Extending the success of cellular immunotherapies against blood cancers to the realm of solid tumors will require improved in vitro models that reveal therapeutic modes of action at the molecular level. Here we describe a system, called BEHAV3D, developed to study the dynamic interactions of immune cells and patient cancer organoids by means of imaging and transcriptomics. We apply BEHAV3D to live-track >150,000 engineered T cells cultured with patient-derived, solid-tumor organoids, identifying a ‘super engager’ behavioral cluster comprising T cells with potent serial killing capacity. Among other T cell concepts we also study cancer metabolome-sensing engineered T cells (TEGs) and detect behavior-specific gene signatures that include a group of 27 genes with no previously described T cell function that are expressed by super engager killer TEGs. We further show that type I interferon can prime resistant organoids for TEG-mediated killing. BEHAV3D is a promising tool for the characterization of behavioral-phenotypic heterogeneity of cellular immunotherapies and may support the optimization of personalized solid-tumor-targeting cell therapies.

Wnt signaling pathways are transmitted via 10 homologous frizzled receptors (FZD1-10) in humans. Reagents broadly inhibiting Wnt signaling pathways reduce growth and metastasis of many tumors, but their therapeutic development has been hampered by the side effect. Inhibitors targeting specific Wnt-FZD pair(s) enriched in cancer cells may reduce side effect, but the therapeutic effect of narrow-spectrum Wnt-FZD inhibitors remains to be established in vivo. Here, we developed a fragment of C. difficile toxin B (TcdBFBD), which recognizes and inhibits a subclass of FZDs, FZD1/2/7, and examined whether targeting this FZD subgroup may offer therapeutic benefits for treating breast cancer models in mice. Utilizing 2 basal-like and 1 luminal-like breast cancer models, we found that TcdBFBD reduces tumor-initiating cells and attenuates growth of basal-like mammary tumor organoids and xenografted tumors, without damaging Wnt-sensitive tissues such as bones in vivo. Furthermore, FZD1/2/7-positive cells are enriched in chemotherapy-resistant cells in both basal-like and luminal mammary tumors treated with cisplatin, and TcdBFBD synergizes strongly with cisplatin in inhibiting both tumor types. These data demonstrate the therapeutic value of narrow-spectrum Wnt signaling inhibitor in treating breast cancers.

SummaryCellular immunotherapies are rapidly gaining clinical importance, yet predictive platforms for modeling their mode of action are lacking. Here, we developed a dynamic immuno-organoid 3D imaging-transcriptomics platform; BEHAV3D, to unravel the behavioral and underlying molecular mechanisms of solid tumor targeting. Applied to an emerging cancer metabolome-sensing immunotherapy: TEGs, we first demonstrate targeting of multiple breast cancer subtypes. Live-tracking of over 120,000 TEGs revealed a diverse behavioral landscape and identified a ‘super engager’ cluster with serial killing capability. Inference of single-cell behavior with transcriptomics identified the gene signature of ‘super engager’ killer TEGs, which contained 27 genes with no previously described T cell function. Furthermore, guided by a dynamic type 1 interferon (IFN-I) signaling module induced by high TEG-sensitive organoids, we show that IFN-I can prime resistant organoids for TEG-mediated killing. Thus, BEHAV3D characterizes behavioral-phenotypic heterogeneity of cellular immunotherapies and holds promise for improving solid tumor-targeting in a patient-specific manner.

Organoid technology has revolutionized the study of human organ development, disease and therapy response tailored to the individual. Although detailed protocols are available for the generation and long-term propagation of human organoids from various organs, such methods are lacking for breast tissue. Here we provide an optimized, highly versatile protocol for long-term culture of organoids derived from either normal human breast tissues or breast cancer (BC) tissues, as well as culturing conditions for a panel of 45 biobanked samples, including BC organoids covering all major disease subtypes (triple-negative, estrogen receptor-positive/progesterone receptor-positive and human epidermal growth receptor 2-positive). Additionally, we provide methods for genetic manipulation by Lipofectamine 2000, electroporation or lentivirus and subsequent organoid selection and clonal culture. Finally, we introduce an optimized method for orthotopic organoid transplantation in mice, which includes injection of organoids and estrogen pellets without the need for surgery. Organoid derivation from tissue fragments until the first split takes 7-21 d; generation of genetically manipulated clonal organoid cultures takes 14-21 d; and organoid expansion for xenotransplantation takes >4 weeks.

