Regulatory Disparities Between US Accelerated Approval and Japanese Authorization of Oncology Drugs: An Analysis of Evidence Quality

ABSTRACT

While the accelerated approval (AA) program in the United States expedites the availability of drugs based on preliminary evidence to fulfill unmet medical needs, it has also raised significant concerns, including a lack of robust evidence of efficacy and subsequent withdrawals. Although drug lag and drug loss are growing regulatory concerns in Japan, considering the stricter withdrawal criteria in Japan, careful evaluation of clinical evidence of accelerated approval and approval timing differences is necessary. Here, we aimed to investigate differences in approval timing between the U.S. and Japan, as well as the clinical evidence of accelerated approval drugs that have not yet been approved in Japan. Using the U.S. Food and Drug Administration (FDA) and Japanese Pharmaceuticals and Medical Devices Agency (PMDA) databases, we examined the Japanese and U.S. regulatory status of cancer drugs granted accelerated approval in the U.S. between 2012 and 2022 and the characteristics of evidence for drugs not yet approved in Japan. Of 132 drug‐indication pairs that received accelerated approval between 2012 and 2022, 72 (54.5%) were approved in Japan by June 2024. Of the remaining 60 (45.5%) drugs not yet approved in Japan, the majority had methodological limitations, including a lack of comparators (93.2%) and scarcity of phase III trials (8.5%), as permitted by the accelerated approval program. Our findings suggest the need for an approach that addresses drug lag while ensuring both careful regulatory review and generation of robust evidence for efficacy and safety.

Study Highlights.

• What is the current knowledge on the topic?

  • ○
    While the accelerated approval (AA) program in the United States expedites the availability of drugs based on preliminary evidence to fulfill unmet medical needs, recent studies have identified significant concerns regarding the US accelerated approval (AA) pathway, including methodological limitations, lack of meaningful outcomes in confirmatory trials, and subsequent withdrawals.
• What question did this study address?

  • ○
    How are AA drugs approved in Japan and what is the underlying evidence for those lacking marketing authorization in Japan?
• What does this study add to our knowledge?

  • ○
    The majority of cancer drugs not yet approved in Japan have not been converted into traditional approval in the US and have had methodological limitations, such as lack of comparators and scarcity of phase III trials.
• How might this change clinical pharmacology or translational science?

  • ○
    Our findings highlight the need for careful regulatory review and generation of robust evidence for efficacy and safety when addressing drug lag and drug loss.

1. Introduction

The accelerated approval (AA) program is a regulatory pathway in the United States (US) that allows conditional marketing authorization of drugs based on preliminary evidence for treating life‐threatening conditions and addressing unmet medical needs [1]. The AA pathway typically requires manufacturers to conduct post‐approval confirmatory trials examining meaningful clinical outcomes while expediting patient access to potentially effective therapies. This mechanism is particularly utilized for novel cancer drugs. However, recent studies have identified significant concerns regarding this pathway, revealing that 41% of AA cancer drugs failed to demonstrate improvements in meaningful outcomes in confirmatory trials [2]. Moreover, many post‐approval confirmatory trials utilized the same surrogate endpoints as their AA review and contained significant methodological limitations, including absence of blinding, randomization, or comparator groups [2, 3]. Approximately 22% of AA cancer drugs were subsequently withdrawn from the market following failure of their confirmatory trials [2].

Nevertheless, the influence of AA extends beyond US regulatory decisions [4]. Many countries lack conditional approval pathways and the authority to mandate rigorous post‐approval trials, often deferring to initial FDA decisions even after US market withdrawal [5]. For instance, in China, certain cancer drug indications that were withdrawn from the US market remained approved or under regulatory review, raising concerns about patient exposure to cancer drugs without validated clinical benefit [6]. Similar concerns have been documented in many low‐ and middle‐income countries, where AA drugs continue to be marketed and promoted even after negative results in confirmatory trials, often without clear disclosure that the approval is conditional [7].

In Japan, two regulatory concerns have emerged: “drug lag”—delays between foreign and Japanese drug approvals—and “drug loss”—absence of drug development in Japan despite foreign approval. For cancer drugs, the median approval delay from US approval is 4.2 years [8]. While policymakers are implementing various strategies to reduce these delays, including waiving domestic clinical trial requirements, such approaches require careful consideration. This is particularly relevant given that Japan maintains more stringent withdrawal criteria, and all cancer drugs approved in Japan following US AA have remained authorized despite subsequent FDA withdrawal [9].

Recent concerns about the AA pathway highlight the need for careful evaluation of approval timing differences between the US and Japan. Although studies have documented discrepancies in regulatory standards between US accelerated approval and the European Medicines Agency (EMA) conditional marketing authorization program [10, 11, 12], the Japanese regulatory status of cancer drugs granted AA in the US remains unexamined. Investigating regulatory disparities between the US and Japan is important, as differences in regulatory frameworks—such as the relatively limited enforcement capacity in Japan compared to the US and Europe [13]—may provide insights for addressing drug lag in other countries. This study compared the regulatory status of AA cancer drugs in the US and Japan and characterized the underlying evidence for those lacking marketing authorization in Japan.

