Functional precision cancer medicine: drug sensitivity screening enabled by cell culture models

Functional precision medicine provides strategies to support clinical decisions on personalized treatment of cancer patients by using functional tests on live patient cells, for example, by testing cell viability after exposure to drugs.For hematological cancers, the methods and test platforms have matured to a stage where drug sensitivity testing is being implemented in prospective clinical trials as a stratification tool to complement genomic and transcriptomic information.For solid tumors, increasing evidence from co-clinical trials using a breadth of test systems to assess the drug sensitivity of patient cells coupled to clinical outcome allows validation of the tests.The field of functional testing in precision cancer medicine is thus moving towards its implementation as a diagnostic method to provide more information to molecular tumor boards. Functional precision medicine is a new, emerging area that can guide cancer treatment by capturing information from direct perturbations of tumor-derived, living cells, such as by drug sensitivity screening. Precision cancer medicine as currently implemented in clinical practice has been driven by genomics, and current molecular tumor boards rely extensively on genomic characterization to advise on therapeutic interventions. However, genomic biomarkers can only guide treatment decisions for a fraction of the patients. In this review we provide an overview of the current state of functional precision medicine, highlight advances for drug-sensitivity screening enabled by cell culture models, and discuss how artificial intelligence (AI) can be coupled to functional precision medicine to guide patient stratification. Functional precision medicine is a new, emerging area that can guide cancer treatment by capturing information from direct perturbations of tumor-derived, living cells, such as by drug sensitivity screening. Precision cancer medicine as currently implemented in clinical practice has been driven by genomics, and current molecular tumor boards rely extensively on genomic characterization to advise on therapeutic interventions. However, genomic biomarkers can only guide treatment decisions for a fraction of the patients. In this review we provide an overview of the current state of functional precision medicine, highlight advances for drug-sensitivity screening enabled by cell culture models, and discuss how artificial intelligence (AI) can be coupled to functional precision medicine to guide patient stratification. Functional precision medicine (see Glossary) is a rapidly advancing strategy to inform personalized treatment decisions for cancer patients based on functional readouts such as direct drug sensitivity testing of patient cancer cells (Box 1) [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar]. This approach was introduced more than four decades ago [2.Salmon S.E. et al.Quantitation of differential sensitivity of human-tumor stem cells to anticancer drugs.N. Engl. J. Med. 1978; 298: 1321-1327Crossref PubMed Scopus (848) Google Scholar,3.Shoemaker R.H. et al.Application of a human tumor colony-forming assay to new drug screening.Cancer Res. 1985; 45: 2145-2153PubMed Google Scholar], but initial reports found that functional assays were too premature for clinical implementation, in part because of the low fraction of tumor samples that could successfully be cultivated and tested for drug sensitivity in the laboratory [4.Selby P. et al.A critical appraisal of the 'human tumor stem-cell assay'.N. Engl. J. Med. 1983; 308: 129-134Crossref PubMed Scopus (320) Google Scholar,5.Von Hoff D.D. Send this patient's tumor for culture and sensitivity.N. Engl. J. Med. 1983; 308: 154-155Crossref PubMed Scopus (65) Google Scholar]. For solid tumors, studies have so far mostly been conducted retrospectively, and prospective evidence that is crucial for adopting such technologies in clinical decision making, is still lacking. The field is more mature for liquid tumors, and some prospective trials have already been published on the performance of functional assays for therapy response prediction [6.Skånland S.S. et al.Functional testing of relapsed chronic lymphocytic leukemia guides precision medicine and maps response and resistance mechanisms. An index case.Haematologica. 2022; 107: 1994-1998PubMed Google Scholar,7.Leonard J.T. et al.Functional and genetic screening of acute myeloid leukemia associated with mediastinal germ cell tumor identifies MEK inhibitor as an active clinical agent.J. Hematol. Oncol. 