Time to Confirmatory Study Initiation After Accelerated Approval of Cancer and Noncancer Drugs in the US

Abstract

This cross-sectional study examines and compares the time taken from the accelerated approval of cancer and noncancer drugs to the initiation of confirmatory studies in the US.

Under the accelerated approval pathway, the US Food and Drug Administration (FDA) may approve a drug based on surrogate end points that are reasonably likely to predict clinical benefit.¹ Although FDA-required studies must be conducted to confirm a drug’s clinical benefit, the accelerated approval pathway has received growing criticism due to manufacturers’ delays in completing these studies.^1,2 One potential source of delays is that confirmatory studies are often not underway at the time of accelerated approval; that is, if studies are ongoing at the time of approval, then time to conversion to full approval or market withdrawal is shorter.^1,3

The Consolidated Appropriations Act of 2023⁴ provided that the FDA “may require” confirmatory studies to be “underway prior to approval, or within a specified time period after” accelerated approval. Since the statute permits, but does not require, the FDA to use this authority, further information regarding the timing of confirmatory study initiation and completion for cancer and noncancer products may benefit both the FDA and policy makers as they implement this legislation and consider future reforms to the accelerated approval pathway.

Methods

In this cross-sectional study, the FDA report “CDER Drug and Biologic Accelerated Approvals Based on a Surrogate Endpoint”⁵ was used to identify drugs granted accelerated approval from 2009 to 2019. Next, we reviewed accelerated approval letters in the Drugs@FDA database⁶ to identify postmarketing requirements that assessed each product’s efficacy and abstracted final report submission dates. (Because cancer drugs could have more than 1 indication, the combination of drug and indication was referred to as a cancer product–indication pair). We then screened ClinicalTrials.gov for corresponding confirmatory studies and extracted study start dates to determine whether studies were underway at the time of accelerated approval and to calculate the time from accelerated approval to study initiation. Last, we considered postmarketing requirements delayed if ongoing confirmatory studies or completed reports were past FDA deadlines indicated in the accelerated approval letters. This study used publicly available, deidentified data and was therefore exempt from institutional board review. It followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

We used χ² tests and Mood median tests to compare the frequency and timing of confirmatory studies for cancer product–indication pairs and noncancer products. All tests were 2-sided; P < .05 was considered statistically significant. Analyses were conducted using Microsoft Excel (Microsoft).

Results

For 103 cancer–indication pairs and noncancer products, 20.31% (26 of 128) of confirmatory studies were not underway at the time of accelerated approval, and the median (IQR) time from approval to study initiation was 1.41 (1.10-2.04) years (Table 1). Noncancer products had a larger proportion of confirmatory studies with initiation occurring after accelerated approval (37.04% [10 of 27] vs 15.84% [16 of 101]; P = .008) and a longer median (IQR) time from approval to study initiation (2.11 [2.00-2.47] years vs 1.20 [0.63-1.45] years; P = .02) compared with cancer product–indication pairs.

Table 1. Time to Initiation of Confirmatory Studies for Products Granted Accelerated FDA Approval From 2009 to 2019^a.

	No. (%)b			P value
	All product-indication pairs	Cancer product–indication pairs	Noncancer products
Confirmatory studies underway at time of accelerated approvalc
Yes	92 (71.88)	79 (78.22)	13 (48.15)	.008
No	26 (20.31)	16 (15.84)	10 (37.04)
Unable to identify	10 (7.81)	6 (5.94)	4 (14.81)
Time to start of confirmatory study, median (IQR), yd	1.41 (1.10-2.04)	1.20 (0.63-1.45)	2.11 (2.00-2.47)	.02

Abbreviation: FDA, US Food and Drug Administration.

^a Instances were excluded if postmarketing requirements solely evaluated a drug’s safety, products were duplicates (eg, approvals were for different product formulations or expanded patient populations), or manufacturers were released from postmarketing requirements (eg, due to safety or submitting other confirmatory studies for a product’s conversion).

