Abstract
This study evaluates whether FDA-approved novel cancer therapeutics supported by pivotal trials with adequate representation of minoritized groups were associated with slower clinical development times than those with inadequate representation.
Ensuring demographic representation in clinical trials supporting new oncology drug approvals by the US Food and Drug Administration (FDA) is essential, considering the disproportionate burden certain diseases pose to older adults (≥65 years) and racial and ethnic minoritized populations.1 Drug sponsors conducting pivotal trials often recruit participants from high-enrolling sites, which may not serve minoritized populations,1 and include international sites, which may not represent the relevant US patient population.2 Women, older adults, and certain racial and ethnic groups often face barriers to trial participation (eg, lack of knowledge about trial availability, distrust of recruitment efforts).1,2 Some have speculated that requirements to improve representation in clinical trials3 may lead to longer recruitment time and thus prolong novel product testing and development.4 We evaluated whether FDA-approved novel cancer therapeutics supported by pivotal trials with adequate representation of minoritized groups were associated with slower clinical development times than those with inadequate representation.
Methods
Drugs@FDA was used to identify key dates (ie, investigational new drug [IND] application submission [ie, when FDA has authorized the sponsor to begin human testing] and new drug application [NDA] or biological license application [BLA] submission), approval pathways, and pivotal trials supporting approval for all new oncology therapeutics approved between January 1, 2015, and December 31, 2021.5 We used in the following order ClinicalTrials.gov results and indexed publications, FDA approval packages, product labels, and FDA Snapshots to identify the proportion of patients in pivotal trials who were reported as women, older adults, Asian, Black, and Hispanic/Latino (eAppendix in Supplement 1).6 The 2016 US Cancer Statistics and American Cancer Society databases were used to estimate demographic distribution of US patients with different cancer diagnoses. We calculated participation to prevalence ratios (PPRs) for each pivotal trial–indication pair by dividing the percentage of each demographic subgroup enrolled in the pivotal trial by the percentage of that subgroup in the US patient population with the approved indication, with PPRs at least 0.8 indicating adequate representation,6 mirroring the FDA’s recommendation to enroll participants who adequately “reflect the population most likely to use the drug.”3 Clinical development times, calculated as time between IND and NDA or BLA submission dates,5 were compared between therapeutic indications with adequate and inadequate representation of female, older adult, Asian, Black, and Hispanic/Latino patients using Mann-Whitney tests in RStudio (version 4.1.2). Two-tailed P < .05 denoted statistical significance. As post hoc analyses, clinical development times were stratified across approval characteristics (NDA vs BLA, accelerated vs traditional approval, and orphan vs nonorphan designation).
Results
The FDA approved 82 novel therapeutics for 85 oncology indications between 2015 and 2021, of which 30 (35%) were for hematologic cancers, 59 (69%) were drugs, 62 (73%) had orphan designations, and 45 (53%) had accelerated approval. Most indications were approved based on pivotal trials with adequate representation for female (82%) and Asian (73%) but not older (46%), Black (11%), and Hispanic/Latino (27%) patients (Table 1).
Table 1. Clinical Development Times for 85 Novel Oncology Therapeutic Indications Approved by the FDA Between 2015 and 2021 With Adequate vs Inadequate Demographic Representation in Pivotal Trials Supporting FDA Approval.
| Measure of demographic representation | Total therapeutic indications with demographic data, No. (%) | Adequate representation in pivotal trials (PPR ≥0.80)a | Inadequate representation in pivotal trials (PPR <0.80)a | P value | ||
|---|---|---|---|---|---|---|
| No. (%) | Clinical development time, median (IQR), y | No. (%) | Clinical development time, median (IQR), y | |||
| Female | 82 (96) | 67 (82) | 5.9 (4.2-7.8) | 15 (18) | 5.4 (4.0-7.5) | .80 |
| Age ≥65 y | 80 (94) | 37 (46) | 5.8 (4.0-7.3) | 43 (54) | 5.9 (4.2-7.8) | .69 |
| Asian | 82 (96) | 60 (73) | 6.0 (4.3-7.7) | 22 (27) | 5.9 (4.0-7.6) | .89 |
| Black | 83 (98) | 9 (11) | 5.8 (4.8-7.4) | 74 (89) | 6.1 (4.0-7.9) | .87 |
| Hispanic/Latino | 67 (79) | 18 (27) | 5.6 (3.9-7.3) | 49 (73) | 5.8 (4.2-7.7) | .60 |
Abbreviations: FDA, US Food and Drug Administration; PPR, participation to prevalence ratio.
For the therapeutics indications with multiple pivotal trials, the highest PPR value available was used.
