Highlights
– In a prespecified analysis of AREN0533 (n=251 evaluable), pulmonary tumor burden (number and size of lung nodules) had limited prognostic value for children with favorable‑histology Wilms tumor (FHWT) when patients were treated per contemporary, response‑adapted protocols.
– Rapid complete responders (RCRs) to initial DD‑4A chemotherapy who were spared lung radiation did not show worse outcomes based on number of metastases; larger nodule size was associated with worse EFS in univariable analysis but not after adjusting for 1q gain.
– In multivariable models that included tumor 1q gain, 1q gain was a statistically significant independent predictor of inferior event‑free survival (EFS, P = .0015) and overall survival (OS, P = .039), while nodule number and size were not.
Background
Wilms tumor is the most common renal malignancy of childhood. Most patients with favorable‑histology Wilms tumor (FHWT) are highly curable, but the presence of pulmonary metastases at diagnosis has historically signaled higher relapse risk and altered therapeutic approach. Current pediatric oncology practice aims to balance optimizing cure rates against acute and late toxicities (notably from thoracic radiation and intensified chemotherapy). Identifying which children with lung‑only metastases genuinely require intensified therapy or thoracic radiation remains an important clinical question.
The Children’s Oncology Group (COG) study AREN0533 used an early response‑adapted strategy: children received two cycles of vincristine/dactinomycin/doxorubicin (Regimen DD‑4A) and underwent chest CT reassessment. Those achieving radiographic rapid complete response (RCR) continued DD‑4A without lung radiation, whereas children with slow incomplete response (SIR) received augmented therapy (Regimen M: DD‑4A plus cyclophosphamide/etoposide) and thoracic radiation. Because pulmonary tumor burden (nodule number, size) is often used intuitively to risk‑stratify, the AREN0533 investigators prospectively evaluated whether nodule number or size added prognostic information beyond response and tumor biology (specifically 1q gain).
Study Design
This analysis included children with FHWT and pulmonary‑only metastases enrolled on COG AREN0533. Lung nodule response was assessed by chest CT after two cycles of Regimen DD‑4A and categorized as either rapid complete response (RCR) or slow incomplete response (SIR). Key features:
– Population: 288 children with stage IV FHWT and pulmonary‑only disease were enrolled; 251 met inclusion criteria for the outcomes analyses reported.
– Intervention and response‑adapted therapy: RCR patients continued DD‑4A without lung radiation therapy (RT). SIR patients were treated with augmented chemotherapy (Regimen M, adding cyclophosphamide/etoposide) plus thoracic RT.
– Pulmonary tumor burden variables: number of lung metastases and maximum nodule size on baseline chest CT.
– Endpoints: Event‑free survival (EFS) and overall survival (OS). Comparisons were performed within the two treatment cohorts (RCR and SIR). Multivariable Cox proportional hazards models for EFS and OS stratified by treatment assessed the impact of number and size of lung metastases, adjusted for tumor 1q gain.
Key Findings
Cohort sizes and primary observations
– Of the 288 enrolled with stage IV pulmonary‑only disease, 251 were included in the outcome analyses.
– RCR cohort: n = 105. These patients achieved complete radiographic pulmonary response after two cycles of DD‑4A and were treated without thoracic radiation.
– SIR cohort: n = 146. These patients had persistent pulmonary disease after two cycles and were treated with augmented chemotherapy plus thoracic radiation.
Univariable outcomes
– In the RCR cohort (n = 105): EFS and OS were not significantly different when patients were grouped by number of lung metastases. However, nodule size was significantly associated with EFS (P = .022) in univariable analysis—larger nodules correlated with worse EFS—but not with OS.
– In the SIR cohort (n = 146): neither number nor size of pulmonary metastases was associated with differences in EFS or OS.
Multivariable analysis including 1q gain
– When number and size of lung metastases were evaluated together with tumor 1q gain in multivariable Cox models (stratified by assigned treatment), neither nodule number nor size remained significantly associated with EFS or OS.
– By contrast, tumor 1q gain remained a significant independent adverse prognostic factor for both EFS (P = .0015) and OS (P = .039) after adjustment for pulmonary tumor burden and treatment assignment.
Interpretation of effect sizes and clinical impact
– The analysis suggests that radiographic pulmonary tumor burden (as measured by number and maximum size of nodules on baseline CT) provides limited independent prognostic information when early chemosensitivity (RCR vs SIR) and tumor 1q status are known.
