The National Cancer Institute ALMANAC: A Comprehensive Screening Resource for the Detection of Anticancer Drug Pairs with Enhanced Therapeutic Activity
TL;DR: A systematic evaluation of the therapeutic activity of over 5,000 pairs of FDA-approved cancer drugs against a panel of 60 well-characterized human tumor cell lines to uncover combinations with greater than additive growth-inhibitory activity demonstrated value in identifying promising combinations of approved drugs with potent anticancer activity.
Abstract: To date, over 100 small-molecule oncology drugs have been approved by the FDA. Because of the inherent heterogeneity of tumors, these small molecules are often administered in combination to prevent emergence of resistant cell subpopulations. Therefore, new combination strategies to overcome drug resistance in patients with advanced cancer are needed. In this study, we performed a systematic evaluation of the therapeutic activity of over 5,000 pairs of FDA-approved cancer drugs against a panel of 60 well-characterized human tumor cell lines (NCI-60) to uncover combinations with greater than additive growth-inhibitory activity. Screening results were compiled into a database, termed the NCI-ALMANAC (A Large Matrix of Anti-Neoplastic Agent Combinations), publicly available at https://dtp.cancer.gov/ncialmanac Subsequent in vivo experiments in mouse xenograft models of human cancer confirmed combinations with greater than single-agent efficacy. Concomitant detection of mechanistic biomarkers for these combinations in vivo supported the initiation of two phase I clinical trials at the NCI to evaluate clofarabine with bortezomib and nilotinib with paclitaxel in patients with advanced cancer. Consequently, the hypothesis-generating NCI-ALMANAC web-based resource has demonstrated value in identifying promising combinations of approved drugs with potent anticancer activity for further mechanistic study and translation to clinical trials. Cancer Res; 77(13); 3564-76. ©2017 AACR.
Citations
More filters
TL;DR: The latest version of SynergyFinder 2.0 is described, which has extensively been upgraded through the addition of novel features supporting especially higher-order combination data analytics and exploratory visualization of multi-drug synergy patterns, along with automated outlier detection procedure, extended curve-fitting functionality and statistical analysis of replicate measurements.
Abstract: SynergyFinder (https://synergyfinder.fimm.fi) is a stand-alone web-application for interactive analysis and visualization of drug combination screening data. Since its first release in 2017, SynergyFinder has become a widely used web-tool both for the discovery of novel synergistic drug combinations in pre-clinical model systems (e.g. cell lines or primary patient-derived cells), and for better understanding of mechanisms of combination treatment efficacy or resistance. Here, we describe the latest version of SynergyFinder (release 2.0), which has extensively been upgraded through the addition of novel features supporting especially higher-order combination data analytics and exploratory visualization of multi-drug synergy patterns, along with automated outlier detection procedure, extended curve-fitting functionality and statistical analysis of replicate measurements. A number of additional improvements were also implemented based on the user requests, including new visualization and export options, updated user interface, as well as enhanced stability and performance of the web-tool. With these improvements, SynergyFinder 2.0 is expected to greatly extend its potential applications in various areas of multi-drug combinatorial screening and precision medicine.
475 citations
Cites background from "The National Cancer Institute ALMAN..."
...combinations that result in a higher than expected effects (8)....
[...]
AstraZeneca1, European Bioinformatics Institute2, University of Sheffield3, Sage Bionetworks4, Hungarian Academy of Sciences5, Semmelweis University6, RWTH Aachen University7, University of Michigan8, Korea University9, SAS Institute10, University of Nevada, Reno11, IBM12, Wellcome Trust Sanger Institute13, Icahn School of Medicine at Mount Sinai14, Heidelberg University15
TL;DR: A large drug combination dataset, consisting of 11,576 experiments from 910 combinations across 85 molecularly characterized cancer cell lines, and results of a DREAM Challenge to evaluate computational strategies for predicting synergistic drug pairs and biomarkers are reported.
