Introduction
Methotrexate (MTX)-associated acute kidney injury (AKI) occurs in 3% to 60% of patients at some point during treatment.1,2 The standard of care for MTX dose derivation and postdose monitoring centers around serum creatinine level as a surrogate for glomerular filtration rate (GFR).3 As the terminal byproduct of skeletal muscle metabolism, nonrenal determinants, including altered muscle mass, deconditioning, and malnutrition, can decrease the accuracy of serum creatinine-based GFR estimation in patients with cancer.4 In addition, creatinine level may not rise until 48 hours after the onset of kidney damage, leaving a missed window of opportunity for early intervention.5
Novel biomarkers that indicate tubular epithelial cell stress or early damage demonstrate the capacity to identify kidney injury before overt AKI is detected by a rise in serum creatinine level or a decline in urine output.6,7 Consensus recommendations support the incorporation of damage and stress biomarkers into the detection and management of AKI.8 Whether these novel biomarkers can predict AKI in patients receiving high-dose MTX (HDMTX) is unclear. Therefore, this study sought to determine whether urinary TIMP2∗IGFBP7 and urinary NGAL measurements are predictive for AKI in patients with lymphoma hospitalized to receive HDMTX. Reporting of this observational study has been done according to the Modified STROBE Statement (Supplementary Appendix 1). Methods, including MTX-inclusive combination chemotherapy protocols (Supplementary Table S1) and the leucovorin dose adjustment schema (Supplementary Table S2) are detailed in Appendix 2 of the Supplementary Material.
Results
Baseline Demographics and Clinical Characteristics
Of 137 patients screened for participation, 77 were enrolled (Supplementary Figure S1). Table 1 details baseline demographics and clinical characteristics. Univariate analysis showed no association between age, sex, race, baseline chronic kidney disease, diabetes mellitus, serum creatinine, estimated GFR (eGFR), or predicted 24-hour urine protein and AKI.
Table 1.
Comparison of baseline demographics and laboratory values between patients with lymphoma who did and did not develop acute kidney injury within 96 hours of HDMTX administration
| Characteristics | All patients (N = 77) | Patients who developed AKI (n = 16) | Patients who did not develop AKI (n = 61) | Odds ratio (95% CI) | P value |
|---|---|---|---|---|---|
| Age (yr), median (IQR) | 68.4 (59–76) | 66.4 (61–69) | 69.9 (59–76) | 1.00 (0.96–1.04) | 0.90 |
| Male, n (%) | 51 (66) | 14 (88) | 37 (61) | 4.54 (0.95–21.78) | 0.059 |
| Race, n (%) | |||||
| Caucasian | 71 (92) | 13 (81) | 58 (95) | Reference | |
| Other/unknown | 6 (8) | 3 (19) | 3 (5) | 4.46 (0.81–24.66) | 0.086 |
| Chronic kidney disease | 4 (5) | 0 (0.0) | 4 (7) | 0.39(0.01–10.64) | 0.57 |
| Diabetes mellitus | 14 (18) | 2 (13) | 12 (20) | 0.58 (0.12–2.92) | 0.51 |
| Body surface area, median (IQR) | 2.0 (1.8–2.1) | 2.1 (2.0–2.2) | 1.9 (1.8–2.1) | 1.22 (0.97–1.54)a | 0.083 |
| Lymphoma diagnosis subtype, n (%) | |||||
| DLBCL | 43 (58) | 9 (60) | 34 (58) | Reference | |
| Primary DLBCL of the CNS | 22 (30) | 3 (20) | 19 (32) | 0.60 (0.14–2.47) | 0.48 |
| EBV-positive DLBCL of elderly | 6 (8) | 2 (13) | 4 (7) | 1.89 (0.30–12.01) | 0.50 |
| Other | 3 (4) | 1 (7) | 2 (3) | 1.89 (0.30–12.01) | 0.50 |
| Bone marrow involvement, n (%) | 10 (13) | 5 (31) | 5 (8) | 5.09 (1.26–20.60) | 0.023 |
| Percent involvement, median (IQR) | 15 (10–30) | 10 (10–10) | 20 (20–30) | ||
| Serum creatinine, mg/dl, median (IQR) | 0.9 (0.74–1.08) | 0.9 (0.8–1.09) | 0.9 (0.72–1.08) | 1.07 (0.87–1.31)a | 0.54 |
| Estimated CrCl, ml/min, median (IQR) | 90 (70–120) | 96 (78–119) | 86 (70–120) | 1.01 (0.90–1.13)b | 0.93 |
| Estimated GFR, ml/min, median (IQR) | 92 (77–106) | 99 (84–116) | 91 (77–106) | 1.09 (0.89–1.34)b | 0.41 |
| Predicted 24-hour urine protein, mg/24 h, median (IQR) | 210 (148–306) | 235 (155–287) | 194 (148–310) | 1.00 (0.97–1.03)b | 0.93 |
| Hemoglobin, g/dl, median (IQR) | 11.4 (10.2–13.2) | 11.6 (10.0–13.5) | 11.3 (10.3–13.1) | 1.06 (0.80–1.40) | 0.69 |
| Albumin, g/dl, median (IQR)c | 3.8 (3.5–4.2) | 3.9 (3.3–4.3) | 3.8 (3.5–4.2) | 0.99 (0.88–1.10)a | 0.80 |
| LDH, median (IQR) | 212 (180–277) | 218 (183–398) | 212 (178–263) | 1.04 (0.99–1.09)b | 0.096 |
AKI, acute kidney injury; CNS, central nervous system; CrCl, creatinine clearance (per Cockcroft Gault equation); DLBCL, diffuse large B-cell lymphoma; EBV, Epstein-Barr virus; GFR, glomerular filtration rate; HDMTX, high-dose methotrexate; IQR, interquartile range; LDH, lactate dehydrogenase.
