Clinical Practice Guideline-Inconsistent Management of Fertility Preservation in Pediatric Cancer Patients in Community Settings: A Children’s Oncology Group Study

Abstract

Background:

The primary objective was to measure adherence to clinical practice guideline (CPG) recommendations for fertility preservation (FP) in pediatric cancer patients treated in National Cancer Institute Community Oncology Research Program (NCORP) sites. Secondary objectives were to describe factors such as site size associated with CPG-inconsistent care delivery and cryopreservation completion.

Methods:

This retrospective, multicenter study included patients 15 to 21 years old with a first cancer diagnosis from January 2014 through December 2015 who were previously enrolled to a Children’s Oncology Group (COG) study and received care at a participating NCORP site. Patients were randomly selected from a list generated by the COG for chart review by participating sites. Primary outcome was care delivery that was inconsistent with a strong CPG recommendation on FP, namely discussion and offering of FP options before cancer treatment initiation, as adjudicated centrally by a panel.

Results:

A total of 129 patients from 25 sites were included. Among these, 48% (62/129) received CPG-inconsistent care. Most CPG-inconsistent care was due to lack of FP discussion documentation (93.5%, 58/62). Small site size, treatment at a pediatric (vs mixed adult/pediatric) site, and female sex were associated with higher odds of CPG-inconsistent care delivery.

Conclusions:

Newly diagnosed pediatric cancer patients often received CPG-inconsistent care for FP, with disproportionate gaps noted for females, and those treated at smaller or pediatric NCORP sites. The primary reason for CPG-inconsistent care is lack of FP discussion from clinicians. Opportunities to improve FP CPG implementation are highlighted.

Introduction

Infertility following cancer treatment is one of the most common late effects experienced by childhood cancer survivors and has a significant impact on quality of life for both male and female patients and their families.¹ Effective fertility preservation (FP) interventions (sperm/oocyte/embryo cryopreservation) are available for pediatric cancer patients and are more effective if completed before receiving gonadotoxic therapies. Thus, longstanding clinical practice guidelines (CPGs) strongly recommend that care teams discuss fertility risks with pediatric cancer patients and offer FP interventions before starting cancer treatment.^2–5 Guidance was first published in 2005 by the American Society of Reproductive Medicine,⁵ with formal guidelines the following year by the American Society of Clinical Oncology.⁴ Since that time, FP guidelines have been published and/or endorsed by numerous organizations worldwide.^6–9 However, significant gaps still remain in discussing or offering FP.^10,11

CPGs are statements that include evidence-based recommendations to improve patient care for a specific clinical condition. They are developed through a systematic review of evidence and include an analysis of benefits and harms of alternative options.¹² CPGs thus serve to translate the best available, published evidence into routine practice and their use improves patient outcomes.^13–15 To promote widespread uptake of CPGs in pediatric cancer care, the Children’s Oncology Group’s (COG) CPG Task Force systematically identifies published supportive care CPGs, endorses those that meet predefined criteria for quality and relevance, disseminates endorsed CPGs widely within the membership, and posts endorsed CPGs on a publicly available website.¹⁶ The Task Force first endorsed a CPG pertaining to FP in pediatric oncology patients in December 2014.¹⁷

However, COG-endorsement of CPGs does not ensure CPG-consistent care delivery, and uptake of a COG-endorsed CPG on FP is the central interest of this study. There is no published information documenting whether pediatric cancer institutions are delivering CPG-consistent care to their patients by discussing FP and offering sperm/oocyte/embryo cryopreservation (thereafter termed cryopreservation) before the start of therapy. Our primary objective was to measure the proportion of patients who received care that was inconsistent with a strong recommendation of the COG-endorsed CPG on FP. Secondary objectives were to identify factors, including site size, associated with CPG-inconsistent care delivery and noncompletion of cryopreservation.

