Engineering Organ-on-a-Chip Systems for Cancer Immunotherapy: Strategies and Assay Integration

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Very well and thoroughly written review on OOAC systems for cancer immunotherapy. I suggest the authors to consider adding the following to their reviews as it may help improve the review and help their readers:

• Discuss HLA-mismatch issue and the need for using both cancer cells (organoids) and immune cells (T cells) derived from the same individuals.
• Discuss steps in immune cells preparation. There was a discussion on Organoid preparation, but no T cells. For example, blood gets collected, the PMBC extracted. Then frozen in LN2. Then woken up in a future date when organoids are ready in the Chip for the co culture. Then, T cells require to be cultured/expanded for a few days, activated IL-2 and other T cell activators before introducing them to the Chip. Our group recently published an article on this workflow (PMID: 41610338). I am not asking you to cite our paper, but instead just providing the workflow as an example which can be found in other recent papers too.
• Maybe helpful to discuss the mechanism of action of Immune Checkpoint inhibitors and how they act either on PD-1 or PD-L1 receptors for a more complete picture on immunotherapy. 

Author Response

Discuss HLA-mismatch issue and the need for using both cancer cells (organoids) and immune cells (T cells) derived from the same individuals.

We thank the reviewer for this insightful comment. We have added a brief discussion on HLA compatibility and the importance of using autologous cancer and immune cells to better recapitulate patient-specific immune responses. Please see lines 204-213 highlighted in yellow.

 

Discuss steps in immune cells preparation. There was a discussion on Organoid preparation, but no T cells. For example, blood gets collected, the PMBC extracted. Then frozen in LN2. Then woken up in a future date when organoids are ready in the Chip for the co culture. Then, T cells require to be cultured/expanded for a few days, activated IL-2 and other T cell activators before introducing them to the Chip. Our group recently published an article on this workflow (PMID: 41610338). I am not asking you to cite our paper, but instead just providing the workflow as an example which can be found in other recent papers too.

We appreciate this constructive comment. We agree that including immune cell preparation improves the completeness of the review. As our primary focus is on cell incorporation strategies in OoC systems, we have added a concise overview of a representative workflow and included two relevant references, including one suggested by the reviewer. Please see lines 224-226 highlighted in yellow.

[1] D.Adnan, N. R.deBarros, L. S.Santovito, et al. “A Patient-Derived Organ-on-Chip Platform for Modeling the Tumor Microenvironment and Drug Responses in Pancreatic Cancer.” Advanced Science13, no. 14 (2026): e08934.

[2] Recaldin, T., Steinacher, L., Gjeta, B., Harter, M.F., Adam, L., Kromer, K., Mendes, M.P., Bellavista, M., Nikolaev, M., Lazzaroni, G. and Krese, R., 2024. Human organoids with an autologous tissue-resident immune compartment. Nature, 633(8028), pp.165-173.

 

Maybe helpful to discuss the mechanism of action of Immune Checkpoint inhibitors and how they act either on PD-1 or PD-L1 receptors for a more complete picture on immunotherapy. 

We thank the reviewer for this suggestion. We have added a concise discussion in the Introduction describing the mechanisms of major immunotherapy classes including checkpoint inhibitors targeting PD-1/PD-L1 pathways. Please see lines 43-50 highlighted in yellow.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper review provides a comprehensive and technically detailed of engineering strategies for developing immune‑competent OoC systems, focussed on cancer immunotherapy research. Authors have covered some platform architectures, immune‑cell incorporation methods, and both classical and emerging immunological readouts. Overall, the review is well‑structured, and relevant to the engineering and immuno‑oncology communities. The authors have integrateed engineering perspectives with immunological requirements. Particularly, the discussion of assay integration and sensors are topics of increasing importance, so it strengthens the paper interest. However, the manuscript requires the following revisions to address key gaps before it can be considered for publication:

