Adult cancer drugs have historically been exempted from pediatric testing requirements. In 2017, Congress passed the Research to Accelerate Cures and Equity (RACE) for Children Act to expand mandatory pediatric testing to cancer drugs; the law took effect in 2020. With this study, we sought to evaluate how the pediatric testing of molecularly targeted adult cancer drugs changed after the RACE Act.
In this retrospective cohort study, we used publicly available Food and Drug Administration data to compare pediatric post-approval requirements, trials, and trial characteristics, including timing, in adult cancer drugs before and after the RACE Act.
Between 2017 and 2024, the Food and Drug Administration approved 61 adult cancer drugs with molecular targets relevant to pediatric cancer; 40 were submitted before 2020, and 21 were submitted after 2020. The 40 pre-RACE Act drugs were associated with no pediatric post-approval requirements, whereas the 21 post-RACE Act drugs were associated with 15 pediatric post-approval testing requirements. Approximately two-thirds (26/40, 65%) of pre-RACE Act drugs and 57% (12/21) of post-RACE Act drugs were evaluated in pediatric trials. Among pre-RACE Act cancer drugs, pediatric trials were initiated a median of 0.04 years after approval (interquartile range: −3.3 to 1.9 years), whereas post-RACE Act trials were initiated a median of 2.8 years before approval (interquartile range: −4.3 to 0.3 years).
The RACE Act has been associated with greater numbers of pediatric post- approval testing requirements and the earlier initiation of pediatric trials, although early pediatric trial rates appear unchanged. Formalizing pediatric testing requirements may lead to the timely completion of pediatric studies to the benefit of pediatric patients with cancer.
The RACE Act facilitated the testing in children of cancer drugs originally approved in adult patients and could serve as an effective model for future legislation seeking to promote organization of timely clinical trials of new drugs in pediatric patients.
Prescription medicines have historically rarely been tested in children, exposing pediatric populations to potentially avoidable adverse effects in exchange for uncertain benefits.1 To increase formal testing of prescription drugs, in 2003, Congress passed the Pediatric Research Equity Act (PREA) to require manufacturers to test all new drugs and indications in pediatric populations before receiving approval.2 PREA studies may be waived for a limited number of statutory reasons, for example, if the drug was indicated for a rare disease, and they are frequently deferred to the post-approval period.3
New drugs indicated for adult patients with cancer are among the most common clinical categories of drug approvals that nearly always have PREA requirements waived or are exempt from PREA because they are indicated for a rare disease under the Orphan Drug Act. However, many modern cancer drugs target molecular pathways that may be shared by pediatric and adult cancers. Recognizing this, in 2017, Congress passed the Research to Accelerate Cures and Equity (RACE) for Children Act, which modified PREA to require that sponsors of new, molecularly targeted oncology drugs conduct trials evaluating “dosing, safety, and preliminary efficacy” in children, regardless of rare disease status, with the same options for waivers in certain circumstances.4 The law went into effect for eligible drug applications submitted for US Food and Drug Administration (FDA) approval on or after August 18, 2020.4
In a previous study, among 78 adult cancer drugs approved from 2007 to 2017, all were exempt from mandatory pediatric testing, with the majority (52/78, 67%) automatically exempted because the drug was approved for a rare type of cancer and the rest (26/78, 33%) waived by the FDA because the cancer did not occur in children.5 In the year following the RACE Act’s effective date, 3 of 5 oncology drugs had pediatric trial requirements.6 A Government Accountability Office report of pediatric study plans submitted by drug sponsors to the FDA in the 2 years following the RACE Act revealed that 25 pipeline drugs had pediatric study plans that could not have been required before the act.7
With dozens of new adult cancer drugs coming to market in the 5 years after the RACE Act, we sought to evaluate pediatric trial requirements and the characteristics of the trials conducted using a cohort of recent molecularly targeted adult cancer drug approvals to understand how pediatric testing rates and trial designs have changed since 2020. Our analysis had 3 parts. First, we analyzed the proportion of adult cancer drugs subject to pediatric testing requirements. We then calculated the proportion of pre- and post-RACE Act cancer drugs that had at least 1 associated industry-funded pediatric trial. Third, we analyzed initiation date, enrollment size, enrollment ages, blinding, and self-reported trial phase to identify potential changes in pediatric trial design before and after the RACE Act.
Methods
The RACE Act went into effect on August 18, 2020.3 We analyzed a cohort of cancer drugs approved between 2017 and 2024, comparing post-approval requirements, pediatric trials, and pediatric trial characteristics among molecularly targeted adult cancer drugs before the RACE Act with pediatric trials in similar drugs after the law’s implementation.
