Cross-comparison of cancer drug approvals at three international regulatory agencies

Background The primary objective of the present study was to examine the drug approval process and the time to approval (tta) for cancer drugs by 3 major international regulatory bodies—Health Canada, the U.S. Food and Drug Administration (fda), and the European Medicines Agency (ema)—and to explore differences in the drug approval processes that might contribute to any disparities. Methods The publicly available Health Canada Drug Product Database was surveyed for all marketed antineoplastic agents approved between 1 January 2005 and 1 June 2013. For the resulting set of cancer drugs, public records of sponsor submission and approval dates by Health Canada, the fda, and the ema were obtained. Results Overall, the tta for the 37 antineoplastic agents that met the study criteria was significantly less for the fda than for the ema (X̄ = 6.7 months, p < 0.001) or for Health Canada (X̄ = 6.4 months, p < 0.001). The tta was not significantly different for Health Canada and the ema (X̄ = 0.65 months, p = 0.89). An analysis of the review processes demonstrated that the primary reason for the identified discrepancies in tta was the disparate use of accelerated approval mechanisms. Summary In the present study, we systematically compared cancer drug approvals at 3 international regulatory bodies. The differences in tta reflect several important considerations in the regulatory framework of cancer drug approvals. Those findings warrant an enhanced dialogue between clinicians and government agencies to understand opportunities and challenges in the current approval processes and to work toward balancing drug safety with timely access.


INTRODUCTION
With the recent evolution of cancer therapy paradigms from pan-cytotoxic therapies toward targeted agents, treatment outcomes for cancer patients have been expected to improve.However, the rate of molecular and genomic advancement in cancer research appears to outpace the processes of regulatory bodies to make new therapies available to patients 1 .Consequently, a salient aspect of cancer drug access is timely approval of drugs that have the potential to improve the clinical course of disease.
Previous studies have demonstrated that differences in outcomes of approval processes by regulatory bodies result in clinically relevant disparities in drug access on an international scale 2 .The relative pace of drug approvals across the 3 main regulatory bodies-Health Canada, the U.S. Food and Drug Administration (fda), and the European Medicines Agency (ema)-can be attributed to a multitude of factors, including specific priorities of the agencies and complexities in regional legislation and processes, among other considerations.However, predictors of regulatory outcomes have not yet been identified.Accordingly, in the present study, we performed the first drug-by-drug analysis of cancer drug approvals by those regulatory bodies, and we dissect the major factors contributing to international disparities in drug approval times.

METHODS
The publicly available Health Canada Drug Product Database (http://www.hc-sc.gc.ca/dhp-mps/prodpharma/ databasdon/index-eng.php) was surveyed for all marketed therapies with the class designation "antineoplastics" approved between 1 January 2005 and 1 June 2013.For the 37 new antineoplastic agents that met the study criteria, the approval dates and the original submission filing dates were extracted from the Summary Basis of Decision documents issued by Health Canada, where available.Similar data about submission and approval dates for the same drugs were obtained for the fda from public fda approval letters and from the ema's Authorization Details and the European Public Assessment Report for each drug.Dates of submission and approval include only the first indication for which the drug was approved and not supplemental submissions or additional approvals.Only active treatment drugs were surveyed and not drugs related to supportive oncology care.
In the statistical analyses, 2-tailed t-tests were used to compare the time to drug approval between two agencies; analysis of variance was used for comparisons involving all 3 agencies.

RESULTS
To facilitate this comparative analysis of times from initial drug submission to approval by each regulatory agency, the period from the filing of a submission by a sponsor until the approval for marketing was granted was evaluated.On average, the time to approval (tta) is approximately 14.0 months for Health Canada and 14.2 months for the ema; it is 6.9 months for the fda (Tables i and ii).
Of the identified drugs assessed by all 3 agencies, cabazitaxel had the shortest tta: only 17 days at the fda.The fda approved cabazitaxel for use in combination with prednisone for the treatment of patients with metastatic hormone-refractory prostate cancer previously treated with a docetaxel-containing regimen.Cabazitaxel was reviewed under the fda's priority review program, designed to expedite the review process for drugs that might offer breakthroughs in treatment (fda press announcement, 17 June 2010).In Canada and Europe, the tta for cabazitaxel was just under 1 year (11.63 and 11.03 months respectively).
The overall tta for the drugs analyzed in the present study was significantly less at the fda than at Health Canada (X ̄ = 6.4 months, p < 0.001); the overall tta at Health Canada and at the ema did not significantly differ (X ̄ = 0.65 months, p = 0.89; Figure 1).As anticipated, the overall tta was also significantly less for the fda than for the ema (X ̄ = 6.7 months, p < 0.001; Figure 1).
It is important to note that, in some instances, approval data were available only for 1 or 2 of the agencies.Differences in ttas were calculated only for drugs that were reviewed by all 3 agencies.Importantly, of the drugs surveyed, 30 of 37 underwent an expedited approval at the fda (Table iii).The differences in drug approvals identified here might therefore be largely related to the disparate use of accelerated drug approval mechanisms by the 3 regulatory agencies.However, the median ttas for the 7 drugs that did not undergo priority review at the fda were still lower than the median ttas for the same drugs at Health Canada and the ema (median: 10.1 months fda, 17.7 months Health Canada, and 15.5 months ema).
Finally, drugs are often filed for fda review before they are submitted for approval at Health Canada or the ema (Table iv).The mean time from fda submission to submission at an additional regulatory body was 28.4 months for Health Canada and 12.9 months for the ema.

