Outcomes of basiliximab vs alemtuzumab induction in kidney allograft recipients with matched immunological Profiles: A retrospective cohort study

Abstract

BACKGROUND

The use of induction immunosuppression agents has improved kidney transplant outcomes, but selecting the optimal agent remains a point of debate.

AIM

To compare the long-term outcomes of kidney transplant recipients receiving alemtuzumab vs basiliximab induction, focusing on graft function, acute rejection, infection, malignancy, post-transplant glomerulonephritis, and survival, using a propensity score matched cohort design.

METHODS

Kidney transplant recipients who received alemtuzumab or basiliximab induction from 2014 to 2019 across two nephrology centres in Northwest England were evaluated. Propensity score matching at a 1:1.5 ratio ensured comparability between cohorts. Baseline characteristics, immunosuppression regimens, and outcomes were analyzed. Linear, binary logistic and Cox proportional hazard regression models.

RESULTS

A total of 436 recipients were included, with a median follow-up of 5.2 years. The matched cohort (n = 262) had a mean age of 51.1 ± 13.5 years; 39% were female and 92% were white. There was no significant difference in the cumulative incidence of acute rejection [odds ratio (OR) = 2.10; 95%CI: 0.9-4.9; P = 0.110]. Compared with basiliximab, alemtuzumab was associated with lower estimated glomerular filtration rate at 12 months (-6.6 mL/minute/1.73 m²; 95%CI: -10.5 to -2.7; P < 0.001) and higher risks of cytomegalovirus viremia (OR = 3.2; 95%CI: 1.6-6.5; P < 0.001), BK viremia (OR = 2.4; 95%CI: 1.1-5.5; P = 0.02), post-transplant malignancy (OR = 6.2; 95%CI: 1.6-29.9; P = 0.013), and death-censored graft loss (hazard ratio = 3.6; 95%CI: 1.2-11.4; P = 0.03). No significant differences were observed in post-transplant glomerulonephritis or recipient mortality.

CONCLUSION

In this propensity score-matched analysis, alemtuzumab induction was associated with lower graft function at 12 months and higher risks of viral infection, post-transplant malignancy, and graft loss compared with basiliximab. These findings highlight the need for further studies to confirm the long-term safety and effectiveness of alemtuzumab in kidney transplantation.

Key Words: Kidney transplantation; Immunosuppression induction; Alemtuzumab; Basiliximab; Graft outcomes

Core Tip: This retrospective cohort study compared real-world outcomes of alemtuzumab and basiliximab induction in kidney transplant recipients from two tertiary centers in North-West England, one routinely using alemtuzumab and the other exclusively basiliximab. Using propensity score matching to reduce confounding, we evaluated graft function, graft and patient survival, acute rejection, infection, and other long-term complications. Alemtuzumab induction was found to be associated with lower estimated glomerular filtration rate, and higher risks of cytomegalovirus and BK viremia, post-transplant malignancy, and death-censored graft loss compared with basiliximab. These findings emphasize the need for individualized selection of induction therapy, balancing the potent immunosuppressive effects of alemtuzumab against its potential for adverse outcomes, and support tailored approaches based on recipient risk profile and center-specific experience.

INTRODUCTION

The use of monoclonal and polyclonal antibody-based induction immunosuppression in kidney transplantation has significantly improved the short-term and long-term outcomes in transplant recipients. It has enabled the utilisation of less-than-optimal donor kidneys, including expanded criteria donors and improved the outcomes of high immunologic risk recipients such as highly sensitised recipients, Black Caribbean recipients, and repeat transplant recipients[1,2]. Two categories of induction agents are currently used for kidney transplantation: (1) Lymphocyte-depleting; and (2) Non-lymphocyte-depleting agents.

Alemtuzumab (Campath) is a recombinant anti-CD52 pan-lymphocyte-depleting monoclonal antibody that is widely used in the United Kingdom.

Basiliximab (Simulect), an interleukin-2 inhibitor, is the main non-lymphocyte-depleting induction agent employed in the United Kingdom.

