Trends in oncology drug lags in Japan from 2001 to 2020: A cross‐sectional study

Abstract

Anticancer drugs are essential in the treatment of serious diseases, but their applications are limited by drug lags. This study investigated the characteristics of anticancer drugs approved in Japan over the past 20 years and compared the drug lag trends between Japan and the US. We assessed the changes in drug lag between Japan and the US and the factors affecting the drug lags using publicly available data for anticancer drugs approved in Japan from January 2001 to December 2020. A total of 299 anticancer drugs were approved in Japan in the last 20 years. The approval lag median between the US and Japan was 498 days (16.6 months), peaking in 2002, and decreasing annually thereafter. The minimum approval lag was 173.5 days (5.7 months) in 2018. Multivariate regression analysis revealed that “global simultaneous strategy,” “catch‐up strategy,” and “immunotherapy” are major factors shortening the drug lag. In the past decade, 226 anticancer drugs were approved in Japan. The drug lag for anticancer drugs between Japan and the US peaked in 2002, after which it declined sharply to less than a year. However, the lag was shortest in 2018.

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

The drug lag between anticancer drugs in Japan and the US is believed to have disappeared. The long‐term trend was unclear.

• WHAT QUESTION DID THIS STUDY ADDRESS?

Is there still a drug lag in the development of Japan's anticancer drugs?

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

The drug lag for anticancer drugs between Japan and the US reached its lowest level in 2018. Our findings suggest that globalization and changes in domestic policies have had strong impacts on changes in drug lag.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

This is the first study to provide long‐term and comprehensive information on the drug lag between Japan and the US, which could be used to elucidate global challenges in the development of pharmaceuticals.

INTRODUCTION

Drug lag relates to the drug approval interval between countries. The lag in approvals is affected by the difference in development start time, development period, and review time. Japan has long had a drug lag problem, and the large lag between Japan and Europe and the US in terms of access to medicine and drug approvals has often caused social problems in Japan. ¹ , ² The problem of drug lag becomes more problematic for serious diseases. Countermeasures against drug lags for anticancer drugs are especially necessary compared to drugs in other fields. ³

The clinical development of anticancer drugs in Japan since 2000 has experienced many promising changes. Molecular‐targeted drugs have emerged in clinical settings, and drugs aimed at precision and personalized medicine have become mainstream. ⁴ Various guidelines were issued from regulatory authorities, such as the 2006 anticancer drugs guideline revision, ⁵ International Conference on Harmonisation (ICH E5) notification, ⁶ 2007 guideline for multinational clinical trials, ⁷ 2012 guideline for first‐in‐human trials, ⁸ and 2018 ICH E17. ⁹ Several development‐promoting systems were also established, such as a public knowledge‐based application ¹⁰ and the Sakigake designation system. ¹¹ The Japan Consortium of Clinical Research Core Hospital was designated as a major medical institution for clinical trials that necessitated the improvement of the institution infrastructure. ⁸ In addition, several Japanese companies moved their development bases to the US. ⁸ These developments have promoted the globalized clinical development of anticancer drugs to be approved in Japan and are believed to have greatly contributed to eliminating drug lag. ¹² , ¹³

Herein, we conducted a historical and comprehensive analysis of trends in anticancer drug approval in Japan from January 2001 to December 2020 and studied the changes in drug lag and the associated drivers. Our study provides new insights into the challenges and advances in the development of anticancer drugs in Japan. This study investigated drug lag over a long period of 20 years and analyzed trends.

METHODS

Selection of drugs for cancer treatment

We investigated initial new drug applications (iNDAs) and supplemental new drug applications (sNDAs) for anticancer drugs intended for systemic administration among all the anticancer drugs approved by the Ministry of Health, Labour and Welfare (MHLW) between January 2001 and December 2020. iNDA was defined as the first time the compound was approved. An sNDA was defined as a case where a compound that obtained an iNDA was subsequently added to the cancer indication. Similarly, a compound that obtained an iNDA for a non‐cancer indication and then obtained a cancer indication for the first time was also defined as an sNDA.

