Incidence, risk factors and prognosis of acute kidney injury in patients treated with immune checkpoint inhibitors: a retrospective study

Machine learning-based risk prediction of acute kidney disease and hospital mortality in older patients

Introduction

Acute kidney injury (AKI) is a prevalent complication in older people, elevating the risks of acute kidney disease (AKD) and mortality. AKD reflects the adverse events developing after AKI. We aimed to develop and validate machine learning models for predicting the occurrence of AKD, AKI and mortality in older patients.

Methods

We retrospectively reviewed the medical records of older patients (aged 65 years and above). To explore the trajectory of kidney dysfunction, patients were categorized into four groups: no kidney disease, AKI recovery, AKD without AKI, or AKD with AKI. We developed eight machine learning models to predict AKD, AKI, and mortality. The best-performing model was identified based on the area under the receiver operating characteristic curve (AUC) and interpreted using the Shapley additive explanations (SHAP) method.

Results

A total of 22,005 patients were finally included in our study. Among them, 4,434 patients (20.15%) developed AKD, 4,000 (18.18%) occurred AKI, and 866 (3.94%) patients deceased. Light gradient boosting machine (LGBM) outperformed in predicting AKD, AKI, and mortality, and the final lite models with 15 features had AUC values of 0.760, 0.767, and 0.927, respectively. The SHAP method revealed that AKI stage, albumin, lactate dehydrogenase, aspirin and coronary heart disease were the top 5 predictors of AKD. An online prediction website for AKD and mortality was developed based on the final models.

Discussion

The LGBM models provide a valuable tool for early prediction of AKD, AKI, and mortality in older patients, facilitating timely interventions. This study highlights the potential of machine learning in improving older adult care, with the developed online tool offering practical utility for healthcare professionals. Further research should aim at external validation and integration of these models into clinical practice.

Interpretable machine learning-based individual analysis of acute kidney injury in immune checkpoint inhibitor therapy

BACKGROUND

Acute kidney injury (AKI) is a critical complication of immune checkpoint inhibitor therapy. Since the etiology of AKI in patients undergoing cancer therapy varies, clarifying underlying causes in individual cases is critical for optimal cancer treatment. Although it is essential to individually analyze immune checkpoint inhibitor-treated patients for underlying pathologies for each AKI episode, these analyses have not been realized. Herein, we aimed to individually clarify the underlying causes of AKI in immune checkpoint inhibitor-treated patients using a new clustering approach with Shapley Additive exPlanations (SHAP).

METHODS

We developed a gradient-boosting decision tree-based machine learning model continuously predicting AKI within 7 days, using the medical records of 616 immune checkpoint inhibitor-treated patients. The temporal changes in individual predictive reasoning in AKI prediction models represented the key features contributing to each AKI prediction and clustered AKI patients based on the features with high predictive contribution quantified in time series by SHAP. We searched for common clinical backgrounds of AKI patients in each cluster, compared with annotation by three nephrologists.

RESULTS

One hundred and twelve patients (18.2%) had at least one AKI episode. They were clustered per the key feature, and their SHAP value patterns, and the nephrologists assessed the clusters' clinical relevance. Receiver operating characteristic analysis revealed that the area under the curve was 0.880. Patients with AKI were categorized into four clusters with significant prognostic differences (p = 0.010). The leading causes of AKI for each cluster, such as hypovolemia, drug-related, and cancer cachexia, were all clinically interpretable, which conventional approaches cannot obtain.

CONCLUSION

Our results suggest that the clustering method of individual predictive reasoning in machine learning models can be applied to infer clinically critical factors for developing each episode of AKI among patients with multiple AKI risk factors, such as immune checkpoint inhibitor-treated patients.

Association of Proton Pump Inhibitor Use and Immune Checkpoint Inhibitor-Mediated Acute Kidney Injury: A Meta-Analysis and a Review of Related Outcomes

INTRODUCTION

Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment. However, they pose the risk of immune-related adverse events, including ICI-mediated acute kidney injury (ICI-AKI). Recent studies have implicated proton pump inhibitors (PPIs) as potential contributors to ICI-AKI development. This meta-analysis examines the association between PPI use and ICI-AKI, exploring a potential modifiable risk factor in ICI therapy while also reviewing the possible outcomes of ICI-AKI.

METHODS

We conducted a comprehensive systematic review and meta-analysis of observational studies, assessing the risk of ICI-AKI in cancer patients concurrently using PPIs and potential outcomes. Odds ratios (ORs) were pooled using random-effects models. Subgroup analyses and sensitivity analyses were performed to evaluate heterogeneity and potential biases.

