FDA is preparing to review an estimated 25 applications for cancer drugs that are being developed based largely—and in many cases entirely—on data from mainland China.
The first of these applications will be considered by the agency’s Oncologic Drugs Advisory Committee Feb. 10. The committee will consider a trial called ORIENT-11, which compares chemotherapy plus sintilimab, a PD-1 monoclonal antibody checkpoint inhibitor, to chemotherapy alone as initial treatment for metastatic non-small cell lung cancer.
The biologics license application for the agent is sponsored by Innovent Biologics (Suzhou) Co., Ltd., which has a global collaboration with Eli Lilly & Co.
As this complex story unfolds, it’s important to keep in mind the rules, strategies, and traditions that will determine the action. The application could be going to ODAC for many reasons, but it’s a safe guess that chief among them is the agency’s desire to spell out the rules in full view of the public.
Richard Pazdur, director of FDA’s Oncology Center of Excellence, has been vocal in academic medical literature about his concerns regarding “me-too” drugs coming from China and the fundamental challenges they pose to the agency’s drug approval criteria.
To lay out this problem, Pazdur and co-authors recently published a paper in The New England Journal of Medicine and sat down for an interview with The Cancer Letter.
“Many of the sponsors performing trials in China did not seek established milestone meetings with the FDA prior to embarking on their regulatory strategies. We need to have a clear understanding why these meetings obtaining our input were not held,” Pazdur said to The Cancer Letter.
We need to have a clear understanding why these meetings obtaining our input were not held.
Richard Pazdur
Sponsors of these PD-1 and PD-L1 drugs say their goals in developing these agents include competing on price with the pharmaceutical companies that sponsored the first wave of the drugs.
However, discussion of price will not be a part of the ODAC agenda on Feb. 10, or at any point in the future. FDA is precluded from considering drug pricing in its approval decisions. Drugs are approved based on their safety, efficacy, and applicability to the U.S. population when considering foreign data.
This is important because, based on FDA descriptions, the drugs coming from China are strictly “me-too,” and their principal advantage is that they may cost less. All of these drugs are going through the BLA approval mechanism, as opposed to the biosimilars mechanism, which opens only after an innovator agent loses exclusivity.
“Most drugs being presently discussed with the FDA are attempts to replicate known advances,” Pazdur said to The Cancer Letter. “Please remember that these drugs are not biosimilar or generic drugs and, hence, are required to have complete clinical and non-clinical studies and manufacturing processes developed.
“Because other checkpoint inhibitors were approved in the indication, we probably would have recommended a trial comparing the new agent to the approved checkpoint inhibitor in the disease, especially in indications where a survival improvement has been demonstrated,” Pazdur said.
It appears that accelerated approval—a pathway where indications hit the market based on intermediate endpoints and receive a full approval after demonstrating patient benefit—may not apply in this setting, FDA officials indicated.
Accelerated approvals can be given in situations where an agent fills an unmet medical need, and many of the drugs in the wave of applications address indications where PD-1 drugs showing a benefit are available.
“Many of the applications we’re seeing that rely on clinical data from China are similar to previously conducted multiregional clinical trials that led to U.S. approval,” Harpreet Singh, director, Division of Oncology 2, said to The Cancer Letter. “So, they don’t fulfill an unmet need in the U.S.”
While many of the drugs being developed in China are “me-too” checkpoint inhibitors, some are, in fact, innovative, and several are focused on indications that are more common in Asia, including, for example, nasopharyngeal cancers, Pazdur said.
Such situations “may warrant a greater degree of regulatory flexibility, but nevertheless require confidence in the generalizability of the data to the U.S. and data integrity,” Pazdur said. “This may be accomplished with more restrictive indications or request for additional trials conducted in the U.S.”
On Feb. 10, ODAC will in effect focus on whether clinical data from one population—patients in mainland China—can be generalized to another—the population of the U.S. This discussion comes at a time when pharma companies and public agencies seek to make clinical trials more inclusive, both in the name of health equity and in order to make the results of clinical trials applicable to broader populations.
Variation in outcomes in China and the U.S. could be caused by standards of care, lifestyle factors, and genetic factors. Pazdur and Singh noted that in the past, regulators in several Asian countries raised questions about the applicability of data from trials conducted in Europe and the U.S. Asian countries addressed these questions through “bridging studies” in indications where there were unmet medical needs.
FDA has encountered issues related to ethnic and regional characteristics before.
The arc of the AstraZeneca drug Iressa (gefitinib) provides a classic example of the differences in efficacy in relation to diversity in population characteristics (The Cancer Letter, May 9, 2003; April 30, 2004; Feb. 11, 2011, July 17, 2015; Aug. 7, 2015).
