Re-defining synthetic lethality by phenotypic profiling for precision oncology

High-throughput functional and genomic screening techniques provide systematic means for phenotypic discovery. Using synthetic lethality (SL) as a paradigm anticancer drug target discovery, we describe how these technologies may offer new possibilities to identify therapeutically relevant selective SL interactions by addressing some of the challenges that have made robust discovery candidates difficult. We further introduce an extended concept interaction, in which simultaneous perturbation two or more cellular components reduces cell viability than expected their individual effects, feel is highly befitting applications. also highlight potential benefits related computational quantification synergistic cancer selectivity. Finally, explore tumoral heterogeneity can be exploited find phenotype-specific precision oncology using high-throughput exciting opportunities methods identification subclonal interactions. Recent advances methods, such those based on comprehensive small-molecule libraries genome-wide CRISPR-based gene editing, provided researchers with cell-based combinatorial profiling. In research, therapeutic targets, lethal partners cancer-related genotypes phenotypes. There are excellent reviews recent developments mapping genetic screens, well factors should considered when developing analysis pipelines seeking implement novel into clinical practice (Shen Ideker, 2018Shen J.P. Ideker T. Synthetic networks oncology: promises pitfalls.J. Mol. Biol. 2018; 430: 2900-2912Crossref PubMed Scopus (14) Google Scholar; Tutuncuoglu Krogan, 2019Tutuncuoglu B. Krogan N.J. Mapping cancer: road rational combination therapies.Genome Med. 2019; 11: 62Crossref (12) Huang et al., 2020Huang A. Garraway L.A. Ashworth Weber engine discovery.Nat. Rev. Drug Discov. 2020; 19: 23-38Crossref (124) Topatana 2020Topatana W. Juengpanich S. Li Cao J. Hu Lee Suliyanto K. Ma D. Zhang Chen M. al.Advances therapy: mechanism translation.J. Hematol. Oncol. 13: 118Crossref (25) Scholar). Here, focus instead very definition, interpretation, usage targeted therapies light experimental techniques, aim maximal output exploit cancer-specific vulnerabilities oncology. outline applications emerging methodologies profiling phenotype-based prioritization phenotypes toward safe effective treatments. believe will improve relevant, Their has suffered past from relatively low validation rate large-scale RNAi screens wide range off-target effects (Brough 2011Brough R. Frankum J.R. Costa-Cabral Lord C.J. Searching cancer.Curr. Opin. Genet. Dev. 2011; 21: 34-41Crossref (84) Nijman, 2011Nijman S.M.B. lethality: general principles, utility detection human cells.FEBS Lett. 585: 1-6Crossref (187) Jaiswal 2017Jaiswal Peddinti G. Akimov Y. Wennerberg Kuznetsov Tang Aittokallio Seed-effect modeling improves consistency loss-of-function identifies cells.Genome 2017; 9: 51Crossref (11) Other biological rare difficult include (1) prevalent context dependency both backgrounds, (2) presence multiple modifiers buffering confounders, (3) intratumoral clonal evolution require single-cell approaches (Ashworth 2011Ashworth Reis-Filho J.S. Genetic progression treatment.Cell. 145: 30-38Abstract Full Text PDF (293) Nijman Friend, 2013Nijman Friend S.H. Cancer. Potential principle.Science. 2013; 342: 809-811Crossref (51) Alleviating through implementation modalities refined map higher-resolution interactions, hence leading treatment options revealing strategies overcoming resistance currently used therapies. For instance, several studies demonstrated widely SL-based poly (ADP-ribose) polymerase (PARP) other DNA damage response inhibitors (Topatana argue that, improved use applications, it important revisit original definition term been applied research. Originally, was described developed model organisms, fruit flies yeast, where genes were interaction if mutation either alone compatible organism viability, but produces phenotype (Dobzhansky, 1946Dobzhansky Genetics natural populations; recombination variability populations Drosophila pseudoobscura.Genetics. 1946; 31: 269-290Crossref Hartman 2001Hartman J.L.I.V. Garvik H. Principles variation.Science. 