Despite their extensive use, the optimal induction agent and regimen remain subjects of ongoing debate amongst researchers and transplant providers. This controversy arises because the comparative effectiveness of commonly used agents such as alemtuzumab and basiliximab has yielded mixed results across studies. The inconsistency is partly due to heterogeneity in study populations, dosing strategies, and maintenance immunosuppressive regimens and partly because the selection of induction therapy requires balancing the competing risks of acute rejection, infection, malignancy, and long-term graft survival[1,3]. Consequently, these agents are usually administered to patients with widely differing baseline clinical, demographic, and immunologic characteristics. Alemtuzumab recipients, for example, tend to be younger and have a higher immunologic risk profile compared with basiliximab recipients. This makes direct comparison between groups with fundamentally dissimilar baseline characteristics inherently challenging[1].

Furthermore, there is a paucity of head-to-head comparison studies[4-6] between alemtuzumab and interleukin-2 receptor antagonist (IL-2RA) in relation to rejection rates, graft and patient survival, and complications such as DNA virus infection and malignancy[7-9].

This study therefore aimed to compare the outcomes of kidney transplant recipients (KTR) who received alemtuzumab vs basiliximab induction, using propensity score matching (PSM) to ensure comparable baseline immunologic risk profiles.

MATERIALS AND METHODS

Study design

This retrospective comparative cohort study evaluated adult KTR who received kidney transplants between January 2014 and December 2019 at two tertiary nephrology centres in northwest England (Hospital A and Hospital B). Subjects received induction immunosuppression with either alemtuzumab (Campath) or basiliximab (Simulect) at the time of transplantation. Hospital A predominantly used alemtuzumab for immunosuppression induction whereas Hospital B uses basiliximab as an induction agent in most kidney transplantations reserving alemtuzumab for simultaneous kidney and pancreas transplantation.

Participants

Subjects were drawn from the transplant recipient databases of two participating hospitals, designated as Hospital A and Hospital B. Hospital A employed a combination of alemtuzumab and basiliximab for induction of immunosuppression, while Hospital B predominantly utilised basiliximab. At Hospital A, the indications for alemtuzumab induction included transplantation of a donation-after-circulatory-death allograft, donation-after-brain-death allografts with a cold ischaemic time exceeding 24 hours, allografts with a 2-DR mismatch, panel reactive antibody (PRA) greater than 20%, positive human leucocyte antigen (HLA) crossmatch, and instances of previous early graft loss due to acute rejection. The dosing regimen for alemtuzumab involved a 30 mg subcutaneous administration at transplantation, with a second dose given a day later for recipients under 60 years old; those over 60 years received a single 30 mg dose at transplantation only. Recipients not meeting these indications received basiliximab induction, dosed at 20 mg at implantation and on days 0 and 4 post-transplant. Conversely, Hospital B employed basiliximab 20 mg intravenously on days 0 and 4 post-transplant, with rare instances of alemtuzumab induction in multi-organ transplant recipients.

Due to evolving immunosuppression regimens and data inconsistency pre-2014, the study focused on KTR from 2014 onwards. In addition, recipients transplanted after 2019 were excluded due to the influence of the coronavirus disease 2019 pandemic on immunosuppression practices.

The study exclusively considered kidney-only transplant recipients who received induction immunosuppression with either alemtuzumab (from Hospital A) or basiliximab (from Hospital B), ensuring no prior exposure to alemtuzumab. Exclusion criteria included recipients with prior alemtuzumab exposure for indications other than transplant immunosuppression induction, previous exposure to other lymphocyte-depleting agents (e.g., rituximab, thymoglobulin), multiorgan transplant recipients, and those experiencing graft loss within three months post-transplantation. Patients who experienced early graft loss (within 3 months) were excluded because their subsequent follow-up data (for outcomes such as acute rejection, viral infections, malignancy, and long-term graft function) would not be available or meaningful in the context of our study endpoints, which sought to assess long-term outcomes. Patients who experienced graft loss within the first three months post-transplantation were excluded from the primary analysis. This decision was made because such patients do not contribute meaningful follow-up time for the outcomes of interest (acute rejection, viral infections, malignancy, and long-term graft function). Our focus was therefore on evaluating medium-term and long-term outcomes among recipients with sustained allograft function beyond the immediate post-transplant period.

