Risk of Kidney Outcomes and Hypertension in Survivors of Wilms Tumor: A Prospective Cohort Study

Abstract

Objective:

To assess the prevalence of therapy-related kidney outcomes in survivors of Wilms tumor (WT).

Study design:

This prospective cohort study included survivors of WT who were ≥5 years old and ≥1 year from completing therapy, excluding those with pre-existing hypertension or prior dialysis or kidney transplant. Participants completed 24-hour ambulatory blood pressure monitoring (ABPM). Abnormal blood pressure (BP) was defined as ≥90^th percentile. Masked hypertension was defined as having normal office BP and abnormal ABPM findings. Urine was analyzed for KIM-1, IL-18, EGF, albumin, and creatinine. Estimated glomerular filtration rate (eGFR) was calculated using the bedside CKiD equation. Recent kidney ultrasounds and echocardiograms were reviewed for contralateral kidney size and left ventricular hypertrophy (LVH), respectively. Clinical follow-up data was collected for approximately 2 years following study enrollment.

Results:

Thirty-two participants (median age 13.6 [IQR: 10.5-16.3] years; 75% ≥Stage 3 WT) were evaluated at a median of 8.7 years (IQR: 6.5-10.8) post-therapy. Twenty-nine participants underwent unilateral radical nephrectomy, 2 bilateral partial nephrectomy, and 1 radical and contralateral partial nephrectomy. 72% received kidney radiotherapy and 75% received doxorubicin. Recent median eGFR was 95.6 ml/min/1.73m² (IQR: 84.6-114.0; 11 (34%) had an eGFR <90). Abnormal ABPM results were found in 22/29 participants (76%), masked hypertension in 10/29 (34%), and microalbuminuria in 2/32 (6%). 22/32 (69%) participants had abnormal EGF; few had abnormal KIM-1 or IL-18. Seven participants with previous unilateral nephrectomy lacked compensatory contralateral kidney hypertrophy. None had LVH.

Conclusion:

In survivors of WT, adverse kidney outcomes were common and should be closely monitored.

Introduction:

Wilms tumor (WT) is the most common primary kidney malignancy in children. Treatment often entails a combination of surgery, chemotherapy, and radiation therapy, depending on the stage and pathologic features of the tumor. Survival rates have steadily increased in the last few decades due to advances in treatment and management. However, with more children surviving WT, the long-term risk of subsequent comorbidities has also risen.^1-3 There is now more focus on preventing kidney disease and optimizing long-term health.

To date, there are limited data on the risk of adverse kidney outcomes, such as hypertension or chronic kidney disease (CKD), in patients after treatment for WT. Nephrotoxic chemotherapy, radiation, and nephrectomy are each associated with potential increased risk of CKD.^4-6 Since children with WT are still growing during completion of their therapy, kidney outcomes may have substantial ramifications later in adulthood.^7-10

Kidney function is typically quantified by estimating glomerular filtration rate (GFR). However, various non-GFR-based characteristics, including overt and masked hypertension, microalbuminuria, and novel urinary biomarkers of kidney injury, have been strongly associated with adverse outcomes.^11-15 Masked hypertension is defined as a normal in-office blood pressure (BP) but elevated BP measured with 24-hour ambulatory BP monitoring (ABPM). Both masked hypertension and microalbuminuria are independently associated with the development of CKD and left ventricular hypertrophy (LVH).^{11, 12} Novel urinary biomarkers such as interleukin-18 (IL-18),¹³ kidney injury molecule-1 (KIM-1),¹⁴ and epidermal growth factor (EGF)^{15, 16} have shown promising results as candidate proteins indicative of kidney injury (IL-18 and KIM-1) or repair (EGF). Collectively, these non-GFR-based biomarkers have not been well-studied in pediatric survivors of WT.

In this study, we prospectively enrolled survivors of WT and measured 24-hour ABPM and urinary albumin, IL-18, KIM-1, and EGF. We quantified the prevalence of masked hypertension, impaired eGFR, and abnormal urine studies. We hypothesized that masked hypertension and CKD would be common among survivors of WT.

