Current Status and Future Directions: The Application of Artificial Intelligence/Machine Learning for Precision Medicine.

Technological innovations, such as artificial intelligence (AI) and machine learning (ML), have the potential to expedite the goal of precision medicine, especially when combined with increased capacity for voluminous data from multiple sources and expanded therapeutic modalities; however, they also present several challenges. In this communication, we first discuss the goals of precision medicine, and contextualize the use of AI in precision medicine by showcasing innovative applications (e.g., prediction of tumor growth and overall survival, biomarker identification using biomedical images, and identification of patient population for clinical practice) which were presented during the February 2023 virtual public workshop entitled "Application of Artificial Intelligence and Machine Learning for Precision Medicine," hosted by the US Food and Drug Administration (FDA) and University of Maryland Center of Excellence in Regulatory Science and Innovation (M-CERSI). Next, we put forward challenges brought about by the multidisciplinary nature of AI, particularly highlighting the need for AI to be trustworthy. To address such challenges, we subsequently note practical approaches, viz., differential privacy, synthetic data generation, and federated learning. The proposed strategies - some of which are highlighted presentations from the workshop - are for the protection of personal information and intellectual property. In addition, methods such as the risk-based management approach and the need for an agile regulatory ecosystem are discussed. Finally, we lay out a call for action that includes sharing of data and algorithms, development of regulatory guidance documents, and pooling of expertise from a broad-spectrum of stakeholders to enhance the application of AI in precision medicine.

Next-generation characterization of the Cancer Cell Line Encyclopedia.

Large panels of comprehensively characterized human cancer models, including the Cancer Cell Line Encyclopedia (CCLE), have provided a rigorous framework with which to study genetic variants, candidate targets, and small-molecule and biological therapeutics and to identify new marker-driven cancer dependencies. To improve our understanding of the molecular features that contribute to cancer phenotypes, including drug responses, here we have expanded the characterizations of cancer cell lines to include genetic, RNA splicing, DNA methylation, histone H3 modification, microRNA expression and reverse-phase protein array data for 1,072 cell lines from individuals of various lineages and ethnicities. Integration of these data with functional characterizations such as drug-sensitivity, short hairpin RNA knockdown and CRISPR-Cas9 knockout data reveals potential targets for cancer drugs and associated biomarkers. Together, this dataset and an accompanying public data portal provide a resource for the acceleration of cancer research using model cancer cell lines.

Resistance mechanisms to TP53-MDM2 inhibition identified by in vivo piggyBac transposon mutagenesis screen in an Arf-/- mouse model.

Inhibitors of double minute 2 protein (MDM2)-tumor protein 53 (TP53) interaction are predicted to be effective in tumors in which the TP53 gene is wild type, by preventing TP53 protein degradation. One such setting is represented by the frequent CDKN2A deletion in human cancer that, through inactivation of p14ARF, activates MDM2 protein, which in turn degrades TP53 tumor suppressor. Here we used piggyBac (PB) transposon insertional mutagenesis to anticipate resistance mechanisms occurring during treatment with the MDM2-TP53 inhibitor HDM201. Constitutive PB mutagenesis in Arf-/- mice provided a collection of spontaneous tumors with characterized insertional genetic landscapes. Tumors were allografted in large cohorts of mice to assess the pharmacologic effects of HDM201. Sixteen out of 21 allograft models were sensitive to HDM201 but ultimately relapsed under treatment. A comparison of tumors with acquired resistance to HDM201 and untreated tumors identified 87 genes that were differentially and significantly targeted by the PB transposon. Resistant tumors displayed a complex clonality pattern suggesting the emergence of several resistant subclones. Among the most frequent alterations conferring resistance, we observed somatic and insertional loss-of-function mutations in transformation-related protein 53 (Trp53) in 54% of tumors and transposon-mediated gain-of-function alterations in B-cell lymphoma-extra large (Bcl-xL), Mdm4, and two TP53 family members, resulting in expression of the TP53 dominant negative truncations ΔNTrp63 and ΔNTrp73. Enhanced BCL-xL and MDM4 protein expression was confirmed in resistant tumors, as well as in HDM201-resistant patient-derived tumor xenografts. Interestingly, concomitant inhibition of MDM2 and BCL-xL demonstrated significant synergy in p53 wild-type cell lines in vitro. Collectively, our findings identify several potential mechanisms by which TP53 wild-type tumors may escape MDM2-targeted therapy.