Said Singh:
“Many people recall the story of gefitinib. It was actually the first EFGR-targeted therapy to be approved for the treatment of NSCLC, but the understanding of the biological role of the target and patient population was limited during the early stages of development.
“The drug was initially approved by the FDA in 2003 based on response rates in an unselected population, however when the confirmatory trial failed to show a statistically significant benefit in overall survival, its use was restricted by the FDA and the company withdrew their application.
“Greater benefit for gefitinib was observed in patients of Asian origin and never smokers, suggesting the potential for improved patient selection based on clinical characteristics, which was explored in conjunction with tumor biology studies. Subsequent trials conducted in Asia showed the presence of an EGFR mutation was the strongest predictor of a more favorable outcome.
“We now understand that the drug is more effective in Asian women who never smoked, and those who harbor an EGFR mutation.
“I think this is a classic example of how intrinsic and extrinsic factors like gender, ethnicity, smoking status, and tumor biology influence patient selection and drug efficacy.”
Pazdur and Singh spoke with Paul Goldberg, editor and publisher of The Cancer Letter.
Paul Goldberg: Let’s talk about the upcoming ODAC meeting on Feb. 10. It involves another PD-1 therapy and a clinical trial conducted entirely in China. Can you tell me more about that?
Harpreet Singh: That’s correct. The committee will discuss ORIENT-11, a clinical trial comparing chemotherapy plus sintilimab, a PD-1 monoclonal antibody checkpoint inhibitor, to chemotherapy alone as initial treatment for metastatic non-small cell lung cancer.
That sounds familiar. Have trials like this been conducted before?
HS: Yes. In fact, the trial design and enrollment criteria very closely resemble landmark trials which established immune checkpoint inhibitors as part of a front-line regimen for lung cancer. ORIENT-11 was conducted exclusively in mainland China, whereas the previous trials were international and multiregional.
That sounds like a first. How many other submissions may be coming with data only from China?
Richard Pazdur: We, of course, can’t speak to the specific issues regarding the application being discussed on Feb. 10, but we can discuss general comments on trials and thinking surrounding trials being submitted from a single country.
We know of at least 25 applications in oncology drug development, planned to be submitted or already submitted to FDA, that are based either solely or predominantly on clinical data from China. Many sponsors of trials performed in China began their clinical development in China after results of other checkpoint inhibitors in the disease were publicly available.
Therefore, their clinical development plans had much less risk than for those sponsors who pioneered checkpoint inhibitors in specific diseases. These sponsors obviously benefitted from this prior knowledge.
You wrote in your recent NEJM commentary about “The Wild West of Checkpoint Inhibitor Development,” and it seems the East is joining the fray. What’s the history of this?
RP: Oncology drug development has become increasingly international over many decades, emphasizing multiregional clinical trials with substantial U.S. enrollment.
With these large trials, we can examine both safety in efficacy in various regions and countries to ensure consistency of results. We want to emphasize that multiregional trials are the preferred route for international drug development—not data from a single country.
Thirty years ago, Asian countries desired additional trials prior to accepting data from the U.S. and Europe. This represented data moving from the “West to the East.” In 1998, the International Council for Harmonization developed a guidance, E5, providing recommendations on these additional trials that would “bridge” the data from Europe and North America.
The U.S., European countries, and Japan signed on to this guidance. These “bridging trials” were conducted in specific countries and were usually based on response rate or pharmacologic endpoints, with the goal of bringing innovation to Asia.
HS: As it happened, because Asian countries were underrepresented in multiregional clinical trials and had to rely on these bridging trials after the larger trials were complete, there was a long lag time in new drugs reaching patients in those countries.
Is that still the case? I’m not an expert on Asia but the world has changed a lot in 30 years.
RP: Many countries in the region, for example, Singapore, Taiwan, South Korea, have rapidly adopted global standards for oncology practice, with improved infrastructure for clinical trials and medical oncology training. There are large, state-of-the-art cancer centers and academic medical centers.
With that increased capacity, over the past 20 years, many Asian countries have participated to a greater extent in multiregional clinical trials, obviating their reliance on bridging trials. These past bridging trials brought innovation to these countries.
The bridging trials allowed acceptance of Western data to the Asian countries to take advantage of important clinical advances developed in the West. This regulatory flexibility by the Asian regulatory authorities allowed patients to have earlier access to novel drugs without having to replicate complete clinical development programs.
The current situation with Chinese drugs, especially “me-too” PD-1/PD-L1 drugs, submitting applications to the U.S. is not the same situation. This current “East to West” movement of clinical data is certainly not about innovation.
Most drugs being presently discussed with the FDA are attempts to replicate known advances. Please remember that these drugs are not biosimilar or generic drugs and, hence, are required to have complete clinical and non-clinical studies and manufacturing processes developed.