– Practically, children with RCR to two cycles of DD‑4A can safely avoid thoracic RT regardless of the number of initial nodules, aligning response‑adapted strategies with attempts to reduce late RT‑related morbidity.
– The data underscore the prognostic importance of molecular markers — specifically 1q gain — in risk stratification of metastatic FHWT.
Expert Commentary and Context
Biological plausibility and prior evidence
– Gain of chromosome 1q has emerged over the past decade as a reproducible adverse prognostic marker in Wilms tumor in multiple cohorts; it likely reflects clonal genomic events associated with more aggressive tumor biology. The AREN0533 analysis reinforces this biological signal within a homogeneous, treatment‑defined population of children with pulmonary‑only metastases.
Clinical implications
– The findings support the broader movement in pediatric oncology towards integrating tumor genomics with early treatment response to refine therapy intensity. Specifically, response‑adapted omission of thoracic RT for RCR patients appears justified and safe across a range of pulmonary tumor burdens when evaluated in the AREN0533 framework.
– Because 1q gain was independently associated with worse outcomes, molecular profiling at diagnosis could inform risk‑adapted intensification strategies or enrollment in trials testing novel agents for patients with 1q gain, even among those who have favorable early radiographic responses.
Limitations and generalizability
– Imaging limitations: Baseline chest CT assessment can vary with image acquisition parameters, radiologist interpretation, and artifact; nodule counts and maximal diameter measurements have intrinsic variability. Although AREN0533 applied structured response assessment after two cycles, residual measurement error may attenuate the prognostic signal of pulmonary burden.
– Central review and measurement standardization: The robustness of nodule quantification depends on whether measurements were centrally reviewed or locally measured; heterogeneity may reduce the power to detect true associations.
– Sample size and event rates: While 251 evaluable patients is substantial for a pediatric metastatic subgroup, event rates (relapses, deaths) remain low overall, which limits power to detect modest effect sizes and interactions.
– Applicability outside the AREN0533 regimen: The conclusions directly apply to children managed with the AREN0533 response‑adapted approach and may not generalize to different frontline regimens or to patients with extrapulmonary metastatic sites.
Opportunities for future work
– Prospective integration of molecular profiling (including 1q gain) into upfront risk stratification algorithms could allow trials to test whether patients with 1q gain benefit from treatment intensification despite radiographic RCR.
– Refinement of imaging biomarkers (e.g., volumetric nodule assessment, functional imaging) could be explored to determine whether more sophisticated measures of burden have independent prognostic value.
– Correlative studies to identify the driver genes and pathways on 1q that confer aggressive behavior may open routes for targeted therapy.
Conclusion
In the AREN0533 cohort of children with stage IV FHWT and pulmonary‑only metastases, baseline pulmonary tumor burden measured by number and size of lung nodules had limited independent prognostic value when patients were treated with a standardized, response‑adapted regimen. Tumor chromosome 1q gain was a stronger and independent predictor of inferior event‑free and overall survival. These findings support integrating tumor genomics (1q status) with early radiographic response to tailor therapy intensity and minimize late toxicities for children with metastatic FHWT.
Funding and clinicaltrials.gov
The study was conducted by the Children’s Oncology Group Renal Tumor Committee and published in the Journal of Clinical Oncology (Dix et al., 2025). Specific grant and sponsor information are reported in the original manuscript. The trial designation is AREN0533 (see the published article for registry identifiers and funding details).
References
– Dix DB, Khanna G, Renfro LA, Tfirn IC, Smith EA, Artunduaga M, Eklund MJ, Sandberg JK, Parsons LN, Kalapurakal JA, Ehrlich PF, Aldrink JH, Glick RD, Benedetti DJ, Fernandez CV, Dome JS, Mullen EA, Geller JI; COG Renal Tumor Committee. Impact of Pulmonary Tumor Burden in Favorable Histology Wilms Tumor Outcomes: A Report From the Children’s Oncology Group Study AREN0533. J Clin Oncol. 2025 Nov 12: JCO2500532. doi: 10.1200/JCO-25-00532. Epub ahead of print. PMID: 41223336; PMCID: PMC12646512.
(Consult the full AREN0533 publication for additional methodological details, exact effect estimates, and full funding/registry statements.)
Author note
This article synthesizes and places in clinical context the AREN0533 findings for clinicians and researchers caring for children with Wilms tumor. Readers should consult the original manuscript for granular data, statistical models, and supplemental analyses.