Methods

This retrospective, multicenter study was developed by the COG and conducted as part of the NCI Community Oncology Research Program (NCORP) cancer care delivery research (CCDR) scientific area. The NCORP is a network of community-based or minority/underserved (defined as serving populations consisting of at least 30% racial/ethnic minorities or rural residents) sites in the United States.¹⁸

This study formed one component of ACCL15N1CD, the COG’s first CCDR study, whose overarching objective was to evaluate CPG-inconsistent supportive care delivery within pediatric oncology NCORP institutions. All 37 COG-member NCORP institutions were invited to participate. The study was approved by the NCI Pediatric Central Institutional Review Board (IRB) and IRBs at participating sites. The need for informed consent or assent was waived given the retrospective nature of the study.

Patients

Patients were considered for study inclusion if they had a first cancer diagnosis from January 1, 2014 through December 31, 2015, were 15 to less than 21 years of age at diagnosis and were enrolled to a COG study (therapeutic or nontherapeutic). Patient sociodemographic and clinical data were obtained from the COG database. The COG created a list of patients meeting these criteria and then randomly selected patients for local health record review to identify episodes of FP. The number of randomly selected patients for chart review per participating site was proportional to the total number of COG enrollments at each participating site during the study period.

Study personnel at participating sites reviewed the list of randomly selected patients to confirm their eligibility for study inclusion. In addition to the eligibility criteria noted above, evaluable patients were required to have received inpatient or outpatient care, including chemotherapy, radiotherapy, or both at the participating institution during the study period (January 1, 2014 through December 31, 2015).

Definitions

CPG-inconsistent FP care was defined as occurring when FP was not discussed with patients or with their families before treatment initiation, as evinced by lack of documentation in the health record, or, when FP was discussed, if cryopreservation was not offered. For patients where FP discussion was not documented, but who completed cryopreservation, FP discussion was assumed to have occurred and CPG-consistent care was deemed to have been delivered.

Site size was defined using a proxy measure of the number of COG therapeutic enrollments within the study period.

FP episode adjudication

The time frame for FP review was limited to the first four weeks after the first cancer diagnosis. Site personnel uploaded specific, deidentified sections (initial hospital history and physical, or clinic new patient consultation; inpatient progress notes; consent conference notes; consult notes; clinic notes; medication administration record; discharge summaries written by any health care professional, including nurses and social workers) of each included patient’s health record into the study database. This information was used to identify FP discussion and referral. One study team member (AV, AS, or NS) reviewed these documents and created a patient summary using a standardized template. A second study team member (LLD) checked each episode summary against the uploaded documents. The adjudication panel (AG, AJS, and LS) reviewed the patient summaries at virtual video conference calls and adjudicated the care delivered as either CPG-consistent or CPG-inconsistent. A proportion of all episode summaries (5%) were randomly selected for readjudication. If any discrepancy between adjudication decisions was identified, the adjudication panel discussed the patient again to finalize the adjudication.

Statistical analyses

Our primary endpoint was the proportion of patients receiving CPG-inconsistent care related to FP. The primary hypothesis was that the true proportion of patients receiving CPG-inconsistent care was greater than 0.50. We tested the primary hypothesis using a modified one-sided, one-sample z-test of proportion with a significance level of 0.025. The study was powered under the alternative hypothesis that the true proportion of CPG-inconsistent care was 0.70 and that the correlation within institutions was 0.2, a fairly conservative estimate of the correlation when considering binary outcomes. Under these assumptions, 100 FP episodes would result in 90% power to conclude that the true proportion was larger than 0.50. Standard error for the hypothesis test was estimated using a hierarchical bootstrap approach, using 5,000 bootstrap samples, to account for correlation between episodes treated at the same institution.¹⁹