Data collection

We investigated anticancer drugs approved in Japan. In addition, we examined the drug lag between Japan and the US in principle, and decided not to investigate the drug lag with other regions because (1) the development of many anticancer drugs was largely carried out in parallel in the US and Europe; ¹⁴ (2) more than 95% of new molecular entity (NME) anticancer drugs were approved in the US for global use, and the gap between Japan and the US is generally larger than between Japan and Europe; ¹² and (3) there are more NMEs approved only in the US than in Europe. ¹⁵ Data were mainly obtained from the Pharmaceuticals and Medical Devices Agency (PMDA) website (http://www.pmda.go.jp/english/index.html). Information from the US Food and Drug Administration (FDA) was primarily obtained from the FDA website (http://www.accessdata.fda.gov/scripts/cder/drugsatfda/). Data collection from the PMDA and FDA websites was conducted as described in our previous studies. ⁵ , ¹⁰ , ¹² , ¹³ , ¹⁶ , ¹⁷ , ¹⁸

Data abstraction

We defined the difference in approval dates, that is, drug lag, between Japan and the US or Japan as the “approval lag.” We also defined “development start lag” as the difference in development start dates in each country. In addition, the “review time lag” was defined as the difference in the review time (the period obtained by subtracting the application date from the approval date) in each country. This study was prepared in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline ¹⁹ for cross‐sectional studies. The characteristics of the oncology drugs, indications, clinical and regulatory details, clinical trials, and data packages were extracted. A.H., M.O., R.S., and Y.M. extracted the regulatory and clinical details of each oncology drug. Any questions regarding regulatory and clinical characteristics were discussed between all authors and were adjusted when necessary; any disagreement was resolved by a third reviewer (T.M. or H.M.).

Statistical analysis

The data were mainly analyzed using descriptive statistics. We also performed a multiple regression analysis to obtain estimated values, partial regression coefficients, and p values when exploring the factors contributing to drug lag. Multiple regression analysis was performed to obtain estimated values, partial regression coefficients, and p values when examining the factors affecting drug lag. The step‐down procedure was applied to select the important factors related to the drug lag from all collected factors. All statistical analyses were performed using JMP® Pro 15.

Ethics statement

This study did not require institutional review board approval or patient informed consent because it was based on publicly available information and involved no patient records.

RESULTS

Characteristics of oncology drugs approved in Japan

There were 299 oncology drugs approved in Japan from January 2001 to December 2020, including 122 iNDA and 177 sNDA approvals (Table 1). The number of oncology drugs approved in Japan annually increased, peaking at 36 approvals in 2020 (Figure S1).

TABLE 1.

Clinical and regulatory characteristics of the 299 approved oncology drugs from 2001 to 2020 in Japan.