Cohort Selection
We used the FDA’s Novel Drug Approval lists8 to identify new molecular entities and new therapeutic biological products approved from January 1, 2017, through March 1, 2024. Using labeling information from the public Drugs@FDA database,9 we used the FDA’s Relevant Molecular Target List10 and Non-Relevant Molecular Target Leading to Waiver List11 located on the FDA’s Pediatric Oncology page12 to identify drugs with molecular targets relevant to children, although these lists are not binding on the FDA.13 We excluded drugs initially approved with a pediatric indication because these would obviously have undergone pediatric testing. Finally, we extracted the submission date of each drug’s application from the FDA’s review dossier to identify which drugs were submitted for approval after August 18, 2020, and thus subject to the RACE Act rules.
Data Extraction
PREA requires testing for all pediatric age groups in the absence of a waiver. We used each drug’s initial approval letter from Drugs@FDA to identify how pediatric testing requirements were met at the time of approval, including whether pediatric testing requirements were waived or deferred and, if so, for which age groups and what reason. We categorized pediatric testing as fully waived if there was no requirement to test the drugs in minors and partially waived if pediatric testing for certain age groups under age 18 was still required. We extracted the pediatric testing post-approval requirements associated with each drug, their projected completion dates, any existing rare disease designation under the Orphan Drug Act, and their average follow-up time as of May 13, 2024.
Trial Identification and Characteristics
For both pre- and post-RACE Act drug cohorts, we searched ClinicalTrials.gov for trials using the generic name of each approved drug, filtering for pediatric trials, defined as any trial enrolling a person <18 years old. To partially standardize the time across the drugs in the cohort, we included only trials initiated within 3 years of each drug’s initial approval date, although this period was necessarily shorter for most of our post-RACE Act cohort, which did not have 3 years of follow-up at the time of data collection. We excluded observational and expanded access studies, as well as trials sponsored exclusively by academic institutions without industry co-sponsors or collaborators because the RACE Act’s requirements apply only to manufacturers. As a subgroup analysis, we also matched post-RACE Act trials to mandatory post-marketing pediatric testing requirements using a combination of active ingredient, sponsor, and inclusion criteria, which has been used in previous research.4
We extracted National Clinical Trials identifiers, sponsor institution, completion status, study completion date (actual, if available, and estimated if not), comparator structure (eg, active comparator, placebo, single arm), blinding, randomization status, self-described trial phase, trial enrollment, study start date, and age criteria for enrollment as of May 13, 2024. We identified FDA-defined pediatric age groups enrolled in each study (eg, infants 0–2 years; children 2–12; adolescents 12–17) and whether trials were pediatric-only, which, similarly to previous studies,5 we defined as having an upper enrollment age limit of ≤21 years old. To evaluate differences in trial timing, we calculated the median time and interquartile range (IQR) between study start and approval, as well as the difference between actual or estimated study completion and approval. Descriptive statistics were generated using Excel version 16 (Microsoft). Differences in trial characteristics were evaluated using χ2 and Wilcoxon rank tests at a significance level of P <.05. This review did not constitute human subjects research and therefore did not require institutional review board approval.
Results
We identified 352 new molecular entities and therapeutic biological products approved between January 1, 2017, and March 1, 2024; 91 of these were cancer drugs. From the Mechanism of Action on each drug’s labeling, we excluded 1 oncology drug without a defined molecular target and used FDA lists to identify drugs with molecular targets relevant to children (n = 71).10,11 We excluded 10 molecularly targeted cancer drugs initially approved for indications including pediatric populations, resulting in 61 molecularly targeted adult cancer drugs in the final cohort (Fig 1). Forty drugs (66%) were submitted for approval before the RACE Act’s effective date; 21 (34%) were submitted to FDA after (Table 1).
Post-Approval Requirements
Of the 40 molecularly targeted adult oncology drugs in our cohort predating the RACE Act, 29 (73%) had rare disease designations and were therefore exempt from pediatric testing requirements. The remaining 11 (28%) had pediatric testing requirements waived due to trials considered “impossible or highly impracticable” because of a low number of children with the given condition. Thus, there were no pediatric testing-related post-approval requirements for any pre-RACE Act drugs (Table 1).
Of the 21 molecularly targeted adult oncology drugs in our cohort subject to the RACE Act, 8 (38%) were granted complete waivers from pediatric testing due to “impossible or highly impracticable” trials (Table 1). Ten drugs (48%) were granted partial waivers and had certain pediatric age groups exempted from testing for the same reason with studies deferred to the post-approval period; 3 drugs (14%) were not granted any testing waivers. In the 3 years after the RACE Act, 13 (62%) of 21 new cancer drugs had 15 pediatric post-approval requirements (some drugs may have multiple requirements); 8 (38%) had no post-approval requirements. Sixteen drugs (57%) had rare disease designations but, because of changes from the RACE Act, were not automatically exempt from pediatric testing.
