Round 1

Reviewer 1 Report

This excellent work explores NK-92 directed by a specific CAR against PSMA in prostate cancer models. The authors show that the CAR-NK-92 cells have a high and specific lytic activity against PSMA positive prostate cancer cells. This effect is not reduced by sublethal irradiation used as a safety measure. The CAR-NK-92 cells also had a very convincing effect in in vivo prostate cancer models, even after systemic application of the CAR-NK-92 cells.

All results are sound and the conclusions drawn are convincing. The linguistic style is very good, making the manuscript easy to read and understand. This work will be the basis of further research, and most likely even of a phase I clinical trial. The manuscript will be of high interest to many readers. In summary, I strongly support publication.

The following issues however should be improved:

1. Please discuss, which indications might be tested in a future phase I clinical trial: local therapy refractory prostate cancer? Metastatic prostate Cancer?
2. The formatting of citation 41 is not correct, please change.
3. Please discuss the CAR2BRAIN Trial, which explores the use of CAR-NK-92 cells in patients suffering from glioblastoma: 

   CAR-Engineered NK Cells for the Treatment of Glioblastoma: Turning Innate Effectors Into Precision Tools for Cancer Immunotherapy

   Michael C. Burger,^1,2,3,4 Congcong Zhang,^3,4,5 Patrick N. Harter,^2,3,4,6 Annette Romanski,⁷ Florian Strassheimer,^1,2 Christian Senft,^2,3,8 Torsten Tonn,^4,9,10,11 Joachim P. Steinbach,^1,2,3,4 and Winfried S. Wels^2,3,4,5,*

Author Response

Reviewer #1:

The following issues however should be improved:

1. Please discuss, which indications might be tested in a future phase I clinical trial: local therapy refractory prostate cancer? Metastatic prostate Cancer?

 

Reply:  Experimentally, we proposed a series of increasingly more complex approaches to provide clear-cut evidence about the feasibility of CAR-NK-92 use even for the most demanding metastatic setting.  Since local therapy still relies on different clinical therapeutic approaches, we envisage that a phase I clinical trial may potentially involve patients with advanced and metastatic cancer.  We have improved our discussion by the insertion of the following considerations (pag.12 see the revised version of the paper): “In the case of the castration-resistant PCa and in its metastatic form, actually no curative interventions are available. So our main goal is to tackle the growth of the castration-resistant PCa and of the metastatic disease.”

2. The formatting of citation 41 is not correct, please change. (Done)
3. Please discuss the CAR2BRAIN Trial, which explores the use of CAR-NK-92 cells in patients suffering from glioblastoma:

“CAR-Engineered NK Cells for the Treatment of Glioblastoma: Turning Innate Effectors Into Precision Tools for Cancer Immunotherapy.”

Michael C. Burger,1,2,3,4 Congcong Zhang,3,4,5 Patrick N. Harter,2,3,4,6 Annette Romanski,7 Florian Strassheimer,1,2 Christian Senft,2,3,8 Torsten Tonn,4,9,10,11 Joachim P. Steinbach,1,2,3,4 and Winfried S. Wels2,3,4,5,*

Reply:     So we have improved our discussion by the insertion of the following considerations (pag.12 see the revised version of the paper): “Regarding Glioblastoma (GB), where patients’ median survival remains less than 1 year, despite multimodal regimes, a phase I clinical trial “Intracranial Injection of NK-92/5.28.z Cells in Patients With Recurrent HER2-positive Glioblastoma” (i.e. CAR2BRAIN,NCT03383978, clinicaltriaols.gov) was started in 2017, in order to evaluate the feasibility, safety and tolerability of the direct injection of irradiated NK-92 CAR recognizing HER2 antigen into the wall of the resection cavity [40]. In this case the local therapy approach was chosen to increase the accumulation of the engineered cells at the site of the lesion overcoming the problems associated with the Blood Brain Barrier passage.” Moreover, as suggested the reference “Michael C. Burger,1,2,3,4 Congcong Zhang,3,4,5 Patrick N. Harter,2,3,4,6 Annette Romanski,7 Florian Strassheimer,1,2 Christian Senft,2,3,8 Torsten Tonn,4,9,10,11 Joachim P. Steinbach,1,2,3,4 and Winfried S. Wels2,3,4,5,*” was inserted as REF.40

Author Response File: Author Response.pdf

Reviewer 2 Report

The work is very well written and addresses a topic of great interest due to the high incidence of prostate cancer among the male population in western countries. The therapeutic method developed by the group has great potential for clinical application, either in the current format or through the inclusion of suicide genes that facilitate clinical management of patients and reduce the number of infusions of the transformed cells. I strongly encourage the authors to develop that second part and run the appropriate trials with the safest suicide genes.

