Cancer immunotherapy has received intense attention over the past two decades owing to the promise of delivering durable cures by harnessing the cytotoxic potential of the immune system against tumor cells (Waldman et al., 2020; Yang, 2015; Gong et al., 2018). However, although impressive improvement in long-term survival has been reported (Hodi et al., 2010; Schadendorf et al., 2015; Wolchok et al., 2017), only a fraction of patients responds. Furthermore, the systemic immunomodulation mediated by these drugs often elicits immune-related adverse events (irAEs), including skin and liver toxicity, colitis, and pneumonitis, limiting their broad clinical application in battling cancer (Champiat et al., 2017; Naidoo et al., 2015; Kennedy and Salama, 2020).

T-cell engaging bispecific antibodies (TCBs) are a novel class of cancer immunotherapeutic agents that have the potential to improve on the clinical efficacy and safety of traditional immunotherapy (Clynes and Desjarlais, 2019; Labrijn et al., 2019). TCBs exert their anti-tumor activity by simultaneously binding to a cancer surface antigen and the CD3 T-cell receptor, thereby both activating the latter and physically crosslinking it to target cells (Bacac et al., 2016a). This synthetic immunity approach is particularly favorable for targeting less immunogenic, neo-antigen-lacking tumors, as T cells can be recruited and activated independently of their T-cell receptor specificity. This strictly tumor-targeted immunomodulation is also expected to reduce the systemic inflammatory toxicities associated with traditional immunotherapies. The therapeutic potential of TCBs is exemplified by the large number of molecules targeting solid and blood tumors, which are currently in various stages of clinical evaluation (Ishiguro et al., 2017; Goebeler and Bargou, 2020).

Although TCBs hold the promise for a safer therapeutic option, they are not risk free. The antigens targeted are rarely exclusive to the tumor, but are also often expressed, albeit at lower levels, in normal tissues, rendering TCBs subject to ‘on-target, off-tumor’ safety liabilities. This is particularly true for epithelial tumor antigens as they are frequently targeted in solid tumor indications. For example, a Bispecific T-cell Engager (BiTE) targeted to the epidermal growth factor receptor (EGFR) produced severe liver and kidney toxicities in non-human primates, in line with EGFR expression in these organs, and led to the termination of the animals (Klinger et al., 2016; Lutterbuese et al., 2010). Clinical adverse events were reported in a recent Phase I study evaluating an epithelial cell adhesion molecule (EpCAM)-targeted BiTE as a therapy for a variety of epithelial carcinomas. Consistent with the expression of EpCAM in the gastrointestinal tract, the molecule triggered severe diarrhea and ultimately prevented escalation to efficacious doses and the identification of a therapeutic window (Kebenko et al., 2018; Trabolsi et al., 2019). Reliable human TCB safety evaluations at the preclinical stage are therefore of vital importance to ensure that well-tolerated and efficacious therapeutics reach patients.

Traditional rodent-based preclinical models are often ill-suited for predicting some cancer immunotherapy-mediated adverse events in humans in part because of the fundamental differences in the immunological responses between the species (Bjornson-Hooper et al., 2019). In the EpCAM example mentioned above, the severity of the diarrhea elicited by the treatment was not predicted by preclinical studies in mice (Brischwein et al., 2006). Moreover, an increasing number of TCBs target human-specific antigens that lack expression in animals, rendering preclinical animal studies uninformative for safety and efficacy assessments (Bacac et al., 2016a). Indeed, the development of preclinical models that better translate to human immunity is regarded as one of the top current challenges of cancer immunotherapy (Hegde and Chen, 2020).

While human-relevant cell-based models of tissues and organs promise to bridge this gap, conventional in vitro two-dimensional (2D) models fail to provide the complexity required to model the biological mechanisms of immunotherapeutic effects. Furthermore, their reductive microenvironment, devoid of essential cellular, biochemical, and biophysical factors found in the native organ, precludes the expression of TCB targets at physiologically relevant levels and patterns, crucial for capturing TCB pharmacology and safety liabilities.

Organ-on-Chip models aim to overcome these limitations by combining micro-engineering with cultured primary human cells to recreate the complex multifactorial microenvironment and functions of native tissues and organs (Huh et al., 2010). The tissue microenvironment in vivo provides the external signals that help drive cellular differentiation toward mature phenotypes. Organs-on-Chips model key functional aspects of tissue-level physiology such as epithelial and microvascular tissue-tissue interfaces, and physiologically relevant mechanical forces, have been shown to more accurately capture in vivo-relevant phenotypes (Kasendra et al., 2020; Kasendra et al., 2018; Gayer and Basson, 2009). The enhanced tissue maturation promoted by Organs-on-Chips could help ensure organ-specific expression of TCB targets, while the modularity of these devices and the possibility for controlled circulation of molecules and immune cells could better capture the functional interactions between TCBs, immune cells, and target-expressing cells that occur in patients. Motivated by these advantages, we set out to evaluate Organs-on-Chips as platforms for the assessment of on-target, off-tumor TCB safety risks in human organs, using a panel of targeting and non-targeting molecules, and leveraging in vivo target expression and toxicity data. We found that these systems could reproduce and predict target-dependent TCB safety liabilities, showing sensitivity to key determinants thereof, such as target expression and antibody affinity.

As a starting point for our method development and validation, we used molecules under current preclinical development. We focused on a T-cell bispecific antibody generated to bring Folate Receptor 1 (FOLR1) expressing tumor cells in close proximity to CD3 expressing cytotoxic T-cells (Figure 1—figure supplement 1A; Geiger et al., 2020). FOLR1 is overexpressed in many solid, epithelial-derived tumors including ovarian, lung, and breast cancer (Scaranti et al., 2020), but is also expressed to a lower degree on normal epithelial cells as found in the lung and kidneys (Parker et al., 2005). While a high-affinity FOLR1-TCB (FOLR1(Hi) TCB) was efficacious in human breast cancer patient-derived xenograft models (Figure 1—figure supplement 1B), severe on-target toxicity in the lung of cynomolgus monkey was observed (AMea, 2016). Clinical signs of severe respiratory inflammation appeared as early as 24 hr post-dosing, and pro-inflammatory cytokines IL-6, IL-2, and IL-8 were elevated in the blood of affected animals, and importantly coincided with an increase of inflammation markers. Histopathology assessment revealed leukocytic infiltrates in lung tissue indicative of immune-mediated toxicity (Figure 1A). Further immunohistochemistry (IHC) studies (Figure 1B–D) indicated low relative expression (compared to high FOLR1 expression in the ovarian carcinoma cell line, HeLa) of the FOLR1 target antigen in lung alveolar epithelial cells of cynomolgus lung tissue suggesting an adverse event largely driven by on-target toxicity. Importantly, IHC analysis of FOLR1 expression in the human lung revealed similar levels and patterns of the antigen as in the cynomolgus (Figure 1E,F), extending the threat of safety liabilities to patients in the clinical setting. Bearing in mind the toxicological profile of FOLR1(Hi) TCB in cynomolgus and the expression of FOLR1 in both species lung tissue, we identified the lung as an at-risk organ in patients and accordingly set out to evaluate a human Alveolus Lung-Chip model as a platform for FOLR1(Hi) TCB toxicology assessment.