• The most widely form used in scientific literature for organ‑on‑a‑chip is OoC. OOAC appears occasionally in other contexts, but it is uncommon and not preferred. So authors should change it or talk about microphysiological systems (MPS), which is also widely accepted.
• Although immune‑specific engineering principles are covered, they are not explained or detailed in engineering terms. This can be improved in the specific section
• Figure 1 mention about the use of deep learning algorithm, but I could not find any discussion about it. Authors should either include some discussion or remove it from the figure. 
• Manufacturing and operability of OoC are not discussed or stated. The paper talks about standards, but does not say which ones. Also no specification on the sensor integration, maintenance or usage in OoC
• Particularly, in Section 2, discussion on emerging OoC Architectures, such as membrane-less or devices without obstacles represent an important next stage in the engineering evolution of immune‑competent OoC systems. OoC architectures that avoids artifacts or synthetic membranes enables more realistic immune infiltration dynamics.
• The manuscript includes sections focused specifically on spheroid and organoid models, which fall outside the primary scope of engineering immune‑competent OoC systems. These parts should be out or reframed, as their distracts from the OoC focus.
• A table summarizing immune assays used in OoC for immunotherapy would be of high value for reader to scan rapidly what has been done. Specifications such as assay measurements, compatibility with OoC architectures, temporal resolution, label-free or destructive assay (important as some OoC models are truly expensive), suitability for specific immunotherapies (e.g. CAR‑T, NK, …). Going through all text might not be very practical for a quick look. So, a table really helps.
• A more appropriate title for section 6 (‘Outlook’) should be chosen.
• Regulatory is discussed briefly, and OoC system are nowadays not use only in research. Authors should elaborate (briefly) on regulatory expectations, qualification pathways and guidelines for translational acceptance.
• Personal opinion from authors or future trends should be improved and stated in the review.

Author Response

The most widely form used in scientific literature for organ‑on‑a‑chip is OoC. OOAC appears occasionally in other contexts, but it is uncommon and not preferred. So authors should change it or talk about microphysiological systems (MPS), which is also widely accepted.

We thank the reviewer for pointing this out. We have replaced “OOAC” with the more widely accepted term “OoC” throughout the manuscript.

 

Although immune‑specific engineering principles are covered, they are not explained or detailed in engineering terms. This can be improved in the specific section

We appreciate this comment. We have revised the manuscript to incorporate clearer engineering descriptions, including key engineering terms such as flow rate, shear stress, diffusion, and transport.

 

Figure 1 mention about the use of deep learning algorithm, but I could not find any discussion about it. Authors should either include some discussion or remove it from the figure. 

We thank the reviewer for this observation. We have added a new subsection (Section 6.5: “Data Integration and Computational Analysis”) to discuss the role of AI and data-driven approaches. Please see lines 718-731 highlighted in yellow.

 

Manufacturing and operability of OoC are not discussed or stated. The paper talks about standards, but does not say which ones. Also no specification on the sensor integration, maintenance or usage in OoC

We appreciate this insightful suggestion. Section 6.4 has been expanded and reframed to address manufacturability and system operation, while sensor integration has been extensively discussed in Section 6.2 and Table 2. Please see lines 705-717 and 601-687 highlighted in yellow.

Particularly, in Section 2, discussion on emerging OoC Architectures, such as membrane-less or devices without obstacles represent an important next stage in the engineering evolution of immune‑competent OoC systems. OoC architectures that avoids artifacts or synthetic membranes enables more realistic immune infiltration dynamics.

We thank the reviewer for this constructive comment. We have highlighted membrane-free and barrier-free architectures in Section 3, while maintaining Section 2’s focus on key historical milestones. Please see line 216 highlighted in yellow.

 

The manuscript includes sections focused specifically on spheroid and organoid models, which fall outside the primary scope of engineering immune‑competent OoC systems. These parts should be out or reframed, as their distracts from the OoC focus.

We agree with the reviewer and have removed the subsection on 3D spheroid co-culture models to maintain focus on OoC systems.

 

A table summarizing immune assays used in OoC for immunotherapy would be of high value for reader to scan rapidly what has been done. Specifications such as assay measurements, compatibility with OoC architectures, temporal resolution, label-free or destructive assay (important as some OoC models are truly expensive), suitability for specific immunotherapies (e.g. CAR‑T, NK, …). Going through all text might not be very practical for a quick look. So, a table really helps.