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

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

High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response.

Profiling candidate therapeutics with limited cancer models during preclinical development hinders predictions of clinical efficacy and identifying factors that underlie heterogeneous patient responses for patient-selection strategies. We established ∼1,000 patient-derived tumor xenograft models (PDXs) with a diverse set of driver mutations. With these PDXs, we performed in vivo compound screens using a 1 × 1 × 1 experimental design (PDX clinical trial or PCT) to assess the population responses to 62 treatments across six indications. We demonstrate both the reproducibility and the clinical translatability of this approach by identifying associations between a genotype and drug response, and established mechanisms of resistance. In addition, our results suggest that PCTs may represent a more accurate approach than cell line models for assessing the clinical potential of some therapeutic modalities. We therefore propose that this experimental paradigm could potentially improve preclinical evaluation of treatment modalities and enhance our ability to predict clinical trial responses.

Gene Expression Ratios Lead to Accurate and Translatable Predictors of DR5 Agonism across Multiple Tumor Lineages.

Death Receptor 5 (DR5) agonists demonstrate anti-tumor activity in preclinical models but have yet to demonstrate robust clinical responses. A key limitation may be the lack of patient selection strategies to identify those most likely to respond to treatment. To overcome this limitation, we screened a DR5 agonist Nanobody across >600 cell lines representing 21 tumor lineages and assessed molecular features associated with response. High expression of DR5 and Casp8 were significantly associated with sensitivity, but their expression thresholds were difficult to translate due to low dynamic ranges. To address the translational challenge of establishing thresholds of gene expression, we developed a classifier based on ratios of genes that predicted response across lineages. The ratio classifier outperformed the DR5+Casp8 classifier, as well as standard approaches for feature selection and classification using genes, instead of ratios. This classifier was independently validated using 11 primary patient-derived pancreatic xenograft models showing perfect predictions as well as a striking linearity between prediction probability and anti-tumor response. A network analysis of the genes in the ratio classifier captured important biological relationships mediating drug response, specifically identifying key positive and negative regulators of DR5 mediated apoptosis, including DR5, CASP8, BID, cFLIP, XIAP and PEA15. Importantly, the ratio classifier shows translatability across gene expression platforms (from Affymetrix microarrays to RNA-seq) and across model systems (in vitro to in vivo). Our approach of using gene expression ratios presents a robust and novel method for constructing translatable biomarkers of compound response, which can also probe the underlying biology of treatment response.

A screen of approved drugs and molecular probes identifies therapeutics with anti-Ebola virus activity.

Currently, no approved therapeutics exist to treat or prevent infections induced by Ebola viruses, and recent events have demonstrated an urgent need for rapid discovery of new treatments. Repurposing approved drugs for emerging infections remains a critical resource for potential antiviral therapies. We tested ~2600 approved drugs and molecular probes in an in vitro infection assay using the type species, Zaire ebolavirus. Selective antiviral activity was found for 80 U.S. Food and Drug Administration-approved drugs spanning multiple mechanistic classes, including selective estrogen receptor modulators, antihistamines, calcium channel blockers, and antidepressants. Results using an in vivo murine Ebola virus infection model confirmed the protective ability of several drugs, such as bepridil and sertraline. Viral entry assays indicated that most of these antiviral drugs block a late stage of viral entry. By nature of their approved status, these drugs have the potential to be rapidly advanced to clinical settings and used as therapeutic countermeasures for Ebola virus infections.

CDK 4/6 inhibitors sensitize PIK3CA mutant breast cancer to PI3K inhibitors.

