Evidence- and Consensus-based European Guideline for Immunosuppressive Therapy After Pediatric Kidney Transplantation

Abstract

Background.

There is a lack of robust evidence regarding immunosuppressive therapy in children and adolescents after kidney transplantation (KTx), and as such, international practice is highly variable. Recent clinical practice recommendations advocating individualized immunosuppressive strategies that incorporate newer agents are often not implemented. This can potentially contribute to reduced patient and renal allograft survival.

Methods.

The guideline was developed between January 1, 2024, and December 12, 2025, according to the Guidance Manual of the German Association of Scientific Medical Societies by the German Societies for Pediatric Nephrology, Nephrology, Transplantation, and Pediatrics, the German Kidney Association, the International Pediatric Transplant Association, the European Society for Pediatric Nephrology and the Members of the Cooperative European Pediatric Renal Transplant Initiative.

Results.

This evidence- and consensus-based guideline provides up-to-date, state-of-the-art recommendations for immunosuppressive therapy after KTx in pediatric kidney transplant recipients. It is based on the best available evidence and the consensus of the relevant German Medical Societies, Members of the Cooperative European Paediatric Renal Transplant Initiative, and the working group on transplantation of the European Society for Paediatric Nephrology, and the International Pediatric Transplant Association.

Conclusions.

The formal consensus reached is particularly significant in cases of weak or inconclusive evidence and where recommendations are based solely on expert opinion.

INTRODUCTION

Kidney transplantation (KTx) is the therapy of choice for children with end-stage kidney disease. Selecting an effective immunosuppressive regimen is crucial to prevent graft rejection. In pediatric patients, this aim is particularly challenging, as immunosuppressive agents may have significant long-term adverse effects, including impaired growth, behavioral changes, osteoporosis, arterial hypertension, hyperlipidemia, diabetes mellitus, and an increased risk of infections and malignancies. These effects can negatively impact development. Consequently, managing immunosuppressive therapy after pediatric KTx requires careful balancing of efficacy and toxicity while accounting for multiple patient-specific factors.

Despite this complexity, immunosuppressive therapy for children is currently guided by results from a few randomized controlled trials (RCTs), extrapolation from data from adult KTx trials, and heterogenous, experience-based standards that vary across centers. To date, limited comprehensive evidence-based guidelines have been available to support the selection and management of immunosuppressive therapy in pediatric KTx recipients.

MATERIALS AND METHODS

The guideline was developed between January 1, 2024, and December 12, 2025, according to the Guidance Manual of the Association of Scientific Medical Societies in Germany (AWMF; version 2.1 of September 5, 2013),¹ which applies to guidelines registered and published in the national German guideline register and which meet international standards. Following the AWMF guidance, the guideline group was representative of the discussed healthcare aspects and comprised German pediatric and adult nephrologists, general pediatricians, and transplant specialists, and a patient representative from a national patient organization. In addition, European and international pediatric nephrologists affiliated with the European professional association and the Cooperative European Paediatric Renal Transplant Initiative (CERTAIN) research network contributed to the development of this guideline (France, Austria, Poland, Ukraine, the United Kingdom, Slovenia, Italy, Spain, and Romania). This broad and diverse representation of pediatric nephrologists was intended to minimize the influence of individual center-specific or national protocols—currently existing only in the United Kingdom—on the guideline development.

All participants were required to disclose their interests, which were reviewed by a designated committee in accordance with AWMF criteria, including assessment with defined consequences. Participants were excluded from voting on specific Patient/Problem, Intervention, Comparison, and Outcome (PICO) questions if moderate thematic conflicts of interest were identified, such as participation in industry-funded advisory boards or management responsibility for industry-funded studies. Evidence review protocols based on PICO questions were developed. For all questions, patients of interest were children at the time of their first KTx with low or moderate immunological risk (panel-reactive antibodies [PRAs] <5% and absence of preformed donor-specific antibodies [DSAs]). High immunological risk has been defined as second or third transplantation or PRA of 5% or the presence of DSAs. This definition has been used because it matched the inclusion criteria for studies on which this guideline is based. We acknowledge that the risk of DSAs for graft loss is much higher than that of nondonor-specific PRAs. This definition of immunological risk was arrived at by panel consensus and lacks evidence (except in the case of retransplantation) of actual alloimmune risk when known DSAs are excluded in the accepted donor. Adolescents are at higher risk of graft loss than younger children and older adults. However, as all pediatric studies include adolescents, we could not differentiate between different age groups. Outcomes across all questions were (1) effectiveness: patient survival, graft survival, quality of life, development of biopsy-proven acute rejection (BPAR), and de novo formation of donor-specific HLA antibodies (HLA-DSA); (2) safety: infections, proteinuria, and posttransplant lymphoproliferative disease (PTLD).

The 5 key questions (KQ) in PICO format are listed below:

1. Initial immunosuppressive regimen

• KQ1: What is the comparative effectiveness and safety of initial immunosuppressive regimens in children at the time of first KTx?

  1. Induction therapy

  2. Maintenance therapy

1. Change of immunosuppressive regimen

• KQ2: What is the effectiveness and safety of changing the immunosuppressive regimen in children after the third month post–first KTx compared with no change in the immunosuppressive regimen?

1. Steering/dosing of immunosuppression

• KQ3: What is the effectiveness and safety of steering immunosuppressive therapy by newer methods (area-under-the-curve [AUC] monitoring, biomarkers) compared with conventional steering of immunosuppressive therapy (ie, fixed dosing, trough level monitoring) in children after a first KTx?

1. Protocol biopsies

• KQ4: What is the effectiveness and safety of protocol (surveillance) biopsies of the kidney allograft in children after a first KTx compared with no protocol biopsies or biopsies for cause?

1. Interventions based on detection of de novo DSAs (dnDSAs)

• KQ5: What is the comparative effectiveness and safety of interventions based on detection of de novo HLA-DSAs (dn-HLA-DSAs; with or without decreased kidney function)?

Literature searches were conducted, including only children at the time of first KTx (at least 80% of the study population) and age 0–18 y (at least 80% of the study population). Due to the very low LoE for KQ5 (interventions based on the detection of dn-HLA-DSA), we conducted an additional systematic literature search on January 22, 2025, to also identify studies in adults (indirect evidence).

The systematic literature search and critical appraisal of the relevant literature according to defined inclusion criteria were performed by the Department for Evidence-based Medicine and Evaluation (University for Continuing Education Krems, Austria) following the Preferred Reporting Items for Systematic reviews and Meta-Analyses and Cochrane guidelines (https://www.cochrane.org). The search was conducted in several databases (MEDLINE via Ovid, Embase via Elsevier, Cochrane Central Register of Controlled Trials (CENTRAL), EuropePMC, Preprint Citation Index). Additionally, a manual search of reference lists was conducted to identify additional studies. Both RCTs and nonrandomized controlled studies (excluding cross-sectional and case-control studies) were considered for evidence assessment. If at least 1 RCT for an intervention of interest reported the prioritized outcomes, other study designs, such as cohort studies, for grading the certainty of evidence, were not considered for that comparison. None of the extracted data reported outcomes regarding quality of life. If no RCTs were identified, cohort studies with contemporary control cohorts were prioritized over those with historical control cohorts. Due to a lack of evidence for KQ5 regarding the prespecified inclusion criteria, we included uncontrolled studies for this KQ and conducted an additional search on January 22, 2025, to identify studies in adults using the same databases as described earlier. The adult studies only allowed indirect evidence and no strong recommendations. Every result underwent abstract screening, and if the inclusion criteria were met, it was accepted for full-text screening. During the screening process, 2 independent team members performed the study selection. Conflicts were resolved through discussion or by consulting a third team member.

After voting on thresholds for important benefits or harms, the certainty of evidence for each outcome was assessed according to the guidance of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group.² Evidence-to-decision tables were developed to facilitate structure and transparent decision-making (see Supplement, SDC, https://links.lww.com/TP/D401). Based on this information, the recommendations were discussed and finalized by the guideline group through a formal consensus process with neutral moderation. Concerning the grade of recommendation, the AWMF guidance was used, which allows for 3 grades.

The recommendations were classified as:

• Strong recommendations, which are shown in the text using the following wording: “we recommend” or “we do not recommend” (grade 1/“A”).

• Conditional recommendations (grade 2/“B”), using the following wording: “we suggest” or “we do not suggest.”

• Open recommendation (grade 0) using the following wording: “may be considered” or “may be omitted.”

According to the AWMF guidance, recommendations not based on results of a systematic search are labeled as expert consensus. No recommendation grade is assigned to these recommendations; the strength of the recommendation is expressed solely through the wording

The strength of the consensus was defined according to the AWMF guidance as the percentage of agreement among voting experts as follows:

• “Strong consensus”: agreement >95%

• “Consensus”: agreement >75%

If no consensus was reached for a recommendation, this was flagged as “majority agreement” (>50%–75%) or no agreement” (<50%) and is only stated in the accompanying text.

Recommendations were adopted through formal consensus. First, a 2-round Delphi technique was applied using an online tool. For final discussion and voting on recommendations that did not reach strong consensus online, an in-person consensus conference, according to the type of the US National Institutes of Health, was held with a neutral moderator from the AWMF, with an agreement of >75% required for adoption.

Levels of Evidence (LoE) have been used as defined in GRADE—1: RCTs, 2: case-control and cohort studies, 3: case series, case studies, and 4: expert opinions. The results of the evidence research and further methodical information are presented in detail in Supplement (SDC, https://links.lww.com/TP/D401).

We acknowledge that there are significant risks associated with making some recommendations based solely on expert opinion. It is important to note that even a consensus of experts may be wrong, and therefore, we have clearly stated where evidence is lacking to support a recommendation. However, in the absence of evidence, a well-reasoned, structured expert consensus developed by a representative group with diverse perspectives is the best available option. Ultimately, even conclusions based on weak evidence may be misleading or even wrong. Therefore, very low evidence is defined in GRADE as “We have very little confidence in the effect estimate. The true effect is likely to be substantially different from the estimate of effect.” It is therefore important to perform clinical studies in these areas in the future.

In some cases, the guideline group considered the evidence too weak and the expert opinions too diverse to make a recommendation, and formulated a statement that no recommendation could be made.

Ethics Statement

For guidelines based on systematic literature research, no ethical approval is needed. No patients were directly involved, so no informed consent was required.