Generalized linear mixed-effects models were used to estimate the association between site size and receipt of CPG-inconsistent care and completion of cryopreservation as outcomes. Models were adjusted for the following a priori selected variables: pediatric site versus mixed adult/pediatric, minority/underserved site versus other, Area Deprivation Index (ADI; as a proxy for household socioeconomic disadvantage),²⁰ and patient sex. A random intercept was included to account for correlation between patients treated at the same institution. Among patients who received CPG-inconsistent care who had no documentation of cryopreservation completion, cryopreservation was assumed not to have been completed. Among patients who received CPG-consistent FP care who had no documentation of cryopreservation completion, the outcome was determined to be missing. Complete case analyses were used for all models. Since we suspected that these associations may differ for males and females, we also conducted exploratory analyses stratified by sex, when possible.

Briefly, a modified version of the ADI of Brokamp, et al. was used to assess socioeconomic disadvantage.²⁰ The ADI is based on the 2015 five-year American Community Survey and was developed for each census tract using principal components analysis and included the following measures: the fraction of households with income below poverty level, the median household income in 2015 inflation-adjusted dollars, the fraction of the population 25 years of age and older with at least a high school graduation or General Education Development equivalency, the fraction of population with no health insurance coverage, the fraction of households receiving public assistance income, food stamps, or Supplemental Nutritional Assistance Program, and the fraction of vacant houses. The ADI for each patient’s residential zip code tabulation area was calculated using the mean of all intersecting census tracts, and higher ADI indicates higher levels of disadvantage.²⁰ Because Puerto Rican sites were included in this study and the original ADI calculations excluded Puerto Rico, we estimated a new ADI using the approach of Brokamp, et al. that included Puerto Rico.²⁰ All analyses were conducted using R version 4.2.2.

Results

Among 37 COG-member NCORP sites, 26 elected to participate and 25 contributed evaluable data. Figure 1 depicts patients assessed for eligibility and delineates those identified for inclusion, and then adjudicated as CPG-consistent or -inconsistent.

FIG. 1.

Diagram of episode identification and adjudications.

Characteristics of the 129 included patients and contributing sites are described in Tables 1 and 2, respectively. The median patient age was 16.8 years (interquartile range: 15.1, 17.7). Most patients were male; one-third were Hispanic. The most common diagnoses overall were acute lymphoblastic leukemia and Hodgkin lymphoma, followed by Ewing sarcoma and osteosarcoma.

Table 1.