Maintenance immunosuppression

In Hospital A, recipients who underwent alemtuzumab induction are managed with a steroid-free, calcineurin inhibitor (CNI)-based regimen, typically involving tacrolimus (or cyclosporin if tacrolimus is contraindicated). The target tacrolimus trough level is 5-8 ng/L. Mycophenolate mofetil (MMF) is also administered at a daily dose of 1000 mg in two divided doses. Recipients classified as high immunological risk, with factors such as a 2-DR mismatch, PRA > 20%, or a history of graft loss from acute rejection, have a higher target tacrolimus trough level of 8-12 ng/L for the initial 6 months, followed by 5-8 ng/L thereafter, also in conjunction with MMF. Additionally, recipients aged over 60 years, who received a single dose of alemtuzumab, received tacrolimus targeting a trough level of 8-12 ng/L for the first 6 months, followed by 5-8 ng/L afterwards, MMF was also given at a dose of 1000 mg daily in two divided doses. In addition, glucocorticoids are given for the initial 3 months post-transplantation.

In Hospital B, recipients who underwent induction with the IL-2RA basiliximab are treated with CNI-based maintenance immunosuppression, using either tacrolimus or cyclosporin. Those without contraindications also receive an anti-proliferative immunosuppressant, either mycophenolic acid or azathioprine. Steroid treatment for recipients with standard immunologic risk is typically restricted to less than two weeks of steroid treatment. However, individuals at high immunologic risk, such as younger recipients and older donor age, calculated PRA greater than 20%, presence of a donor-specific antibody, blood group incompatibility, delayed onset of graft function, or cold ischemia time (CIT) exceeding 24 hours, receive a more prolonged course of corticosteroid maintenance. The necessity for continued steroid administration is re-evaluated after 3 months and 6 months, with the potential for discontinuation if the patient is deemed to be at low risk of acute rejection or high risk of infections.

Although Hospital A had a steroid avoidance regimen, a significant number of patients were commenced on steroids as a replacement for MMF due to severe lymphopenia linked to the combination of alemtuzumab induction and MMF maintenance or post-transplant DNA viremia.

Data collection

The data for this study were extracted from the hospital's electronic patient records, comprising relevant information obtained from clinical letters, records, and laboratory results.

Follow-up

Data collection continued until clinical endpoints of graft loss, patient death, loss to follow-up, or study conclusion on December 31, 2021. Subjects were censored at the occurrence of any of the above endpoints, whichever came first.

Variables

Baseline demographic variables, comorbidities, transplant factors, maintenance immunosuppression, and post-transplant complications were recorded.

Exposure variable

The exposure variable was induction type, specifically alemtuzumab induction vs basiliximab induction.

Confounding variables

Potential confounders and effect modifiers included recipient age, ethnicity, the primary cause of end-stage kidney disease, donor type, donor and recipient cytomegalovirus (CMV) serostatus, degree of HLA mismatch, crossmatch status, transplant year, CIT, steroid regime, baseline estimated glomerular filtration rate (eGFR), PRA, recipients were also categorised into age groups < 60 years and ≥ 60 years (with recipients aged 60 years and above receiving half the dose of alemtuzumab compared to those less than 60 years) and CIT categories (≤ 24 hours and > 24 hours).

Outcome variables

Outcome variables included change in eGFR at 1 year, history of acute rejection, history of post-transplant DNA virus infections CMV, Epstein-Barr virus and BK viremia respectively, history of post-transplant malignancy, history of post-transplant glomerulonephritis (GN), death censored graft survival and recipient survival (RS).

Graft loss was defined as a return to dialysis or re-transplantation. Post-transplant malignancy included all de-novo cancers after transplantation, excluding non-melanoma skin cancers. Baseline eGFR was defined as eGFR at 3 months post-transplantation and was calculated using the modification of diet in the renal disease equation. Acute rejection was confirmed through allograft biopsy.

