Long-term outcomes and patterns of relapse in patients with bilateral Wilms tumor or bilaterally-predisposed unilateral Wilms tumor, a report from the COG AREN0534 study

Andrew J Murphy; Jack Brzezinski; Lindsay A Renfro; Brett Tornwall; Marcus M Malek; Daniel J Benedetti; Nicholas G Cost; Ethan A Smith; Jennifer Aldrink; Rodrigo LP Romao; Jeffrey S Dome; Andrew M Davidoff; Amy L Treece; Lauren N Parsons; Elizabeth A Mullen; Robert C Shamberger; Arnold C Paulino; Andrea C Lo; James I Geller; Peter F Ehrlich

doi:10.1002/ijc.35080

. Author manuscript; available in PMC: 2025 Nov 15.

Published in final edited form as: Int J Cancer. 2024 Jul 8;155(10):1824–1831. doi: 10.1002/ijc.35080

Long-term outcomes and patterns of relapse in patients with bilateral Wilms tumor or bilaterally-predisposed unilateral Wilms tumor, a report from the COG AREN0534 study

Andrew J Murphy ^1,^*, Jack Brzezinski ^2,^*, Lindsay A Renfro ³, Brett Tornwall ^3,⁴, Marcus M Malek ⁵, Daniel J Benedetti ⁶, Nicholas G Cost ⁷, Ethan A Smith ⁸, Jennifer Aldrink ⁹, Rodrigo LP Romao ², Jeffrey S Dome ¹⁰, Andrew M Davidoff ¹, Amy L Treece ¹¹, Lauren N Parsons ¹², Elizabeth A Mullen ¹³, Robert C Shamberger ¹³, Arnold C Paulino ¹⁴, Andrea C Lo ¹⁵, James I Geller ⁸, Peter F Ehrlich ¹⁶

^1.St Jude Children’s Research Hospital, Memphis, TN, USA

^2.Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada

^3.Division of Biostatistics, University of Southern California and Children’s Oncology Group, Los Angeles, CA, USA

^4.Glaukos Corp., San Clemente, California, USA.

^5.Children’s Hospital of Pittsburgh, University of Pittsburgh, Pittsburgh, PA, USA

^6.Monroe Carell Jr Children’s Hospital at Vanderbilt, Vanderbilt University Medical Center, Nashville, TN, USA

^7.The Surgical Oncology Program at the Children’s Hospital of Colorado, University of Colorado, Denver, CO, USA

^8.Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA

^9.Division of Pediatric Surgery, Department of Surgery, Nationwide Children’s Hospital, The Ohio State University College of Medicine, Columbus, OH, USA

^10.Children National Hospital, George Washington University School of Medicine and Health Sciences, Washington, DC, USA

^11.Ann &Robert H Lurie Children’s Hospital, Chicago, Illinois, USA

^12.Department of Pathology, Children’s Hospital of Wisconsin, Milwaukee, WI, USA.

^13.Boston Children’s Hospital and Dana Farber Cancer Institute, Boston, Mass, USA

^14.MD Anderson Cancer Center Houston Texas, USA

^15.University of British Columbia, Vancouver Canada

^16.Section of Pediatric Surgery CS Mott Children’s Hospital University of Michigan, Ann Arbor, Michigan, USA

Co first authors/equal contribution to work

Author Contributions: The work reported in the paper has been performed by the authors, unless clearly specified in the text. Each author contributed in the following ways to the manuscript:

AJM – conceptualization, visualization, writing – original draft, writing- review and editing

JB – conceptualization, visualization, writing – original draft, writing- review and editing

LAR – conceptualization, data curation, formal analysis, investigation, methodology, supervision, validation, visualization, writing- original draft, writing – review and editing

BT – data curation, formal analysis, investigation, validation, visualization, writing – review and editing.

MMM – conceptualization, visualization, writing- review and editing

DJB - conceptualization, visualization, writing- review and editing

NGC - conceptualization, visualization, writing- review and editing

EAS - conceptualization, visualization, writing- review and editing

JA - conceptualization, visualization, writing- review and editing

RLPR - conceptualization, visualization, writing- review and editing

JSD – conceptualization, funding acquisition, investigation, supervision, writing- review and editing

AMD - conceptualization, visualization, writing- review and editing

ALT - conceptualization, visualization, writing- review and editing

LNP - conceptualization, visualization, writing- review and editing

EAM - conceptualization, visualization, writing- review and editing

RCS - conceptualization, visualization, writing- review and editing

ACP - conceptualization, visualization, writing- review and editing

ACL - conceptualization, visualization, writing- review and editing

JIG - conceptualization, visualization, supervision, resources, funding acquisition, project administration, writing- review and editing

PFE – conceptualization, data curation, investigation, methodology, project administration, resources, supervision, writing – original draft, writing- review and editing

^✉

Corresponding Author: Andrew J. Murphy, MD, Department of Surgery, St. Jude Children’s Research Hospital, 262 Danny Thomas Place, Mail Stop 133, Memphis, Tennessee 38103, Andrew.murphy@stjude.org, X/Twitter: @ajacksonmurphy

PMCID: PMC11570340 NIHMSID: NIHMS2010785 PMID: 38973574

Abstract

The objective of this study is to report the long-term timing and patterns of relapse for children enrolled on Children’s Oncology Group (COG) AREN0534, a multicenter phase III clinical trial conducted from 2009–2015. Participants included children with bilateral Wilms tumor (BWT) or unilateral WT with genetic predisposition to develop BWT followed for up to 10 years. Smoothed hazard (risk) functions for event-free survival (EFS) were plotted so that timing of events could be visualized, both overall and within pre-specified groups. 222 children (190 BWT and 32 unilateral WT with BWT predisposition) were followed for a median of 8.6 years. Fifty events were reported of which 48 were relapse/progression. The overall 8-year EFS was 75% (95%CI: 69%−83%). The highest risk for an EFS event was immediately after diagnosis with a declining rate over two years. A second peak of events was observed around 4 years after diagnosis, and a small number of events were reported until the end of the follow-up period. In subset analyses, later increases in risk were more commonly observed in patients with female sex, anaplastic histology, negative lymph nodes or margins, and FHWT patients with post-chemotherapy intermediate risk. Among relapses that occurred after 2 years, most were to the kidney. These patterns suggest that late events may be second primary tumors occurring more commonly in females, although more investigation is required. Clinicians may consider observation of patients with BWT beyond 4 years from diagnosis.

