Pazdur steers ODAC away from PD-1s tested only in China

Ahead of Tyvyt panel, FDA oncology chief pushes for multiregional trials, argues studies must represent the U.S. population

One week ahead of an advisory committee meeting for the first PD-1 inhibitor from a China company seeking U.S. approval, FDA’s oncology chief Richard Pazdur has forcefully outlined his concerns over trials conducted exclusively in China.

The ramifications extend well beyond PD-1s developed in China, including the first of these to come before FDA, Tyvyt sintilimab. FDA is signaling a preference for multiregional trials and placing heavy emphasis on clinical trial diversity that is not limited to oncology and could affect how sponsors construct trials in other regions of the world to support U.S. submissions. 

The decision by Pazdur, director of FDA’s Oncology Center of Excellence, to highlight the issue is part of a drive to meet FDA’s commitment to ensuring that clinical trials for drugs to be marketed in the U.S. represent the U.S. population.

“Single-country trials done in China or any other country that has a homogeneous population that is markedly different from the U.S. is antithetical to the concept of racial or ethnic diversity,” Pazdur told BioCentury. 

In a comment in The Lancet published Friday, Pazdur and Harpreet Singh, director of FDA’s Division of Oncology 2, argue for the use of multiregional clinical trials (MRCTs), and warn that the wave of oncology applications in progress that use only or mainly clinical data from China may not find favor with FDA if they merely recapitulate trials performed for drugs already approved and marketed in the U.S.

The comment notes that the Oncologic Drugs Advisory Committee (ODAC) will meet Feb. 10 to discuss the application for Tyvyt, an anti-PD-1 mAb from Innovent Biologics Inc. (HKEX:1801) that is partnered with  Eli Lilly and Co. (NYSE:LLY). The registrational trial, ORIENT-11, is in non-small cell lung cancer (NSCLC) and tests Tyvyt in combination with chemotherapy. Pazdur and Singh note that the trial is “done exclusively in China,” and that it closely resembles the “landmark NSCLC trials” that established the early PD-1 inhibitors (see Table 1).

On Lilly’s earnings call Thursday, Jacob Van Naarden, SVP and CEO of Lilly’s Loxo Oncology and president of Lilly Oncology, hinted that FDA’s review raised concerns about the lack of diversity in trials of Tyvyt and that the issue might be raised at the ODAC meeting. He said the risk/benefit of Tyvyt is supported by a “well-conducted” study. “We believe the results of the study are indeed applicable to a U.S. population, and we’ll make our case in that respect a week from today.”

Van Naarden said there may have been mixed or wrongly read signals on what would be required in the trial data. “We understand the stance of the agency may have changed, or maybe we misinterpreted it a few years ago, and so we’ll await the FDA’s presentation on the topic and the feedback from the ODAC members.”

The timing of Pazdur’s commentary suggests it is intended as a strong message for ODAC. FDA, however, is not bound to follow the committee’s recommendations.

Pazdur told BioCentury his concern is not about approval of a specific drug. He indicated that he is raising the issue of trials conducted solely or largely in one country because FDA expects a deluge of applications for PD-1s that have been developed using China-only trials. There are at least 25 applications from China in the pipeline, he said.   

In contrast, all the PD-1 inhibitors approved by FDA were tested in multiregional trials, said Pazdur. 

Even though multiregional trials may not properly represent different racial and ethnic groups, they are preferred over China-only trials where the patient population is highly homogeneous. “Lack of diversity versus no diversity are two different issues,” said Pazdur. “In multiregional trials, the numbers may not be what we like. It’s a different issue not having any diversity than a trial where we need to work on improvement of ethnic and racial minority groups.” 

Pazdur said FDA’s thinking has evolved since his comments at the 2019 American Association for Clinical Research (AACR) meeting, where he suggested companies could reduce their risk by running trials in China that mimicked the trials behind approved PD-1 mAbs. In remarks that were widely circulated in China, Pazdur said: “If I was developing a PD-1 in China and I wanted to go after the first-line lung cancer indication that Pfizer had, I would simply do that study in China and duplicate it.” He also said FDA would “accept only Chinese data” if the data were high quality.

“Since that meeting,” Pazdur told BioCentury, “there has been a huge change in the world due to COVID, and demand for ethnic and racial diversification of clinical trials, as well as congressional demands that we have better representation in clinical trials.”

He added that regulatory flexibility will depend on how innovative the drug is, and that the issue is primarily about where FDA wants the field to go. Moreover, it is not an anti-China position, but an effort to bring China innovators “into the fold” and enable them to be successful.

“We are not against China, Chinese companies or Chinese patients. We want them to participate in multiregional trials,” said Pazdur.

Limitations of bridging studies

Multiregional trials have been common practice for more than 20 years, but historically included few patients in Asian countries. FDA’s oncology division records that NDA submissions since 2010 include under 20% of patients from Asia, with less than about 5% from China.

Because of the low representation of Asian patients, Japan’s regulatory agency sought a way to add data that could extrapolate the results from the multiregional clinical trials, via bridging studies that presented clinical data relevant to its population and medical practices.

