Patient-derived micro-organospheres enable clinical precision oncology

•Clinical-biopsy-derived MOSs enable rapid drug testing in 14 days•MOSs maintain patient tumor microenvironment•MOSs capture patient tumor response to immunotherapy•MOS assay to test T cell potency against patient tumor for adoptive cell therapy Patient-derived xenografts (PDXs) and patient-derived organoids (PDOs) have been shown to model clinical response to cancer therapy. However, it remains challenging to use these models to guide timely clinical decisions for cancer patients. Here, we used droplet emulsion microfluidics with temperature control and dead-volume minimization to rapidly generate thousands of micro-organospheres (MOSs) from low-volume patient tissues, which serve as an ideal patient-derived model for clinical precision oncology. A clinical study of recently diagnosed metastatic colorectal cancer (CRC) patients using an MOS-based precision oncology pipeline reliably assessed tumor drug response within 14 days, a timeline suitable for guiding treatment decisions in the clinic. Furthermore, MOSs capture original stromal cells and allow T cell penetration, providing a clinical assay for testing immuno-oncology (IO) therapies such as PD-1 blockade, bispecific antibodies, and T cell therapies on patient tumors. Patient-derived xenografts (PDXs) and patient-derived organoids (PDOs) have been shown to model clinical response to cancer therapy. However, it remains challenging to use these models to guide timely clinical decisions for cancer patients. Here, we used droplet emulsion microfluidics with temperature control and dead-volume minimization to rapidly generate thousands of micro-organospheres (MOSs) from low-volume patient tissues, which serve as an ideal patient-derived model for clinical precision oncology. A clinical study of recently diagnosed metastatic colorectal cancer (CRC) patients using an MOS-based precision oncology pipeline reliably assessed tumor drug response within 14 days, a timeline suitable for guiding treatment decisions in the clinic. Furthermore, MOSs capture original stromal cells and allow T cell penetration, providing a clinical assay for testing immuno-oncology (IO) therapies such as PD-1 blockade, bispecific antibodies, and T cell therapies on patient tumors. IntroductionThe success of precision oncology relies on models that capture the morphological, molecular, and functional characteristics of patient tumors to accurately predict drug response and resistance. The development of various patient-derived models of cancer (PDMC) has provided important tools in this effort. Drug sensitivity assays using PDMC have recapitulated antitumor response in the clinic, underscoring their potential for guiding personalized care (Barretina et al., 2012Barretina J. Caponigro G. Stransky N. Venkatesan K. Margolin A.A. Kim S. Wilson C.J. Lehár J. Kryukov G.V. Sonkin D. et al.The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity.Nature. 2012; 483: 603-607Crossref PubMed Scopus (4668) Google Scholar; Gao et al., 2015Gao H. Korn J.M. Ferretti S. Monahan J.E. Wang Y. Singh M. Zhang C. Schnell C. Yang G. Zhang Y. et al.High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response.Nat. Med. 2015; 21: 1318-1325Crossref PubMed Scopus (745) Google Scholar; Lu et al., 2017Lu M. Zessin A.S. Glover W. Hsu D.S. Activation of the mTOR pathway by oxaliplatin in the treatment of colorectal cancer liver metastasis.PLoS One. 2017; 12: e0169439PubMed Google Scholar; Vlachogiannis et al., 2018Vlachogiannis G. Hedayat S. Vatsiou A. Jamin Y. Fernández-Mateos J. Khan K. Lampis A. Eason K. Huntingford I. Burke R. et al.Patient-derived organoids model treatment response of metastatic gastrointestinal cancers.Science. 