Abstract: The effectiveness of most cancer targeted therapies is short-lived. Tumors often develop resistance that might be overcome with drug combinations. However, the number of possible combinations is vast, necessitating data-driven approaches to find optimal patient-specific treatments. Here we report AstraZeneca's large drug combination dataset, consisting of 11,576 experiments from 910 combinations across 85 molecularly characterized cancer cell lines, and results of a DREAM Challenge to evaluate computational strategies for predicting synergistic drug pairs and biomarkers. 160 teams participated to provide a comprehensive methodological development and benchmarking. Winning methods incorporate prior knowledge of drug-target interactions. Synergy is predicted with an accuracy matching biological replicates for >60% of combinations. However, 20% of drug combinations are poorly predicted by all methods. Genomic rationale for synergy predictions are identified, including ADAM17 inhibitor antagonism when combined with PIK3CB/D inhibition contrasting to synergy when combined with other PI3K-pathway inhibitors in PIK3CA mutant cells.
227 citations
TL;DR: The ubiquitin proteasome system (UPS) degrades individual proteins in a highly regulated fashion and is responsible for the degradation of misfolded, damaged, or unneeded cellular proteins as mentioned in this paper.
Abstract: The ubiquitin proteasome system (UPS) degrades individual proteins in a highly regulated fashion and is responsible for the degradation of misfolded, damaged, or unneeded cellular proteins. During the past 20 years, investigators have established a critical role for the UPS in essentially every cellular process, including cell cycle progression, transcriptional regulation, genome integrity, apoptosis, immune responses, and neuronal plasticity. At the center of the UPS is the proteasome, a large and complex molecular machine containing a multicatalytic protease complex. When the efficiency of this proteostasis system is perturbed, misfolded and damaged protein aggregates can accumulate to toxic levels and cause neuronal dysfunction, which may underlie many neurodegenerative diseases. In addition, many cancers rely on robust proteasome activity for degrading tumor suppressors and cell cycle checkpoint inhibitors necessary for rapid cell division. Thus, proteasome inhibitors have proven clinically useful to treat some types of cancer, especially multiple myeloma. Numerous cellular processes rely on finely tuned proteasome function, making it a crucial target for future therapeutic intervention in many diseases, including neurodegenerative diseases, cystic fibrosis, atherosclerosis, autoimmune diseases, diabetes, and cancer. In this review, we discuss the structure and function of the proteasome, the mechanisms of action of different proteasome inhibitors, various techniques to evaluate proteasome function in vitro and in vivo, proteasome inhibitors in preclinical and clinical development, and the feasibility for pharmacological activation of the proteasome to potentially treat neurodegenerative disease.
192 citations
Cites background from "The National Cancer Institute ALMAN..."
...Rechsteiner’s group later went on to purify the ATP-dependent 26S proteasome responsible for ubiquitin conjugate degradation (Hough et al., 1986, 1987)....
[...]
15 Jun 2020
TL;DR: This review provides an overview of the computational challenges and advances in drug response prediction, and focuses on comparing the machine learning techniques to be of utmost practical use for clinicians and machine learning non-experts.
Abstract: Cancer is a leading cause of death worldwide. Identifying the best treatment using computational models to personalize drug response prediction holds great promise to improve patient's chances of successful recovery. Unfortunately, the computational task of predicting drug response is very challenging, partially due to the limitations of the available data and partially due to algorithmic shortcomings. The recent advances in deep learning may open a new chapter in the search for computational drug response prediction models and ultimately result in more accurate tools for therapy response. This review provides an overview of the computational challenges and advances in drug response prediction, and focuses on comparing the machine learning techniques to be of utmost practical use for clinicians and machine learning non-experts. The incorporation of new data modalities such as single-cell profiling, along with techniques that rapidly find effective drug combinations will likely be instrumental in improving cancer care.
149 citations
TL;DR: It was shown that linear regression approaches, when considering chemical fingerprints as predictors, have the potential to achieve high accuracy of predicting the sensitivity of drug combinations and are freely available in DrugComb.
