learning skills in multi-omics studies.

Personalized drug testing in a patient with non-small-cell lung cancer using cultured cancer cells from pleural effusion

Objective

Patients with non-small-cell lung cancer (NSCLC) and primary or acquired resistance do not respond to targeted drugs. We explored whether cancer cells can be cultured from liquid biopsies from patients with primary resistance to tyrosine kinase inhibitors (TKIs). We aimed to predict patients’ responses to drugs according to in vitro drug testing results.

Methods

Cancer cell cultures were established from the pleural effusion of a patient with TKI-resistant NSCLC using a conditional reprogramming technique. Phenotypic drug sensitivity tests were performed using the Cell Counting Kit-8 assay. We tested individual drugs and compared the synergistic and inhibitory effects of drug combinations.

Results

The results of our in vitro sensitivity test using the combination of cisplatin and pemetrexed were correlated with the patient’s response.

Conclusion

This represents the first successful report of predictive testing for combination therapy in patients with epidermal growth factor receptor-mutant NSCLC and primary TKI resistance. This strategy should be applicable to both chemotherapies and targeted therapies, and it will significantly improve the clinical treatment and management of patients with NSCLC and primary or acquired resistance to targeted therapies, as well as patients lacking targetable mutations.

Bioinformatics analysis of potential core genes for glioblastoma

Background: Glioblastoma (GBM) has a high degree of malignancy, aggressiveness and recurrence rate. However, there are limited options available for the treatment of GBM, and they often result in poor prognosis and unsatisfactory outcomes.

Materials and methods: In order to identify potential core genes in GBM that may provide new therapeutic insights, we analyzed three gene chips (GSE2223, GSE4290 and GSE50161) screened from the GEO database. Differentially expressed genes (DEG) from the tissues of GBM and normal brain were screened using GEO2R. To determine the functional annotation and pathway of DEG, Gene Ontology (GO) and KEGG pathway enrichment analysis were conducted using DAVID database. Protein interactions of DEG were visualized using PPI network on Cytoscape software. Next, 10 Hub nodes were screened from the differentially expressed network using MCC algorithm on CytoHubba software and subsequently identified as Hub genes. Finally, the relationship between Hub genes and the prognosis of GBM patients was described using GEPIA2 survival analysis web tool.

Results: A total of 37 up-regulated and 187 down-regulated genes were identified through microarray analysis. Amongst the 10 Hub genes selected, SV2B appeared to be the only gene associated with poor prognosis in glioblastoma based on the survival analysis.

Conclusion: Our study suggests that high expression of SV2B is associated with poor prognosis in GBM patients. Whether SV2B can be used as a new therapeutic target for GBM requires further validation.

Screening for Combination Cancer Therapies With Dynamic Fuzzy Modeling and Multi-Objective Optimization

Combination therapies proved to be a valuable strategy in the fight against cancer, thanks to their increased efficacy in inducing tumor cell death and in reducing tumor growth, metastatic potential, and the risk of developing drug resistance. The identification of effective combinations of drug targets generally relies on costly and time consuming processes based on in vitro experiments. Here, we present a novel computational approach that, by integrating dynamic fuzzy modeling with multi-objective optimization, allows to efficiently identify novel combination cancer therapies, with a relevant saving in working time and costs. We tested this approach on a model of oncogenic K-ras cancer cells characterized by a marked Warburg effect. The computational approach was validated by its capability in finding out therapies already known in the literature for this type of cancer cell. More importantly, our results show that this method can suggest potential therapies consisting in a small number of molecular targets. In the model of oncogenic K-ras cancer cells, for instance, we identified combination of up to three targets, which affect different cellular pathways that are crucial for cancer proliferation and survival.

Sequential filtering for clinically relevant variants as a method for clinical interpretation of whole exome sequencing findings in glioma

Background

In the clinical setting, workflows for analyzing individual genomics data should be both comprehensive and convenient for clinical interpretation. In an effort for comprehensiveness and practicality, we attempted to create a clinical individual whole exome sequencing (WES) analysis workflow, allowing identification of genomic alterations and presentation of neurooncologically-relevant findings.

Methods

The analysis workflow detects germline and somatic variants and presents: (1) germline variants, (2) somatic short variants, (3) tumor mutational burden (TMB), (4) microsatellite instability (MSI), (5) somatic copy number alterations (SCNA), (6) SCNA burden, (7) loss of heterozygosity, (8) genes with double-hit, (9) mutational signatures, and (10) pathway enrichment analyses. Using the workflow, 58 WES analyses from matched blood and tumor samples of 52 patients were analyzed: 47 primary and 11 recurrent diffuse gliomas.

