Next Article in Journal
Feline Morbillivirus in Southern Italy: Epidemiology, Clinico-Pathological Features and Phylogenetic Analysis in Cats
Previous Article in Journal
Characterization of the Roles of SGT1/RAR1, EDS1/NDR1, NPR1, and NRC/ADR1/NRG1 in Sw-5b-Mediated Resistance to Tomato Spotted Wilt Virus
 
 
Article
Peer-Review Record

Effects of Pegylated Interferon Alpha and Ribavirin (pegIFN-α/RBV) Therapeutic Approach on Regulatory T Cells in HCV-Monoinfected and HCV/HIV-Coinfected Patients

Viruses 2021, 13(8), 1448; https://doi.org/10.3390/v13081448
by Kamil Grubczak 1,*, Anna Grzeszczuk 2, Monika Groth 3, Anna Hryniewicz 4, Anna Kretowska-Grunwald 1, Robert Flisiak 5 and Marcin Moniuszko 1,3,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Viruses 2021, 13(8), 1448; https://doi.org/10.3390/v13081448
Submission received: 17 June 2021 / Revised: 10 July 2021 / Accepted: 20 July 2021 / Published: 25 July 2021
(This article belongs to the Section Viral Immunology, Vaccines, and Antivirals)

Round 1

Reviewer 1 Report

Grubezak et al. reported that Peg-interferon plus ribavirin treatment induces significant effect on regulatory T cells in the course of HIV and/or HCV infection, indicating that these play a role in antiviral response. Manuscript seems well written.

 

  1. In DAA era, enhanced host immune status may play an additive role on HCV RNA clearance by DAA (Sasaki R, et al. J Med Virol. 2019 Mar;91(3):411-418. doi: 10.1002/jmv.25310.). Please discuss about the difference between interferon-free treatment and interferon-including treatment.
  2. Authors also used ribavirin. How the role of ribavirin? Authors should discuss more (Kanda T, et al. J Viral Hepat. 2004 Nov;11(6):479-87. doi: 10.1111/j.1365-2893.2004.00531.x.).

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 2 Report

In their submission, Effects of pegylated interferon alpha and ribavirin (pegIFN-α/RBV) therapeutic approach on regulatory T cells in HCV-monoinfected and HCV/HIV-coinfected patients, Grubczak et al. explore different Treg phenotypes in HCV monoinfected and HCV/HIV coinfected patients before and after treatment with pegIFN-a/RBV therapy. They found that pegIFN-a/RBV administration was associated with higher levels of Tregs only in HCV monoinfected patients. HCV/HIV coinfection seemed to inhibit such expansion of Tregs suggesting that HIV infection and/or HIV-ART treatment might contribute to this limited expansion.

I recommend that this manuscript be accepted after major revision.

Comments:

1) I recommend the authors to clarify in Material and Methods-Patients whether HCV/HIV coinfected patients were treated with HIV-ART treatment or not during the study and indicate which combined treatment was administered.

2) In Material and Methods-Patients, which was the limit of detection for plasma HIV RNA measurement?

3) In Material and Methods-Statistical analysis, the authors need to include how the correlation coefficients were determined.

4) In Results-Characteristic of the studied subjects based on immunological, biochemical and virological laboratory results, the authors point that HCV-infected patients seemed to have well-controlled HCV viral load, but a median value of 17.100 seems too high to be considered well-controlled. Did pegIFN-a/RBV administration result in lower HCV viral loads in both groups of patients? I recommend the authors to include extra columns in Supp. Tab. 1 and present immunological, biochemical and virological laboratory results after treatment as well.

5) In Results- Alterations in frequencies and absolute numbers of regulatory T cells in HCV-infected and HCV/HIV-coinfected patients in the course of anti-retroviral therapy, the authors mentioned “pegIFN-α/RBV administration in HCV-infected patients led to a significant decrease in total number of CD4+ T cells. In HCV/HIV-coinfected patients” but the p-value given is not significant, this should be noted as a slightly decrease or tendency.

6) In Results- Association between regulatory T cells and immunological, biochemical and virological laboratory parameters, in HCV-infected and HCV/HIV-coinfected patients. In order to support author’s conclusions, I recommend the authors to include as well associations found to be significant after pegIFN-α/RBV administration.

7) In Discussion, the authors need to include some discussion regarding the influence of HCV genotype on the differential response of Tregs to pegIFN-α/RBV administration. Were these results expected?

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have addressed all my concerns and therefore I support publication without further changes.

Back to TopTop