Abstract: Drug combination therapy has the potential to enhance efficacy, reduce dose-dependent toxicity and prevent the emergence of drug resistance. However, discovery of synergistic and effective drug combinations has been a laborious and often serendipitous process. In recent years, identification of combination therapies has been accelerated due to the advances in high-throughput drug screening, but informatics approaches for systems-level data management and analysis are needed. To contribute toward this goal, we created an open-access data portal called DrugComb (https://drugcomb.fimm.fi) where the results of drug combination screening studies are accumulated, standardized and harmonized. Through the data portal, we provided a web server to analyze and visualize users' own drug combination screening data. The users can also effectively participate a crowdsourcing data curation effect by depositing their data at DrugComb. To initiate the data repository, we collected 437 932 drug combinations tested on a variety of cancer cell lines. We showed that linear regression approaches, when considering chemical fingerprints as predictors, have the potential to achieve high accuracy of predicting the sensitivity of drug combinations. All the data and informatics tools are freely available in DrugComb to enable a more efficient utilization of data resources for future drug combination discovery.
131 citations
Cites methods from "The National Cancer Institute ALMAN..."
...The sources of the data include: i) The NCI ALMANAC dataset (26), ii) The ONEIL dataset (27), iii) The FORCINA dataset (28) and iv) The CLOUD dataset (29) (Table 1)....
[...]
References
More filters
TL;DR: A quantitative analysis of the toxicity of drugs or poisons applied jointly requires that they be administered at several dosages in mixtures containing fixed proportions of the ingredients, and the presence of synergism is indicated.
Abstract: Summary
A quantitative analysis of the toxicity of drugs or poisons applied jointly requires that they be administered at several dosages in mixtures containing fixed proportions of the ingredients. From a study of the dosage-mortality curves for several such mixtures, preferably in comparison with equivalent curves for the isolated active ingredients, most cases of combined action can be classified into one of three types:
(1) The first type is that in which the constituents act independently and diversely, so that the toxicity of any combination can be predicted from that of the isolated components and from the association of susceptibilities to the two components. The coefficient of association can be measured experimentally and should be constant at all proportions of the ingredients. When high, the toxicity of the mixture is reduced. The form of the dosage-mortality curve has been examined for several hypothetical mixtures. Whenever the curves for the two constituents were assumed to differ in slope, there was a relatively abrupt bend in the curve for the mixture, the rectilinear segments above and below the break approaching in slope the values for the original constituents. This observation indicates that in homogeneous populations the slope of a dosage-mortality curve is of toxicological significance. Since the same numerical relations would be expected if a single poison were to have two independent lethal effects within the animal, there is theoretical basis for fitting the linear segments of a dosage-mortality curve separately when a break occurs after transformation to probits and logarithms. This argument has been extended to time-mortality experiments to explain the smoothly concave curves characteristic of natural mortality.
(2) The second type of joint action is that in which the constituents act independently but similarly, so that one ingredient can be substituted at a constant ratio for any proportion of a second without altering the toxicity of the mixture. With homogeneous populations, dosage-mortality curves for the separate ingredients and for all mixtures should be parallel. Although by hypothesis the susceptibility to one ingredient is completely correlated with that to the other, mixtures in this category are more toxic than in the preceding class where association may vary from 0 to 1. The numerical relations have been illustrated by an experiment on the toxicity to the house-fly of solutions containing pyrethrin and rotenone. A mixture with a little less than four equitoxic units of pyrethrin to one of rotenone agreed closely with the definition but one in which the ingredients were about equally balanced showed a significantly greater toxicity than expected on the hypothesis of independent action, indicating the presence of synergism.
(3) Synergism forms the third type of joint action, characterized by a toxicity greater than that predicted from studies on the isolated constituents. It is the reverse of antagonism, which has not been considered directly. Two methods are proposed for the analysis of synergism. The more direct is to relate equitoxic dosages of mixture to its percentage composition in terms of the more active ingredient. When both are in logarithms the relation is linear over a useful range of compositions. This procedure preserves the original structure of the experiment, can be extended readily to three or more ingredients and leads to a convenient practical result. Theoretically it is less satisfactory than a second method in which for equitoxic dosages of each mixture the content of one ingredient (A) is related to the content of the other (B.) The equation which satisfies this relation most completely is (1 +k1A) Bi=k2, where the three constants are computed from the experimental data. When the exponent i is equal to 1, only two constants need be determined and their product, k1k2, is proposed as a measure of the intensity of synergism.
The synergism between a nitro-phenol and petroleum oil has been computed by both methods. For mixtures containing from 0·5 to 5% of the nitrophenol, the deposit of mixture (Dc) killing 98 % of the eggs of a plant bug could be expressed adequately in terms of the percentage of the phenol (Q) as log Dc= 0·687-0·307 log Q, for 98% of overwintering San Jose scale as log Dc= 0·472 -0·363 log Q. All observations, including those for a 0·1 % mixture and for oil alone which were omitted in the first method, could be fitted satisfactorily in terms of the separate ingredients. For plant bug eggs at LD50, (1+25·6A) B= 4·29 and for San Jose scale (1 + 66·7 A) B= 2·73: in both cases i= 1 and the intensity of synergism 110 and 182 respectively.
The full procedure has also been applied to the constituents of seven samples of Derris root. One sample gave an unaccountably low toxicity and was omitted. The log LD 50 of ether extract for the remaining six was related to the percentage composition of two components in the extract, rotenone (A) and dehydro mixture (B.) Since the toxicity of extract could be expressed almost entirely in terms of these particular two constituents, they were then related to each other by the second method. None of the samples contained a very small proportion of one ingredient, so that several equations were equally applicable, one of them being (1+0–714A) B= 56·1, from which the intensity of synergism was 40.
The problem of measuring synergism in fumigants has been discussed briefly.
1,970 citations
"The National Cancer Institute ALMAN..." refers methods in this paper
...Development and use of the NCI ComboScore Determination of combination benefit ("ComboScore") utilized a modification of Bliss independence (18), as follows: Let Yi ApBq be the growth fraction for the i cell line exposed to the p concentration of drug A and the q concentration of drug B, defined as:...
[...]
TL;DR: For 39 agents with both xenograft data and Phase II clinical trials results available, in vivo activity in a particular histology in a tumour model did not closely correlate with activity in the same human cancer histology, casting doubt on the correspondence of the pre-clinical models to clinical results.
Abstract: An analysis of the activity of compounds tested in pre-clinical in vivo and in vitro assays by the National Cancer Institute's Developmental Therapeutics Program was performed. For 39 agents with both xenograft data and Phase II clinical trials results available, in vivo activity in a particular histology in a tumour model did not closely correlate with activity in the same human cancer histology, casting doubt on the correspondence of the pre-clinical models to clinical results. However, for compounds with in vivo activity in at least one-third of tested xenograft models, there was correlation with ultimate activity in at least some Phase II trials. Thus, an efficient means of predicting activity in vivo models remains desirable for compounds with anti-proliferative activity in vitro. For 564 compounds tested in the hollow fibre assay which were also tested against in vivo tumour models, the likelihood of finding xenograft activity in at least one-third of the in vivo models tested rose with increasing intraperitoneal hollow fibre activity, from 8% for all compounds tested to 20% in agents with evidence of response in more than 6 intraperitoneal fibres (P< 0.0001). Intraperitoneal hollow fibre activity was also found to be a better predictor of xenograft activity than either subcutaneous hollow fibre activity or intraperitoneal plus subcutaneous activity combined. Since hollow fibre activity was a useful indicator of potential in vivo response, correlates with hollow fibre activity were examined for 2304 compounds tested in both the NCI 60 cell line in vitro cancer drug screen and hollow fibre assay. A positive correlation was found for histologic selectivity between in vitro and hollow fibre responses. The most striking correlation was between potency in the 60 cell line screen and hollow fibre activity; 56% of compounds with mean 50% growth inhibition below 10–7.5 M were active in more than 6 intraperitoneal fibres whereas only 4% of compounds with a potency of 10–4 M achieved the same level of hollow fibre activity (P< 0.0001). Structural parameters of the drugs analysed included compound molecular weight and hydrogen-bonding factors, both of which were found to be predictive of hollow fibre activity. © 2001 Cancer Research Campaign www.bjcancer.com
1,448 citations
"The National Cancer Institute ALMAN..." refers background in this paper
...Although these mechanistic details remain to be elucidated, the in vivo efficacy of the clofarabine– bortezomib combination, along with a prior NCI review of the relationship between efficacy in xenograft models and activity in human phase II trials (45) and the development of a validated multiplex assay for apoptotic pathway components (19), provided support for initiating a phase I trial of this drug pair in patients with advanced, refractory cancers (clinicaltrials....
[...]
TL;DR: It was found that 73–75% of identified ccRCC driver aberrations were subclonal, confounding estimates of driver mutation prevalence, and the proportion of C>T transitions at CpG sites increased during tumor progression.
Abstract: Clear cell renal carcinomas (ccRCCs) can display intratumor heterogeneity (ITH). We applied multiregion exome sequencing (M-seq) to resolve the genetic architecture and evolutionary histories of ten ccRCCs. Ultra-deep sequencing identified ITH in all cases. We found that 73-75% of identified ccRCC driver aberrations were subclonal, confounding estimates of driver mutation prevalence. ITH increased with the number of biopsies analyzed, without evidence of saturation in most tumors. Chromosome 3p loss and VHL aberrations were the only ubiquitous events. The proportion of C>T transitions at CpG sites increased during tumor progression. M-seq permits the temporal resolution of ccRCC evolution and refines mutational signatures occurring during tumor development.
1,105 citations
"The National Cancer Institute ALMAN..." refers background in this paper
...Days post implantation Vehicle Nilotinib (75) Paclitaxel (15) Paclitaxel (10) Nilotinib (75) + paclitaxel (15) Nilotinib (75) + paclitaxel (10) MDA-MB-468 C B...
[...]
...Paclitaxel has previously been shown to induce both apoptosis and necrosis in a cell-cycle stage–specific manner (49), and a number of kinase inhibitors are known to induce necrosis (50), suggesting a potential role for nilotinib in shifting the apoptotic–necroptotic balance in paclitaxel-induced cell death....
[...]
TL;DR: This technology enabled quantitative analysis of nearly 8,000 proteins and more than 20,000 phosphorylation, 15,000 ubiquitination and 3,000 acetylation sites per experiment, generating a holistic view of cellular signal transduction pathways as exemplified by analysis of bortezomib-treated human leukemia cells.
Abstract: We report a mass spectrometry-based method for the integrated analysis of protein expression, phosphorylation, ubiquitination and acetylation by serial enrichments of different post-translational modifications (SEPTM) from the same biological sample. This technology enabled quantitative analysis of nearly 8,000 proteins and more than 20,000 phosphorylation, 15,000 ubiquitination and 3,000 acetylation sites per experiment, generating a holistic view of cellular signal transduction pathways as exemplified by analysis of bortezomib-treated human leukemia cells.
538 citations
"The National Cancer Institute ALMAN..." refers background in this paper
...A whole-proteome analysis in Jurkat cells revealed that bortezomib treatment yielded a reproducible 12- to 13-fold upregulation of nucleoside diphosphate kinase, which catalyzes the final step of clofarabine activation, as well as more modest upregulation of the clofarabineactivating enzyme deoxycytidine kinase and the clofarabine targets DNA polymerase-a and e (27, 44)....
[...]
TL;DR: It is shown that wild-type p53 represses survivin expression at both mRNA and protein levels, suggesting that loss of survivin mediates, at least, in part the p53-dependent apoptotic pathway.
Abstract: Survivin is an inhibitor of apoptosis protein, which is over-expressed in most tumors. Aberrant expression of survivin and loss of wild-type p53 in many tumors prompted us to investigate a possible link between these two events. Here we show that wild-type p53 represses survivin expression at both mRNA and protein levels. Transient transfection analyses revealed that the expression of wild-type p53, but not mutant p53, was associated with strong repression of the survivin promoter in various cell types. The over-expression of exogenous survivin protein rescues cells from p53-induced apoptosis in a dose-dependent manner, suggesting that loss of survivin mediates, at least, in part the p53-dependent apoptotic pathway. In spite of the presence of two putative p53-binding sites in the survivin promoter, deletion and mutation analyses suggested that neither site is required for transcriptional repression of survivin expression. This was confirmed by chromatin immunoprecipitation assays. Further analyses suggested that the modification of chromatin within the survivin promoter could be a molecular explanation for silencing of survivin gene transcription by p53.
533 citations
"The National Cancer Institute ALMAN..." refers background in this paper
...The mechanism whereby clofarabine–bortezomib treatment reduces survivin levels in HCT116 xenografts could be either p53-dependent (41), as indicated by our Western blot analysis (Supplementary Fig....
[...]