Author Response

The work is very well written and addresses a topic of great interest due to the high incidence of prostate cancer among the male population in western countries. The therapeutic method developed by the group has great potential for clinical application, either in the current format or through the inclusion of suicide genes that facilitate clinical management of patients and reduce the number of infusions of the transformed cells. I strongly encourage the authors to develop that second part and run the appropriate trials with the safest suicide genes.

Reply: We thank the Reviewer for supportive comments. We agree with his/her suggestion, and actually we are already pursuing the development of new vectors coding for both the CAR and a suicide gene.

Author Response File: Author Response.pdf

Reviewer 3 Report

Authors present a high-quality and very well-written experimental manuscript that describes anti-PSMA CAR-engineered NK-92 cells as a prospective off-the-shelf cell therapy for prostate cancer.

Authors used a wide range of experimental methods that include generation of CAR-expressing NK-92 cells, irradiation of NK-92 cells, immunoblotting, ELISA cytokine release assay, degranulation assay, cytotoxicity Cr-release assay and in vivo studies using xenograft mice.

 

Authors demonstrated that:
- NK-92/CAR cells are able to specifically and effectively lyse PSMA-expressing PCa cells, in contrast to parental NK-92 cells.
- adoptive transfer of NK-92/CAR cells reduced tumor growth in different PCa mouse tumor models, and significantly enhanced survival.

Based on these results they conclude that:
- PSMA-targeted NK-92/CAR cells acquire antigen-specific cytotoxic activity and can be irradiated as a safety measure to prevent cell proliferation in the patient.
- Irradiated NK-92/CAR cells undergo degranulation and IFN-γ release upon interaction with PSMA-expressing targets and retain a high and specific target killing activity.
- CAR-mediated cytotoxicity is independent on NKG2D activity in NK-92/CAR cells.
- NK-92/CAR cells control the growth of locally implanted prostate carcinoma.
- The adoptive transfer of NK-92/CAR cells is therapeutic in an orthotopic and metastatic prostate tumor model

 

Questions:
- Authors mention that 5-10 Gy γ-irradiation of NK-92/CAR cells resulted in their eradication within 5-7 days. The long-term safety concern of this procedure is clear. However, how this might affect the efficacy? Are 5-7 days enough to obtain the required therapeutic effect?
- Data in Fig 4A, 4B suggests that K562 and PC3 are primarily targeted via NKG2D-mediated mechanism. Whereas, LNCaP and PC3-PSMA are targeted mostly via anti-PSMA CAR. Any explanation for such difference?
-  CAR-T cells often fail to penetrate the solid tutor stroma in in vivo experiments. Any potential explanation why NK-92/CAR cells might be a better choice for targeting solid tumors?
- How did multiple NK-92/CAR injections affect cell persistence in your study?
- A general question: is it correct to expect that for autologous NK-92/CAR cells the intrinsic NK receptor-mediated cytotoxicity will be silenced by MHC in patients, whereas CAR-mediated cytotoxicity will remain?
- Do the same signalling pathways trigger cytotoxic degranulation for NKG2D- and CAR-mediated mechanisms?

 

Other comments:
- Typo. Section 3.5 “independent oN NKG2D” ->“independent oF NKG2D”.
- Authors are kindly requested to cite the following article (DOI 10.3390/cancers12010125) that reviews the use of CAR-T cells against solid tumors. 

 

Overall, the manuscript is highly valuable for the CAR-T scientific community and should be accepted for publication.

Author Response

Reviewer #3:

Questions:

- Authors mention that 5-10 Gy γ-irradiation of NK-92/CAR cells resulted in their eradication within 5-7 days. The long-term safety concern of this procedure is clear. However, how this might affect the efficacy? Are 5-7 days enough to obtain the required therapeutic effect?

Reply:    NK-92 cells are a tumor cell line and regulatory agencies require that they are lethally irradiated before infusion in humans, a safety method employed in the clinical setting to block proliferation and prevent their permanent engraftment in transferred patients. While the approach has minimal effects on their short-term functionality, it negatively impacts on their in vivo persistence and long-term antitumor efficacy. Hence, repeated infusions of irradiated NK-92 cells are required as a means to circumvent their limited survival. This has been repeatedly reported both in preclinical and clinical setting, as clearly stated in the manuscript (please, see citations [9-11], [36, 37],[39, 40]). Therefore, we adopted a therapeutic protocol involving multiple treatments. We are aware that this approach may appear more complex as compared to a single infusion of anti-CD19 CAR-T cells (which, however, have to be manufactured for any single different patient). On the other hand, this aspect is positively counterbalanced by the fact that NK-92 cells are a continuously growing cell line that, upon CAR transduction, can be propagated indefinitely and at low cost, and is suitable for treatment of potentially all patients carrying the same cancer. Thus, the loss of functionality after irradiation is not a major matter for clinical application (as already demonstrated) because it can be replaced by repeated infusions. Additionally, current approaches are pursuing the modification of NK-92 cells with suicide genes, which in principle could guarantee long-term survival and efficacy, and at the same time the possibility to switch-off the population when required.