2018; 359: 920-926Crossref PubMed Scopus (795) Google Scholar).However, the promise of clinical translation of PDMC—using drug sensitivity assays on patient-derived tissue to drive clinical decision-making—remains largely unrealized, in part, due to technical limitations with each model. Patient-derived cell lines exhibit genetic and morphological changes over time that make them unsuitable for clinical screening (Iorio et al., 2016Iorio F. Knijnenburg T.A. Vis D.J. Bignell G.R. Menden M.P. Schubert M. Aben N. Gonçalves E. Barthorpe S. Lightfoot H. et al.A landscape of pharmacogenomic interactions in Cancer.Cell. 2016; 166: 740-754Abstract Full Text Full Text PDF PubMed Scopus (888) Google Scholar; Stein et al., 2004Stein W.D. Litman T. Fojo T. Bates S.E. A serial analysis of gene expression (SAGE) database analysis of chemosensitivity: comparing solid tumors with cell lines and comparing solid tumors from different tissue origins.Cancer Res. 2004; 64: 2805-2816Crossref PubMed Scopus (100) Google Scholar). Patient-derived xenografts (PDXs) can be repeatedly passaged with retention of intratumoral clonal architecture and genetic diversity (Bruna et al., 2016Bruna A. Rueda O.M. Greenwood W. Batra A.S. Callari M. Batra R.N. Pogrebniak K. Sandoval J. Cassidy J.W. Tufegdzic-Vidakovic A. A biobank of breast cancer explants with preserved intra-tumor heterogeneity to screen anticancer compounds.Cell. 2016; 167: 260-274.e22Abstract Full Text Full Text PDF PubMed Scopus (268) Google Scholar), although serving as a model to study patient therapeutic response (Gao et al., 2015Gao H. Korn J.M. Ferretti S. Monahan J.E. Wang Y. Singh M. Zhang C. Schnell C. Yang G. Zhang Y. et al.High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response.Nat. Med. 2015; 21: 1318-1325Crossref PubMed Scopus (745) Google Scholar; Hidalgo et al., 2014Hidalgo M. Amant F. Biankin A.V. Budinská E. Byrne A.T. Caldas C. Clarke R.B. de Jong S. Jonkers J. Mælandsmo G.M. et al.Patient-derived xenograft models: an emerging platform for translational cancer research.Cancer Discov. 2014; 4: 998-1013Crossref PubMed Scopus (1026) Google Scholar). However, PDX are expensive and slow to generate, limiting their utility for diagnostic drug screens and precision medicine. In comparison, patient-derived organoids (PDOs) offer a lower cost and higher throughput model for clinical applications (Jenkins et al., 2018Jenkins R.W. Aref A.R. Lizotte P.H. Ivanova E. Stinson S. Zhou C.W. Bowden M. Deng J. Liu H. Miao D. et al.Ex vivo profiling of PD-1 blockade using organotypic tumor spheroids.Cancer Discov. 2018; 8: 196-215Crossref PubMed Scopus (249) Google Scholar; Neal et al., 2018Neal J.T. Li X. Zhu J. Giangarra V. Grzeskowiak C.L. Ju J. Liu I.H. Chiou S.H. Salahudeen A.A. Smith A.R. et al.Organoid modeling of the tumor immune microenvironment.Cell. 2018; 175: 1972-1988.e16Abstract Full Text Full Text PDF PubMed Scopus (527) Google Scholar; Yuki et al., 2020Yuki K. Cheng N. Nakano M. Kuo C.J. Organoid models of tumor immunology.Trends Immunol. 2020; 41: 652-664Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar). With large-scale biobanks of breast (Sachs et al., 2018Sachs N. de Ligt J. Kopper O. Gogola E. Bounova G. Weeber F. Balgobind A.V. Wind K. Gracanin A. Begthel H. A living biobank of breast cancer organoids captures disease heterogeneity.Cell. 2018; 172: 373-386.e10Abstract Full Text Full Text PDF PubMed Scopus (767) Google Scholar), colorectal (Sato et al., 2011Sato T. Stange D.E. Ferrante M. Vries R.G. Van Es J.H. Van Den Brink S. Van Houdt W.J. Pronk A. Van Gorp J. Siersema P.D. Clevers H. Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett's epithelium.Gastroenterology. 2011; 141: 1762-1772Abstract Full Text Full Text PDF PubMed Scopus (2010) Google Scholar; van de Wetering et al., 2015van de Wetering M. Francies H.E. Francis J.M. Bounova G. Iorio F. Pronk A. van Houdt W. van Gorp J. Taylor-Weiner A. Kester L. et al.Prospective derivation of a living organoid biobank of colorectal cancer patients.Cell. 2015; 161: 933-945Abstract Full Text Full Text PDF PubMed Scopus (1284) Google Scholar), ovarian (Kopper et al., 2019Kopper O. de Witte C.J. Lõhmussaar K. Valle-Inclan J.E. Hami N. Kester L. Balgobind A.V. Korving J. Proost N. Begthel H. et al.An organoid platform for ovarian cancer captures intra-and interpatient heterogeneity.Nat. Med. 2019; 25: 838-849Crossref PubMed Scopus (285) Google Scholar), pancreatic (Driehuis et al., 2019bDriehuis E. van Hoeck A. Moore K. Kolders S. Francies H.E. Gulersonmez M.C. Stigter E.C.A. Burgering B. Geurts V. Gracanin A. et al.Pancreatic cancer organoids recapitulate disease and allow personalized drug screening.Proc. Natl. Acad. Sci. USA. 2019; 116: 26580-26590Crossref Scopus (159) Google Scholar), brain (Jacob et al., 2020Jacob F. Salinas R.D. Zhang D.Y. Nguyen P.T.T. Schnoll J.G. Wong S.Z.H. Thokala R. Sheikh S. Saxena D. Prokop S. A patient-derived glioblastoma organoid model and biobank recapitulates inter-and intra-tumoral heterogeneity.Cell. 2020; 180: 188-204.e22Abstract Full Text Full Text PDF PubMed Scopus (282) Google Scholar), kidney (Calandrini et al., 2020Calandrini C. Schutgens F. Oka R. Margaritis T. Candelli T. Mathijsen L. Ammerlaan C. van Ineveld R.L. Derakhshan S. de Haan S. et al.An organoid biobank for childhood kidney cancers that captures disease and tissue heterogeneity.Nat. Commun. 2020; 11: 1310Crossref PubMed Scopus (115) Google Scholar), head and neck (Driehuis et al., 2019aDriehuis E. Spelier S. Beltrán Hernández I. de Bree R. M Willems S. Clevers H. Oliveira S. Patient-derived head and neck cancer organoids recapitulate EGFR expression levels of respective tissues and are responsive to EGFR-targeted photodynamic therapy.J. Clin. Med. 2019; 8: 1880Crossref Scopus (45) Google Scholar), and gastric cancers (Seidlitz et al., 2021Seidlitz T. Koo B.-K. Stange D.E. Gastric organoids—an in vitro model system for the study of gastric development and road to personalized medicine.Cell Death Differ. 2021; 28: 68-83Crossref PubMed Scopus (25) Google Scholar; Yan et al., 2018Yan H.H.N. Siu H.C. Law S. Ho S.L. Yue S.S.K. Tsui W.Y. Chan D. Chan A.S. Ma S. Lam K.O. A comprehensive human gastric cancer organoid biobank captures tumor subtype heterogeneity and enables therapeutic screening.Cell Stem Cell. 2018; 23: 882-897.e11Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar), PDO have been shown to capture patient diversity. Additionally, broad-based drug screens have reproduced known associations between genetic mutations and sensitivity to targeted therapies (Gao et al., 2014Gao D. Vela I. Sboner A. Iaquinta P.J. Karthaus W.R. Gopalan A. Dowling C. Wanjala J.N. Undvall E.A. Arora V.K. et al.Organoid cultures derived from patients with advanced prostate cancer.Cell. 2014; 159: 176-187Abstract Full Text Full Text PDF PubMed Scopus (883) Google Scholar; Skardal et al., 2015Skardal A. Devarasetty M. Rodman C. Atala A. Soker S. Liver-tumor hybrid organoids for modeling tumor growth and drug response in vitro.Ann. Biomed. Eng. 2015; 43: 2361-2373Crossref PubMed Scopus (92) Google Scholar). Thus, PDO have been exploited as a potential functional precision medicine technology for guiding treatment decisions