2016; 9: 31Crossref PubMed Scopus (23) Google Scholar] (Tables 1 and 2). These were preceded by retrospective studies in hematological cancers demonstrating that ex vivo drug sensitivity is associated with clinical responses to therapy [8.Lin L. et al.Ex-vivo drug testing predicts chemosensitivity in acute myeloid leukemia.J. Leukoc. Biol. 2020; 107: 859-870Crossref PubMed Scopus (10) Google Scholar, 9.Kuusanmäki H. et al.Phenotype-based drug screening reveals association between venetoclax response and differentiation stage in acute myeloid leukemia.Haematologica. 2020; 105: 708-720Crossref PubMed Scopus (53) Google Scholar, 10.Spinner M.A. et al.Ex vivo drug screening defines novel drug sensitivity patterns for informing personalized therapy in myeloid neoplasms.Blood Adv. 2020; 4: 2768-2778Crossref PubMed Scopus (14) Google Scholar]. Although functional precision medicine has proved to be valuable in patient stratification for treatments in clinical trials and in some real-world case reports, considerable efforts are still required before it can be implemented in routine clinical practice. However, improved cell culture protocols and technology have now advanced functional precision medicine to a point where we need to explore its potential to guide treatment decisions in clinical trials (Table 1) [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar,11.Malani D. et al.Implementing a functional precision medicine tumor board for acute myeloid leukemia.Cancer Discov. 2022; 12: 388-401Crossref PubMed Scopus (25) Google Scholar,12.Kornauth C. et al.Functional precision medicine provides clinical benefit in advanced aggressive hematologic cancers and identifies exceptional responders.Cancer Discov. 2022; 12: 372-387Crossref PubMed Scopus (27) Google Scholar]. In this review we provide an overview of the current state of functional precision medicine, highlight advances for drug sensitivity screening enabled by cell culture models, and discuss how AI can be coupled to functional precision medicine to guide patient stratification.Box 1Functional precision medicineFunctional precision medicine is a diagnostic discipline that takes into account cell and tissue responses to perturbations. This is in contrast to traditional pathology diagnostics which focuses on static conditions of cells and tissues at specific timepoints and locations of the disease [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar,73.Letai A. Functional precision cancer medicine-moving beyond pure genomics.Nat. Med. 2017; 23: 1028-1035Crossref PubMed Scopus (186) Google Scholar]. Perturbations represent controlled modulation of culture conditions, and can include drug exposure, immune stimulation, temperature control, and gene expression modulation. Readouts span all measurements that can reliably be collected from cultures, and can include viability assays as well as proteomic, transcriptomic, and metabolomic analyses [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar]. Cells under study can include single cell types (e.g., cancer cells), specified multiple cell types (e.g., cancer–immune cell cocultures), or unspecified multicell cultures such as microtissue collections derived directly from tumors or other tissues of interest [34.Wang Y. Jeon H. 3D cell cultures toward quantitative high-throughput drug screening.Trends Pharmacol. Sci. 2022; 43: 569-581Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar].Table 1Prospective studies in hematological cancers using functional assays to guide cancer therapyCancer typePatients included in the studyFunctional approachClinical response to treatmentRefsHematological cancers143 patients; 56 (39%) patients received treatmentImage-based single-cell drug profiling30 patients (54%) achieved more than 1.3-fold enhanced progression-free survival compared with their previous line of therapy[12.Kornauth C. et al.Functional precision medicine provides clinical benefit in advanced aggressive hematologic cancers and identifies exceptional responders.Cancer Discov. 2022; 12: 372-387Crossref PubMed Scopus (27) Google Scholar]AML186 patients; 37 patients (20%) received treatmentDrug sensitivity testingClinically meaningful complete or partial responses in 17 of 29 patients (59% objective response rate)[11.Malani D. et al.Implementing a functional precision medicine tumor board for acute myeloid leukemia.Cancer Discov. 2022; 12: 388-401Crossref PubMed Scopus (25) Google Scholar]Mediastinal germ cell tumor and AMLOne relapsed/refractory patientDrug sensitivity testingStable disease (AML), relapse of metastatic germ cell tumor after 5 months of therapy[7.Leonard J.T. et al.Functional and genetic screening of acute myeloid leukemia associated with mediastinal germ cell tumor identifies MEK inhibitor as an active clinical agent.J. Hematol. Oncol. 2016; 9: 31Crossref PubMed Scopus (23) Google Scholar]CLLOne relapsed/refractory patientDrug sensitivity testingPartial response[6.Skånland S.S. et al.Functional testing of relapsed chronic lymphocytic leukemia guides precision medicine and maps response and resistance mechanisms. An index case.Haematologica. 