- Data in Fig 4A, 4B suggests that K562 and PC3 are primarily targeted via NKG2D-mediated mechanism. Whereas, LNCaP and PC3-PSMA are targeted mostly via anti-PSMA CAR. Any explanation for such difference?

Reply:  This is an interesting observation. Apart K562 that is a classical target of NK activity, whether we consider solid tumors (e.g., PC3) the level of NK-92 cell cytotoxicity is not very prominent. Therefore, blocking NKG2D very likely completely abolishes such effect. On the other hand, the functional activity provided by the CAR leads to an enormously higher cytotoxicity when the specific target is encountered; likely, such lytic activity, being already maximal, is poorly affected by NKGD2 blocking.

-  CAR-T cells often fail to penetrate the solid tumor stroma in in vivo experiments. Any potential explanation why NK-92/CAR cells might be a better choice for targeting solid tumors?

Reply:  There are no evidences that NK-92 CAR cells penetrate the solid tumors better than CAR-T cells, however in our work, as in other preclinical settings (please see citations [36,37]), we have shown that specific recognition of tumor cells and antitumor activity of NK-92/CAR cells were retained in vivo, with a reduction of the tumor size and more importantly with a significant enhance of the survival period of the mice treated. Very probably NK-92/CAR cells might be a better choice for targeting solid tumors than CAR-T cells for their ability to bypass the immunosuppressive effects of the tumor microenvironment which is a major issue in case of solid tumors. Indeed the inhibitory receptors are expressed at low levels on the surface of NK-92 cells making them more resistant to solid tumors than other immune cells. Moreover even if the targeted antigen on the tumor is rapidly lost, the CAR-NK-92 cells can still be activated by their activating receptors. Last but not least NK-92 cells are immortalized cell lines with a uniform phenotype, allowing them to be cultured in vitro for a long time to proliferate which can compensate for the decline in immune cell viability in patients with advanced cancer, in fact,  depending on the type of tumor, corresponding CAR-NK-92 cells can be directly expanded for treatment, thereby shortening the treatment cycle and reducing the cost of CAR therapy with the possibility to treat the patients with repeated infusions.

- How did multiple NK-92/CAR injections affect cell persistence in your study?

Reply:   This point is strictly connected to point n. 1. If you substitute persistence with efficacy, similar consideration can be advanced. In other words, irradiation must be performed for safety reasons and this affects both long-term functionality and persistence. Either aspects can be overcome by providing the system with multiple infusions of irradiated effectors.

- A general question: is it correct to expect that for autologous NK-92/CAR cells the intrinsic NK receptor-mediated cytotoxicity will be silenced by MHC in patients, whereas CAR-mediated cytotoxicity will remain?

Reply:     NK-92/CAR cells are not produced from autologous NK cells. NK-92 is a human tumor cell line originally established from a patient, which has been subsequently validated in GMP conditions. NK-92 cells express a series of activating receptors and lack almost all inhibitory killing receptors, thus not being affected by silencing activity of recipient’s MHC molecules.

- Do the same signalling pathways trigger cytotoxic degranulation for NKG2D- and CAR-mediated mechanisms?

Reply:      NKG2D receptors usually recognize stress-induced ligands expressed on tumor cells under the pressure of immune cells or long-term treatment during tumor evolution.  The specific binding of NKG2D to the highly expressed MICA, MICB, and ULBPs ligands on tumor cell surface can trigger multiple signaling pathways, such as phosphatidylinositol 3-hydroxy kinase (PI3K), phospholipase C Gamma 2 (PLCγ2), c-Jun-NH(2)-terminal kinase (JNK), and others. On the other hand, CAR constructs have an intracellular CD3ζ domain, which contains three ITAM motifs, and the co-stimulatory motif CD28. Ligation of the CAR to tumor antigen provokes aggregation of the CAR polypeptide. CD3ζ‐mediated activation proceeds via classical Zap70‐mediated pathway resulting in Ca++ influx and release from intracellular stores, and translocation of de‐phosphorylated NFAT to the nucleus. Costimulatory domains preferentially recruit PI3 kinase and TRAF to enhance cytokine and cell survival gene transcription, particularly through AP‐1 and NF‐κB translocation to the nucleus. Signalling by the costimulatory domain activates cytoskeletal mobilisation, enabling colocalisation of CAR to membrane rafts. In conclusion, both NKG2D and CAR signalling patways lead to cytotoxic granule exocytosis but employ partially different mechanisms.

Other comments:

- Typo. Section 3.5 “independent oN NKG2D” -> “independent oF NKG2D”. (done)

- Authors are kindly requested to cite the following article (DOI 10.3390/cancers12010125) that reviews the use of CAR-T cells against solid tumors. (done-Ref.#4)

Author Response File: Author Response.pdf