in the clinic (Ganesh et al., 2019Ganesh K. Wu C. O'Rourke K.P. Szeglin B.C. Zheng Y. Sauvé C.G. Adileh M. Wasserman I. Marco M.R. Kim A.S. et al.A rectal cancer organoid platform to study individual responses to chemoradiation.Nat. Med. 2019; 25: 1607-1614Crossref PubMed Scopus (192) Google Scholar; Ooft et al., 2019Ooft S.N. Weeber F. Dijkstra K.K. McLean C.M. Kaing S. van Werkhoven E. Schipper L. Hoes L. Vis D.J. van de Haar J. et al.Patient-derived organoids can predict response to chemotherapy in metastatic colorectal cancer patients.Sci. Transl. Med. 2019; 11: eaay2574Crossref PubMed Scopus (257) Google Scholar; Pauli et al., 2017Pauli C. Hopkins B.D. Prandi D. Shaw R. Fedrizzi T. Sboner A. Sailer V. Augello M. Puca L. Rosati R. et al.Personalized in vitro and in vivo cancer models to guide precision medicine.Cancer Discov. 2017; 7: 462-477Crossref PubMed Scopus (505) Google Scholar; Tiriac et al., 2018Tiriac H. Belleau P. Engle D.D. Plenker D. Deschênes A. Somerville T.D.D. Froeling F.E.M. Burkhart R.A. Denroche R.E. Jang G.H. et al.Organoid profiling identifies common responders to chemotherapy in pancreatic cancer.Cancer Discov. 2018; 8: 1112-1129Crossref PubMed Scopus (437) Google Scholar; Vlachogiannis et al., 2018Vlachogiannis G. Hedayat S. Vatsiou A. Jamin Y. Fernández-Mateos J. Khan K. Lampis A. Eason K. Huntingford I. Burke R. et al.Patient-derived organoids model treatment response of metastatic gastrointestinal cancers.Science. 2018; 359: 920-926Crossref PubMed Scopus (795) Google Scholar; Yao et al., 2020Yao Y. Xu X. Yang L. Zhu J. Wan J. Shen L. Xia F. Fu G. Deng Y. Pan M. et al.Patient-derived organoids predict chemoradiation responses of locally advanced rectal cancer.Cell Stem Cell. 2020; 26: 17-26.e6Abstract Full Text Full Text PDF PubMed Scopus (223) Google Scholar).However, expanding sufficient numbers of PDO for drug screening remains too slow and inefficient for adoption into the clinic (; van de Wetering et al., 2015van de Wetering M. Francies H.E. Francis J.M. Bounova G. Iorio F. Pronk A. van Houdt W. van Gorp J. Taylor-Weiner A. Kester L. et al.Prospective derivation of a living organoid biobank of colorectal cancer patients.Cell. 2015; 161: 933-945Abstract Full Text Full Text PDF PubMed Scopus (1284) Google Scholar). Since clinical treatment decisions are typically made within 14 days of diagnosis, the existing timeframe for PDO generation would result in unacceptable treatment delays. Thus, to develop a clinically useful diagnostic assay, it is necessary to both accelerate PDO generation and functional testing as well as develop automated procedures from a core biopsy.Furthermore, given the growing clinical importance of immuno-oncology (IO), there is significant interest to reproduce physiological immune activity in organoid cultures. For example, peripheral blood lymphocyte and tumor organoid coculture models have been used to test tumor-reactive T cells (Dijkstra et al., 2018Dijkstra K.K. Cattaneo C.M. Weeber F. Chalabi M. van de Haar J. Fanchi L.F. Slagter M. van der Velden D.L. Kaing S. Kelderman S. Generation of tumor-reactive T cells by co-culture of peripheral blood lymphocytes and tumor organoids.Cell. 2018; 174: 1586-1598.e12Abstract Full Text Full Text PDF PubMed Scopus (409) Google Scholar). In addition to organoid culture, patient-derived organotypic spheroids and minced primary tissue fragments embedded in collagen gels in air-liquid interface cultures have been developed to study interactions between tumor and local immune cells (Jenkins et al., 2018Jenkins R.W. Aref A.R. Lizotte P.H. Ivanova E. Stinson S. Zhou C.W. Bowden M. Deng J. Liu H. Miao D. et al.Ex vivo profiling of PD-1 blockade using organotypic tumor spheroids.Cancer Discov. 