Statistical analysis

PSM: We conducted a PSM analysis to compare the outcomes between alemtuzumab and basiliximab recipients. We aimed to achieve a matched ratio of 1:2. The propensity scores were estimated using logistic regression, incorporating covariates such as recipient’s age, donor type, CIT, CIT more or less than 24 hours, year of transplantation, age more than 60 years vs age less than 60 years, HLA-DR mismatch and the percentage of the PRA. These covariates were selected based on the indications for alemtuzumab induction in Hospital A. MatchIt software in R was used for PSM[10]. Nearest neighbour greedy matching was performed using a calliper width of 0.25 standard deviations of the logit of the propensity score. The matched sample consisted of 101 alemtuzumab recipients and 160 basiliximab recipients resulting in a matching ratio of approximately 1:1.5. Despite not achieving a perfect 1:2 ratio, the matched groups were well-balanced in terms of observed covariates, indicating successful PSM. Covariate balance was assessed using standardised mean differences (SMDs), with an absolute SMD of ≤ 0.10 considered to indicate balance. Following PSM, two variables with significant residual imbalance (pre-emptive transplantation and primary renal disease) were additionally adjusted for in all multivariable regression models used to estimate treatment effects.

Covariate balance after PSM: Overall, the propensity matching improved the balance of baseline characteristics between the Simulect group and Campath group, evidenced by a well-balanced SMDs on all the variables on which the cohorts were matched as shown in the Supplementary Figure 1. Before matching, there were notable differences in baseline characteristics between the alemtuzumab and basiliximab groups, with some variables having SMDs greater than 0.1. However, after PSM, all variables achieved balance, with SMD approximately 0.1 for all matched variables.

Inferential analysis was conducted on the matched cohort. Categorical variables were expressed as frequencies and percentages, while continuous variables as means with standard deviations or medians with interquartile ranges depending on the distribution. Propensity scores were used as covariates in linear regression for eGFR change at 12 months, logistic regression for binary outcomes (acute rejection, post-transplant GN, post-transplant viral infections), and Cox proportional hazard regression for death-censored graft survival (DCGS) and RS. Statistical significance was set at P < 0.05, and R software (R Foundation for Statistical Computing, Vienna, Austria, version 4.2.2) was used for analysis. Further analyses of the treatment effects were then carried out on the matched cohort.

RESULTS

Between January 2014 and December 2019, a total of 665 patients underwent kidney transplantation across two hospitals, with 325 recipients from Hospital A and 330 from Hospital B. Among them, 149 recipients received alemtuzumab induction at Hospital A, while 329 received basiliximab induction at Hospital B. After excluding recipients who met the predefined exclusion criteria, 436 eligible recipients [basiliximab (n = 319), alemtuzumab (n = 117)] remained for PSM. Ultimately, 262 patients [alemtuzumab (n = 102) and basiliximab (n = 160)] were successfully matched. Twelve patients experienced early graft loss (< 3 months) and were excluded from the matched analysis [alemtuzumab (n = 5) and basiliximab (n = 7)]. The study flow and matching process are summarized in Figure 1.

Figure 1 Flow chart showing the selection of kidney transplant recipients who received alemtuzumab or basiliximab induction between 2014 and 2019 at two transplant centers. Exclusions were based on predefined criteria. Propensity score matching was performed using recipient age, donor type, cold ischemia time (< 24 hours or > 24 hours), year of transplantation, age group (≥ 60 years vs < 60 years), human leucocyte antigen-DR mismatch, and percentage of panel reactive antibody: (1) Cold ischemia time; (2) Human leucocyte antigen; and (3) Royal Liverpool Hospital. CIT: Cold ischemia time; HLA: Human leucocyte antigen; PRA: Panel reactive antibody; RLH: Royal Liverpool Hospital.

The baseline characteristics

The baseline characteristics of the recipients before propensity matching are shown in Supplementary Table 1. Whereas the characteristics of the matched cohort are presented in Table 1, and the distribution of the outcome variables are shown in Table 2. The average age of the total cohort was 51.6 ± 14.2 years, which remained consistent at 51.1 ± 13.5 years in the matched cohort. The proportion of women was 38% in the total cohort and 39% in the matched cohort, with 80% of the total and 92% of the matched cohort being of white ethnicity. GN was the primary cause of kidney failure in 26% of both cohorts, while diabetic kidney disease occurred in 15% of the total and 13% of the matched cohort. The median follow-up period was 5.20 years (interquartile range: 4.17-6.20), during which 31 (7.1%) of the total cohort and 22 (8.4%) of the matched cohort experienced graft loss, and 37 (8.5%) of the total and 18 (6.9%) of the matched cohort died with functioning grafts.