The goal was to “bring innovative therapies from the West to a homogeneous population” without needing to duplicate the entire portfolio, Pazdur told BioCentury,

The 1998 International Council for Harmonisation E5 guidance document outlined how such bridging studies should be performed.

But Pazdur and Singh state in their Comment that the influx of applications with trials using China-only or China-mainly patients “might have limitations not envisioned in E5.” Apart from the different make-up of the populations, the trials can’t be well-generalized to the U.S. population because of how they are constructed.

Specifically, “they are smaller, tend to be non-randomized, and rely on response rates or pharmacodynamic comparisons rather than the endpoint used in the MRCT, such as overall survival,” they wrote.

However, the criticism is not relevant for all China PD-1 studies. For example, Innovent’s ORIENT-11 trial is a randomized, double-blind Phase III study in about 400 patients. Its primary endpoint is progression-free survival, with OS a secondary outcome.

“Our degree of flexibility on the issue will be dictated by the degree of innovation the product offers.”

Richard Pazdur, FDA

The PD-1 situation today is different from Asian countries seeking to provide their patients access to drugs developed in the U.S. or Europe, Pazdur told BioCentury. “This is not about innovation. This is at best about me-too drugs,” which calls for a different degree of regulatory flexibility.

“That is much different from the situation envisioned originally with bridging studies, where you were thinking of bridging between heterogeneous populations in the West to homogeneous populations in the East with innovation. Here it is the opposite, going from a homogeneous population to a heterogeneous population, and these drugs for the most part are not innovative.”

Pazdur added that how regulators consider China-only data, and the degree of flexibility they exercise, will depend on the compound in question. “If there are applications for innovative products, we might take a much different viewpoint,” he said. He added: “Our degree of flexibility on the issue will be dictated by the degree of innovation the product offers.”

It may also depend on the disease prevalence in the U.S. compared with a foreign region, he said. An application for a disease more common in Asia such as hepatocellular carcinoma or nasopharyngeal cancer may be afforded more flexibility, especially if there is an improvement in overall survival.

Multiregional push

Pazdur’s push for multiregional trials is based on his view — and, he says, international consensus — that they are better for drug development.

The ICH provided an additional document, E17, on multiregional clinical trials in 2017 because of growing agreement that international collaboration is preferred over single-country trials. The reasoning is that the trials avoid duplication, allow earlier access to innovation, enhance infrastructure development, and are a faster route to establishing new standards of care that can influence further drug development.

E17 advises an exploratory phase before a multiregional clinical trial to iron out issues related to intrinsic and extrinsic factors that can differ in specific countries. Intrinsic factors are those related to an individual’s characteristics — such as genetic and physiological variables — whereas extrinsic factors are related to the specific region, such as the medical practice there, available therapies and social and cultural factors, including diet.

Multiregional clinical trials seek to use a single prospectively planned analysis to evaluate a treatment effect across an entire population, including patients from diverse national, ethnic and racial backgrounds. Encompassing different intrinsic and extrinsic factors, they allow investigation of treatment effects among different groups in a single trial, rather than trying to glean the information from cross-trial comparisons of studies.

Though the trials aren’t usually statistically powered to formally test different responses among races or minority ethnic groups, by pooling analyses from regions with similar intrinsic and extrinsic features, they can yield results that serve as a starting point for hypothesis generation and further drug development.

Multiregional clinical trials are in line with the goals of FDA’s Project Orbis, which aims to enable simultaneous regulatory submissions in different countries to expedite access to novel therapies. Project Orbis is a partnership of FDA and six other regulatory authorities that enables concurrent review of select clinically significant cancer products for approval in their respective countries.

Pazdur argues that embracing MRCTs would benefit everyone.

He is also critical of sponsors that don’t seek FDA advice during development.

Pazdur speculated that some Chinese companies planning to apply for FDA approval of PD-1 inhibitors have not solicited the agency’s advice because they know the agency will recommend comparative trials that test investigational compounds against an approved drug.  

“Trials are being done outside the U.S. that couldn’t be done in the U.S.” because investigators would not randomize patients to an investigational therapy that mimics one that is already marketed, said Pazdur. 

His concern is largely focused on PD-1s, where, he previously told BioCentury, there is too much duplication and too little collaboration among sponsors.

Pazdur contends that such sponsors are trying to create “almost an alternative drug development pathway.” Though the single-country path has lower costs than MRCTs, “that is not a regulatory consideration,” he said.

Despite the lower cost of conducting trials only in China, multiregional clinical trials would benefit Chinese patients and sponsors in the long run, according to Pazdur.

“Everyone is better off having a more uniform regulatory standard versus countries taking advantage of a loophole by doing trials that couldn’t be done in the U.S. or Western Europe,” Pazdur told BioCentury.

Sponsors should expect that FDA’s focus on ensuring proper racial and ethnic representation in clinical trials will grow.

Pazdur said that the oncology division is writing a guidance document and that companies should come with a plan for ethnic diversity in clinical trials. “This is not just about registrational trials, it is about the whole development program, starting in Phase I.”