2022; 107: 1994-1998PubMed Google Scholar]CLLOne relapsed/refractory patientDrug sensitivity testingPartial response[19.Yin Y. et al.Functional testing of PI3K inhibitors stratifies responders to idelalisib and identifies treatment vulnerabilities in idelalisib-refractory/intolerant chronic lymphocytic leukemia..BioRxiv. 2022; (Published online April 15, 2022. https://doi.org/10.1101/2022.04.14.488428)Google Scholar] Open table in a new tab Table 2Co-clinical trials in patients with solid tumors and matching functional testing with clinical outcomeaAbbreviations: FOLFIRI, folinic acid, fluorouracil, and irinotecan; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; 5-FU, 5-fluorouracil., bWhere available, test sensitivity and specificity are reported for the performance of the organoid platform in predicting clinically observed patient responses.Cancer typePatients included in the studyFunctional approachClinical response to treatmentRefsCRC32 patients; 12 received FOLFIRI, ten received irinotecan, and ten received FOLFOXDrug sensitivity testingFor predicting patient responses to FOLFIRI, the test had 100% specificity, 83% sensitivity. The test was not effective for predicting responses for the other tested chemotherapy regimens[38.Ooft S.N. et al.Patient-derived organoids can predict response to chemotherapy in metastatic colorectal cancer patients.Sci. Transl. Med. 2019; 11eaay2574Crossref PubMed Scopus (321) Google Scholar]CRC30 patients in pilot; 71 patients in blinded studyDrug sensitivity testing63% sensitivity, 94% specificity[74.Wang T. et al.Accuracy of using a patient-derived tumor organoid culture model to predict the response to chemotherapy regimens in stage IV colorectal cancer: a blinded study.Dis. Colon Rectum. 2021; 64: 833-850Crossref PubMed Scopus (17) Google Scholar]CRC11 patientsDrug sensitivity testing50% sensitivity, 100% specificity[75.Chalabi M. et al.Neoadjuvant immunotherapy leads to pathological responses in MMR-proficient and MMR-deficient early-stage colon cancers.Nat. Med. 2020; 26: 566-576Crossref PubMed Scopus (476) Google Scholar]Locally advanced rectal cancer80 patients; all received chemoradiotherapy in a neoadjuvant setting; organoids were tested against 5-FU, irinotecan, radiation, or chemoradiationDrug- and radiotherapy sensitivity testingFor chemoradiation 78% sensitivity and 92% specificity[76.Yao Y. et al.Patient-derived organoids predict chemoradiation responses of locally advanced rectal cancer.Cell Stem Cell. 2020; 26: 17-26Abstract Full Text Full Text PDF PubMed Scopus (277) Google Scholar]Esophageal adenocarcinomasTen patients; patients received ECX (epirubicin, oxaliplastin, and capecitabine), CF (cisplatin and 5-FU), or no chemotherapyDrug sensitivity testingFor organoid cultures that were considered to be insensitive to drugs prescribed to the patients, the drug resistance matched the observed high tumor regression grade (TRG)[77.Li X. et al.Organoid cultures recapitulate esophageal adenocarcinoma heterogeneity providing a model for clonality studies and precision therapeutics.Nat. Commun. 2018; 9: 2983Crossref PubMed Scopus (157) Google Scholar]Breast cancer12 patients with clinical follow-up dataDrug sensitivity screeningTamoxifen was the only drug for which differential responses were recorded (one sensitive, one insensitive, the remainder were undetermined), and the observed organoid drug sensitivity matched clinical observations[78.Sachs N. et al.A living biobank of breast cancer organoids captures disease heterogeneity.Cell. 2018; 172: 373-386Abstract Full Text Full Text PDF PubMed Scopus (885) Google Scholar]Gastrointestinal cancer11 patients with CRC; four patients with gastroesophageal cancerDrug sensitivity screening for several chemotherapies and targeted drugs, including paclitaxel, regorafenib, cetuximab, and investigational compoundsFor the therapies administered to organoids and to patients, the organoid assays demonstrated a predictive performance of 100% sensitivity, 93% specificity, 88% positive predictive value, and 100% negative predictive value when compared to clinical response data[79.Vlachogiannis G. et al.Patient-derived organoids model treatment response of metastatic gastrointestinal cancers.Science. 2018; 359: 920-926Crossref PubMed Scopus (918) Google Scholar]Head and neck squamous cell carcinomaSeven patients for which radiotherapy testing was performed and compared to clinical responsesDrug and radiotherapy sensitivity screeningCorrelation between relapses and therapy sensitivity; of four organoids least sensitive to therapy, three patients experienced a relapse, whereas for the three most sensitive organoids no patients experienced a relapse within the observed period[80.Driehuis E. et al.Oral mucosal organoids as a potential platform for personalized cancer therapy.Cancer Discov. 