US Food and Drug Administration Approval of Drugs Not Meeting Pivotal Trial Primary End Points, 2018-2021

Abstract

This cross-sectional study determines the frequency of and rationale for US Food and Drug Administration (FDA) approval of drugs not meeting pivotal trial primary efficacy end points.

In June 2021, the US Food and Drug Administration (FDA) granted accelerated approval to aducanumab for the treatment of Alzheimer disease despite both pivotal trials being stopped for futility with respect to the primary trial end point: change from baseline clinical dementia rating score.¹ Instead, FDA approval was based on an exploratory, surrogate marker: amyloid plaque reduction.² Prior studies have shown an FDA history of approving new drugs and high-risk medical devices when evidence of efficacy is uncertain.^3,4 We sought to systematically determine the frequency of and rationale for FDA approval of drugs based on pivotal trials with null findings for 1 or more primary efficacy end points.

Methods

Using the Drugs@FDA database, we conducted a cross-sectional study of all FDA-approved New Drug Applications between 2018 and 2021. From their approval package, we identified all pivotal trials, associated primary efficacy end points, and whether each end point met the prespecified criteria. For each drug not meeting at least 1 prespecified primary end point, we extracted drug, pivotal study, and end point characteristics, then characterized the FDA’s rationale for approval described in approval packages, whether a postmarket study was required or requested, and whether the postmarket study addressed the unmet primary end point.

This study used public, nonidentifiable data that did not constitute human participants research (45 CFR §46.102) and was not submitted for institutional review board review. The study followed the STROBE reporting guideline.

Results

Between 2018 and 2021, the FDA approved 210 new drugs, 21 (10.0%) based on pivotal studies with null findings for 1 or more primary efficacy end points (Table 1). These 21 drugs were approved for 21 unique clinical indications. Of these drugs, 11 (52.4%) were first in class, 10 (47.6%) received orphan designation, and 14 (66.7%) received an expedited review designation. Before approval, an advisory committee was convened for 3 (14.3%) of the drugs.

Table 1. Characteristics of 21 Drugs Approved by the US Food and Drug Administration (FDA) Supported by Pivotal Studies With Null Findings for 1 or More Primary Efficacy End Points, 2018-2021.

Drug characteristic	No. (%)
Drug type
Biologic	7 (33.3)
Small molecule	14 (66.7)
Expedited pathway
Any	14 (66.7)
Priority review	10 (47.6)
Accelerated approval	4 (19.0)
Breakthrough therapy	2 (9.5)
Fast track	9 (42.9)
None	7 (33.3)
Orphan drug	10 (47.6)
First in class
Yes	11 (52.4)
No	10 (47.6)
Therapeutic area
Oncology	4 (19.0)
Infectious disease	3 (14.3)
Psychiatry	3 (14.3)
Neurology	2 (9.5)
Hematology	2 (9.5)
Imaging and radiation	2 (9.5)
Pulmonology	2 (9.5)
Rheumatology	2 (9.5)
Gastroenterology and inborn error	1 (4.8)
FDA advisory committee convened
Yes	3 (14.3)
No	18 (85.7)
FDA approval rationale
Post hoc effectiveness analysis	7 (33.3)
Near-miss or nonsignificant numerical benefit	2 (9.5)
Success of other primary end points in pivotal study with unmet end point	4 (19.0)
Success of other pivotal studies	13 (61.9)
Benefit suggested by secondary or exploratory end points	10 (47.6)
FDA-required or -requested postmarketing studies addressing unmet end points
Yes	7 (33.3)
No	14 (66.7)

The 21 drug approvals were supported by 55 pivotal trials; 6 approvals (28.6%) were supported by a single pivotal trial. These 55 trials included 68 primary efficacy end points, 27 of which (39.7%) were null; among these 27 end points, 13 (48.1%) were clinical outcomes, 7 (25.9%) were surrogate markers, 4 (14.8%) were clinical scales, and 3 (11.1%) were composite end points. Four drug approvals—naxitamab-gqgk for high-risk refractory or relapsed neuroblastoma, tazemetostat hydrobromide for epitheloid sarcoma, migalastat hydrochloride for adults with Fabry disease, and asparaginase erwinia chrysanthemi (recombinant) for acute lymphoblastic leukemia and lymphoblastic lymphoma—were based on a single pivotal study with null results for all pivotal primary end points.