2. Methods

This study compared the Japanese and US regulatory status of cancer drugs granted AA in the US between 2012 and 2022 and evaluated the characteristics of underlying evidence used for AA review among drugs not yet approved in Japan. All AA cancer drugs granted approval between 2012 and 2022 were identified from the published US Food and Drug Administration (FDA) list [1]. Their approval status in Japan was assessed through the Pharmaceuticals and Medical Devices Agency (PMDA) database as of June 2024 [14]. Approval status was evaluated at the drug‐indication level.

For each AA cancer drug not approved in Japan by June 2024, clinical trial information was extracted from three sources in order of priority: FDA review reports, manufacturer press releases, and the ClinicalTrials.gov database. Consistent with the methodology employed in previous studies [2, 3, 15], we collected clinical trial data on study design, trial phase, setting, location (whether Japan was included as a trial site), overall trial participation, allocation, interventional model, masking, control status, and primary endpoint. The drug selection process is presented in Figure 1.

FIGURE 1.

Flow chart of drug selection and approval status of cancer drugs in the US and Japan.

Descriptive statistics were calculated for the clinical trial data. We conducted a Jonckheere–Terpstra trend test to evaluate ordered trends in US regulatory outcomes (verified clinical benefit, ongoing AA, or withdrawal) across the approval status categories in Japan. A p‐value less than 0.05 was considered statistically significant. All statistical analyses were conducted using Microsoft Excel 365 (Microsoft Corp., Redmond, WA, USA) and Stata MP version 17 (Stata Corp., College Station, TX, USA). The number of drugs was counted at the drug‐indication level.

As all data were publicly available, institutional review board approval was not required in accordance with the Ethical Guidelines for Medical Research Involving Human Participants at Tohoku University, Japan.

3. Results

Of 132 drug‐indication pairs granted AA between 2012 and 2022, 72 (54.5%) were approved in Japan by June 2024 (Figure 1). Among drugs approved in Japan (n = 72), 48 (66.7%) had demonstrated clinical benefit and converted to traditional approval in the US, 19 (26.4%) remained under ongoing confirmatory trials, and 5 (6.9%) had been withdrawn from the US market by June 2024. In contrast, among drugs not yet approved in Japan (n = 60, 45.5%), 16 (26.7%) had converted to traditional approval in the US after demonstrating clinical benefit, 26 (43.3%) were under ongoing confirmatory trials, and 18 (30.0%) had their AA withdrawn (Figure 1). The Jonckheere–Terpstra trend test demonstrated a significant ordered trend in US regulatory outcomes according to Japanese approval status (Z = −4.43, p < 0.001), with AA cancer drugs not yet approved in Japan showing significantly higher rates of AA withdrawal than those approved in Japan.

Table 1 summarizes the characteristics of evidence supporting the 60 AA drugs not yet approved in Japan. The evidence comprised 59 interventional studies and one non‐interventional study—a retrospective chart review of alpelisib [16]. Among the interventional studies, the majority were multicenter (n = 55, 93.2%) and international trials (n = 50.5, 85.6%), with 29 trials (49.2%) enrolling 100 or more patients. However, few studies incorporated robust methodological features: only 9 trials (15.3%) were randomized, 5 (8.5%) were phase III studies, and just one trial each employed double‐blinding or placebo controls. No trials evaluated true endpoints such as overall survival and quality of life as their primary outcome measure, consistent with the surrogate endpoint focus permitted under the AA program.

TABLE 1.

Characteristics of the evidence supporting US accelerated approval drugs not yet approved in Japan (n = 60).

Characteristic variables	Number of drug‐indication pairs, no. (%)
Study design
Interventional study	59 (98.3)
Non‐interventional retrospective study a	1 (1.7)
Characteristics of interventional studies (n = 59) a
Trial phase
Phase I	3.5 (5.9)
Phase II	50.5 (85.6)
Phase III	5 (8.5)
Setting
Multicenter	55 (93.2)
Single‐center	4 (6.8)
Location
Multiple countries (two or more countries)	50.5 (85.6)
Single‐country	8.5 (14.4)
Inclusion of Japan in the trial location
Yes	8 (13.6)
No	51 (86.4)
Allocation
Randomized	9 (15.3)
Non‐randomized	50 (84.7)
Interventional model
Parallel assignment	22.5 (38.1)
Sequential assignment	9.5 (16.1)
Single‐arm	27 (45.8)
Masking
Open label	58 (98.3)
Double‐blind	1 (1.7)
Control status
Placebo‐controlled	1 (1.7)
Controlled	3 (5.1)
Uncontrolled	55 (93.2)
Primary endpoint
True endpoint	0 (0)
Surrogate endpoint b	59 (100)
Progression‐free survival	3 (5.1)
Complete/partial response rate	53.5 (90.7)
Adverse events	1.5 (2.5)
Other surrogate endpoint	1 (1.7)
Number of patients included in the trial
100 or more patients	29 (49.2)
Less than 100 patients	30 (50.8)

^a EPIK‐P1 study was a retrospective chart review of alpelisib (Vijoice). Alpelisib was excluded from subsequent analysis of trial characteristics, as it was not an interventional clinical study.