RESULTS

A total of 14 studies involving 12,694 patients were included. In total, we analyzed 639 patients with all-cause AKI and 779 patients with ICI-AKI. The pooled OR for the overall incidence of AKI from all-causes was 1.57 (95% confidence interval [CI] 1.02-2.40) among patients on PPIs. Specifically, the risk of ICI-AKI associated with PPI use was significantly higher, with a pooled OR of 1.84 (95% CI 1.16-2.90). This indicates approximately 84% higher likelihood of developing ICI-AKI with concurrent use of PPIs. Additionally, among patients with ICI-AKI, 67% had complete or partial recovery of renal function, 32% progressed to chronic kidney disease (CKD), and about 36% died during a follow-up period of at least 3 months.

CONCLUSION

This meta-analysis highlights the importance of cautious PPI prescription in cancer patients undergoing ICI therapy. Clinicians are advised to evaluate the risks and benefits of PPI use and consider alternative therapies when feasible.

Immune checkpoint inhibitors and acute kidney injury

As a new type of anti-tumor immunotherapy, immune checkpoint inhibitors (ICIs) have improved the prognosis of multiple malignancies. However, renal complications are becoming more frequent. Nephrotoxicity often manifests as acute kidney injury (AKI), and the most common histopathological type is acute tubulointerstitial nephritis (ATIN). Based on previous studies of the incidence and potential risk factors for nephrotoxicity, in this review, we describe the mechanism of AKI after ICIs treatment, summarize the incidence, risk factors, and outcomes of AKI, and discuss the diagnosis and management of immune checkpoint inhibitors-associated acute kidney injury (ICI-AKI). In addition, we review the current status of ICIs rechallenge and the therapeutic strategies of ICIs applied in kidney transplant recipients. Finally, we emphasize the importance of collaboration between nephrologists and oncologists to guide the treatment of ICIs and the management of renal complications.

All-cause and immune checkpoint inhibitor-associated acute kidney injury in immune checkpoint inhibitor users: a meta-analysis of occurrence rate, risk factors and mortality

Background

Immune checkpoint inhibitors (ICIs) have been associated with acute kidney injury (AKI). However, the occurrence rate of ICI-related AKI has not been systematically examined. Additionally, exposure to proton pump inhibitors (PPIs) and non-steroidal anti-inflammatory drugs (NSAIDs) were considered as risk factors for AKI, but with inconclusive results in ICI-related AKI. Our aim was to analyse the occurrence rate of all-cause AKI and ICI-related AKI and the occurrence rates of severe AKI and dialysis-requiring AKI, and to determine whether exposure to PPIs and NSAIDs poses a risk for all-cause and ICI-related AKI.

Methods

This study population was adult ICI recipients. A systematic review was conducted by searching MEDLINE, Embase and PubMed through October 2023. We included prospective trials and observational studies that reported any of the following outcomes: the occurrence rate of all-cause or ICI-related AKI, the relationship between PPI or NSAID exposure and AKI development or the mortality rate in the AKI or non-AKI group. Proportional meta-analysis and pairwise meta-analysis were performed. The evidence certainty was assessed using the Grading of Recommendations Assessment, Development and Evaluation framework.

Results

A total of 120 studies comprising 46 417 patients were included. The occurrence rates of all-cause AKI were 7.4% (14.6% from retrospective studies and 1.2% from prospective clinical trials). The occurrence rate of ICI-related AKI was 3.2%. The use of PPIs was associated with an odds ratio (OR) of 1.77 [95% confidence interval (CI) 1.43-2.18] for all-cause AKI and an OR of 2.42 (95% CI 1.96-2.97) for ICI-related AKI. The use of NSAIDs was associated with an OR of 1.77 (95% CI 1.10-2.83) for all-cause AKI and an OR of 2.57 (95% CI 1.68-3.93) for ICI-related AKI.

Conclusions

Our analysis revealed that approximately 1 in 13 adult ICI recipients may experience all-cause AKI, while 1 in 33 adult ICI recipients may experience ICI-related AKI. Exposure to PPIs and NSAIDs was associated with an increased OR risk for AKI in the current meta-analysis.

Nephrotoxicity in the Age of Immune Checkpoint Inhibitors: Mechanisms, Diagnosis, and Management

Immune checkpoint inhibitors (ICI) revolutionized cancer therapy by augmenting anti-tumor immunity via cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and programmed death-1/programmed death-ligand 1 (PD-1/PD-L1). However, this breakthrough is accompanied by immune-related adverse effects (irAEs), including renal complications. ICI-related nephritis involves complex mechanisms like auto-reactive T cells, auto-antibodies, reactivation of drug-specific T cells, and cytokine-driven inflammation culminating in AKI. ICI-AKI typically manifests weeks to months into treatment, often with other irAEs. Timely detection relies on monitoring creatinine levels and urine characteristics. Biomarkers, like soluble interleukin-2 receptor (sIL-2R) and urine cytokine levels, provide non-invasive insights, while renal biopsy remains the gold standard for confirmation. Management of ICI-AKI requires a balance between discontinuing ICI therapy and prompt immunosuppressive intervention, typically with corticosteroids. Some cases permit ICI therapy resumption, but varying renal recovery rates highlight the importance of vigilant monitoring and effective therapy. Beyond its clinical implications, the potential of irAEs to predict positive treatment responses in certain cancers raises intriguing questions. Data on nephritis-treatment response links are limited, and ongoing research explores this complex interaction. In summary, ICI therapy's transformative impact on cancer treatment is counterbalanced by irAEs, including nephritis. Early recognition and management are vital, with ongoing research refining diagnostic and treatment strategies.