Activation of the phosphoinositide 3-kinase (PI3K) pathway occurs frequently in breast cancer. However, clinical results of single-agent PI3K inhibitors have been modest to date. A combinatorial drug screen on multiple PIK3CA mutant cancers with decreased sensitivity to PI3K inhibitors revealed that combined CDK 4/6-PI3K inhibition synergistically reduces cell viability. Laboratory studies revealed that sensitive cancers suppress RB phosphorylation upon treatment with single-agent PI3K inhibitors but cancers with reduced sensitivity fail to do so. Similarly, patients' tumors that responded to the PI3K inhibitor BYL719 demonstrated suppression of pRB, while nonresponding tumors showed sustained or increased levels of pRB. Importantly, the combination of PI3K and CDK 4/6 inhibitors overcomes intrinsic and adaptive resistance leading to tumor regressions in PIK3CA mutant xenografts.

A meta-analysis approach for characterizing pan-cancer mechanisms of drug sensitivity in cell lines.

Understanding the heterogeneous drug response of cancer patients is essential to precision oncology. Pioneering genomic analyses of individual cancer subtypes have begun to identify key determinants of resistance, including up-regulation of multi-drug resistance (MDR) genes and mutational alterations of drug targets. However, these alterations are sufficient to explain only a minority of the population, and additional mechanisms of drug resistance or sensitivity are required to explain the remaining spectrum of patient responses to ultimately achieve the goal of precision oncology. We hypothesized that a pan-cancer analysis of in vitro drug sensitivities across numerous cancer lineages will improve the detection of statistical associations and yield more robust and, importantly, recurrent determinants of response. In this study, we developed a statistical framework based on the meta-analysis of expression profiles to identify pan-cancer markers and mechanisms of drug response. Using the Cancer Cell Line Encyclopaedia (CCLE), a large panel of several hundred cancer cell lines from numerous distinct lineages, we characterized both known and novel mechanisms of response to cytotoxic drugs including inhibitors of Topoisomerase 1 (TOP1; Topotecan, Irinotecan) and targeted therapies including inhibitors of histone deacetylases (HDAC; Panobinostat) and MAP/ERK kinases (MEK; PD-0325901, AZD6244). Notably, our analysis implicated reduced replication and transcriptional rates, as well as deficiency in DNA damage repair genes in resistance to TOP1 inhibitors. The constitutive activation of several signaling pathways including the interferon/STAT-1 pathway was implicated in resistance to the pan-HDAC inhibitor. Finally, a number of dysregulations upstream of MEK were identified as compensatory mechanisms of resistance to the MEK inhibitors. In comparison to alternative pan-cancer analysis strategies, our approach can better elucidate relevant drug response mechanisms. Moreover, the compendium of putative markers and mechanisms identified through our analysis can serve as a foundation for future studies into these drugs.

Inhibiting Tankyrases sensitizes KRAS-mutant cancer cells to MEK inhibitors via FGFR2 feedback signaling.

Tankyrases (TNKS) play roles in Wnt signaling, telomere homeostasis, and mitosis, offering attractive targets for anticancer treatment. Using unbiased combination screening in a large panel of cancer cell lines, we have identified a strong synergy between TNKS and MEK inhibitors (MEKi) in KRAS-mutant cancer cells. Our study uncovers a novel function of TNKS in the relief of a feedback loop induced by MEK inhibition on FGFR2 signaling pathway. Moreover, dual inhibition of TNKS and MEK leads to more robust apoptosis and antitumor activity both in vitro and in vivo than effects observed by previously reported MEKi combinations. Altogether, our results show how a novel combination of TNKS and MEK inhibitors can be highly effective in targeting KRAS-mutant cancers by suppressing a newly discovered resistance mechanism.

Characterization of the novel and specific PI3Kα inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials.

Somatic PIK3CA mutations are frequently found in solid tumors, raising the hypothesis that selective inhibition of PI3Kα may have robust efficacy in PIK3CA-mutant cancers while sparing patients the side-effects associated with broader inhibition of the class I phosphoinositide 3-kinase (PI3K) family. Here, we report the biologic properties of the 2-aminothiazole derivative NVP-BYL719, a selective inhibitor of PI3Kα and its most common oncogenic mutant forms. The compound selectivity combined with excellent drug-like properties translates to dose- and time-dependent inhibition of PI3Kα signaling in vivo, resulting in robust therapeutic efficacy and tolerability in PIK3CA-dependent tumors. Novel targeted therapeutics such as NVP-BYL719, designed to modulate aberrant functions elicited by cancer-specific genetic alterations upon which the disease depends, require well-defined patient stratification strategies in order to maximize their therapeutic impact and benefit for the patients. Here, we also describe the application of the Cancer Cell Line Encyclopedia as a preclinical platform to refine the patient stratification strategy for NVP-BYL719 and found that PIK3CA mutation was the foremost positive predictor of sensitivity while revealing additional positive and negative associations such as PIK3CA amplification and PTEN mutation, respectively. These patient selection determinants are being assayed in the ongoing NVP-BYL719 clinical trials.