Methods:

Study Design

This was a prospective cohort study of survivors of WT. All participants were recruited from the cancer survivorship clinics at the Children’s Hospital of Philadelphia (CHOP) from November 2016 through June 2018. Data were collected from the participants’ charts retrospectively and prospectively up to 2 years after they completed the study visit. The CHOP Institutional Review Board approved this study (IRB #16-013208). Written informed consent was obtained from all legal guardians or participants 18 years or older, along with assent, when appropriate, from children under 18 years of age.

Participants

Inclusion criteria were: a history of WT, age ≥5years, and ≥1 year since completion of all therapy for WT. Potential participants were excluded if they had a history of bilateral radical nephrectomy.

Study Procedures

On the day of the study visit, participants collected a first morning voided urine specimen at home and brought the sample to the visit (kept refrigerated at home prior to the visit). Aliquots of the urine specimen were collected and stored at −80 degrees Celsius, and all assays were performed in batch. Urine microalbumin was processed using fluorescence commercial kits from Siemens (Tarrytown, NY). Urine creatinine was measured using IDMS-calibrated enzymatic commercial kits from Roche (Indianapolis, IN). Urine IL-18 and KIM-1 were measured using electrochemiluminescence commercial kits from Meso Scale Discovery (Baltimore, MD). Urine EGF was measured using ELISA commercial kits from R&D Systems (Minneapolis, MN).

Casual clinic BP readings obtained using an automated sphygmomanometer as part of standard of care for the clinical visit were recorded. ABPM was performed using the Spacelabs 90217 device according to manufacturers’ instructions. Appropriate cuff size was determined according to the circumference of the non-dominant upper arm. Instructions on operating the ABPM device were reviewed with participants and their guardians. Written instructions were also provided. Participants were given the option to start the 24-hour monitoring period at the research visit or another more convenient time.

Outcomes

Abnormal ABPM systolic and diastolic BPs were defined using the 90^th percentile for age, sex, and height based on normative data from Wuhl et al.¹⁷ At our institution, standard clinical practice is to use the 90^th percentile BP as the cut-off for patients with underlying renal risk.^{18, 19} BP loads, defined as the percentage of BP readings ≥90^th percentile, were categorized as: normal <25%, borderline 25-50%, and elevated >50%. Nocturnal dipping of <10% was considered diminished. BP variability was quantified using average real variability, defined as the mean of the absolute difference of consecutive BP measurements during the ABPM period.²⁰ All ABPM reports were evaluated by a pediatric nephrologist on the study team and referred for further evaluation in nephrology clinic based on clinical judgement.

Casual clinic BP measurements were classified based on the 2017 AAP Clinical Practice Guideline²¹ and were defined as abnormal using the 90^th percentile for age, sex, and height. Masked hypertension was diagnosed if casual clinic BP readings were normal but ABPM loads or nocturnal dipping were abnormal.

Urine microalbumin and EGF were indexed to urine creatinine. Microalbuminuria and macroalbuminuria were defined as a urine albumin-to-creatinine ratio of 30-300 mg/g and >300 mg/g, respectively. Cut-offs for urinary KIM-1, IL-18, and EGF were defined from previous studies,^{22, 23} as these tests are not used clinically and do not have well-established normative ranges.

As most participants were children and not adults, estimated GFR (eGFR) was calculated using the revised bedside CKiD equation²⁴ with k=0.413 using the most recent serum creatinine and height (before or after study visit). Post-treatment imaging (ultrasound) and cardiac echocardiography data (if performed) were also collected from the most recent studies (before or after the study visit). Compensatory contralateral hypertrophy was documented based on kidney length on renal ultrasound using age-based reference data²⁵ and was defined as >1 standard deviation above normal, mean, age-adjusted kidney length.

We collected relevant follow-up data from subsequent survivorship clinic and nephrology clinic visits for approximately two years following the study visit. This included data on repeat ABPM and treatment with antihypertensive medications.

Covariates

Demographic information and clinical histories were obtained in person at the research visit and from chart review. Treatment histories for WT included details on past surgeries, chemotherapy, and radiotherapy.

Statistical Analysis

Descriptive statistics included median values and interquartile ranges or frequencies and proportions as appropriate. ABPM results were analyzed by median BP loads and variability and proportions with abnormal loads or diminished nocturnal dipping. The Fisher’s exact test was used for categorical data analysis, and the student’s t-test or Wilcoxon ranksum test was used for group comparisons of continuous data as appropriate. The Wilcoxon signed rank test was used to compare awake versus sleep period BP variability.

