PREreview of Tumor-specific antibodies elicited by engineered bacteria promote bladder cancer immunotherapy

Brief summary of the study - a sentence summarizing the study and general comments that apply across the full paper

• The study demonstrates that engineered E. coli Nissle 1997, which expresses the human cytokine CXCL13, significantly enhances the efficacy of PD-1 blockade in bladder cancer by promoting tumor-specific antibody production and fosters adaptive immunity by elevating germline complex formation. 

• The study also demonstrated the safety and efficacy of its combination approach in orthotopic mice models of bladder cancer. It has leveraged the adaptive immune response and improved protection upon tumor rechallenge. The paper stresses the importance of localized immune modulation.

Major comments  - Comments on the validity or strength of the methodology, experiments and analyses, strength of the conclusions

• The study presents statistical design and analysis concerns. 

  1. The determination of sample size lacks a formal power analysis, which is particularly critical for survival studies. 

  2. The choice of statistical tests appears appropriate, but the rationale for selecting specific tests needs further clarification. 

  3. Survival analysis would benefit from the inclusion of hazard ratio calculations and testing of the proportional hazards assumption.

• The rationale for selecting CXCL13-expressing EcN for testing is not sufficiently clear. Although CXCL13 may have therapeutic implications as a chemokine, previous studies have also explored CXCL10 (PMID 38720149), blockage of CXCL12 (PMID 38001512), and CX3CL1 (PMID 37771579). Therefore, the choice of CXCL13 should be discussed in more detail. 

• Furthermore, the manuscript should provide a comprehensive discussion of its advancements. Although the authors cite previous publications on engineered EcN, they do not adequately compare these studies to their approach. For example, EcN has been engineered to increase intratumoral L-arginine concentrations to synergize with PD-L1 blockade (Canale et al., Nature 2021) and to release nanobodies targeting PD-L1 and CTLA-4 simultaneously (Gurbatri et al., Science Translational Medicine 2020). There has even been research on the expression of chemokine CXCL16 by engineered bacteria (Savage et al., Sci Adv 2023). While none of these have been tested in bladder cancer before, without a comparison to these studies, it remains unclear whether the only novelty and therapeutic benefit of the presented EcN-CXCL13 lies in its application to bladder cancer.

• In their MB49 model, the authors do not observe a therapeutic benefit from using EcN-CXCL13 alone; however, they observe tumoral immune microenvironment reorganization when treating an UPPL1541 model. Do they observe the same in the MB49 model? And is there a survival benefit in the UPPL1541 model? Extending the analysis of the second model could demonstrate that the authors’ approach is not only relevant in one specific bladder cancer model but also likely applicable to various models, thereby enhancing its therapeutic relevance.

Minor comments - Clarifications to statements in the text, interpretation of the results, presentation of the data/figures

• We have difficulties drawing conclusions from the rechallenge experiment (Fig. 2E, F). Comparing previously unchallenged mice as controls to mice that have been “cured” of bladder cancer does not seem like an appropriate control, as mice that cleared the tumor would of course naturally have developed immunity to protect against rechallenge, unlike the naive mice, regardless of their previous treatment. This issue is further illustrated in the single treatment groups, where a high proportion of mice cleared the tumor; however, the limited sample size of only three mice for single treatment hampers interpretation. Therefore, I suggest slightly rephrasing the conclusions from this rechallenge experiment for clarity.

• While combining EcN-CXCL13 with PD-1 inhibition does increase mouse survival, the observed effect on survival appears to be relatively small. Additionally, when investigating T follicular helper (Tfh) cells, the addition of EcN does not significantly increase Tfh cells number. Can the authors elaborate on the reasons for this lack of increase and discuss whether further improvements are necessary to make this approach clinically feasible?

• Providing gating strategies for the flow cytometry data in the supplementary could improve interpretation of the results

• The drug used for PD-1 blockade was not identified

• The authors mentioned that they used wild-type C57BL/6NJ. Please explain the characters of this strain to be chosen for the study or add a reference

• Please explain why the difference in the sex of the animals used in Intravesical tumor implantation (female) & Generation of bone marrow chimeric mice (male)

Comments on reporting - information on the statistical analyses or availability of data.

