All patients who received a kidney transplant between April 2021 and March 2022 at Imperial College Renal and Transplant Centre London were assessed for inclusion. Patients were eligible for participation if they had received 2 or 3 doses of a SARS-CoV-2 vaccine pre-transplantation, Figure 1 . During the study period, it was routine clinical practice for patients to have SARS-CoV-2 serological assessment at the time of transplantation. Additional assessments were also undertaken following booster doses of vaccine. For the purposes of this report, only serological assessments performed between days 10-62 post-boost were considered. The median follow up of the study patients post-transplant was 11 (8–14) months. A cohort of haemodialysis patients receiving vaccination whilst awaiting a kidney transplant were used as controls, Supplemental Information, Table S1 . Clinical and vaccine data were obtained from electronic patient records and vaccine database, respectively. The study was approved by the Health Research Authority, Research Ethics Committee (Reference: 20/WA/0123).

The immunosuppression protocol at our centre consists of monoclonal antibody induction for all patients with maintenance tacrolimus together with a steroid minimisation protocol. Patients receiving long-term corticosteroid treatment at the time of transplant are continued on their established dose post-transplant. In addition, maintenance mycophenolate mofetil (MMF) is prescribed for patients who received basiliximab or who are highly HLA sensitised, with a calculated reaction frequency of ≥85%.

Infection was confirmed through viral detection via reverse-transcriptase polymerase chain reaction (RT-PCR) assays, or by positive lateral flow antigen tests from April 2022. Infection outcome data was obtained from centrally collected data held by the National Health Service Blood and Transplant (NHSBT) Service via The UK Health Security Agency (UKHSA). Prior infection was defined by a history of viral detection or via serological assessment; nucleocapsid protein antibodies (anti-NP) at any time, or the presence of spike protein antibodies (anti-S) pre-vaccination. Anti-NP was tested using the Abbott Architect SARS-CoV-2 IgG 2 step chemiluminescent immunoassay (CMIA). Samples were interpreted as positive or negative with a threshold index value of 1.4. Anti-S IgG concentrations ([anti-S]) were assessed using the Abbott Architect SARS-CoV-2 IgG Quant II CMIA. A quantitative assay with a threshold value of 7.1 BAU/ml for positivity, and an upper level of detection of 5680 BAU/ml. Boosted status was defined as a post-vaccination level ≥700 BAU/ml, based on arbitrary internal observations of lower-level limits in healthy populations. Patients who were diagnosed with infection post-vaccination but prior to serological testing, were considered as infection-experienced.

Treatment of COVID-19 infection was clinically based, and aligned with the UK Chief Medical Officer’s Interim Clinical Commissioning Policies (11). Prior to 20^th December 2021, all treatment was in-patient based, and the monoclonal antibody (mAb) utilised in algorithms was casirivimab plus imdevimab (Ronapreve^®) in seronegative patients (2.4g). From 20^th December 2021, the mAb sotrovimab (Xevudy^®) at 500mg doses, was available for people considered at highest risk of progression to severe disease in the community. Other outpatient treatments available during this period for this population included remdesivir, whilst for inpatients dexamethasone and tocilizumab were also considered.

Statistical analysis was conducted using Prism 9.0 (GraphPad Software Inc., San Diego, California). Unless otherwise stated, all data are reported as median with interquartile range (IQR). The Chi-squared test was used for proportional assessments. The Mann-Whitney and Kruskal-Wallis tests were used to assess the difference between 2 or >2 groups, with Dunn’s post-hoc test to compare individual groups. Comparison of paired samples was assessed using the Wilcoxon test. Survival analysis was assessed using the log-rank test. A p value of <0.05 was deemed statistically significant.

Post-V3 serological assessment was undertaken in 95 V2-group patients who received a third vaccine post-transplantation (PT-V3), 66(69.5%) and 29(30.5%) were infection-naïve and infection-experienced respectively. The median time to PT-V3 was 116(78-150) days post-transplant. Seventeen (25.8%) infection-naïve patients and 22 (75.6%) infection-experienced patients reached the definition of boosted status following PT-V3, p<0.0001. Median [anti-S] PT-V3 was 166(38-720) and 2207(684-5586) BAU/ml in the infection-naïve and infection-experienced individuals respectively, p<0.0001. Infection-naïve patients who had been primed with ChAdOx1 had lower [anti-S] compared with those who have received BNT162b2, following a PT-V3, with median [anti-S] of 77 (38-178) BAU/ml and 315 (194-955) respectively, p=0.015.

A paired comparison of pre-transplant and PT-V3 [anti-S] was undertaken in 60 infection-naïve V2-group patients. There was no difference in median [anti-S] pre-transplant and PT-V3, at 122(32-574) and 111(34-682) BAU/ml respectively, with a median difference of -1 (-182-153) BAU/ml, p=0.78, Figure 2A . Additional testing was performed in 51/60(85.0%) patients prior to PT-V3 at a median time of 44(32-64) days post-transplant; [anti-S] waned in the absence of infection or vaccination to 40(15-271) BAU/ml, significantly lower than both the corresponding pre-transplant level and following PT-V3 level, p<0.01, Supplemental Information, Figure S1 .

A further comparison of pre-transplant and PT-V3 [anti-S] was undertaken in 29 infection-experienced patients. There was no difference in median [anti-S] pre-transplant and PT-V3, at 1090(133-3667) and 2207(650-5618) BAU/ml respectively, with a median difference of 4 (-625-2160) BAU/ml, p=0.26, Figure 2B .

Anti-S concentrations following a 3^rd vaccine dose were analysed in a control group of 63 haemodialysis patients (HD-V3) who contemporaneously received the two vaccines (V1+V2) whilst remaining on the transplant waitlist, of whom 36(57.1%) and 27(42.9%) were infection-naïve and infection-experienced respectively. A comparison of clinical characteristics between this comparator group and the V2-group can be found in the Supplemental Information, Table S1 .

There was no difference in [anti-S] between the infection-naïve waitlist and pre-transplant V2-group following 2 vaccine doses, with median values of 277(32-952) BAU/ml and 236(48-1046) BAU/ml respectively, p=0.97, Figure 3A . However, following a 3^rd vaccine dose, [anti-S] was significantly higher in the HD-V3 group, who remained on the waitlist, compared with the PT-V3 group, at 1982(936-5593) and 71(30-516) BAU/ml respectively, p<0.0001. Similarly, whilst there was no difference in [anti-S] in infection-experienced waitlist and pre-transplant V2-group, with median values of 2695(462-5680) and 696(173-3830) BAU/ml respectively, p=0.22; following a 3^rd vaccine, [anti-S] was significantly higher in the HD-V3 versus PT-V3 patients at 5680(2681-5680) and 983(427-5214) BAU/ml respectively, p=0.0006, Figure 3B .

From 75 V2-group patients who received 2 vaccines (V3 and V4) post-transplant, paired serological testing was available in 36/75 (48.0%) patients at the time of transplantation, following -V3 and following -V4. The median time to PT-V4 in these 36 patients was 226(208-295) days. The corresponding median time to serological testing following PT-V3 and PT-V4 was 33(25-42) and 30(22-40) days respectively, p=0.56. At the time of testing PT-V4, 24/36(66.7%) patients remained infection-naïve, 16(66.7%) of whom met the serological criteria for a boosted status, compared with only 7(29.2%) following PT-V3, p=0.01. Median [anti-S] in infection-naïve patients following PT-V4 versus PT-V3 was 5134 (229-5680) versus 97(34-1074) BAU/ml respectively, p=0.0016, Figure 4 . Only 1/12 (8.3%) infection-experienced patients did not meet boosted status PT-V4.