Figure 1 with 1 supplement see all

Download asset Open asset

Alveolus Lung-Chips were seeded with human adult lung primary alveolar epithelial cells on top of an extra-cellular matrix-coated porous membrane that separates two parallel, fluidic microchannels (Figure 1G, Figure 2—figure supplement 1). On the opposite side of the membrane, human primary lung microvascular cells were seeded to form a lower tubular vascular channel as described previously (Jain et al., 2018). A mature Alveolus Lung-Chip model was obtained after 5 days of liquid–liquid culture (LLI) followed by establishment of an air–liquid–culture (ALI) for a further 5 days. To evaluate FOLR1 expression in chips, we combined RNA sequencing, immunofluorescence and flow cytometry analyses. FOLR1 gene expression was even over time as shown by quantification of RNA transcripts (Figure 1H). We also confirmed FOLR1 protein expression in mature chips (Figure 1I). Flow cytometry-mediated quantification allowed us to estimate the cell surface expression of FOLR1 within the Alveolus Lung-Chip at an average of ~10,000 molecules expressed per cell (Figure 1J). For comparison, the high FOLR1 expressing ovarian carcinoma HeLa cell line displayed an average of ~450,000 molecules per cell when cultured on chip (Figure 2—figure supplement 1B), confirming the difference observed in IHC between healthy and tumor cells.

To render the device immunocompetent and capable of simulating on-target TCB-mediated immunomodulation, we added peripheral mononuclear blood cells (PBMC) isolated from human whole blood to the epithelial channel in direct contact with the mature alveolar epithelium (Figure 2—figure supplement 1). Introduction of T cells is required to engage the CD3 arm of FOLR1(Hi) TCB thereby allowing its mode of action. The sequence of events we aimed to reproduce in chips are the crosslinking of T cells to the FOLR1 expressing target cells mediated by the TCB, subsequent T-cell activation and cytolytic synapse formation resulting in cytotoxic granules release (granzymes and perforin) and consequent target cell apoptosis. Early T-cell cytokine release (TNFα, IFNγ) should be followed by later cytokine release from epithelial cells and monocytes (IL-6, IL1β, IL-8) combined with strong physical attachment of immune cells to the FOLR1-expressing lung epithelium via the TCB. Thus, we selected and optimized experimental readouts that would enable us to monitor these steps in the Alveolar Lung-Chip.

Figure 2A shows representative brightfield and fluorescent images of the Alveolus Lung-Chip at 48 hr after the administration of immune cells. Compared to the chips treated with a non-targeting (NT) TCB control antibody, FOLR1(Hi) TCB treated chips presented an increased apoptosis of the alveolar epithelium (Figure 2B). Consistently, FOLR1(Hi) TCB treatment led to increased T-cell activation (as evidenced by CD69 upregulation in CD8⁺ T cells) in the presence of target-expressing cells (Figure 2C), but not in PBMC only (Figure 2C, right panel). Supernatants collected from the outlet reservoir of the epithelial channels at 24 and 48 hr post-treatment were measured for multiplex cytokines (Figure 2D), revealing a significantly increased secretion of IFNγ at 24 hr and 48 hr that correlated with increased granzyme B and IL-6 at 48 hr in response to FOLR1(Hi) TCB.

Figure 2 with 2 supplements see all

Download asset Open asset

Interestingly, we noticed a higher number of immune cells in the FOLR1(Hi) TCB condition compared to the NT control, possibly due to a combination of T-cell proliferation and increased attachment of T cells (Figure 2A). Quantification of the immune cell presence in fixed chips confirmed increased attachment of both T cells and non CD3⁺ cells to the target epithelium in the FOLR1(Hi) TCB condition (Figure 2E,F), which is consistent with the TCB mode of action, whereby immune cells are crosslinked to target cells. Together, these data suggest that the Alveolus Lung-Chip successfully replicates aspects of the FOLR1 (Hi) TCB-mediated toxicity observed in cynomolgus, and suggests that the human lung would be subject to similar safety liabilities.

In light of the toxicity risk predicted above, and hoping to define a potential therapeutic window of FOLR1(Hi) TCB, we performed the same study with chips seeded with the high FOLR1 expressing ovarian carcinoma cell line, HeLa, previously used to assess drug efficacy (Geiger et al., 2020). Although no effects were seen at the lowest concentration, all concentrations starting from 2 ng/mL induced significant T-cell activation (measured at 48 hr), cancer cell apoptosis (from 24 hr onwards) and strong cytokine release, as expected from the high level of FOLR1 expression in HeLa cells (Figure 2—figure supplement 2). Thus, we efficiently killed tumor cells at a much lower concentration than that needed to induce damage to the healthy alveolar epithelial cells. Of note, FOLR1(Hi) TCB EC50 was estimated at 1.1 pM, which was close to the value obtained from standard 2D in vitro killing experiments (2.2 pM).

Although the data described above suggested that a therapeutic window for FOLR1(Hi) TCB could be determined, we leveraged the chip to instead identify a safer molecule. In particular, we utilized an antibody with lower monovalent affinity for the FOLR1 target, referred to as FOLR1(Lo) TCB making use of avidity mediated selectivity gain (Figure 1—figure supplement 1C,D). As a result of that design, FOLR1(Lo) TCB presented a lower binding to FOLR1-expressing HeLa cells while retaining a potent killing activity in coculture assays and an in vivo tumor control efficacy (Figure 1—figure supplement 1E–G). We profiled the two TCBs, FOLR1(Hi) and FOLR1 (Lo), in the immunocompetent Alveolus Lung-Chip following the workflow and readouts described above, and found that FOLR1(Hi) TCB induced a significant increase in all the readouts starting at the 0.2 μg/mL concentration whereas FOLR1(Lo) TCB showed a response only at the highest concentration of 20 μg/mL and to a much lower magnitude than the high-affinity molecule (Figure 3A–D, Figure 3—figure supplement 1). These results demonstrated that the cellular responses on the platform are sensitive to differences in antibody affinity and recapitulates the biology associated with the mode of action of TCBs. Following these in vitro observations, the lower affinity FOLR1 TCB was tested in a cynomolgus toxicology study and none of the animals experienced the lung inflammation observed with FOLR1(Hi) TCB (Figure 3E), confirming the safer profile predicted by the Alveolus Lung-Chip.

Figure 3 with 2 supplements see all

Download asset Open asset

To compare the chip format to its 2D equivalent, we benchmarked the Alveolus Lung-Chip against transwell inserts coated with alveolar epithelial cells and endothelial cells on opposing sides (Figure 3—figure supplement 2). In the transwell environment, quantification of TCB-dependent immune cell attachment and apoptosis did not show differences between the control and FOLR1 TCB treatment conditions. T-cell activation and T-cell-specific cytokines were elevated, possibly due to prolonged interactions of the immune cells with the epithelium and static media supply. Also, the transwell version did not capture differences between FOLR1 antibodies: FOLR1(Lo) TCB led to a higher amount of granzyme B and IFNγ release than the chips, which is inconsistent with the absence of toxicity observed in vivo. Given the higher concordance of the results produced by the Alveolar Lung-Chip compared with the transwell counterpart, we propose that the novel immunocompetent Alveolus Lung-Chip platform can faithfully evaluate TCB on-target, off-tumor risks and presents a superior value to the existing alternatives.

To demonstrate the broad applicability of the model for testing target-mediated TCB safety risks, we extended the methodology to a second target and a second Organ-Chip model. In this example, we focused on TCBs targeting carcinoembryonic antigen (CEA), which is overexpressed in a range of solid tumors, including colorectal cancer (Hammarström, 1999). We have created TCBs binding to CEA with high or low affinity – CEA(Hi) and CEA(Lo) TCB, respectively (Figure 4—figure supplement 1), and are currently evaluating them as therapies for a range of solid tumors. Indeed, we have found that CEA(Hi) and CEA(Lo) TCB are potent mediators of tumor cell lysis and T-cell activation in vitro (Figure 4—figure supplement 1B,C), and exhibit robust anti-tumor activity in humanized mice engrafted with CEA-expressing tumors (Figure 4—figure supplement 1C).