Per 0.1 unit increase.
Per 10 unit increase.
Only available in 61 (n = 14 who developed AKI and n = 47 who did not develop AKI).
Acute Kidney Injury
AKI occurred in 16 patients (21%) within 96 hours of HDMTX. Overall, 5 patients (31%) developed stage 1 AKI, 5 patients (31%) had stage 2 AKI, and 6 patients (38%) experienced stage 3 disease. The median time from the start of HDMTX infusion to AKI onset was 49 (interquartile range: 26–56) hours. No patient required renal replacement therapy for AKI.
The incidence of AKI varied according to the MTX dose prescribed. AKI occurred in 4 patients (11.8%) prescribed 8 g/m2 as central nervous system lymphoma treatment. In addition, AKI developed in 10 patients (32%) who received 3.5 g/m2 (treatment: n = 4; prophylaxis: n = 8) and 2 patients (22%) given 1.5 g/m2 as prophylaxis against central nervous system relapse. Patients prescribed MTX dose of 3.5 g/m2 were significantly more likely to develop AKI than those prescribed 8 g/m2 (odds ratio = 3.75, 95% CI: 1.07–13.12, P = 0.039).
Urinary Biomarkers
Median baseline TIMP2∗IGFBP7 concentration was 0.3 (interquartile range: 0.1–0.8). The distribution of baseline TIMP2∗IGFBP7 included 40 patients (52%) at <0.3, 23 (30%) between 0.3 and 0.99, and 14 (18%) at ≥1.0. There were 3 patients who had a baseline TIMP2∗IGFBP7 ≥ 2.0. Baseline TIMP2∗IGFBP7 did not demonstrate any correlation with baseline serum creatinine at r = −0.08 (95% CI: −0.30 to 0.15, P = 0.48) or eGFR at r = 0.13 (95% CI: −0.10 to 0.34, P = 0.27).
Patients with a baseline TIMP2∗IGFBP7 concentration of ≥0.3 were significantly more likely to develop any AKI (odds ratio = 6.68, 95% CI: 1.72–25.94, P = 0.006) or AKI stage 2 or 3 (odds ratio = 6.11, 95% CI: 1.22–30.49, P = 0.027). AKI occurred in 7.5%, 30%, and 54% of patients with a baseline TIMP2∗IGFBP7 <0.3, between 0.3 and 0.99, and ≥1.0, respectively. The area under the receiver operating characteristic curve (AUC) of TIMP2∗IGFBP7 to predict AKI is depicted in Figure 1 (AUC = 0.70, 95% CI: 0.54–0.86, P = 0.008).
Figure 1.
AUC demonstrating the diagnostic performance for baseline urinary TIMP2∗IGFBP7 to predict acute kidney injury (AUC = 0.70, 95% CI: 0.54–0.86, P = 0.008). AUC, area under the receiver operating characteristic curve.
The median baseline urinary NGAL concentration was 21 ng/ml (interquartile range: 13–44). Urinary NGAL was not significantly associated with any AKI or severe AKI development. Furthermore, the AUC of urinary NGAL to predict AKI was 0.54 (95% CI: 0.39–0.69).