FDA-approved selective estrogen receptor modulators inhibit Ebola virus infection.

Ebola viruses remain a substantial threat to both civilian and military populations as bioweapons, during sporadic outbreaks, and from the possibility of accidental importation from endemic regions by infected individuals. Currently, no approved therapeutics exist to treat or prevent infection by Ebola viruses. Therefore, we performed an in vitro screen of Food and Drug Administration (FDA)- and ex-US-approved drugs and selected molecular probes to identify drugs with antiviral activity against the type species Zaire ebolavirus (EBOV). From this screen, we identified a set of selective estrogen receptor modulators (SERMs), including clomiphene and toremifene, which act as potent inhibitors of EBOV infection. Anti-EBOV activity was confirmed for both of these SERMs in an in vivo mouse infection model. This anti-EBOV activity occurred even in the absence of detectable estrogen receptor expression, and both SERMs inhibited virus entry after internalization, suggesting that clomiphene and toremifene are not working through classical pathways associated with the estrogen receptor. Instead, the response appeared to be an off-target effect where the compounds interfere with a step late in viral entry and likely affect the triggering of fusion. These data support the screening of readily available approved drugs to identify therapeutics for the Ebola viruses and other infectious diseases. The SERM compounds described in this report are an immediately actionable class of approved drugs that can be repurposed for treatment of filovirus infections.

RAD001 enhances the potency of BEZ235 to inhibit mTOR signaling and tumor growth.

The mammalian target of rapamycin (mTOR) is regulated by oncogenic growth factor signals and plays a pivotal role in controlling cellular metabolism, growth and survival. Everolimus (RAD001) is an allosteric mTOR inhibitor that has shown marked efficacy in certain cancers but is unable to completely inhibit mTOR activity. ATP-competitive mTOR inhibitors such as NVP-BEZ235 can block rapamycin-insensitive mTOR readouts and have entered clinical development as anti-cancer agents. Here, we show the degree to which RAD001 and BEZ235 can be synergistically combined to inhibit mTOR pathway activation, cell proliferation and tumor growth, both in vitro and in vivo. RAD001 and BEZ235 synergized in cancer lines representing different lineages and genetic backgrounds. Strong synergy is seen in neuronal, renal, breast, lung, and haematopoietic cancer cells harboring abnormalities in PTEN, VHL, LKB1, Her2, or KRAS. Critically, in the presence of RAD001, the mTOR-4EBP1 pathway and tumorigenesis can be fully inhibited using lower doses of BEZ235. This is relevant since RAD001 is relatively well tolerated in patients while the toxicity profiles of ATP-competitive mTOR inhibitors are currently unknown.

The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity.

The systematic translation of cancer genomic data into knowledge of tumour biology and therapeutic possibilities remains challenging. Such efforts should be greatly aided by robust preclinical model systems that reflect the genomic diversity of human cancers and for which detailed genetic and pharmacological annotation is available. Here we describe the Cancer Cell Line Encyclopedia (CCLE): a compilation of gene expression, chromosomal copy number and massively parallel sequencing data from 947 human cancer cell lines. When coupled with pharmacological profiles for 24 anticancer drugs across 479 of the cell lines, this collection allowed identification of genetic, lineage, and gene-expression-based predictors of drug sensitivity. In addition to known predictors, we found that plasma cell lineage correlated with sensitivity to IGF1 receptor inhibitors; AHR expression was associated with MEK inhibitor efficacy in NRAS-mutant lines; and SLFN11 expression predicted sensitivity to topoisomerase inhibitors. Together, our results indicate that large, annotated cell-line collections may help to enable preclinical stratification schemata for anticancer agents. The generation of genetic predictions of drug response in the preclinical setting and their incorporation into cancer clinical trial design could speed the emergence of 'personalized' therapeutic regimens.