Associated Pediatric Trials
Among the 40 pre-RACE Act drugs, within the first 3 years after approval, we identified 58 pediatric oncology studies (mean: 1.45 trials/drug), of which 21 were pediatric-only and were not designed to recruit adult patients (mean: 0.53 trials/drug). Fourteen drugs (35%) were associated with no pediatric trials (Table 1).
Among the 21 drugs subject to the RACE Act, with an average follow-up of 1.63 years after approval, we identified 19 pediatric oncology studies (mean: 0.90 trials/drug), of which 10 were pediatric-only (mean: 0.48 trials/drug). Nine drugs (43%) were associated with no pediatric trials.
Among the same 21 drugs, we also identified 11 pediatric trials fulfilling mandatory post-approval pediatric testing requirements for 10 different drugs, of which 8 trials (73%) were pediatric-only (Table 2). The remaining 4 mandatory pediatric post-approval studies (relating to 3 drugs) were not yet initiated according to the FDA’s post-marketing commitment database as of May 13, 2024.
Trial Characteristics
The 58 pre-RACE Act pediatric trials enrolled a median of 57 patients (IQR, 20–99; Table 2). Fifty-five (95%) were open-label, and 3 (5%) were blinded. Fourteen trials (24%) had any comparator, and 44 (76%) had no comparison group. Nineteen trials (33%) were Phase 1, 11 (19%) were Phase 1/2, 23 (40%) were Phase 2, and 5 (9%) were Phase 2/3 or 3. Twenty-seven trials (47%) enrolled infants, 32 (55%) enrolled children 2 to 12 years of age, and all enrolled adolescents.
The 19 post-RACE Act studies enrolled a median of 65 patients (IQR, 35–108; Table 2). All trials were open-label. Two trials (11%) had a comparator, and 17 (89%) had no comparison group. Five trials (26%) were Phase 1, 7 (37%) were Phase 1/2, 6 (32%) were Phase 2, and 1 (5%) was Phase 2/3 or Phase 3. Eleven trials (58%) enrolled infants, 13 (68%) enrolled children 2 to 12 years of age, and all 19 studies enrolled adolescents.
The 11 required post-RACE Act studies enrolled a median of 50 patients (IQR, 35–65; Table 2). All trials were open-label. None of the 11 trials had a comparison group. Two trials (11%) were Phase 1, 7 (37%) were Phase 1/2, 2 (11%) were Phase 2, and none were Phase 2/3 or Phase 3. Eight trials (73%) enrolled infants, 10 (91%) enrolled children 2 to 12 years of age, and all 11 studies enrolled adolescents.
Trial Timing
Among the 58 pre-RACE Act pediatric trials, the median trial was initiated 0.04 years after each drug’s approval (IQR, 3.3 years before approval to 1.9 years after approval). Among the same group, the median time between drug approval and actual or projected study completion was 5.4 years after approval (IQR, 1.4–7.7 years).
Among the 19 post-RACE Act trials, after a median follow-up of 1.63 years after approval, the median pediatric trial was initiated 2.8 years before each drug’s initial approval (IQR, 4.3–0.3 years before approval). Among the same drugs, the median time between drug approval and projected study completion was 3.3 years (IQR, 0.6–4.3 years).
Discussion
The RACE Act was designed to close a loophole in PREA through which drugs indicated to treat rare cancers in adults were automatically exempted from otherwise mandatory pediatric testing. As expected, compared with cancer drugs approved before the RACE Act, more recently approved molecularly targeted adult cancer therapies had more mandatory pediatric post-market studies and, despite shorter follow-up time, post-RACE Act drugs had similar pediatric testing rates as pre-RACE Act drugs. In addition, we found that pediatric trials for drugs subject to the RACE Act have been initiated earlier and have earlier anticipated completion dates. This may aid the timely acquisition of valuable dosing, safety, and efficacy data for children living with cancer and the clinicians treating them.
Our cohort of pre-RACE Act, molecularly targeted adult cancer drugs appears to have had similar regulatory treatment to the cohort of adult cancer drugs approved between 2007 and 2017 reported by Hwang and colleagues in earlier work.5 Approximately two-thirds of the drugs in that group treated a rare cancer, similar to the approximately three-quarters of our drugs that did the same. Like the earlier cohort of drugs, our pre-RACE Act cohort also had no pediatric testing requirements.
By contrast, our cohort of post-RACE Act adult cancer drugs makes it clear that the RACE Act has been effective at increasing pediatric testing requirements. Although all the drugs in our pre-RACE Act cohort lacked pediatric testing requirements, only about 40% (8/21) were free of such requirements after the law’s enactment. Three drugs subject to the law received no waivers and were obliged to be tested in all pediatric age groups.