Discussion
This prospective observational study of patients with lymphoma revealed a 21% incidence of AKI after HDMTX. Notably, 69% of the cases were stage 2 or 3 AKI. Patients with baseline urinary TIMP2∗IGFBP7 concentrations of ≥0.3 were significantly more likely to develop any AKI and stage 2 or 3 AKI. Urinary NGAL failed to demonstrate any utility in this cohort.
We did not observe an association between urinary NGAL concentrations and AKI. In addition, the AUC suggests no ability of urinary NGAL to predict AKI in patients with lymphoma prescribed HDMTX. NGAL for assessment of kidney injury after HDMTX has currently been limited to pediatric patients with acute lymphoblastic leukemia or osteosarcoma with variable results.9,S6,S7 One study of 20 pediatric patients with acute lymphoblastic leukemia prescribed HDMTX at 5 g/m2 over 24 hours found no utility for plasma or urinary NGAL measurements.9 Conversely, serum NGAL values were significantly higher after a 24-hour MTX infusion of 3 g/m2 in children with acute lymphoblastic leukemia who experienced AKI than those who did not.S6 Furthermore, 12 patients with osteosarcoma receiving HDMTX 12 g/m2 demonstrated that elevations in AKI biomarkers could reflect HDMTX-associated nephrotoxicity; however, the regimen prescribed also contained cisplatin, a known nephrotoxin.S7 A younger patient population, longer MTX infusion, different supportive care processes, and different biomarker sample timing preclude direct comparisons to our study. Future studies should confirm our results and investigate whether serial urinary NGAL measurements can identify optimal parameters to predict renal insult after HDMTX.
Our results identified an association between baseline urinary TIMP2∗IGFBP7 and any AKI, including stage 2 or 3 AKI, after HDMTX. This differs from studies in critically ill or postsurgical patients where baseline concentrations were similar and increases in those with AKI occurred shortly after a known inciting event.6,7,S8,S9 Little is known about the role of urinary TIMP2 and IGFBP7 concentrations for AKI prediction in patients with lymphoma receiving HDMTX. TIMP2∗IGFBP7 has currently only been studied in adult patients with cancer receiving platinum chemotherapy.S10,S11 One study in patients with lung cancer demonstrated no TIMP2 and IGFBP7 concentration differences in those with or without AKI after cisplatin.S11 However, another study in patients receiving cisplatin concluded that urinary TIMP2∗IGFBP7 was a useful tool for AKI prediction.S10 Elevations in baseline urinary TIMP2∗IGFBP7 in our population may represent subclinical AKI given the normal serum creatinine and eGFR values.8,S12,S13 A cancer diagnosis at hospital admission has been associated with moderate-to-severe hospital-acquired AKI.S14 This risk would only be compounded by administering HDMTX for cancer treatment. In addition, patients with cancer have concurrent, acute risk factors for AKI, such as hypotension and hypovolemia.4 Given that the AUC demonstrated acceptable discrimination of TIMP2∗IGFBP7 to predict AKI, these hypothesis-generating data compel investigations that explore the utility of this biomarker as a component within a screening tool for AKI risk estimation in adult patients with lymphoma who require HDMTX therapy.S15,S16
Limitations of this study include the single-center setting and moderate sample size. Second, a full characterization of baseline kidney health, including baseline urine albumin or baseline urine protein, was not part of our protocol and was not always available near the HDMTX administration on electronic health record review; although, assessment beyond serum creatinine is atypical before HDMTX in these patients, especially when the serum creatinine value and corresponding estimated creatinine clearance or eGFR represent normal kidney function. Importantly, few patients in our study had evidence of pre-existing chronic kidney disease based on eGFR, and none fit AKI criteria at admission. In addition, kidney assessment and management after HDMTX-associated AKI lacks standardization with testing and imaging at the discretion of the primary service and requires optimization. Last, three different, protocol-defined doses of HDMTX were prescribed. However, our patient population possessed similar baseline demographics among the dosing tiers. To our knowledge, this is the largest study of urinary TIMP2∗IGFBP7 and urinary NGAL in adult patients with lymphoma and the first to attempt evaluating these biomarkers in adult patients with lymphoma receiving HDMTX.
Conclusion
AKI occurred frequently in patients with lymphoma receiving HDMTX, and most cases were moderate or severe. Urinary NGAL lacked utility in predicting AKI. Increased baseline urinary TIMP2∗IGFBP7 was associated with AKI and severe AKI development.
Disclosure
EFB provides consultation for FAST Biomedical and Wolters Kluwer, unrelated to this work. All the other authors declared no competing interests.