Keywords: Bilateral Wilms tumor, predisposition, relapse, events, long-term outcomes

Graphical Abstract

graphic file with name nihms-2010785-f0010.jpg

INTRODUCTION

The Children’s Oncology Group (COG) Study AREN0534, “Treatment for Patients with Bilateral, Multicentric, or Bilaterally-Predisposed Unilateral Wilms Tumor” was a registered phase 3 clinical trial and the first prospective study designed specifically to improve outcomes in this unique population.¹ Prior to this study, outcomes for children with bilateral Wilms tumors (BWT) were sub-optimal compared to those with unilateral Wilms tumor (WT). For example, on the National Wilms Tumor Study 5 (NWTS-5) that ran from 1995 to 2002, 4-year Event Free Survival (EFS) for all children with BWT was 56% (95% CI: 44.8% −66.6%) compared to >85% in children with unilateral WT.² The AREN0534 trial represented a major advance with a reported 4-year EFS of 82.1% (95% CI: 73.5%−90.8%) and 4-year overall survival (OS) of 94.9% (95% CI: 90.1%−99.7%) for children with BWT; and a 4-year EFS of 94.0% (95% CI: 85.2% - 100%) and 4-year OS of 100% for children with multicentric tumors or unilateral Wilms tumor who were predisposed to bilateral tumor development.^{1, 3}

These standard-term oncologic outcomes are encouraging, but children with BWT still present distinct treatment challenges compared to those with unilateral tumors, including the potential for late relapse and, in the case of cancer predisposition syndromes, the potential for second primary tumors. Such secondary events, whether a relapse or a second primary tumor, depending on tumor location, present additional risks inherent in overall treatment burden, not limited to concerns for overall kidney parenchymal preservation. For unilateral tumors, most relapses occur “early” before 18 months; however, in reports from both the NWTSG and the International Society of Pediatric Oncology Renal Tumor Study Group (SIOP-RTSG), some relapses occurred later than 3 years in children with BWT. Longer term EFS/OS rates were also reported to be appreciably lower than for children with Stage I-IV unilateral favorable histology WT. Specifically, SIOP investigators observed a lower 10-year OS of 69% for children with synchronous BWT, noting several recurrences occurring more than 3 years after diagnosis. A limitation of these reports was that there was no defined treatment or protocol for children with BWT or multicentric/bilaterally predisposed unilateral WT and follow-up was inconsistent.⁴ These outcomes contrast with those published from the AREN0532 and AREN0533 studies, where most relapses occurred “early” within 18 months to 2 years of diagnosis and treatment initiation. Prior evidence described above suggested that these patterns may not hold in bilateral disease.^5–8

A goal of the AREN0534 study was to follow patients for up to 10 years after protocol therapy to document patterns and timing of recurrence in this well-characterized and consistently treated cohort of children with BWT or bilaterally-predisposed unilateral WT. Understanding the timing of EFS events and related factors could yield implications for treatment, follow-up duration, and patient/family counseling. Further, understanding if there are populations at risk for metachronous new primary renal tumors could also enable more opportunity for personalized medicine approaches aiming to identify lesions early, and limit the impact of total therapy, particularly regarding preserving healthy renal parenchyma. The purpose of this report is to describe how the risk of relapse and other EFS events changed over time among children who enrolled on the AREN0534 protocol, characterize any late relapses that occurred, and explore whether any patient or disease characteristics are associated with differences in the timing of relapse, second malignant neoplasm (SMN), or death.

MATERIALS AND METHODS

The AREN0534 trial was conducted from 2009 until 2015. 195 children with BWT and 54 with unilateral WT and predisposition to BWT were enrolled. Briefly, all patients were treated with neoadjuvant chemotherapy for a minimum of 6 and a maximum of 12 weeks after which surgery was mandated. Postoperative therapy was determined based on pathologic stage and post-chemotherapy histology.⁹ Definitions for high, intermediate, and low risk post-treatment histology were defined using criteria published as part of the SIOP WT 2001 trial.^{10, 11} Notably, the primary histologic classification in the AREN0534 trial was made according to the COG system in which tumors were characterized as favorable histology Wilms tumor (FHWT) or unfavorable (diffuse or focal anaplasia) histology. Following neoadjuvant chemotherapy, information from post-treatment tumor histology was also used to guide adjuvant therapy. Favorable histology post-treatment tumors were classified as low-risk, intermediate-risk, or blastemal subtype and thus information from the SIOP histology system was incorporated into the AREN0534 treatment algorithm. In AREN0534, patients with positive margins (whether viable tumor or necrotic) were assigned to receive radiotherapy. In this study, only one local stage was reported per patient. The reported local stage was the highest local stage of any kidney per patient. This is because the highest local stage drove the selection of systemic therapy for a given patient. Primary AREN0534 trial results have been published and detailed information about histology classification and chemotherapy regimens can be found in those published manuscripts.^{1, 3, 12}

Eligible AREN0534 patients included in the present analysis cohort were those with BWT or bilaterally-predisposed unilateral WT histologically confirmed by central pathology review to be WT (anaplastic or non-anaplastic); patients with other histologic diagnoses were excluded.