2019; 9: 852-871Crossref PubMed Scopus (169) Google Scholar]Ovarian cancerFive patients with drug sensitivity testing and clinical follow-up data. For two patients, two organoids were derivedDrug sensitivity testingThree patients with organoids sensitive to therapy achieved stable disease; the two patients with the least sensitive organoids had progressive disease[81.de Witte C.J. et al.Patient-derived ovarian cancer organoids mimic clinical response and exhibit heterogeneous inter- and intrapatient drug responses.Cell Rep. 2020; 31107762Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar]Pancreatic cancer11 patients with matched organoid and clinical outcome dataDrug sensitivity testingOrganoids from four patients were found to be insensitive to all tested drugs and patients from whom these were derived experienced progressive disease upon therapy with the same drug cocktail. For the organoids from seven patients that were sensitive to at least one drug in the tested drug combination, all seven patients experienced stable disease or better[82.Grossman J.E. et al.Organoid sensitivity correlates with therapeutic response in patients with pancreatic cancer.Clin. Cancer Res. 2022; 28: 708-718Crossref PubMed Scopus (23) Google Scholar]Locally advanced rectal cancer17 patients with matched organoid and clinical outcome dataFunctional immunotoxicity assayAll six patients who were classified as complete responders were correctly classified based on tumor-infiltrating lymphocyte scores[83.Kong J.C.H. et al.Tumor-infiltrating lymphocyte function predicts response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer.JCO Precis Oncol. 2018; 2: 1-15Crossref PubMed Scopus (39) Google Scholar]a Abbreviations: FOLFIRI, folinic acid, fluorouracil, and irinotecan; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; 5-FU, 5-fluorouracil.b Where available, test sensitivity and specificity are reported for the performance of the organoid platform in predicting clinically observed patient responses. Open table in a new tab Functional precision medicine is a diagnostic discipline that takes into account cell and tissue responses to perturbations. This is in contrast to traditional pathology diagnostics which focuses on static conditions of cells and tissues at specific timepoints and locations of the disease [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar,73.Letai A. Functional precision cancer medicine-moving beyond pure genomics.Nat. Med. 2017; 23: 1028-1035Crossref PubMed Scopus (186) Google Scholar]. Perturbations represent controlled modulation of culture conditions, and can include drug exposure, immune stimulation, temperature control, and gene expression modulation. Readouts span all measurements that can reliably be collected from cultures, and can include viability assays as well as proteomic, transcriptomic, and metabolomic analyses [1.Letai A. et al.Functional precision oncology: testing tumors with drugs to identify vulnerabilities and novel combinations.Cancer Cell. 2022; 40: 26-35Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar]. Cells under study can include single cell types (e.g., cancer cells), specified multiple cell types (e.g., cancer–immune cell cocultures), or unspecified multicell cultures such as microtissue collections derived directly from tumors or other tissues of interest [34.Wang Y. Jeon H. 3D cell cultures toward quantitative high-throughput drug screening.Trends Pharmacol. Sci. 2022; 43: 569-581Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar]. The evidence level for functional precision medicine is now being transformed from retrospective trials to prospective study designs (see Tables 1 and 2 for prospective trials in hematological malignancies and retrospective trials in solid tumors, respectively). This is exemplified by the study by Malani et al. where a multidisciplinary functional precision medicine molecular tumor board was created to guide clinical decisions for patients with acute myeloid leukemia (AML) (Table 1) [11.Malani D. et al.Implementing a functional precision medicine tumor board for acute myeloid leukemia.Cancer Discov. 2022; 12: 388-401Crossref PubMed Scopus (25) Google Scholar]. The authors reported that actionable drugs were identified for 97% of the patients, and treatment recommendations were implemented for 37 individuals with a 59% objective response rate [11.Malani D. et al.Implementing a functional precision medicine tumor board for acute myeloid leukemia.Cancer Discov. 2022; 12: 388-401Crossref PubMed Scopus (25) Google Scholar]. In another example, the EXALT trial (Extended Analysis for Leukemia/Lymphoma Treatment), treatment of 56 heavily treated patients with advanced hematological malignancies was guided based on drug testing (Table 1) [12.Kornauth C. et al.Functional precision medicine provides clinical benefit in advanced aggressive hematologic cancers and identifies exceptional responders.Cancer Discov. 