Aside from solid tumors, however, CEA is expressed in the gastrointestinal tract (Benchimol et al., 1989; Thomas et al., 1995; Zhou et al., 1993). Immunohistochemistry analysis of primary human intestinal samples confirmed high expression of CEA in the colon, whereas small intestinal expression was lower. In both tissues, CEA was enriched on the apical surface of the barrier (Figure 4A). The substantial target presence in the gastrointestinal tract implicates this system as an at-risk organ, motivating us to assess potential intestinal toxicities triggered by CEA-engaging TCBs. Our antibodies recognize a human-specific epitope within the CEA protein,, making mice and cynomolgus preclinical toxicology models unsuitable for the assessment of toxicities caused by CEA-targeting TCBs. Indeed, our antibodies showed lack of cross-reactivity to cynomolgus monkey CEA, which underscores the need for human-relevant models in addressing this question (Figure 4—figure supplement 1E). Therefore, we leveraged the recently developed and characterized Colon and Duodenum Intestine-Chips, which combine the two most advanced approaches in the field of intestinal modeling – primary human organoids and Organs-on-Chips (Kasendra et al., 2020; Kasendra et al., 2018). Briefly, primary human colon intestinal organoids are dissociated and seeded within an Organ-Chip (Figure 4—figure supplements 2 and 3A), where they form a tight, polarized barrier, containing the full range of mature intestinal cell types. We have previously shown that the inclusion of luminal flow, peristaltic motion, and an endothelial layer enhances the maturation and physiological fidelity of the barrier, compared with organoids in conventional 3D culture (Apostolou et al., 2020).

Figure 4 with 3 supplements see all

Download asset Open asset

To qualify the Intestine Chip as a platform for TCB safety assessment, we set out to determine whether it (1) supports physiologically relevant target expression and (2) can successfully capture target-mediated TCB toxicity. Immunofluorescence analysis revealed robust expression of CEA in the Colon Intestine-Chip epithelium, whereas expression in the Duodenum Intestine-Chip appeared weaker and localized to the apical surface (Figure 4B).

Flow cytometry-mediated quantification verified the cell surface expression of CEA within the Intestine-Chip and confirmed the higher target abundance in the Colon-Chip compared with the Duodenum-chip (Figure 4C), in line with the regional differences observed in vivo. Based on previous data suggesting that CEA-targeting TCBs trigger immune cell activation and cancer cell killing above a threshold of 10,000 CEA molecules (Bacac et al., 2016b), we expect to detect on-target TCB toxicity in the Intestine-Chip. To our surprise, colon and duodenum organoids cultured in the conventional 3D format featured target expression levels similar to those measured in the chips (Figure 4C). Nonetheless, despite the similarity in target abundance, we believe that the Intestine-Chip is more suitable for the assessment of on-target TCB toxicities, given the more physiologically accurate CEA expression patterns compared with those observed in organoids. Immunofluorescence analysis revealed the familiar apical enrichment of CEA in the Colon-Chip (strong) and Duodenum-chip (weak) (Figure 4—figure supplement 3B), whereas CEA appeared to be junctional and non-polarized in colon organoids. Indeed, the incorporation of microenvironmental cues, including flow and peristalsis, has shown to lead to enhanced maturation within the Intestine-Chip (Kasendra et al., 2020; Kasendra et al., 2018). Importantly, immunohistochemistry analysis of a commercial 2D model comprising primary intestinal epithelial cells showed patchy CEA expression, which appeared weaker than that observed in the native colon (Figure 4—figure supplement 3C). Consistently, we estimated an average of about 2000 CEA surface binding sites per cell in this system (Figure 4—figure supplement 3D), which is dramatically lower than the expression recorded in the Colon Intestine-Chip.

Next, we evaluated the Intestine-Chip for its ability to capture the toxic effects CEA(Hi) and CEA(Lo) TCB, and expected differences therein, owing to differential binding affinity. We focused first on the Colon Intestine-Chip, bearing in mind the higher levels of target in the native organ and the chip model. To render the Intestine-Chips capable of simulating an immune response, we took an approach analogous to that of the Lung-Chip: PBMC and TCBs were introduced using the top fluidic channel of the Colon Intestine-Chip, affording direct contact with the epithelium and enabling engagement with the target (Figure 4—figure supplement 2). Epithelial cell death, immune cell attachment, and activation were monitored as readouts of on-target TCB safety liabilities. Unlike the Alveolus Lung-Chip, TCB treatment did not lead to increased epithelial cell killing. Nonetheless, we did observe significant changes in all of the other readouts of TCB-mediated toxicity. Both CEA(Hi) and CEA(Lo) TCB induced dose-dependent increase in PBMC attachment (Figure 4D) and activation, as evidenced by CD69 upregulation (Figure 4E) and the release of pro-inflammatory cytokines, including granzyme B, IFNγ, and IL-6 (Figure 4F). As expected, CEA(Hi) TCB triggered higher PBMC attachment, activation and cytokine release compared with CEA(Lo) TCB, confirming that the model is sensitive to differences in antibody affinity, which would be expected to translate into differential toxicity outcomes in the clinic. Importantly, we observed no activation upon treatment with an antibody that engages the CD3 receptor on T cells but cannot bind to CEA (NT TCB). Likewise, the CEA-targeting TCBs failed to induce activation of PBMC only, in the absence of target tissue (Figure 4—figure supplement 3E). Together, these data confirm that the effects observed in the chips are target dependent and mode of action dependent, ruling out non-specific PBMC activation and cytokine release mediated by CD3 engagement only.

We then took advantage of the Duodenum Intestine-Chip, which faithfully mimics the lower CEA expression observed in the native small intestine (Figure 4B,C), to explore whether the system is sensitive to variations in target expression. Indeed, we observed differences in on-target toxicity governed by target abundance: CD69 expression and cytokine release induced by both TCBs were significantly attenuated in the Duodenum Intestine-Chip, in line with the lower target expression (Figure 4G–H). The reduction was more extensive in the case of CEA(Lo) TCB, which induced minimal increase in CD69 expression and no cytokine release, relative to the non-targeting TCB. CEA(Hi) TCB induced PBMC activation and cytokine release which, while lower than those observed in the Colon Intestine-Chip, were significantly elevated compared with the non-targeting control, suggesting that the high-affinity molecule may pose safety risks even in tissues with low target expression as described for the high-affinity FOLR1 molecule in the lung.

Here, we describe an Organs-on-Chips based approach for the assessment of TCB toxicity in the lung and the intestine. We demonstrate that the Alveolus Lung-Chip successfully recapitulated FOLR1 TCB-mediated lung toxicities observed in cynomolgus monkeys and instructed the design of a second-generation molecule, whose favorable lung safety profile predicted by the chip model was verified in vivo. The Intestine-Chip model captured the liabilities of a TCB targeting a human-specific antigen, thus filling the gap left by the lack of cross-reactivity in animals and suitable animal models overall. The model likewise displayed sensitivity to TCB affinity and predicted differential, target expression-dependent toxicity outcomes between different intestinal regions. Both models were able to shed light into the toxicity mechanisms of the molecules, clearly decoupling target-mediated effects from T cell activation through CD3 engagement only, which has been shown to be an additional mode of TCB-induced adverse events (Segal et al., 1999). It is worth mentioning that both the Lung-Chip and the Intestine-Chip demonstrated an advantage over conventional models for this application: the Alveolus Lung-Chip was found to report more specific TCB-responses and provide additional important readouts compared with transwell-based approaches, whereas the Intestine-Chip supported more physiologically relevant target expression, compared with both 3D organoids and primary intestinal barrier grown in transwells.

