An international panel of experts prepared an evidenced-based guideline for vaccination of immunocompromised adults and children. These guidelines are intended for use by primary care and ...subspecialty providers who care for immunocompromised patients. Evidence was often limited. Areas that warrant future investigation are highlighted.
Obesity is a worldwide epidemic in children and adolescents. Adult cohort studies have reported an association between higher body mass index (BMI) and increased leukemia-related mortality; whether a ...similar effect exists in childhood leukemia remains controversial.
We conducted a meta-analysis to determine whether a higher BMI at diagnosis of pediatric acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) is associated with worse event-free survival (EFS), overall survival (OS), and cumulative incidence of relapse (CIR).
We searched 4 electronic databases from inception through March 2015 without language restriction and included studies in pediatric ALL or AML (0-21 y of age) reporting BMI as a predictor of survival or relapse. Higher BMI, defined as obese (≥95%) or overweight/obese (≥85%), was compared with lower BMI nonoverweight/obese (<85%). Summary risk estimates for EFS, OS, and CIR (ALL only) were calculated with random- or fixed-effects models according to tests for between-study heterogeneity.
Of 4690 reports identified, 107 full-text articles were evaluated, with 2 additional articles identified via review of citations; 11 articles were eligible for inclusion in this meta-analysis. In ALL, we observed poorer EFS in children with a higher BMI (RR: 1.35; 95% CI: 1.20, 1.51) than in those at a lower BMI. A higher BMI was associated with significantly increased mortality (RR: 1.31; 95% CI: 1.09, 1.58) and a statistically nonsignificant trend toward greater risk of relapse (RR: 1.17; 95% CI: 0.99, 1.38) compared with a lower BMI. In AML, a higher BMI was significantly associated with poorer EFS and OS (RR: 1.36; 95% CI: 1.16, 1.60 and RR: 1.56; 95% CI: 1.32, 1.86, respectively) than was a lower BMI.
Higher BMI at diagnosis is associated with poorer survival in children with pediatric ALL or AML.
While low socioeconomic status (SES) has been associated with inferior cancer outcome among adults, its impact in pediatric oncology is unclear. Our objective was therefore to conduct a systematic ...review to determine the impact of SES upon outcome in children with cancer.
We searched Ovid Medline, EMBASE and CINAHL from inception to December 2012. Studies for which survival-related outcomes were reported by socioeconomic subgroups were eligible for inclusion. Two reviewers independently assessed articles and extracted data. Given anticipated heterogeneity, no quantitative meta-analyses were planned a priori.
Of 7,737 publications, 527 in ten languages met criteria for full review; 36 studies met final inclusion criteria. In low- and middle-income countries (LMIC), lower SES was uniformly associated with inferior survival, regardless of the measure chosen. The majority of associations were statistically significant. Of 52 associations between socioeconomic variables and outcome among high-income country (HIC) children, 38 (73.1%) found low SES to be associated with worse survival, 15 of which were statistically significant. Of the remaining 14 (no association or high SES associated with worse survival), only one was statistically significant. Both HIC studies examining the effect of insurance found uninsured status to be statistically associated with inferior survival.
Socioeconomic gradients in which low SES is associated with inferior childhood cancer survival are ubiquitous in LMIC and common in HIC. Future studies should elucidate mechanisms underlying these gradients, allowing the design of interventions mediating socioeconomic effects. Targeting the effect of low SES will allow for further improvements in childhood cancer survival.
New therapeutic strategies are needed for pediatric acute myeloid leukemia (AML) to reduce disease recurrence and treatment-related morbidity. The Children's Oncology Group Phase III AAML1031 trial ...tested whether the addition of bortezomib to standard chemotherapy improves survival in pediatric patients with newly diagnosed AML. AAML1031 randomized patients younger than 30 years of age with
AML to standard treatment with or without bortezomib. All patients received the identical chemotherapy backbone with either four intensive chemotherapy courses or three courses followed by allogeneic hematopoietic stem cell transplantation for high-risk patients. For those randomized to the intervention arm, bortezomib 1.3 mg/m
was given on days 1, 4 and 8 of each chemotherapy course. For those randomized to the control arm, bortezomib was not administered. In total, 1,097 patients were randomized to standard chemotherapy (n=542) or standard chemotherapy with bortezomib (n=555). There was no difference in remission induction rate between the bortezomib and control treatment arms (89%
91%,
=0.531). Bortezomib failed to improve 3-year event-free survival (44.8±4.5%
47.0±4.5%,
=0.236) or overall survival (63.6±4.5
67.2±4.3,
=0.356) compared with the control arm. However, bortezomib was associated with significantly more peripheral neuropathy (
=0.006) and intensive care unit admissions (
=0.025) during the first course. The addition of bortezomib to standard chemotherapy increased toxicity but did not improve survival. These data do not support the addition of bortezomib to standard chemotherapy in children with
AML. (Trial registered at
).
Numerous faces scales have been developed for the measurement of pain intensity in children. It remains unclear whether any one of the faces scales is better for a particular purpose with regard to ...validity, reliability, feasibility, and preference.