The median clinical development times for approvals based on pivotal trials with adequate vs inadequate representation of female patients was 5.9 years (IQR, 4.2-7.8) vs 5.4 years (IQR, 4.0-7.5; P = .80); older adults, 5.8 years (IQR, 4.0-7.3) vs 5.9 years (IQR, 4.2-7.8; P = .69); Asian patients, 6.0 years (IQR, 4.3-7.7) vs 5.9 years (IQR, 4.0-7.6; P = .89); Black patients, 5.8 years (IQR, 4.8-7.4) vs 6.1 years (IQR, 4.0-7.9; P = .87); and Hispanic/Latino patients, 5.6 years (IQR, 3.9-7.3) vs 5.8 years (IQR, 4.2-7.7; P = .60) were not statistically significantly different (Table 1). Median clinical development times were largely consistent when stratified across approval characteristics (NDA vs BLA, accelerated vs traditional approval, and orphan vs nonorphan designation; Table 2).
Table 2. Stratified Clinical Development Times for 85 Novel Oncology Therapeutic Indications Approved by the FDA Between 2015 and 2021 With Adequate vs Inadequate Demographic Representation in Pivotal Trials Supporting FDA Approval.
| Measure of demographic representation | Total therapeutic indications with demographic data, No. (%) | Adequate representation in pivotal trials (PPR ≥0.80)a | Inadequate representation in pivotal trials (PPR <0.80)a | Total therapeutic indications with demographic data, No. (%) | Adequate representation in pivotal trials (PPR ≥0.80)a | Inadequate representation in pivotal trials (PPR <0.80)a | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| No. (%) | Clinical development time, median (IQR), y | No. (%) | Clinical development time, median (IQR), y | No. (%) | Clinical development time, median (IQR), y | No. (%) | Clinical development time, median (IQR), y | |||
| New drug approval (n = 59) | Biologics license approval (n = 26) | |||||||||
| Female | 57 (97) | 47 (82) | 5.8 (4.2-7.5) | 10 (18) | 5.1 (4.0-6.3) | 25 (96) | 20 (80) | 6.6 (3.9-8.9) | 5 (20) | 6.0 (4.8-11.3) |
| Age ≥65 y | 58 (98) | 26 (45) | 5.9 (4.1-7.2) | 32 (55) | 5.3 (4.2-8.0) | 22 (85) | 11 (50) | 5.8 (4.2-8.0) | 11 (50) | 6.0 (4.2-7.8) |
| Asian | 57 (97) | 44 (77) | 5.4 (4.2-7.4) | 13 (23) | 5.9 (4.2-7.4) | 25 (96) | 16 (64) | 6.9 (4.6-8.9) | 9 (36) | 5.9 (4.0-10.1) |
| Black | 57 (97) | 5 (9) | 6.1 (5.8-7.4) | 52 (91) | 5.7 (4.0-7.5) | 26 (100) | 4 (15) | 4.5 (3.0-6.6) | 22 (85) | 6.9 (4.4-9.2) |
| Hispanic/Latino | 46 (78) | 12 (26) | 5.8 (4.0-7.4) | 34 (74) | 5.0 (4.0-6.5) | 21 (81) | 6 (29) | 5.3 (3.5-6.4) | 15 (71) | 7.7 (5.3-9.1) |
| Accelerated approval (n = 45) | Traditional approval (n = 40) | |||||||||
| Female | 44 (98) | 39 (89) | 4.9 (4.0-7.2) | 5 (11) | 4.8 (4.8-5.4) | 38 (95) | 28 (74) | 7.6 (4.6-10.6) | 10 (26) | 6.0 (4.0-10.6) |
| Age ≥65 y | 44 (98) | 18 (41) | 5.6 (4.6-7.0) | 26 (59) | 4.8 (3.8-6.9) | 36 (90) | 19 (53) | 5.9 (4.0-8.1) | 17 (47) | 7.6 (5.4-12.3) |
| Asian | 45 (100) | 34 (76) | 5.0 (4.1-7.0) | 11 (24) | 4.9 (4.1-6.4) | 37 (93) | 26 (70) | 7.3 (4.9-9.7) | 11 (30) | 7.6 (4.1-10.7) |
| Black | 43 (96) | 3 (7) | 6.1 (6.0-6.8) | 40 (93) | 4.9 (4.0-7.0) | 40 (100) | 6 (15) | 5.3 (3.6-8.1) | 34 (85) | 7.5 (4.6-11.0) |
| Hispanic/Latino | 32 (71) | 11 (34) | 4.8 (3.9-6.5) | 21 (66) | 4.9 (4.3-6.1) | 35 (88) | 7 (20) | 7.3 (4.6-13.1) | 28 (80) | 6.5 (4.1-9.8) |
| Orphan designated (n = 62) | Nonorphan designated (n = 23) | |||||||||
| Female | 62 (100) | 50 (81) | 5.6 (4.2-7.9) | 12 (19) | 5.4 (4.0-7.0) | 20 (87) | 17 (85) | 6.3 (4.0-7.6) | 3 (15) | 5.4 (5.1-9.2) |
| Age ≥65 y | 58 (97) | 25 (43) | 5.8 (4.0-7.3) | 33 (57) | 5.1 (4.0-7.7) | 22 (96) | 12 (55) | 6.2 (4.3-7.3) | 10 (45) | 7.4 (6.0-9.2) |