Results

The median mean read depths were 199.88 for tumor and 110.955 for normal samples. For germline variants, a median of 22 (14–33) variants per patient was reported. There was a median of 6 (0–590) reported somatic short variants per tumor. A median of 19 (0–94) broad SCNAs and a median of 6 (0–12) gene-level SCNAs were reported per tumor. The gene with the most frequent somatic short variants was TP53 (41.38%). The most frequent chromosome-/arm-level SCNA events were chr7 amplification, chr22q loss, and chr10 loss. TMB in primary gliomas were significantly lower than in recurrent tumors (p = 0.002). MSI incidence was low (6.9%).

Conclusions

We demonstrate that WES can be practically and efficiently utilized for clinical analysis of individual brain tumors. The results display that NOTATES produces clinically relevant results in a concise but exhaustive manner.

Patient-derived non-small cell lung cancer xenograft mirrors complex tumor heterogeneity

Objective

Patient-derived xenograft (PDX) models have shown great promise in preclinical and translational applications, but their consistency with primary tumors in phenotypic, genetic, and pharmacodynamic heterogeneity has not been well-studied. This study aimed to establish a PDX repository for non-small cell lung cancer (NSCLC) and to further elucidate whether it could preserve the heterogeneity within and between tumors in patients.

Methods

A total of 75 surgically resected NSCLC specimens were implanted into immunodeficient NOD/SCID mice. Based on the successful establishment of the NSCLC PDX model, we compared the expressions of vimentin, Ki67, EGFR, and PD-L1 proteins between cancer tissues and PDX models using hematoxylin and eosin staining and immunohistochemical staining. In addition, we detected whole gene expression profiling between primary tumors and PDX generations. We also performed whole exome sequencing (WES) analysis in 17 first generation xenografts to further assess whether PDXs retained the patient heterogeneities. Finally, paclitaxel, cisplatin, doxorubicin, atezolizumab, afatininb, and AZD4547 were used to evaluate the responses of PDX models to the standard-of-care agents.

Results

A large collection of serially transplantable PDX models for NSCLC were successfully developed. The histology and pathological immunohistochemistry of PDX xenografts were consistent with the patients' tumor samples. WES and RNA-seq further confirmed that PDX accurately replicated the molecular heterogeneities of primary tumors. Similar to clinical patients, PDX models responded differentially to the standard-of-care treatment, including chemo-, targeted- and immuno-therapeutics.

Conclusions

Our established PDX models of NSCLC faithfully reproduced the molecular, histopathological, and therapeutic characteristics, as well as the corresponding tumor heterogeneities, which provides a clinically relevant platform for drug screening, biomarker discovery, and translational research.

Molecular Profiling-Based Assignment of Cancer Therapy (NCI-MPACT): A Randomized Multicenter Phase II Trial

This trial assessed the utility of applying tumor DNA sequencing to treatment selection for patients with advanced, refractory cancer and somatic mutations in one of four signaling pathways by comparing the efficacy of four study regimens that were either matched to the patient's aberrant pathway (experimental arm) or not matched to that pathway (control arm).

MATERIALS AND METHODS

Adult patients with an actionable mutation of interest were randomly assigned 2:1 to receive either (1) a study regimen identified to target the aberrant pathway found in their tumor (veliparib with temozolomide or adavosertib with carboplatin [DNA repair pathway], everolimus [PI3K pathway], or trametinib [RAS/RAF/MEK pathway]), or (2) one of the same four regimens, but chosen from among those not targeting that pathway.

RESULTS

Among 49 patients treated in the experimental arm, the objective response rate was 2% (95% CI, 0% to 10.9%). One of 20 patients (5%) in the experimental trametinib cohort had a partial response. There were no responses in the other cohorts. Although patients and physicians were blinded to the sequencing and random assignment results, a higher pretreatment dropout rate was observed in the control arm (22%) compared with the experimental arm (6%; P = .038), suggesting that some patients may have had prior tumor mutation profiling performed that led to a lack of participation in the control arm.

CONCLUSION

Further investigation, better annotation of predictive biomarkers, and the development of more effective agents are necessary to inform treatment decisions in an era of precision cancer medicine. Increasing prevalence of tumor mutation profiling and preference for targeted therapy make it difficult to use a randomized phase II design to evaluate targeted therapy efficacy in an advanced disease setting.

Opportunities for Gene and Environment Research in Cancer: An Updated Review of NCI's Extramural Grant Portfolio

BACKGROUND

The study of gene-environment (GxE) interactions is a research priority for the NCI. Previously, our group analyzed NCI's extramural grant portfolio from fiscal years (FY) 2007 to 2009 to determine the state of the science in GxE research. This study builds upon our previous effort and examines changes in the landscape of GxE cancer research funded by NCI.

