DISCUSSION
In this cross-sectional study, we found a high incidence of AKI in children diagnosed with ALL, with 33% of subjects having at least one AKI episode and the majority of AKI episodes noted at time of ALL diagnosis. Most patients had mild AKI (KDIGO stage 1). Patients with high-risk ALL had a higher prevalence of AKI compared to patients with standard-risk ALL. This may have been due to greater likelihood of leukemic infiltration in the kidneys at diagnosis, greater risk of tumor lysis syndrome with initial chemotherapy, and/or the more toxic chemotherapy regimens used in high-risk ALL which can be directly nephrotoxic and increase risk for complications such as sepsis.
No patients were found to have CKD in this early follow-up study, a better short-term outcome than that reported by studies from previous decades, and perhaps a reflection of better supportive care now available for children with ALL.4,6,7 Almost a third of patients were noted to have hyperfiltration, however, and in other populations this identifies increased risk for subsequent renal injury and progressive CKD.33,34 The mean last recorded GFR of the overall group was supraphysiologic, worrisome for potential future renal complications. Additionally, 24% of patients who completed a urinalysis were noted to have either proteinuria, albuminuria or both.
The low prevalence of CKD in our cohort is likely related to short duration of follow-up time. In a recent Israeli study evaluating the risk of ESRD in childhood cancer survivors with median follow-up time of 30 years, participants with history of malignancy in childhood had adjusted HR of 3.2 of developing ESRD compared to participants without no history of tumors/malignancy in childhood.35 In the 1990’s, Krawczuk-Rybak et al. found 5/37 (13.5%) childhood ALL survivors had abnormal creatinine clearance when followed 3.9±3.7 years after completion of chemotherapy6. Yetgin et al. detected reduced GFR in 22/116 (19%) of pediatric ALL patients treated mostly in the 1990’s who were assessed 18-96 months after completion of chemotherapy4. Yetgin also performed urinalysis on all subjects post chemotherapy and found no evidence of proteinuria in these patients. However, 7/116 (6%) of patients had high urine B2-Microglobulin, a marker of renal tubular proteinuria. Hovi et al. reported GFR < 85ml/min/1.73m2in 6/60 (10%) of children with ALL followed 1-9 years after completion of chemotherapy, but this study presents data from over 25 years ago.7
The high prevalence of hyperfiltration, proteinuria, and/or albuminuria noted in our cohort suggests that these children with ALL, treated since 2015, are at risk of developing CKD in the long term. This phenomenon is well-described in patients with diabetic nephropathy. Glomerular hyperfiltration commonly occurs early in diabetes secondary to increased glomerular capillary pressure. Over time, sustained elevated GFR can result in mesangial and podocyte injury and eventually glomerular scarring leading to CKD progression and ESRD. 33,34,36The presence of albuminuria/proteinuria in patients with diabetes or CKD has been strongly associated with increased risk of subsequent renal function decline and measures that are successful at reducing albuminuria/proteinuria have been shown to slow the renal disease progression in these populations.36,37
Our study also found that children with ALL have a high prevalence of HTN.
Almost half of our study patients (42%) fulfilled criteria for outpatient diagnosis of HTN based on clinic BP measurements, termed casual BP readings. Our findings are consistent with prior ALL survivorship research, which has reported incidences of HTN in children with ALL ranging from 5-65%, with variations likely affected by study methodology and HTN definitions.5,10,15,20,21 The high incidence of HTN is likely multifactorial, related to past or present corticosteroid therapy, chemotherapy, prior AKI, and/or renal leukemic infiltration. Prior studies have also proposed metabolic syndrome and obesity as risk factors for developing chronic HTN in adults who survive childhood leukemia.5,8,9,12-14
In the 21 patients who completed an ABPM study, we found that the percentage of individuals with HTN was much higher, since 67% had an abnormal high BP value. The majority of patients with abnormal ABPM had abnormal nocturnal dipping pattern. Ociepa et al. is the only other study to report ABPM results as a method to diagnose HTN in children with ALL. This study detected a higher prevalence of childhood survivors of ALL with HTN via ABPM (37%, 30/81), although it is perhaps noteworthy that their study had a longer mean time of follow-up (60 months) compared to our study (36 months). Similar to our study, Ociepa et al. found that childhood survivors of ALL were more likely to have abnormal nocturnal dipping pattern (non-dippers, reverse-dippers, and extreme-dippers) compared to control population. In healthy children during sleep, a 10-20% physiologic decrease of BP is normally noted. The lack of this physiologic decrease is termed non-dipping and BP decrease in excess of 20% is considered extreme dipping. Both non-dipping and extreme dipping have been associated with increased risk of cardiovascular or cerebral ischemic/hemorrhagic events in adults.38-42 In pediatric subjects with systemic lupus erythematosus, non-dipping has been associated with endothelial dysfunction and atherosclerotic changes.43 Similarly, in adolescents with diabetes, non-dipping pattern has been associated with LVH and increased carotid intimal media thickness44,45. Further studies of these ABPM abnormalities and outcomes in children with ALL will help understand the prognostic implications and identify opportunities for useful interventions in these children.
Limitations of this study include its single-center design with relatively small sample size and short duration of follow-up. A multi-center study with larger sample size could usefully confirm the findings of this study and help determine distinct risk factors for developing renal dysfunction and hypertension. It should also be noted in this study that urine samples were collected at the time of clinic appointment for convenience and were not first morning samples. Thus, orthostatic proteinuria (a benign phenomenon occurring in up to 7% of children and adolescents) could account for some of the proteinuria/albuminuria noted in our findings46.
In conclusion, our study showed that, among children with ALL, there is a high prevalence of HTN based on casual BP readings obtained in the clinic setting and a high prevalence of impaired nocturnal dipping based on ABPM. Both abnormalities are associated with increased risk of cardiovascular disease and cerebrovascular events later in life. While the prevalence of CKD is very low, children with ALL have a high prevalence of hyperfiltration and/or proteinuria/albuminuria suggesting increased risk of developing CKD in the future.
Interestingly, screening urinalysis has been removed from the Children’s Oncology Group long-term follow-up guidelines for children who survive ALL.47 The value of a screening urinalysis in detection of early evidence of renal disease is highlighted by the Prevention of Renal and Vascular End-Stage Disease (PREVEND) study conducted of over 40,000 adult participants which indicated the usefulness of albuminuria in early detection of CKD. PREVEND found that of individuals who required renal replacement therapy after 9-years follow-up, 58% had moderately increased albuminuria (≥20mg/L) on early screening urine sample. Increased albuminuria was an independent risk factor for progression to ESRD independent of GFR.48Based on our findings and other recent studies,3,7,20we propose that children with ALL should undergo regular (yearly) renal function testing, urinalysis and BP measurements to screen for early signs of renal abnormalities. Longer term and larger studies will be needed to confirm our findings and define predictors for these adverse outcomes in children with ALL.
As the long-term cure rate of ALL has improved to greater than 90%, the challenge now extends to achieving the same cure rates with reduced short-term and long-term toxicities for these children. One treatment modality which shows promise in this goal is immunologic therapy. There have been several recent trials demonstrating the safety and efficacy of antibody-based and T cell-based therapies for ALL.49,50 Our findings demonstrate that current conventional therapies for childhood ALL are accompanied by concerning frequencies of early renal dysfunction and hypertension. The shift towards targeted immunologic therapies may mitigate some of these long-term renal consequences of conventional therapy and provide better quality of life in survivors of childhood ALL.
Conflicts of Interest Statement: The authors have no conflicts of interest
Acknowledgements: This work was supported by an intramural grant from Nationwide Children’s Hospital