Immune Checkpoint Inhibitor-Associated Nephrotoxicity

CONTEXT

The clinical indications for immune checkpoint inhibitors (ICIs) are rapidly expanding. However, adverse events affecting multiple organs, including kidneys leading to ICI-associated acute kidney injury (AKI), remain a significant challenge with ICI therapy. Although AKI is considered a rare complication, it can be severe and result in treatment interruption or discontinuation of ICIs. Despite a generally favorable kidney prognosis, the possibility of re-challenging ICI therapy remains a subject of debate, particularly for patients who have exhausted other treatment options or experienced severe AKI. Subject of Review: In a recent review article, Sprangers et al. provide a comprehensive overview of the possible mechanisms and clinical manifestations of ICI-associated AKI [Nat Rev Nephrol. 2022;18(12):794-805]. The authors propose a practical strategy for diagnosing and managing suspected cases of ICI-associated AKI, which includes identifying a subset of eligible patients who may be re-exposed to ICIs following an episode of AKI. Second Opinion: The authors of the review article offer several recommendations on the diagnosis and treatment of ICI-associated nephrotoxicity. While we generally agree with the recommendations proposed by the authors, it is important to acknowledge that the available data primarily rely on small retrospective studies, as the authors have recognized. In addition, there are two key questions that need be carefully addressed in future studies: (1) the optimal dose and duration of corticosteroids and the use of alternative immunosuppressive agents in patients with ICI-associated nephrotoxicity and (2) a clear guideline for restarting ICI treatment in patients with AKI who have not fully recovered their kidney function.

Incidence and risk factors of acute kidney injury in cancer patients treated with immune checkpoint inhibitors: a systematic review and meta-analysis

Background

The incidence and risk factors of acute kidney injury (AKI) in patients with malignancies receiving immune checkpoint inhibitors (ICIs) are being extensively reported with their widespread application.

Objective

This study aimed to quantify the incidence and identify risk factors of AKI in cancer patients treated with ICIs.

Methods

We searched the electronic databases of PubMed/Medline, Web of Science, Cochrane and Embase before 1 February 2023 on the incidence and risk factors of AKI in patients receiving ICIs and registered the protocol in PROSPERO (CRD42023391939). A random-effect meta-analysis was performed to quantify the pooled incidence estimate of AKI, identify risk factors with pooled odds ratios (ORs) and 95% confidence intervals (95% CIs) and investigate the median latency period of ICI-AKI in patients treated with ICIs. Assessment of study quality, meta-regression, and sensitivity and publication bias analyses were conducted.

Results

In total, 27 studies consisting of 24048 participants were included in this systematic review and meta-analysis. The overall pooled incidence of AKI secondary to ICIs was 5.7% (95% CI: 3.7%-8.2%). Significant risk factors were older age (OR: 1.01, 95% CI: 1.00-1.03), preexisting chronic kidney disease (CKD) (OR: 2.90, 95% CI: 1.65-5.11), ipilimumab (OR: 2.66, 95% CI: 1.42-4.98), combination of ICIs (OR: 2.45, 95% CI: 1.40-4.31), extrarenal immune-related adverse events (irAEs) (OR: 2.34, 95% CI: 1.53-3.59), and proton pump inhibitor (PPI) (OR: 2.23, 95% CI: 1.88-2.64), nonsteroidal anti-inflammatory drug (NSAID) (OR: 2.61, 95% CI: 1.90-3.57), fluindione (OR: 6.48, 95% CI: 2.72-15.46), diuretic (OR: 1.78, 95% CI: 1.32-2.40) and angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin-receptor blockers (ARBs) (pooled OR: 1.76, 95% CI: 1.15-2.68) use. Median time from ICIs initiation to AKI was 108.07 days. Sensitivity and publication bias analyses indicated robust results for this study.

Conclusion

The occurrence of AKI following ICIs was not uncommon, with an incidence of 5.7% and a median time interval of 108.07 days after ICIs initiation. Older age, preexisting chronic kidney disease (CKD), ipilimumab, combined use of ICIs, extrarenal irAEs, and PPI, NSAID, fluindione, diuretics and ACEI/ARB use are risk factors for AKI in patients receiving ICIs.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier CRD42023391939.