A third injection of the BNT162b2 mRNA COVID-19 vaccine in kidney transplant recipients improves the humoral immune response.

To the editor:

Several reports have highlighted the poor humoral immune response of kidney transplant recipients following coronavirus disease 2019 (COVID-19) mRNA vaccination compared with immunocompetent patients.1, 2, 3 Therefore, the French National Authority for Health has recommended the use of a third vaccine dose for immunosuppressed patients, such as solid organ transplant recipients. We retrospectively assessed the humoral response of all kidney and pancreas transplant recipients vaccinated with the BNT162b2 mRNA (Pfizer BioNTech) COVID-19 vaccine between January and May 2021 in our center. Patients with previous COVID-19 infection or positive prevaccination serology were excluded. The methods for detection of the anti-spike protein responses were dependent on laboratories’ practices and included chemiluminescent microparticle immunoassay (Abbott Architect), chemiluminescence immunoassay (Siemens Atellica), and electrochemiluminescence immunoassay (Roche Elecsys). Anti-spike (IgG) responses were assessed 1 month after the second and third injection, and patients were considered positive if their anti-spike level was above the laboratory threshold. A total of 456 patients had a serologic assessment 1 month after the second injection, of whom 227 were positive, representing 49.7% of our cohort (Table 1 ). A total of 10.7% of these patients had a positive anti-spike protein 1 month after the first injection. Multivariate regression analysis (Table 2 ) revealed that there was an increased likelihood of being a nonresponder after the second mRNA injection for patients treated with antimetabolite drugs (odds ratio [OR], 5.74; 95% confidence interval [CI], 2.99–11.48; P < 0.0001) or steroids (OR, 3.68; 95% CI, 2.10–6.66; P < 0.0001), older recipients (OR, 1.03; 95% CI, 1.01–1.06; P = 0.0010), those with impaired allograft function (OR, 0.98; 95% CI, 0.96–0.99; P = 0.0011), and those with a transplant ≤4 years (OR, 2.91; 95% CI, 1.70–5.08; P = 0.0001). A total of 136 patients had a serologic assessment 1 month after the third injection (median, 30 days; quartile 1, 28 days; quartile 3, 32 days). The average time between the second and the third injection was 50 days. A total of 94 patients were positive, representing a 69.2% serologic conversion following the third mRNA injection (Figure 1 a and Table 3 ). Among patients receiving a third dose, 85 had a serologic assessment after both second and third injections, and 34 of them (40%) became seropositive between the second and the third dose. The magnitude of immune response was investigated in 71 patients who had serologic assessment using electrochemiluminescence immunoassay (Roche Elecsys) after the second and third injections (Figure 1b and c). Nearly all patients with a positive serology after the second mRNA vaccine had a high titer of anti-spike antibody (>250 UI/L). Multivariate regression analysis (Table 4 ) determined that lymphocyte count <1500/mm3 increased the likelihood of being a nonresponder after the third mRNA injection (OR, 3.84; 95% CI, 1.58–19.96; P = 0.0039), as impairment of allograft function (OR, 0.97; 95% CI, 0.94–0.99; P = 0.0232). Of note, the use of antiproliferative drugs and steroids no longer seemed to significantly impact the serologic conversion after the third mRNA vaccine injection. Male recipients were more likely to respond to the third mRNA vaccine injection in our cohort, without any clear explanation so far. Kidney transplant recipients respond poorly to COVID-19 mRNA vaccination, and although cellular responses to the vaccine seem better than humoral responses, severe COVID-19 pneumonia can occur following 2 mRNA vaccine injections. Our data support the use of a third mRNA injection to improve the humoral response to vaccination from about 50% to 70%, reducing the negative impact of antimetabolite drugs and steroids on seroconversion. Moreover, for previously seropositive patients, the third mRNA vaccine largely improved the intensity of humoral response, reaching titers suggestive of neutralizing antibody activity. Indeed, seropositive assessment (especially weak titers in immunocompromised patients) is not constantly associated with protective antibodies titers. However, severe lymphopenia and impaired graft function remained as risk factors for a nonserologic response. In this particular situation, a fourth dose, with or without immunomodulation, could be discussed to improve the humoral response of this highly immunosuppressed population. These data will need confirmation from other series.