METHODS

The NCI grant portfolio was examined from FY 2010 to 2018 using the iSearch application. A time-trend analysis was conducted to explore changes over the study interval.

RESULTS

A total of 107 grants met the search criteria and were abstracted. The most common cancer types studied were breast (19.6%) and colorectal (18.7%). Most grants focused on GxE using specific candidate genes (69.2%) compared with agnostic approaches using genome-wide (26.2%) or whole-exome/whole-genome next-generation sequencing (NGS) approaches (19.6%); some grants used more than one approach to assess genetic variation. More funded grants incorporated NGS technologies in FY 2016-2018 compared with prior FYs. Environmental exposures most commonly examined were energy balance (46.7%) and drugs/treatment (40.2%). Over the time interval, we observed a decrease in energy balance applications with a concurrent increase in drug/treatment applications.

CONCLUSIONS

Research in GxE interactions has continued to concentrate on common cancers, while there have been some shifts in focus of genetic and environmental exposures. Opportunities exist to study less common cancers, apply new technologies, and increase racial/ethnic diversity.

IMPACT

This analysis of NCI's extramural grant portfolio updates previous efforts and provides a review of NCI grant support for GxE research.

Application of Artificial Intelligence Technology in Oncology: Towards the Establishment of Precision Medicine

In recent years, advances in artificial intelligence (AI) technology have led to the rapid clinical implementation of devices with AI technology in the medical field. More than 60 AI-equipped medical devices have already been approved by the Food and Drug Administration (FDA) in the United States, and the active introduction of AI technology is considered to be an inevitable trend in the future of medicine. In the field of oncology, clinical applications of medical devices using AI technology are already underway, mainly in radiology, and AI technology is expected to be positioned as an important core technology. In particular, "precision medicine," a medical treatment that selects the most appropriate treatment for each patient based on a vast amount of medical data such as genome information, has become a worldwide trend; AI technology is expected to be utilized in the process of extracting truly useful information from a large amount of medical data and applying it to diagnosis and treatment. In this review, we would like to introduce the history of AI technology and the current state of medical AI, especially in the oncology field, as well as discuss the possibilities and challenges of AI technology in the medical field.

Organotypic primary blood vessel models of clear cell renal cell carcinoma for single-patient clinical trials

Clear cell renal cell carcinoma (ccRCC) is a common genitourinary cancer associated with the development of abnormal tumor angiogenesis. Although multiple anti-angiogenic therapies have been developed, responses to individual treatment are highly variable between patients. Thus, the use of one-patient clinical trials has been suggested as an alternative to standard trials. We used a microfluidic device to generate organotypic primary patient-specific blood vessel models using normal (NEnC) and tumor-associated primary CD31+ selected cells (TEnC). Our model was able to recapitulate differences in angiogenic sprouting and vessel permeability that characterize normal and tumor-associated vessels. We analyzed the expression profile of vessel models to define vascular normalization in a patient-specific manner. Using this data, we identified actionable targets to normalize TEnC vessel function to a more NEnC-like phenotype. Finally, we tested two of these drugs in our patient-specific models to determine the efficiency in restoring vessel function showing the potential of the model for single-patient clinical trials.

A Middle-Out Modeling Strategy to Extend a Colon Cancer Logical Model Improves Drug Synergy Predictions in Epithelial-Derived Cancer Cell Lines

Cancer is a heterogeneous and complex disease and one of the leading causes of death worldwide. The high tumor heterogeneity between individuals affected by the same cancer type is accompanied by distinct molecular and phenotypic tumor profiles and variation in drug treatment response. In silico modeling of cancer as an aberrantly regulated system of interacting signaling molecules provides a basis to enhance our biological understanding of disease progression, and it offers the means to use computer simulations to test and optimize drug therapy designs on particular cancer types and subtypes. This sets the stage for precision medicine: the design of treatments tailored to individuals or groups of patients based on their tumor-specific molecular cancer profiles. Here, we show how a relatively large manually curated logical model can be efficiently enhanced further by including components highlighted by a multi-omics data analysis of data from Consensus Molecular Subtypes covering colorectal cancer. The model expansion was performed in a pathway-centric manner, following a partitioning of the model into functional subsystems, named modules. The resulting approach constitutes a middle-out modeling strategy enabling a data-driven expansion of a model from a generic and intermediate level of molecular detail to a model better covering relevant processes that are affected in specific cancer subtypes, comprising 183 biological entities and 603 interactions between them, partitioned in 25 functional modules of varying size and structure. We tested this model for its ability to correctly predict drug combination synergies, against a dataset of experimentally determined cell growth responses with 18 drugs in all combinations, on eight cancer cell lines. The results indicate that the extended model had an improved accuracy for drug synergy prediction for the majority of the experimentally tested cancer cell lines, although significant improvements of the model's predictive performance are still needed. Our study demonstrates how a tumor-data driven middle-out approach toward refining a logical model of a biological system can further customize a computer model to represent specific cancer cell lines and provide a basis for identifying synergistic effects of drugs targeting specific regulatory proteins. This approach bridges between preclinical cancer model data and clinical patient data and may thereby ultimately be of help to develop patient-specific in silico models that can steer treatment decisions in the clinic.