Patient and Treatment Characteristics

Characteristic	All patients (n = 129)	Males (n = 95)	Females (n = 34)
Patient characteristics
Median age in years (IQR)	16.80 (15.07, 17.67)	16.67 (15.72, 17.62)	17.06 (15.46, 18.23)
Cancer diagnosis (n; %)
Acute lymphoblastic leukemia	34 (26.4)	29 (30.5)	5 (14.7)
Acute myelogenous leukemia	12 (9.3)	7 (7.4)	5 (14.7)
Ewing sarcoma	14 (10.9)	10 (10.5)	4 (11.8)
Hodgkin Lymphoma	25 (19.4)	15 (15.8)	10 (29.4)
Non-Hodgkin Lymphoma	7 (5.4)	6 (6.3)	1 (2.9)
Osteosarcoma	12 (9.3)	10 (10.5)	2 (5.9)
Othera	25 (19.4)	18 (19.0)	7 (20.6)
Race/Ethnicity (n; %)
Hispanic/Latino	43 (33.3)	29 (30.5)	14 (41.2)
Non-Hispanic American Indian/Alaska Native	2 (1.9)	2 (2.1)	0 (0.0)
Non-Hispanic Asian	4 (3.1)	4 (4.2)	0 (0.0)
Non-Hispanic Black	9 (7.0)	6 (6.3)	3 (8.8)
Non-Hispanic Native Hawaiian/Other Pacific Islander	1 (0.8)	1 (1.1)	0 (0.0)
Non-Hispanic White	67 (51.9)	50 (52.6)	17 (50.0)
Unknown/not reported	3 (2.3)	3 (3.2)	0 (0.0)
Median Income by Zip Code; U.S.$
20th percentile	38,005.0	38,245.33	31,708.78
40th percentile	49,458.00	51,673.67	42,183.89
60th percentile	59,814.00	61,596.80	55,625.00
80th percentile	75,666.8	76,208.22	71,879.80
100th percentile	138,419.20	138,419.20	98,199.82
Median proportion with assisted income (IQR)	0.13 (0.07, 0.22)	0.12 (0.07, 0.20)	0.20 (0.09, 0.24)
Median proportion with a high school education (IQR)	0.87 (0.79, 0.93)	0.88 (0.81, 0.93)	0.85 (0.76, 0.91)
Median proportion with no health insurance (IQR)	0.13 (0.08, 0.18)	0.12 (0.07, 0.18)	0.15 (0.10, 0.21)
Median proportion in poverty (IQR)	0.15 (0.08, 0.24	0.13 (0.08, 0.23)	0.18 (0.11, 0.25)
Median proportion in vacant housing (IQR)	0.10 (0.07, 0.15)	0.10 (0.07, 0.14)	0.11 (0.06, 0.19)
Median Area Deprivation Index (IQR)	0.38 (0.29, 0.49)	0.37 (0.29, 0.46)	0.43 (0.32, 0.51)
Treatment characteristics
Planned treatment protocol contains alkylating agents; n (%)	98 (76.0)	74 (77.9)	24 (70.6)
Planned protocol contains cisplatin (males only); n (%)	NA	13 (13.7)	NA
Planned treatment protocol contains radiation with potential inclusion of the gonads; n (%)	8 (6.2)	6 (6.3)	2 (5.9)

^a Central nervous system tumors, rhabdosarcoma, other leukemia.

IQR, interquartile range; n, number; NA, not applicable.

Table 2.

Characteristics of 25 Sites That Contributed Data

Characteristic	n = 25
Type (n; %)
Minority/Underserved Community (vs. Other)	11 (44.0)
Pediatric (vs. Mixed Adult/Pediatric)	16 (64.0)
Private (vs. Academic)	18 (72.0)
Location (n; %)
Western U.S.	3 (12.0)
Southwestern U.S.	5 (20.0)
Northeastern U.S.	1 (4.0)
Southeastern U.S.	10 (40.0)
Midwestern U.S.	5 (20.0)
Puerto Rico	1 (4.0)
Size
Median number of patients registered on COG treatment studies (no banking, biology, or registry) during study period (IQR)	38 (26, 48)

n, number; IQR, interquartile range.

Participating sites were located across the United States and Puerto Rico. A majority of sites cared for pediatric patients only rather than both adults and pediatric patients and were private rather than academic centers. Almost half of participating sites were minority/underserved designated NCORP sites.

CPG-inconsistent care

Among 129 patients, nearly half (48%) received CPG-inconsistent care for FP (Table 3). The proportion of patients who received CPG-inconsistent FP care was not significantly different from 0.5 (p = 0.618). Most cases of CPG-inconsistent care (93.5%) arose from a lack of FP discussion by health care professionals before the start of therapy. In one male patient, FP was discussed after treatment started. In three patients (1 male, 2 females) FP was discussed, but cryopreservation was not offered.

Table 3.

Care Delivered and Patient Outcomes

	All patients (n = 129) proportion (%)	Male (n = 95) proportion (%)	Female (n = 34) proportion (%)
FP CPG-inconsistent care	62/129 (48.1)	40/95 (42.1)	22/34 (64.7)
Fertility preservation not discussed	58/62 (93.5)	38/40 (95.0)	20/22 (91.0)
Fertility preservation discussed after chemotherapy given	1/62 (1.6)	1/40 (2.5)	0
Fertility preservation discussed but cryopreservation not offered/discussed	3/62 (4.8)	1/40 (2.5)	2 /22 (9.0)
Cryopreservationa
Cryopreservation completed	26/106 (24.5)	24/78 (30.8)	2/28 (7.1)
Cryopreservation declined	12/106 (11.3)	9/78 (11.5)	3/28 (10.7)
Cryopreservation attempted but failed	3/106 (2.8)	3/78 (3.8)	0
Cryopreservation not completed due to illness severity	1/106 (0.9)	0	1/28 (3.6)
Cryopreservation not completed, reason unknown	2/106 (1.9)	2/78 (2.6)	0

^a Cryopreservation completion data missing for 23 patients (17 males, 6 females).