Owing to the absence of predictive early-stage assays, ‘on-target, off-tumor’ TCB safety liabilities have in some cases been only detected either in late-stage preclinical models (non-human primates) or as life-threatening adverse events in the clinic. Advanced human cellular models that capture the immunopathology of TCB-induced adverse events would aid the iterative antibody design at the early stage, thus ensuring favorable safety profiles before entering clinical trials, reducing attrition rates, and ultimately expediting the application of these potentially life-saving therapies. Importantly, the mechanistic insights into on-target, off-tumor toxicities and design opportunities afforded by these platforms are not restricted to T-cell engagers like TCBs, but also apply to chimeric antigen receptor T (CART) cell therapy, bearing in mind their similar modes of action. For instance, a human epidermal growth factor receptor 2 (Her2) CART therapy, intended to treat a patient with colorectal cancer, led to lethal toxicity through off-tumor cardiopulmonary targeting (Morgan et al., 2010). To improve its efficacy-safety profile, this therapy was affinity-tuned to detect tumor cells with a high density of surface antigens, while sparing normal cells with lower antigen expression (Zhao et al., 2009). A mouse model expressing human Her2 was used to confirm the safer profile of the low-affinity CART (Castellarin et al., 2020). Using the Lung- and Intestine-Chips, we similarly identified a reduced risk of healthy tissue targeting with a low-affinity TCB for both the FOLR1 and CEA targets. However, in contrast to the humanized mouse models, these in vitro tools are fully human, applicable to various targets and faster to generate, making them a promising alternative for antibody/CART preclinical safety testing, format selection, and optimization.

Furthermore, the immune-competent Organs-Chips described here can bridge preclinical research to clinical application, by aiding the discovery of early predictive biomarkers of TCB/CART toxicity in patients, which would certainly help to anticipate and manage life-threatening adverse events. While no universally predictive markers of adverse events associated with these therapies are currently accepted and validated, the phenotypic outcomes observed in the chips closely match some of the few clinical indicators that have been proposed. For example, early elevation of specific serum cytokines, including IFN-γ and IL-6, was found to precede the development of severe cytokine release syndrome in response to CART therapy (Teachey et al., 2016; Hay, 2018). The release of IFN-γ and IL-6 was consistently observed in both the Alveolus Lung and Colon Intestine-Chip upon treatment with high-affinity FOLR1 and CEA TCB, which led to on-target toxicity. The detection of such clinically relevant biomarkers exemplifies the translational value of our platform. Going forward, the system could be coupled with, for example, unbiased proteomic analyses of chip supernatants and transcriptomic dissection of the effector and target cell pools and used to uncover early novel predictors or TCB-mediated adverse events. It may also be of interest to other types of molecules such as super agonist CD28 antibodies, which induced a severe cytokine storm in the clinics (Hünig, 2012).

Another attractive future development of our platform is its use to evaluate the therapeutic window of a therapy. We tested the same molecule in a healthy Lung-Chip versus a cancer-Chip and found a 1000-fold difference in the concentration of TCB required to kill target cells (Figure 2—figure supplement 2). As a next step, an all-in-one platform could combine a healthy population with a tumor one in the same chip as previously described (Hassell et al., 2017), thus capturing safety liabilities triggered by tumor lysis itself. Although efficacy–safety interactions are clinically relevant, these two aspects are often evaluated independently prior to phase I clinical studies, owing largely to the absence of fitting preclinical tools. We believe that using a chip comprising only healthy tissue is a reasonable safety evaluation approach in the case when the organ of concern is different from the cancer-affected organ. However, the tumor microenvironment and anti-tumor efficacy effects might influence the toxicity outcome when the target-expressing organ and the tumor co-localize, and should therefore be taken into account. For example, treating colorectal cancer with CEA-targeting TCBs may lead to an increased risk of intestinal toxicities, taking into consideration that tumor cell lysis may lead to the release of inflammatory cytokines and tumor antigens, thus further potentiating immune cell activation and cytotoxicity. The coupling of healthy and tumor biopsy-derived organoids with immunocompetent Organ-Chip technology may provide the ideal setup for a combined efficacy and safety assessment. Aside from safety questions, the presence of a relevant tumor compartment would open possibilities for tuning of compound pharmacology and efficacy testing. For example, a matrix layer below a tumor cell layer could be used to better recapitulate the immunosuppressive tumor environment via the introduction of regulatory T cells and tumor-associated macrophages or fibroblasts, taking it into account when designing therapies.

In conclusion, we describe novel human immunocompetent models of the lung and intestine and validate them as platforms for TCB safety profiling, outlining how these systems could reduce the reliance on animal-based safety assessments, enable educated antibody format selection, shed light into the mechanistic underpinnings of toxicities, and support the identification of clinical biomarkers. Going forward, the concepts we introduced here can be expanded to address the persisting gap in modeling immune-related toxicities, associated with, for example, immune checkpoint blockade (Brahmer et al., 2018; Ramos-Casals et al., 2020). Considering their systemic and multifactorial immunopathology, modeling these processes accurately would likely require the incorporation of resident immune cells and lymphoid structures, as well as modalities that support the simulation of T-cell trafficking and tissue infiltration.

RNA sequencing data have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus (GEO) under accession number GSE175821.

1. 1. Kerns SJ
   2. Manatakis DV

   (2021) NCBI Gene Expression Omnibus

   ID GSE175821. Genome-wide transcriptome profiling of human organoids, colon intestine-chip and human alveolus-chips using RNA-seq.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE175821

1. Conference

   1. AMea G

   (2016)

   Adverse or not adverse—assessment and consequences

   In 14th European Congress of Toxicologic Pathology (ESTP) Barcelona Spain.

   • Google Scholar
2. Preprint

   1. Apostolou A
   2. Panchakshari RA
   3. Banerjee A
   4. Manatakis DV
   5. Paraskevopoulou MD
   6. Luc R

   (2020) A Micro-engineered human Colon Intestine-Chip platform to study leaky barrier

   bioRxiv.

   https://doi.org/10.1101/2020.08.28.271759

   • Google Scholar
3. 1. Bacac M
   2. Klein C
   3. Umana P

   (2016a) CEA TCB: a novel head-to-tail 2:1 T cell bispecific antibody for treatment of CEA-positive solid tumors

   OncoImmunology 5:e1203498.

   https://doi.org/10.1080/2162402X.2016.1203498

   • PubMed
   • Google Scholar
4. 1. Bacac M
   2. Fauti T
   3. Sam J
   4. Colombetti S
   5. Weinzierl T
   6. Ouaret D
   7. Bodmer W
   8. Lehmann S
   9. Hofer T
   10. Hosse RJ
   11. Moessner E
   12. Ast O
   13. Bruenker P
   14. Grau-Richards S
   15. Schaller T
   16. Seidl A
   17. Gerdes C
   18. Perro M
   19. Nicolini V
   20. Steinhoff N
   21. Dudal S
   22. Neumann S
   23. von Hirschheydt T
   24. Jaeger C
   25. Saro J
   26. Karanikas V
   27. Klein C
   28. Umaña P

   (2016b) A novel carcinoembryonic antigen T-Cell bispecific antibody (CEA TCB) for the treatment of solid tumors

   Clinical Cancer Research 22:3286–3297.

   https://doi.org/10.1158/1078-0432.CCR-15-1696

   • PubMed
   • Google Scholar
5. 1. Benchimol S
   2. Fuks A
   3. Jothy S
   4. Beauchemin N
   5. Shirota K
   6. Stanners CP