Given the limited sample size, we decided a priori not to perform multivariable regression analyses. The results are considered hypothesis-generating. All statistical analyses were performed with Stata (version 15, StataCorp, LLC, College Station, Texas).

Results:

A total of 32 survivors of WT participated (50% female, median age 13.6 years (interquartile range [IQR]: 10.5-16.3) at a median of 8.7 years (IQR: 6.5-10.8) after completion of treatment (Table 1). None were diagnosed with genetic syndromes associated with WT. WT stage classification was 6%, 19%, 31%, 38%, and 6%, with stages 1, 2, 3, 4, and 5, respectively. Unilateral radical nephrectomy was performed in 29 (91%) patients; two underwent bilateral partial nephrectomy, and one participant had a radical and contralateral partial nephrectomy. Radiotherapy to the kidney for either initial therapy or for recurrence was received by 72% of participants, with 48% of these further receiving radiotherapy to other organs (64% to the lung). Chemotherapy was received by all patients: 100% received vincristine/Actinomycin-D; 75% received doxorubicin; 16% received cyclophosphamide, 16% received etoposide; 13% received carboplatin. Median most recent eGFR was 95.6 ml/min/1.73m² (IQR: 84.6-114.0), with 34% having an eGFR <90 ml/min/1.73m². No patients had an eGFR <60 ml/min/1.73m².

Table 1.

Characteristics of study subjects.

Total subjects (M/F)	32 (16/16)
Median age in years (IQR)	13.6 (10.5-16.3; range 7.4-21.3)
Race, n (%)	White	25 (78%)
	Black	5 (16%)
	Other	2 (6%)
Wilms Tumor Stage, n (%)	Stage 1	2 (6%)
	Stage 2	6 (19%)
	Stage 3	10 (31%)
	Stage 4	12 (38%)
	Stage 5	2 (6%)
Median years from completion of treatment (IQR)*	8.7 (6.5-10.8; range 1.0-18.0)
Radiotherapy to the kidney, n (%)	23/32 (72%)
Radiotherapy to any other organ, n (%)	11/23 (48%)
	Lung	7/11 (64%)
	Abdomen	2/11 (18%)
	IVC	1/11 (9%)
	Whole Body	1/11 (9%)
Received chemotherapy	32/32 (100%)
Type of chemotherapy, n (%)	Vincristine	32 (100%)
	Doxorubicin	24 (75%)
	Actinomycin-d	32 (100%)
	Cyclophosphamide	5 (16%)
	Etoposide	5 (16%)
	Carboplatin	4 (13%)
Type of surgery, n (%)	Radical Nephrectomy	29 (91%)
	Bilateral Partial Nephrectomy	2 (6%)
	Radical Plus Partial Nephrectomy	1 (3%)
eGFR median (IQR)	95.6 (84.6-114.0) ml/min/1.73m2
GFR <90 ml/min/1.73m2	11/32 (34%)

GFR = glomerular filtration rate; IQR = interquartile range; IVC = inferior vena cava

^*median years from completion of treatment available only in 29 patients who completed ABPM

Of 32 participants, 17 (53%) had elevated casual clinic BP readings. Of these, nine had elevated BP, six met criteria for stage 1 hypertension, and two had readings consistent with stage 2 hypertension. Twenty-nine participants returned completed and usable ABPM results (Table 2). Of these participants, 31% and 14%, had either borderline or elevated systolic BP loads, respectively, while awake. While asleep, 34% and 14% of participants had borderline or elevated systolic BP loads, respectively. Systolic nocturnal dipping was abnormal in 52%, and 28% had abnormal diastolic nocturnal dipping. Median values of BP variability ranged from 7.0-7.4 mm Hg for nighttime BPs and 8.4-8.8 mm Hg for daytime BPs. Daytime diastolic BP variability was significantly different from nighttime diastolic BP variability (p = 0.008). Spearman correlations between respective BP load and BP variability were all statistically significant (rho ranging 0.41-0.69, all p <0.03) except for nighttime systolic BP. Any ABPM abnormality was found in 22 of 29 participants (76%). Stratification of ABPM results by casual clinic BPs showed 10 of 29 (34%) participants had masked hypertension (Table 3). Only 1 of these 10 with masked hypertension had an eGFR<90 ml/min/1.73m².