High-grade serous ovarian cancer (HG-SOC)—often referred to as a “silent killer”—is the most lethal gynecological malignancy. The fallopian tube (murine oviduct) and ovarian surface epithelium (OSE) are considered the main candidate tissues of origin of this cancer. However, the relative contribution of each tissue to HG-SOC is not yet clear. Here, we establish organoid-based tumor progression models of HG-SOC from murine oviductal and OSE tissues. We use CRISPR-Cas9 genome editing to introduce mutations into genes commonly found mutated in HG-SOC, such as Trp53 , Brca1 , Nf1 and Pten . Our results support the dual origin hypothesis of HG-SOC, as we demonstrate that both epithelia can give rise to ovarian tumors with high-grade pathology. However, the mutated oviductal organoids expand much faster in vitro and more readily form malignant tumors upon transplantation. Furthermore, in vitro drug testing reveals distinct lineage-dependent sensitivities to the common drugs used to treat HG-SOC in patients.

Microscopy analysis of tumour samples is commonly performed on fixed, thinly sectioned and protein-labelled tissues. However, these examinations do not reveal the intricate three-dimensional structures of tumours, nor enable the detection of aberrant transcripts. Here, we report a method, which we name DIIFCO (for diagnosing in situ immunofluorescence-labelled cleared oncosamples), for the multimodal volumetric imaging of RNAs and proteins in intact tumour volumes and organoids. We used DIIFCO to spatially profile the expression of diverse coding RNAs and non-coding RNAs at the single-cell resolution in a variety of cancer tissues. Quantitative single-cell analysis revealed spatial niches of cancer stem-like cells, and showed that the niches were present at a higher density in triple-negative breast cancer tissue. The improved molecular phenotyping and histopathological diagnosis of cancers may lead to new insights into the biology of tumours of patients.

There remains an unmet need for preclinical models to enable personalized therapy for ovarian cancer (OC) patients. Here we evaluate the capacity of patient-derived organoids (PDOs) to predict clinical drug response and functional consequences of tumor heterogeneity. We included 36 whole-genome-characterized PDOs from 23 OC patients with known clinical histories. OC PDOs maintain the genomic features of the original tumor lesion and recapitulate patient response to neoadjuvant carboplatin/paclitaxel combination treatment. PDOs display inter- and intrapatient drug response heterogeneity to chemotherapy and targeted drugs, which can be partially explained by genetic aberrations. PDO drug screening identifies high responsiveness to at least one drug for 88% of patients. PDOs are valuable preclinical models that can provide insights into drug response for individual patients with OC, complementary to genetic testing. Generating PDOs of multiple tumor locations can improve clinical decision making and increase our knowledge of genetic and drug response heterogeneity.

BRCA1 germ-line mutations are a major cause of hereditary breast cancer and BRCA1-deficient breast cancer shares many characteristics as sporadic basal-like breast cancer (BLBC). Effective therapeutic targets for BRCA1-deficient BLBC remain lacking. By utilizing a BRCA1-deficient BLBC mouse model based on intraductal injection of Krt8-Cre adenovirus to inactivate Brca1 and Trp53 in luminal mammary epithelial cells, here we report that the Wnt receptor Frizzled 7 (FZD7) serves as a biomarker and therapeutic target in the resulting mammary tumor cells and is particularly enriched in cancer stem cells / tumor-initiating cells (CSCs/TICs). Inhibiting FZD7-mediated Wnt signaling using a nontoxic FZD-binding fragment of C. difficile toxin B (TcdBFBD) attenuates growth of BRCA1-deficient tumor organoids and xenografted tumors, without damaging Wnt-sensitive tissues such as bones in vivo. Finally, FZD1/2/7-positive cells are enriched in chemotherapy-resistant cells in both BLBC and luminal breast tumors treated with cisplatin, and TcdBFBD synergizes strongly with cisplatin in inhibiting both tumor types. These findings demonstrate the therapeutic value for targeting FZD1/2/7 in treating breast cancers and establish TcdBFBD as a potential therapeutic agent targeting TICs and chemotherapy-resistant cancer cells.

Abstract Over the last few years several studies led to a significant change in the field of ovarian cancer and it is currently believed that the fallopian tube and not the ovary surface epithelium (OSE) is the main origin of high-grade serous ovarian cancer (HG-SOC). Nevertheless, due to the lack of unique markers and adequate model systems the relative contribution of each tissue is not yet clear and the notion that OSE has a role in HG-SOC development was not cast aside altogether. In this work we have established novel organoid systems derived from both mouse OSE and oviduct (Ovi, the equivalent of human fallopian tube). These systems recapitulate their tissue of origin and demonstrate differences in medium requirements as well as gene expression. To establish comparable tumor progression models for both OSE and Ovi we used CRISPR-Cas9 technique and targeted commonly mutated genes in ovarian cancer (Trp53, Brca1, Nf1 and Pten). Thus, we were able to establish clones with different combinations of mutations. Histological, metaphase spread and gene expression analysis of the mutated organoid clones from both OSE and Ovi demonstrated different degrees of deviation from their wild type counterpart. This deviation became more evident as the amount of introduced mutations increased. Preliminary transplantation experiments showed that only triple mutants gave rise to tumors (Trp53, Brca1 and Pten or Trp53, Brca1 and Nf1). Moreover, Ovi mutant clones were more prone to grow into tumors in comparison to OSE as no tumors have been detected in mice transplanted with the OSE equivalent clones so far. Taken together, in this study we present the first comparable Ovi/OSE research platform that enables addressing questions related to origin and early stages of HG-SOC development. Citation Format: Kadi Lõhmussaar, Oded Kopper, Hans C. Clevers. BUILDING AN ORGANOID-BASED MODEL FOR OVARIAN CANCER [abstract]. In: Proceedings of the 12th Biennial Ovarian Cancer Research Symposium; Sep 13-15, 2018; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2019;25(22 Suppl):Abstract nr GMM-039.

Introduction/Background Epithelial ovarian cancer is characterized by intratumour genetic heterogeneity, chemotherapy resistance and poor survival. The advent of ovarian cancer organoids, that recapitulate histological and genetic features of the original lesions, has providedopportunities for personalized assessment of drug response. Here, we assessed if the ovarian cancer organoid system can predict patients9 clinical response to chemotherapeutics. In addition, we evaluated the link between intratumour genetic heterogeneity and drug response using multiple organoids derived from individual patients. Methodology In total, we included 34 organoid lines derived from 22 patients with ovarian cancer at primary surgery or interval debulking. Organoid drug responses to chemotherapeutics as well as targeted drugs were determined. The IC50 and area under the curve values were computed from the drug response curves. Patients9 recurrence-free and overall survival were compared to the organoid drug response. For patients who received carboplatin and paclitaxel chemotherapy prior to interval debulking (N=7), we additionally compared organoid drug response to biochemical (CA125), pathological, radiological outcomes. The organoids and corresponding tumour tissues were subjected to whole genome sequencing for genomic analysis. Results Organoid drug response correlated to clinical response as defined by histopathological tumour assessment (chemotherapy response score (CRS)). Organoids of patients with no or minimal pathological response (CRS=1) were less sensitive to carboplatin and paclitaxelcompared to organoids of the patients with appreciable pathological response (CRS=2). The presumed additive effect of paclitaxel on carboplatin was only observed in a subset of organoids. Organoid exposure to eight additional drugs revealed that multiple tumour locations of the same patients exhibited varying responses to specific drugs. Genomic analyses are being performed to identify underlying mechanisms of differential drug responses. Conclusion Ovarian cancer organoids provide a valuable preclinical system with the potential to guide treatment choice through prediction of drug response, avoid unnecessary overtreatment and provide insights in response mechanisms. Disclosure No competing interests. Applicable funding sources: EIF | Stand Up To Cancer (SU2C) [Clevers, Kopper] KWF Kankerbestrijding (Dutch Cancer Society) - UU2015-7743 [de Witte, Witteveen, Zweemer, Kloosterman] Gieskes Strijbis Foundation (1816199) [Clevers, Witteveen, Kloosterman, Zweemer, Lõhmussaar, Espejo Valle-Inclan, Hami, Snippert]

Ovarian cancer (OC) is a heterogeneous disease usually diagnosed at a late stage. Experimental in vitro models that faithfully capture the hallmarks and tumor heterogeneity of OC are limited and hard to establish. We present a protocol that enables efficient derivation and long-term expansion of OC organoids. Utilizing this protocol, we have established 56 organoid lines from 32 patients, representing all main subtypes of OC. OC organoids recapitulate histological and genomic features of the pertinent lesion from which they were derived, illustrating intra-and interpatient heterogeneity, and can be genetically modified. We show that OC organoids can be used for drug-screening assays and capture different tumor subtype responses to the gold standard platinum-based chemotherapy, including acquisition of chemoresistance in recurrent disease. Finally, OC organoids can be xenografted, enabling in vivo drug-sensitivity assays. Taken together, this demonstrates their potential application for research and personalized medicine.

Breast cancer (BC) comprises multiple distinct subtypes that differ genetically, pathologically, and clinically. Here, we describe a robust protocol for long-term culturing of human mammary epithelial organoids. Using this protocol, >100 primary and metastatic BC organoid lines were generated, broadly recapitulating the diversity of the disease. BC organoid morphologies typically matched the histopathology, hormone receptor status, and HER2 status of the original tumor. DNA copy number variations as well as sequence changes were consistent within tumor-organoid pairs and largely retained even after extended passaging. BC organoids furthermore populated all major gene-expression-based classification groups and allowed in vitro drug screens that were consistent with in vivo xeno-transplantations and patient response. This study describes a representative collection of well-characterized BC organoids available for cancer research and drug development, as well as a strategy to assess in vitro drug response in a personalized fashion.