2018; 8: 196-215Crossref PubMed Scopus (249) Google Scholar; Neal et al., 2018Neal J.T. Li X. Zhu J. Giangarra V. Grzeskowiak C.L. Ju J. Liu I.H. Chiou S.H. Salahudeen A.A. Smith A.R. et al.Organoid modeling of the tumor immune microenvironment.Cell. 2018; 175: 1972-1988.e16Abstract Full Text Full Text PDF PubMed Scopus (527) Google Scholar).In the current study, we report the development of an automatic microfluidics droplet platform that can generate patient-derived micro-organospheres (MOS) in a high-throughput fashion to facilitate drug screening and personalized medicine treatment. A clinical study of eight metastatic colorectal cancer (CRC) patients showed that MOS assay readout within 14 days (10 days on average) correlates with later clinical outcomes. Furthermore, MOS preserved stromal cells of the original tumor tissue and potency of immune cells, providing a powerful tool for testing IO therapy.ResultsMOS generation and establishmentTo establish a precision medicine pipeline that can be used to guide patient care, we developed droplet-based microfluidics technology to rapidly generate PDMC in a reliable manner (Figure 1A). The core principle involves adding suspended cells from primary tissue to a 3D-extracellular matrix (Matrigel) followed by mixing with a biphasic liquid (oil) to generate microfluidic-based droplet MOS. The generated MOS are demulsified to remove excess oil and then cultured as suspension droplets.The basis of our pipeline is a benchtop machine for the generation of MOS (Figures 1B and S1A; Table S1; Video S1). Important design features of our device include reservoirs for loading both the oil and sample phases directly onto a custom microfluidic chip followed by positioning of the sample outlet on the backside of the chip for direct dispensing into a MOS recovery vessel. Attached pressure sources (e.g., Fluigent FlowEZ) are used to control the flow of oil and sample fluids into the custom microfluidic chip through tubing connected via a clamped manifold. The sample and oil meet at a “T” junction (Figure 1B) where the sample is “pinched” into droplets by the oil phase as it enters a collection channel. The system is compatible with the temperature-sensitive Matrigel. Both the 4°C sample and 37°C collection blocks are integrated into the device, which allows the Matrigel to flow through microfluidic channels and then quickly solidify at higher temperatures. The channel and chamber heights were engineered to generate MOS that on average were 250–450 μm in diameter, as these dimensions provide a 3D environment that is well suited for a variety of cell numbers and sizes. The device can generate MOS from as few as 15,000 cells from 18-gauge core biopsies, a sample size typically too small for reliable generation of conventional organoids for therapeutic profiling within the clinical time constraint.eyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI5MGNkNmYzZGFhOWZlN2YxMTY1ZDEwZTU2ODIwMTI4NiIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjcwMzQ2MTk0fQ.n4iU1tZkG0J9lATDiPSovccNA_kUReMYW5LKM85JzrwNT_1xsSAKDochIkbePSn1RsLDmjWzTh0xNpvBbp-mLVzvSmJDVA1cO7ltxbawsoGRLnF79WIUMBIA6L0vg-QG1coybe7qfrPYiPW5t09BFG845lu0DBEGjW50wGbu4YDiM9Le7v2Zw9sTz7FJxxIpXbxDmWEzHCJm6npXoDLGEJ4VVuJcUvdjI5dRPo2lbzDt-DF8g7tT9cyAGfzVP4_12fwddKacEwpsIVwqXid7yyIMClh7cTLghRvmLwDhg70j8BxuyPQusMxve4fBpyTGZ-p3D0V1rlr2kG_o4Tv8nA Download .mp4 (22.94 MB) Help with .mp4 files Video S1. MOS generation, related to Figure 1As a proof of concept, we first used our device to generate MOS from CRC PDX cells. We monitored CRC MOS growth at different