FDA buries the lede on a key benefit of multiregional clinical trials: More trustworthy data
Life Sciences
| By Alexander Gaffney, MS, RAC
Earlier this month, the FDA published a long-awaited draft guidance document on a topic long top-of-mind for key oncology regulators: Companies should be using so-called multiregional clinical trials (MRCTs) in oncology development to ensure clinical data are more representative of the U.S. patients who will ultimately receive the products. However, the guidance fails to mention one of the best cases for MRCTs: Their ability to build trust and confidence between sponsors and regulators by reducing the risk of fraudulent or incorrect clinical trial data.
The design of each clinical trial for a new medicine is reflective of the questions a sponsor is trying to answer – and the constraints it is under
- Those constraints often lead to tradeoffs. While a perfectly designed clinical trial might involve thousands of patients enrolled in trial sites across dozens of different countries operating under a flawless trial protocol, in practice trial sponsors must often make tradeoffs to make the trial realistic from both an enrollment perspective and a cost perspective.
- These tradeoffs are often about ensuring one thing: Spending less on the trial to obtain the evidence necessary to obtain FDA approval more quickly. This makes a certain amount of intuitive sense. Consider two options: In Option A, a sponsor spends $50 million and three years to obtain evidence necessary to obtain FDA approval for a new drug; in Option B, the sponsor spends $100 million and five years for the same approval.
- Given these two options, most companies would likely choose Option A – the cheaper, faster option.
- There are still scenarios in which a company might logically choose Option B – the more expensive, slower option. These might include when the additional time and expense might result in even better odds for approval, or allow a sponsor to obtain an even more expansive indication which allows it to dominate a market segment.
- And there is another important reason why companies might choose Option B: Because they believe that while regulators might approve a drug based on the evidence collected right now, the company has reason to believe the FDA might change its stance in the near future and demand additional evidence that would cause it to need to re-do its study (or conduct an additional study), causing it to spend even more time – and potentially even more money as well.
- Reading the regulatory tea leaves is an essential activity that most life sciences companies large and small engage in for this very reason. Because developing a drug is a profoundly expensive pursuit, the costs of failure can be enormous. A good, well-reasoned strategy is therefore essential.
We mention all this because of a new draft guidance document published by the FDA that could easily be titled: Why you should spend more money on more complex clinical trials for oncology drugs
- The actual guidance, Considerations for Generating Clinical Evidence from Oncology Multiregional Clinical Development Programs, is focused on “improving the evidence obtained from one or more multiregional clinical trials (MRCTs) intended to support a marketing application” for an oncology drug in order to help “facilitate FDA’s assessment of applicability of the data to the U.S. population with the cancer being investigated and to U.S. medical practice.”
- In the guidance, the FDA explains its problem : “In oncology MRCTs, there has been decreasing proportion of U.S. participants included in these trials; this can limit the assessment of treatment effect consistency between U.S. enrolled participants and the effect observed for the overall study population in the MRCT. Additionally, the distribution of demographic characteristics or clinical characteristics of participants enrolled in these trials may differ significantly from the U.S. population such that foreign data may not be appropriate to support an FDA regulatory decision. These and other factors can impact FDA’s assessment of the generalizability and applicability of the results from such trials to the U.S. population or to U.S. medical practice.”
- Put differently: An increasing number of sponsors have wanted to decrease clinical trials costs and increase the speed at which they could obtain data. To do this, they sought to obtain evidence from clinical trials located outside of the United States, and sometimes in concentrated ways resulting in trial demographics that were not representative of the demographics of the United States.
- This isn’t the first time we’ve heard this concern from the FDA. In an interview earlier this year, AgencyIQ asked RICK PAZDUR, the Director of FDA’s Oncology Center of Excellence to comment on how his agency viewed foreign clinical data, and in particular data from oncology trials run in China following the agency’s decision not to approve sintilimab, a drug with clinical data exclusively from China.
- As Pazdur explained: “It is not about China. It is not about China bashing. It is not anti-Chinese patients. It’s not about the Chinese government, etc. It’s really about what we want for drug development on a global basis, and that is multi-regional trials with all of the countries participating in this. Having a multi-regional trial brings in countries that we don’t have regulatory experience with. For example, we don’t have a lot of regulatory experience with clinical sites in China. So bringing them into a multi-regional trial, we will then take a look at this data and interrogate it, okay, because we get the raw data. How did patients do in China in all of these different sites compared to North America? What was the safety reporting? What was the results of response rates, survival, analysis, et cetera, in these different areas? Looking for consistency, not necessarily differences here. So, it builds that confidence.”
- Confidence is a key theme in FDA’s new draft guidance. While the specific word (“confidence”) doesn’t appear, FDA speaks of the importance of “generating the data necessary to evaluate the trial’s results in the context of U.S. patients who have the cancer for which the drugs are being developed.” And while FDA also concedes that “not all clinical development programs may be appropriate to conduct in a multiregional fashion,” trials that are appropriate may benefit, including by facilitating the confidence of international regulators, who would be more likely to “enable parallel marketing application submissions to global regulatory authorities and expedite access to innovative drugs for all.”
- The document alludes to key regulatory themes and trends we have heard from FDA in recent years, including the benefits of MRCTs to ensuring study diversity. For example, FDA writes that it “recognizes that a thoughtfully designed multiregional [clinical development program] CDP can facilitate enrollment of foreign participants with diverse genetic ancestral backgrounds (e.g., Indigenous American, African, Asian, and European ancestries), that may be scientifically and clinically relevant to the U.S. population.”