We thank the reviewer for this valuable suggestion. We have added Table 2 summarizing representative immunological assays, including measurement type and applicability to OoC platforms. Please see lines 367-368 highlighted in yellow.

 

A more appropriate title for section 6 (‘Outlook’) should be chosen.

We thank the reviewer for this suggestion. We have renamed Section 6 from “Outlook” to “Translational Considerations and Future Engineering Directions”. Please see line 576 highlighted in yellow.

 

Regulatory is discussed briefly, and OoC system are nowadays not use only in research. Authors should elaborate (briefly) on regulatory expectations, qualification pathways and guidelines for translational acceptance.

We appreciate this suggestion. Section 6.3 has been revised to include regulatory expectations, qualification pathways, and considerations for translational adoption. Please see lines 688-704 highlighted in yellow.

 

Personal opinion from authors or future trends should be improved and stated in the review.

We have integrated forward-looking perspectives and author insights into Section 6 “Translational Considerations and Future Engineering Directions”.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript presents a comprehensive overview of engineering strategies for developing immune-competent organ-on-a-chip (OOAC) systems for cancer immunotherapy research. The authors summarize platform architectures, immune cell incorporation strategies, and both classical and emerging assay modalities, with particular emphasis on integrated sensing approaches such as cytokine, oxygen, and impedance monitoring. The topic is timely and highly relevant, given the growing interest in microphysiological systems and new approach methodologies (NAMs) in translational immuno-oncology.

The manuscript itself is considered to be well structured and clearly written. The figures are generally informative, and the progression from architectural design to readout integration is logically organized. However, in its current form, the manuscript remains largely descriptive and does not yet fully achieve the depth expected of an engineering-focused review. Addressing the points below would enhance the rigor, impact, and translational relevance of the manuscript.

 

Specific Comments:

• Clarification of “Immune-Competent” Definition. Throughout the manuscript, the term “immune-competent OOAC” is used; however, it is not clearly defined. What constitutes minimal immune competence? (e.g., sustained immune viability, antigen-specific cytotoxicity, cytokine gradient preservation, trafficking capability)
• Whether the authors distinguish between bolus immune exposure versus recirculating systems?
• How chronic versus acute immune activation is handled in different architectures?
• Providing operational criteria or a short framework would improve conceptual clarity.
• The review describes architectural categories (vascular perfusion, 3D spheroids, compartmentalized systems), but does not sufficiently link platform design to specific immunotherapy mechanisms. For example, CAR-T therapy requires evaluation of trafficking under flow conditions, tumor infiltration through extracellular matrix barriers, and longitudinal monitoring of exhaustion-associated phenotypes. These distinct biological processes impose different architectural and assay requirements, which are not explicitly analyzed in the current manuscript.
• Similarly, immune checkpoint blockade primarily modulates T cell functional state rather than trafficking, and therefore demands platforms capable of resolving functional activation, cytokine dynamics, and exhaustion reprogramming over time. In contrast, bispecific antibodies depend critically on immune-tumor synapse formation under controlled effector-to-target ratios, necessitating microenvironmental control over spatial proximity and cellular stoichiometry. The manuscript would benefit from a clearer articulation of how these mechanistic differences translate into distinct architectural and assay design considerations.
• While transport limitations (e.g., diffusion constraints in thick constructs) are mentioned, the discussion remains largely qualitative. Given the engineering focus of the review, please explicitly address how diffusion-convection balance under perfusion influences immune-tumor interactions within OOAC platforms.
• How do gradient formation and Peclet number considerations shape cytokine distribution, immune activation, and spatial heterogeneity in these systems? A more quantitative discussion of transport regimes would strengthen the mechanistic interpretation.
• Additionally, how do shear stress ranges relevant to immune cell activation and dilution effects under continuous flow impact functional readouts such as cytokine measurements? Even a conceptual quantitative framework would elevate the manuscript beyond descriptive summarization.
• Section 4 (Fluorescence, ELISA, Flow Cytometry) reads largely as a general immunology methods overview. To strengthen this section, the authors should more explicitly discuss OOAC-specific limitations, such as PDMS adsorption of cytokines and small molecules, shear-induced alterations in cytokine profiles, and perturbation of phenotypes during single-cell extraction.
• The manuscript would benefit from discussion of tumor heterogeneity, including cold vs. hot tumors, stromal stiffness and immune exclusion, hypoxia-driven immune suppression beyond PD-L1 upregulation. This would align the engineering design discussion more closely with current immunotherapy challenges.
• Although NAMs are mentioned, the regulatory discussion remains general. The authors should clarify whether immune-competent OOAC platforms are being used in pre-IND contexts. Also, please discuss reproducibility, benchmarking, and inter-laboratory variability.
• The manuscript would benefit from inclusion of a schematic explicitly linking immune incorporation strategies to corresponding measurable endpoints.
• In addition, a figure highlighting key engineering tradeoffs (e.g., platform complexity vs. physiological fidelity vs. throughput) would enhance readers’ understanding of design strategy in immunotherapy-oriented OOAC development.
• The quality of Figure 6 is insufficient for clear interpretation. The text and graphical elements are difficult to read. Please provide a high-resolution version to ensure readability.