CLINICAL PRACTICE RECOMMENDATIONS

KQ1a: Induction Therapy

Interleukin-2 Receptor Antagonists (Literature Pediatric KTx)

1	Recommendation		2025		Grade of recommendation/LoE
a) We suggest using interleukin-2 receptor antagonist (IL2RA) induction therapy as part of an early steroid withdrawal regimen (ie, according to the TWIST trial). b) We do not suggest using induction with IL2RA in children receiving a first KTx, receiving a maintenance immunosuppressive regimen comprising tacrolimus, mycophenolate mofetil (MMF), and prednisolone.					2 ⇑ 2 ⇑
Offner G, Toenshoff B, Hocker B, et al. Efficacy and safety of basiliximab in pediatric renal transplant patients receiving cyclosporine, mycophenolate mofetil, and steroids. Transplantation. 2008;86(9):1241–1248. 3 Webb NJ, Prokurat S, Vondrak K, et al. Multicentre prospective randomised trial of tacrolimus, azathioprine and prednisolone with or without basiliximab: two-year follow-up data. Pediatr Nephrol. 2009;24(1):177–182. 4 Shemshadi M, Hoseini R, Zareh R, et al. Use of basiliximab with the standard immunosuppressive protocol in pediatric renal transplantation: a double-blind randomized clinical trial. Int J Organ Transplant Med. 2020;11(1):8–14. 5 Duzova A, Buyan N, Bakkaloglu M, et al. Triple immunosuppression with or without basiliximab in pediatric renal transplantation: acute rejection rates at one year. Transplant Proc. 2003;35(8):2878–2880. 6 Michael M, Minard CG, Kale AS, et al. Outcomes of two-drug maintenance immunosuppression for pediatric renal transplantation: 10-yr follow-up in a single center. Pediatr Transplant. 2016;20(1):49–58. 7 Baron PW, Ojogho ON, Yorgin P, et al. Comparison of outcomes with low-dose anti-thymocyte globulin, basiliximab or no induction therapy in pediatric kidney transplant recipients: a retrospective study. Pediatr Transplant. 2008;12(1):32–39. 8 Ojogho O, Sahney S, Cutler D, et al. Mycophenolate mofetil in pediatric renal transplantation: non-induction vs. induction with basiliximab. Pediatr Transplant. 2005;9(1):80–83. 9 Mincham CM, Wong G, Teixeira-Pinto A, et al. Induction therapy, rejection, and graft outcomes in pediatric and adolescent kidney transplant recipients. Transplantation. 2017;101(9):2146–2151. 10 Sharifian M, Arad B, Simfroosh N, et al. Effects of interleukin 2 receptor blockers on patient and graft survival in renal-transplanted children. Nephrourol Mon. 2014;6(4):e18641. 11 Brunkhorst LC, Fichtner A, Hocker B, et al. Efficacy and safety of an everolimus- vs. mycophenolate mofetil-based regimen in pediatric renal transplant recipients. PLoS ONE. 2015;10(9):e0135439. 12 Grenda R, Watson A, Vondrak K, et al. A prospective, randomized, multicenter trial of tacrolimus-based therapy with or without basiliximab in pediatric renal transplantation. Am J Transplant. 2006;6(7):1666–1672. 13 Sampaio MS, Poommipanit N, Kuo HT, et al. Induction therapy in pediatric kidney transplant recipients discharged with a triple drug immunosuppressive regimen. Pediatr Transplant. 2010;14(6):770–778. 14 Billing H, Burmeister G, Plotnicki L, et al. Longitudinal growth on an everolimus- versus an MMF-based steroid-free immunosuppressive regimen in paediatric renal transplant recipients. Transpl Int. 2013;26(9):903–909. 15 Moudgil A, Martz K, Stablein DM, et al. Variables affecting estimated glomerular filtration rate after renal transplantation in children: a NAPRTCS data analysis. Pediatr Transplant. 2010;14(2):288–294. 16 Pape L, Strehlau J, Henne T, et al. Single centre experience with basiliximab in paediatric renal transplantation. Nephrol Dial Transplant. 2002;17(2):276–280. 17 Martinez-Mier G, Enriquez-De Los Santos H, Mendez-Lopez MT, et al. Rejection is a strong graft survival predictor in live donor pediatric renal transplantation using cyclosporine, mycophenolate mofetil, and steroids: 5-year outcomes in a single Mexican center. Transplant Proc. 2013;45(4):1442–1444. 18					Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IL-2 receptor antibodies on patient mortality. Graft loss: low ⊕⊕⊝⊝ IL-2 receptor antibodies may result in a slight reduction in graft loss compared with no induction. Graft function: moderate ⊕⊕⊕⊝ IL-2 receptor antibodies probably result in little to no difference in graft function compared with no induction. Incidence of BPAR: low ⊕⊕⊝⊝ IL-2 receptor antibodies may slightly reduce incidence of BPAR compared with no induction. Incidence of malignancies: low ⊕⊕⊝⊝ IL-2 receptor antibodies may result in a slight reduction in incidence of malignancies compared with no induction. Incidence of infection (including opportunistic infections): moderate ⊕⊕⊕⊝ IL-2 receptor antibodies probably slightly increased incidence of infection (including opportunistic infections) compared with no induction. Incidence of opportunistic infections: very low ⊕⊝⊝⊝ The evidence is very unclear about the effect of IL-2 receptor antibodies on incidence of opportunistic infections compared with no induction. Quality of life, incidence of serious adverse events, and incidence of hospitalization: no evidence was identified in the systematic evidence search.
Strength of consensus 1a) 93% 1b) 90%

2		Recommendation		2025		Grade of recommendation/LoE
c) The use of an IL2RA induction therapy (basiliximab) as part of an immunosuppression regimen comprising cyclosporine A (CsA), azathioprine (AZA), and prednisolone may be considered. d) No recommendation can be made on the use of induction with IL2RA as part of an immunosuppression regimen comprising low-dose calcineurin inhibitors (CNIs) in combination with everolimus. e) No recommendation can be made on the use of induction with IL2RA in children and adolescents receiving a first KTx with a maintenance immunosuppressive regimen in combination with tacrolimus, AZA, and prednisolone.						Expert consensus
Strength of consensus 2a) 92.86% 2b) 75.86%

BPAR is a major risk factor for graft dysfunction (adjusted hazard ratio, 2.48, 95% confidence interval [CI], 1.44-4.30) and the development of antibody-mediated rejection (AMR).¹⁹

Early randomized trials in the era of CsA- and AZA-based immunosuppressive therapy showed that induction with rabbit antithymocyte globulin (rATG) or IL2RA reduces early acute rejection,²⁰ leading to recommendations by the Kidney Disease Improving Global Outcomes (KDIGO)²¹ group that IL2RA induction be used routinely in first-line therapy after KTx, with lymphocyte-depleting induction with rATG reserved for high-risk patients. However, these studies mainly used outdated maintenance regimens. No large, randomized trial has examined the effect of IL2RA versus no induction in adult or pediatric patients receiving tacrolimus, MMF, and steroids.

The only IL2RA currently available is basiliximab. A cross-study comparison revealed that MMF reduces basiliximab clearance and prolongs CD25 saturation duration from approximately 5 wk in the absence of MMF to 10 wk in its presence.²² Basiliximab led to a marked reduction of CD25⁺ T-cell fraction during the first 8–10 wk posttransplant but did not specifically affect CD122⁺ T cells. The majority of BPARs occurred after IL-2 receptor desaturation, whereas about a quarter occurred despite adequate IL-2R blockade.²²

With triple maintenance therapy consisting of tacrolimus, MMF, and steroids, the addition of induction therapy may achieve an absolute risk reduction of only 1%–4% for acute rejection in standard-risk adult patients, without improving graft or patient survival.²³ Pooled analysis of pediatric study data indicated that IL2RA, compared with no induction, may result in little to no difference in graft loss after 6 mo, a slight reduction in graft loss after 1 to 2 y,^4,5 a slight reduction in BPAR at 6 mo but not after 2 y,^3,4,13 and a slight reduction in malignancy.^3,4 There was moderate certainty of evidence for a likely increased incidence of infection and a slight decrease in graft function after 6 mo but not after 1 to 2 y, posttransplant in children and adolescents receiving IL2RA.^3,4 The group considered the low rates of acute rejection in children and adolescents receiving tacrolimus, mycophenolic acid (MPA), and steroids, with graft survival and function similar to those receiving IL2RA induction and early steroid withdrawal in the TWIST trial.²⁴ Another important factor to consider in this discussion is that basiliximab treatment is expensive. This needs to be considered, as the resource implications of a change in practice in Europe are unclear. It was noted that in 2 prospective randomized trials, the authors concluded that basiliximab provided no additional benefit in preventing acute rejection compared with tacrolimus, AZA, and steroids^4,13 or a regimen of CsA, MMF, and steroids.³ However, target tacrolimus levels in these studies were higher (10–20 ng/mL until day 21) than in current practice, and study patients may have been at lower immunological risk than many children and adolescents undergoing KTx.

Induction With Polyclonal Antibodies (Literature Pediatric KTx)

3	Recommendation	2025	Grade of recommendation
a) In children and adolescents receiving a first KTx, we suggest induction with polyclonal antibodies in patients with high immunological risk. b) In children and adolescents receiving a first KTx, we suggest that induction with polyclonal antibodies could be reserved if planned maintenance therapy comprises early steroid withdrawal. c) No agreement. Therefore, no recommendation can be made regarding induction with polyclonal antibodies in children and adolescents receiving maintenance therapy with tacrolimus, MMF, and long-term prednisolone.			2 ⇑ 2 ⇑
Baron PW, Ojogho ON, Yorgin P, et al. Comparison of outcomes with low-dose anti-thymocyte globulin, basiliximab or no induction therapy in pediatric kidney transplant recipients: a retrospective study. Pediatr Transplant. 2008;12(1):32–39. 8 Mota C, Martins L, Costa T, et al. Nineteen years of experience utilizing anti-T-lymphocyte globulin induction in pediatric kidney transplantation. Ann Transplant. 2010;15(4):84–91. 25 Sampaio MS, Poommipanit N, Kuo HT, et al. Induction therapy in pediatric kidney transplant recipients discharged with a triple drug immunosuppressive regimen. Pediatr Transplant. 2010;14(6):770–778. 14 Puliyanda DP, Stablein DM, Dharnidharka VR. Younger age and antibody induction increase the risk for infection in pediatric renal transplantation: a NAPRTCS report. Am J Transplant. 2007;7(3):662–666. 26			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of polyclonal antibodies on patient mortality compared with no induction. Graft loss: moderate ⊕⊕⊕⊝ Polyclonal antibodies probably result in little to no difference in graft loss compared with no induction. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of polyclonal antibodies on graft function at 12 mo compared with no induction. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of polyclonal antibodies on incidence of BPAR compared with no induction. Incidence of malignancies: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of polyclonal antibodies on incidence of malignancies compared with no induction. Incidence of infection (including opportunistic infections): low ⊕⊕⊝⊝ Polyclonal antibodies may result in little to no difference in incidence of infection (including opportunistic infections) compared with no induction. Quality of life, incidence of serious adverse events, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 3a) 95% 3b) 80% 3c) 57%

While the efficacy of ATG is not disputed, the numerous short- and long-term side effects should be considered in the risk–benefit assessment compared with other less toxic therapies. ATG is associated with higher rates of cytomegalovirus (CMV) disease, other opportunistic infections, cytokine-related infusion reactions, and a markedly increased risk of posttransplant malignancy, particularly lymphoproliferative disease (see below).^27-30

Evidence accumulated during the past 2 decades has reshaped guideline recommendations. The 2009 KDIGO guidelines in adults advise reserving ATG for recipients with high immunological risk (sensitization, repeat graft, or African ancestry).^21,31 Long-term outcomes of ATG have been compared with those of IL2R antibodies. A 2010 Cochrane meta-analysis in adult KTx recipients showed better 1-y allograft survival with IL2RA induction versus no induction, but when IL2RA was compared with ATG (16 studies, 2211 participants), whereas there was no difference in graft loss there was a benefit for ATG in BPAR at 1 y (8 studies: relative risk (RR), 1.30; [95% CI, 1.01-1.67]). However, this was at the cost of a 75% increase in malignancy (7 studies: RR, 0.25 [95% CI, 0.07-0.87]) and a 32% increase in CMV disease (13 studies: RR, 0.68 [95% CI, 0.50-0.93]). Interestingly, serum creatinine was significantly lower in IL2RA-treated patients at 6 mo (4 studies; mean difference, −11.2 μmol/L [95% CI, −19.94 to −2.09]). ATG patients experienced significantly more fever, cytokine release syndrome, leukopenia, and other adverse reactions associated with ATG administration. The results were independent of the CNI, the antimetabolite, and the baseline immunological risk of the study population. There was no evidence that effects differed between equine and rATG.²⁰ The authors concluded that, compared with IL2RA induction therapy, ATG may prevent more acute rejections, but compared with IL2RA, 1 of 16 patients will develop an additional CMV infection, and 1 of 58 patients will develop an additional malignancy. While the meta-analysis focused on 1-y outcomes, other recent articles reported long-term outcomes showing similar efficacy and safety for ATG and basiliximab.³² Hellemans et al³³ demonstrated the benefit of ATG only for 5-y BPAR in high-immunological-risk patients. In contrast, no benefit regarding acute rejection or allograft survival was demonstrated in low-risk patients.³⁴

Overall, longer-term studies have not shown superior graft survival with ATG, and any advantage is confined to high-risk cohorts. Conversely, in low-risk recipients—and in strategies that delay the start of CNIs or aim for steroid avoidance—T cell–depleting induction confers no durable benefit. Overall, current evidence supports the use of ATG only when the immunological benefits clearly outweigh its infectious and oncologic risks and favors IL2RA or no induction in standard-risk patients. Furthermore, robust evidence is lacking to support ATG induction when CNIs are intentionally deferred to facilitate graft recovery after ischemic injury.³⁵ In the context of steroid-sparing regimens, induction with polyclonal antibodies may be considered for early administration of low-dose CNIs in combination with everolimus; however, there is no scientific evidence to support this approach. The long-term superiority of these regimens remains unproven, but T cell–depleting induction therapy has been shown to reduce early acute rejection in this context.³⁶ As there is no clear evidence on benefits versus harms using ATG induction in low-risk patients, no agreement on a recommendation could be reached.