Table 2.

Ambulatory Blood Pressure Monitoring results.

	Median BP Load (IQR), %	Median BP variability (IQR), mmHg	Correlation between BP Load and variability, rho (p-value)	Subjects with BP load ≥25%, n (%)	Subjects with BP load >50%, n (%)
Systolic awake	11.4 (2.9-26.5)	8.4 (7.4-9.2)	0.41 (0.03)	9 (31%)	4 (14%)
Diastolic awake	11.5 (6.5-20.0)	8.8 (7.5-9.5)	0.59 (<0.001)	3 (10%)	0 (0%)
Systolic asleep	12.5 (5.9-26.7)	7.4 (6.2-9.4)	0.27 (0.16)	10 (34%)	4 (14%)
Diastolic asleep	15.4 (10.0-30.0)	7.0 (5.6-8.2)	0.69 (<0.001)	10 (34%)	1 (3%)
	Median decline (IQR), %	Subjects with nocturnal dipping <10%, n (%)
Systolic dipping	9.0 (6.0-13.3)	15 (52%)
Diastolic dipping	17.2 (9.4-21.5)	8 (28%)

BP = blood pressure; IQR = interquartile range; N/A = not applicable. n=29.

Table 3.

Casual clinic blood pressure measurements compared with ambulatory blood pressure monitoring measurements.

	Normal clinic BP	Abnormal clinic BP
N	13	16
Systolic awake load ≥25%	4 (31%)	5 (31%)
Diastolic awake load ≥25%	1 (8%)	2 (13%)
Systolic sleep load ≥25%	4 (31%)	6 (38%)
Diastolic sleep load ≥25%	4 (31%)	6 (38%)
Systolic dip <10%	8 (62%)	7 (44%)
Diastolic dip <10%	4 (31%)	4 (25%)
Any abnormality on ABPM	10 (77%)	12 (75%)

ABPM = ambulatory blood pressure monitoring; BP = blood pressure. n=29.

Thirty participants had routine post-treatment clinical imaging with kidney ultrasound, and 25 had transthoracic echocardiography performed (Table 4). The most recent kidney ultrasound was performed a median of 4.6 years (IQR 3.1-6.2) post-treatment. Of the 29 patients who underwent radical nephrectomy, contralateral compensatory hypertrophy was noted in 22 participants, with 15 having contralateral kidney length >2 standard deviations above the mean. The most recent cardiac echocardiography was performed a median of 9.4 years (IQR 8.1-11.7) post-treatment. Median left ventricular mass index was 31 (IQR 29-37) g/m^2.7; no participants met criteria for LVH.

Table 4.

Kidney and cardiac imaging at most recent evaluation.

Variable	N (%) or median (IQR)
Kidney imaging post-treatment, n (%)	30 (94%)
Median duration for ultrasound post-treatment in years, (IQR)	4.6 (3.1-6.2)
Median Kidney percentile % (IQR)	98.1 (87.2-99.8)
Median Kidney size SD (IQR)	2.1 (1.1-2.8)
Compensatory Hypertrophy >1 SD above normal	22 (of 29 with unilateral radical nephrectomy, 76%)
Compensatory Hypertrophy >2 SD above normal	15 (of 29 with unilateral radical nephrectomy, 52%)
Echo report available, n (%)	25 (78%)
Median time for echo post-treatment in years, (IQR)	9.4 (8.1-11.7)
Median time for echo post-ABPM in years, (IQR)	1.0 (0.0-1.1)
Median LVMI in g/m2.7, (IQR)	31.0 (29.0-37.0)

ABPM = ambulatory blood pressure monitoring; IQR = interquartile range; LVMI = left ventricular mass index; SD = standard deviation

Abnormal ABPM results were not significantly associated with eGFR<90 ml/min/1.73m², WT stage, nor contralateral kidney compensatory hypertrophy (data not shown).