- It also notes that it “may be acceptable for a sponsor to conduct an MRCT that will enroll a substantial proportion of participants in a single foreign geographical region” as long as the sponsor plans to conduct an additional pivotal trial “that will enroll a population that is representative of the U.S. population.”
But there’s one important point that we were surprised FDA didn’t mention more prominently in its guidance: Data integrity
- Between the aforementioned two different approaches – Options A and B – there are some additional reasons why a sponsor might want to choose a slower, more expensive option. One of those reasons could be to act as a sort of insurance against problems or fraud. A quicker, cheaper option may be the result of a trial site cutting corners or falsifying data, for example. Conversely, a more thorough, expensive and diversified option makes it easier to detect potential fraud and reduce the potential for a single trial site to derail an entire clinical development program.
- This is far from a hypothetical concern. Due to the cost and complexity of clinical development, some companies and individuals have been known to try to take shortcuts, including falsifying clinical data. Because the FDA does not itself run clinical trials, it necessasrily relies on submitted data to come to conclusions about the safety and efficacy of drug products.
- To help guard against fraud, the FDA engages in Bioresearch Monitoring (BIMO), primarily run through its Office of Regulatory Affairs (ORA) with oversight from FDA’s various product centers. The FDA conducts hundreds of BIMO inspections of study sites each year, with inspectors typically looking for information that would indicate that the investigational plan wasn’t followed, study records weren’t kept, inaccurate data were recorded, patients did not consent to the research, the product wasn’t properly controlled, or adverse events weren’t properly reported.
- While the overwhelming majority of clinical investigator inspections under the BIMO program find no serious issues, about 1% of all inspections conducted in 2023 resulted in an Official Action Indicated (OAI) finding, and another 20% resulted in FDA advising voluntary changes. Looking at the data another way, for sponsors and contract research organizations, there were just 18 OAI findings between 2017 and 2023. he FDA and Department of Justice have been known to prosecute people and companies for violating clinical practice regulations, with several convictions in just the last few years alone.
- But this high rate of compliance is potentially misleading. The FDA had only 109 dedicated BIMO inspectors as of FY 2023, according to a Government Accountability Office (GAO) report, and the agency has long had a problem with conducting international inspections of both manufacturing and trial sites. For example, its June 2024 guidance document on BIMO inspections notes that FDA will commonly pre-announce its inspection of a foreign trial site before the inspection takes place. That can make it easier for companies to hide or clean up problematic data.
- While we weren’t able to find any data on the number of FDA BIMO inspections by country, we are aware of concerns that FDA has recently raised about fraudulent data coming from foreign research organizations. For example, FDA’s Center for Devices and Radiological Health (CDRH) has raised concerns about “fraudulent and unreliable laboratory testing data in premarket submissions” that had been generated by Chinese and Indian facilities, and said it was part of an “increase in submissions containing unreliable data.” An Indian contract research organization called Synapse Labs was found to have likely falsified clinical and bioanalytical studies conducted on behalf of sponsors.
- FDA’s guidance does briefly mention one factor implicitly related to data integrity. As it explains: “Sponsors should consider whether individual clinical trial sites have a record for complying with applicable laws and regulations covering good clinical practice.” However, the context of the remark is more related to the experience of the trial site (also important), rather than an explicit commentary on data-related shenanigans. Looping back to the question of BIMO inspections, the guidance also reminds sponsors that trial sites are “subject to onsite inspections should FDA deem such inspections to be necessary.”
FDA may wish to consider making this point about data integrity more prominent, because it’s also a potential key benefit to both sponsors and regulators
- For a sponsor considering a multiregional clinical trial, one understated benefit is that the trial is less susceptible to a finding or concern about data integrity issues at any one trial site. As demonstrated by the recent medical device laboratory testing and Synapse Labs case examples, if the FDA finds evidence of fraud in clinical development, the onus is then immediately placed on the sponsor to either re-do the study in its entirety or provide convincing evidence of the effects of the products that does not rely upon the suspect data.
- Because data fraud can occur without the knowledge or intent of the sponsor, an MRCT provides the sponsor with a sort of insurance for data integrity. When trials are multiregional, the potential for every trial site to be contaminated by fraud is significantly lessened; further, sponsors have enhanced ability to detect anomalous, and possibly fraudulent, data.
- That latter point is also important to the FDA. More trial sites make it easier for regulators to detect outlier data which may be indicative of fraud. The essence of fraud is delivering a result with the intent of masking false data as believable and reliable information. In the case of clinical trials, if a company concentrates all of its clinical development efforts into a single country (and just a handful of trial sites), it is comparatively more difficult to detect implausible data because of the relatively small sample size. However, if data are collected across multiple countries and dozens (or more) trial sites, fraudulent data are more likely to become evident.
- It is worth noting that FDA has recently received new authority that could also aid in its ability to crack down on trial-related fraud. As part of the passage of the Consolidated Appropriations Act of 2023, FDA received authority to request records in advance of or in lieu of a BIMO inspection – authority it has long had for drug manufacturing inspections. This could help it to collect additional data from international trial sites, making it still more critical for companies to pursue a diversified strategy.
- FDA may also find this argument resonates with policymakers. In effect: Better trials can be better for national security and patients, reducing the opportunity for certain countries’ trial sites to engage in fraud beyond U.S. borders.
To contact the author of this analysis, please email Alexander Gaffney ( agaffney@agencyiq.com)
To contact the editors of this analysis, please email Laura DiAngelo ( ldiangelo@agencyiq.com) or Kari Oakes ( koakes@agencyiq.com).