Descriptive statistics for demographics and patient and disease characteristics were presented as medians and interquartile ranges for continuous variables and counts and percentages for categorical variables. Smoothed hazard (risk) functions for EFS were plotted overall and by pre-specified patient subgroups using the method of Müller and Wang,¹³ including visual comparisons of EFS event timing by unilateral vs. bilateral disease, age (< 2 vs ≥2 years), sex, presence or absence of anaplasia, grouped stage (I-II vs III-IV), post-chemotherapy histologic risk classification (low/necrotic, intermediate, or high/blastemal subtype), and nodal and margin status for children with local stage 3 disease.

For children with EFS events, we additionally tabled the types of first event (relapse/progression, SMN, or death), locations of relapses, and timing of relapse and deaths, by time from study enrollment (< 2 years vs ≥2 years). Throughout, EFS was defined as the time from enrollment on AREN0534 to the earliest of relapse, progression, second malignancy, or death due to any cause; children not experiencing an event by the last known disease status were right-censored. All analyses were prospectively planned by the current study authorship before analyses were conducted.

RESULTS

AREN0534 enrolled 242 eligible children of any diagnosis. Excluding 20 children with diffuse hyperplastic perilobar nephroblastomatosis and other non-WT histologies, 222 children contributed to the present analyses whose descriptive statistics are shown in Table 1. The median age at diagnosis was 2.5 years; 60% of children were female, 86% had bilateral disease, 88% had FHWT, 86% of FHWT had post-chemotherapy intermediate risk histology. Fifty-four percent of the overall cohort had local stage III disease, and 17% had overall stage IV disease. Fifty-six percent of the included cohort had lymph node sampling at the time of first surgical procedure (biopsy or nephrectomy) with a positivity rate of 8.9%. Among children with local stage III disease and known margin status at nephrectomy, 29% had positive margins.

Table 1.

Descriptive Statistics

Characteristic	N = 222¹
Age (Years)	2.5 (1.2, 3.7)
Sex
Female	133 (60%)
Male	89 (40%)
Unilateral or Bilateral
Unilateral	32 (14%)
Bilateral	190 (86%)
Central Review Histology
DAWT	17 (7.7%)
FAWT	10 (4.5%)
FHWT	195 (88%)
Local Stage
Stage 1	72 (33%)
Stage 2	29 (13%)
Stage 3	118 (54%)
(Missing)	3
Maximum Stage
Stage I	70 (32%)
Stage II	27 (12%)
Stage III	87 (39%)
Stage IV	37 (17%)
(Missing)	1
Lymph Nodes Sampled	124 (56%)
Lymph Node Status
Negative	113 (91%)
Positive	11 (8.9%)
(Not Sampled)	98
Margin Status
Negative	123 (71%)
Positive	51 (29%)
(Missing)	48
FHWT Post-Chemo Risk
Low	7 (4.2%)
Intermediate	142 (86%)
Blastemal subtype	17 (10%)
(Missing or Not Applicable)	56

Open in a new tab

Median (IQR); n (%);

FHWT Post Chemotherapy Risk does not include any DAWT or FAWT; DAWT – diffuse anaplastic Wilms tumor, FAWT – focal anaplastic Wilms tumor, FHWT – favorable histology Wilms tumor

At the time of this analysis, median follow-up among children without an event was 8.6 years. Eight-year EFS for BWT (all histologies) was 75% (95% CI: 69%−83%). Fifty EFS events were reported, of which 48 were relapse/progression, one was a SMN, and one was a patient death. Accordingly, we will refer generally to patterns of EFS risk as “risk of relapse”.

In the overall cohort (Figure 1), a smoothed estimate of the hazard function showing risk of relapse over time (with 95% confidence bands) is superimposed over an estimated piecewise exponential distribution showing the annual risk of relapse up to 10 years after enrollment on AREN0534. The highest period of risk for EFS events is from diagnosis to two years (i.e. “early”), when 68% of events occurred. A second peak occurred between years three and four, risk then dropped further to become nearly zero between years 4–5 after enrollment. A third low, flat peak corresponding to increased risk of EFS events was present from approximately years 5 to 9 after enrollment, subsiding again for years 9–10 (where data also becomes sparse).

Risk of relapse over time is shown by pre-specified subgroups in Figures 2A–H. In Figure 2A, it appears that bilaterally-predisposed unilateral WT and bilateral patients follow rather different relapse patterns, with most bilateral patients relapsing early or during the year 3–4 peak and with very low risk subsequently. There were only three events among the bilaterally-predisposed unilateral WT patients; one occurred early before 2 years and two occurred late (around years 6–8). When examining patterns by age (Figure 2B), we find that younger (age < 2 years) and older children (age ≥2 years) follow similar patterns with most relapsing early, and no differences in late relapses by age are apparent. Patient sex appears to be associated with timing of relapse (Figure 2C); while the risk of relapse in males is highest in the first two years and decreases gradually over time to nearly zero by year 4 post-enrollment, female patients comprise the subsequent peaks in risk around years 3–4 and 6–9.