Comments for author File: Comments.pdf

Author Response

Clarification of “Immune-Competent” Definition. Throughout the manuscript, the term “immune-competent OOAC” is used; however, it is not clearly defined. What constitutes minimal immune competence? (e.g., sustained immune viability, antigen-specific cytotoxicity, cytokine gradient preservation, trafficking capability)

We thank the reviewer for this suggestion. A clear definition of “immune-competent” has been added to the manuscript. Please see lines 102-105 highlighted in yellow.

 

Whether the authors distinguish between bolus immune exposure versus recirculating systems?

We appreciate this point. We have clarified these delivery modes and included a dedicated discussion of perfusion-based systems in Section 3.1. Please see lines 214-215 highlighted in yellow.

 

How chronic versus acute immune activation is handled in different architectures?

We thank the reviewer for this constructive suggestion. We have incorporated this discussion in Section 6.2.1, highlighting how different OoC architectures support transient versus sustained immune activation. Please see lines 614-620 highlighted in yellow.

 

Providing operational criteria or a short framework would improve conceptual clarity.

We thank the reviewer for this suggestion. The manuscript has been organized with design considerations around three coupled engineering variables, delivery mode, compartment geometry, and barrier design, which together define the operational space of immune-competent OoC systems. Please see lines 214-216 highlighted in yellow.

 

The review describes architectural categories (vascular perfusion, 3D spheroids, compartmentalized systems), but does not sufficiently link platform design to specific immunotherapy mechanisms. For example, CAR-T therapy requires evaluation of trafficking under flow conditions, tumor infiltration through extracellular matrix barriers, and longitudinal monitoring of exhaustion-associated phenotypes. These distinct biological processes impose different architectural and assay requirements, which are not explicitly analyzed in the current manuscript.

We appreciate this important suggestion. A new subsection (Section 6.2.2) has been added to explicitly connect immunotherapy mechanisms with OoC design requirements. Please see lines 623-642 highlighted in yellow.

 

Similarly, immune checkpoint blockade primarily modulates T cell functional state rather than trafficking, and therefore demands platforms capable of resolving functional activation, cytokine dynamics, and exhaustion reprogramming over time. In contrast, bispecific antibodies depend critically on immune-tumor synapse formation under controlled effector-to-target ratios, necessitating microenvironmental control over spatial proximity and cellular stoichiometry. The manuscript would benefit from a clearer articulation of how these mechanistic differences translate into distinct architectural and assay design considerations.

We thank the reviewer for this suggestion. We have included a new subsection (Section 6.2.2) to discuss the relationship between immunotherapies and OoC requirements. Please see lines 623-642 highlighted in yellow.

 

While transport limitations (e.g., diffusion constraints in thick constructs) are mentioned, the discussion remains largely qualitative. Given the engineering focus of the review, please explicitly address how diffusion-convection balance under perfusion influences immune-tumor interactions within OOAC platforms.

We thank the reviewer for this comment. We acknowledge the importance of transport phenomena (e.g., diffusion–convection balance, Peclet number, shear stress). However, a detailed quantitative treatment would shift the focus toward transport theory, which is beyond the intended scope of this review. We instead maintain a conceptual discussion aligned with the review’s objectives.