These drugs cannot rely on information generated by the already-approved checkpoint inhibitors. There is a lot of redundancy and expense for questionable benefit.
But what about the trials from China? Will FDA require bridging studies to include U.S. patients?
HS: The ICH E5 guidance didn’t exactly envision this situation. It provides a framework for assessment, but it may not be adequate in this case. Do you mind if we nerd out a bit on that guidance?
Please do!
HS. Okay. So, the ICH E5 document tells you how to evaluate the acceptability and generalizability of clinical data from outside a region by evaluating intrinsic and extrinsic factors.
Could I ask you to back up and tell me what these intrinsic and extrinsic factors might be?
HS: Sure. Intrinsic factors include genetic and physiologic variables, such as polymorphisms, lean body mass, organ dysfunction, etiology, histologically and molecularly defined disease subtypes, and known determinants of response to therapy.
Extrinsic factors include medical practice, available therapies, supportive care medications, subsequent oncologic treatments, social and cultural factors such as diet, and herbal medications.
If these differences are small, the guidance states that there may be no further studies required, but larger differences may require bridging studies or controlled clinical trials.
RP: The E5 was written for a situation where regulators were looking at data from large trials in a heterogenous Western population and extrapolating to a less ethnically diverse population, usually in Japan or other Asian countries.
Now we’re looking at data from a trial from a single Asian country and considering how to extrapolate the trial to the heterogenous U.S. population with its multiple ethnic and racial constituents. This brings up its own set of issues, especially regarding well-publicized efforts to encourage diversity in registration trials over the past two years.
HS: It’s not that simple to use a bridging study to get adequate data to enable us to make an appropriate risk-benefit decision on the efficacy and safety of a drug for the U.S. population.
Have there been any examples demonstrating the impact of intrinsic and extrinsic factors?
HS: Many people recall the story of gefitinib. It was actually the first EGFR-targeted therapy to be approved for the treatment of NSCLC, but the understanding of the biological role of the target and patient population was limited during the early stages of development.
The drug was initially approved by the FDA in 2003 based on response rates in an unselected population, however when the confirmatory trial failed to show a statistically significant benefit in overall survival, its use was restricted by the FDA and the company withdrew their application.
Greater benefit for gefitinib was observed in patients of Asian origin and never smokers, suggesting the potential for improved patient selection based on clinical characteristics, which was explored in conjunction with tumor biology studies.
Subsequent trials conducted in Asia showed the presence of an EGFR mutation was the strongest predictor of a more favorable outcome. We now understand that the drug is more effective in Asian women who never smoked, and those who harbor an EGFR mutation.
I think this is a classic example of how intrinsic and extrinsic factors like gender, ethnicity, smoking status, and tumor biology influence patient selection and drug efficacy.
As an aside, is there an issue with data quality in trials from China?
HS: There was a report in the British Medical Journal in 2016 that pointed to data quality issues in 80% of trials conducted in China. The Chinese FDA stated that data from 1308 of the 1622 applications should be withdrawn because they contained fabricated, flawed, or inadequate data from clinical trials.
But that was five years ago, and we know there have been efforts to address these issues. In general, we would still want to explore whether any of the sites in a particular trial were involved in these withdrawals, and what was done to rectify the specific situation.
RP: Confidence in data quality is built into clinical trials—so called “quality by design.” Clinical site inspections are random and involve a small number of sites and have been a challenge in our current COVID pandemic.
First, this confidence can be established by having sites selected that have track records of contributing significant number of patients to international, multiregional trials that were submitted to regulatory agencies, particularly the FDA. As previously stated, multiregional trials provide regulators a data set to compare safety and efficacy between multiple regions to determine consistency and, if that consistency is not demonstrated, to determine underlying reasons for disparate results.
A second factor in establishing confidence is the individual clinical site’s past track record regarding data quality. If there have been challenges regarding data quality or withdrawal of clinical data because of data quality issues from specific sites these sites would require closer scrutiny and may render data from a site unacceptable.
We would have to have a thorough understanding of what occurred and specific measures to address these deficiencies.
Okay, so you have to look at differences between the China and U.S. patient populations, with data from a small trial in China, for a PD-1 drug that is similar to one already on the market in the U.S. for this indication. Is that the challenge you’re asking ODAC to consider on Feb. 10?
RP: I don’t want to get into too many specifics on an application until the ODAC meeting. We’ve been discussing this topic in general terms. What’s happening is a coming wave of similar applications and the question of acceptability of these trials.
I have expressed concern about the large number of PD-1 and PD-L1 drugs under development worldwide, estimated to be at least 33 drugs with over 2,000 clinical trials, without better collaboration. I’m concerned about redundant trials, lack of uniform biomarker development, too many “me-too” drugs, and the potential for less than optimal use of our most important resource for clinical trials—the confidence of our patients.