2022; 12: 372-387Crossref PubMed Scopus (27) Google Scholar]. In this study clinical benefit was defined as minimum 1.3-fold prolonged progression-free survival relative to that obtained with the previous line of therapy. Thirty patients (54%) achieved this at a median follow-up of 23.9 months [12.Kornauth C. et al.Functional precision medicine provides clinical benefit in advanced aggressive hematologic cancers and identifies exceptional responders.Cancer Discov. 2022; 12: 372-387Crossref PubMed Scopus (27) Google Scholar]. The ongoing EXALT-2 trial (NCT04470947) is comparing treatment guided by functional drug screening, genomic profiling, and physician’s choice. The results from this study promise to add new insights into the strategies for next-generation clinical decision support. Suspension-based techniques, tissue architecture-preserving models, and emerging in vivo methods that include implantable devices to maintain the multi-cell composition of the cancer are used in drug sensitivity screening. Current efforts have focused on applying suspension-based techniques to solid tumor organoids, which can potentially advance throughput in drug screening, bypassing some of the cumbersome steps associated with solid substrates for organoid growth [13.Capeling M.M. et al.Suspension culture promotes serosal mesothelial development in human intestinal organoids.Cell Rep. 2022; 38110379Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar,14.Hirokawa Y. et al.Low-viscosity matrix suspension culture enables scalable analysis of patient-derived organoids and tumoroids from the large intestine.Commun. Biol. 2021; 4: 1067Crossref PubMed Scopus (9) Google Scholar]. Conversely, because tumor cells are known to communicate extensively with a variety of host cells, coculture methods, originally developed in solid tumor assays, are now implemented in suspension-based protocols [15.Scielzo C. Ghia P. Modeling the leukemia microenvironment in vitro.Front. Oncol. 2020; 10607608Crossref PubMed Scopus (16) Google Scholar,16.Athanasiadis, P. et al. Computational pipeline for rational drug combination screening in patient-derived cells. In Data Mining Techniques for the Life Sciences (3rd edn) (Carugo, O. and Eisenhaber, F. eds), Humana Press (in press).Google Scholar] and tissue architecture-preserving models (Figure 1). We discuss and highlight recent developments in suspension-based, tissue architecture-preserving models and in vivo protocols below. Suspension-based models were initially developed for cultivating primary cells from hematological cancers in growth-supporting solutions. Hematological cancers appear as single-cell suspensions when obtained from blood samples or bone marrow draws. This appearance makes them readily dispensable and highly compatible with existing high-throughput drug screening protocols for characterization using larger compound libraries [17.Dietrich S. et al.Drug-perturbation-based stratification of blood cancer.J. Clin. Invest. 2018; 128: 427-445Crossref PubMed Scopus (86) Google Scholar,18.Schmidl C. et al.Combined chemosensitivity and chromatin profiling prioritizes drug combinations in CLL.Nat. Chem. Biol. 2019; 15: 232-240Crossref PubMed Scopus (27) Google Scholar]. One group recently carried out ex vivo drug sensitivity testing of 63 drugs on blood cancer samples from 246 patients. They showed that the malignancies, which could not be categorized based on genomic biomarkers for target vulnerabilities (Box 2) , could be stratified into subgroups based on therapy responses. The ex vivo drug sensitivities in drug response-defined subgroups were associated with treatment outcome [17.Dietrich S. et al.Drug-perturbation-based stratification of blood cancer.J. Clin. Invest. 2018; 128: 427-445Crossref PubMed Scopus (86) Google Scholar]. Another group identified treatment-induced changes in vulnerabilities which could inform individualized combination regimens using single-cell, image-based sensitivity profiling (pharmacoscopy) of paired samples. In this study the samples were collected from chronic lymphocytic leukemia (CLL) patients before and during treatment with the Bruton’s tyrosine kinase (BTK) inhibitor ibrutinib [18.Schmidl C. et al.Combined chemosensitivity and chromatin profiling prioritizes drug combinations in CLL.Nat. Chem. Biol. 2019; 15: 232-240Crossref PubMed Scopus (27) Google Scholar]. However, the study was performed on non-proliferating cells without stimuli from the tumor microenvironment in the bone marrow, and with only 18 h of drug exposure, because it has remained a challenge to cultivate CLL cells for longer periods of time, and this could reduce the clinical relevance of drug sensitivity readouts. Because CLL cells undergo spontaneous cell death when they are cultured ex vivo, several culture models have been developed to mimic the CLL tumor microenvironment. A key component for culture success is signals provided by T cells (CD40L, cytokines) and nurse-like cells (APRIL, BAFF), both of which are found in the microenvironment of CLL [15.Scielzo C. Ghia P. Modeling the leukemia microenvironment in vitro.Front. Oncol. 