Of the urinary biomarkers, 2 participants had microalbuminuria, both of whom had eGFR >90 ml/min/1.73m². Median urinary albumin:creatinine ratio was 5.5 (IQR 3.9-8.0) mg/g. Four and 3 participants had abnormal levels of IL-18 and KIM-1, respectively. 22 (69%) participants had abnormal creatinine-indexed, age-adjusted EGF (Table 5). When stratified by eGFR, creatinine-indexed EGF was significantly lower (i.e., more abnormal) in those with eGFR<90 ml/min/1.73m² (14.6 ± 6.6 vs. 22.1 ± 7.4, p = 0.009).

Table 5.

Urinary albumin, IL-18, KIM-1, and EGF levels (indexed to creatinine when indicated; n=32).

Urine biomarker	Cut-off used to define abnormal	Median (IQR)	N (%) abnormal
Albumin/Creatinine (mg/g)	≥30	5.5 (3.9-8.0)	2 (6%)
IL-18 (pg/mL)	Ages 5-9 years: ≥54.3	39.9 (21.2-67.6)	4 (12.5%)
	Ages 10-14 years: ≥88.8
	Ages 15-18 years*: ≥138.9
KIM-1 (pg/mL)	Ages 5-9 years: ≥1239.5	618.3 (365.4-720.4)	3 (9%)
	Ages 10-14 years: ≥1141.1
	Ages 15-18 years*: ≥1876.2
EGF/Creatinine (ng/mg)	Ages 7-12 years: ≤24	19.2 (12.7-26.1)	22 (69%)
	Ages 13-15 years: ≤26
	Ages 16-18 years*: ≤23

IQR = interquartile range

^*cut-off applied to participants >18 years of age

On clinical follow-up after the study visit, 14 and 23 of the 32 participants were seen in nephrology and survivorship clinics, respectively. Six participants underwent a total of 7 repeat ABPM studies. One participant was started on anti-hypertensive therapy with an ACE inhibitor. All urinalyses obtained for clinical care had negative or trace protein.

Discussion:

In this prospective study of pediatric survivors of WT, a substantial proportion of children screened for hypertension by ABPM had abnormal BP loads, especially during sleep. Diminished nocturnal dipping was the most frequent abnormality. One participant was started on anti-hypertensive therapy on clinical follow-up. Our findings suggest that long-term ABPM studies should be considered for routine post-treatment surveillance of CKD in survivors of WT.

Our results add to prior knowledge. Though the differential effect of the type of surgical approach (i.e., unilateral radical nephrectomy versus unilateral partial nephrectomy) on kidney outcomes is inconclusive, the detrimental effects of radiotherapy to the kidney⁶ are well known. In our study, we found that 34% of participants had an eGFR <90 ml/min/1.73m², presumably because of the additional nephrotoxic therapies received, including radiotherapy to the kidney in 72% of our participants. To date, multiple studies have assessed the prevalence of kidney injury as defined by eGFR in survivors of WT, but with heterogenous cohorts and results.^{4, 5, 26, 27} One retrospective study of 75 patients who underwent a unilateral radical nephrectomy for WT without radiotherapy or nephrotoxic chemotherapy reported that at a median follow-up of 20 years, 21% and 0% had an eGFR <90 ml/min/1.73m² and <60 ml/min/1.73m², respectively.⁴ The authors concluded that survivors of WT are at low risk of developing significant long-term kidney dysfunction.⁴ However, another study of longitudinal GFR measurements using technetium 99m-DTPA scans in 12 patients during therapy for WT found the greatest decrease in GFR to occur after nephrectomy (38% decline), compared to chemotherapy or radiotherapy.⁵