To summarize and systematically review faces pain scales most commonly used to obtain self-report of pain intensity in children for evaluation of reliability and validity and to compare the scales for preference and utility.
Five major electronic databases were systematically searched for studies that used a faces scale for the self-report measurement of pain intensity in children. Fourteen faces pain scales were identified, of which 4 have undergone extensive psychometric testing: Faces Pain Scale (FPS) (scored 0-6); Faces Pain Scale-Revised (FPS-R) (0-10); Oucher pain scale (0-10); and Wong-Baker Faces Pain Rating Scale (WBFPRS) (0-10). These 4 scales were included in the review. Studies were classified by using psychometric criteria, including construct validity, reliability, and responsiveness, that were established a priori.
From a total of 276 articles retrieved, 182 were screened for psychometric evaluation, and 127 were included. All 4 faces pain scales were found to be adequately supported by psychometric data. When given a choice between faces scales, children preferred the WBFPRS. Confounding of pain intensity with affect caused by use of smiling and crying anchor faces is a disadvantage of the WBFPRS.
For clinical use, we found no grounds to switch from 1 faces scale to another when 1 of the scales is in use. For research use, the FPS-R has been recommended on the basis of utility and psychometric features. Data are sparse for children below the age of 5 years, and future research should focus on simplified measures, instructions, and anchors for these younger children.
Objective was to determine whether prophylactic low level laser therapy (LLLT) reduces the risk of severe mucositis as compared to placebo or no therapy.
MEDLINE, EMBASE, and Cochrane Central ...Register of Controlled Trials were searched until February 2014 for randomized controlled trials (RCTs) comparing prophylactic LLLT with placebo or no therapy in patients with cancer or undergoing hematopoietic stem cell transplantation (HSCT). All analyses used random effects models.
Eighteen RCTs (1144 patients) were included. Prophylactic LLLT reduced the overall risk of severe mucositis (risk ratio (RR) 0.37, 95% confidence interval (CI) 0.20 to 0.67; P = 0.001). LLLT also reduced the following outcomes when compared to placebo/no therapy: severe mucositis at the time of anticipated maximal mucositis (RR 0.34, 95% CI 0.20 to 0.59), overall mean grade of mucositis (standardized mean difference -1.49, 95% CI -2.02 to -0.95), duration of severe mucositis (weighted mean difference -5.32, 95% CI -9.45 to -1.19) and incidence of severe pain (RR 0.26, 95% CI 0.18 to 0.37).
Prophylactic LLLT reduced severe mucositis and pain in patients with cancer and HSCT recipients. Future research should identify the optimal characteristics of LLLT and determine feasibility in the clinical setting.
Summary Background Sodium thiosulfate is an antioxidant shown in preclinical studies in animals to prevent cisplatin-induced hearing loss with timed administration after cisplatin without ...compromising the antitumour efficacy of cisplatin. The primary aim of this study was to assess sodium thiosulfate for prevention of cisplatin-induced hearing loss in children and adolescents. Methods ACCL0431 was a multicentre, randomised, open-label, phase 3 trial that enrolled participants at 38 participating Children's Oncology Group hospitals in the USA and Canada. Eligible participants aged 1–18 years with newly diagnosed cancer and normal audiometry were randomly assigned (1:1) to receive sodium thiosulfate or observation (control group) in addition to their planned cisplatin-containing chemotherapy regimen, using permuted blocks of four. Randomisation was initially stratified by age and duration of cisplatin infusion. Stratification by previous cranial irradiation was added later as a protocol amendment. The allocation sequence was computer-generated centrally and concealed to all personnel. Participants received sodium thiosulfate 16 g/m2 intravenously 6 h after each cisplatin dose or observation. The primary endpoint was incidence of hearing loss 4 weeks after final cisplatin dose. Hearing was measured using standard audiometry and reviewed centrally by audiologists masked to allocation using American Speech-Language-Hearing Association criteria but treatment was not masked for participants or clinicians. Analysis of the primary endpoint was by modified intention to treat, which included all randomly assigned patients irrespective of treatment received but restricted to those assessable for hearing loss. Enrolment is complete and this report represents the final analysis. This trial is registered with ClinicalTrials.gov , number NCT00716976. Findings Between June 23, 2008, and Sept 28, 2012, 125 eligible participants were randomly assigned to either sodium thiosulfate (n=61) or observation (n=64). Of these, 104 participants were assessable for the primary endpoint (sodium thiosulfate, n=49; control, n=55). Hearing loss was identified in 14 (28·6%; 95% CI 16·6–43·3) participants in the sodium thiosulfate group compared with 31 (56·4%; 42·3–69·7) in the control group (p=0·00022). Adjusted for stratification variables, the likelihood of hearing loss was significantly lower in the sodium thiosulfate group compared with the control group (odds ratio 0·31, 95% CI 0·13–0·73; p=0·0036). The most common grade 3–4 haematological adverse events reported, irrespective of attribution, were neutropenia (117 66% of 178 participant cycles in the sodium thiosulfate group vs 145 65% of 224 in the control group), whereas the most common non-haematological adverse event was hypokalaemia (25 17% of 149 vs 22 12% of 187). Of 194 serious adverse events reported in 26 participants who had received sodium thiosulfate, none were deemed probably or definitely related to sodium thiosulfate; the most common serious adverse event was decreased neutrophil count: 26 episodes in 14 participants. Interpretation Sodium thiosulfate protects against cisplatin-induced hearing loss in children and is not associated with serious adverse events attributed to its use. Further research is needed to define the appropriate role for sodium thiosulfate among emerging otoprotection strategies. Funding US National Cancer Institute.
Temporal dataset shift associated with changes in healthcare over time is a barrier to deploying machine learning-based clinical decision support systems. Algorithms that learn robust models by ...estimating invariant properties across time periods for domain generalization (DG) and unsupervised domain adaptation (UDA) might be suitable to proactively mitigate dataset shift. The objective was to characterize the impact of temporal dataset shift on clinical prediction models and benchmark DG and UDA algorithms on improving model robustness. In this cohort study, intensive care unit patients from the MIMIC-IV database were categorized by year groups (2008-2010, 2011-2013, 2014-2016 and 2017-2019). Tasks were predicting mortality, long length of stay, sepsis and invasive ventilation. Feedforward neural networks were used as prediction models. The baseline experiment trained models using empirical risk minimization (ERM) on 2008-2010 (ERM08-10) and evaluated them on subsequent year groups. DG experiment trained models using algorithms that estimated invariant properties using 2008-2016 and evaluated them on 2017-2019. UDA experiment leveraged unlabelled samples from 2017 to 2019 for unsupervised distribution matching. DG and UDA models were compared to ERM08-16 models trained using 2008-2016. Main performance measures were area-under-the-receiver-operating-characteristic curve (AUROC), area-under-the-precision-recall curve and absolute calibration error. Threshold-based metrics including false-positives and false-negatives were used to assess the clinical impact of temporal dataset shift and its mitigation strategies. In the baseline experiments, dataset shift was most evident for sepsis prediction (maximum AUROC drop, 0.090; 95% confidence interval (CI), 0.080-0.101). Considering a scenario of 100 consecutively admitted patients showed that ERM08-10 applied to 2017-2019 was associated with one additional false-negative among 11 patients with sepsis, when compared to the model applied to 2008-2010. When compared with ERM08-16, DG and UDA experiments failed to produce more robust models (range of AUROC difference, - 0.003 to 0.050). In conclusion, DG and UDA failed to produce more robust models compared to ERM in the setting of temporal dataset shift. Alternate approaches are required to preserve model performance over time in clinical medicine.
Diagnosis, treatment, response monitoring, and outcome of pediatric acute myeloid leukemia (AML) have made enormous progress during the past decades. Because AML is a rare type of childhood cancer, ...with an incidence of approximately seven occurrences per 1 million children annually, national and international collaborative efforts have evolved. This overview describes these efforts and includes a summary of the history and contributions of each of the main collaborative pediatric AML groups worldwide. The focus is on translational and clinical research, which includes past, current, and future clinical trials. Separate sections concern acute promyelocytic leukemia, myeloid leukemia of Down syndrome, and relapsed AML. A plethora of novel antileukemic agents that have emerged, including new classes of drugs, are summarized as well. Finally, an important aspect of the treatment of pediatric AML--supportive care--and late effects are discussed. The future is bright, with a wide range of emerging innovative therapies and with more and more international collaboration that ultimately aim to cure all children with AML, with fewer adverse effects and without late effects.
Temporal distribution shift negatively impacts the performance of clinical prediction models over time. Pretraining foundation models using self-supervised learning on electronic health records (EHR) ...may be effective in acquiring informative global patterns that can improve the robustness of task-specific models. The objective was to evaluate the utility of EHR foundation models in improving the in-distribution (ID) and out-of-distribution (OOD) performance of clinical prediction models. Transformer- and gated recurrent unit-based foundation models were pretrained on EHR of up to 1.8 M patients (382 M coded events) collected within pre-determined year groups (e.g., 2009-2012) and were subsequently used to construct patient representations for patients admitted to inpatient units. These representations were used to train logistic regression models to predict hospital mortality, long length of stay, 30-day readmission, and ICU admission. We compared our EHR foundation models with baseline logistic regression models learned on count-based representations (count-LR) in ID and OOD year groups. Performance was measured using area-under-the-receiver-operating-characteristic curve (AUROC), area-under-the-precision-recall curve, and absolute calibration error. Both transformer and recurrent-based foundation models generally showed better ID and OOD discrimination relative to count-LR and often exhibited less decay in tasks where there is observable degradation of discrimination performance (average AUROC decay of 3% for transformer-based foundation model vs. 7% for count-LR after 5-9 years). In addition, the performance and robustness of transformer-based foundation models continued to improve as pretraining set size increased. These results suggest that pretraining EHR foundation models at scale is a useful approach for developing clinical prediction models that perform well in the presence of temporal distribution shift.