 

How do gradient formation and Peclet number considerations shape cytokine distribution, immune activation, and spatial heterogeneity in these systems? A more quantitative discussion of transport regimes would strengthen the mechanistic interpretation.

We thank the reviewer for this suggestion. We acknowledge the importance of dimensionless analysis (e.g., Peclet number) in defining transport regimes in microfluidic systems. However, incorporating such quantitative frameworks would move the manuscript toward a transport-theory-oriented focus, which is beyond the intended scope of this review. We instead maintain a conceptual discussion aligned with the review’s objectives.

 

Additionally, how do shear stress ranges relevant to immune cell activation and dilution effects under continuous flow impact functional readouts such as cytokine measurements? Even a conceptual quantitative framework would elevate the manuscript beyond descriptive summarization.

We thank the reviewer for this suggestion. We acknowledge that shear stress and dilution effects under perfusion can influence immune cell behavior and cytokine measurements. However, a detailed quantitative framework describing shear-dependent activation or cytokine transport would require extensive discussion of fluid dynamics and cell mechanobiology, which is beyond the intended scope of this review. We instead maintain a conceptual discussion aligned with the review’s objectives.

 

Section 4 (Fluorescence, ELISA, Flow Cytometry) reads largely as a general immunology methods overview. To strengthen this section, the authors should more explicitly discuss OOAC-specific limitations, such as PDMS adsorption of cytokines and small molecules, shear-induced alterations in cytokine profiles, and perturbation of phenotypes during single-cell extraction.

We appreciate this suggestion. We agree that OoC-specific factors such as material adsorption, flow conditions, and sample processing can influence assay readouts. However, a detailed discussion of material-specific effects (e.g., PDMS adsorption) and flow-induced perturbations falls beyond the intended scope of this review, which focuses on assay selection and integration rather than device material science or fluid-mechanical effects.

 

The manuscript would benefit from discussion of tumor heterogeneity, including cold vs. hot tumors, stromal stiffness and immune exclusion, hypoxia-driven immune suppression beyond PD-L1 upregulation. This would align the engineering design discussion more closely with current immunotherapy challenges.

We thank the reviewer for this insightful comment. We have added a discussion in Section 6.2.1 addressing tumor heterogeneity, including immune dynamics and microenvironmental variability. Please see lines 603-607 highlighted in yellow.

 

Although NAMs are mentioned, the regulatory discussion remains general. The authors should clarify whether immune-competent OOAC platforms are being used in pre-IND contexts. Also, please discuss reproducibility, benchmarking, and inter-laboratory variability.

We thank the reviewer for this suggestion. We have revised Sections 6.1 (see lines 585-600) and 6.3 (see lines 689-704) to discuss reproducibility, benchmarking, and inter-laboratory variability. We have included that the OoC systems serve as complementary tools for preclinical evaluation, particularly in mechanistic studies and safety assessment (see lines 693-694).

 

The manuscript would benefit from inclusion of a schematic explicitly linking immune incorporation strategies to corresponding measurable endpoints.

We thank the reviewer for this suggestion. While we agree that such a schematic would be valuable, instead, we have included Table 2 to connect immunotherapies to relevant readouts. Please see lines 367-368 highlighted in yellow.

 

In addition, a figure highlighting key engineering tradeoffs (e.g., platform complexity vs. physiological fidelity vs. throughput) would enhance readers’ understanding of design strategy in immunotherapy-oriented OOAC development.

We appreciate the reviewer for this suggestion. While a dedicated figure illustrating engineering tradeoffs would be informative, we have instead revised Section 6.2 to include these tradeoffs in the text. Please see lines 601-687 highlighted in yellow.

 

The quality of Figure 6 is insufficient for clear interpretation. The text and graphical elements are difficult to read. Please provide a high-resolution version to ensure readability.

We thank the reviewer for this suggestion. We have replaced Figure 6 with a high-resolution version to ensure clarity and readability.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

After evaluating authors’ revised manuscript, I confirm that the previously raised comments have been addressed appropriately. No further changes are required at this stage. The manuscript is suitable for publication in its current form.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed well the points raised in the previous review stage.