^b For drugs supported by multiple trials, each trial was weighted equally as 1/n (n = total number of trials).

4. Discussion

This analysis identified several critical patterns in the regulatory landscape of AA cancer drugs between Japan and the US. First, drugs approved in Japan predominantly included those that successfully converted to traditional approval in the US (66.7%), while notably excluding the majority (18 of 23) of drugs withdrawn from the US market. Cancer drug indications not yet approved in Japan but subsequently withdrawn from the US market include nivolumab monotherapy for patients with hepatocellular carcinoma previously treated with sorafenib and pembrolizumab monotherapy for patients with metastatic small cell lung cancer, both of which were withdrawn after their confirmatory trials failed to demonstrate prolonged overall survival. All AA drugs were approved based on surrogate endpoints, as permitted by the AA program. In contrast, Japan utilizes surrogate endpoints such as overall response rates and progression‐free survival within the framework of traditional approval. Approximately two‐thirds of cancer drug approvals in Japan are reportedly based on surrogate endpoints, whereas overall survival is used in about 30% of cancer drug approvals [17, 18]. No AA drug was approved based on quality of life, despite its recognized importance in evaluating clinical benefit [19]. Furthermore, trials supporting AA drugs not yet approved in Japan demonstrated methodological limitations. The predominance of uncontrolled studies (93.2%) and the scarcity of phase III trials (8.5%) raise concerns about the robustness of efficacy evidence. This conservative approval pattern suggests that Japan's regulatory framework may have effectively filtered out drugs less likely to demonstrate meaningful clinical benefit, albeit at the potential cost of delayed access to potentially beneficial treatments.

Although this study focused on regulatory disparities between US accelerated approval and Japanese authorization of oncology drugs, the issue is not unique to Japan or the US. For instance, Chinese immuno‐oncology therapies have faced challenges in obtaining FDA approval, primarily due to reliance on single‐country clinical data [20, 21]. This highlights that drug lag is a global issue affecting therapies developed across various regions and underscores the need for global drug development and regulatory harmonization.

To address drug lag, the Japanese Ministry of Health, Labour and Welfare recently issued a ministerial ordinance expanding the application of the conditional early approval program [22]. This revision enables the approval of AA‐granted drugs and those with planned confirmatory trials in the US and Europe, even with limited evidence. While this policy change aims to expedite drug access, previous studies have highlighted limitations in the design and endpoint setting of confirmatory trials for AA‐approved drugs [3], necessitating careful evaluation of the applicability of Western confirmatory trials to Japanese patients. This concern extends to healthcare expenditure considerations. Unlike the US, where drug approval and public insurance coverage decisions undergo separate review processes, Japan conventionally provides national health insurance coverage for virtually all approved drugs. Consequently, there is a high likelihood that costs for potentially less effective drugs approved based on preliminary evidence will be covered by public insurance, leading to increased financial burden on both the national healthcare system and patients.

Several limitations should be acknowledged. First, since clinical trial information was obtained from multiple sources, potential discrepancies in trial and approval information among databases may exist. Second, this study could not quantify the impact of delayed drug access on patient outcomes, such as the number of patients who could have benefited from earlier drug availability during the delay period. This represents a critical limitation when evaluating drug lag, as the fundamental tension inherent in accelerated approval pathways involves balancing the risk of approving therapies based on limited efficacy data against the potential benefit of expedited access for patients with life‐threatening conditions. Therefore, further studies are needed to evaluate the clinical consequences of regulatory delays on patient outcomes.

In conclusion, while strategies to address drug lag are necessary in Japan, it is crucial to recognize that novel drugs in other countries are sometimes approved based on limited evidence and may subsequently be withdrawn from the market. A balanced approach is therefore required: one that addresses drug lag while simultaneously ensuring careful regulatory review, generation of robust evidence of efficacy and safety both pre‐ and post‐approval, and consideration of healthcare cost implications.

Author Contributions

M.I. and A.M. wrote the manuscript; M.I. and A.M. designed the research; M.I. and A.M. performed the research; M.I. analyzed the data. A.M. supervised the entire study project.

Conflicts of Interest

A.M. received a grant from Japan Society for the Promotion of Science (KAKENHI Grant Number 25K20514) outside of this study. M.I. declared no conflicts of interest.

Iwata M. and Murayama A., “Regulatory Disparities Between US Accelerated Approval and Japanese Authorization of Oncology Drugs: An Analysis of Evidence Quality,” Clinical and Translational Science 18, no. 9 (2025): e70339, 10.1111/cts.70339.

Data Availability Statement

This study utilized information from publicly accessible databases. Data are available upon request.