KQ1b: Initial Maintenance Immunosuppressive Therapy (Literature Pediatric KTx)

Standard initial maintenance immunosuppressive therapy after pediatric KTx typically involves a combination of a CNI and an antimetabolite with or without corticosteroids in steroid minimization or withdrawal protocols. This triple therapy aims to prevent acute rejection while balancing the risk of infection, malignancy, and long-term toxicity.

Antimetabolites

4	Recommendation	2025	Grade of recommendation/LoE
d) In children and adolescents after KTx we suggest using MMF rather than AZA in combination with a CNI with or without corticosteroids as initial maintenance immunosuppression. Potential side effects might affect clinician’s choice.			2 ⇑
Cransberg K, Marlies Cornelissen EA, Davin JC, et al. Improved outcome of pediatric kidney transplantations in the Netherlands—effect of the introduction of MMF mofetil? Pediatr Transplant. 2005;9(1):104–111. 37 Almasi-Hashiani A, Rajaeefard A, Hassanzade J, et al. Graft survival rate in pediatric renal transplantation: a single center experience. Pak J Med Sci. 2012;28(5):9. 38 Jungraithmayr TC, Wiesmayr S, Staskewitz A, et al. Five-year outcome in pediatric patients with MMF mofetil-based renal transplantation. Transplantation. 2007;83(7):900–905. 39 Ojogho ON, Sahney S, Cutler D, et al. Mycophenolate mofetil without antibody induction in pediatric renal transplantation. Transplant Proc. 2002;34(5):1953–1954. 40 Velez C, Zuluaga G, Ocampo C, et al. Clinical description and evolution of renal transplant pediatric patients treated with alemtuzumab. Transplant Proc. 2011;43(9):3350–3354. 41 Garcia CD, Schneider L, Barros VR, et al. Pediatric renal transplantation under tacrolimus or cyclosporine immunosuppression and basiliximab induction. Transplant Proc. 2002;34(7):2533–2534. 42 Staskewitz A, Kirste G, Tonshoff B, et al. Mycophenolate mofetil in pediatric renal transplantation without induction therapy: results after 12 months of treatment. German Pediatric Renal Transplantation Study Group. Transplantation. 2001;71(5):638–644. 43 Jungraithmayr T, Staskewitz A, Kirste G, et al. Pediatric renal transplantation with MMF mofetil-based immunosuppression without induction: results after three years. Transplantation. 2003;75(4):454–461. 44 Otukesh H, Sharifian M, Basiri A, et al. Mycophenolate mofetil in pediatric renal transplantation. Transplant Proc. 2005;37(7):3012–3015. 45			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on patient mortality compared with AZA. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on graft loss compared with AZA. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on graft function compared with AZA. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on incidence of BPAR compared with AZA. Incidence of malignancies: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on incidence of malignancies compared with AZA. Incidence of infection (including opportunistic infections): very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF on incidence of infections (including opportunistic infections) compared with AZA. Quality of life, incidence of serious adverse events, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 86%

The comparative safety and effectiveness of antimetabolites were considered as part of the initial maintenance immunosuppressive regimen for children undergoing a first KTx. The focus is on comparing MMF with AZA across relevant outcomes. The evidence synthesis primarily relies on observational studies and cohort data concerning CNIs and steroids as comedications. Despite the very low certainty of evidence, MMF has emerged as the more frequently used agent, in combination with a CNI, because of evidence in adults that MMF is superior to AZA for the prevention of BPAR.⁴⁶ Nevertheless, conclusions are limited by the lack of RCTs in pediatric KTx recipients and the high risk of bias in the included observational studies.

The main reason for the recommendation below to use MMF rather than AZA was 3 observational studies showing a reduced incidence of BPAR in children and adolescents treated with MMF.^37,40,43 Although the certainty of evidence is low due to the critical risk of bias, the group decided on a conditional recommendation in favor of MMF based on the consistency of the positive effect on BPAR across all 3 studies. These findings are consistent with evidence in adults that MMF is superior to AZA for the prevention of BPAR.⁴⁶

However, the side effects of MMF and AZA differ: MMF is associated with more bone marrow suppression and gastrointestinal side effects. This can influence the clinician’s choice of antimetabolite.

CNIs (Literature Pediatric KTx)

5	Recommendation	2025	Grade of recommendation
In children and adolescents after KTx we recommend using tacrolimus rather than CsA in combination with an antimetabolite with or without corticosteroids as initial maintenance immunosuppression.			1 ⇑⇑
Filler G, Webb NJ, Milford DV, et al. Four-year data after pediatric renal transplantation: a randomized trial of tacrolimus vs. cyclosporin microemulsion. Pediatr Transplant. 2005;9(4):498–503. 47 Trompeter R, Filler G, Webb NJ, et al. Randomized trial of tacrolimus versus cyclosporin microemulsion in renal transplantation. Pediatr Nephrol. 2002;17(3):141–149. 48 Filler G, Trompeter R, Webb NJ, et al. One-year glomerular filtration rate predicts graft survival in pediatric renal recipients: a randomized trial of tacrolimus vs cyclosporine microemulsion. Transplant Proc. 2002;34(5):1935–1938. 49 Kizilbash SJ, Rheault MN, Bangdiwala A, et al. Infection rates in tacrolimus versus cyclosporine-treated pediatric kidney transplant recipients on a rapid discontinuation of prednisone protocol: 1-year analysis. Pediatr Transplant. 2017;21(4). 50 Neu AM, Ho PL, Fine RN, et al. Tacrolimus vs. cyclosporine A as primary immunosuppression in pediatric renal transplantation: a NAPRTCS study. Pediatr Transplant. 2003;7(3):217–222. 51 Moudgil A, Martz K, Stablein DM, et al. Variables affecting estimated glomerular filtration rate after renal transplantation in children: a NAPRTCS data analysis. Pediatr Transplant. 2010;14(2):288–294. 16 Hocker B, van Gelder T, Martin-Govantes J, et al. Comparison of MMF efficacy and safety in paediatric vs. adult renal transplantation: subgroup analysis of the randomised, multicentre FDCC trial. Nephrol Dial Transplant. 2011;26(3):1073–1079. 52 Lancia P, Aurich B, Ha P, et al. Adverse events under tacrolimus and cyclosporine in the first 3 years post-renal transplantation in children. Clin Drug Invest. 2018;38(2):157–171. 53 Velez C, Zuluaga G, Ocampo C, et al. Clinical description and evolution of renal transplant pediatric patients treated with alemtuzumab. Transplant Proc. 2011;43(9):3350–3354. 41 Aoun B, Decramer S, Vitkevic R, et al. Protocol biopsies in pediatric renal transplant recipients on cyclosporine versus tacrolimus-based immunosuppression. Pediatr Nephrol. 2013;28(3):493–498. 54			Mortality: low ⊕⊕⊝⊝ Tacrolimus may result in little to no difference in patient mortality. Graft loss: moderate ⊕⊕⊕⊝ Tacrolimus likely reduces graft loss. Graft function: moderate ⊕⊕⊕⊝ Tacrolimus likely increases graft function. Incidence of BPAR: low ⊕⊕⊝⊝ Tacrolimus may reduce incidence of BPAR. Incidence of malignancies: moderate ⊕⊕⊕⊝ Tacrolimus likely results in little to no difference in incidence of malignancies. Incidence of infection (including opportunistic infections): low ⊕⊕⊝⊝ Tacrolimus may slightly increase incidence of infection (including opportunistic infections). Quality of life, incidence of serious adverse events, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 93%

Tacrolimus versus CsA as initial maintenance immunosuppression after pediatric KTx leads to improved graft survival and graft function in the first 4 y and a reduced incidence of BPARs in the first 6 mo posttransplant.^47,48 The group acknowledged a slightly increased risk of infection with tacrolimus compared with CsA; however, they judged this to be substantially outweighed by the benefits and agreed to make this a strong recommendation.

Steroid Minimization (Literature Pediatric KTx)