Table 1 Characteristics of the matched cohort, n (%)/median (interquartile range).

Variables	Basiliximab (n = 160)	Alemtuzumab (n = 102)	Total (n = 262)	P value
Age(years) [mean (SD)]1	51.1 (13.6)	51.2 (13.4)	51.1 (13.5)	0.952
Sex (female)	62 (38.8)	41 (40.2)	103 (39.3)	0.815
Ethnicity1
White	146 (91.2)	96 (94.1)	242 (92.4)	0.773
Asian	10 (6.2)	4 (3.9)	14 (5.3)
Black	3 (1.9)	1 (1.0)	4 (1.5)
Other	1 (0.6)	1 (1.0)	2 (0.8)
Primary renal disease
Adult polycystic kidney disease	39 (24.2)	14 (13.9)	53 (20.2)	0.048
Glomerulonephritis	42 (26.1)	27 (26.7)	69 (26.3)
Diabetic kidney disease	19 (11.8)	14 (13.9)	33 (12.6)
Reflux/chronic pyelonephritis	18 (11.2)	14 (13.9)	32 (12.2)
Other	28 (17.4)	29 (28.7)	57 (21.8)
Unknown	15 (9.3)	3 (3.0)	18 (6.9)
Recipient diabetes	28 (17.6)	17 (17.3)	45 (17.5)	0.957
Body mass index (kg/m2) [mean (SD)]	27.6 (4.4)	27.0 (4.9)	27.4 (4.6)	0.318
Pre-emptive transplant	67 (41.9)	20 (19.6)	87 (33.2)	< 0.001
Donor type1
Living donor	63 (39.4)	38 (37.3)	101 (38.5)	0.346
Donation after brain death	41 (25.6)	20 (19.6)	61 (23.3)
Donation after circulatory death	56 (35.0)	44 (43.1)	100 (38.2)
Donor CMV	93 (58.1)	50 (49.0)	143 (54.6)	0.149
Recipient CMV	89 (55.6)	51 (50.0)	140 (53.4)	0.373
Recipient Epstein-Barr virus	45 (90.0)	87 (96.7)	132 (94.3)	0.103
HLA-DR mismatch [mean (SD)]1	0.7 (0.6)	0.7 (0.6)	0.7 (0.6)	0.979
Total HLA mismatch [mean (SD)]1	2.6 (1.4)	2.6 (1.5)	2.6 (1.4)	0.987
Crossmatch positive1	1 (0.6)	1 (1.0)	2 (0.8)	0.738
Pannel reactive antibodies [mean (SD)]1	22.4 (33.8)	25.6 (35.8)	23.7 (34.5)	0.471
Cold ischemia time1	11.0 (4.0-17.0)	12.0 (5.0-15.0)	11.5 (5.0-15.8)	0.906
Calcineurin inhibitor
Tacrolimus	156 (97.5)	102 (100.0)	258 (98.5)	0.274
Cyclosporin	3 (1.9)	0 (0.0)	3 (1.1)
Other	1 (0.6)	0 (0.0)	1 (0.4)
Antimetabolite
Mycophenolic acid	150 (93.8)	97 (95.1)	247 (94.3)	0.123
Azathioprine	9 (5.6)	2 (2.0)	11 (4.2)
None	1 (0.6)	3 (2.9)	4 (1.5)
Steroid maintenance	30 (22.2)	22 (21.6)	52 (21.9)	0.904
Baseline estimated glomerular filtration rate	47.0 (36.0-57.0)	43.0 (33.0-53.0)	45.5 (35.0-54.8)	0.093

¹Variables used in the propensity score matching.