Histology also appears to play a role in timing of relapse (Figure 2D). Overall, most children with either FHWT or tumors with diffuse anaplasia who relapse tend to do so early before two years. No patients with diffuse anaplasia relapsed late but 2 patients with focal anaplasia had a relapse at year 3 and year 4, respectively. In contrast to patients with anaplastic tumors, a slight late risk peak remains for patients with FHWT around year 7. Overall stage of disease (Figure 2E) does not appear to be a major factor in timing of relapse, though our investigation shows that risk is higher for higher-staged patients at early timepoints. Timing of relapse in FHWT by post-chemotherapy risk classification is shown in Figure 2F. While no FHWT patients with post-chemotherapy low-risk histology had EFS events and thus could not be plotted, we find that FHWT patients with intermediate-risk histology (the majority group) follow a pattern of high initial risk followed by a late peak around 6–8 years, while patients with FHWT blastemal subtype histology who relapse do so exclusively between years 1–4 and have a substantially higher risk of relapse in general. Among patients with local stage III disease, those with positive lymph nodes who relapse do so by 3 years, while those with negative nodes follow a bimodal pattern with some early risk (years 1–2) and some late risk (years 5–7) post-enrollment, shown in Figure 2G. In Figure 2H, a similar pattern is observed by surgical margin status; those with positive margins who relapse tend to do so early, with risk becoming near zero after 3 years post-enrollment, while those with negative margins are just as likely to relapse late as relapse early, with a second risk peak occurring late from years 3–10.

Table 2 shows anatomic locations and timing of relapse (early versus late). Most relapses overall (65% combined early and late) occur locally in the kidney or tumor bed. Of those that occur before two years, 59% occurred in the kidney or tumor bed. In addition, almost 80% of late events occurred in the kidney. Abdominal, liver, and lung relapses were always within 2 years, while relapses to the chest wall/pleura and multiple sites at once (annotated below Table 2) more commonly occurred later than 2 years.

Table 2.

Types of EFS Events and Location/Timing of Relapses - All Patients

	<2 Years (N=34)	2+ Years (N=16)	Total (N=50)
Timing by Type of First EFS Event
Relapse/Progression	34 (100%)	14 (88%)	48 (96%)
SMN	0 (0%)	1 (6%)	1 (2%)
Death	0 (0%)	1 (6%)	1 (2%)
Relapse/Progression Sites
Abdomen (Excluding Liver)	2 (6%)	0 (0%)	2 (4%)
Kidney	17 (50%)	11 (79%)	28 (58%)
Liver	1 (3%)	0 (0%)	1 (2%)
Lung	8 (24%)	0 (0%)	8 (17%)
Multiple Sites^*	1 (3%)	2 (14%)	3 (6%)
Pelvis	2 (6%)	0 (0%)	2 (4%)
Thorax	0 (0%)	1 (7%)	1 (2%)
Tumor Bed	3 (9%)	0 (0%)	3 (6%)
Missing/Unknown	0	2	2

Open in a new tab

EFS – event-free survival; SMN – subsequent malignant neoplasm;

Patients with multiple sites of relapse include one patient with abdomen (not including liver), right kidney, and retroperitoneal nodule; one patient with abdomen (not including liver), left kidney, liver, and lung; and one patient with left kidney, liver, and retroperitoneal lymph nodes.

DISCUSSION

In this report we described the longitudinal timing of tumor relapse/recurrence in children enrolled on the AREN0534 trial who had BWT or bilaterally-predisposed unilateral WT. We found that more than 25% of first recurrences occur after 2 years from diagnosis. These late events tend to occur at year three or four as well as a smaller number of events after 6 years. We also noted that the risk of late recurrences was higher for females and in children with negative margins or negative lymph nodes. In this long-term follow up study, the 8-year EFS was 75% (95% CI: 69%−83%) which is still lower than for children with non-syndromic unilateral tumors although improved compared to historical trials.

The locations of relapse between those with early (< 2 years) and late (≥ 2 years) events suggest different driving mechanisms within each group. Those with early events were more likely to have distant relapses – 41% of these children relapsed in the lung, liver, or abdomen. This is in line with outcomes in contemporaneous COG trials for children with non-syndromic unilateral WT, where risk of metastatic recurrence was more common than locoregional recurrence (76.5% of recurrences were metastatic in AREN0532).^{5, 6, 14} On the other hand, in this study, 79% of recurrences after 2 years occurred in remaining kidney tissue. Although it is a possibility that these local recurrences are related to residual viable tumor tissue from the primary tumor, the inverse association of positive margins with late relapses is evidence against this hypothesis. It is not possible to evaluate the role of radiotherapy in preventing late recurrence in patients with positive margins as all patients on this study with positive margins were assigned to receive radiation to the margin and flank. It is reasonable to presume that radiotherapy would have a similar role in the reduction of early and late recurrences and therefore does not account for this discrepancy. However, this hypothesis cannot be directly answered with these data. An alternative hypothesis would suggest that non-malignant kidney tissue in children with BWT is primed to develop WT even after initial therapy and that therefore many late events observed on this trial represent new primary tumors (metachronous bilateral disease in a cancer predisposed child) rather than relapses of the initial tumor.

The hypothesis that many late events are new primary tumors rather than relapses of the initial tumor is supported by the factors associated with these events in this study. Clinical features that have classically been associated with tumor recurrence such as positive margins, positive regional lymph nodes, blastemal-predominant histology, and hematogenous metastases were not associated with late events on this trial.^15–17 The only factor that was strongly associated with late events other than BWT itself was female sex, suggesting that constitutional features rather than tumor-related features are the primary drivers of this phenomenon.