6	Recommendation	2025	Grade of recommendation/LoE
We suggest early glucocorticoid withdrawal or avoidance in pediatric renal transplant recipients with low immunologic risk, if potent immunosuppressive therapy with tacrolimus and MMF in combination with anti-interleukin-2 receptor antibodies induction is used.			2 ⇑
Mericq V, Salas P, Pinto V, et al. Steroid withdrawal in pediatric kidney transplant allows better growth, lipids and body composition: a randomized controlled trial. Horm Res Paediatr. 2013;79(2):88–96. 55 Sarwal MM, Ettenger RB, Dharnidharka V, et al. Complete steroid avoidance is effective and safe in children with renal transplants: a multicenter randomized trial with three-year follow-up. Am J Transplant. 2012;12(10):2719–2729. 56 Grenda R, Watson A, Trompeter R, et al. A randomized trial to assess the impact of early steroid withdrawal on growth in pediatric renal transplantation: the TWIST study. Am J Transplant. 2010;10(4):828–836. 24 Li L, Chang A, Naesens M, et al. Steroid-free immunosuppression since 1999: 129 pediatric renal transplants with sustained graft and patient benefits. Am J Transplant. 2009;9(6):1362–1372. 57 Riad S, Jackson S, Chinnakotla S, et al. Primary pediatric live-donor-kidney transplant-recipients’ outcomes by immunosuppression induction received in the United States. Pediatr Transplant. 2021;25(5):e13925. 58 Riad S, Jackson S, Chinnakotla S, et al. Primary pediatric deceased-donor kidney transplant recipients outcomes by immunosuppression induction received in the United States. Pediatr Transplant. 2021;25(5):e13928. 59 Delucchi A, Valenzuela M, Lillo AM, et al. Early steroid withdrawal in pediatric renal transplant: five years of follow-up. Pediatr Nephrol. 2011;26(12):2235–2244. 60 Nehus E, Goebel J, Abraham E. Outcomes of steroid-avoidance protocols in pediatric kidney transplant recipients. Am J Transplant. 2012;12(12):3441–3448. 61 Delucchi BA, Valenzuela AM, Ferrario BM, et al. Early steroid withdrawal in pediatric renal transplantation. Rev Med Chil. 2006;134(11):1393–1401. 62 Weaver DJ Jr, Selewski D, Janjua H, et al. Improved cardiovascular risk factors in pediatric renal transplant recipients on steroid avoidance immunosuppression: a study of the Midwest Pediatric Nephrology Consortium. Pediatr Transplant. 2016;20(1):59–67. 63 Li L, Weintraub L, Concepcion W, et al. Potential influence of tacrolimus and steroid avoidance on early graft function in pediatric renal transplantation. Pediatr Transplant. 2008;12(6):701–707. 64 Verghese PS, Chinnakotla S, Berglund D, et al. Re-hospitalization after pediatric kidney transplant: a single-center study. Pediatr Transplant. 2020;24(5):e13717. 65 Bhakta N, Marik J, Malekzadeh M, et al. Can pediatric steroid-free renal transplantation improve growth and metabolic complications? Pediatr Transplant. 2008;12(8):854–861. 66 Mosaad AM, Farouk Aziz Hamdy A, Abd El-Fattah Hassan NM, et al. Evaluation of live-donor kidney transplant survival in low body weight Egyptian children: 25 year-experience. Dialisis y Trasplante. 2012;33(1):1–8. 67 Naesens M, Salvatierra O, Benfield M, et al. Subclinical inflammation and chronic renal allograft injury in a randomized trial on steroid avoidance in pediatric kidney transplantation. Am J Transplant. 2012;12(10):2730–2743. 68 Nehus EJ, Liu C, Lu B, et al. Graft survival of pediatric kidney transplant recipients selected for de novo steroid avoidance—a propensity score-matched study. Nephrol Dial Transplant. 2017;32(8):1424–1431. 69 Webb NJ, Douglas SE, Rajai A, et al. Corticosteroid-free kidney transplantation improves growth: 2-year follow-up of the TWIST randomized controlled trial. Transplantation. 2015;99(6):1178–1185. 70			Mortality: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid–avoidance, or steroid-free regimens may result in little to no difference in patient mortality. Graft loss: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid-avoidance, or steroid-free regimens may reduce graft loss slightly. Graft function: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid-avoidance, or steroid-free regimens may result in little to no difference in graft function. Incidence of BPAR: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid-avoidance or steroid-free regimens may result in little to no difference in incidence of BPAR. Incidence of serious adverse events: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid–avoidance, or steroid-free regimens may result in little to no difference in incidence of serious adverse events. Incidence of malignancies: low ⊕⊕⊝⊝ Early steroid withdrawal, steroid–avoidance, or steroid-free regimens may result in little to no difference in incidence of malignancies. Incidence of hospitalization: moderate ⊕⊕⊕⊝ Early steroid withdrawal, steroid–avoidance, or steroid-free regimens likely result in little to no difference in incidence of hospitalization. Incidence of infection (including opportunistic infections): low ⊕⊕⊝⊝ Early steroid withdrawal, steroid-avoidance, or steroid-free regimens may result in little to no difference in incidence of infection (including opportunistic infections). Quality of life and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 93%

Steroid minimization protocols are implemented primarily to mitigate or avoid steroid side effects, such as growth impairment, metabolic disorders (including overweight/obesity, diabetes, arterial hypertension, hyperlipidemia), skeletal (osteopenia, osteoporosis), ophthalmological (cataracts, glaucoma), and cosmetic side effects.

The group considered the evidence from the TWIST trial²⁴ that early steroid withdrawal significantly improves growth and lipid profiles without increasing rejections or graft loss despite similar exposure to tacrolimus and MMF. The group therefore recommended early steroid withdrawal for children and adolescents undergoing a first KTx if a comparable immunosuppressive regimen including IL2RA induction therapy is used. The group also noted that the adoption of the TWIST protocol has become established practice in many countries. However, it should be taken into account that in the TWIST publication, only BPAR with T cell–mediated rejection (TCMR) grade IA or higher according to the Banff classification was reported, whereas borderline rejection and treated rejection were not included, which may have biased the results in favor of early steroid withdrawal. In addition, the TWIST trial used induction therapy with 2 doses of daclizumab, which are no longer available on the market. Although immunosuppressive efficacy and safety of daclizumab and basiliximab appear to be comparable, some uncertainty remains.

The group voted for a conditional recommendation because some patients might not benefit from early steroid withdrawal, especially patients with higher immunological risk. Although these risk states are ill-defined, we decided to limit the recommendation to low-immunological-risk patients (as described earlier). In the TWIST trial, all patients had PRAs <50% and almost all had no prior KTx. The effect of steroid sparing on growth has been shown to be especially beneficial in prepubertal patients.⁷⁰

Mechanistic Target of Rapamycin Inhibitors (Literature Pediatric KTx)

7	Recommendation	2025	Grade of recommendation/LoE
In children and adolescents after KTx, with a higher risk of CMV infection, we suggest using either an immunosuppressive regimen of low-dose CNIs and everolimus, or a standard regimen with CNIs and MMF.			2 ⇑
Tonshoff B, Tedesco-Silva H, Ettenger R, et al. Three-year outcomes from the CRADLE study in de novo pediatric kidney transplant recipients receiving everolimus with reduced tacrolimus and early steroid withdrawal. Am J Transplant. 2021;21(1):123–137. 71 Hocker B, Zencke S, Pape L, et al. Impact of everolimus and low-dose cyclosporin on cytomegalovirus replication and disease in pediatric renal transplantation. Am J Transplant. 2016;16(3):921–929. 72 Brunkhorst LC, Fichtner A, Hocker B, et al. Efficacy and safety of an everolimus- vs. mycophenolate mofetil-based regimen in pediatric renal transplant recipients. PLoS ONE. 2015;10(9):e0135439. 12 Tonshoff B, Ettenger R, Dello Strologo L, et al. Early conversion of pediatric kidney transplant patients to everolimus with reduced tacrolimus and steroid elimination: results of a randomized trial. Am J Transplant. 2019;19(3):811–822. 73 Billing H, Burmeister G, Plotnicki L, et al. Longitudinal growth on an everolimus- versus an MMF-based steroid-free immunosuppressive regimen in paediatric renal transplant recipients. Transpl Int. 2013;26(9):903–909. 15 Kanzelmeyer NK, Ahlenstiel T, Drube J, et al. Protocol biopsy-driven interventions after pediatric renal transplantation. Pediatr Transplant. 2010;14(8):1012–108. 74			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of mechanistic target of rapamycin inhibitors (mTORi) on patient mortality compared with antimetabolites. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of mTORi on graft loss compared with antimetabolites. Graft function: low ⊕⊕⊝⊝ mTORi may result in little to no difference in graft function compared with antimetabolites. Incidence of BPAR: low ⊕⊕⊝⊝ mTORi may result in little to no difference in incidence of BPAR compared with antimetabolites. Incidence of serious adverse events: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of mTORi on the incidence of serious adverse events compared with antimetabolites. Incidence of malignancies: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of mTORi on the incidence of malignancy compared with antimetabolites. Incidence of infection (including opportunistic infections): low ⊕⊕⊝⊝ mTORi may result in little to no difference in incidence of infection (including opportunistic infections) compared with antimetabolites. Quality of life, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 76%
8	Recommendation	2025	Grade of recommendation
We recommend that mTORi, if used as initial maintenance immunosuppression, should only be started after graft function has been established and wound healing is complete.			Expert consensus
Strength of consensus 76%

Due to its relevance for the therapy with mTORi in pediatric KTx recipients, the CRADLE study published in 2021⁷¹ was included in the discussion and recommendation after discussion with the guideline group, although it did not fully meet the a priori-defined eligibility criteria for an initial immunosuppressive regimen, because randomization to either everolimus or standard of care was performed only at weeks 4–6 posttransplant. The overall certainty of the evidence for the benefits and harms of using everolimus instead of MMF, combined with reduced tacrolimus exposure, is very low. In the CRADLE RCT, which suggests that mTORi may achieve rates of infection, graft function, and rejection similar to MMF-based regimens, patients in the everolimus and reduced tacrolimus groups had a higher incidence of study drug discontinuation and adverse events leading to discontinuation. This may reflect inferior tolerability of everolimus in conjunction with reduced tacrolimus compared with standard therapy.^71,75 Thus, the group voted against a general recommendation for everolimus in conjunction with reduced-dose tacrolimus.

However, although there was no difference in overall infections in the CRADLE study, CMV replication and CMV syndrome/disease were numerically less common in patients treated with everolimus and reduced-dose tacrolimus.⁷¹ Therefore, in patients with high-risk CMV serostatus (donor CMV seropositive, recipient CMV seronegative), the benefits of everolimus and reduced tacrolimus might outweigh the harms, although the certainty of the evidence was very low, as statistical significance was not achieved. Based on these considerations, the group voted for a conditional recommendation for either everolimus and reduced tacrolimus or standard therapy in patients at high risk of CMV infection after KTx.

If mTORi are used as initial maintenance immunosuppression after KTx, the standard of care is to initiate therapy after wound healing is complete, given some evidence of impaired wound healing under mTORi. This is also recommended by the KDIGO guidelines.²¹

KQ2: Late Change of Immunosuppressive Therapy (Literature Pediatric KTx)

9	Recommendation	2025	Grade of recommendation
a) In the context of growth velocity and cardiovascular side effects, we suggest late steroid withdrawal. b) We do not suggest an intensified immunosuppression before steroid withdrawal. c) When changing immunosuppression, contraindications and the tolerability of individual drugs, including CNI plus MMF or mTORi, ought to be taken into account.			2 ⇑ 2 ⇑ 2 ⇑
Hocker B, Weber LT, Feneberg R, et al. Improved growth and cardiovascular risk after late steroid withdrawal: 2-year results of a prospective, randomised trial in paediatric renal transplantation. Nephrol Dial Transplant. 2010;25(2):617–624. 76 Benfield MR, Bartosh S, Ikle D, et al. A randomized double-blind, placebo controlled trial of steroid withdrawal after pediatric renal transplantation. Am J Transplant. 2010;10(1):81–88. 77 McDonald RA, Smith JM, Ho M, et al. Incidence of PTLD in pediatric renal transplant recipients receiving basiliximab, calcineurin inhibitor, sirolimus and steroids. Am J Transplant. 2008;8(5):984–989. 78 Hocker B, Weber LT, Feneberg R, et al. Prospective, randomized trial on late steroid withdrawal in pediatric renal transplant recipients under cyclosporine microemulsion and mycophenolate mofetil. Transplantation. 2009;87(6):934–941. 79 Chakrabarti P, Wong HY, Scantlebury VP, et al. Outcome after steroid withdrawal in pediatric renal transplant patients receiving tacrolimus-based immunosuppression. Transplantation. 2000;70(5):760–764. 80 Ellis D. Growth and renal function after steroid-free tacrolimus-based immunosuppression in children with renal transplants. Pediatr Nephrol. 2000;14(7):689–694. 81 Laube GF, Falger J, Kemper MJ, et al. Selective late steroid withdrawal after renal transplantation. Pediatr Nephrol. 2007;22(11):1947–1952. 82 Hocker B, John U, Plank C, et al. Successful withdrawal of steroids in pediatric renal transplant recipients receiving cyclosporine A and mycophenolate mofetil treatment: results after four years. Transplantation. 2004;78(2):228–234. 83			Mortality: low ⊕⊕⊝⊝ Late steroid withdrawal may reduce patient mortality compared with steroid maintenance. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of late steroid withdrawal on graft loss compared with steroid maintenance. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of late steroid withdrawal on graft function compared with steroid maintenance. Incidence of BPAR: low ⊕⊕⊝⊝ Late steroid withdrawal may reduce incidence of BPAR compared with steroid maintenance. Incidence of malignancies: moderate ⊕⊕⊕⊝ Late discontinuation of glucocorticoid therapy is likely to result in a higher incidence of malignancies compared with glucocorticoid maintenance therapy. Incidence of opportunistic infections: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of late steroid withdrawal on incidence of opportunistic infections compared with steroid maintenance. Quality of life, incidence of serious adverse events, incidence of hospitalization, and incidence of overall infections: no evidence was identified in the systematic evidence search.
Strength of consensus 9a) 79.31% 9b) 85.71% 9c) 92.1%

10	Recommendation	2025	Grade of recommendation/LoE
A recommendation on conversion from MMF to sirolimus in pediatric patients with moderate to severe interstitial fibrosis and tubular atrophy (IF/TA) and CNI minimization cannot be given (no consensus).			0
Blydt-Hansen TD, Gibson IW, Birk PE. Histological progression of chronic renal allograft injury comparing sirolimus and mycophenolate mofetil-based protocols. A single-centre, prospective, randomized, controlled study. Pediatr Transplant. 2010;14(7):909–918. 84			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on patient mortality compared with continuation of MMF. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on graft loss compared with continuation of MMF. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on graft function compared with continuation of MMF. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on incidence of BPAR compared with continuation of MMF. Quality of life, incidence of serious adverse events, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 38.46%

11	Recommendation	2025	Grade of recommendation
No recommendation can be given on the discontinuation of MMF in pediatric patients with significant side effects as there was no consensus.			0
Moudgil A, Sgambat K, Benoit E, et al. Prevalence of mycophenolate mofetil discontinuation and subsequent outcomes in pediatric kidney transplant recipients: a PNRC study. Pediatr Transplant. 2024;28(1):e14628. 85			Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF discontinuation on graft function compared with MMF continuation. Incidence of hospitalization: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of MMF discontinuation on incidence of hospitalization compared with MMF continuation. Mortality, graft loss, incidence of BPAR, quality of life, incidence of serious adverse events, incidence of malignancies, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 48%

12	Recommendation	2025	Grade of recommendation
We do not recommend CNI withdrawal and switch to sirolimus-based therapy compared with CNI minimization in pediatric KTx recipients with declining graft function because of CNI-induced nephrotoxicity.			1 ⇓⇓
Hocker B, Feneberg R, Kopf S, et al. SRL-based immunosuppression vs. CNI minimization in pediatric renal transplant recipients with chronic CNI nephrotoxicity. Pediatr Transplant. 2006;10(5):593–601. 86			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on patient mortality compared with CNI minimization. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on graft mortality compared with CNI minimization. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on graft function compared with CNI minimization. Incidence of biopsy-proven acute rejection: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on incidence of BPAR compared with CNI minimization. Incidence of opportunistic infections: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of conversion to sirolimus on incidence of opportunistic infections compared with CNI minimization. Quality of life, incidence of serious adverse events, incidence of malignancies, incidence of hospitalization, and incidence of infection (including opportunistic infections): no evidence was identified in the systematic evidence search.
Strength of consensus 80.77%

The effect of late change of immunosuppressive therapy, beyond 3 mo posttransplant was considered with the term “change of immunosuppression,” comprising either late steroid withdrawal or conversion to sirolimus versus MMF continuation in patients with IF/TA, or MMF discontinuation, or conversion to sirolimus versus CNI minimization in chronic CNI nephrotoxicity.

It has to be emphasized that the change in height z-score before and after steroid withdrawal was the primary outcome in all randomized studies evaluated in this section. Patient mortality and allograft failure were evaluated only as secondary outcomes. It is therefore not surprising that the evidence on the effect of late steroid withdrawal on the main outcomes of this guideline is low or very low.

Höcker et al^76,79 reported in an RCT that there were no differences in patient mortality or graft failure after steroid withdrawal compared with maintaining steroids. Notably, the study cohort comprised low- or standard-immunological-risk patients receiving an immunosuppressive regimen consisting of CsA and MMF. In the cohort reported by Benfield et al⁷⁷ and McDonald et al⁷⁸ (n = 274), 10 cases of PTLD occurred in the prerandomization phase and were ascribed to the intensive immunosuppression regimen (induction therapy with an anti-CD25 antibody, maintenance therapy with standard-dose tacrolimus and sirolimus) applied in the initial 6 mo posttransplant to facilitate the prospective steroid withdrawal. These malignancies should not be considered the result of steroid withdrawal itself, as there is no pathophysiologic rationale for this. Therefore, the conclusion from this trial is that in patients in whom late steroid withdrawal is anticipated, the initial immunosuppressive regimen should not be intensified. In addition, an increase in the incidence of malignancies after steroid withdrawal was not observed in the literature from adult studies.⁸⁷

Based on clinical experience, the guideline group concluded that late withdrawal of steroids is usually easily accepted and even desired by patients. Late steroid withdrawal (as discussed earlier) may be especially beneficial in prepubertal patients, particularly in the context of optimization of growth.

The development of IF/TA and glomerulosclerosis is associated with shorter allograft survival. The pathomechanism is multifactorial, including donor-derived (dd) lesions, ischemia/reperfusion injury, and posttransplant complications such as urinary tract infection and rejection. CNI exposure is a well-known contributor to chronic lesions, justifying strategies to minimize exposure to CNI.

Regarding CNI switch to sirolimus in patients with IF/TA, no benefit in short-term allograft survival or preservation of glomerular filtration rate was observed, but there was a higher risk of rejection and lower tolerance to the regimen, leading to proteinuria and oral ulcers. Some members of the panel, however, had positive experiences with this switch despite the absence of evidence. The expert panel did not reach consensus, and therefore, no recommendation can be given.

The main reasons for discontinuing MMF were gastrointestinal disturbances, infections, and other side effects. In 85% of children with MMF discontinuation, MMF was replaced with other immunosuppressive agents, such as AZA, enteric-coated mycophenolate, sirolimus, or leflunomide. The certainty of the evidence regarding a beneficial effect on graft function and hospitalization incidence was very low, as no controlled trials were available. Moreover, there was no evidence to guide the immunosuppressive treatment for patients with significant side effects from MMF. We did not identify any pediatric studies that directly compared MMF with enteric-coated MPA. In practice, clinicians may explore alternative contributing factors and consider adjusting other medications. If required, the MMF dose can be reduced or divided into 3 doses a day before stopping it altogether. Important indications for stopping MMF are severe neutropenia, severe infections, bronchiectasis, and severe gastrointestinal symptoms leading to weight loss in the absence of other causes. Some experts had positive experiences with reducing/discontinuing MMF in these cases, but without clear evidence, no consensus could be reached.

KQ3: Steering of Immunosuppression (Literature Pediatric KTx)

13	Recommendation	2025	Grade of recommendation/LoE
We suggest monitoring of virus-specific T cells (Tvis) in addition to trough level monitoring to avoid overimmunosuppression and provide an “effect-related drug monitoring” in children after KTx.			2 ⇑
Ahlenstiel-Grunow T, Liu X, Schild R, et al. Steering transplant immunosuppression by measuring virus-specific T cell levels: the randomized, controlled IVIST trial. J Am Soc Nephrol. 2021;32(2):502–516. 88			Mortality: low ⊕⊕⊝⊝ Tvis in combination with trough level steering may result in little to no difference in patient mortality compared with trough level steering alone. Graft loss: low ⊕⊕⊝⊝ Tvis in combination with trough level steering may result in little to no difference in graft loss compared with trough level steering alone. Graft function: low ⊕⊕⊝⊝ Tvis in combination with trough level steering may result in little to no difference in graft function compared with trough level steering alone. Incidence of BPAR: moderate ⊕⊕⊕⊝ Tvis in combination with trough level steering likely reduces incidence of BPAR slightly compared with trough level steering alone. Incidence of serious adverse events: low ⊕⊕⊝⊝ Tvis in combination with trough level steering may result in little to no difference in incidence of serious adverse events compared with trough level steering alone. Incidence of infection (including opportunistic infections): moderate ⊕⊕⊕⊝ Tvis in combination with trough level steering likely results in little to no difference in incidence of infection (including opportunistic infections) compared with trough level steering alone. Quality of life, incidence of malignancies, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 76.92%

It is crucial to achieve an optimal individual balance between over- and underimmunosuppression, thereby avoiding unnecessary exposure to immunosuppressive drugs and lowering the risk of side effects.

Pharmacokinetic blood level monitoring and fixed-dose regimens, which are used as the standard of care, are insufficient to estimate the individual intensity of immunosuppression. Accordingly, there is an urgent need for methods to improve the steering of immunosuppression. Pharmacodynamic approaches focus on the direct effects of immunosuppressive agents and may provide a more personalized assessment of immunosuppression. Recent methodologies consider viral and donor-specific immunity to guide the administration of immunosuppressive therapy. Only 1 RCT in pediatric KTx, addressing this issue in a relatively small cohort of patients (n = 31 in the intervention group, n = 33 in the control group), was identified in the literature search.⁸⁸ Given the moderate certainty of the evidence from 1 RCT and a thorough weighing of benefits and harms, as explained below, the group decided on a conditional recommendation.

There are no negative effects of Tvis-based immune monitoring on mortality, graft loss, glomerular filtration rate, or the incidence of serious adverse events and infections, whereas Tvis-guided immunosuppressive therapy showed some positive effects in the IVIST trial (the randomized, controlled trial in steering transplant Immunosuppression by measuring Virus-Specific T cell levels).⁸⁸ For example, additional Tvis-based immune monitoring reduced exposure to immunosuppressive drugs without increasing the risk of acute rejection. These data indicate that Tvis-based monitoring allowed the identification of overimmunosuppressed patients receiving more immunosuppression than needed to prevent rejections. Therefore, the balance probably favors intervention.

After pediatric KTx, adenovirus-specific CD4⁺ T cells appear to be most suitable for Tvis-based immune monitoring due to their stability and high prevalence, even in children. In the RCT mentioned earlier (the IVIST trial), Tvis were analyzed by intracellular cytokine staining followed by flow cytometry.⁸⁸ Currently, analysis of virus-specific CD4⁺ T cells by cytokine flow cytometry is performed only in research laboratories. Therefore, Tvis analysis can be considered an emerging clinical tool.

Based on the literature examined, we cannot make any further recommendations regarding the management of immunosuppressive therapy in children after their first KTx. One prospective randomized trial also included pediatric patients to compare a fixed-dose MMF regimen with an interventional concentration-controlled regimen in KTx recipients.⁸⁹ There was no difference in the incidence of treatment failure between the 2 groups because MPA exposure was comparable in both. In contrast, the APOMYGRE study in 137 adult KTx recipients on CsA used Bayesian estimation of MPA-AUC on days 7, 14, and months 1, 3, and 6 with a target AUC_0–12h = 40 mg × h/L and a recommended dose to reach it in the concentration-controlled dosing arm, as compared with a 2 g fixed daily dose in the comparative arm. This study showed a statistically significant and clinically important reduction in a patient’s adverse outcome at 1 y in the concentration-controlled dosing arm, largely due to a highly significant reduction in acute rejection.⁹⁰ Thus, these results favor monitoring MPA using a limited sampling strategy.

Table 1 provides target trough levels and AUC concentrations of immunosuppressants for children after KTx.

TABLE 1.

Target trough levels and AUC concentrations of immunosuppressants for children after kidney transplantation (Tönshoff et al. Management of the paediatric kidney transplant recipient, heiBooks; 2026)

Tacrolimus target trough level (ng/mL) in combination with everolimus	Tacrolimus target trough level (ng/mL) in combination with MMF or AZA	Tacrolimus target AUC (ng × h/mL) in combination with MMF or AZA	Cyclosporine A target trough level (ng/mL) in combination with MMF or AZA	Everolimus target trough levels (ng/mL) in combination with a low-dose calcineurin inhibitor	Everolimus target trough levels (ng/mL) without calcineurin inhibitor	Target MPA-AUC (µg × h/mL) in combination with calcineurin inhibitor
Week 0–3: 5–8 Week 4–month 6: 4–6 >Month 6: 2–4	Week 0–3: 8–12 Week 4–month 6: 7–10 >Month 6: 5–8	Week 1–4: 150–200 Months 1–3: 120–150 >Month 3: 75–150	Months 0–3: 120–200 ng/mL >Month 4: 80–160 ng/mL	Months 2–6: 3–8 >Month 6: 2–5	6–8		>40

AUC, area under the curve; AZA, azathioprine; MMF, mycophenolate mofetil; MPA, mycophenolic acid.

Additional Methods for Steering of Immunosuppressive Therapy (Literature Pediatric and Adult KTx)

Table 2 provides an overview of additional methods for steering immunosuppressive therapy, based on a literature review of adult and pediatric kidney recipients.

TABLE 2.

Tabular summary of recommendations and expert opinions for steering immunosuppressive therapy in children after kidney transplantation

Parameter	Advice from working group	Basis for the recommendation	Practical advice and further information	Referencesa
Therapeutic drug monitoring	TDM of immunosuppressive drugs can help to optimize treatment by avoiding under- and overexposure	Open recommendation of the working group (expert opinion)	Indicated -To reach and control target drug levels -When there is a change in medication or patient status that may affect blood levels -When there is a decline in kidney function -When drug-related toxicity is suspected -In case of (suspected) drug-drug interaction
Pharmacokinetic drug monitoring – standard of care
Calcineurin inhibitors	Measuring calcineurin inhibitor blood levels	Strong advice from the working group (expert opinion)
Cyclosporine A	Monitoring cyclosporine A using either 12-h trough (C0), 2-h postdose (C2), or abbreviated area-under-the-concentration-vs-time-curve (AUC)	Advice from working group (expert opinion supported by literature)		5, S1–S9
Tacrolimus	Monitoring tacrolimus using 12-h trough (C0) levels	Advice from working group (expert opinion supported by literature)	Tacrolimus monitoring by abbreviated AUC may be useful in patients in whom trough levels in the target range cannot be achieved with a relatively high body weight-based dose	5, S8–S16
MPA	TDM of MPA to avoid MPA underexposure and to optimize efficacy in pediatric patients after kidney transplantation. We suggest abbreviated AUC monitoring for MPA	Advice from working group (expert opinion supported by literature)	Predose MPA plasma levels are a rather imprecise marker of MPA exposure. Abbreviated AUCs based on a limited sampling strategy are preferable to estimate MPA exposure. The use of distinct algorithms needs to take into account the cohorts in which they were generated and validated (special focus on concomitant medication). TDM of free MPA may be useful in special situations such as MPA-related toxicity despite MPA concentrations within the assumed therapeutic window. The upper threshold for MPA-AUC is not well defined but may be a useful parameter to facilitate the decision to reduce the dose when the patient has potential MPA-related side effects	38, 39, S17–S117
mTOR inhibitors	Monitoring mTOR inhibitor levels using 12-h trough (C0)	Advice from working group (expert opinion supported by literature)		S11, S118
Pharmacodynamic drug monitoringb			Pharmacodynamic monitoring of immunosuppressive drugs is a valuable research tool but is not yet part of the clinical routine to optimize therapy
dd-cfDNA (emerging clinical tool)	Monitoring of dd-cfDNA for early detection of graft injury (not graded)	Advice from working group (expert opinion supported by literature)	dd-cfDNA is a valuable biomarker approaching clinical use in kidney transplantation. Its use provides information about allograft status, allowing early detection of graft damage, especially in subclinical AMR, both with and without DSA; meanwhile, it is less likely to identify TCMR. An increase in dd-cfDNA has also been observed in cases of pyelonephritis, BKPyV viremia, and viruria Both relative and absolute dd-cfDNA can be measured; changes in recipient cfDNA (eg, by infection) can affect the results of dd-cfDNA fractional determination. This limitation can be overcome using absolute dd-cfDNA quantification. Very preliminary data suggest that absolute values are more sensitive for detecting early allograft injury or rejection, whereas relative quantification may be more useful for longitudinal monitoring. The combination of fractional and absolute determination, including total cfDNA, is recommended for meaningful interpretation of the results Although data in children are limited compared with adults, results follow the same direction Factors such as donor-recipient size mismatch can influence dd-cfDNA levels, and further studies are needed to determine the values of dd-cfDNA that allow early identification of graft dysfunction and subclinical rejection in pediatric recipients	S119–S125
HLA-DSAs (standard of care)	Measuring of HLA-DSAs assesses the risk of AMR	Advice from working group (expert opinion supported by literature)	De novo HLA-DSA formation is predictive of the development of AMR and downstream results in inferior graft survival. Insufficient exposure to immunosuppression is an important factor leading to de novo HLA-DSA formation. Therefore, serial monitoring of DSAs can have a role in patients where there is a reduced exposure to immunosuppression. Once persistent de novo HLA-DSA with high MFI values are detected, early intervention with optimization of immunosuppression can be considered, but this needs to be proven in interventional studies	S125–S130
Tvis (emerging clinical tool)	We suggest monitoring of virus-specific T cells in addition to pharmacokinetic TDM (trough/predose levels) to avoid overimmunosuppression and provide an “effect-related drug monitoring” in children after kidney transplantation	See recommendation 13	After pediatric kidney transplantation, adenovirus-specific CD4+ T cells appear to be most suitable for T cell–based immune monitoring due to their stability and high prevalence, even in children. In the IVIST trial, Tvis were analyzed using intracellular cytokine staining followed by flow cytometry	37
Torque Teno virus (emerging clinical tool)	We suggest monitoring Torque Teno virus plasma load in addition to pharmacokinetic TDM to assess the degree of individual overall immunosuppressive load	Advice from working group (expert opinion supported by literature)	Torque Teno virus plasma load assessment for clinical purposes should be started after month 3 posttransplantation (after initial stabilization of this biomarker). The biomarker Torque Teno virus plasma load shows promising results to assess the individual extent of immunosuppression in CNI-based regimens in adult kidney transplant recipients and for risk stratification of rejection and infection. In pediatric kidney transplantation, information is still limited to a few retrospective studies showing that Torque Teno virus may also be used to assess the individual extent of immunosuppression	S131–S145
Pharmacogenetic screening			Pharmacogenetic screening of patients is a promising research topic and needs further clinical proof before it is suitable for broad clinical use
CYP3A5 (emerging clinical tool)	Pharmacogenetic screening of CYP3A5 to calculate the tacrolimus starting dose based on a dosing algorithm	Advice from working group (expert opinion supported by literature)	The starting dose of tacrolimus is based solely on body weight, and subsequent doses are adjusted using TDM. With this strategy, it may take up to 3 wk before target concentrations are reached. It has been shown that not only body weight but also other covariates influence early tacrolimus exposure, such as co-medication, CYP3A5 genotype, hematocrit, ethnicity, and age. Using a dosing algorithm that includes the CYP3A5 genotype, it has been shown that target tacrolimus concentrations are reached earlier In immunosuppressive regimens where adequate tacrolimus exposure is needed within the first week, CYP3A5 genotype screening and a dosing algorithm may be implemented	S146, S147

^aSee Supplement (SDC, https://links.lww.com/TP/D401) for references.

^bIn addition to pharmacokinetic measurement.

AMR, antibody-mediated rejection; AUC, area under the curve; dd-cfDNA, donor-derived cell-free DNA; HLA-DSA, donor-specific HLA antibody; MFI, mean fluorescence intensity; MPA, mycophenolic acid; mTOR, mechanistic target of rapamycin; TCMR, T cell–mediated rejection; TDM, therapeutic drug monitoring; Tvis, virus-specific T cell.

For this overview, an additional comprehensive literature search using PubMed was carried out with the search terms: mycophenolate mofetil, mycophenolic acid in conjunction with therapeutic drug monitoring, pediatric transplantation, pharmacokinetics, pharmacodynamics, pharmacogenetics, cyclosporine A, tacrolimus, calcineurin inhibitor, mTOR inhibitor, everolimus, efficacy, toxicity, analytic methodology, donor derived cell-free DNA, donor specific antibodies, virus specific T-cells, Torque Teno virus, and future perspectives of immunosuppression. The reference sections of the identified articles were also screened, and relevant articles were extracted. RCTs, clinical studies with sufficient case numbers, and informative reviews were considered to be relevant.

KQ4: Protocol Biopsies (Literature Pediatric KTx)

14	Recommendation	2025	Grade of recommendation
f) We suggest that protocol biopsies may be considered in children and adolescents after the first KTx in selected cases, such as patients at high immunological risk or with underlying disease prone to recurrence, if local resources and expertise allow for safe implementation. g) We do not suggest routine protocol biopsies in all pediatric KTx recipients due to very low certainty of evidence and unclear balance of benefits and harms.			2 ⇑ 2 ⇓
Kanzelmeyer NK, Ahlenstiel T, Drube J, et al. Protocol biopsy-driven interventions after pediatric renal transplantation. Pediatr Transplant. 2010;14(8):1012–1018. 74 Birk PE, Blydt-Hansen TD, Dart AB, et al. Low incidence of adverse events in outpatient pediatric renal allograft biopsies. Pediatr Transplant. 2007;11(2):196–200. 91 Vidhun J, Masciandro J, Varich L, et al. Safety and risk stratification of percutaneous biopsies of adult-sized renal allografts in infant and older pediatric recipients. Transplantation. 2003;76(3):552–557. 92			Mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of protocol biopsies on patient mortality. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of protocol biopsies on graft loss. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of protocol biopsies on graft function. Incidence of serious adverse events: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of protocol biopsies on incidence of serious adverse events. Quality of life, incidence of BPAR, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence search.
Strength of consensus 100%

Scheduled (protocol) biopsies were compared with no biopsies or indication-driven biopsies, aiming to improve outcomes in children after their first KTx. The rationale for protocol biopsies is to detect subclinical pathologies such as acute rejection, recurrence of the primary disease, or CNI nephrotoxicity, which would otherwise remain undetected until clinical deterioration (eg, an increase in serum creatinine) occurs. In adult transplant recipients, protocol biopsies have long been an integral part of routine surveillance protocols in some centers in the CsA era but to a lesser extent in the tacrolimus and MMF era. However, in pediatric KTx, the implementation is variable, and the supporting evidence is sparse.

A single-center pediatric study demonstrated that protocol-based biopsy-guided adjustments in immunosuppressive therapy may stabilize graft function.⁷⁴ Two pediatric studies reported complication rates of 9% and 16%, respectively, with gross hematuria being the most common adverse event.^91,92 No graft loss or graft nephrectomy due to biopsy-related complications was reported. Overall safety appears acceptable in centers with appropriate infrastructure.

Previous adult data from the pre–tacrolimus era support the diagnostic yield and clinical benefit of protocol biopsies, particularly in detecting subclinical rejection and early recurrence of glomerulonephritis; however, these benefits were no longer observed after the introduction of tacrolimus into the immunosuppressive regimen. In addition, these findings cannot be directly extrapolated to pediatric patients. The evidence base in children is of very low certainty, and most desirable and undesirable effects remain uncertain.

Given the very low certainty of the available evidence and the potential harms associated with biopsy complications, the group did not recommend routine implementation of protocol biopsies for all pediatric KTx recipients.

In selected cases, such as patients with C3 glomerulopathy, complement-mediated hemolytic uremic syndrome, or high HLA sensitization protocol, biopsies may support the optimization of immunosuppressive therapy because pathologies can be diagnosed early in biopsies before kidney function has begun deteriorating.

There are no clear KDIGO recommendations for protocol biopsies in adults.

KQ5: dn–HLA-DSAs (Literature Pediatric and Adult KTx)

15		Recommendation	2025							Grade of recommendation
We suggest using C1q complement-binding status combined with anti-dn-HLA-DSA screening and allograft biopsy to stratify individual therapeutic interventions of AMR in children after KTx.										2 ⇑
Sigurjonsdottir VK, Purington N, Chaudhuri A, et al. Complement-binding donor-specific anti-HLA antibodies: biomarker for immunologic risk stratification in pediatric kidney transplantation recipients. Transpl Int. 2022;35:10158. 93										Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of using C1q-dnDSAs for risk stratification of graft loss in children with KTx and the presence of dnDSAs. Mortality, graft function, incidence of BPARs, quality of life, incidence of serious adverse events, incidence of malignancy, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 90%

16		Recommendation			2025			Grade of recommendation/LoE
In KTx recipients with de novo HLA donor-specific antibodies but without decline of renal function or AMR, we do not suggest using preemptive treatment with high-dose IVIG alone or in combination with rituximab.								2 ⇓
Kim HW, Lee J, Heo SJ, et al. Comparison of high-dose IVIG and rituximab versus rituximab as a preemptive therapy for de novo donor-specific antibodies in kidney transplant patients. Sci Rep. 2023;13(1):7682. 94 Knight RJ, Loucks-Devos J, Khan NA, et al. Long-term follow-up of renal transplant recipients treated with IVIG for de novo donor-specific antibodies. Transplant Proc. 2021;53(6):1865–1871. 95 Matignon M, Pilon C, Commereuc M, et al. Intravenous immunoglobulin therapy in kidney transplant recipients with de novo DSA: results of an observational study. PLoS ONE. 2017;12(6):e0178572. 96								IVIG+rituximab vs rituximab alone: Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG plus rituximab on graft function compared with rituximab alone. Incidence of serious adverse events: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG plus rituximab on incidence of serious adverse events compared with rituximab alone. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG plus rituximab on incidence of BPAR compared with rituximab alone. Mortality, graft loss, quality of life, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence IVIG vs no IVIG Patient mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG on patient mortality compared with no IVIG. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG on graft loss compared with no IVIG. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG on graft function compared with no IVIG. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of IVIG on incidence of BPAR compared with no IVIG. Quality of life, incidence of serious adverse events, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 93%

17		Recommendation				2025				Grade of recommendation/LoE
We suggest considering a combination treatment of IVIG and rituximab with or without plasmapheresis for pediatric KTx recipients with dn-HLA DSAs and AMR to prolong graft survival.										2 ⇑
Cioni M, Nocera A, Tagliamacco A, et al. Failure to remove de novo donor-specific HLA antibodies is influenced by antibody properties and identifies kidney recipients with late antibody-mediated rejection destined to graft loss—a retrospective study. Transpl Int. 2019;32(1):38–48. 97 Billing H, Rieger S, Susal C, et al. IVIG and rituximab for treatment of chronic antibody-mediated rejection: a prospective study in paediatric renal transplantation with a 2-year follow-up. Transpl Int. 2012;25(11):1165–1173. 98										Plasmapheresis, low-dose IVIG, and rituximab vs high-dose IVIG and rituximab alone: Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of plasmapheresis+low-dose IVIG+rituximab on graft loss compared with high-dose IVIG+rituximab. Mortality, graft function, incidence of biopsy-proven acute rejections, quality of life, incidence of serious adverse events, incidence of malignancy, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence. Unmodified immunosuppressive regimen in combination with antihumoral therapy vs CNI conversion with or without high-dose steroids in combination with antihumoral therapy: Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of an unmodified immunosuppressive regimen in combination with antihumoral therapy on graft function compared with CNI conversion with or without high-dose steroids in combination with antihumoral therapy. Mortality, graft loss, incidence of BPARs, quality of life, incidence of serious adverse events, incidence of malignancy, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 80%

18	Recommendation				2025					Grade of recommendation/LoE
No recommendation can be given concerning the use of daratumumab (anti-CD38 monoclonal antibody [mAb]) for the treatment of difficult cases of or resistant chronic AMR, as there was no consensus.										0
Zhu L, Guo Z, Zhao D, et al. Case report: daratumumab for treatment of refractory late or chronic active antibody-mediated rejection in renal allograft recipients with high levels of de novo donor-specific antibodies. Front Immunol. 2023;13. 99										Patient mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of plasmapheresis, IVIG, and daratumumab on mortality in children with KTx and presence of dnDSAs. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of plasmapheresis, IVIG, and daratumumab on graft loss in children with KTx and the presence of dnDSAs. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of plasmapheresis, IVIG, and daratumumab on graft function in children with KTx and the presence of dnDSAs. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of plasmapheresis, IVIG, and daratumumab on incidence of BPAR in children with KTx and the presence of dnDSAs. Quality of life, incidence of serious adverse events, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 37%

19		Recommendation					2025		Grade of recommendation/LoE
h) We suggest that the use of carfilzomib (a proteasome inhibitor) for treating AMR or de novo HLAdn-HLA-DSAs in pediatric KTx recipients is not recommended because of the associated nephrotoxicity and its use requires careful consideration. i) We do not suggest using eculizumab therapy for KTx recipients with dn-HLA-DSAs and AMR.									2 ⇓ 2 ⇓
Cody EM, Varnell C Jr, Lazear D, et al. Carfilzomib-based antibody mediated rejection therapy in pediatric kidney transplant recipients. Pediatr Transplant. 2023;27(7):e14534. 100 Kulkarni S, Kirkiles-Smith NC, Deng YH, et al. Eculizumab therapy for chronic antibody-mediated injury in kidney transplant recipients: a pilot randomized controlled trial. Am J Transplant. 2017;17(3):682–691. 101									Carfilzomib: Patient mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about a carfilzomib-based AMR therapy. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about a carfilzomib-based AMR therapy. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about a carfilzomib-based AMR therapy. Quality of life, incidence of BPAR, incidence of serious adverse events, incidence of malignancies, incidence of hospitalization, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence Eculizumab: Patient mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of eculizumab on patient mortality. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of eculizumab on graft loss. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of eculizumab on graft function. Incidence of hospitalization: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of eculizumab on incidence of hospitalization. Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of eculizumab on incidence of BPAR. Quality of life, incidence of serious adverse events, incidence of malignancies, incidence of infection (including opportunistic infections), and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 76%

20	Recommendation			2025						Grade of recommendation/LoE
Optimization of immunosuppression (encouragement of medication adherence, tailored optimization of immunosuppressive therapy) in patients with persistent dn-HLA-DSAs can be considered.										0
Stringer D, Gardner L, Shaw O, et al. Optimized immunosuppression to prevent graft failure in renal transplant recipients with HLA antibodies (OuTSMART): a randomised controlled trial. EClinicalMed. 2023;56:101819. 102 Stringer D, Gardner L, Shaw O, et al. Preventing kidney transplant failure by screening for antibodies against human leucocyte antigens followed by optimised immunosuppression: OuTSMART RCT. Southampton, UK: National Institute for Health and Care Research. 2023. 103										Patient mortality: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on patient mortality. Graft loss: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on graft loss. Graft function: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on graft function. Incidence of malignancies: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on incidence of malignancies. Incidence of infection (including opportunistic infections): very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on incidence of infection (including opportunistic infections). Incidence of BPAR: very low ⊕⊝⊝⊝ The evidence is very uncertain about the effect of biomarker-led care on incidence of BPAR. Quality of life, incidence of serious adverse events, incidence of hospitalization, and incidence of opportunistic infections: no evidence was identified in the systematic evidence.
Strength of consensus 90%

21		Recommendation	2025							Grade of recommendation/LoE
We suggest regular and periodic monitoring for dn-HLA-DSAs in pediatric KTx recipients.										Expert consensus
Strength of consensus 87%

22		Guideline	2025							Grade
We suggest performing renal biopsy when dn-HLA-DSAs are persistently found in pediatric KTx recipients without renal dysfunction.										Expert consensus
Strength of consensus 87%ss

23		Recommendation	2025							Grade of recommendation/LoE
We suggest establishing uniformity in the mean fluorescence intensity (MFI) assays used and in cutoff values for the definition of positive dn-HLA-DSAs in KTx recipients.										Expert consensus
Strength of consensus 87%

The development of dn-HLA-DSA after pediatric KTx is associated with an increased risk of AMR and graft loss.¹⁹ In the multicenter observational study from the CERTAIN research network, the cumulative incidence of dn-HLA-DSA posttransplant was 4.5% in year 1, 8.3% in year 3, and 13% in year 5; the corresponding data for de novo class II HLA-DSA were 10%, 22.5%, and 30.6%, respectively.¹⁹ In adult patients, it is estimated that 2% to 35% of patients develop dn-HLA-DSA posttransplant.¹⁰⁴ The cumulative incidence of acute AMR and chronic active AMR in children at 5 y posttransplant is around 10% and 6%, respectively.¹⁹ It is important to prevent acute AMR or treat it early, before irreversible damage occurs. However, some patients with dn-HLA-DSA escape rejection, graft dysfunction, or graft failure. It is challenging to identify clinically relevant dn-HLA-DSA. Due to differences in HLA-DSA positivity cutoff values across transplantation centers, it is difficult to compare studies.

Factors determining the pathogenicity of HLA-DSAs are thought to be related to the MFI, in vitro complement-binding capacity (C1q, C3d), HLA class locus, and HLA-IgG subclasses.¹⁰⁵ Less invasive tests, such as dd cell-free deoxyribonucleic acid (dd-cfDNA) level tests and gene expression profiling kits, may have a future role in the clinical management of patients with dn-HLA-DSA; however, these tests are costly, and their exact utility needs to be established.¹⁰⁶

C1q is a component of the classical complement pathway and binds to the FC region of IgG and IgM antibodies. The C1q-binding ability of HLA-DSAs is the bridge to complement activation. The prevalence of C1q-binding HLA-DSAs in DSA-positive patients ranges from 14% to 64%.¹⁰⁷ Patients with C1q-binding HLA-DSAs are considered more likely to develop AMR and graft failure than those with non-C1q-binding HLA-DSAs, in both pediatric and adult KTx recipients.

The group reviewed the available evidence on the use of C1q-binding HLA-DSA stratification for individualized treatment of AMR and acknowledged that the certainty of the evidence of effect was very low. In the selected study in adult KTx recipients,⁹³ the presence of C1q-binding DSAs was determined using the C1qScreen assay (One Lambda Inc). Patients with C1q-binding DSAs were not divided into 2 (treatment) groups, but all received gradual intensification of immunosuppressive treatment (increased target trough levels of CNI, MMF, or IVIG, rituximab, plasmapheresis, bortezomib). Therefore, the evidence of the benefit of C1q-binding HLA-DSA stratification is difficult to establish. However, 14 of 21 patients with persisting C1q-HLA-DSAs after treatment lost their graft, compared with 0 of 31 patients with C1q-HLA-DSAs that disappeared after treatment.⁹³

The treatment of dn-HLA-DSA with early AMR (<6 mo posttransplant) consists mainly of IVIG, plasma exchange, or rituximab. The treatment of patients with subclinical AMR and dn-HLA-DSA with plasma exchange, IVIG, and rituximab results in better graft survival than in untreated patients with subclinical AMR or in patients with clinical AMR treated with plasma exchange, IVIG, and rituximab.¹⁰⁸ These findings highlight the importance of early diagnosis and treatment.

The following treatment recommendation is based on the effectiveness of preemptive treatment with IVIG alone or in combination with plasma exchange and rituximab in patients with dn-HLA-DSA but without graft dysfunction or rejection. IVIG is generally well tolerated but can cause infusion reactions or more serious side effects, such as allergic (anaphylactic) reactions, transfusion-related acute lung injury (very rare), thrombotic events, and hemolytic anemia. The group reviewed the available evidence on IVIG and/or rituximab treatment in patients who developed dn-HLA-DSA but had no signs of renal function decline or AMR and acknowledged that the certainty of the evidence of effect was very low. The group noted that the studies had conflicting results. One study⁹⁴ showed a significant decline of dn-HLA-DSA MFI values at 12 mo after treatment with rituximab and IVIG, as well as with rituximab alone. AMR was not observed in either group. Therefore, no conclusion can be drawn on whether treatment with rituximab ± IVIG prevents AMR. Another study showed that high-dose IVIG did not prevent acute AMR,⁹⁶ whereas the study by Knight et al⁹⁵ showed a borderline significant difference in acute rejection between the IVIG⁺ (8%) and IVIG^– (42%) cohorts (P = 0.06). The group concluded that more randomized studies with larger sample sizes are needed to assess whether preemptive treatment with rituximab and/or IVIG in patients with dn-HLA-DSA without graft dysfunction or AMR reduces the risk of AMR.

Plasma exchange is relatively invasive because placement of a central venous line is required. The rationale for using plasma exchange and IVIG is to combine the removal of dn-HLA-DSA by plasma exchange with the immunomodulatory effects of IVIG. Rituximab has the potential to increase the risk of infection by blocking CD20⁺ B cells. The removal of dn-HLA-DSA with plasma exchange was a significant factor in reducing graft loss (hazard ratio, 0.11; 95% CI, 0.02-0.46; P = 0.003).⁹⁷ These results, together with the 2019 expert consensus from The Transplantation Society,¹⁰⁹ confirm that plasma exchange and IVIG can be considered standard of care for the treatment of early acute AMR, despite the limited evidence. Because plasma exchange, IVIG, and rituximab are used simultaneously in patients with dn-HLA-DSA and AMR, the evidence for each treatment alone is difficult to define.

CD38 is a transmembrane glycoprotein that is expressed by immune and hematopoietic cells, particularly plasma cells and natural killer cells. Anti-CD38 mAb have been approved for the treatment of patients with multiple myeloma.¹¹⁰ Daratumumab (an anti-CD38 targeted drug) induces complement-dependent cytotoxicity and apoptotic signaling in CD38-expressing cells.¹¹¹ Anti-CD38 mAb therapy is now studied in patients with lupus nephritis, IgA nephropathy, and membranous nephropathy.¹¹² Anti-CD38 mAb treatment might be a promising therapy for AMR by inhibiting natural killer cells and depleting plasma cells that produce dn-HLA-DSA, thereby reducing microvascular inflammation.¹¹³ Few case reports have been published on the use of daratumumab for the treatment of AMR or for desensitization before transplantation. Experimental studies in nonhuman primate models have shown that daratumumab reduces HLA-DSAs when combined with the CXCR4 antagonist plerixafor.¹¹⁴

One study demonstrated that in patients with dn-HLA-DSA and AMR in whom standard therapy failed, the addition of the CD38 antibody daratumumab therapy decreased HLA-DSA levels substantially and stabilized kidney function. A few more case reports have been published on the positive effect of daratumumab on the treatment and prevention treatment of AMR in solid organ transplant recipients.^115,116 Also, the CD38⁻targeting antibody felzartamab has been evaluated in a phase 2 study and is now being tested in a phase 3 trial.^117,118 The group recommended that treatment with daratumumab be reserved for patients with dn-HLA-DSA and AMR who are refractory to standard of care. It must be noted that the safety of daratumumab has not been established in pediatric KTx recipients; therefore, some members of the guideline group voted against a recommendation based on adult data, and no consensus was reached.

The proteasome is an intracellular protein complex that degrades misfolded proteins. Proteasome inhibitors lead to apoptotic cell death. Plasma cells are especially vulnerable to proteasome inhibition. The proteasome inhibitor bortezomib has been used with rituximab, plasma exchange, and IVIG for the treatment of AMR, but it can cause serious side effects, including peripheral neuropathy, gastrointestinal symptoms, anemia, leukopenia, and thrombocytopenia.¹¹⁹ Carfilzomib is a second-generation irreversible proteasome inhibitor with less neurotoxicity and is approved for the treatment of multiple myeloma. Carfilzomib is now being evaluated for desensitization in KTx candidates.¹²⁰ It may provide HLA-DSA elimination or reduction, but it is also associated with nephrotoxicity and thrombotic microangiopathy.¹²¹ These 2 case reports both showed increased serum creatinine levels that returned to baseline values in 1 of the 2 patients. The mechanism by which carfilzomib causes renal toxicity is not well understood, and this has also been demonstrated in patients with multiple myeloma.¹²² Due to the nephrotoxic side effects, the utility of carfilzomib in children with dn-HLA-DSA with or without AMR is limited.

Eculizumab is a terminal complement inhibitor (anti-C5-Ab) and is mainly used to treat complement-mediated diseases, such as complement-mediated hemolytic uremic syndrome. Eculizumab has been shown to reduce the incidence of early AMR in patients with a positive crossmatch at the time of transplantation.^123,124 But it did not affect the long-term outcome, suggesting that complement blockade alone might not be an effective treatment for AMR. Most studies on eculizumab in AMR are case reports. Patients treated with eculizumab showed stabilization of renal function compared with the gradual decline in the control group, but acute humoral injury was not reduced. Therefore, eculizumab seems ineffective for the treatment of dn-HLA-DSA and AMR. The group recommended not administering carfilzomib or eculizumab in patients with dn-HLA-DSA and AMR.

The results of the OuTSMART study show that monitoring the development of dn-HLA-DSA has prognostic value. However, no evidence has been found to support the hypothesis that optimization of maintenance immunosuppressive therapy can prevent graft failure. An evidence-based expert article recommended performing HLA-DSA monitoring and surveillance based on the patient’s immunological risk for HLA-DSAs and rejection.¹⁰⁴ The panel recommends at least 1 HLA-DSA determination between 3 and 12 mo posttransplant in all patients. Periodic monitoring is advised in patients whose immunosuppressive treatment is changed or who exhibit significant variability in CNI blood levels, poor adherence, or graft dysfunction. The group recommends optimizing immunosuppressive maintenance therapy (including avoiding reductions in immunosuppression) in patients who develop dn-HLA-DSAs and are at risk for AMR. Mostly, graft dysfunction is the indicator for performing a KTx biopsy. In addition, routine screening for the development of dn-HLA-DSA is not a common practice among pediatric transplantation centers. According to the Banff classification, the most widely used definition of AMR combines histological findings of AMR with the presence of HLA-DSAs. dn-HLA-DSA can emerge without clinical signs of graft dysfunction.¹²⁵ Subclinical rejection is diagnosed by surveillance of renal biopsies in patients with stable kidney function. It has been demonstrated that subclinical rejection in the first year posttransplant occurs in 20%–40% of children.¹²⁶ Yamamoto et al have demonstrated that 40% of (mostly adult) patients (18/43) developed biopsy-proven subclinical rejection when dn-HLA-DSA were detected^.125 Without a surveillance biopsy protocol, these rejections were not diagnosed until later, when graft dysfunction occurred. Subclinical rejection is associated with long-term reduced kidney function, increased IF/TA, and the appearance of dn-HLA-DSA.^127-129 There are potential practical difficulties with surveillance biopsies, and less invasive methods to identify subclinical rejection using biomarkers are under development. Currently, serial monitoring for dn-HLA-DSA is the most widely used biomarker.¹³⁰ It is assumed that early treatment of rejection results in better graft function and lower graft failure than late treatment. The group recommended performing a renal biopsy in patients who develop persistent and/or dn-HLA-DSA with an MFI value of ≥3000. This can help stratify patients for additional immunosuppressive therapy.

There are different assays for HLA-antibody testing. The Luminex assay is the most commonly used test for detecting HLA antibodies. These assays detect IgG but not IgM, and HLA antibodies against HLA class I or HLA class II. Results of this assay are described as MFI units. The test is semiquantitative. The Luminex single-antigen bead (LSAB) assay has increased the sensitivity and specificity of detecting HLA antibodies. Manufacturers of LSAB assays are: Lifecodes, Immucor (United States), and One Lambda (Canoga Park, CA). The LSAB screening can be guided by the MFI of dn-HLA-DSA. The higher the MFI, the higher the risk of AMR.¹³¹ The results of the Lifecodes Immucor assay cannot be interpreted using the usual One Lambda Labscreen criteria. One Lambda Labscreen detects low-intensity HLA-DSAs better but has lower specificity.^132,133 Literature on HLA antibodies, HLA-DSA detection, and intensity cutoffs are mostly based on the One Lambda Labscreen test. A very good summary of the clinical utility of posttransplant monitoring of dn-HLA-DSA can be found in a recent publication by the European Society of Transplantation.¹¹³

CONCLUSIONS

During recent years, immunosuppressive therapy has evolved, and substantial knowledge has been gained regarding the most commonly used immunosuppressive agents. In this guideline, we have compiled evidence-based and consensus-based recommendations intended to support pediatric nephrologists and transplant physicians in selecting and managing optimal immunosuppressive regimens for children after KTx. Interestingly, recommendations on immunosuppression are largely consistent with those proposed by KDIGO for adults.²¹ The main exception relates to steroid avoidance and early steroid withdrawal, which differ in immunologically low-risk pediatric populations, where longitudinal growth and long-term steroid-related adverse effects are of particular importance. Many of the recommendations—especially those concerning basic immunosuppressive regimens—can be implemented in low-resource settings; however, not all the medications discussed are readily available in developing countries. It should be noted that recent advances in drug development, such as once-daily tacrolimus formulations (Advagraf and Envarsus) and costimulation blockers such as belatacept, may become important additional options for improving patient adherence, particularly when they are licensed for use in children.¹¹³

However, because of the limited number of pediatric KTx recipients, the overall evidence base remains limited. Further well-designed controlled trials are needed in the areas of immunosuppression monitoring and adjustment, biomarker-guided therapy, protocol biopsies, and AMR to enable more robust, evidence-based recommendations in the future.

ACKNOWLEDGMENTS

The authors thank the team of Cochrane Austria (Isabel Moser, Isolde Sommer, Andreea Dobrescu, Brigitte Piso) for the support in literature research and grading.

Supplementary Material

tpa-110-e1889-s001.pdf ^{(7.7MB, pdf)}