Characteristic	N (%)
Number	299 (100.0)
NDA
iNDA	122 (40.8)
sNDA	177 (59.2)
Tumor type
Major cancer	87 (29.1)
Non‐small cell lung cancer	34 (11.4)
Breast cancer	27 (9.0)
Colorectal cancer	17 (5.7)
Gastric cancer	9 (3.0)
Not major cancer	212 (70.9)
Multiple myeloma	15 (5.0)
Prostate cancer	13 (4.3)
Melanoma	13 (4.3)
Renal cell carcinoma	11 (3.7)
Non‐Hodgkin lymphoma	10 (3.3)
Chronic lymphocytic leukemia	9 (3.0)
Ovarian cancer	8 (2.7)
Cervical cancer	8 (2.7)
Hepatocellular carcinoma	8 (2.7)
Acute lymphatic leukemia	8 (2.7)
Other solid tumor	59 (19.7)
Other hematological cancer	50 (16.7)
Solid cancer/hematologic cancer
Solid cancer	208 (69.6)
Hematologic cancer	91 (30.4)
Limitation of indication
Resistant or second‐line<=	98 (32.8)
Nothing or first‐line	201 (67.2)
Mode of action
Cytotoxic drug	95 (31.8)
Molecularly targeted drug	142 (47.5)
Immunotherapy	37 (12.4)
Hormonal drug	14 (4.7)
Antibody drug conjugate	3 (1.0)
Others	8 (2.7)
Companion diagnostics
Yes	82 (27.4)
No	217 (72.6)
Special designation by PMDA
Orphan drug designation	111 (37.1)
Normal application	104 (34.8)
Priority review	53 (17.7)
Public knowledge‐based application	30 (10.0)
Expedited review	15 (5.0)
Sakigake designation	6 (2.0)
Pediatric disease designation	6 (2.0)
Conditional early approval system	3 (1.0)
Strategy for clinical development
Global simultaneous strategy	127 (42.5)
Bridging strategy	109 (36.5)
Domestic strategy	30 (10.0)
Other style (public knowledge‐based application etc)	33 (11.0)
Using foreign clinical data for NDA package in Japan
Yes	246 (82.3)
No	53 (17.7)
Special committee on unapproved drug in Japan
Yes	58 (19.4)
No	241 (80.6)
All cases investigated after approval in Japan
Yes	115 (38.5)
No	229 (76.6)
Postmarketing clinical study requirement in Japan
Yes	18 (6.0)
No	281 (94.0)
Special designation by FDA
Priority review and/or orphan designation	81 (27.1)
Fast track designation	32 (10.7)
Accelerated approval	37 (12.4)
Breakthrough therapy designation	11 (3.7)
FDA approval when Japan approval
Yes	218 (72.9)
No	81 (27.1)
Patients in pivotal clinical study (n)
<100	63 (21.1)
>100	236 (78.9)
Domestic company/foreign company
Domestic company	108 (36.1)
Foreign company	191 (63.9)
Origin of product
Japan	59 (19.7)
Foreign country	240 (80.3)
Biological drug
Yes	88 (29.4)
No	211 (70.6)
Formulation
Oral	121 (40.5)
Injection	170 (56.9)
Others	8 (2.7)

Abbreviations: FDA, Food and Drug Administration; iNDA, initial new drug application; NDA, new drug application; sNDA, supplemental new drug application; PMDA, Pharmaceuticals and Medical Devices Agency.

Eighty‐seven of the 299 approved oncology drugs (29.1%) were for major cancers (non‐small cell lung cancer, breast cancer, colorectal cancer, and gastric cancer). Solid cancer was also common (208 approvals, 69.6%). The most common types of cancer were non‐small cell lung cancer (34 approvals, 11%), breast cancer (27 approvals, 9%), and colorectal cancer (17 approvals, 6%) (Figure S2).

Changes in clinical characteristics and regulatory background were examined every 5 years; those with statistically significant changes are shown in Figure 1. Molecularly targeted drugs were the most common (142 approvals, 47.5%) and annually increased. Refractory/relapse annually increased for the treatment lines. The bridging strategy was the most common clinical development strategy, accounting for 36.5% over the past 20 years, but declined after peaking in 2006–2010, at which time the global simultaneous strategy and catch‐up strategy annually increased. There were 30 approvals (10.0%) of public knowledge‐based applications without clinical trials. In addition, the 246 approvals (82.3%) made using overseas data annually increased.

FIGURE 1.

Changes in drug type, clinical development strategy, and treatment lines for approved oncology drugs from 2001 to 2020 in Japan. (a) Mode of action of the anticancer drug. (b) Treatment lines in each indication. (c) Strategy for clinical development. (d) Use of foreign data in the Japanese approval process.

Changes in drug lag between Japan and the US

Table S1 shows each drug lag data set used in this study. Of these, 64 (21.4%) were not approved in the US at the time of approval in Japan. The breakdown of unapproved drugs in the US was developing (11 approvals), under review (2 approvals), withdrawn (4 approvals), or not developing (47 approvals).

Initially, we examined the drug lag between Japan and the US. The median drug lag [IQR] from 2001 to 2020 was 498 days (16.6 months) [181.5–1302.0 days], and the mean value was 901.9 days (30.1 months) ± 1282.6 days. This indicated earlier approval in the US than in Japan. This study examined drug lag over 20 years, with some years having a very small number of anticancer drugs approved; therefore, we divided the period from 2001 to 2020 into four 5‐year periods (2001–2005, 2006–2010, 2011–2015, 2016–2020) and examined the changes over time in approval lag, development start lag, and review time lag (Figure 2).