Table 1 presents the baseline characteristics of patients before and after propensity matching, stratified by treatment group (basiliximab and alemtuzumab). CMV: Cytomegalovirus; HLA: Human leucocyte antigen.

Table 2 Unadjusted outcomes in basiliximab vs alemtuzumab, n (%)/median (interquartile range).

Outcomes	Basiliximab (n = 160)	Alemtuzumab (n = 102)	Total (n = 262)	P value
One-year estimated glomerular filtration rate	53.5 (42.0-65.0)	46.5 (34.0-54.8)	50.0 (40.0-62.0)	< 0.001
Cytomegalovirus viremia	21 (13.2)	31 (30.4)	52 (19.9)	< 0.001
Epstein-Barr virus viremia	6 (3.8)	11 (10.8)	17 (6.5)	0.025
BK viremia	18 (11.3)	21 (20.6)	39 (14.9)	0.040
Acute rejection throughout the follow up period	10 (7.6)	19 (18.6)	29 (12.4)	0.012
Post-transplant glomerulonephritis	5 (3.2)	4 (3.9)	9 (3.5)	0.744
Post transplant malignancy	6 (4.3)	12 (11.9)	18 (7.5)	0.027

In the pre-matched cohort, there was a higher proportion of white recipients in the Campath group compared to the Simulect group (94.9% vs 74.9%). However, after matching, the distribution of ethnicity became similar between the two groups, with a White majority (94.1% vs 91.2%). Pre-emptive transplantation was more common in the basiliximab group both before matching (39.8% vs 17.2%) and after matching (41.9% vs 19.6%). Additionally, basiliximab recipients exhibited a lower average PRA and a higher baseline eGFR.

Treatment effects

The treatment effects of alemtuzumab vs basiliximab were assessed across various outcome measures in the propensity-matched cohort, as depicted in Figure 2 and Supplementary Table 2. Recipients of alemtuzumab induction demonstrated a significantly lower eGFR at 12 months post-transplant compared with those who received basiliximab, with an adjusted mean difference of –6.6 mL/minute/1.73 m² (95%CI: -10.5 to -2.7; P = 0.001), after adjustment for primary renal disease, baseline eGFR, pre-emptive transplantation, and propensity scores. Similarly, alemtuzumab induction was associated with a significantly higher risk of CMV viremia [odds ratio (OR) = 3.15; 95%CI: 1.58-6.48; P < 0.001] and BK viremia (OR = 2.40; 95%CI: 1.10-5.50; P = 0.024), following adjustment for pre-emptive transplantation, primary renal disease, donor and recipient CMV status, and propensity scores. In contrast, no significant difference was observed in the odds of Epstein-Barr virus viremia between the two groups. Furthermore, alemtuzumab induction was associated with a sixfold higher risk of post-transplant malignancy compared with basiliximab (OR = 6.17; 95%CI: 1.62-29.93; P = 0.013), after adjustment for age, pre-emptive transplantation, primary renal disease, and propensity scores. Finally, although the incidence of acute rejection was numerically higher among recipients of alemtuzumab, the difference did not reach statistical significance after adjustment for pre-emptive transplantation, PRA, and propensity scores (OR = 2.10; 95%CI: 0.88-4.93; P = 0.110).

Figure 2 The treatment effects of alemtuzumab induction compared to basiliximab induction (adjusted for confounding factors including propensity scores, primary kidney disease and pre-emptive transplant rate) along with the corresponding 95%CI. Estimates were calculated using various statistical methods based on the nature of the outcome variable. CMV: Cytomegalovirus; DCGL: Death-censored graft loss; EBV: Epstein-Barr virus; eGFR: Estimated glomerular filtration rate; GN: Glomerulonephritis; NS: Not significant. ¹Estimated using linear regression. ²Odd ratio by binary logistic regression. ³Hazard ratio by Cox proportional hazard model.

Graft and RS

The hazard ratio (HR) of death-censored graft loss and recipient death is shown in Figure 2, whereas the all-cause graft survival, DCGS, and RS curves comparing alemtuzumab and basiliximab induction are presented in Figure 3. Alemtuzumab induction was associated with a significantly higher risk of death-censored graft loss compared with basiliximab (HR = 3.6; 95%CI: 1.2-11.4; P = 0.03). In contrast, no significant difference in overall RS was observed between the two groups (HR = 0.61; 95%CI: 0.23-1.58; P = 0.31).