DISCUSSION
The present study highlights variation in the processes involved in new oncology drug submissions to the main regulatory agencies and differences in the ttas subsequently observed, to which use of expedited drug approvals might be the most significant contributor.Close examination of differences between the regulatory frameworks can provide insight into the varying drug approval times identified in the present study and in previous literature.Consistent with the findings in the present study, a previous report of the regulatory review of novel therapeutics by Health Canada, the ema, and the fda also noted that, on average, drug applications are reviewed more quickly by the fda than by the ema or Health Canada 3 .The findings in the present study are also consistent with other reports indicating that access to new cancer drugs is associated with greater use of expedited review procedures in the United States than in Europe 4,5 .
Before accelerated approval mechanisms can be implemented in a meaningful way, a number of issues have to be addressed, including data quality, completeness, and clear guidelines for post-marketing surveillance of drugs approved through such pathways 6 .In the present study, the drug with the shortest tta was cabazitaxel.The time from final sponsor submission to fda approval was only 17 days.On closer inspection, that tta was a consequence of the fda's rolling review of the application, meaning that the sponsor was permitted to submit data to the fda as it gradually accumulated during the development process 7 .The application for cabazitaxel was eventually approved on the basis of one randomized open-label trial of 755 patients, which demonstrated an overall survival advantage of 2.4 months and also reported cabazitaxel-associated deaths.The approval was contingent on the requirement of the sponsor to complete 6 essential post-marketing studies pertaining to sustained demonstration of efficacy, pharmacokinetics, safety, and toxicity.
Although expedient drug approvals are needed to deliver drugs to patients sooner, it is essential to balance the approval pace with assurance that the drugs are sufficiently safe.After accelerated approvals, the fda has demonstrated a much higher than expected rate of label revision, suggesting that the rigour of the process has to be revisited 8 .Additionally, one thorough report of accelerated approval of cancer drugs by the fda demonstrated that post-approval black-box warnings were added to the labels of 4 oncology drugs (17%) that received accelerated approval and 2 oncology drugs (9%) that received regular approval 9 .Recognizing that the fda should raise its standards for granting accelerated approval to experimental cancer drugs, the Oncologic Drugs Advisory Committee in 2011 reached a consensus that the fda must, for more definitive demonstration of efficacy, require that at least 2 controlled trials be actively under way 10 .
In an effort to improve patient outcomes, concern about safety standards should not be dissipated.It is clear that vigilance in ascertaining clinical benefit in postmarketing studies is an essential cornerstone of successful accelerated approval processes 11 .

Limitations
Some limitations of our study should be noted.First, the analysis was not designed to identify cancer drugs that were not approved in Canada, perhaps for valid reasons.For example, ponatinib, which was approved by the fda in December 2012 through priority review and by the ema in July 2013, has not yet been approved by Health Canada.Another salient consideration not addressed in the study is the fact that approval times are not the only dimension to drug access.Regulation of drug costs and coverage on regional levels constitute another important aspect of cancer drug access.

SUMMARY
Understanding the pace of cancer drug approvals, as well as the underlying factors, is a necessary dimension of the continuum of cancer drug access.The findings of the present study contribute to the published evidence that ongoing monitoring and inquiry into international cancer drug approval times is essential.Faster drug approval times do not always translate into a direct path to safe therapies and drug access.A global discussion about the methods and criteria for fast-track approvals is needed.We anticipate that this cross-comparison of drug approval times in Canada, the United States, and Europe can enhance the existing dialogue between clinicians and government agencies to understand the deficiencies and strengths in the various approval models and to work toward improving them.

FIGURE 1
FIGURE 1 Spectrum of drug approval times at the U.S. Food and Drug Administration (FDA), Health Canada (HC), and the European Medicines Agency (EMA).Each box on the horizontal axis represents the set of times to approval (TTAs) for all drugs surveyed in the present study.The vertical axis depicts the TTA in months.Horizontal bars in each box correspond to the median TTA for the respective agency.Circles outside the box and the whiskers of each box plot denote outliers that are not within the 95th percentile of the other values in the dataset.**Statistically significant difference in TTA between the FDA and HC, and between the FDA and the EMA (p < 0.001).
cell acute lymphoblastic leukemia, T-cell lymphoblastic lymphoma When disease has not responded or has relapsed after treatment with at least 2 Accelerated-phase Philadelphia chromosome-positive (Ph+) disease in adults resistant or tolerant to at least 1 prior therapy with imatinib (as of 9 June 2011 was also approved for the treatment of newly diagnosed Ph+ disease in chronic Metastatic clear cell disease who have received no prior therapy or prior therapy with cytokines for metastatic disease in patients who have not received prior anti-HER2 therapy or chemotherapy; in combination with trastuzumab and disease in patients who have been previously treated with fluoropyrimidine-based chemotherapy, oxaliplatin, irinotecan, and an anti-VEGF therapy, and, if wild-type KRAS, with antilymphoma Low-grade CD20-positive relapsed or refractory, follicular or transformed disease, including in patients refractory to rituximab 825 Trabectedin Intravenous Ovarian cancer Sarcoma In combination with pegylated liposomal doxorubicin hydrochloride in patients with platinum-sensitive disease

TABLE I
Parameters derived from a comparison of the time from an initial drug submission by a pharmaceutical company to the date of approval for marketing in months, by regulatory

body Parameter Time to approval (months) Health Canada FDA EMA
aValues in boldface type are statistically significant by two-tailed t-test.NS = nonsignificant.

TABLE IV
Time in months between submissions made by the U.S. Food and Drug Administration (FDA) and submissions made by Health Canada and the European Medicines Agency (EMA)