In this study, female biological sex was associated with a later pattern of relapse than male biological sex. We made the a priori decision to analyze data according to biological sex because several prior lines of evidence indicated that female children may have a unique tumor biology pertaining to BWT.^{18, 19} First, epidemiologic data have shown that the female-to-male ratio in BWT is 1.6:1 whereas unilateral WT cases are evenly distributed.^{20, 21} In keeping with these epidemiologic data, 143/242 (59%) of children with BWT enrolled on AREN0534 were female. While the biological basis of this female predominance is not yet completely understood, it may involve an association between BWT predisposition due to constitutional epigenetic 11p15.5 H19/ICR1 gain of methylation (also known as loss of imprinting) and female biological sex. Fiala et. al. demonstrated a pattern of female children with BWT having epigenetic somatic mosaicism for 11p15.5 H19/ICR1 gain of methylation detectable in the bloodstream. In fact, 7 of 7 children with gain of methylation at H19/ICR1 in this study were female.²² Furthermore, a recent study of 24 children from a Dutch 5-year, nationwide prospective cohort demonstrated that 57% of females with WT were found to have a genetic or epigenetic predisposition, which was notably higher than the proportion of predisposition detected in males from the cohort.¹⁸

Our data suggest that children treated for BWT are at risk for late events even when optimal and rigorous therapeutic approaches are used. Such late events are often intra-renal, suggesting a second primary tumor. Understanding such risks could potentially enable a more tailored approach to renal sparing surgery early on, as cumulative chemotherapy toxicity, and surgical toxicity (nephron loss), are relevant for such children. Although these data suggest that late events are related to constitutional risk, the specific molecular drivers of that risk are yet to be elucidated. Nevertheless, clinicians treating children with BWT should consider continuing frequent tumor surveillance beyond 2 years from diagnosis. A long-term future direction of this work may be consideration of maintenance targeted therapies in children with BWT or BWT predisposition identified to be at elevated risk of late relapse. A reasonable approach may be to offer surveillance on a schedule like those used for children with WT predisposition syndromes such as Beckwith-Wiedemann syndrome – namely every 3 months until 7 or 8 years of age. Given the absence of late pulmonary events on this trial it may be sufficient to offer abdominal ultrasonography without chest x-ray for surveillance after 2 years from diagnosis.²³

There are challenges to this study. First, detailed reporting of syndromic features in children with BWT that could be suggestive of a constitutional risk could have been improved. Second, capturing of clinically relevant WT biomarkers such as combined LOH of 1p/16q or 1q gain was not routinely performed during the time of this study. Examining outcomes according to these tumor-related or constitutional biomarkers will be the subject of future investigation. The rate of lymph node sampling reported in this study is low due to this information being collected from the first surgery form only, for both nephrectomies and biopsies, where lymph node sampling may not be expected for biopsies. On future COG trials, sampling of at least one lymph node will be mandatory for trial enrollment/participation, and lymph node sampling will be collected for all complete and partial nephrectomies regardless of timing. Reporting on the laterality of relapse and how it related to the type of surgery performed (nephron-sparing surgery versus radical nephroureterectomy) was incomplete. Lastly, this is a rare disease and large numbers would be more beneficial. This is the largest prospective cohort of patients with BWT or bilaterally-predisposed unilateral WT; however, increased power may be achievable by future international collaboration.

In conclusion, children with BWT have an increased risk of events after 2 years that may represent metachronous bilateral tumors. This risk appears to be higher in girls. Future directions should focus on understanding the risk factors associated with late relapse and the role of mosaicism and molecular risk markers at 11p15.5 and WT1. Further advances in understanding the genetic and biological landscape of BWT and WT predisposition hold promise to improve personalized care delivery, with the goals of optimizing survival and limiting cumulative medical and surgical toxicity.

Novelty and Impact:

In this analysis of long-term outcomes from a phase III multicenter Children’s Oncology Group clinical trial, risk of an EFS event was elevated at later timepoints for patients with female sex, anaplastic histology, negative lymph nodes or margins, and in FHWT patients with post-chemotherapy intermediate risk. Most relapses after 2 years occurred in the kidney rather than at distant sites. Due to late patterns of relapse observed in this cohort, clinicians may consider surveillance of patients with BWT beyond 4 years from diagnosis.

whats-new.

Children with bilateral Wilms tumor (BWT) pose a treatment challenge compared with those who have unilateral tumors, including the potential for late relapse or a second primary tumor. Here, the authors report on the patterns of relapse in children with BWT observed in the Children’s Oncology Group AREN0534 study. Girls had a higher risk of relapse than boys, they found. Other features that correlated with increased relapse risk included anaplastic histology and negative lymph nodes or margins. Distant relapses, such as in lung, liver, or abdomen, tended to occur early, while relapses after 2 years were usually in kidney tissue.

Funding Statement:

This project was supported by grants St. Baldrick’s Foundation, NCTN Operations Center Grant U10CA180886, and NCTN Statistics & Data Center Grant U10CA180899.

Abbreviations:

BWT: bilateral Wilms tumor
CI: confidence interval
COG: Children’s Oncology Group
EFS: event-free survival
FHWT: favorable histology Wilms tumor
IRB: institutional review board
NAWT: Non-anaplastic Wilms tumor
NWTS: National Wilms Tumor Study
OS: overall survival
SIOP: International Society of Pediatric Oncology
SIOP-RTSG: International Society of Pediatric Oncology Renal Tumor Study Group
SMN: second malignant neoplasm
WT: Wilms tumor

Footnotes

Ethics Statement: The AREN0534 trial was conducted from 2009 until 2015. Enrollment on trial required local IRB approval at each participating center. Written informed consent was obtained from each participant’s parent(s) or legal guardian(s). The Children’s Oncology Group AREN0534 trial is registered on Clinicaltrials.gov ID: NCT00945009

Previous Presentation: Results from this study were presented in part at the International Society of Pediatric Oncology/International Pediatric Surgical Oncology Association Meeting in Ottawa, Canada in October 2023.