   (1989) Carcinoembryonic antigen, a human tumor marker, functions as an intercellular adhesion molecule

   Cell 57:327–334.

   https://doi.org/10.1016/0092-8674(89)90970-7

   • PubMed
   • Google Scholar
6. Preprint

   1. Bjornson-Hooper ZB
   2. Fragiadakis GK
   3. Spitzer MH
   4. Madhireddy D
   5. McIlwain D
   6. Nolan GP

   (2019) A comprehensive atlas of immunological differences between humans mice and non-human primates

   bioRxiv.

   https://doi.org/10.1101/574160

   • Google Scholar
7. 1. Brahmer JR
   2. Lacchetti C
   3. Schneider BJ
   4. Atkins MB
   5. Brassil KJ
   6. Caterino JM
   7. Chau I
   8. Ernstoff MS
   9. Gardner JM
   10. Ginex P
   11. Hallmeyer S
   12. Holter Chakrabarty J
   13. Leighl NB
   14. Mammen JS
   15. McDermott DF
   16. Naing A
   17. Nastoupil LJ
   18. Phillips T
   19. Porter LD
   20. Puzanov I
   21. Reichner CA
   22. Santomasso BD
   23. Seigel C
   24. Spira A
   25. Suarez-Almazor ME
   26. Wang Y
   27. Weber JS
   28. Wolchok JD
   29. Thompson JA
   30. National Comprehensive Cancer Network

   (2018) Management of Immune-Related adverse events in patients treated with immune checkpoint inhibitor therapy: american society of clinical oncology clinical practice guideline

   Journal of Clinical Oncology 36:1714–1768.

   https://doi.org/10.1200/JCO.2017.77.6385

   • PubMed
   • Google Scholar
8. 1. Brischwein K
   2. Schlereth B
   3. Guller B
   4. Steiger C
   5. Wolf A
   6. Lutterbuese R
   7. Offner S
   8. Locher M
   9. Urbig T
   10. Raum T
   11. Kleindienst P
   12. Wimberger P
   13. Kimmig R
   14. Fichtner I
   15. Kufer P
   16. Hofmeister R
   17. da Silva AJ
   18. Baeuerle PA

   (2006) MT110: a novel bispecific single-chain antibody construct with high efficacy in eradicating established tumors

   Molecular Immunology 43:1129–1143.

   https://doi.org/10.1016/j.molimm.2005.07.034

   • PubMed
   • Google Scholar
9. 1. Castellarin M
   2. Sands C
   3. Da T
   4. Scholler J
   5. Graham K
   6. Buza E
   7. Fraietta JA
   8. Zhao Y
   9. June CH

   (2020) A rational mouse model to detect on-target, off-tumor CAR T cell toxicity

   JCI Insight 5:136012.

   https://doi.org/10.1172/jci.insight.136012

   • PubMed
   • Google Scholar
10. 1. Champiat S
    2. Dercle L
    3. Ammari S
    4. Massard C
    5. Hollebecque A
    6. Postel-Vinay S
    7. Chaput N
    8. Eggermont A
    9. Marabelle A
    10. Soria JC
    11. Ferté C

    (2017) Hyperprogressive disease is a new pattern of progression in Cancer patients treated by Anti-PD-1/PD-L1

    Clinical Cancer Research 23:1920–1928.

    https://doi.org/10.1158/1078-0432.CCR-16-1741

    • PubMed
    • Google Scholar
11. 1. Clynes RA
    2. Desjarlais JR

    (2019) Redirected T cell cytotoxicity in Cancer therapy

    Annual Review of Medicine 70:437–450.

    https://doi.org/10.1146/annurev-med-062617-035821

    • PubMed
    • Google Scholar
12. 1. Gayer CP
    2. Basson MD

    (2009) The effects of mechanical forces on intestinal physiology and pathology

    Cellular Signalling 21:1237–1244.

    https://doi.org/10.1016/j.cellsig.2009.02.011

    • PubMed
    • Google Scholar
13. 1. Geiger M
    2. Stubenrauch KG
    3. Sam J
    4. Richter WF
    5. Jordan G
    6. Eckmann J
    7. Hage C
    8. Nicolini V
    9. Freimoser-Grundschober A
    10. Ritter M
    11. Lauer ME
    12. Stahlberg H
    13. Ringler P
    14. Patel J
    15. Sullivan E
    16. Grau-Richards S
    17. Endres S
    18. Kobold S
    19. Umaña P
    20. Brünker P
    21. Klein C

    (2020) Protease-activation using anti-idiotypic masks enables tumor specificity of a folate receptor 1-T cell bispecific antibody

    Nature Communications 11:3196.

    https://doi.org/10.1038/s41467-020-16838-w

    • PubMed
    • Google Scholar
14. 1. Goebeler ME
    2. Bargou RC

    (2020) T cell-engaging therapies - BiTEs and beyond

    Nature Reviews Clinical Oncology 17:418–434.

    https://doi.org/10.1038/s41571-020-0347-5

    • PubMed
    • Google Scholar
15. 1. Gong J
    2. Chehrazi-Raffle A
    3. Reddi S
    4. Salgia R

    (2018) Development of PD-1 and PD-L1 inhibitors as a form of Cancer immunotherapy: a comprehensive review of registration trials and future considerations

    Journal for ImmunoTherapy of Cancer 6:8.

    https://doi.org/10.1186/s40425-018-0316-z

    • PubMed
    • Google Scholar
16. 1. Hammarström S

    (1999) The carcinoembryonic antigen (CEA) family: structures, suggested functions and expression in normal and malignant tissues

    Seminars in Cancer Biology 9:67–81.

    https://doi.org/10.1006/scbi.1998.0119

    • PubMed
    • Google Scholar
17. 1. Hassell BA
    2. Goyal G
    3. Lee E
    4. Sontheimer-Phelps A
    5. Levy O
    6. Chen CS
    7. Ingber DE

    (2017) Human organ chip models recapitulate orthotopic lung Cancer growth, therapeutic responses, and tumor dormancy in Vitro

    Cell Reports 21:508–516.

    https://doi.org/10.1016/j.celrep.2017.09.043

    • PubMed
    • Google Scholar
18. 1. Hay KA

    (2018) Cytokine release syndrome and neurotoxicity after CD19 chimeric antigen receptor-modified (CAR-) T cell therapy

    British Journal of Haematology 183:364–374.

    https://doi.org/10.1111/bjh.15644

    • PubMed
    • Google Scholar
19. 1. Hegde PS
    2. Chen DS

    (2020) Top 10 challenges in Cancer immunotherapy

    Immunity 52:17–35.

    https://doi.org/10.1016/j.immuni.2019.12.011

    • PubMed
    • Google Scholar
20. 1. Hodi FS
    2. O'Day SJ
    3. McDermott DF
    4. Weber RW
    5. Sosman JA
    6. Haanen JB
    7. Gonzalez R
    8. Robert C
    9. Schadendorf D
    10. Hassel JC
    11. Akerley W
    12. van den Eertwegh AJ
    13. Lutzky J
    14. Lorigan P
    15. Vaubel JM
    16. Linette GP
    17. Hogg D
    18. Ottensmeier CH
    19. Lebbé C
    20. Peschel C
    21. Quirt I
    22. Clark JI
    23. Wolchok JD
    24. Weber JS
    25. Tian J
    26. Yellin MJ
    27. Nichol GM
    28. Hoos A
    29. Urba WJ

    (2010) Improved survival with ipilimumab in patients with metastatic melanoma

    New England Journal of Medicine 363:711–723.