Predicting synthetic lethal interactions using heterogeneous data sources

MOTIVATION

A synthetic lethal (SL) interaction is a relationship between two functional entities where the loss of either one of the entities is viable but the loss of both entities is lethal to the cell. Such pairs can be used as drug targets in targeted anticancer therapies, and so, many methods have been developed to identify potential candidate SL pairs. However, these methods use only a subset of available data from multiple platforms, at genomic, epigenomic and transcriptomic levels; and hence are limited in their ability to learn from complex associations in heterogeneous data sources.

RESULTS

In this article, we develop techniques that can seamlessly integrate multiple heterogeneous data sources to predict SL interactions. Our approach obtains latent representations by collective matrix factorization-based techniques, which in turn are used for prediction through matrix completion. Our experiments, on a variety of biological datasets, illustrate the efficacy and versatility of our approach, that outperforms state-of-the-art methods for predicting SL interactions and can be used with heterogeneous data sources with minimal feature engineering.

AVAILABILITY AND IMPLEMENTATION

Software available at https://github.com/lianyh.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Personalized cancer therapy prioritization based on driver alteration co-occurrence patterns

Identification of actionable genomic vulnerabilities is key to precision oncology. Utilizing a large-scale drug screening in patient-derived xenografts, we uncover driver gene alteration connections, derive driver co-occurrence (DCO) networks, and relate these to drug sensitivity. Our collection of 53 drug-response predictors attains an average balanced accuracy of 58% in a cross-validation setting, rising to 66% for a subset of high-confidence predictions. We experimentally validated 12 out of 14 predictions in mice and adapted our strategy to obtain drug-response models from patients’ progression-free survival data. Our strategy reveals links between oncogenic alterations, increasing the clinical impact of genomic profiling.

Strategies to Enhance Logic Modeling-Based Cell Line-Specific Drug Synergy Prediction

Discrete dynamical modeling shows promise in prioritizing drug combinations for screening efforts by reducing the experimental workload inherent to the vast numbers of possible drug combinations. We have investigated approaches to predict combination responses across different cancer cell lines using logic models generated from one generic prior-knowledge network representing 144 nodes covering major cancer signaling pathways. Cell-line specific models were configured to agree with baseline activity data from each unperturbed cell line. Testing against experimental data demonstrated a high number of true positive and true negative predictions, including also cell-specific responses. We demonstrate the possible enhancement of predictive capability of models by curation of literature knowledge further detailing subtle biologically founded signaling mechanisms in the model topology. In silico model analysis pinpointed a subset of network nodes highly influencing model predictions. Our results indicate that the performance of logic models can be improved by focusing on high-influence node protein activity data for model configuration and that these nodes accommodate high information flow in the regulatory network.

Comparison of Large, Medium, and Small Solid Tumor Gene Panels for Detection of Clinically Actionable Mutations in Cancer

BACKGROUND

Next-generation sequencing of gene panels has supplanted single-gene testing for cancer molecular diagnostics in many laboratories. Considerations for the optimal number of genes to assess in a panel depend on the purpose of the testing.

OBJECTIVE

To address the optimal size for the identification of clinically actionable variants in different-sized solid tumor sequencing panels.

PATIENTS AND METHODS

Sequencing results from 480 patients with a large, 315 gene, panel were compared against coverage of a medium, 161 gene, and small, 50 gene, panel.

RESULTS

The large panel detected a total of 2072 sequence variants in 480 patient specimens; 61 (12.7%) contained variants for which there is therapy approved by the US Food and Drug Administration, 89 (18.5%) had variants associated with an off-label therapy, and 312 (65.0%) contained variants eligible for a genomically matched clinical trial. The small panel covered only 737 of the 2072 variants (35.5%) and somewhat fewer therapy-related variants (on-label 88.5%, off-label 60.7%). The medium-size panel included 1354 of the 2072 (65.3%) variants reported by the large panel. All 318 patients with a clinically actionable variant would have been identified by the medium panel.

CONCLUSIONS

The results demonstrate that a carefully designed medium size gene panel is as effective as a large panel for the detection of clinically actionable variants and can be run by most molecular pathology laboratories.