seeding densities (20–100 cells per droplet) and discovered that MOS establish tumorsphere-like structures (Figure 1C). The number and size of tumorspheres increased with the seeding density per droplet. MOS were then generated from clinical CRC biopsies (Figure 1D) and shown to have various morphologies (Figure 1E). The number of MOS are determined by the number of viable cells divided by the number of cells per droplet.MOSs rapidly assess patient drug response in a prospective clinical studySince clinical treatment decisions are often made within 10–14 days of diagnosis, the ideal diagnostic assay should give results within 14 days and use minimal tissue (i.e., core biopsies) to predict clinical outcome. As a proof-of-concept study, we obtained a biopsy from a patient who presented with metastatic rectal cancer and established MOS (30 tumor cells per MOS) within 8 days of biopsy. We performed an in vitro high-throughput drug screen by treating the MOS with the Approved Oncology Set VI panel (provided by the NCI Developmental Therapeutics Program), which contained 119 different Food and Drug Administration (FDA)-approved small molecule inhibitors at 1 μM concentrations, and then analyzed treatment responses. The MOS were sensitive to oxaliplatin (% killing > 50%) and resistant to irinotecan (% killing < 50%) (Figure 1F). The entire process was performed within 11 days of biopsy acquisition. Consistent with the MOS drug response, the patient’s tumor still responded to oxaliplatin-based therapy 6 months later (Figure 1G).We next designed and conducted a prospective clinical study, wherein we obtained 18-gauge core biopsies from seven additional patients presenting with metastatic CRC, generated MOS, and performed drug testing (Figures 1H and 1I). Patient demographic information and mutation status from these two protocols are shown in Table 1. We generated MOS (30 tumor cells per MOS) and tested responses to oxaliplatin within 13 days (9.9 days on average) from time of biopsy for all eight biopsy samples with a success rate of 100% (8/8) (Table 2). Given the limited tissue volume, dosages such as 1 and 10 μM were selected based on previous literature (Vlachogiannis et al., 2018Vlachogiannis G. Hedayat S. Vatsiou A. Jamin Y. Fernández-Mateos J. Khan K. Lampis A. Eason K. Huntingford I. Burke R. et al.Patient-derived organoids model treatment response of metastatic gastrointestinal cancers.Science. 2018; 359: 920-926Crossref PubMed Scopus (795) Google Scholar; Ooft et al., 2019Ooft S.N. Weeber F. Dijkstra K.K. McLean C.M. Kaing S. van Werkhoven E. Schipper L. Hoes L. Vis D.J. van de Haar J. et al.Patient-derived organoids can predict response to chemotherapy in metastatic colorectal cancer patients.Sci. Transl. Med. 2019; 11: eaay2574Crossref PubMed Scopus (257) Google Scholar; Ganesh et al., 2019Ganesh K. Wu C. O'Rourke K.P. Szeglin B.C. Zheng Y. Sauvé C.G. Adileh M. Wasserman I. Marco M.R. Kim A.S. et al.A rectal cancer organoid platform to study individual responses to chemoradiation.Nat. Med. 2019; 25: 1607-1614Crossref PubMed Scopus (192) Google Scholar; Yao et al., 2020Yao Y. Xu X. Yang L. Zhu J. Wan J. Shen L. Xia F. Fu G. Deng Y. Pan M. et al.Patient-derived organoids predict chemoradiation responses of locally advanced rectal cancer.Cell Stem Cell. 2020; 26: 17-26.e6Abstract Full Text Full Text PDF PubMed Scopus (223) Google Scholar). We used the same cutoff as measured via Cell Titer Glo. Among the eight patients, four had oxaliplatin-sensitive MOS and four had oxaliplatin-resistant MOS (Figure 1J).Table 1CRC patient demographics and clinical diagnosisIDAgeGenderRaceHistologyGradePrimary siteMetastatic