Figure 3 Kaplan-Meier survival curves of all-cause graft survival, death-censored graft survival and recipient survival in patients who received alemtuzumab vs recipients of basiliximab induction. A: All-cause graft survival; B: Death-censored graft survival; C: Recipient survival. Data derived from a propensity-matched cohort (events = allograft loss and death with functioning graft; allograft loss censored for death and recipient death with functioning graft, respectively). ACGS: All-cause graft survival; DCGS: Death-censored graft survival; RS: Recipient survival.

Sensitivity analysis

To evaluate the robustness of our findings, we performed sensitivity analyses using full matching as an alternative to the main propensity score greedy matching procedure. Effect estimates remained consistent across approaches, supporting the reliability of our main results (Supplementary Table 2 and Supplementary Figure 2).

DISCUSSION

Comparisons between alemtuzumab and IL-2RA induction therapies in kidney transplantation have produced inconsistent results, often reflecting differences in study populations and follow-up durations. Assessing long-term outcomes has been particularly challenging because alemtuzumab is frequently reserved for higher-risk recipients, whereas basiliximab is widely used in standard-risk settings. Therefore, obtaining a cohort of recipients with comparable immunologic risks has been challenging. By leveraging center-specific induction practices and applying PSM, our study was able to compare recipients with similar immunologic and demographic profiles. While prior investigations[6,7] have predominantly focused on short-term to medium-term outcomes (6-36 months)[11-14], our study's median follow-up duration of 62 months (5.2 years) provided valuable long-term insights.

Our findings indicate that alemtuzumab induction was associated with significantly lower eGFR at 12 months, a threefold higher risk of CMV viremia, a twofold higher risk of BK viremia, a sixfold higher risk of post-transplant malignancy, and inferior DCGS. In contrast, rates of acute rejection, post-transplant GN, and patient survival did not differ significantly.

The elevated risks of viral infections and malignancy are biologically plausible given alemtuzumab’s potent lymphocyte-depleting effects, which impair immune reconstitution and viral surveillance. Previous studies have reported mixed findings regarding infection risk[5,7,15]. The 3C study reported no difference in the occurrence of serious infections between alemtuzumab and basiliximab induction, including CMV viremia, but a higher BK virus infection in the alemtuzumab group[5] with shorter follow-up and universal CMV prophylaxis. The 3C study may have underestimated later complications.

The higher incidence of malignancy is similarly consistent with impaired immune surveillance. The depletion of peripheral B cells and T cells, particularly regulatory T cells, in the early post-transplant period, followed by a skewed immune system reconstitution, may contribute to impaired viral surveillance and control mechanisms.

The decline in early graft function at 12 months parallels earlier reports and may reflect the cumulative impact of opportunistic infections, higher rates of chronic immunologic injury, and increased de novo donor-specific antibody formation associated with alemtuzumab[15-18]. These mechanisms[6,7,13] cause late immunological injury to the graft and likely contributed to the lower graft survival observed in our cohort, a finding also supported by several registry and single-center studies[19-23].

The effect on acute rejection remains less clear. Randomized controlled trials have reported both reduced[7,8,21] and similar[12,13,24] rates of rejection with alemtuzumab, and evidence suggests that any protective effect diminishes with time[7,16,25]. In our study, cumulative incidence over five years did not differ, but the absence of data on timing, type, severity, and treatment of rejection episodes limits interpretation. It remains possible that an early benefit was obscured.

To mitigate potential risks associated with alemtuzumab induction, alemtuzumab recipients often receive minimal exposure to corticosteroids and lower CNI trough levels, but the increased risk of cytopenias often leads to reduced or omitted antiproliferative agents. This may inadvertently increase long-term immunologic risk, particularly once lymphocyte reconstitution occurs after the first post-transplant year. Careful risk stratification and individualised immunosuppression adjustments are therefore crucial in these recipients.