FIGURE 2.

Changes in lags between the US and Japan over 5‐year periods from 2001 to 2020 for oncology drugs. The horizontal line in each box shows the median. The x‐axis shows the average. The line at the top edge and bottom edge of each box shows the 75th percentile and the 25th percentile, respectively. The upper and lower limit of the vertical line is the maximum value and minimum value, respectively.

The median and average approval lag decreased with time. The development start lag, and review time lag were the largest between 2006 and 2010 and decreased thereafter. The annual changes in approval lag peaked in 2002 and started to decrease thereafter; the minimum value was 173.5 days (5.7 months, also the median) [78.5–342.0 days] in 2018 (Figure 3). The median was 220 days (7.3 months) as of 2020; thus, a considerable approval lag remained. In addition, we analyzed 191 anticancer drugs approved in Japan between 2001 and 2020, for which the development period in the US (“approval in the US” –“development start date in the US”) was identified to examine the impact of the expedited programs (including Priority Review/Orphan designation and other expedited programs in the US) on the development period of the US during the past 20 years. The median developmental period [IQR] from 2001 to 2020 was 2019.0 days (67.3 months) [1458.0–3110.0 days], and the mean value was 2358.4 days (78.6 months) ± 1153.7 days. The development times of 117 drugs that received either Fast Track designation, Priority Review/Orphan designation, Accelerated Approval, or Breakthrough Therapy designation from the FDA (with expedited program) and 74 drugs that did not receive either of the expedited programs (without expedited program) were compared. The median developmental period [IQR] was 1966.0 (65.5 months) [1428.0–2922.0 days], and the mean value was 2212.1 days (73.3 months) ± 1032.2 days with the expedited drug program. The median developmental period [IQR] was 2373.0 (79.1 months) [1544.8–3234.3 days], and the mean value was 2589.8 days (86.3 months) ± 1297.2 days without the expedited drug program. The difference in development time with and without the expedited drug program was significant (p = 0.0271, Student's t‐test).

FIGURE 3.

Annual changes in approval lag between the US and Japan from 2001 to 2020 for oncology drugs. The horizontal line in each box shows the median. The x‐axis shows the average. The line at the top edge and bottom edge of each box shows the 75th percentile and the 25th percentile, respectively. The upper and lower limit of the vertical line is the maximum value and minimum value, respectively.

Examination of factors affecting drug lag in Japan and the US

We performed an exploratory multiple regression analysis to find the factors contributing to drug lag. The analysis was performed with approval lag, development start lag, and review time lag as objective variables to examine the factors affecting the drug lag of oncology drug approval in Japan, and relevant factors as explanatory variables; an effect level was estimated for each factor for the size of the approval lag. Table S2 shows the details for the objective and explanatory variables.

The step‐down procedure was used to evaluate the roles of the significant variables in the regression analysis. Table 2 shows the results of the partial regression coefficient, t value, and p value for each of the factors affecting “approval lag,” development start lag,” and “review time lag.” “Global simultaneous strategy,” “Catch‐up strategy,” “Immunotherapy drug,” and “Priority Review/Orphan” designation in FDA significantly contributed to shortening the approval lag.

TABLE 2.

Multivariate regression analysis on approval lag, development start lag, and review time lag between the US and Japan.

Parameter	Estimate (days)	Standard error	t value	P value
Approval lag (adjusted R 2 = 0.340, intercept = 1372.1, N = 235)
Global simulutaneous strategy	−1124.13	75.94039	−2.32	0.0214
Catch‐up strategy	−1024.129	169.571	−6.63	<0.0001
Immunotherapy drug	−452.2106	327.8876	2.84	0.0049
Priority review/orphan in FDA	−175.9423	183.2861	−5.59	<0.0001
Cytotoxic drug	424.09049	188.6693	−2.40	0.0174
Public knowledge‐based application	930.73948	161.5278	2.63	0.0092
Development start lag (adjusted R 2 = 0.364, intercept = 2873.0, N = 200)
Utilization of foreign clinical trial data	−1868.468	357.3676	−5.23	<0.0001
Immunotherapy drug	−799.0144	221.577	−2.96	0.0035
Global simultaneous strategy	−656.1854	189.8808	3.70	0.0003
Compounds originated overseas	390.63822	150.5595	2.59	0.0102
Bridging strategy	701.82619	269.2207	−2.97	0.0034
Cytotoxic drug	1044.2737	243.8281	4.28	<0.0001
Review time lag (adjusted R 2 = 0.383, intercept = 74.5, N = 233)
Orphan drug designation	−33.39841	0.011374	−2.94	0.0037
Number of patients in pivotal trial	−0.033428	0.134259	3.14	0.0019
Number of patients in Japanese clinical trials	0.4219466	30.39678	2.64	0.0089
Major cancer	31.425658	14.84498	2.50	0.0132
All‐case surveillance	37.104656	15.73582	−2.12	0.0349
Bridging strategy	80.301676	15.73417	2.00	0.0471
Hormonal drug	169.92339	49.92805	3.40	0.0008

Abbreviation: FDA, Food and Drug Administration.

Multiple regression analysis showed that “Compounds originated overseas,” “Bridging strategy,” and “Cytotoxic drug” significantly delayed the development start. Meanwhile, “Utilization of foreign clinical trial data,” “Immunotherapy drug,” and “Global simultaneous strategy” significantly accelerated the development start.

Factors that significantly shortened the review time lag included “Orphan drug designation” and “Number of patients in pivotal trial.”

DISCUSSION

This study compared Japan's drug lag with that of the US for anticancer drugs approved in Japan from 2001 to 2020, a very long period. We also showed the drug lag and development trends of anticancer drugs over a 20‐year period.

The number of anticancer drugs approved in Japan annually increased; the 36 approvals recorded in 2020 were the largest ever documented. Anticancer drug trends over the past 20 years showed that molecularly targeted drugs were the most common mode of action (142 approvals, 47.5%), with annual increases (Table 1, Figure 1). Refractory/relapse annually increased, although first‐line or naïve indications did not in terms of treatment lines (Figure 1). A bridging strategy was the most common from 2006 to 2015 in terms of a clinical development strategy; however, a global simultaneous strategy and catch‐up strategy showed increasing trends in recent years (Figure 1), accounting for 42.5% over the past 20 years (Table 1). In addition, the 246 approvals (82.3%) generated using overseas data annually increased (Table 1, Figure 1).

The drug lag between Japan and the US for anticancer drugs peaked in 2002 and then significantly decreased to less than 1 year. ¹² , ¹³ Our study showed that the lag was the shortest in 2018 and slightly increased in 2019 and 2020. Tachibana et al. reported that the 2017–2022 drug lag in oncology drugs was significantly higher than the 2011–2016 drug lag. ²⁰ Our study concurs with the findings of Tachibana et al., suggesting that the bottom of the drag lag was in 2018. The problem of drug loss was recently noted since drugs developed in the US were not developed in Japan. ²¹ “Drug lag” means that it takes time for a new drug to become available in Japan, and the country is behind foreign countries where the drug was first approved (the US or other countries). “Drug loss” is a recent hot topic in Japan. ²² , ²³ It is a situation wherein a drug that is already approved overseas (in the US or other countries) is not approved and unavailable in Japan. Recent data show that only approximately 60% of drugs approved in the US are approved in Japan, and the percentage of unapproved drugs is increasing every year. ²⁴ , ²⁵ , ²⁶ This is believed to be a consequence of pharmaceutical companies not having domestic legal entities in Japan or Japan branch offices. ²¹ We postulate that anticancer drugs developed and approved in Japan may have increased the drug lag, although further investigation is needed to confirm this.