CPG, clinical practice guideline; n, number.

In adjusted models, the odds of CPG-inconsistent care delivery were estimated to be lower at larger sites (OR: 0.79 95% CI: 0.68, 0.91). Furthermore, the odds of CPG-inconsistent care delivery were higher among female patients (OR: 2.50, 95% CI: 1.03, 6.05) and for patients receiving care at pediatric versus mixed adult/pediatric sites (OR: 3.57; 95% CI: 1.25, 10.25) (Table 4). Unadjusted estimates are presented in Supplementary Table S1. These findings were consistent with exploratory analyses stratified by sex (Supplementary Table S2).

Table 4.

Adjusted Models of CPG-Inconsistent Care and Completion of Cryopreservation Based on Site Characteristics, Patient Sex, and Area Deprivation Index

	Receipt of CPG-inconsistent carea	Noncompletion of cryopreservationb
	Estimated odds ratio (95% CI)
Site size (per 5 patients)	0.79 (0.68, 0.91)	0.85 (0.67, 1.07)
Pediatric site (ref: Mixed)	3.57 (1.25, 10.25)	1.72 (0.34, 8.66)
Minority/Underserved site (ref: No)	1.44 (0.57, 3.62)	0.77 (0.17, 3.39)
Area Deprivation Index (per 0.1)	1.12 (0.81, 1.55)	1.60 (1.01, 2.55)
Sex (ref: Male)	2.50 (1.03, 6.05)	5.78 (1.05, 31.69)

^a Based on 128 observations: 1 patient missing Area Deprivation Index.

^b Based on 105 observations: 1 patient missing Area Deprivation Index, 23 patients missing information regarding the completion of cryopreservation.

CI, confidence interval.

Cryopreservation completion

Among patients for whom information regarding cryopreservation completion was available, higher ADI of household (OR 1.60; 95% CI: 1.01, 2.55) and female sex (5.8; 95% CI: 1.05, 31.69) were associated with cryopreservation noncompletion (Table 4). The associations between noncompletion of cryopreservation and site size, site type (pediatric vs. mixed adult/pediatric), population served (minority/underserved vs. other) were not statistically significant. A comparison of factors influencing completion of cryopreservation in males versus females was not possible due to very low numbers of females completing cryopreservation (n = 2) (Supplementary Table S2). However, among males, noncompletion of cryopreservation was not associated with site size, site type (pediatric vs. mixed adult/pediatric; minority/underserved vs. other) or ADI.

Discussion

Among our cohort of pediatric patients diagnosed with cancer at NCORP sites, almost half received CPG-inconsistent FP care. FP discussion with these patients or their families before treatment initiation was not documented or, if discussed, cryopreservation was not offered. Patients receiving care at smaller or pediatric-only NCORP sites, particularly female patients, had a higher likelihood of receiving CPG-inconsistent care. When offered, the rate of completion of cryopreservation was low, with disparities for female patients and those living in areas of greater socioeconomic disadvantage being particularly large. The findings that cryopreservation noncompletion is associated with female sex and living in areas that are more highly disadvantaged further emphasize the gap in care for these patients.