| Asian | 62 (100) | 44 (71) | 5.6 (4.2-7.7) | 18 (29) | 5.4 (4.0-8.4) | 20 (87) | 16 (80) | 6.6 (4.6-8.1) | 4 (20) | 6.7 (5.4-7.5) |
| Black | 60 (97) | 7 (12) | 6.1 (5.3-8.1) | 53 (88) | 5.4 (4.0-7.9) | 23 (100) | 2 (9) | 4.5 (3.8-5.2) | 21 (91) | 6.6 (4.8-7.6) |
| Hispanic/Latino | 50 (81) | 14 (28) | 5.0 (3.9-6.9) | 36 (72) | 5.0 (4.0-7.7) | 17 (74) | 4 (24) | 7.0 (5.7-7.4) | 13 (76) | 6.6 (5.4-7.6) |
Abbreviations: FDA, US Food and Drug Administration; PPR, participation to prevalence ratio.
For the therapeutics indications with multiple pivotal trials, the highest PPR value available was used.
Discussion
FDA oncology approvals had similar clinical development times regardless of whether their pivotal trials had adequate or inadequate representation of female, older adult, Asian, Black, or Hispanic/Latino patients. These findings suggest that efforts to further increase representation in pivotal trials may not be associated with prolonged trial enrollment or clinical development times, as commonly feared.
Study limitations included reliance on US patient demographic data, which were not uniformly reported, and inability to account for marketing applications that were not approved owing to recruitment failure. Although many pivotal trials include international sites, FDA guides sponsors to submit data for approval of oncology therapeutics that reflect the US population.3
Better reporting of demographic data by country and monitoring success of efforts to promote enrollment of diverse populations in clinical trials supporting drug development are needed.
Section Editors: Jody W. Zylke, MD, Deputy Editor; Karen Lasser, MD, and Kristin Walter, MD, Senior Editors.
eAppendix. eMethods
Data Sharing Statement
References
- 1.AACR cancer disparities progress report. American Association for Cancer Research. Accessed September 15, 2023. https://cancerprogressreport.aacr.org/disparities/
- 2.National Academies of Sciences, Engineering, and Medicine. Improving Representation in Clinical Trials and Research: Building Research Equity for Women and Underrepresented Groups. National Academies Press; 2022. doi: 10.17226/26479 [DOI] [PubMed] [Google Scholar]
- 3.Guidance document: enhancing the diversity of clinical trial populations—eligibility criteria, enrollment practices, and trial designs. US Food and Drug Administration . Published November 2020. Accessed September 15, 2021. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/enhancing-diversity-clinical-trial-populations-eligibility-criteria-enrollment-practices-and-trial
- 4.The Future of Medicine: Legislation to Encourage Innovation and Improve Oversight, Hearing Before the Subcommittee on Health; Committee on Energy and Commerce, 117th Cong, 2nd session (2022) (testimony of Lucy Vereshchagina, vice president, science and regulatory advocacy, Pharmaceutical Research and Manufacturers of America). Accessed October 1, 2023. https://docs.house.gov/meetings/IF/IF14/20220317/114510/HHRG-117-IF14-Wstate-VereshchaginaL-20220317.pdf
- 5.Wong AK, Mooghali M, Ramachandran R, Ross JS, Wallach JD. Use of expedited regulatory programs and clinical development times for FDA-approved novel therapeutics. JAMA Netw Open. 2023;6(8):e2331753. doi: 10.1001/jamanetworkopen.2023.31753 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Varma T, Mello M, Ross JS, Gross C, Miller J. Metrics, baseline scores, and a tool to improve sponsor performance on clinical trial diversity: retrospective cross sectional study. BMJ Med. 2023;2(1):e000395. doi: 10.1136/bmjmed-2022-000395 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eAppendix. eMethods
Data Sharing Statement