siteMolecular profilingCRC-MOS-00187FCaucasianadenocarcinomamoderately differentiatedrectallungMSSCRC-MOS-00268MCaucasianadenocarcinomamoderately differentiatedcolonliverMSS, TMB 1 Muts/Mb, KRAS (G12V), APC (L674fs), p53 (I254S)CRC-MOS-00371MCaucasianadenocarcinomamoderately differentiatedcolonliverMSS, TMB 3 Mut/Mb, KRAS (G12V), APC (Q1303), SMAD4 loss, TP53 (L257Q)CRC-MOS-00462FCaucasianadenocarcinomamoderately differentiatedcolonliverMSS, TMB 3 Mut/Mb, KRAS WT APC (R1450), ATM (R805), PIK3CA (E545K)CRC-MOS-00573MCaucasianadenocarcinomamoderately differentiatedcolonliverMSS, TMB 4 Mut/Mb, KRAS (G12D), APC (R876), TP53 (R158fs), PIK3CA (E545Q)CRC-MOS-00631FAsianadenocarcinomapoorly differentiatedrectalpelvisMSS, TMB (4 Muts/Mb) KRAS (G12D), TP53 (R175H)CRC-MOS-00737MCaucasianadenocarcinoma–colonliverMSS, TMB (1 Mut/Mb), KRAS (G12D), APC (R216), TP53 (G226E)CRC-MOS-00868MCaucasianadenocarcinomamoderately differentiatedcolonliverMSS, TMB (6 Mut/Mb) KRAS WT, APC (Q1367), EGFR amplification TP53 (R248W) Open table in a new tab Table 2MOS correlation to CRC patient outcomePatient IDMOS predictionClinical outcomeResponse timeDrug screen001Sensitiveresponse46 weaks8 days002Resistanceresponse28 weaks9 days003Sensitiveresponse38 weaks12 days004Resistanceno response8 weaks8 days005Sensitiveresponse36 weaks8 days006Resistanceno response3 weaks10 days007Sensitiveno response24 weaks11 days008Resistanceno response6 weaks13 days Open table in a new tab All eight patients received oxaliplatin-based therapy per usual treatment guidelines. Patient outcomes were subsequently evaluated by CT scan before and after oxaliplatin treatment (Figures S1B–S1H), and we compared time on treatment with MOS oxaliplatin sensitivity. The four patients whose MOS were sensitive to oxaliplatin all responded clinically and stayed on treatment past 20 weeks (and three of four still remained on treatment close to 40 weeks), whereas three of the four patients with resistant MOS did not respond to oxaliplatin treatment and were taken off treatment within 10 weeks (Figure 1J). The remaining patient (Case #1, ID #002) from the resistant MOS group on initial CT scan had a mixed response to therapy, but given clinical benefit, the patient was continued on therapy. Subsequent CT scan showed response to therapy, and the patient continue to remain on treatment until 28 weeks when liver section was performed to remove the metastatic lesion (Figure S1B; Table 2).This proof-of-concept clinical study suggests that MOS can be reliably generated from 18-gauge biopsies of metastatic CRC tumors and then be used to test sensitivity to frontline chemotherapy within 14 days. Our initial results demonstrate that this workflow largely correlate with patient outcomes, albeit larger trials are needed to further validate clinical applicability.We further measured cell death in each MOS by imaging the caspase-3/-7 fluorescence signal and normalizing it by the cell surface area inside each MOS. Treatment of two available CRC MOS lines (20 cells per MOS) that are resistant to oxaliplatin showed that only the highest dosages induced significant cell killing with heterogeneity among different MOS (Figures S2A and S2B).Tumor stromal and immune cells in MOSAs the tumor microenvironment, in particular immune components, has been shown to affect cancer therapy, we sought to characterize the stromal components of patient-derived MOS. We focused on lung tumor due to its response to immunotherapy but also characterized renal, breast, CRC, and ovarian tumors to lesser degrees. We generated MOS at a density of 30 tumor cells per MOS in 70% Matrigel diluted in culture medium and