The FDA’s most common reasons for approval of the 21 drugs were success in at least 1 other pivotal study (n = 13; 61.9%), positive findings from secondary or exploratory end points in the pivotal study (n = 10; 47.6%), and favorable post hoc analysis (n = 7; 33.3%) (Table 1). The FDA required or requested postmarketing studies for 7 (33.3%) of the drugs to address the null end point or a related clinical end point. Table 2 provides additional details for the 21 drug approvals.

Table 2. Descriptions of the 21 Drugs Approved by the US Food and Drug Administration (FDA) Supported by Pivotal Studies With Null Findings for 1 or More Primary Efficacy End Points, 2018-2021.

Drug name	FDA-approved indication within study period	No. of pivotal studies	No. of primary end points	Missed primary end points	Missed primary end points	FDA rationale for approval
Margetuximab-CMKB (Margenza; MacroGenics)	To treat ERBB2-positive breast cancer	1	2	1	OS (superiority to trastuzumab)	Success of other primary end points in pivotal study Othera
68Ga PSMA-11	For detection and localization of prostate cancer	2	3	2	Sensitivity and specificity of 68Ga-PSMA-11 PET for the detection of regional nodal metastases vs pathology at radical prostatectomy on a per-patient basis	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points Near-miss or numeric benefit in missed outcome measure
Naxitamab-gqgk (Danyelza; Y-mAbs Therapeutics)	To treat high-risk refractory or relapsed neuroblastoma	1	1	1	ORR	Near-miss or numeric benefit in outcome measure Benefit suggested by secondary or exploratory end points Otherb
Tazemetostat hydrobromide (Tazverik; Epizyme)	To treat epithelioid sarcoma	1	1	1	ORR	Otherc
Lumateperone tosylate (Caplyta; Intra-Cellular Therapies)	To treat schizophrenia	3	3	1	Change from baseline to end of treatment period on the Positive and Negative Syndrome Symptom scale	Success of primary effectiveness end points in at least 1 of multiple pivotal studies
Istradefylline (Nourianz; Kyowa Kirin Co, Ltd)	To treat adult patients with Parkinson disease experiencing off episodes	8	12	6	1. Change from baseline in daily off time based on 24-h diaries completed by patients (4 missed end points) 2. Change from baseline to the last available postbaseline value for the number of hours of awake time spent in the off state per day, recorded in the patient diary (2 missed end points)	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Post hoc effectiveness analysis Benefit suggested by secondary or exploratory end points
Cilastatin sodium, imipenem, and relebactam (Recarbrio; Merck)	To treat complicated urinary tract and complicated intra-abdominal infections	2	2	1	Combined clinical and microbiological response at discontinuation of intravenous therapy visit (noninferiority to imipenem) in patients with complicated urinary track infections	Otherd
Solriamfetol hydrochloride (Sunosi; Axsome Therapeutics)	To treat excessive sleepiness in adult patients with narcolepsy or obstructive sleep apnea	3	10	1	Change in Maintenance of Wakefulness Test from baseline to week 12	Success of other primary effectiveness end point in pivotal study Success of primary effectiveness end points in at least 1 of multiple pivotal studies
Triclabendazole (Egaten; Novartis)	To treat fascioliasis, a parasitic infestation caused by 2 species of flatworms or trematodes that mainly affect the liver, sometimes referred to as liver flukes	8	1	1	Resolution of abdominal pain at hospital discharge (day 10)	Post hoc effectiveness analysis Othere
Baloxavir marboxil (Xofluza; Genentech)	To treat acute uncomplicated influenza in patients who have been symptomatic for no more than 48 h	2	2	1	Time to alleviation of influenza signs and symptoms	Post hoc effectiveness analysis Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points
Migalastat hydrochloride (Galafold; Amicus Therapeutics)	To treat adults with Fabry disease	1	1	1	Proportion of patients with ≥50% reduction from baseline to month 6 in the average number of globotriaosylceramide inclusions per kidney interstitial capillary in biopsy samples	Post hoc effectiveness analysis
Fostamatinib disodium (Tavalisse; Rigel Pharmaceuticals)	To treat thrombocytopenia in adult patients with persistent or chronic immune thrombocytopenia	2	2	1	Stable platelet response by week 24, defined as having a platelet count of at least 50 000/μL on at least 4 of the 6 visits over weeks 14-24	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Otherf