Despite strong recommendations from several iterations of CPGs that FP be discussed with patients about to receive chemotherapy, low FP discussion rates have been reported. Quinn et al. reported that among 18- to 45-year-old cancer patients receiving care in four American cancer centers, only 26% had a note documenting a discussion of infertility risk documented in their health record, 24% had a discussion of FP options documented, and 13% had a documented referral to a fertility specialist.²¹ Lewin et al. describe a baseline rate of FP discussion documentation of 54% among patients 17 to 39 years of age receiving care at Canadian cancer centers.²² Other investigators evaluating access to FP services in minority-based NCORP practices found that among interview participants, only 40% acknowledged routine use of FP guidelines, whereas 30% did not consider the guidelines a priority and another 30% were unfamiliar with the guidelines.²³ Similarly, the main source of CPG inconsistency in our study was the omission of FP in the documented discussions between clinician and patients and families before initiation of therapy. Clinicians may avoid FP discussion due to lack of education, uncertainty regarding the level of infertility risk presented by a patient’s treatment, discomfort with the topic of infertility, a desire to protect the patient or family from information overload, assumptions that FP interventions are too expensive for the patient and family to entertain, or lack of access to the required services.^23,24

In our cohort, patients receiving care at pediatric-only or smaller sites had a lower likelihood of receiving CPG-consistent FP care. We speculate that FP services may be more readily available at pediatric sites that are partnered with adult services (mixed sites). In addition, pediatric sites with fewer patients eligible for FP interventions, may struggle to petition for improved access to the required services. Nonetheless, our findings underscored the challenges that pediatric and smaller NCORP sites may face to offer their patients CPG-consistent FP care. One solution may be for smaller and pediatric sites to forge relationships with larger centers, adult centers or regional FP referral centers to ensure that their patients have access to CPG-consistent FP interventions.

We also found that female patients were 2.5 times less likely to be informed of FP options (i.e., receive CPG-inconsistent care) than males. Other investigators have observed a similar disparity between sexes and have speculated that the gap may be attributed to the higher cost of FP interventions in women, decreased access to FP interventions, and the requirement to defer cancer treatment.^10,21,25 It is nevertheless important to prepare patients of both sexes and their families to make an informed choice regarding future fertility as this may improve quality of life and reduce fertility-related distress in survivorship, whatever decision is made.^26–29

Previously we elicited the perspectives of clinicians at NCORP institutions regarding barriers and facilitators to FP CPG-consistent care delivery.³⁰ Facilitators included prioritization of FP care by NCORP institutions and developing a program where clinicians are trained to discuss FP, resources are made available and the pathway to accessing them is clear. Several participating sites did have a systematic approach to cueing clinicians to discuss FP with newly diagnosed patients. This was obvious in their new diagnosis checklists and templated progress notes. They were the exception. The instigation of programs focused on infertility risk discussions and offering FP interventions increases the proportion of patients informed of FP options.^22,31 Recent tools, such as the Pediatric Initiation Network Risk Stratification System and others, will provide guidance to clinicians regarding the gonadotoxic risk of various treatments and will increase their comfort in having these discussions.^32–34

Few patients in our cohort completed cryopreservation. Cryopreservation noncompletion was associated with female sex and living in an area of higher deprivation. Certainly, FP interventions are more logistically difficult in female than male patients. They may require hormone administration, surgical intervention, and chemotherapy deferral. Success is not guaranteed. Furthermore, oocyte cryopreservation was newly recommended for routine use immediately before our study period.³⁵ Thus, resources to support its incorporation into routine care may not have been widely available initially. For either sex, cryopreservation is expensive and rarely covered by insurance. For example, in the United States, current oocyte cryopreservation costs are approximately 20 to 30 times higher than sperm cryopreservation costs.³⁶ These findings again underscore the gap in care experienced by female patients and those living with socioeconomic hardship. Notably, in 2018, ovarian tissue cryopreservation was recommended as an option for selected patients by the American Society of Clinical Oncology.³⁷ Consequently, the number of FP intervention options available to females has increased since our study period. Importantly, ovarian tissue cryopreservation, unlike oocyte cryopreservation, does not require ovarian stimulation and, thus, may be a more expedient option for cancer patients. However, expense and logistics continue to be barriers to completion of FP interventions.³⁸

The strength of this study lies in the rigorous methods used to prepare and adjudicate the patient summaries. However, it is possible that FP discussions and offers of cryopreservation may have taken place despite the lack of documentation in the health record. Given the consistency of our findings with similar studies of FP discussion documentation rates, we believe the gap between discussion and documentation to be slight. However, our rates of FP documentation may be lower than the rates of FP discussion. It is also possible that, particularly at sites that used electronic charting or checklists, FP discussions may have been charted in error as having taken place. Due to data sparsity, precision of some coefficient estimates is low, particularly in the stratified analyses. Nevertheless, the observed estimates were clinically meaningful and provide important information regarding FP practices. It should also be noted that confidence intervals were not adjusted for multiplicity; however, our results are consistent with those of previous studies. We did not collect information regarding the time between cancer diagnosis disclosure and initiation of treatment. Future studies should evaluate this to better understand potential reasons for CPG-inconsistent care. Lastly, although the data in this retrospective review reflect practice in 2014–2015, a scoping review of practice patterns regarding adherence to FP CPGs across pediatric community cancer centers is lacking. Evidence regarding patient-level and practice-level factors associated with CPG-inconsistent care will inform development and implementation of interventions to improve FP CPG uptake.

Pediatric oncology patients in this cohort often received CPG-inconsistent FP care, with disproportionate gaps noted particularly for females and patients treated at smaller or pediatric NCORP sites. The primary reason for CPG-inconsistent care was lack of FP discussion by clinicians. This information establishes an important baseline for future work to improve the quality of supportive care of pediatric patients receiving care at NCORP sites through trustworthy CPG implementation. Such work should focus on surmounting barriers to FP discussion initiation as well as on barriers to cryopreservation access and completion. Increased health care professional support and education in community cancer centers alongside improved financial support, advocacy at the legislative level to improve insurance coverage for FP services, and increasing participation and access to oncofertility consortia at NCORP sites may be targetable areas for future progress.

Authors’ Contributions

Conception and design: A.C.G., A.J.S., M.P.B., A.C.G., P.D.R., N.S., A.M.W., A.R.W., L.C.Y., L.S., and L.L.D. Collection and assembly of data: A.C.G., A.J.S., M.M.N., S.R., L.S., and L.L.D. Data analysis: A.C.G., M.M.N., S.R., and L.L.D. Data interpretation: All authors. Article writing: All authors. Final approval of the article: All authors. Accountable for all aspects of the work: All authors.

Author Disclosure Statement

Potential conflicts of interest: A.C.G. received grant funding to their institution from the National Institutes of Health, Canadian Institute of Health Research, American Cancer Society and Cancer Prevention Research Institution of Texas. She also received payment from Servier Pharmaceuticals for services as a content expert, and the Clark Hill Law Firm for expert testimony. A.C.G. also received salary support from the Children’s Oncology Group; funds were received by their institution.

Potential conflicts of interest: A.J.S. received salary support from the Children’s Oncology Group. Funds were received by his institution.

Potential conflicts of interest: M.M.N. received salary support to their institution from the National Cancer Institute and grant funding to their institution from the National Cancer Institute, Department of Defense and St Baldrick’s Foundation.

Potential conflicts of interest: S.K.P. has received consulting fees from Seagen unrelated to the work described in this article and is an external member of the Dana-Farber/Harvard Cancer Center Data Safety Monitoring Board.

Potential conflicts of interest: L.L.D. received salary support from the Children’s Oncology Group; funds were received by her institution.

No conflicts: M.P.B., P.D.R., N.S., L.S., L.C.Y., A.R.W., and A.M.W.

Funding Information

Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under NCORP Grant UG1CA189955 to the Children’s Oncology Group. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.