concurrently established bulk organoids using the same density of cells for comparison. Representative pictures of MOS generated from each tumor type as well as hematoxylin and eosin (H&E) from each tumor tissue and MOS are shown in Figures 2A, S2C, and S2D. Formation and growth of MOS and bulk organoids at days 2, 5, and 7 were comparable (Figure S2E).Figure 2Genomic and transcriptomic characterization of MOS generated from patient lung tumorShow full caption(A) Representative images of generated MOS from patient lung tumor tissue and H&E staining of the primary lung tumor tissue and derived MOS. Scale bar: 100 μm.(B) High-throughput drug screen demonstrates feasibility of using lung MOS to identify other targets in cancer therapy.(C) Copy-number variation (CNV) profiles with correlations of lung tumor tissue and derived MOS.(D) Driver mutations in commonly mutated genes for lung cancer is largely preserved in MOS compared with respective original tissues. Gray, driver mutations present; white, driver mutations absent.(E) UMAP of cells from primary lung tumor tissue or derived MOS labeled by cell types.(F) Comparison of log-transformed relative abundance of each cell type for lung tumor samples and derived MOS.(G) Relative abundance of cell types represented in either tissue (n = 3) or MOS (n = 3) samples. Abundances reported as log1p(percentage out of 1).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Overgrowth of fibroblasts is often a challenge for establishing organoids from clinical samples of certain cancer types. We compared the number of fibroblasts in MOS and bulk organoid cultures between days 7 and 9. Fewer fibroblasts were observed in MOS compared with bulk organoid cultures (Figure S2F), as confirmed by flow cytometry analysis of Vimentin expression (Figure S3A). We then performed rapid, high-throughput chemotherapeutic drug screening on MOS generated from lung, ovarian, and kidney cancer patients and measured sensitivities to commonly used agents in the treatment of these cancers (Figures 2B and S3B).We compared whole-exome sequencing of MOS to the matched original tumor specimen to determine if genomic alterations were maintained (Table S2). First, we characterized copy-number variations (CNVs). A similar pattern of amplifications and losses was seen in MOS and original tissue from lung cancer (Figure 2C) and other cancer types (Figure S3C). Second, we characterized somatic mutations in the genomes of matched MOS and original tumor samples. For each cancer type, mutation profiles for matched tissue specimen and MOS were highly correlated, whereas unmatched samples were not (Figure S3D). Variants were common between tissue specimens and matched MOS (Figure S3E). Driver mutations were largely consistent between tissue specimens and MOS among commonly affected genes for each cancer type (Figures 2D and S3F), with a sensitivity (mutations detected in tumor tissue also detected in MOS) of 85% ± 0.7% and a specificity (mutations absent in tumor tissue also absent in MOS) of 95% ± 0.3% (Figure S3G). These findings suggest that MOS largely capture the genomic profiles of patient tumors from which they are derived.To compare the tumor and stromal cell types between tissue and derived MOS, we performed single-cell RNA sequencing (RNA-seq) (Macosko et al., 2015Macosko E.Z. Basu A. Satija R. Nemesh J. Shekhar K. Goldman M. Tirosh I. Bialas A.R. Kamitaki N. Martersteck E.M. et al.Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets.Cell. 2015; 161: 1202-1214Abstract Full Text Full Text PDF PubMed Scopus (3473) Google Scholar) on six pairs of matched patient tumor specimens (three lung cancers, one kidne