In addition to GFR-based assessment of kidney injury, hypertension is a well-established risk factor for and complication of CKD. In our study, we found that 53% of our participants had abnormal casual clinic BPs, 31-34% had borderline BP loads on ABPM, and >50% had abnormal nocturnal dipping. On repeat ABPM performed clinically for 6 patients, only one had ambulatory hypertension requiring anti-hypertensive therapy. BP variability has been shown to add prognostic value to ABPM results.²⁰ While our participants had low variability, BP variability did correlate significantly with corresponding BP load. In a similar recent study of 37 survivors of WT evaluated at a median of 22.5 years post-diagnosis, 40.5% were diagnosed with pre-hypertension or hypertension based on in-office BP measurements.²⁸ Similarly, a retrospective study of 30 survivors of WT at a median follow-up of 10 years found hypertension (BP ≥90^th percentile) in 33% of participants.²⁹ Among 40 adult survivors who underwent treatment for unilateral, nonsyndromic WT, 12 (30%) had hypertension.²⁷ The role of ABPM has been increasingly recognized as a useful adjunctive tool in managing pediatric CKD. One retrospective study compared ABPM results in 44 children with a solitary kidney and 25 age-matched controls and found no differences.³⁰ However, their group of children with solitary kidney constituted a heterogenous cohort, with 34% having acquired solitary kidney due to nephrectomy (only 4 due to WT). Another study compared ABPM in 15 survivors of WT to 20 age-, weight-, and height-matched healthy children and found that 24-hour, daytime, and night-time systolic BPs and night-time diastolic BPs were significantly higher in the WT group compared to controls.¹ Based on the high prevalence of abnormal ABPM findings in our study and others, we advocate for consideration of ABPM as part of routine post-treatment surveillance among survivors of WT.

Masked hypertension specifically is associated with LVH in children with CKD.¹² We found 34% of our participants had masked hypertension. While we did not find any participants with LVH on transthoracic echocardiography, this may be because of the opposing effects of anthracycline chemotherapy and systemic hypertension on left ventricular mass. Chemotherapy-related cardiotoxicity reduces left ventricular mass³¹ and causes heart failure,³² but systemic hypertension increases left ventricular mass.¹² Our abnormal BP findings may also not have reached the threshold to cause LVH.

Urinary biomarkers provide additional, promising non-GFR-based indices of structural kidney injury and repair. Among children with CKD without diabetes, microalbuminuria is significantly associated with CKD progression.¹¹ We found two children with microalbuminuria. In our study, we also tested novel urinary biomarkers including IL-18, KIM-1, and EGF. We found that most of the participants had normal IL-18 and KIM-1, but 69% had abnormal creatinine-indexed EGF values, which were significantly lower in those with eGFR <90 ml/min/1.73m². EGF is a pro-proliferative protein which is expressed by the ascending loop of Henle and distal convoluted tubule. EGF can mediate tubule cell regeneration and is a marker of functional tubular cell mass. Urine EGF has been found to be significantly lower in children with CKD compared to healthy controls¹⁶ and is directly correlated with eGFR,¹⁵ as confirmed in our study. In a study of 80 children with congenital anomalies of the kidney and urinary tract, children with a nephrectomy were found to have the lowest urine EGF/Cr levels as compared with other children with congenital anomalies or healthy controls.³³ In our study, the low urine EGF/Cr concentrations may reflect the reduction in functional tubular mass after nephrectomy. A plausible but hypothetical reason for the mostly normal findings for IL-18 and KIM-1 is that, overall, our cohort has mostly preserved eGFR (median 96 ml/min/1.73m²), few comorbidities, and limited ongoing tubular injury.

Our study does have limitations. Our sample size is small, which limits statistical power and prevents stratifying analyses by surgical approach or specific therapies, especially with a wide range of follow-up duration. However, our study is prospective in nature with robust, comprehensive, and granular assessment of non-GFR-based kidney injury, including ABPM and urine biomarkers. Our study was single-center, but since participants were treated with standard Children’s Oncology Group protocols, results may be generalizable to other survivors of WT. Lastly, most of our participants had advanced stage WT, so our findings may not reflect the risk in patients with earlier stage WT. However, since all survivors of WT come to our survivorship clinic regardless of stage or risks for late effects, and we did not include those with known CKD or hypertension since those were the outcomes of interest, selection bias is mitigated.

In light of our findings, and consistent with the long-term follow-up guidelines from the Children’s Oncology Group,³⁴ we recommend annual outpatient visits with blood tests for kidney function and electrolytes, blood pressure check and urinalysis. Prior recipients of anthracyclines should be evaluated by echocardiogram dependent on anthracycline dosage received. In addition, however, we recommend considering screening with ABPM after completion of all therapy.

In summary, in a cohort of pediatric survivors of WT, we observed a high prevalence of abnormal results of kidney function, kidney injury, and blood pressure. Urinary biomarkers of IL-18 and KIM-1 were generally within normal limits, but urinary EGF was abnormal in many patients. Additional research and longer-term follow-up are needed to assess the overall hypertension and CKD risk facing survivors of WT.