From 43 V3-group patients who received a 4^th vaccine post-transplant (having received V1-3 pre-transplant), paired serological testing at the time of transplant and post-V4 was available in 35(81.3%). The median time to PT-V4 in the V3-group was 82(49-101) days, which was significantly shorter than the comparator V2-group who received PT-V4 at median time of 226(208-295) days, p<0.0001 post-transplant, Figure 4 . Only 12/35(34.3%) V3-group patients were considered infection-naïve at the time of sampling PT-V4, at a median of 34 (21-42) days post vaccination. Seven of 12 (58.3%) patients were considered to have boosted status, although there was no significant difference in [anti-S] PT-V4 compared with at the time of transplant for these patients, with median values of 871(242-2092) and 981(396-2666) BAU/ml respectively, p=0.62, Figure 4 . Twenty-three (65.7%) V3-group patients were infection-experienced at the time of testing PT-V4, of whom 17(68.0%) were considered boosted. There was no difference in [anti-S] PT-V4 compared with pre-transplantation, at 5477(690-5680) and 239(1089-5622) BAU/ml respectively, p=0.31.

Ninety of all 204 (44.1%) patients were infection-experienced at the time of transplant. Seventy-seven patients (37.8%) were diagnosed with infection post-transplant, of which 24/77 (31.2%) cases were re-infections. Despite high reinfection rates, patients with infection exposure pre-transplantation were less likely to become infected post-transplant compared with infection-naïve patients, with 24/90(26.7%) and 53/114(46.5%) infections respectively; OR: 0.42 (0.23-0.76), p=0.004. The majority of the V2-group, 87/129 (67.4%), were transplanted whilst the Delta variant was dominant. In the V3-group, 47/75 (62.7%) patients were transplanted during the Omicron period. There was no difference in infection-free survival post-transplant between the 2 groups at the end of follow up, p=0.09, Supplemental Information, Figure S2 . However, 51/55 (92.7%) of infected V2-group patients were diagnosed during the Omicron period, as were all V3-group patients.

Patients who were diagnosed with infection post-transplant had lower [anti-S] at the time of transplant compared with those who remained infection-free, with concentrations of 515(38-1847) and 1281(178-3925) BAU/ml respectively, p=0.004. Of patients who were diagnosed with infection post-transplant, 51/77 (66.2%) had received an additional vaccine dose during the post-transplant period. Five of 77 (6.5%) patients who were diagnosed with infection post-transplant died, of whom 3 (3.9%) were within 28 days of COVID infection. All five cases had PCR confirmed infection, all occurring during the Omicron period, 4 with available results confirming Omicron by genotyping. Twenty-three of 77 (29.9%) patients with post-transplant infection were hospitalised at the time or within 28 days following diagnosis; 5/23(21.7%) were infected following nosocomial transmission. Primary indication for admission may be found in the Supplementary Information, Table S2 . By comparison, of the waitlist controls, 18/63 (28.6%) were diagnosed with infection post at least 3 doses of vaccine, of which there were no admissions and no deaths recorded.

Fifty patients had paired serology pre- and post-infection. Thirty-three (66.0%) had no interval vaccination, of whom 13 (39.4%) received monoclonal antibody treatment (mAb). Seventeen (34.0%) had an interval vaccine dose in addition to infection, of whom 5/17 (29.4%) received monoclonal antibody treatment, two casirivimab plus imdevimab and 15 sotrovimab. There was no significant difference in time to anti-S testing post-infection in the vaccine+mAb+, vaccine+mAb-, vaccine-mAb+ and vaccine-mAb- patients at a median of 35 (14-143), 72 (45-99), 65 (28-89) and 61 (36-143) days post-infection respectively, p=0.77. There was also no difference in [anti-S] between the groups, with median levels of 2370(1418-5680), 3703(129-5680), 3386 (2487-5680) and 1937(673-4903) BAU/ml respectively, p=0.27, Supplemental Information, Figure S3 . All groups had a significant increase in anti-S compared with pre-transplant except for the vaccine+mAb+ group, in whom [anti-S] rose from 39 (7.1-879) to 2370 (1418-5680), p=0.13, Supplemental Information, Figure S4 .