• Statistical methods section requires:

  • Detailed description of randomization procedures

  • Specifics of multiple comparison corrections

  • Software and packages used for analysis

• Raw data availability statement needed

• Quality control metrics for key assays should be reported

• In the abstract section, a structured summary of the study that includes the background, the aim of the work, the methods, the results, and the conclusion should be illustrated

• A keyword section should be added after the abstract section.

• The summary and conclusion section of the study should be added

Suggestions for future studies

• Investigate the detailed mechanism by which CXCL13 influences immune cell recruitment in tumor environment

• Combination of probiotics with other immune checkpoint inhibitors 

Conflicts of interest of reviewers

• There are no conflicts of interest to declare.

Inline commenting section

Please add comments on the preprint below via comments. You can add comments on the full paper, sections or only individual fragments. Any comments added here will be reviewed for inclusion in the public review section if relevant, but will not be posted publicly in any way that can identify the commenter for individual comments.

• Result. “To effectively deliver the human chemokine CXCL13 (hCXCL13) to the bladder TME, we engineered a well-characterized probiotic strain, Escherichia coli Nissle 1917 (EcN)”

  • This section needs significant improvement:

    • The stability of the SLIC system in vivo needs to be better characterized

    • Quantitative data showing the expression levels and kinetics of CXCL13 release in vivo is missing

    • The authors should provide data on plasmid retention rates over multiple bacterial generations

    • Controls using non-SLIC versions of the construct would strengthen the findings

• Result. “First, we confirmed that release of hCXCL13 was SLIC dependent and EcNhCXCL13 demonstrated no difference in growth rate relative to EcNWT in non-lysing conditions ”

  • In the material and methods section the strains used are EcNhCXCL13 and EcN– not EcNWT. please revise

• Result. “Compared to control EcN lysate, human B cells significantly migrated in response to lysates of both EcNhCXCL13 and EcNmCXCL13 (fig.S1D)”

  • The generation of this strain and its exclusion were not discussed in the materials and methods section

• Supplementary Figure 1. (b). two-tailed unpaired Mann Whitney test)

  • This test was not mentioned in the statistical analysis (the test of normality of data was not mentioned in the statistical analysis)

• Figure 2. “No tumor engraftment was observed in surviving mice previously treated with EcNslichCXCL13 plus PD-1 blockade (n=0/7) (right panel).”

  • While this is an impressive result, it's interesting that there was complete protection (0/7) rather than a partial response which would be more typically expected in immunotherapy studies. Could environmental or housing conditions have influenced this unusually strong protection? Were the rechallenge sites pre-conditioned in any way by previous bacterial treatment? Some discussion of potential confounding factors would strengthen the interpretation of this striking result.

• Result. “ The SU-DHL-4 cell line was cultured in complete RPMI supplemented with 20% FCS and 10% DMSO. Cultures were maintained within a humidified 5% O2/CO2 atmosphere at 37°C inside an incubator. Cells were split twice per week, and cell viability was measured using trypan blue staining.”

  • Please identify this media as Roswell Park Memorial Institute (RPMI) 1640 Medium, fetal calf serum (FCS), and dimethyl sulfoxide (DMSO)

• Result. “CXCL13-expressing vectors were transformed into electrocompetent EcN-SLIC strains and cultured in LB media with 50 μg/ml kanamycin with 0.2% glucose, in a 37°C shaking incubator. ”

  • Please identify this as Lysogeny broth (LB)

• Result. “Bacillus Calmette-Guerin (BCG, TICE strain, Merck) was obtained by resuspension in PBS of the lyophilized content of clinically available MSD vials”

  • please explain this abbreviation (MSD vials)

• Methodology. Generation of bone marrow chimeric mice

  • The antibody analysis needs strengthening:

    • Detailed characterization of antibody isotypes

    • Specificity controls using other cancer cell lines

    • Functional assays to demonstrate antibody-dependent cell cytotoxicity

    • Time course of antibody development

• Methodology. Fluorescent microscopy

  • It would be great to add the settings used for imaging, like wavelength, exposure time.

• Methodology. Detection of antitumor antibodies

  • antibodies bought from (company), good to write all details to bring reproducibility in results for readers and followers. Specifically, the line “fluorescently labelled antibodies to mouse IgG, IgA and IgM ”

Competing interests

The authors declare that they have no competing interests.