Ivacaftor and ivacaftor and tezacaftor (Symdeko; Vertex Pharmaceuticals)	To treat cystic fibrosis in patients aged 12 y or older	3	3	1	Absolute change in forced expiratory volume in 1 s (percent predicted) from baseline to week 3	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points Otherg
Tezepelumab-ekko (Tezspire; Amgen and AstraZeneca)	To treat severe asthma as an add-on maintenance therapy	3	3	1	Categorized percent reduction from baseline in the daily oral corticosteroid dose at week 48 while not losing asthma control	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points
Ropeginterferon alfa-2b-njft (Besremi; PharmaEssentia)	To treat polycythemia vera	2	2	1	Disease response rate after 12 mo defined by No. of patients who met all 4 components in the PROUD-PV study: Hematocrit <45% without phlebotomy (at least 3 mo since last phlebotomy) Platelet count <400 × 109/L White blood cell count <10 × 109/L (all 3 hematologic parameters measured by blinded central laboratory) Normal spleen size (by imaging, central, blinded assessment where spleen length <12 cm for females and <13 cm for males)	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Otherh
Avacopan (Tavneos; ChemoCentryx)	To treat severe active antineutrophil cytoplasmic autoantibody–associated vasculitis (granulomatosis with polyangiitis and microscopic polyangiitis) in combination with standard therapy, including glucocorticoids	1	4	1	The proportion of participants achieving disease remission at week 26. Disease remission at week 26 was defined by the following criteria: Achieving a Birmingham Vasculitis Activity Score (BVAS) as determined by the adjudication committee No administration of glucocorticoids for treatment of ANCA-associated vasculitis within 4 wk prior to week 26 No BVAS >0 during the 4 wk prior to week 26 (if collected for an unscheduled assessment) (Met end point for noninferiority to active control; did not meet end point for superiority)	Success of other primary end points in pivotal study Benefit suggested by secondary or exploratory end points
Anifrolumab-fnia (Saphnelo; AstraZeneca)	To treat moderate to severe systemic lupus erythematosus along with standard therapy	3	3	1	Systemic Lupus Erythematosus Responder Index-4 at week 52, which is a composite of the following: A reduction in baseline Systemic Lupus Erythematosus Disease Activity Index 2000 disease activity score of ≥4 points No worsening in Physician’s Global Assessment defined as an increase of ≥0.3 from baseline on a 0-3 visual analog scale No worsening in British Isles Lupus Assessment Group 2004 activity index defined as ≥1 new A or ≥2 new B items compared with baseline No discontinuation of investigational product No initiation of restricted medication beyond the protocol-allowed threshold before week 52	Success of primary effectiveness end points in at least one of multiple pivotal studies Benefit suggested by secondary or exploratory end points Post hoc effectiveness analysis Otheri
Asparaginase erwinia chrysanthemi (recombinant) (Rylaze; Jazz Pharmaceuticals)	To treat acute lymphoblastic leukemia and lymphoblastic lymphoma in patients who are allergic to Escherichia coli–derived asparaginase products as a component of a chemotherapy regimen	1	1	1	Response rate, defined as the proportion of patients with the last 72-h nadir serum asparaginase activity level ≥0.1 IU/mL during the first course of Rylaze	Post hoc effectiveness analysis
Aducanumab-avwa (Aduhelm; Biogen)	To treat Alzheimer disease	2	2	1	Change from baseline in Clinical Dementia Rating Scale Sum of Boxes at week 78	Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points Post hoc effectiveness analysis Otherj
Piflufolastat F-18 (Pylarify; Lantheus)	To identify prostate-specific membrane antigen-positive lesions in prostate cancer	2	3	1	Sensitivity of piflufolastat F-18 PET/CT for detection of pelvic lymph node prostate cancer metastases against a histopathology reference standard derived from the tissues removed at pelvic lymph node dissection	Success of other primary end point in pivotal study with missed end point Success of primary effectiveness end points in at least 1 of multiple pivotal studies Benefit suggested by secondary or exploratory end points
Viloxazine hydrochloride (Qelbree; Supernus Pharmaceuticals)	To treat ADHD	4	7	1	Change from baseline in the ADHD Rating Scale-5 total score at the end of study	Success of primary effectiveness end points in at least 1 of multiple pivotal studies