The implementation of post-approval requirements is a necessary intermediate step for the RACE Act to have practical impact. However, the primary goal of the law was to increase the rates of “pediatric studies of cancer drugs…[to] determine whether a drug is safe and effective in children.”14 Thus, the translation of post-approval pediatric testing requirements into actual pediatric testing, which often faces significant delays in trial conduct and reporting,4 and the dissemination and translation of those trial results into clinical practice will be a key area of follow-up to evaluate the long-term impacts of the law.
Our post-RACE Act cohort had approximately half the median follow-up period (1.6 years versus 3 years) of our pre-RACE Act cohort. Despite this limitation, we found that nearly the same proportion of RACE Act-eligible adult cancer drugs had initiated pediatric trials (12/21, 57%), compared with before the Act (26/40, 65%). When identifying mandatory pediatric trials, we found that nearly half (48%, 10/21) of all RACE Act-eligible adult cancer drugs had initiated mandatory pediatric trials, compared with no mandatory pediatric trials before the law. The overall pediatric testing rate of pre-RACE Act adult cancer drugs was 65% (26/40) within 3 years of FDA approval, which is consistent with previous studies.5 Although our pre- and post-RACE Act cohorts are necessarily different in some ways, the experience with post-RACE Act testing, especially the increase in FDA-mandated trials, points to signs of progress.
The RACE Act appears to be successful at increasing the rate of dedicated pediatric-only trials, with a pre- to post-RACE Act increase from one-third (34%, 20/58) to more than half (53%, 10/19), although this change did not reach statistical significance (P = .16). As expected, mandatory post-RACE Act trials also enrolled infants <2 years old and children between 2 and 12 years at higher rates than pre-RACE Act pediatric trials, although these results did not reach statistical significance. Although enrolling a range of pediatric age groups is important, the FDA could also consider criteria such as geographic, racial, and socioeconomic diversity to enhance the generalizability of study results in required post-approval pediatric trials to enhance access to novel therapies for these groups. Although we were unable to analyze those characteristics of trial diversity, our analysis of characteristics such as comparator structure, blinding, trial phase and trial size did not highlight any differences with pediatric trials of cancer drugs approved in the years immediately preceding the RACE Act.
The RACE Act appears to have had an effect on the timing of pediatric trials. Post-RACE Act pediatric trials have been initiated earlier than pre-RACE Act trials relative to each drug’s initial approval date, with a statistically significant difference in median trial initiation date of almost 3 years. The addition of a formal regulatory requirement to complete pediatric studies may therefore lead manufacturers to organize trials earlier. The earlier start date of these trials is also reflected in a significantly earlier median expected completion date of post-RACE Act trials, estimated at 3.3 years after approval, compared with 5.4 years for pre-RACE Act drugs. However, post-approval studies are often delayed,15 even for required studies, like those under PREA.16 In addition, the RACE Act’s effects on earlier trial initiation may decrease the length of time when pediatric patients are using a cancer drug “off-label” with unknown adverse effects or suboptimal dosing regimens, and may therefore hasten the reporting of evidence of benefit for pediatric populations.
Our study has limitations. First, while the pre-RACE Act cohort of drugs all had 3 full years after approval to initiate pediatric trials, our post-RACE Act cohort had an average follow-up time of 1.6 years, likely undercounting pediatric trials that had not yet been initiated or registered at ClinicalTrials.gov for our post-RACE Act cohort at the time of data collection. Second, the pre-RACE Act cohort included only industry-funded trials in which a pharmaceutical company or other business were listed as sponsors or collaborators, excluding trials run by universities or other non-profits because the RACE Act requirement was only applicable to manufacturers. This makes it possible that we excluded trials sponsored by industry if industry ties were not adequately disclosed.
Policy Implications
Since passage of the RACE Act, some have proposed eliminating the Orphan Drug Act exemption from PREA entirely,17 effectively extending the RACE Act from cancer drugs to all drugs. Regardless, policymakers should ensure that any pediatric study requirements are met in a timely manner, for example, by providing automatic civil monetary penalties when trials are delayed without cause.
Legislators may consider using the RACE Act as a model for areas in which pediatric drug testing has lagged. However, because post-approval pediatric studies suffer from low completion rates,4 long-term follow-up and enforcement is still needed to ensure the translation of pediatric post-approval testing requirements into useful knowledge for pediatric patients with cancer and the clinicians who treat them.
Conclusions
The RACE Act has been associated with higher rates of post-approval pediatric testing requirements. Although the characteristics of pediatric trials appear to be unaffected by the law and early trial rates are so far unchanged, emerging evidence suggests that the RACE Act may be successful at prompting manufacturers to initiate and potentially complete pediatric trials sooner.
Dr Liu conceptualized and designed the study, collected and analyzed data, and drafted the initial manuscript; Dr Kesselheim conceptualized and designed the study, supervised data collection, and acquired funding; and all authors critically reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
FUNDING: This article was funded by Arnold Ventures, Commonwealth Fund. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest relevant to this article to disclose.