Finally, treatment crossover complicates interpretation. In our cohort, more than one-third (42 of 117) of alemtuzumab recipients required prolonged steroids (≥ 6 months) within the first year, despite an intended steroid-free regimen (Supplementary Table 3). These recipients had a higher degree of HLA mismatch and a higher frequency of CMV viremia (48% vs 20%). It is worth noting, however, that this association between steroid use and occurrence of CMV viremia may partly reflect substitution of steroids for antiproliferative agents following CMV infection rather than preceding it. Nonetheless, such complexities highlight the challenges of long-term immunosuppression management and the importance of tailoring therapy to individual risk profiles.

In summary, our findings add to the evidence that alemtuzumab induction, while effective in some contexts, may be associated with increased risks of viral infections, malignancy, and inferior graft survival compared with basiliximab. These results underscore the need for cautious use, close monitoring, and future studies designed to disentangle the effects of induction therapy from subsequent treatment modifications.

Strengths and limitations of the study

The main strength of this retrospective investigation lies in its use of PSM, which enabled the identification of two cohorts of transplant recipients with comparable immunological and demographic profiles. This approach enabled a more balanced comparison of outcomes associated with alemtuzumab and basiliximab induction, thereby reducing confounding bias that has limited prior retrospective analyses.

Nevertheless, several limitations must be acknowledged. First, the retrospective design also carries inherent limitations, including the risk of selection bias and incomplete data capture. Although PSM was employed to mitigate these concerns, residual confounding cannot be completely excluded[26].

Second, the matching process excluded a substantial proportion of patients from the original cohort, potentially reducing statistical power and limiting the generalizability of the findings. However, the use of the full matching method in the sensitivity analysis allowed all patients to be included by creating matched groups.

Third, the two-center design may introduce systematic bias related to institutional differences in clinical practice. While both centers adhered to National British Transplantation Society guidelines for key aspects of post-transplant management (including CNI trough level monitoring, infection prophylaxis, and rejection treatment), unmeasured differences, such as local antibiotic protocols, cardiovascular risk management strategies, and biopsy practices, may have influenced outcomes. These variations could have introduced residual confounding and should be considered when interpreting the results.

Fourth, the study period (2014-2019) coincided with evolving transplant practices, including refinements in immunosuppressive regimens and the implementation of a new kidney allocation scheme in the United Kingdom. Although these changes were unlikely to directly alter induction strategies, their potential impact on overall outcomes cannot be excluded.

Fifth, patients who experienced early graft loss (within three months post-transplantation) were excluded from the analysis. This decision was made to focus on long-term outcomes; however, because early graft failure may be influenced by the choice of induction agent, its exclusion could have underestimated risks associated with one or both agents. While the numbers were few and the characteristics of excluded patients were reported, these criteria changes in maintenance limits the interpretation of results, particularly regarding graft survival.

Additionally, detailed data on the timing, type, and severity of acute rejection episodes were not available, nor were histological assessments of chronic rejection or interstitial fibrosis and tubular atrophy, owing to the absence of routine protocol biopsies in both centers. This represents an important limitation in evaluating the full impact of induction therapy on graft durability.

Finally, changes in maintenance immunosuppression during follow-up were not systematically accounted for. These modifications, such as steroid reintroduction or adjustments in antiproliferative therapy, though difficult to capture in a retrospective analysis, may have influenced long-term outcomes and represent a potential source of confounding.

CONCLUSION

While no difference was observed in acute rejection rates, alemtuzumab induction was associated with several adverse outcomes, including inferior renal function, higher risks of viral infections and malignancy, and reduced graft survival compared with basiliximab induction. These findings highlight the need for careful long-term monitoring, risk stratification beyond the first post-transplant year, and individualized adjustments to maintenance immunosuppression in alemtuzumab recipients. Future prospective studies with extended follow-up are required to validate these findings and provide stronger evidence to guide induction therapy choices in kidney transplantation.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the transplant teams, nephrology nursing staff, and data management personnel at both participating centers for their invaluable support in data collection and patient care. We also thank the clinical governance departments for facilitating access to institutional databases. The authors are particularly grateful to the patients whose clinical data made this study possible.