Identification of synergistic combinations of F508del cystic fibrosis transmembrane conductance regulator (CFTR) modulators.

Cystic fibrosis (CF) is an inherited, life-threatening disease caused by mutations in the gene encoding cystic fibrosis transmembrane conductance regulator (CFTR), an ABC transporter-class protein and ion channel that transports ions across epithelial cell membranes. The most common mutation leads to the deletion of a single phenylalanine, and the resulting protein, F508del-CFTR, shows reduced trafficking to the membrane and defective channel gating. The ideal therapeutic approach would address both of these defects and restore channel function at the same time. We describe here the application of a combination high-throughput screening to search for synergistic modulators of F508del-CFTR. With the adapted Fischer rat thyroid-yellow fluorescent protein halide flux assay to the combination high-throughput screening platform, we identified many interesting single agents as CFTR modulators from a library of approved drugs and mechanistic probe compounds, and combinations that synergistically modulate F508del-CFTR channel function in Fischer rat thyroid cells, demonstrating the potential for combination therapeutics to address the defects that cause CF.

Chemical combinations elucidate pathway interactions and regulation relevant to Hepatitis C replication.

The search for effective Hepatitis C antiviral therapies has recently focused on host sterol metabolism and protein prenylation pathways that indirectly affect viral replication. However, inhibition of the sterol pathway with statin drugs has not yielded consistent results in patients. Here, we present a combination chemical genetic study to explore how the sterol and protein prenylation pathways work together to affect hepatitis C viral replication in a replicon assay. In addition to finding novel targets affecting viral replication, our data suggest that the viral replication is strongly affected by sterol pathway regulation. There is a marked transition from antagonistic to synergistic antiviral effects as the combination targets shift downstream along the sterol pathway. We also show how pathway regulation frustrates potential hepatitis C therapies based on the sterol pathway, and reveal novel synergies that selectively inhibit hepatitis C replication over host toxicity. In particular, combinations targeting the downstream sterol pathway enzymes produced robust and selective synergistic inhibition of hepatitis C replication. Our findings show how combination chemical genetics can reveal critical pathway connections relevant to viral replication, and can identify potential treatments with an increased therapeutic window.

Synergistic drug combinations tend to improve therapeutically relevant selectivity.

Drug combinations are a promising strategy to overcome the compensatory mechanisms and unwanted off-target effects that limit the utility of many potential drugs. However, enthusiasm for this approach is tempered by concerns that the therapeutic synergy of a combination will be accompanied by synergistic side effects. Using large scale simulations of bacterial metabolism and 94,110 multi-dose experiments relevant to diverse diseases, we provide evidence that synergistic drug combinations are generally more specific to particular cellular contexts than are single agent activities. We highlight six combinations whose selective synergy depends on multitarget drug activity. For one anti-inflammatory example, we show how such selectivity is achieved through differential expression of the drugs' targets in cell types associated with therapeutic, but not toxic, effects and validate its therapeutic relevance in a rat model of asthma. The context specificity of synergistic combinations creates many opportunities for therapeutically relevant selectivity and enables improved control of complex biological systems.

Chemical combination effects predict connectivity in biological systems.

Efforts to construct therapeutically useful models of biological systems require large and diverse sets of data on functional connections between their components. Here we show that cellular responses to combinations of chemicals reveal how their biological targets are connected. Simulations of pathways with pairs of inhibitors at varying doses predict distinct response surface shapes that are reproduced in a yeast experiment, with further support from a larger screen using human tumour cells. The response morphology yields detailed connectivity constraints between nearby targets, and synergy profiles across many combinations show relatedness between targets in the whole network. Constraints from chemical combinations complement genetic studies, because they probe different cellular components and can be applied to disease models that are not amenable to mutagenesis. Chemical probes also offer increased flexibility, as they can be continuously dosed, temporally controlled, and readily combined. After extending this initial study to cover a wider range of combination effects and pathway topologies, chemical combinations may be used to refine network models or to identify novel targets. This response surface methodology may even apply to non-biological systems where responses to targeted perturbations can be measured.