Conflict of Interest Disclosure: Brett Tornwall reports full-time employment by Glaukos Corp at the time of submission; however, affiliation at the time of the work was Children’s Oncology Group Statistics and Data Center. The other authors have no conflicts of interest to declare.

Other Disclosures/Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data Availability Statement:

Primary patient enrollment and outcomes data are publicly available on clinicaltrial.gov at: https://clinicaltrials.gov/study/NCT00945009?tab=results#results-overview

Further information is available from the corresponding author upon request.

REFERENCES

1.Ehrlich P, Chi YY, Chintagumpala MM, Hoffer FA, Perlman EJ, Kalapurakal JA, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, Paulino AC, et al. Results of the First Prospective Multi-institutional Treatment Study in Children With Bilateral Wilms Tumor (AREN0534): A Report From the Children’s Oncology Group. Ann Surg 2017;266: 470–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Grundy P, Breslow N, Norkool P, al e, National Wilms Tumor Study Five Statistical Report, 2002.
3.Ehrlich PF, Chi YY, Chintagumpala MM, Hoffer FA, Perlman EJ, Kalapurakal JA, Tornwall B, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, et al. Results of Treatment for Patients With Multicentric or Bilaterally Predisposed Unilateral Wilms Tumor (AREN0534): A report from the Children’s Oncology Group. Cancer 2020;126: 3516–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Coppes MJ, de Kraker J, van Dijken PJ. Bilateral Wilms’ tumor: long-term survival and some epidemiological features. J Clin Oncol 1989;7: 310–5. [DOI] [PubMed] [Google Scholar]
5.Fernandez CV, Mullen EA, Chi YY, Ehrlich PF, Perlman EJ, Kalapurakal JA, Khanna G, Paulino AC, Hamilton TE, Gow KW, Tochner Z, Hoffer FA, et al. Outcome and Prognostic Factors in Stage III Favorable-Histology Wilms Tumor: A Report From the Children’s Oncology Group Study AREN0532. J Clin Oncol 2018;36: 254–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Fernandez CV, Perlman EJ, Mulle nEA, Ehrlich PF, Dome JS, Geller JI, Shamberger RC. Clinical outcome and biological predictors of relapse following nephrectomy only for very low-risk Wilms tumor (VLRWT): A report from Children’s Oncology Group AREN0532. Ann Surg In Press. Ann Surg 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Dix DB, Fernandez CV, Chi YY, al e. Augmentation of therapy for favorable-histology Wilms Tumor with combined loss of heterozygosity of chromosomes 1p and 16q: A report from the Children’s Oncology Group studies AREN0532 and AREN0533. J Clin Oncol 2015;33: 1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Dix DB, Gratias EJ, Seibel NL, Anderson JR, Mullen E, Geller JI, Khanna G, Kalapurakal JA, Perlman EJ, Ehrlich PF, Malogolowkin MH, Grundy PE, et al. Treatment of stage IV favorable histology Wilms tumor with incomplete lung metastasis response after chemotherapy: A report from Children’s Oncology Group study AREN0533. J Clin Oncol 2014;32: 10001. [Google Scholar]
9.Ehrlich PF, Dome JS, Shamberger RC, Ritchey M, Chintagumpala MM, Warwick A, Rosen N. Treatment for Patients with Bilateral, Multicentric, or Bilaterally-Predisposed Unilateral Wilms 2006.
10.Vujanić GM, Sandstedt B, Harms D, Kelsey A, Leuschner I, de Kraker J, Committee SNS. Revised International Society of Paediatric Oncology (SIOP) working classification of renal tumors of childhood. Med Pediatr Oncol 2002;38: 79–82. [DOI] [PubMed] [Google Scholar]
11.Graf N, van Tinteren H, Bergeron C, Pein F, van den Heuvel-Eibrink MM, Sandstedt B, Schenk JP, Godzinski J, Oldenburger F, Furtwangler R, de Kraker J. Characteristics and outcome of stage II and III non-anaplastic Wilms’ tumour treated according to the SIOP trial and study 93–01. Eur J Cancer 2012;48: 3240–8. [DOI] [PubMed] [Google Scholar]
12.Ehrlich PF, Tornwall B, Chintagumpala MM, Chi YY, Hoffer FA, Perlman EJ, Kalapurakal JA, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, et al. Kidney Preservation and Wilms Tumor Development in Children with Diffuse Hyperplastic Perilobar Nephroblastomatosis: A Report from the Children’s Oncology Group Study AREN0534. Ann Surg Oncol 2022;29: 3252–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Muller HG, Wang JL. Hazard rate estimation under random censoring with varying kernels and bandwidths. Biometrics 1994;50: 61–76. [PubMed] [Google Scholar]
14.Dome JS, Mullen EA, Dix DB, Gratias EJ, Ehrlich PF, Daw NC, Geller JI, Chintagumpala M, Khanna G, Kalapurakal JA, Renfro LA, Perlman EJ, et al. Impact of the First Generation of Children’s Oncology Group Clinical Trials on Clinical Practice for Wilms Tumor. J Natl Compr Canc Netw 2021;19: 978–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Malogolowkin MH, Cotton CA, Green DM, al e. Treatment of Wilms tumor relapsing after initial treatment with vincristine, actinomycin D, and doxorubicin. A report from the National Wilms Tumor Study Group. Pediatr Blood Cancer 2008;50: 236–41. [DOI] [PubMed] [Google Scholar]
16.DM G, CA C, M M, al e. Treatment of Wilms tumor relapsing after inital treatment with vincristine snd actinomycin D: a report form the National Wilms Tumor Study. Pediatr Blood and Cancer 2007;48: 493–9. [DOI] [PubMed] [Google Scholar]
17.Malogolowkin M, Cotton CA, Green DM, Breslow NE, Perlman E, Miser J, Ritchey ML, Thomas PR, Grundy PE, D’Angio GJ, Beckwith JB, Shamberger RC, et al. Treatment of Wilms tumor relapsing after initial treatment with vincristine, actinomycin D, and doxorubicin. A report from the National Wilms Tumor Study Group. Pediatr Blood Cancer 2008;50: 236–41. [DOI] [PubMed] [Google Scholar]
18.Stiller CA, Olshan AF. Epidemiology of Renal Tumours of Childhood. In: Pritchard-Jones K, Dome JS. Renal Tumors of Childhood: Biology and Therapyed. Berlin, Heidelberg: Springer Berlin Heidelberg, 2014: 1–17. [Google Scholar]
19.Stoltze UK, Hildonen M, Hansen TVO, Foss-Skiftesvik J, Byrjalsen A, Lundsgaard M, Pignata L, Gronskov K, Tumer AZ, Schmiegelow K, Brok JS, Wadt KAW. Germline (epi)genetics reveals high predisposition in females: a 5-year, nationwide, prospective Wilms tumour cohort. J Med Genet 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Stiller CA, Olshan A. Epidemiology of renal tumours of childhood. ed. Berlin: Springer- Verlag, 2014. [Google Scholar]
21.Charlton J, Irtan S, Bergeron C, Pritchard-Jones K. Bilateral Wilms tumour: a review of clinical and molecular features. Expert Rev Mol Med 2017;19: e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Fiala EM, Ortiz MV, Kennedy JA, Glodzik D, Fleischut MH, Duffy KA, Hathaway ER, Heaton T, Gerstle JT, Steinherz P, Shukla N, McNeer N, et al. 11p15.5 epimutations in children with Wilms tumor and hepatoblastoma detected in peripheral blood. Cancer 2020;126: 3114–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Mullen EA, Chi YY, Hibbitts E, Anderson JR, Steacy KJ, Geller JI, Green DM, Khanna G, Malogolowkin MH, Grundy PE, Fernandez CV, Dome JS. Impact of Surveillance Imaging Modality on Survival After Recurrence in Patients With Favorable-Histology Wilms Tumor: A Report From the Children’s Oncology Group. J Clin Oncol 2018: JCO1800076. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Primary patient enrollment and outcomes data are publicly available on clinicaltrial.gov at: https://clinicaltrials.gov/study/NCT00945009?tab=results#results-overview

