Humoral and cellular immune responses to COVID-19 mRNA vaccines in immunosuppressed liver transplant recipients

Background Liver transplant recipients (LTRs) are at a high risk of severe COVID-19 owing to immunosuppression and comorbidities. LTRs are less responsive to mRNA vaccines than healthy donors (HDs) or other immunosuppressed patients. However, the disruption mechanism in humoral and cellular immune memory responses is unclear. Methods We longitudinally collected peripheral blood mononuclear cells and plasma samples from HDs (n = 44) and LTRs (n = 54) who received BNT162b2 or mRNA-1273 vaccines. We measured the levels of anti-receptor-binding domain (RBD) antibodies and spike-specific CD4+ and CD8+ T-cell responses. Results Here, we show that the induction of anti-RBD IgG was weaker in LTRs than in HDs. The use of multiple immunosuppressive drugs is associated with lower antibody titers than only calcineurin inhibitor, and limits the induction of CD4+ T-cell responses. However, spike-specific CD4+ T-cell and antibody responses improved with a third vaccination. Furthermore, mRNA vaccine-induced spike-specific CD8+ T cells are quantitatively, but not qualitatively, limited to LTRs. Both CD4+ and CD8+ T cells react to omicron sublineages, regardless of the presence in HDs or LTRs. However, there is no boosting effect of spike-specific memory CD8+ T-cell responses after a third vaccination in HDs or LTRs. Conclusions The third mRNA vaccination improves both humoral responses and spike-specific CD4+ T-cell responses in LTRs but provides no booster effect for spike-specific memory CD8+ T-cell responses. A third mRNA vaccination could be helpful in LTRs to prevent severe COVID-19, although further investigation is required to elicit CD8+ T-cell responses in LTRs and HDs.

The time between vaccinations (1,2 and 3rd dose) is not mentioned anywhere in the manuscript.It may be possible that individuals who receive early booster (shorter gap between 2nd and 3rd dose) after 2nd dose have a high response.Or did everybody in LTR and HD receive all vaccinations at the same time?I assume everyone was a homologous recipient of vaccination.Future studies could evaluate impact of heterologous vaccination, which has been reported to be more beneficial.
The authors report lack of CD8 response after booter.Figure 1E/H: The authors mention that anti-RBD and pVNT50 in LTR reach comparable levels with HD after 3rd dose.(1 month after 2nd dose in the healthy group vs. 1 month after 3rd dose in LTR group.)The endpoint are different.What was the comparison at 1 month post 3rd dose among these groups?I may have missed it probably.
Line 259/260 authors claimed that "Furthermore, Th1/Th2 ratio in LTRs was significantly lower compared to HDs (Figure 2G), suggesting that LTRs are more susceptible to the induction of Th2biased CD4+T-cell responses".In general, LTRs with immune suppressive drugs usually have Th2 lineage and more Th2 cytokines compare to Th1.
Throughout the study authors have compared all the parameters between healthy and LTRs (CNI/CNI+other drugs) at different time points.This is important although it would have been interesting to learn the comparison between two groups CNI alone (n=23) vs. CNI +other drugs (n=31).Data can be analysed within the same group at different time (5 time points) points with statistical analysis.Also, mention this n of each group somewhere in the results.i.e., 23 vs. 31 vs. 44.Similarly, I don't see a number (sample size) anywhere for all sub-group anaylsis.For example: comparing LDLT vs. DDLT; recipients 12 years after and less.Why this 12 years was chosen?MMF and mTORi have been reported to affect the immune response and it is suggested to stop this drug in infected individuals (PMID: 32750442; PMID: 36374707).Although this hypothesis is contentious it would be interesting to know if there was any differential expression of immune response in those on MMF/mTORi.Did you analyse Th1/Th2 ratio between LTR and HD pre and post which would have added more value to the current study.CD 4 and 8 naïve cells in the marrow of LTR are unaffected, while the CD4 effector memory and CD4 and 8 effector cells are lower in LTRs which has been reported in the literature.Similarly, the authors report that after booster dose CM changes to EM.However, on Page no 24, Figure 3, data for CD8 memory cells of 3rd post 1-month vaccination is missing both.Did the authors perform the analysis or missed?
There are several panels (A-H).Please number them sequentially in the manuscript in bold letters.Line 289: over-r is missing

Comments:
Elisa's assay is commercial?How was the cut-off established?Reference 20 is associated with it, I think that is not correct.

Reply:
The ELISA used in this study targets only the RBD region of spike protein.Therefore, instead of using commercial products, we constructed in-house ELISA.We adopted a value that added twice the standard deviation to the average OD value of plasma from unvaccinated individuals used as negative controls as the cutoff.

Revision (Page 6, Lines 144-146):
The cutoff value of OD = 0.3 was determined based on the OD values of plasma from unvaccinated individuals used as a negative control, specifically by adding twice the standard deviation to the average OD value.

How much time passed between the blood sample of the 6 months post-2nd dose and the 3rd dose of vaccine?
Reply: Thank you for this comment.The third vaccination for liver transplant recipients was administered a median of 41 days with an interquartile range (IQR) of 32.5-57.75days from 6 months after the second vaccination.The shortest interval was 5 days, and the longest was 87 days.For healthy individuals, the median was 42.5 days with an IQR of 42.5-79.5days.The shortest interval was 4 days, and the longest was 102 days.We illustrated the relationship between the days from 6 months after 2nd vaccination to 3rd vaccination and the plasma anti-RBD antibody titer one month after the third vaccination in a graph.As the interval between vaccine doses increased, there was a slight tendency for the plasma anti-RBD antibody titer to decrease, but the difference was not statistically significant.
Therefore, we determined that the difference in interval days did not have a significant impact on our analysis.

What dose of steroids were the patients receiving?
Reply: The patients were taking a median dose of 5 mg/day of steroids (range: 0.5 -10).Specifically, one patient was taking 0.5 mg/day, six patients were taking 2.5 mg/day, nine patients were taking 5 mg/day, and one patient was taking 10 mg/day.We plotted the steroid dose against the antibody titer and CD4, CD8 T-cell responses one month after the third vaccination on graphs.The antibody titer showed a significant negative correlation with the steroid dose, but no significant correlation was observed for CD4 and CD8 T-cell responses.In a previous study analyzing T-cell responses after mRNA vaccine administration in patients with systemic autoimmune rheumatism, it was shown that those taking a high dose of steroid (>10 mg/day) had significantly weaker CD4 T-cell responses compared to those not taking or those taking a low dose of steroid (< 10 mg/day) (Maliah et al, Sci. rep., 2022).The doses taken by the patients in our study did not exceed 10 mg/day, and the lack of effect on T-cell responses is consistent with the previous study.We have uploaded the graphs as Supplementary Figure 7.
Revision (Supplementary Figure 7 and Page 18, Lines 446-448): Furthermore, the median steroid dose administered to the liver transplant recipients in this study was 5 mg/day (range: 0.5-10 mg/day).Within this range, an effect was observed on the antibody titer, but no effect on the T-cell response was noted (Supplementary Fig. 7A).

Reviewer # 1 (
Remarks to the Author): Dear Dr. Yamamoto Thank you for submitting your paper.Nogimori et al. reported an prospective study on the impact of mRNA SarsCov2 vaccination in a cohort of liver transplant recipients.They assessed the humoral and cellular biological response to vaccination, after the second and third dose of vaccine.