Table 1

Characteristics associated with risk of nonhumoral response to a second dose of mRNA COVID-19 vaccine after univariate analysis (n = 456)

Characteristics	Global (n = 456)			Negative (n = 229)			Positive (n = 227)			P value
	NA	n	%	NA	n	%	NA	n	%
Transplantation ≤4 yr	0	336	73.7	0	146	63.8	0	190	83.7	<0.0001
Male recipient	0	275	60.3	0	130	56.8	0	145	63.9	0.1455
Transplant rank ≥2	0	81	17.8	0	40	17.5	0	41	18.1	0.9653
Transplant type (kidney vs. SPK/pancreas)	0	421	92.3	0	214	93.4	0	207	91.2	0.3845
Primitive kidney disease	0			0			0			0.5491
Unknown		39	8.6		15	6.6		24	10.6
Glomerulonephritis		143	31.4		71	31		72	31.7
Other		205	45		109	47.6		96	42.3
Vascular disease		28	6.1		13	5.7		15	6.6
Diabetes		41	9		21	9.2		20	8.8
Deceased donor	0	376	82.5	0	196	85.6	0	180	79.3	0.1002
ABDR incompatibilities >4	0	102	22.4	0	59	25.8	0	43	18.9	0.0920
Blood type	0			0			0			0.4043
O		192	42.1		102	44.5		90	39.6
A		206	45.2		103	45		103	45.4
B		46	10.1		18	7.9		28	12.3
AB		12	2.6		6	2.6		6	2.6
Depleting induction	1	239	52.5	1	129	56.6	0	110	48.5	0.1009
Lymphocytes <1500/mm3	36	220	52.4	20	126	60.3	16	94	44.5	0.0017
Calcineurin inhibitor treatment	20	369	84.6	11	191	87.6	9	178	81.7	0.1110
Belatacept treatment	18	11	2.6	10	9	4.4	8	2	0.9	0.0343
mTOR inhibitor treatment	20	68	15.6	11	20	9.2	9	48	22	<0.0001
Antimetabolite treatment	20	325	74.5	11	180	82.6	9	145	66.5	0.0002
Steroid treatment	20	150	34.4	11	94	43.1	9	56	25.7	0.0002
Diabetes history	0	79	17.3	0	46	20.1	0	33	14.5	0.1493
Cardiovascular history	0	179	39.3	0	94	41	0	85	37.4	0.4890
Neoplasia history	0	89	19.5	0	54	23.6	0	35	15.4	0.0375
DSA before transplant	136	21	6.6	47	16	8.8	89	5	3.6	0.1050
DSA de novo	6	56	12.4	5	26	11.6	1	30	13.3	0.6944
Episode of rejection	0	51	11.2	0	30	13.1	0	21	9.3	0.2479
	NA	Mean	SD	NA	Mean	SD	NA	Mean	SD
Recipient age, yr	0	61.4	12.1	0	62.7	11.1	0	60.2	13.0	0.0289
Recipient BMI, kg/m2	31	25.4	4.9	16	25.5	4.8	15	25.4	5.0	0.7514
Time from transplantation, yr	0	10.5	8.5	0	8.5	7.6	0	12.4	8.9	<0.0001
Allograft function by MDRD, ml/min	24	48.7	19.4	14	43.4	16.8	10	53.9	20.3	<0.0001
Total leukocyte count, G/L	30	32.1	376.2	18	57.9	534.0	12	6.7	2.0	0.1653
C0 cyclosporine, ng/ml	419	100.6	35.9	218	99.5	50.3	201	101.1	29.0	0.9217
C0 tacrolimus, ng/ml	193	5.9	1.9	83	6.1	2.0	110	5.6	1.7	0.0239

BMI, body mass index; C0, trough level; COVID-19, coronavirus disease 2019; DSA, donor-specific antibody; MDRD, Modification of Diet in Renal Disease; mTOR, mechanistic target of rapamycin; NA, not available; SPK, simultaneous pancreas–kidney.

Values in bold are significant (P < 0.05).

Table 2

Characteristics associated with risk of nonhumoral response to a second dose of mRNA COVID-19 vaccine after multivariate analysis (n = 394)

Characteristics	OR	95% CI	P value
Recipient age at vaccination, yr	1.03	1.01–1.06	0.0010
Transplantation ≤4 yr	2.91	1.70–5.08	0.0001
Allograft function by MDRD, ml/min	0.98	0.96–0.99	0.0011
Calcineurin inhibitor treatment	1.55	0.73–3.30	0.2555
mTOR inhibitor treatment	0.73	0.33–1.62	0.4402
Antimetabolite treatment	5.74	2.99–11.48	<0.0001
Steroid treatment	3.68	2.10–6.66	<0.0001
Lymphocytes <1500/mm3	1.48	0.93–2.35	0.0961

CI, confidence interval; COVID-19, coronavirus disease 2019; MDRD, Modification of Diet in Renal Disease; mTOR, mechanistic target of rapamycin; OR, odds ratio.

Values in bold are significant (P < 0.05).

Figure 1

(a) Prevalence of anti-spike IgG seroconversion following 1, 2, and 3 mRNA injections in kidney transplant recipients. (b) Antibodies titers (electrochemiluminescence immunoassay by Roche Elecsys) 1 month after the second and third mRNA injections. (c) The evolution of antibody titers in previously positive patients who received a third mRNA injection.