Further information is available from the corresponding author upon request.

[R1] 1.Ehrlich P, Chi YY, Chintagumpala MM, Hoffer FA, Perlman EJ, Kalapurakal JA, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, Paulino AC, et al. Results of the First Prospective Multi-institutional Treatment Study in Children With Bilateral Wilms Tumor (AREN0534): A Report From the Children’s Oncology Group. Ann Surg 2017;266: 470–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Grundy P, Breslow N, Norkool P, al e, National Wilms Tumor Study Five Statistical Report, 2002.

[R3] 3.Ehrlich PF, Chi YY, Chintagumpala MM, Hoffer FA, Perlman EJ, Kalapurakal JA, Tornwall B, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, et al. Results of Treatment for Patients With Multicentric or Bilaterally Predisposed Unilateral Wilms Tumor (AREN0534): A report from the Children’s Oncology Group. Cancer 2020;126: 3516–25. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Coppes MJ, de Kraker J, van Dijken PJ. Bilateral Wilms’ tumor: long-term survival and some epidemiological features. J Clin Oncol 1989;7: 310–5. [DOI] [PubMed] [Google Scholar]

[R5] 5.Fernandez CV, Mullen EA, Chi YY, Ehrlich PF, Perlman EJ, Kalapurakal JA, Khanna G, Paulino AC, Hamilton TE, Gow KW, Tochner Z, Hoffer FA, et al. Outcome and Prognostic Factors in Stage III Favorable-Histology Wilms Tumor: A Report From the Children’s Oncology Group Study AREN0532. J Clin Oncol 2018;36: 254–61. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Fernandez CV, Perlman EJ, Mulle nEA, Ehrlich PF, Dome JS, Geller JI, Shamberger RC. Clinical outcome and biological predictors of relapse following nephrectomy only for very low-risk Wilms tumor (VLRWT): A report from Children’s Oncology Group AREN0532. Ann Surg In Press. Ann Surg 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Dix DB, Fernandez CV, Chi YY, al e. Augmentation of therapy for favorable-histology Wilms Tumor with combined loss of heterozygosity of chromosomes 1p and 16q: A report from the Children’s Oncology Group studies AREN0532 and AREN0533. J Clin Oncol 2015;33: 1009. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Dix DB, Gratias EJ, Seibel NL, Anderson JR, Mullen E, Geller JI, Khanna G, Kalapurakal JA, Perlman EJ, Ehrlich PF, Malogolowkin MH, Grundy PE, et al. Treatment of stage IV favorable histology Wilms tumor with incomplete lung metastasis response after chemotherapy: A report from Children’s Oncology Group study AREN0533. J Clin Oncol 2014;32: 10001. [Google Scholar]

[R9] 9.Ehrlich PF, Dome JS, Shamberger RC, Ritchey M, Chintagumpala MM, Warwick A, Rosen N. Treatment for Patients with Bilateral, Multicentric, or Bilaterally-Predisposed Unilateral Wilms 2006.