Many of the applications we’re seeing that rely on clinical data from China are similar to previously conducted multiregional clinical trials that led to U.S. approval. So, they don’t fulfill an unmet need in the U.S.
Harpreet Singh
In particular, when other drugs studied in multinational trials are already approved based on survival, we do not want to lose efficacy with a “me-too” drug studied from a single country where there may be questions on generalizability.
Not all of these drugs being developed in China have the same challenges. Many are “me-too” checkpoint inhibitors developed in indications with checkpoint inhibitors already approved in the U.S. Other may be innovative. Others address indications that are more common in Asia, such as nasopharyngeal cancers, where enrollment in the U.S. may be more difficult.
These latter two situations may warrant a greater degree of regulatory flexibility, but nevertheless require confidence in the generalizability of the data to the U.S. and data integrity.
This may be accomplished with more restrictive indications or request for additional trials conducted in the U.S.
HS: In effect, E5 encouraged Asian regulators to use regulatory flexibility to accept “West to East” bridging studies, rather than require another controlled clinical trial, for situations where there were no alternatives and the drug in question filled an unmet medical need. But many of the applications we’re seeing that rely on clinical data from China are similar to previously conducted multiregional clinical trials that led to U.S. approval.
So, they don’t fulfill an unmet need in the U.S.
RP: Many of the sponsors performing trials in China did not seek established milestone meetings with the FDA prior to embarking on their regulatory strategies. We need to have a clear understanding why these meetings obtaining our input were not held.
Because other checkpoint inhibitors were approved in the indication, we probably would have recommended a trial comparing the new agent to the approved checkpoint inhibitor in the disease, especially in indications where a survival improvement has been demonstrated.
We want to make sure we are not losing efficacy with a “me-too” drug. The sponsors of these trials in China feel justified in conducting the trials in China where there was a delay in approving the already U.S.-approved agent.
A superior approach would have been the involvement of China in the same large multiregional trials that led to U.S. drug approval and rapid submission to the Chinese regulatory authorities. We believe that all patients throughout the world would benefit with a more efficient regulatory approach and that was the genesis of OCE’s Project Orbis.
Project Orbis involves other regulatory agencies; correct?
HS: Yes, Project Orbis has the goal of concurrent approval of oncology drugs world-wide. This would set up a uniform standard of care not only immediately benefiting patients but also allowing rapid acceptance of novel therapies as control arms in subsequent trials quickly.
Is China interested in being part of Project Orbis?
HS: We have established meetings with China on a quarterly basis to discuss general regulatory issues in oncology—not specific applications. To be a member of Project Orbis, China would have to have confidentiality agreements with all members.
This is currently lacking. Also, China would have to accept the submission in English. The confidentiality challenge could potentially be “worked around” by having a limited agreement for a specific application and the Orbis members and commercial sponsor would have to agree.
We would hope that China moves in the direction toward more fully participating in international, multiregional trials and I believe their regulatory agencies share this desire.
China did join the ICH in 2017 as its eighth regulatory member, with a pledge to actively promote the faster entry of international drugs into the Chinese market.
The same year, the ICH issued the E17 guidance, which doubled down on the value of multiregional clinical trials, outlining concerns regarding population differences amongst regions. E17 provides a framework for addressing intrinsic and extrinsic differences in an exploratory phase prior to embarking on a multiregional international trial.
Would these drugs from China potentially address the high prices of drugs?
RP: FDA does not take “price” into consideration in making regulatory decisions. From a societal viewpoint, we need to address drug pricing head-on and not attempt to address the issue with convoluted approaches.
Drug pricing can rapidly change, and one does not have a guarantee of a price. With at least seven checkpoint inhibitors approved in the U.S., there has been little movement on price. Single-country trials raise many more questions, especially with regards to generalizability of their results to the U.S. population, data quality assurance, preservation of survival gains, than simply looking at this at a drug’s expense.
There has been no direct comparison of any of these checkpoint inhibitors to allow extrapolation of results from one drug to another. Each drug must stand alone on the data submitted.
Is there anything we’ve missed?
RP: At every major cancer meeting during the past two years there has been an expressed interest and commitment to ensure ethnic and racial diversity in clinical trials by all members of the cancer community.
Almost all major international drug companies have confirmed a commitment to this principle. Some of these same companies have conducted “China only” trials or are co-sponsors of these regulatory submissions to the FDA with Chinese counterparts.
I would call on them to reconcile their public comments on their commitment to racial and ethnic diversity in clinical trials with their actions in submitting applications that not only lack any consideration of U.S. ethnic and racial diversity, but completely lack any U.S. patients.