    https://doi.org/10.1056/NEJMoa1003466

    • PubMed
    • Google Scholar
21. 1. Huh D
    2. Matthews BD
    3. Mammoto A
    4. Montoya-Zavala M
    5. Hsin HY
    6. Ingber DE

    (2010) Reconstituting organ-level lung functions on a chip

    Science 328:1662–1668.

    https://doi.org/10.1126/science.1188302

    • PubMed
    • Google Scholar
22. 1. Huh D
    2. Torisawa YS
    3. Hamilton GA
    4. Kim HJ
    5. Ingber DE

    (2012) Microengineered physiological biomimicry: organs-on-chips

    Lab on a Chip 12:2156–2164.

    https://doi.org/10.1039/c2lc40089h

    • PubMed
    • Google Scholar
23. 1. Hünig T

    (2012) The storm has cleared: lessons from the CD28 superagonist TGN1412 trial

    Nature Reviews Immunology 12:317–318.

    https://doi.org/10.1038/nri3192

    • PubMed
    • Google Scholar
24. 1. Ishiguro T
    2. Sano Y
    3. Komatsu SI
    4. Kamata-Sakurai M
    5. Kaneko A
    6. Kinoshita Y
    7. Shiraiwa H
    8. Azuma Y
    9. Tsunenari T
    10. Kayukawa Y
    11. Sonobe Y
    12. Ono N
    13. Sakata K
    14. Fujii T
    15. Miyazaki Y
    16. Noguchi M
    17. Endo M
    18. Harada A
    19. Frings W
    20. Fujii E
    21. Nanba E
    22. Narita A
    23. Sakamoto A
    24. Wakabayashi T
    25. Konishi H
    26. Segawa H
    27. Igawa T
    28. Tsushima T
    29. Mutoh H
    30. Nishito Y
    31. Takahashi M
    32. Stewart L
    33. ElGabry E
    34. Kawabe Y
    35. Ishigai M
    36. Chiba S
    37. Aoki M
    38. Hattori K
    39. Nezu J

    (2017) An anti-glypican 3/CD3 bispecific T cell-redirecting antibody for treatment of solid tumors

    Science Translational Medicine 9:eaal4291.

    https://doi.org/10.1126/scitranslmed.aal4291

    • PubMed
    • Google Scholar
25. 1. Jain A
    2. Barrile R
    3. van der Meer AD
    4. Mammoto A
    5. Mammoto T
    6. De Ceunynck K
    7. Aisiku O
    8. Otieno MA
    9. Louden CS
    10. Hamilton GA
    11. Flaumenhaft R
    12. Ingber DE

    (2018) Primary human lung Alveolus-on-a-chip model of intravascular thrombosis for assessment of therapeutics

    Clinical Pharmacology & Therapeutics 103:332–340.

    https://doi.org/10.1002/cpt.742

    • PubMed
    • Google Scholar
26. 1. Kasendra M
    2. Tovaglieri A
    3. Sontheimer-Phelps A
    4. Jalili-Firoozinezhad S
    5. Bein A
    6. Chalkiadaki A
    7. Scholl W
    8. Zhang C
    9. Rickner H
    10. Richmond CA
    11. Li H
    12. Breault DT
    13. Ingber DE

    (2018) Development of a primary human small Intestine-on-a-Chip using biopsy-derived organoids

    Scientific Reports 8:2871.

    https://doi.org/10.1038/s41598-018-21201-7

    • PubMed
    • Google Scholar
27. 1. Kasendra M
    2. Luc R
    3. Yin J
    4. Manatakis DV
    5. Kulkarni G
    6. Lucchesi C
    7. Sliz J
    8. Apostolou A
    9. Sunuwar L
    10. Obrigewitch J
    11. Jang KJ
    12. Hamilton GA
    13. Donowitz M
    14. Karalis K

    (2020) Duodenum Intestine-Chip for preclinical drug assessment in a human relevant model

    eLife 9:e50135.

    https://doi.org/10.7554/eLife.50135

    • PubMed
    • Google Scholar
28. 1. Kebenko M
    2. Goebeler ME
    3. Wolf M
    4. Hasenburg A
    5. Seggewiss-Bernhardt R
    6. Ritter B
    7. Rautenberg B
    8. Atanackovic D
    9. Kratzer A
    10. Rottman JB
    11. Friedrich M
    12. Vieser E
    13. Elm S
    14. Patzak I
    15. Wessiepe D
    16. Stienen S
    17. Fiedler W

    (2018) A multicenter phase 1 study of solitomab (MT110, AMG 110), a bispecific EpCAM/CD3 T-cell engager (BiTE(R)) antibody construct, in patients with refractory solid tumors

    OncoImmunology 7:e1450710.

    https://doi.org/10.1080/2162402X.2018.1450710

    • PubMed
    • Google Scholar
29. 1. Kennedy LB
    2. Salama AKS

    (2020) A review of Cancer immunotherapy toxicity

    CA: A Cancer Journal for Clinicians 70:86–104.

    https://doi.org/10.3322/caac.21596

    • PubMed
    • Google Scholar
30. 1. Klein C
    2. Sustmann C
    3. Thomas M
    4. Stubenrauch K
    5. Croasdale R
    6. Schanzer J
    7. Brinkmann U
    8. Kettenberger H
    9. Regula JT
    10. Schaefer W

    (2012) Progress in overcoming the chain association issue in bispecific heterodimeric IgG antibodies

    mAbs 4:653–663.

    https://doi.org/10.4161/mabs.21379

    • PubMed
    • Google Scholar
31. 1. Klinger M
    2. Benjamin J
    3. Kischel R
    4. Stienen S
    5. Zugmaier G

    (2016) Harnessing T cells to fight Cancer with BiTE antibody constructs--past developments and future directions

    Immunological Reviews 270:193–208.

    https://doi.org/10.1111/imr.12393

    • PubMed
    • Google Scholar
32. 1. Labrijn AF
    2. Janmaat ML
    3. Reichert JM
    4. Parren P

    (2019) Bispecific antibodies: a mechanistic review of the pipeline

    Nature Reviews Drug Discovery 18:585–608.

    https://doi.org/10.1038/s41573-019-0028-1

    • PubMed
    • Google Scholar
33. 1. Lutterbuese R
    2. Raum T
    3. Kischel R
    4. Hoffmann P
    5. Mangold S
    6. Rattel B
    7. Friedrich M
    8. Thomas O
    9. Lorenczewski G
    10. Rau D
    11. Schaller E
    12. Herrmann I
    13. Wolf A
    14. Urbig T
    15. Baeuerle PA
    16. Kufer P

    (2010) T cell-engaging BiTE antibodies specific for EGFR potently eliminate KRAS- and BRAF-mutated colorectal Cancer cells

    PNAS 107:12605–12610.

    https://doi.org/10.1073/pnas.1000976107

    • PubMed
    • Google Scholar
34. 1. Morgan RA
    2. Yang JC
    3. Kitano M
    4. Dudley ME
    5. Laurencot CM
    6. Rosenberg SA

    (2010) Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2

    Molecular Therapy 18:843–851.

    https://doi.org/10.1038/mt.2010.24

    • PubMed
    • Google Scholar
35. 1. Naidoo J
    2. Page DB
    3. Li BT
    4. Connell LC
    5. Schindler K
    6. Lacouture ME
    7. Postow MA
    8. Wolchok JD

    (2015) Toxicities of the anti-PD-1 and anti-PD-L1 immune checkpoint antibodies

    Annals of Oncology 26:2375–2391.

    https://doi.org/10.1093/annonc/mdv383

    • PubMed
    • Google Scholar
36. 1. Parker N
    2. Turk MJ
    3. Westrick E
    4. Lewis JD
    5. Low PS
    6. Leamon CP