[R10] 10.Vujanić GM, Sandstedt B, Harms D, Kelsey A, Leuschner I, de Kraker J, Committee SNS. Revised International Society of Paediatric Oncology (SIOP) working classification of renal tumors of childhood. Med Pediatr Oncol 2002;38: 79–82. [DOI] [PubMed] [Google Scholar]

[R11] 11.Graf N, van Tinteren H, Bergeron C, Pein F, van den Heuvel-Eibrink MM, Sandstedt B, Schenk JP, Godzinski J, Oldenburger F, Furtwangler R, de Kraker J. Characteristics and outcome of stage II and III non-anaplastic Wilms’ tumour treated according to the SIOP trial and study 93–01. Eur J Cancer 2012;48: 3240–8. [DOI] [PubMed] [Google Scholar]

[R12] 12.Ehrlich PF, Tornwall B, Chintagumpala MM, Chi YY, Hoffer FA, Perlman EJ, Kalapurakal JA, Warwick A, Shamberger RC, Khanna G, Hamilton TE, Gow KW, et al. Kidney Preservation and Wilms Tumor Development in Children with Diffuse Hyperplastic Perilobar Nephroblastomatosis: A Report from the Children’s Oncology Group Study AREN0534. Ann Surg Oncol 2022;29: 3252–61. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Muller HG, Wang JL. Hazard rate estimation under random censoring with varying kernels and bandwidths. Biometrics 1994;50: 61–76. [PubMed] [Google Scholar]

[R14] 14.Dome JS, Mullen EA, Dix DB, Gratias EJ, Ehrlich PF, Daw NC, Geller JI, Chintagumpala M, Khanna G, Kalapurakal JA, Renfro LA, Perlman EJ, et al. Impact of the First Generation of Children’s Oncology Group Clinical Trials on Clinical Practice for Wilms Tumor. J Natl Compr Canc Netw 2021;19: 978–85. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Malogolowkin MH, Cotton CA, Green DM, al e. Treatment of Wilms tumor relapsing after initial treatment with vincristine, actinomycin D, and doxorubicin. A report from the National Wilms Tumor Study Group. Pediatr Blood Cancer 2008;50: 236–41. [DOI] [PubMed] [Google Scholar]

[R16] 16.DM G, CA C, M M, al e. Treatment of Wilms tumor relapsing after inital treatment with vincristine snd actinomycin D: a report form the National Wilms Tumor Study. Pediatr Blood and Cancer 2007;48: 493–9. [DOI] [PubMed] [Google Scholar]

[R17] 17.Malogolowkin M, Cotton CA, Green DM, Breslow NE, Perlman E, Miser J, Ritchey ML, Thomas PR, Grundy PE, D’Angio GJ, Beckwith JB, Shamberger RC, et al. Treatment of Wilms tumor relapsing after initial treatment with vincristine, actinomycin D, and doxorubicin. A report from the National Wilms Tumor Study Group. Pediatr Blood Cancer 2008;50: 236–41. [DOI] [PubMed] [Google Scholar]

[R18] 18.Stiller CA, Olshan AF. Epidemiology of Renal Tumours of Childhood. In: Pritchard-Jones K, Dome JS. Renal Tumors of Childhood: Biology and Therapyed. Berlin, Heidelberg: Springer Berlin Heidelberg, 2014: 1–17. [Google Scholar]

[R19] 19.Stoltze UK, Hildonen M, Hansen TVO, Foss-Skiftesvik J, Byrjalsen A, Lundsgaard M, Pignata L, Gronskov K, Tumer AZ, Schmiegelow K, Brok JS, Wadt KAW. Germline (epi)genetics reveals high predisposition in females: a 5-year, nationwide, prospective Wilms tumour cohort. J Med Genet 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Stiller CA, Olshan A. Epidemiology of renal tumours of childhood. ed. Berlin: Springer- Verlag, 2014. [Google Scholar]

[R21] 21.Charlton J, Irtan S, Bergeron C, Pritchard-Jones K. Bilateral Wilms tumour: a review of clinical and molecular features. Expert Rev Mol Med 2017;19: e8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Fiala EM, Ortiz MV, Kennedy JA, Glodzik D, Fleischut MH, Duffy KA, Hathaway ER, Heaton T, Gerstle JT, Steinherz P, Shukla N, McNeer N, et al. 11p15.5 epimutations in children with Wilms tumor and hepatoblastoma detected in peripheral blood. Cancer 2020;126: 3114–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Mullen EA, Chi YY, Hibbitts E, Anderson JR, Steacy KJ, Geller JI, Green DM, Khanna G, Malogolowkin MH, Grundy PE, Fernandez CV, Dome JS. Impact of Surveillance Imaging Modality on Survival After Recurrence in Patients With Favorable-Histology Wilms Tumor: A Report From the Children’s Oncology Group. J Clin Oncol 2018: JCO1800076. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Long-term outcomes and patterns of relapse in patients with bilateral Wilms tumor or bilaterally-predisposed unilateral Wilms tumor, a report from the COG AREN0534 study

Andrew J Murphy, MD

Jack Brzezinski, MD

Lindsay A Renfro, PhD

Brett Tornwall

Marcus M Malek, MD

Daniel J Benedetti, MD

Nicholas G Cost, MD

Ethan A Smith, MD

Jennifer Aldrink, MD

Rodrigo LP Romao, MD

Jeffrey S Dome, MD

Andrew M Davidoff, MD

Amy L Treece, MD

Lauren N Parsons, MD

Elizabeth A Mullen, MD

Robert C Shamberger, MD

Arnold C Paulino, MD

Andrea C Lo, MD

James I Geller, MD

Peter F Ehrlich, MD

Abstract

Graphical Abstract

INTRODUCTION

MATERIALS AND METHODS

RESULTS

Table 1.

Figure 1.

Figure 2.

Table 2.

DISCUSSION

Novelty and Impact:

whats-new.

Funding Statement:

Abbreviations:

Footnotes

Data Availability Statement:

REFERENCES

Associated Data

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases