#6652 acute kidney injury in patients with multiple myeloma submitted to autologous hematopoietic cell transplant: a cohort analysis

Abstract Background and Aims The incidence of most haematological neoplasms has been increasing and with them hematopoietic cell transplantation (HCT) as it provides better survival rates to several of these malignancies. There is a wide range of acute kidney injury (AKI) reported incidence in HCT patients, mostly due to the heterogeneity of AKI definitions and study populations that include different underlying illnesses. We aimed to evaluate incidence, severity and risk factors of AKI by KDIGO classification considering creatinine and urinary output (UO) criteria in patients with multiple myeloma (MM) submitted to autologous HCT. Method We conducted a retrospective cohort study including patients with MM submitted to autologous HCT admitted in a tertiary hospital between Jan./2005-Dec./2015 with a 3 year follow-up period. To evaluate AKI cumulative incidence, risk factors and impact on relapse, survival analysis methods considering death as competing were used. Multivariate stepwise analysis using logistic regression was used to evaluate risk factors associated with AKI. Results 143 patients were included [median age was 50.5 (59.2-63.6) years, 61.5% male and 90.9% Caucasian]. 37.1% had hypertension, 12.6% diabetes mellitus and 10.5% chronic kidney disease (CKD) [median overall baseline eGFR 100.4 (89.5-110.8) mL/min/1.73 m2]. Median percentage of plasma cells on bone marrow aspirate was 30 (17-67) %, serum M-protein 3.3 (2.2-6.6) g/dL and serum titre of the affected light chain 3560 (890-7800) mg/dL. 49 patients had cytogenetic abnormalities. 49% of patients had R-ISS stage I at diagnosis. Median number of previous lines of therapy was 1 (1-2) and the median number of cycles 4 (3-4). 34.5% were previously submitted to radiotherapy. At admission, mean haemoglobin was 11.7 (10.8-12.4) g/dL, neutrophils 3375 (2330-4765) /mm3, serum urea 33 (28-44) mg/dL, uric acid 5 (4.1-6) mg/dL, phosphate 3.4 (3-2.9) mg/dL, LDH 333 (297-390) U/L, albumin 3.9 (3.7-4.2) mg/dL and alanine transaminase 18 (14-27) U/L. Fever (77.6%), exposure to nephrotoxins (74.1%), low grade mucositis (55.9%) and sepsis (26.6%) were common during hospital admission. Cumulative incidence of AKI at 100 days was 49.7% (71 patients) with median time to event of 8 (5-13) days. 85.9% of patients had stage 1 AKI at diagnosis, although 16.4% of these (10 patients) degraded to worse stages. On univariable analysis, the predictor variables with impact on AKI were higher age (p=0.057), higher body mass index (BMI) (p=0.048), HCT-comorbidity index ≥2 (p=0.005), lower baseline eGFR (p<0.001), previous CKD (p=0.001), presence of amyloidosis (p=0.078), higher R-ISS (p=0.002), presence of cytogenetic abnormalities (p=0.055), development of sepsis (p=0.016), fever (p=0.044) and higher grade mucositis (p<0.001), exposure to nephrotoxins (p=0.007), higher neutrophils count (p=0.021), serum urea (p<0.001) and LDH (p=0.002) and lower phosphate (p=0.049), and alanine transaminase (p=0.077). Multivariable analysis showed an independent association of AKI with BMI [HR 1.08 (1.03-1.14), p=0.004], CKD [HR 3.00 (1.55-5.78), p=0.001], amyloidosis [HR 2.73 (1.40-5.33), p=0.003], high grade mucositis [HR 2.35 (1.37-4.02), p=0.002], exposure to nephrotoxins [HR 2.20 (1.08-4.47), p=0.029], leukocyte count [HR 1.01 (1.01-1.01), p<0.001] and serum phosphate [HR 0.57 (0.39-0.84), p=0.004] maintained an independent association with AKI. Conclusion Our results show that AKI occurs in almost half of the patients with MM submitted to HCT and reinforce the need for prevention, monitoring and early approach to AKI in these patients particularly in those with higher BMI, CKD, secondary amyloidosis, higher leukocyte count and lower serum phosphate at admission, as well as in those that develop mucositis and are exposed to nephrotoxins. We also believe considering both SCr and UO KDIGO criteria for AKI may have increased the reliability of our data, this being the first study to include both criteria in this population of patients.