    (2005) Folate receptor expression in carcinomas and normal tissues determined by a quantitative radioligand binding assay

    Analytical Biochemistry 338:284–293.

    https://doi.org/10.1016/j.ab.2004.12.026

    • PubMed
    • Google Scholar
37. 1. Ramos-Casals M
    2. Brahmer JR
    3. Callahan MK
    4. Flores-Chávez A
    5. Keegan N
    6. Khamashta MA
    7. Lambotte O
    8. Mariette X
    9. Prat A
    10. Suárez-Almazor ME

    (2020) Immune-related adverse events of checkpoint inhibitors

    Nature Reviews Disease Primers 6:38.

    https://doi.org/10.1038/s41572-020-0160-6

    • PubMed
    • Google Scholar
38. 1. Ridgway JB
    2. Presta LG
    3. Carter P

    (1996) 'Knobs-into-holes' engineering of antibody CH3 domains for heavy chain heterodimerization

    Protein Engineering 9:617–621.

    https://doi.org/10.1093/protein/9.7.617

    • PubMed
    • Google Scholar
39. 1. Sato T
    2. Stange DE
    3. Ferrante M
    4. Vries RG
    5. Van Es JH
    6. Van den Brink S
    7. Van Houdt WJ
    8. Pronk A
    9. Van Gorp J
    10. Siersema PD
    11. Clevers H

    (2011) Long-term expansion of epithelial organoids from human Colon, adenoma, adenocarcinoma, and Barrett's epithelium

    Gastroenterology 141:1762–1772.

    https://doi.org/10.1053/j.gastro.2011.07.050

    • PubMed
    • Google Scholar
40. 1. Scaranti M
    2. Cojocaru E
    3. Banerjee S
    4. Banerji U

    (2020) Exploiting the folate receptor α in oncology

    Nature Reviews Clinical Oncology 17:349–359.

    https://doi.org/10.1038/s41571-020-0339-5

    • PubMed
    • Google Scholar
41. 1. Schadendorf D
    2. Hodi FS
    3. Robert C
    4. Weber JS
    5. Margolin K
    6. Hamid O
    7. Patt D
    8. Chen TT
    9. Berman DM
    10. Wolchok JD

    (2015) Pooled analysis of Long-Term survival data from phase II and phase III trials of ipilimumab in unresectable or metastatic melanoma

    Journal of Clinical Oncology 33:1889–1894.

    https://doi.org/10.1200/JCO.2014.56.2736

    • PubMed
    • Google Scholar
42. 1. Schlothauer T
    2. Herter S
    3. Koller CF
    4. Grau-Richards S
    5. Steinhart V
    6. Spick C
    7. Kubbies M
    8. Klein C
    9. Umaña P
    10. Mössner E

    (2016) Novel human IgG1 and IgG4 Fc-engineered antibodies with completely abolished immune effector functions

    Protein Engineering Design and Selection 29:457–466.

    https://doi.org/10.1093/protein/gzw040

    • PubMed
    • Google Scholar
43. 1. Seckinger A
    2. Delgado JA
    3. Moser S
    4. Moreno L
    5. Neuber B
    6. Grab A
    7. Lipp S
    8. Merino J
    9. Prosper F
    10. Emde M
    11. Delon C
    12. Latzko M
    13. Gianotti R
    14. Lüoend R
    15. Murr R
    16. Hosse RJ
    17. Harnisch LJ
    18. Bacac M
    19. Fauti T
    20. Klein C
    21. Zabaleta A
    22. Hillengass J
    23. Cavalcanti-Adam EA
    24. Ho AD
    25. Hundemer M
    26. San Miguel JF
    27. Strein K
    28. Umaña P
    29. Hose D
    30. Paiva B
    31. Vu MD

    (2017) Target expression, generation, preclinical activity, and pharmacokinetics of the BCMA-T cell bispecific antibody EM801 for multiple myeloma treatment

    Cancer Cell 31:396–410.

    https://doi.org/10.1016/j.ccell.2017.02.002

    • PubMed
    • Google Scholar
44. 1. Segal DM
    2. Weiner GJ
    3. Weiner LM

    (1999) Bispecific antibodies in Cancer therapy

    Current Opinion in Immunology 11:558–562.

    https://doi.org/10.1016/S0952-7915(99)00015-1

    • PubMed
    • Google Scholar
45. 1. Teachey DT
    2. Lacey SF
    3. Shaw PA
    4. Melenhorst JJ
    5. Maude SL
    6. Frey N
    7. Pequignot E
    8. Gonzalez VE
    9. Chen F
    10. Finklestein J
    11. Barrett DM
    12. Weiss SL
    13. Fitzgerald JC
    14. Berg RA
    15. Aplenc R
    16. Callahan C
    17. Rheingold SR
    18. Zheng Z
    19. Rose-John S
    20. White JC
    21. Nazimuddin F
    22. Wertheim G
    23. Levine BL
    24. June CH
    25. Porter DL
    26. Grupp SA

    (2016) Identification of predictive biomarkers for cytokine release syndrome after chimeric antigen receptor T-cell therapy for acute lymphoblastic leukemia

    Cancer Discovery 6:664–679.

    https://doi.org/10.1158/2159-8290.CD-16-0040

    • PubMed
    • Google Scholar
46. 1. Thomas P
    2. Gangopadhyay A
    3. Steele G
    4. Andrews C
    5. Nakazato H
    6. Oikawa S
    7. Jessup JM

    (1995) The effect of transfection of the CEA gene on the metastatic behavior of the human colorectal Cancer cell line MIP-101

    Cancer Letters 92:59–66.

    https://doi.org/10.1016/0304-3835(95)03764-N

    • PubMed
    • Google Scholar
47. 1. Trabolsi A
    2. Arumov A
    3. Schatz JH

    (2019) T Cell-Activating bispecific antibodies in Cancer therapy

    The Journal of Immunology 203:585–592.

    https://doi.org/10.4049/jimmunol.1900496

    • PubMed
    • Google Scholar
48. 1. Waldman AD
    2. Fritz JM
    3. Lenardo MJ

    (2020) A guide to Cancer immunotherapy: from T cell basic science to clinical practice

    Nature Reviews Immunology 20:651–668.

    https://doi.org/10.1038/s41577-020-0306-5

    • PubMed
    • Google Scholar
49. 1. Wolchok JD
    2. Chiarion-Sileni V
    3. Gonzalez R
    4. Rutkowski P
    5. Grob JJ
    6. Cowey CL
    7. Lao CD
    8. Wagstaff J
    9. Schadendorf D
    10. Ferrucci PF
    11. Smylie M
    12. Dummer R
    13. Hill A
    14. Hogg D
    15. Haanen J
    16. Carlino MS
    17. Bechter O
    18. Maio M
    19. Marquez-Rodas I
    20. Guidoboni M
    21. McArthur G
    22. Lebbé C
    23. Ascierto PA
    24. Long GV
    25. Cebon J
    26. Sosman J
    27. Postow MA
    28. Callahan MK
    29. Walker D
    30. Rollin L
    31. Bhore R
    32. Hodi FS
    33. Larkin J

    (2017) Overall survival with combined nivolumab and ipilimumab in advanced melanoma

    New England Journal of Medicine 377:1345–1356.

    https://doi.org/10.1056/NEJMoa1709684

    • PubMed
    • Google Scholar
50. 1. Yang Y

    (2015) Cancer immunotherapy: harnessing the immune system to battle Cancer

    Journal of Clinical Investigation 125:3335–3337.

    https://doi.org/10.1172/JCI83871

    • PubMed
    • Google Scholar
51. 1. Zhao Y
    2. Wang QJ
    3. Yang S
    4. Kochenderfer JN
    5. Zheng Z
    6. Zhong X
    7. Sadelain M
    8. Eshhar Z
    9. Rosenberg SA
    10. Morgan RA

    (2009) A herceptin-based chimeric antigen receptor with modified signaling domains leads to enhanced survival of transduced T lymphocytes and antitumor activity

    The Journal of Immunology 183:5563–5574.

    https://doi.org/10.4049/jimmunol.0900447

    • PubMed
    • Google Scholar
52. 1. Zhou H
    2. Stanners CP
    3. Fuks A

    (1993)

    Specificity of anti-carcinoembryonic antigen monoclonal antibodies and their effects on CEA-mediated adhesion

    Cancer Research 53:3817–3822.