Acknowledgments
The authors thank Ms. Holly A. Thorson and the Mayo Clinic Clinical Research Trials Unit staff for their assistance with patient enrollment and sample collection. This project was supported in part by the Hematology Oncology Pharmacy Association Foundation Research Grant Program (JNB), the CTSA Grant Number UL1 TR002377 from the National Center for Advancing Translational Science (JNB), and the National Institute Of Allergy And Infectious Diseases of the National Institutes of Health under Award Number K23AI143882 (EFB). Its contents are solely the authors' responsibility and do not necessarily represent the official views of the National Institutes of Health.
Data Availability Statement
The deidentified data supporting this study’s findings are available on request from the corresponding author. The data are not publicly available owing to privacy or ethical restrictions.
Author Contributions
JNB, KBK, KCM, ADR, JCL, CDG, CAT, NL, TEW, and EFB wrote the manuscript. JNB, KBK, KCM, ADR, JCL, CAT, NL, TEW, and EFB designed the research. JNB, KBK, KCM, ADR, CDG, and EFB performed the research. JNB, JMR, JNB, KBK, KCM, ADR, JCL, CDG, CAT, NL, TEW, and EFB analyzed the data. JNB, KBK, KCM, ADR, JCL, CDG, CAT, NL, TEW, and EFB played an important role in interpreting the results.
Footnotes
Appendix 1. Modified STROBE statement.
Appendix 2. Methods.
Table S1. MTX-inclusive combination chemotherapy regimens.
Table S2. Leucovorin dose adjustment protocol.
Figure S1. Flow diagram of patient enrollment.
Supplementary References.
Supplementary Material
Appendix 1. Modified STROBE statement.
Appendix 2. Methods.
Table S1. MTX-inclusive combination chemotherapy regimens.
Table S2. Leucovorin dose adjustment protocol.
Figure S1. Flow diagram of patient enrollment.
Supplementary References .
References
- 1.May J., Carson K.R., Butler S., Liu W., Bartlett N.L., Wagner-Johnston N.D. High incidence of methotrexate associated renal toxicity in patients with lymphoma: a retrospective analysis. Leuk Lymphoma. 2014;55:1345–1349. doi: 10.3109/10428194.2013.840780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barreto J.N., Peterson K.T., Barreto E.F., et al. Early, empiric high-dose leucovorin rescue in lymphoma patients treated with sequential doses of high-dose methotrexate. Support Care Cancer. 2021;29:5293–5301. doi: 10.1007/s00520-021-06106-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Batchelor T., Carson K., O’Neill A., et al. Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96–07. J Clin Oncol. 2003;21:1044–1049. doi: 10.1200/JCO.2003.03.036. [DOI] [PubMed] [Google Scholar]
- 4.Lameire N., Van Biesen W., Vanholder R. Acute renal problems in the critically ill cancer patient. Curr Opin Crit Care. 2008;14:635–646. doi: 10.1097/MCC.0b013e32830ef70b. [DOI] [PubMed] [Google Scholar]
- 5.Kashani K., Rosner M.H., Ostermann M. Creatinine: from physiology to clinical application. Eur J Intern Med. 2020;72:9–14. doi: 10.1016/j.ejim.2019.10.025. [DOI] [PubMed] [Google Scholar]
- 6.Kashani K., Al-Khafaji A., Ardiles T., et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit Care. 2013;17:R25. doi: 10.1186/cc12503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bihorac A., Chawla L.S., Shaw A.D., et al. Validation of cell-cycle arrest biomarkers for acute kidney injury using clinical adjudication. Am J Respir Crit Care Med. 2014;189:932–939. doi: 10.1164/rccm.201401-0077OC. [DOI] [PubMed] [Google Scholar]
- 8.Ostermann M., Zarbock A., Goldstein S., et al. Recommendations on acute kidney injury biomarkers from the acute disease quality initiative consensus conference: a consensus statement [published correction appears in JAMA Netw Open. 2020;3:e2029182] JAMA Netw Open. 2020;3 doi: 10.1001/jamanetworkopen.2020.19209. [DOI] [PubMed] [Google Scholar]
- 9.Ylinen E., Jahnukainen K., Saarinen-Pihkala U.M., Jahnukainen T. Assessment of renal function during high-dose methotrexate treatment in children with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2014;61:2199–2202. doi: 10.1002/pbc.25137. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The deidentified data supporting this study’s findings are available on request from the corresponding author. The data are not publicly available owing to privacy or ethical restrictions.