Abbreviations: ADHD, attention deficit/hyperactivity disorder; CT, computed tomography; ORR, overall response rate; OS, overall survival; PET, positron emission tomography; PROUD-PV, Pegylated Interferon Alpha-2b Versus Hydroxyurea in Polycythemia Vera.

^a The OS was immature (70% of targeted OS events); final OS will be provided as part of the agreed postmarketing commitment.

^b Data from a second, nonpivotal study were included in the primary efficacy population and cited as additional evidence of efficacy; although the lower bound of the 95% CI was on the cutoff value for ORR, ORR in the pivotal study and supporting study was considered clinically meaningful by the FDA.

^c Although the clinical team leader and clinical reviewer recommended against FDA approval through the accelerated approval pathway due to what they considered insufficient evidence of efficacy, the Oncologic Drug Advisory Committee voted 11-0 recommending accelerated approval, citing periods of stable disease observed in some treated patients, prolonged responses observed in some patients, and the rarity of the disease and paucity of treatment options.

^d The FDA determined that the 2 pivotal studies were not appropriately designed for testing the noninferiority hypotheses prespecified by the trial. The FDA also noted that due to their design, the studies had significant limitations, making it difficult for the agency to come to reliable conclusions based on the studies. Ultimately, the FDA did not approve the drug based on these pivotal studies but instead on in vitro and animal studies.

^e Favorable analysis of cure rates in single-arm studies (World Health Organization/CIBA-Geigy studies and second published study) against historical control generated from a third published study provided additional evidence of effectiveness).

^f Missed primary end point in 1 pivotal study was attributed to a patient in the control arm who had a highly variable platelet count, extension study with findings supportive of efficacy.

^g Pivotal study with null primary end point was conducted in patient population featuring mutations not expected to be responsive to drug.

^h Single-arm, phase 2 study with primary end point of complete hematologic response was the primary basis of approval of New Drug Application, and result for primary end point was determined to be clinically meaningful by the FDA; objective laboratory measures from phase 3 study, such as change in white blood cell and platelet count, were supportive of phase 2 findings.

ⁱ A phase 2 study, though not properly controlled for multiplicity, met its primary end point and was considered by the FDA as additional evidence of safety and efficacy.

^j The FDA stated that reduction in brain amyloid plaque is reasonably likely to predict clinical benefit and is therefore an appropriate surrogate. Additionally, a safety and tolerability study showed statistically significant benefit compared with placebo in Clinical Dementia Rating Scale Sum of Boxes improvement from baseline.

Discussion

From 2018 to 2021, 10% of drugs approved by the FDA were based on pivotal studies with null findings for 1 or more primary efficacy end points. Study limitations include a focus on pivotal trials, although the agency may have considered other studies or other regulators’ decisions. We were unable to analyze withdrawn or rejected New Drug Applications, as this information is not public.

Our findings underscore the complexity of regulatory decision-making, as exemplified by evidence of effectiveness despite a null primary end point finding (eg, tezacaftor-ivacaftor and triclabendazole). For other drugs (eg, cilastatin sodium, imipenem, and relebactam and istradefylline), the evidence of efficacy was less clear. Greater transparency regarding FDA decision-making could increase clinician, patient, and payer confidence in novel drugs and improve clinical use. Timely completion of postapproval studies addressing areas of clinical uncertainty is also necessary.⁵

Supplement.

Data Sharing Statement