    • PubMed
    • Google Scholar

Author details

1. S Jordan Kerns

   Emulate Inc, Boston, United States

   Contribution

   Formal analysis, Supervision, Validation, Investigation, Visualization, Writing - review and editing

   Contributed equally with

   Chaitra Belgur

   Competing interests

   Is a current employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

2. Chaitra Belgur

   Emulate Inc, Boston, United States

   Contribution

   Software, Formal analysis, Validation, Investigation, Visualization, Writing - review and editing

   Contributed equally with

   S Jordan Kerns

   Competing interests

   Is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

   "This ORCID iD identifies the author of this article:" 0000-0001-5670-6166

3. Debora Petropolis

   Emulate Inc, Boston, United States

   Contribution

   Data curation, Software, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology

   Competing interests

   Is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers Method for Assessing a Compound Interacting with a Target on Epithelial Cells,

4. Marianne Kanellias

   Emulate Inc, Boston, United States

   Contribution

   Software, Validation, Investigation, Visualization

   Competing interests

   Is a current or former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc).

5. Riccardo Barrile

   1. Emulate Inc, Boston, United States
   2. Department of Biomedical Engineering, University of Cincinnati, Cincinnati, United States

   Contribution

   Conceptualization, Formal analysis, Supervision, Investigation, Methodology

   Competing interests

   Is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers, Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

   "This ORCID iD identifies the author of this article:" 0000-0002-7301-3959

6. Johannes Sam

   Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

   Contribution

   Formal analysis, Visualization, Methodology

   Competing interests

   Employment, patents and ownership of stock with Roche.

7. Tina Weinzierl

   Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

   Contribution

   Formal analysis, Supervision, Investigation, Visualization

   Competing interests

   Employment, patents and ownership of stock with Roche.

8. Tanja Fauti

   Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

   Contribution

   Formal analysis, Investigation, Visualization

   Competing interests

   Employment, patents and ownership of stock with Roche.

9. Anne Freimoser-Grundschober

   Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

   Contribution

   Resources, Supervision, Investigation, Visualization, Methodology

   Competing interests

   Employment, patents and ownership of stock with Roche.

10. Jan Eckmann

    Roche Pharma Research & Early Development, Roche Innovation Center Munich, Penzberg, Germany

    Contribution

    Supervision, Investigation, Visualization, Methodology

    Competing interests

    Employment, patents and ownership of stock with Roche.

11. Carina Hage

    Roche Pharma Research & Early Development, Roche Innovation Center Munich, Penzberg, Germany

    Contribution

    Investigation, Visualization, Methodology

    Competing interests

    Employment and ownership of stock with Roche.

12. Martina Geiger

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources

    Competing interests

    Employment, patents and ownership of stock with Roche.

13. Patrick Ray Ng

    Emulate Inc, Boston, United States

    Contribution

    Data curation, Investigation

    Competing interests

    is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc).

    "This ORCID iD identifies the author of this article:" 0000-0003-4566-9173

14. William Tien-Street

    Emulate Inc, Boston, United States

    Contribution

    Data curation, Investigation

    Competing interests

    is a current employee of and hold equity interests or options to obtain equity interests in (Emulate Inc).

15. Dimitris V Manatakis

    Emulate Inc, Boston, United States

    Contribution

    Software, Visualization

    Competing interests

    Is a current employee of and hold equity interests or options to obtain equity interests in (Emulate Inc).

16. Virginie Micallef

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Investigation, Visualization

    Competing interests

    Employment and ownership of stock with Roche.

17. Regine Gerard

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Investigation, Visualization

    Competing interests

    Employment and ownership of stock with Roche.

18. Michael Bscheider

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Supervision

    Competing interests

    Employment and ownership of stock with Roche.

19. Ekaterina Breous-Nystrom

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Project administration

    Competing interests

    Employment and ownership of stock with Roche.

20. Anneliese Schneider

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources

    Competing interests

    Employment, patents and ownership of stock with Roche.

21. Anna Maria Giusti

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources, Investigation, Visualization, Methodology

    Competing interests

    Employment, patents and ownership of stock with Roche.

22. Cristina Bertinetti-Lapatki

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Conceptualization, Supervision

    Competing interests

    Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'. Employment, patents and ownership of stock with Roche.

23. Heather Shannon Grant

    Emulate Inc, Boston, United States

    Contribution

    Project administration, Writing - review and editing

    Competing interests

    Is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

24. Adrian B Roth

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Project administration

    Competing interests

    Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'. Employment, patents and ownership of stock with Roche.

25. Geraldine A Hamilton

    Emulate Inc, Boston, United States

    Contribution

    Project administration

    Competing interests

    Is a current or former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

26. Thomas Singer

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Funding acquisition, Project administration

    Competing interests

    Employment, patents and ownership of stock with Roche.

27. Katia Karalis

    Emulate Inc, Boston, United States

    Contribution

    Conceptualization, Supervision, Writing - review and editing

    Competing interests

    Is a former employee of and hold equity interests or options to obtain equity interests in (Emulate Inc). Is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers 'Method for Assessing a Compound Interacting with a Target on Epithelial Cells'

28. Annie Moisan

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Conceptualization, Supervision, Writing - review and editing

    Competing interests

    is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers Method for Assessing a Compound Interacting with a Target on Epithelial Cells'.

29. Peter Bruenker

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources

    Competing interests

    No competing interests declared

30. Christian Klein

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources, Supervision, Writing - review and editing

    Competing interests

    Employment, patents and ownership of stock with Roche.

31. Marina Bacac

    Roche Pharma Research & Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland

    Contribution

    Resources, Supervision, Writing - review and editing

    Competing interests

    Employment, patents and ownership of stock with Roche.

32. Nikolce Gjorevski

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Conceptualization, Data curation, Formal analysis, Supervision, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

    Contributed equally with

    Lauriane Cabon

    For correspondence

    nikolche.gjorevski@roche.com

    Competing interests

    is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers Method for Assessing a Compound Interacting with a Target on Epithelial Cells'. Employment, patents and ownership of stock with Roche.

    "This ORCID iD identifies the author of this article:" 0000-0001-8320-0443

33. Lauriane Cabon

    Roche Pharma Research & Early Development, Roche Innovation Center Basel, Basel, Switzerland

    Contribution

    Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

    Contributed equally with

    Nikolce Gjorevski

    For correspondence

    lauriane.cabon@roche.com

    Competing interests

    is an inventor on a patent application (P16451EP00/16/912,391) submitted by Hoffmann-LaRoche and Emulate that covers Method for Assessing a Compound Interacting with a Target on Epithelial Cells'. Employment, patents and ownership of stock with Roche.

    "This ORCID iD identifies the author of this article:" 0000-0001-8472-2227

• 6,102

  Page views

• 986

  Downloads

• 31

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.