An Increase in Baseline Level of Urinary Liver-Type Fatty Acid-Binding Protein Linked with Poor Prognosis after Allogeneic Hematopoietic Stem Cell Transplantation

Shingai, Naoki; Morito, Taku; Ikegawa, Shuntaro; Yamamoto, Keita; Hino, Yutaro; Harada, Kaito; Igarashi, Aiko; Watanabe, Ken; Senoo, Yasushi; Najima, Yuho; Doki, Noriko; Kobayashi, Takeshi; Kakihana, Kazuhiko; Ohashi, Kazuteru; Sakamaki, Hisashi; Ando, Minoru

doi:10.1182/blood.V124.21.3861.3861

Abstract

Introduction: Stem cell transplantation (SCT) involves a great risk of acute kidney injury (AKI), which is largely committed to fatality derived from SCT. Liver-type fatty acid-binding protein (L-FABP) is a 15kDa protein expressed in human renal proximal tubular cells, and is shed into urine in response to renal tissue ischemia or oxidative stress. Urinary L-FABP (uL-FABP) has been reported to be not only a sensitive biomarker to detect incipient AKI but also a potential marker for something more than a measure of renal insult. This study addressed utility of uL-FABP, as it has not been fully understood on the clinical platform of SCT.

Patients and Methods: A prospective cohort study was conducted in 81 patients who received allogeneic SCT in our hospital from April 2009 to March 2012 (median age, 49 years; median [range] follow-up period, 771 [2-1923] days). All had hematological malignancies, and were recipients of the first SCT. Associations of increased uL-FABP before executing conditioning therapy (at baseline) and clinical outcomes including incidence of AKI, non-relapse mortality (NRM), and overall survival (OS) were investigated. AKI was defined, based on the creatinine (Cr) criteria of the 2013 Kidney Disease Global Outcomes. AKI that developed within the first 30 days after SCT was considered in this study. The recipients were stratified into an elevated and a normal uL-FABP group at baseline, according to the reference value for healthy adult individuals (8.5µg/gCr). Hazard ratio (HR) with its 95% confidence interval (CI) of uL-FABP elevation was calculated for OS using the Cox proportional hazard model and for NRM using the Fine-Gray proportional hazards model, adjusted for confounders which had P -values ≤ 0.2 in univariate analyses.

Results: Patients had been diagnosed with acute myeloid leukemia (n = 36), acute lymphoid leukemia (n = 23), myelodysplastic syndrome (n = 17), and myeloproliferative neoplasm (n = 5). The elevated and normal uL-FABP groups included 17 (21%) and 64 (79%) patients, respectively. The proportion of high disease risk was significantly greater in the elevated than in the normal uL-FAB

P group (70.6 % vs. 31.3%, P = 0.005). There were no significant differences in age, serum Cr level, and the proportion of gender, hematopoietic cell transplantation comorbidity index (HCT-CI) ≥3, myeloablative conditioning regimen, stem-cell sources (bone marrow, peripheral blood, cord blood), related donor type and matched human leukocyte antigen compatibility. AKI emerged in 28 recipients (34.6%) at day 30? following SCT (median [range], 14 [0-30] days). The cumulative incidence [95% CI] of AKI was significantly greater in the elevated than in the normal uL-FABP group (64.7% [36-83.1 %] vs. 26.6% [16.4-37.8 %], P<0.001), shown in Figure 1. A total of 35 patients (43.2%) died within 2 years after SCT: 13 (37.1%) died of relapse or progression of disease and 22 (62.9%) died of non-relapse death. The OS at the time of 2 year was significantly lower in the elevated than in the normal uL-FABP group (23.5% vs. 64.7%, P<0.001), shown in Figure 2. NRM was significantly greater in the elevated than in the normal uL-FABP group (58.8% vs. 19.2%, P<0.001) at the time of 2 year, as shown in Figure 3. Multivariate analysis revealed that the presence of elevated uL-FABP at baseline was significantly associated with OS (HR [95% CI]; 2.99 [1.3-6.87], P =0.01) and NRM (HR [95% CI]; 4.75 [1.83-12.3], P=0.001). In addition, high age, unrelated donor type, high disease risk, HCT-CI ≥3, and the development of veno-occlusive disease were involved in either OS or NRM. There was no significant difference in the relapse or progression rate within 2 years between the elevated and normal uL-FABP groups (20.7% vs. 29.4%, P=0.43)

Conclusions: uL-FABP is a simple and non-invasive measurement to target patients at increased risk of AKI, NRM and OS following allogeneic SCT. The presence of antecedent subclinical renal injury, which can be detected by uL-FABP elevation at baseline, may provide prognostic information beyond renal outcome after allogeneic SCT.