2020; 10607608Crossref PubMed Scopus (16) Google Scholar,16.Athanasiadis, P. et al. Computational pipeline for rational drug combination screening in patient-derived cells. In Data Mining Techniques for the Life Sciences (3rd edn) (Carugo, O. and Eisenhaber, F. eds), Humana Press (in press).Google Scholar]. Drug sensitivity testing on CLL cells that were precultured with these microenvironmental factors over longer time-periods (≥72 h, allowing for the relatively slow proliferation rate of CLL cells) has been used to guide personalized treatment of relapsed CLL (Table 1) [6.Skånland S.S. et al.Functional testing of relapsed chronic lymphocytic leukemia guides precision medicine and maps response and resistance mechanisms. An index case.Haematologica. 2022; 107: 1994-1998PubMed Google Scholar,19.Yin Y. et al.Functional testing of PI3K inhibitors stratifies responders to idelalisib and identifies treatment vulnerabilities in idelalisib-refractory/intolerant chronic lymphocytic leukemia..BioRxiv. 2022; (Published online April 15, 2022. https://doi.org/10.1101/2022.04.14.488428)Google Scholar], and protocols have also been developed to sustain the viability and proliferation of multiple myeloma (MM) cells for functional testing [20.Giliberto M. et al.Ex vivo drug sensitivity screening in multiple myeloma identifies drug combinations that act synergistically.Mol. Oncol. 2022; 16: 1241-1258Crossref PubMed Scopus (3) Google Scholar,21.Wang D. et al.Autologous bone marrow Th cells can support multiple myeloma cell proliferation in vitro and in xenografted mice.Leukemia. 2017; 31: 2114-2121Crossref PubMed Scopus (8) Google Scholar].Box 2Genomic precision medicine cannot advise on therapy for all patientsThe Human Genome Project [84.Lander E.S. et al.Initial sequencing and analysis of the human genome.Nature. 2001; 409: 860-921Crossref PubMed Scopus (17994) Google Scholar], which was declared completed in 2001, greatly accelerated both drug discovery targeting specific molecular aberrations in cancer and efforts to identify biomarkers that predict drug responses [85.Mateo J. et al.Delivering precision oncology to patients with cancer.Nat. Med. 2022; 28: 658-665Crossref PubMed Scopus (57) Google Scholar]. Several single genetic biomarkers have since been identified and approved for clinical decision support. For example, the BCR–ABL1 (breakpoint cluster region gene – Abelson proto-oncogene) fusion in chronic myelogenous leukemia (CML) is strongly linked to sensitivity to imatinib and its derivatives. For solid tumors, BRAF (v-Raf murine sarcoma viral oncogene homolog B1) V600E mutations in malignant melanoma and lung cancer are prominent examples of biomarkers that can select patients for therapy with drugs that inhibit BRAF/MEK (mitogen-activated protein kinase kinase) signaling [86.Planchard D. et al.Dabrafenib plus trametinib in patients with previously treated BRAF(V600E)-mutant metastatic non-small cell lung cancer: an open-label, multicentre phase 2 trial.Lancet Oncol. 2016; 17: 984-993Abstract Full Text Full Text PDF PubMed Scopus (611) Google Scholar,87.Druker B.J. et al.Efficacy and safety of a specific inhibitor of the BCR–ABL tyrosine kinase in chronic myeloid leukemia.N. Engl. J. Med. 2001; 344: 1031-1037Crossref PubMed Scopus (4487) Google Scholar]. In general, overall therapy responses to a particular drug have proved to be difficult to infer based on single genomic alterations. The basis for effective personalized treatment decisions is therefore shifting from single, 'static' genetic biomarkers to encompass global assessment of cancer omics data. Examples of more complex genetics-based biomarkers include genome-wide assessments of tumor mutation burden (TMB) and microsatellite instability (MSI).With decreasing costs and higher throughput, transcriptomic-based readouts have been tested for their ability not only to report static traits of tumor cells and tissues but also to capture the phenotypic properties of a tumor sample. Gene expression profiling has revealed subtypes within tumor types that do not reflect histologically recognized entities. The readouts of multiple gene transcripts can therefore yield 'signatures' or 'profiles' that correlate with clinical behavior and/or treatment responses beyond what can be predicted from histology-based d