Search

Pediatric cancer treatment, especially for brain tumors, can have profound and complicated late effects. With the survival rates increasing because of improved detection and treatment, a more comprehensive understanding of the impact of current treatments on neurocognitive function and brain structure is critically needed. A frontline medulloblastoma clinical trial (SJMB03) has collected data, including treatment, clinical, neuroimaging, and cognitive variables. Advanced methods for modeling and integrating these data are critically needed to understand the mediation pathway from the treatment through brain structure to neurocognitive outcomes. We propose an integrative Bayesian mediation analysis approach to model jointly a treatment exposure, a high‐dimensional structural neuroimaging mediator, and a neurocognitive outcome and to uncover the mediation pathway. The high‐dimensional imaging‐related coefficients are modeled via a binary Ising–Gaussian Markov random field prior (BI‐GMRF), addressing the sparsity, spatial dependency, and smoothness and increasing the power to detect brain regions with mediation effects. Numerical simulations demonstrate the estimation accuracy, power, and robustness. For the SJMB03 study, the BI‐GMRF method has identified white matter microstructure that is damaged by cancer‐directed treatment and impacts late neurocognitive outcomes. The results provide guidance on improving treatment planning to minimize long‐term cognitive sequela for pediatric brain tumor patients. [ABSTRACT FROM AUTHOR]

The development of normal tissue radiation dose-response models for children with cancer has been challenged by many factors, including small sample sizes; the long length of follow-up needed to observe some toxicities; the continuing occurrence of events beyond the time of assessment; the often complex relationship between age at treatment, normal tissue developmental dynamics, and age at assessment; and the need to use retrospective dosimetry. Meta-analyses of published pediatric outcome studies face additional obstacles of incomplete reporting of critical dosimetric, clinical, and statistical information. This report describes general methods used to address some of the pediatric modeling issues. It highlights previous single- and multi-institutional pediatric dose-response studies and summarizes how each PENTEC taskforce addressed the challenges and limitations of the reviewed publications in constructing, when possible, organ-specific dose-effect models. [ABSTRACT FROM AUTHOR]

Purpose: Clinical trials data from National Cancer Institute (NCI)-funded cooperative oncology group trials could be enhanced by merging with external data sources. Merging without direct patient identifiers would provide additional patient privacy protections. We sought to develop and validate a matching algorithm that uses only indirect patient identifiers. Methods: We merged the data from two Phase III Children’s Oncology Group (COG) trials for de novo acute myeloid leukemia (AML) with the Pediatric Health Information Systems (PHIS). We developed a stepwise matching algorithm that used indirect identifiers including treatment site, gender, birth year, birth month, enrollment year and enrollment month. Results from the stepwise algorithm were compared against the direct merge method that used date of birth, treatment site, and gender. The indirect merge algorithm was developed on AAML0531 and validated on AAML1031. Results: Of 415 patients enrolled on the AAML0531 trial at PHIS centers, we successfully matched 378 (91.1%) patients using the indirect stepwise algorithm. Comparison to the direct merge result suggested that 362 (95.7%) matches identified by the indirect merge algorithm were concordant with the direct merge result. When validating the indirect stepwise algorithm using the AAML1031 trial, we successfully matched 157 out of 165 patients (95.2%) and 150 (95.5%) of the indirectly merged matches were concordant with the directly merged matches. Conclusions: These data demonstrate that patients enrolled on COG clinical trials can be successfully merged with PHIS administrative data using a stepwise algorithm based on indirect patient identifiers. The merged data sets can be used as a platform for comparative effectiveness and cost effectiveness studies. [ABSTRACT FROM AUTHOR]

Increased understanding of the underlying mechanisms of cognitive remediation is needed to facilitate development of intervention strategies for childhood cancer survivors experiencing cognitive late effects. Accordingly, a pilot functional magnetic resonance imaging (fMRI) study was conducted with 14 cancer survivors (12.02 ± 0.09 years old), who participated in a cognitive remediation clinical trial, and 28 healthy children (12.7 ± 0.6 years old). The ventral visual areas, cerebellum, supplementary motor area, and left inferior frontal cortex were significantly activated in the healthy participants during a continuous performance task. In survivors, brain activation in these regions was diminished at baseline, and increased upon completion of remediation and at a 6-month follow-up. The fMRI activation index for each region of interest was inversely associated with the Conners' Clinical Competence Index (p<.01). The pilot study suggests that fMRI is useful in evaluating neural responses to cognitive remediation. [ABSTRACT FROM PUBLISHER]

Introduction: Childhood cancer survivors (CCS) are at risk for radiotherapy (RT) late effects, including second malignancies. Optimal screening for differentiated thyroid cancer (DTC) in CCS post-RT remains controversial. We assessed the outcome of thyroid ultrasound (US) surveillance in CCS exposed to RT. Methods: 306 CCS were surveilled with thyroid US between 2002–2021. Surveillance was dependent on age at the time of primary diagnosis, interval from receipt of RT, and individual provider. Thyroid US, clinicopathologic features, and outcomes were described. Cutpoints of CCS RT age associated with varying risk of nodule presentation were explored. The selected cutpoints were used to define age categories, which were then used to compare thyroid nodule-related outcomes. Risk factors for thyroid nodule(s) were evaluated using multivariate logistic regression (odds ratio [OR] [95% confidence interval]). Results: The most common CCS diagnoses were leukemia (32%), CNS tumor (26%), and neuroblastoma (18%). Patients received TBI (45%) and/or RT to craniospinal (44%), chest (11%), and neck regions (6%). About 49% (n = 150) of patients had thyroid nodule(s). Forty-four patients underwent surgery, and 28 had DTC: 19 with American Thyroid Association (ATA) low-risk classification, 2 with ATA intermediate-risk, and 7 with ATA high-risk disease. Age cutpoint analyses identified cutpoints 3 and 10; hence, ≤3, >3 to ≤10, and >10 years were used. Of the 9 patients with intermediate- or high-risk disease, 8 were ≤10 years and 1 was >10 years at the time of RT. Female sex (OR = 1.62 [1.13–2.12] p = 0.054) and greater interval between RT and first US (OR = 1.10 [1.04–1.16] p = 0.001) were independent risk factors for nodule presentation. Conclusions: Thyroid US surveillance may be beneficial for CCS exposed to RT at younger ages (≤10 years) for earlier detection of DTC, prior to developing advanced metastatic disease. [ABSTRACT FROM AUTHOR]

Purpose: To accurately map the retinal area covered by tumor in a prospectively enrolled cohort of children diagnosed with retinoblastoma. Methods: Orbital MRI in 106 consecutive retinoblastoma patients (44 bilateral) was analyzed. For MRI-visible tumors, the polar angle and angle of eccentricity of points defining tumor perimeter on the retina were determined by triangulation from images in three orthogonal planes. The centroid of the mapped area was calculated to approximate tumor origin, and the location and cumulative tumor burden were analyzed in relation to mutation type (germline vs. somatic), tumor area, and patient age at diagnosis. Location of small tumors undetected by MRI was approximated with fundoscopic images. Results: Mapping was successful for 129 tumors in 91 eyes from 67 patients (39 bilateral, 43 germline mutation). Cumulative tumor burden was highest within the macula and posterior pole and was asymmetrically higher within the inferonasal periphery. Tumor incidence was lowest in the superotemporal periphery. Tumor location varied with age at diagnosis in a complex pattern. Tumor location was concentrated in the macula and superonasal periphery in patients <5.6 months, in the inferotemporal quadrant of the posterior pole in patients 5.6-8.8 months, in the inferonasal quadrant in patients 8.8-13.2 months, and in the nasal and superotemporal periphery in patients >13.2 months. The distribution of MRI-invisible tumors was consistent with the asymmetry of mapped tumors. Conclusions: MRI-based mapping revealed a previously unrecognized pattern of retinoblastoma localization that evolves with age at diagnosis. The structured spatiotemporal distribution of tumors may provide valuable clues about cellular or molecular events associated with tumorigenesis in the developing retina. [ABSTRACT FROM AUTHOR]

Background Infections are a leading cause of morbidity and mortality in children with cancer. Although data are limited, the impact of infection in this population appears to be amplified in low- and middle-income countries. Defining the epidemiology of infection in a specific region is paramount to developing effective interventions. This study aims to define the epidemiology of and outcomes from infection in children with leukemia in the Dominican Republic. Methods A retrospective cohort was assembled of children newly diagnosed with leukemia between July 1, 2015 and June 30, 2017 at Hospital Infantil Dr. Robert Reid Cabral in Santo Domingo. Patients were identified from the Pediatric Oncology Network Database and hospital admissions from the Oncology admissions logbook. Medical records and microbiology results were reviewed to identify all inpatient invasive infections. Distance from a child's home to the hospital was determined using ArcGIS 10.5.1. Infection rates were described in discrete time periods after diagnosis and risk factors for invasive infection were explored using Poisson regression. Results The cohort included 68 patients; 54 (79.4%) with acute lymphoblastic leukemia and 14 (20.6%) with acute myeloblastic leukemia. The cohort was 48.5% female, had a median age at diagnosis of 7.3 years (range 1.1–16.6), and a median weight-for-age of 45.5 percentile (range 0–99.9). There were 1.2 invasive infections per 100 days at risk in the first 60 days after diagnosis, 0.8 from > 60–100 days, and 0.4 from > 100–180 days. Gastroenteritis, skin/soft-tissue infection, and pneumonia were most frequent, with bacteremia most common in the first 60 days. In a multivariate Poisson regression model, age ≥10 years (IRR 0.49, 95% CI 0.28–0.85) and distance from the hospital >100 km (IRR 0.34, 95% CI 0.14 – 0.81) were each protective against infection in the first 180 days after diagnosis. During the 2-year period, 8 of 22 (36%) patient deaths were related to infection. Conclusion Invasive infections were common and a prominent source of death in this cohort. Interventions aimed at reducing infection should target the first 60 days after diagnosis. Decreased infection incidence among children of older age and farther distance from the hospital were unexpected and warrant further investigation. Disclosures All authors: No reported disclosures. [ABSTRACT FROM AUTHOR]

This article discusses regression analysis of mixed interval-censored failure time data. Such data frequently occur across a variety of settings, including clinical trials, epidemiologic investigations, and many other biomedical studies with a follow-up component. For example, mixed failure times are commonly found in the two largest studies of long-term survivorship after childhood cancer, the datasets that motivated this work. However, most existing methods for failure time data consider only right-censored or only interval-censored failure times, not the more general case where times may be mixed. Additionally, among regression models developed for mixed interval-censored failure times, the proportional hazards formulation is generally assumed. It is well-known that the proportional hazards model may be inappropriate in certain situations, and alternatives are needed to analyze mixed failure time data in such cases. To fill this need, we develop a maximum likelihood estimation procedure for the proportional odds regression model with mixed interval-censored data. We show that the resulting estimators are consistent and asymptotically Gaussian. An extensive simulation study is performed to assess the finite-sample properties of the method, and this investigation indicates that the proposed method works well for many practical situations. We then apply our approach to examine the impact of age at cranial radiation therapy on risk of growth hormone deficiency in long-term survivors of childhood cancer. [ABSTRACT FROM AUTHOR]

Equal access to clinical trial enrollment is important to ensure that findings are generalizable to the broader population. This study aimed to evaluate disparities in enrollment on pediatric oncology clinical trials. We assessed the relationship between patient characteristics and enrollment on COG trial AAML1031 in a cohort of pediatric patients with AML in the Pediatric Health Information System. The associations of enrollment with outcomes were evaluated. Non-Hispanic Black patients, infants, and patients from zip codes with a lower proportion of poverty were less likely to enroll (30% vs. 61%, p =.004; 34% vs. 58%, p =.003; 46% vs. 58%, p =.02). On-therapy mortality was similar among enrolled and nonenrolled patients (7.3% vs. 8.9%, p =.47). Differences in early mortality were more pronounced among nonenrolled patients compared to enrolled patients (3.0% vs. 0.5%, p =.03). Understanding the etiology of these disparities will inform strategies to ensure balanced access to clinical trials across patient populations. [ABSTRACT FROM AUTHOR]

Purpose: Imaging descriptions of posterior fossa ependymoma in children have focused on magnetic resonance imaging (MRI) signal and local anatomic relationships with imaging location only recently used to classify these neoplasms. We developed a quantitative method for analyzing the location of ependymoma in the posterior fossa, tested its effectiveness in distinguishing groups of tumors, and examined potential associations of distinct tumor groups with treatment and prognostic factors. Methods: Pre-operative MRI examinations of the brain for 38 children with histopathologically proven posterior fossa ependymoma were analyzed. Tumor margin contours and anatomic landmarks were manually marked and used to calculate the centroid of each tumor. Landmarks were used to calculate a transformation to align, scale, and rotate each patient's image coordinates to a common coordinate space. Hierarchical cluster analysis of the location and morphological variables was performed to detect multivariate patterns in tumor characteristics. The ependymomas were also characterized as 'central' or 'lateral' based on published radiological criteria. Therapeutic details and demographic, recurrence, and survival information were obtained from medical records and analyzed with the tumor location and morphology to identify prognostic tumor characteristics. Results: Cluster analysis yielded two distinct tumor groups based on centroid location The cluster groups were associated with differences in PFS ( p = .044), 'central' vs. 'lateral' radiological designation ( p = .035), and marginally associated with multiple operative interventions ( p = .064). Conclusions: Posterior fossa ependymoma can be objectively classified based on quantitative analysis of tumor location, and these classifications are associated with prognostic and treatment factors. [ABSTRACT FROM AUTHOR]