2001; 291: 1001-1004Crossref (591) Boone 2007Boone C. Bussey Andrews B.J. Exploring yeast.Nat. 2007; 8: 437-449Crossref (433) Brough Costanzo 2016Costanzo VanderSluis Koch E.N. Baryshnikova Pons Tan Wang Usaj Hanchard S.D. al.A global network maps wiring diagram function.Science. 2016; 353: aaf1420Crossref (547) Due driving role mutations many types, therefore seen strategy treatments patients harboring oncogenes tumor suppressors yeast genetics system, Hartwell Stephen proposed seminal article unique molecular changes distinguish cells wild-type (WT) counterparts could (Hartwell 1997Hartwell L.H. Szankasi P. Roberts Murray A.W. Integrating drugs.Science. 1997; 278: 1064-1068Crossref (568) Later, William Kaelin Jr, 2019 Nobel Prize recipient medicine physiology, this application famous his landmark paper (Kaelin, 2005Kaelin Jr., W.G. The therapy.Nat. 2005; 5: 689-698Crossref (1044) Scholar) he inhibition products exhibit should, selectively kill harbor mutations, while sparing WT “normal” cells. According principle, cancer-associated opportunities, even “undruggable” not accessible small molecules To historical outlook almost 20 years active research cancer, make projections field heading, carried out literature review abstract analysis. Notably, number papers focusing growing rapidly since 2007 (Figure 1A), 2 after between PARP BRCA (Farmer 2005Farmer McCabe N. Tutt A.N.J. Johnson D.A. Richardson T.B. Santarosa Dillon K.J. Hickson I. Knights al.Targeting repair defect mutant strategy.Nature. 434: 917-921Crossref (4269) This subsequent BRCA-mutant exquisitely sensitive enzyme, led era biomarker-driven various cancers (Pilié 2019Pilié P.G. Mills G.B. Yap T.A. State-of-the-art targeting cancer.Nat. Clin. 16: 81-104Crossref (354) Co-occurrence suggests BRCA1/2, TP53, PARP1/2 most-studied 1B). Over years, searched approaches. chemical libraries, engineered isogenic line pairs genetically characterized panels observed patients, possible systematically compounds inhibit genotype-specific manner (Nijman, rapid adoption RNAi-based started 1C). Although still most popular date searching hits, grow popularity 1D). particular, CRISPR-Cas9 single-gene knockout recently enabled search targets lines (Huang originally defined phenomenon inactivation no effect leads death 2A). however, actual diffused encompass events Such post-genetic protein microRNA expression, pathway activation rewiring, microenvironment states, responses chemotherapeutic agents irradiation, among others (Asim 2017Asim Tarish F. Zecchini H.I. Sanjiv Gelali E. Massie C.E. Baridi Warren A.Y. Zhao Ogris al.Synthetic androgen receptor signalling prostate Commun. 374Crossref (93) Sizemore 2018Sizemore S.T. Mohammad G.M. Nowsheen Yu Ostrowski M.C. Chakravarti Xia ionizing radiation p53-dependent.Mol. Cancer Res. 1092-1102Crossref (20) Noack 2018Noack Raab Matthess Sanhaji Krämer Gy?rffy Kaderali L. El-Balat Becker Strebhardt -amplified high grade serous ovarian combined Polo-like kinase 1 microtubule dynamics.Oncotarget. 25842-25859Crossref (19) Zhu 2018Zhu Xu Deyanat-Yazdi Peng S.X. Barnes Narla R.K. Tran Mikolon Ning Shi potent TTK CLK1/2 inhibitor triple-negative breast compromised G-S checkpoint.Mol. Ther. 17: 1727-1738Crossref Kerins 2018Kerins M.J. Milligan Wohlschlegel J.A. Ooi Fumarate hydratase hereditary leiomyomatosis renal ferroptosis induction.Cancer Sci. 109: 2757-2766Crossref Behan 2019Behan F.M. Iorio Picco Gonçalves Beaver C.M. Migliardi Santos Rao Sassi Pinnelli al.Prioritization screens.Nature. 568: 511-516Crossref (377) Chan 2019Chan E.M. Shibue McFarland J.M. Gaeta Ghandi Dumont Gonzalez McPartlan al.WRN helicase microsatellite unstable cancers.Nature. 551-556Crossref (122) 2019Chen Liang X. Chang Henry V.K. Lan Z. Spring D.J. al.Symbiotic macrophage-glioma reveal PTEN-null glioma.Cancer Cell. 35: 868-884.e6Abstract (86) Hinze 2019Hinze Pfirrmann Karim Degar McGuckin Vinjamur Sacher Stevenson K.E. Neuberg D.S. Orellana Wnt asparaginase drug-resistant acute leukemias.Cancer 664-676.e7Abstract (40) 2019Lee H.Y. Cha Kim S.K. Park J.H. Song K.H. M.-Y. c-MYC drives metastasis brain, promotes TRAIL.Mol. 544-554Crossref (32) Serebrennikova 2019Serebrennikova O.B. Paraskevopoulou M.D. Aguado-Fraile Taraslia V. Ren Thapa Roper Du Croce Tsichlis P.N. knockdown TNF? causes via caspase-8 carcinoma lines.Proc. Natl. Acad. U S 116: 14039-14048Crossref (4) Srinivasan 2019Srinivasan Williamson E.A. Kong A.S. H.-S. Schärer O. Burma Sung al.MiR223-3p BRCA1-deficient cancers.Proc. 17438-17443Crossref Sun 2019Sun Bandi Lofton Smith Bristow C.A. Carugo Rogers Leonard Q. Mullinax al.Functional genomics reveals phosphogluconate dehydrogenase oxidative phosphorylation.Cell Rep. 26: 469-482.e5Abstract (21) Wéra 2019Wéra A.-C. Lobbens Stoyanov Lucas Michiels Radiation-induced poly(ADP-ribose) RAD51 sensitize proton irradiation.Cell Cycle. 18: 1770-1783Crossref (17) Yang 2019Yang Feng Fu Jiang Bi Lin Zhou AT-101 idarubicin myeloid leukemia blockade intrinsic apoptotic pathway.Cancer 461: 31-43Crossref (5) extend far beyond inactivation, they activating environmental factors, epigenetic changes, making somewhat ambiguous. Another reason confusion often reduction does match binary dead/alive definition. partly independent context-dependent challenging Scholar), especially metrics. account discrepancies, propose (see Figure 2B). note requires quantitative borrows synergy scoring drug-induced (details references given elsewhere relation Quantitative combinations). implies less dramatic classification principle 2A), depending non-interaction background single perturbations defined. Even though extends nature ways, think actually reflects accurately its current field. Our includes special cases varying subclasses (e.g., “synthetic sickness”; allowing purely results, thus minimizing uncertainty associated mechanism-driven concepts “induced essentiality,” “non-cell-autonomous SL,” “metabolic forms classes 2020Li Cai Development classification.Signal Transduct. Target. 241Crossref Regardless formally defined, main motivation remains same; is, only one uniquely present cells, healthy principal objective (i.e., increasing window; forgotten over selectivity implicitly assumed specific oncogenes. argued BRCA/PARP so successfully translated clinic (Nijman Prof. opposing view 2009, suggesting indices all drugs “almost certainly relate poorly understood” 2009Kaelin framework development wiser therapeutics.Genome 2009; 1: 99Crossref (72) Indeed, differences normal distinctive feature makes them susceptible treatment. However, imposes different elements perturbed, oncogene-addiction acting direct cancer-promoting 2C); therefore, cannot work mechanism. contrast, act mechanisms Immunotherapy programmed death-ligand (PD-L1) inhibitors, specifically display neoantigens-MHC complex appropriate immune machinery; hence, PD-L1 neoantigen-MHC dependent context, or, formally, three-component interaction. chemotherapeutics, like cisplatin, proliferative Since do proliferate cisplatin exhibits greater activity generating required window. Overall, commonly working and, needed. Combining enables synergize (or to) approved pre-clinical compounds. Besides straightforward well-described chemogenetic integrated approach provides mediating example, Durocher coworkers utilized whole-genome CRISPR pathways survival 27 genotoxic agents, unbiased types (Olivieri 2020Olivieri Cho Álvarez-Quilón Schellenberg Zimmermann Hustedt Rossi S.E. Adam Melo cells.Cell. 182: 481-496.e21Abstract (102) Similarly, combining rich resource clues underpinnings putative mode action, prove great importance refining strategies. These possibilities, methodological challenges, experiments strong selection pressure doses drugs), standard show suboptimal performance identifying differentially represented sequencing tags guide RNAs) (Akimov 2020Akimov Bulanova Timonen Improved barcodes phenomics.Mol. Syst. e9195Crossref Furthermore, variable dynamics, vivo vitro (Gerrits 2010Gerrits Dykstra Kalmykowa O.J. Klauke Verovskaya Broekhuis M.J.C. de Haan Bystrykh L.V. Cellular barcoding tool hematopoietic system.Blood. 2010; 115: 2610-2618Crossref (154) Porter 2014Porter S.N. Baker L.C. Mittelman Porteus M.H. Lentiviral ongoing dynamics vivo.Genome 2014; 15: R75Crossref (44) Merino 2019Merino T.S. Serrano Vaillant Liu Pal Di Stefano Schreuder al.Barcoding behavior patient-derived xenografts metastatic triple negative 10: 766Crossref (49) significantly affect read-count variance chemogenomic followed re-growth. depend strength pressure, accounting practical resemble clinically concentrations). Combinatorial represents orthogonal method mitigates non-druggable targets. Initially Wong 2016