^b Seventy-two drugs (55 cancer, 17 noncancer) with 103 indications (86 cancer; 17 noncancer) were granted accelerated approval from 2009 to 2019.

^c χ² Tests were used to compare the confirmatory studies underway at the time of accelerated approval for cancer product–indication pairs and noncancer products. If multiple confirmatory studies were being conducted for a postmarketing requirement, then we selected the earliest confirmatory study.

^d Time from accelerated approval to start of the confirmatory study was calculated for the postmarketing requirements that were not underway at the time of accelerated approval. Mood median tests were used to compare median time to start confirmatory studies between cancer product–indication pairs and noncancer products.

When assessing delays in confirmatory study completion, studies underway at the time of accelerated approval were less likely to be submitted past FDA-designated deadlines relative to studies initiated after approval (37.04% [30 of 81] vs 61.11% [11 of 18]) (Table 2). Similar patterns were observed for cancer and noncancer products; however, delays in confirmatory study completion were more pronounced among studies initiated after approval for noncancer products than for cancer product–indication pairs (100% [6 of 6] vs 41.67% [5 of 12]; P = .04).

Table 2. Delays in Completing or Submitting Confirmatory Studies for Products Granted Accelerated FDA Approval From 2009 to 2019^a.

	No. (%)
	Studies underway at approval				Studies not underway at approval
	All products	Cancer product–indication pairs	Noncancer products	P value	All products	Cancer product–indication pairs	Noncancer products	P value
Within FDA deadlinesb
Yesc	49 (60.49)	41 (58.57)	8 (72.73)	.62	6 (33.3)	6 (50.00)	0	.04
Nod	30 (37.04)	27 (38.57)	3 (27.27)		11 (61.11)	5 (41.67)	6 (100)
Unable to identify	2 (2.47)	2 (2.86)	0		1 (5.56)	1 (8.33)	0

Abbreviation: FDA, US Food and Drug Administration.

^a Instances were excluded if postmarketing requirements solely evaluated a drug’s safety, that involved duplicates (in instances in which approvals were for different product formulations or expanded patient populations), in which manufacturers were released from postmarketing requirements (eg, due to safety or submitting other confirmatory studies for a product’s conversion), or that were for cancer product–indication pairs and noncancer products that were not yet converted and within FDA deadlines (eg, they may end up being converted to full approval or withdrawn from the market within or past FDA deadlines or their confirmatory studies may still be ongoing at the FDA deadline).

^b χ² Tests were used to compare the confirmatory studies for cancer product–indication pairs and noncancer products. If multiple confirmatory studies were being conducted for a postmarketing requirement, the earliest confirmatory study was selected.

^c Postmarketing requirements were considered on time if completed reports (converted or withdrawn products) were within FDA submission deadlines indicated in accelerated approval letters.

^d Postmarketing requirements were considered delayed if ongoing confirmatory studies (not-yet-converted products) or completed reports (converted or withdrawn products) were past the FDA submission deadlines indicated in accelerated approval letters.

Discussion

Building on prior evidence^1,3 that suggests that initiation lags appear to increase the overall time to conversion to full approval or market withdrawal, our findings suggest that initiation lags may increase delays in confirmatory study completion.

This analysis has limitations. We were unable to identify some confirmatory studies due to limitations in publicly available data³ and to determine the reasons why confirmatory studies were or were not underway at the time of accelerated approval.

Although the FDA has endorsed the timely initiation of confirmatory studies,¹ the agency must choose to use its expanded statutory authority⁴ and require that these studies be underway at the time of accelerated approval. Such actions could support the simplification (eg, streamlined patient recruitment and enrollment) and completion of confirmatory studies. However, additional reforms (eg, linking payer reimbursement to FDA-designated confirmatory study deadlines) may be necessary for timely completion of confirmatory studies and verification of products’ clinical benefit.

Supplement 1.

Data Sharing Statement