Table 3

Characteristics associated with risk of nonhumoral response to a third dose of mRNA COVID-19 vaccine after univariate analysis (n = 136)

Characteristics	Global (n = 136)			Negative (n = 42)			Positive (n = 94)			P value
	NA	n	%	NA	n	%	NA	n	%
Transplantation ≤4 yr	0	46	33.8	0	20	47.6	0	26	27.7	0.0378
Male recipient	0	86	63.2	0	17	40.5	0	69	73.4	<0.0001
Transplant rank ≥2	0	20	14.7	0	8	19.0	0	12	12.8	0.4879
Transplant type (kidney vs. SPK/pancreas)	0	124	91.2	0	38	90.5	0	86	91.5	1.0000
Primitive kidney disease	0			0			0			0.6077
Unknown		10	7.4		5	11.9		5	5.3
Glomerulonephritis		44	32.4		12	28.6		32	34
Other		54	39.7		15	35.7		39	41.5
Vascular disease		12	8.8		4	9.5		8	8.5
Diabetes		16	11.8		6	14.3		10	10.6
Deceased donor	0	114	83.8	0	37	88.1	0	77	81.9	0.5142
ABDR incompatibilities >4	0	33	24.3	0	12	28.6	0	21	22.3	0.5167
Blood type	0			0			0			0.4838
O		52	38.2		20	47.6		32	34
A		66	48.5		18	42.9		48	51.1
B		14	10.3		3	7.1		11	11.7
AB		4	2.9		1	2.4		3	3.2
Depleting induction	0	80	58.8	0	29	69	0	51	54.3	0.1525
Lymphocytes <1500/mm3	6	65	50	2	29	72.5	4	36	40	0.0012
Calcineurin inhibitor treatment	2	115	85.8	1	36	87.8	1	79	84.9	0.8662
mTOR inhibitor treatment	2	20	14.9	1	2	4.9	1	18	19.4	0.0569
Antimetabolite treatment	2	101	75.4	1	32	78	1	69	74.2	0.7950
Steroid treatment	2	43	32.1	1	18	43.9	1	25	26.9	0.0811
Diabetes history	0	28	20.6	0	11	26.2	0	17	18.1	0.3950
Cardiovascular history	0	56	41.2	0	17	40.5	0	39	41.5	1.0000
Neoplasia history	0	29	21.3	0	12	28.6	0	17	18.1	0.2490
DSA before transplant	35	8	7.9	8	4	11.8	27	4	6	0.4370
DSA de novo	5	17	13	2	6	15	3	11	12.1	0.8615
Episode of rejection	0	13	9.6	0	6	14.3	0	7	7.4	0.2207
	NA	Mean	SD	NA	Mean	SD	NA	Mean	SD
Recipient age, yr	0	63.7	11.7	0	65.3	10.9	0	63.0	11.9	0.2679
Recipient BMI, kg/m2	3	25.5	4.7	1	24.5	3.7	2	25.9	5.0	0.0642
Time from transplantation, yr	0	9.4	8.1	0	7.0	6.7	0	10.4	8.5	0.0143
Allograft function by MDRD, ml/min	3	49.1	19.3	1	41.1	15.1	2	52.6	20.0	0.0004
Total leukocyte count, G/L	5	6.6	2.2	1	6.7	2.4	4	6.5	2.2	0.7306
C0 cyclosporine, ng/ml	125	103.0	25.0	40	129.0	29.7	85	97.2	21.5	0.3526
C0 tacrolimus, ng/ml	55	6.0	1.8	13	6.2	1.9	42	5.9	1.8	0.5758

BMI, body mass index; C0, trough level; COVID-19, coronavirus disease 2019; DSA, donor-specific antibody; MDRD, Modification of Diet in Renal Disease; mTOR, mechanistic target of rapamycin; NA, not available; SPK, simultaneous pancreas–kidney.

Values in bold are significant (P < 0.05).

Table 4

Characteristics associated with risk of nonhumoral response to a third dose of mRNA COVID-19 vaccine after multivariate analysis (n = 129)

Characteristics	OR	95% CI	P value
Male recipient	0.25	0.10–0.61	0.0027
Allograft function by MDRD, ml/min	0.97	0.94–0.99	0.0232
Antimetabolite treatment	1.76	0.59–5.64	0.3237
Steroid treatment	2.45	0.91–6.81	0.0795
Lymphocytes <1500/mm3	3.84	1.58–9.96	0.0039

CI, confidence interval; COVID-19, coronavirus disease 2019; MDRD, Modification of Diet in Renal Disease; OR, odds ratio.

Values in bold are significant (P < 0.05).