There are no major differences regarding factors affecting the drug lag between Japan and the US. ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² International joint development methods such as the “Global simultaneous strategy” and “Catch‐up strategy” are new modes of action represented by “Immunotherapy drug” and “Priority review/orphan in FDA” expedited programs in the US that shorten the drug lag between Japan and the US. Ueno et al. examined the drug lag in drugs approved in Japan and the US over a 5‐year period from 2007 to 2012 and found that Japanese participation in global clinical trials reduced the drug lag. ³¹ Although the results of our study are limited to anticancer drugs, our study of drug lag over 20 years also showed that the international development approaches, the “Global simultaneous strategy,” and the “Catch‐up strategy” reduced the drug lag between Japan and the US. Additionally, by studying all drugs from 2008 to 2019, Tanaka et al. found that drugs with first approval in Japan reduced drug lag more than drugs with first approval in other countries (world). They also found that Japanese pharmaceutical companies were more likely to have first approval in Japan than in other foreign‐capital companies. ³² In our multiple regression analysis, we also considered Japanese companies as an explanatory variable, but the results were not significant. Our results might differ due to the analyses being limited to anticancer drugs; however, further studies will be needed. We also found that expedited programs in the FDA shortened the development period in the US for oncology drugs approved in Japan. Processes such as FDA Priority Review/Orphan designation have a particular impact on oncology drugs that influences their expedited approval by the FDA. ³³ , ³⁴ Interestingly, these processes have also influenced the reduction of drug lag in Japan. FDA Priority Review/Orphan designation had an impact on this study; however, Breakthrough Therapy designation and Accelerated Approval did not significantly help reduce drug lag. The impact was previously reported ²⁷ and will require further elucidation. In Japan, the Sakigake designation system was started in 2015 ³⁵ to obtain approval earlier than in other countries, such as the US. However, our study showed that the system did not help shorten the drug lag up to 2020. In addition, this study did not consider the economic impact, such as the 2008 financial recession and other factors. However, it is difficult to conclude that there was a significant impact from the annual drug lag results (Figure 3). Immune checkpoint inhibitors generally have indications across multiple cancer therapeutics. ³⁶ Therefore, they may play a significant role in shortening drug lag. We experimentally analyzed the approval lag between Japan and the US for immunotherapy drugs (n = 37). The median approval lag [IQR] of these drugs was 197.5 days (6.6 months) [132.3–447.5 days] from 2001 to 2020, and the mean value was 170.5 days (5.7 months) ± 650.3 days (data not shown). The overall approval lag from 2002 to 2020 was 498 days (16.4 months) [181.5–1302.0 days], and the mean value was 901.9 days (30.1 months) ± 1282.6 days, with a possibly short drug lag. This study did not consider the economic impact such as the 2008 financial recession and other factors; however, it is difficult to conclude that there was a significant impact from the annual drug lag results (Figure 3).

This study has some limitations. It targeted anticancer drugs approved in Japan; thus, anticancer drugs approved only in the US were excluded. Therefore, we did not investigate drug loss. Furthermore, we did not include anticancer drugs for which approval was withdrawn or development was discontinued. We discussed the drug lag between Japan and the US, although approval information was not examined for other countries and the European Union. We experimentally used multiple regression analysis to investigate the factors contributing to drug lag. We have also examined considerable explanatory variables that might affect the drug lag of anticancer drugs in Japan based on previous studies ² , ³ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ and other considerable factors; however, there might be other important potential factors.

CONCLUSION

The drug lag for anticancer drugs between Japan and the US peaked in 2002, after which it sharply declined to less than 1 year, and its lowest value was observed in 2018.

AUTHOR CONTRIBUTIONS

H.M., A.H., and M.O. wrote the manuscript. H.M., A.H., and Y. M. designed the research. H.M., A.H., M.O., R.S., and T.M. performed the research. A.H., A.H., M.O., R.S., and Y.M. analyzed the data. H.M. contributed new reagents/analytical tools.

FUNDING INFORMATION

This study was partially funded by a grant from JSPS KAKENHI with grant numbers JP20K20251 and 23K11940 (Dr. Maeda).

CONFLICT OF INTEREST STATEMENT

The authors declared no competing interests regarding this work.

Supporting information

Figure S1

Figure S2

Table S1

Table S2

Maeda H, Hara A, Ofuchi M, Shingai R, Misumi T, Murai Y. Trends in oncology drug lags in Japan from 2001 to 2020: A cross‐sectional study. Clin Transl Sci. 2023;16:2665‐2674. doi: 10.1111/cts.13660

DATA AVAILABILITY STATEMENT

Data are available on request from the researchers after approval by the corresponding author.