Data Integrity Concerns Discovered in Gene Therapy Product Submissions

Gene therapies are an emerging and promising field of medicine that offers cures for rare and debilitating diseases, often at exorbitant cost.  The science supporting them is complex and serious side effects are possible.  Consequently, manufacturers of gene therapies must be forthcoming and transparent with information provided to regulatory agencies and patients.  When that transparency or the integrity of the data submitted to get a product approved is compromised, the company, its product, the field of gene therapy—and the potential patients—can all suffer.

[Related: Data integrity is one of many hot pharma GMP topics at the moment. Learn more by downloading the FREE report, “The Current Pharmaceutical Quality Landscape.”]

At the FDLI Enforcement, Litigation, and Compliance Conference held virtually in December 2020, Spark Therapeutics Associate General Counsel Amy Garabedian spoke in a session on cell and gene therapies where she highlighted issues with data integrity in FDA drug applications, emerging trends resulting from what took place in those case studies, and best practices for gene therapy manufacturers.

Case 1: AveXis Zolgensma Data Integrity

The first case involving data integrity issues that she discussed was related to the AveXis spinal muscular atrophy (SMA) product Zolgensma.  In March of 2019, there was an internal allegation of data manipulation in the Biologics License Application (BLA) that was reported to AveXis senior management. At that time, AveXis opened an internal investigation that consisted of two phases.  The investigation was still ongoing when FDA approved Zolgensma in May of 2019 (Figure 1).

About a month later the internal investigation was closed and AveXis voluntarily disclosed to FDA that part of the data that was submitted in the BLA was inaccurate. Both FDA and AveXis agreed that there was an issue with the animal studies that were submitted as part of the BLA, which were not supportive of what was ultimately commercialized. It was also agreed that there was no issue of safety or efficacy with the Zolgensma product that was on the market.

Later that year, in August, CBER Director Peter Marks issued a press release, pointing out that if data are mishandled, patients can be harmed, and if patients are harmed, the field of gene therapy could be set back. He also emphasized that even if patients are not harmed, “transparency in data reporting is an industry cornerstone.” He further stated that “It is the manufacturer’s responsibility to submit complete and accurate information in marketing applications for evaluation by the FDA.”

The statement also pointed out that “AveXis became aware of the issue of the data manipulation that created inaccuracies in their BLA before the FDA approved the product yet did not inform the FDA until after the product was approved. The agency will use its full authorities to take action, if appropriate, which may include civil or criminal penalties.”

The press release from FDA was “likely a signal to the industry about the agency’s thinking with regards to data integrity when it comes to gene therapies,” Garabedian said.

The Novartis AveXis case was closed March 30, 2020 with voluntary actions by AveXis advised.  The remediation plan that was ultimately accepted by FDA included: 

• Firing two senior executives who were implicated in the data manipulation
• Retraining all employees
• Hiring a data integrity officer and instituting employee awareness through a “Speak Up” program
• Broadening assessment of AveXis’ data oversight by a third party manufacturing consultant

The company also agreed that it would notify FDA within five days of any credible allegation regarding data integrity tied to a submitted BLA.

Case 2: BioMarin Roctavian CMC Issues

Another case study presented involved issues around the CMC portion of an FDA submission.  It came to light in August 2020 with BioMarin’s product Roctavian, formerly Valrox, an investigational gene therapy product for Hemophilia A.

In August 2020, FDA issued a Complete Response Letter to BioMarin’s BLA. It indicated that the agency review was complete and that the BLA was not ready to be approved in its current form.  What BioMarin had submitted was an application that relied on Phase 1/2 data.  This is an instance of using shorter studies and earlier data in a submission.  The firm relied on Phase 1/2 data for “durability of effect.”  FDA recommended that it go back and complete Phase 3 and gather two years of data to provide substantial evidence of the durability of effect.

Products in the respective trials used different manufacturing techniques and facilities, which, combined with slightly different results in efficacy, “likely underscored FDA’s concern that manufacturing changes may result in different product characteristics over time,” Garabedian commented.

It is the manufacturer’s responsibility to submit complete and accurate information

In addition, the assays used changed as the trials progressed.  There was a change from a one-stage assay to a chromogenic assay, and a change to the commercial-grade product, each of which may have affected the durability of effect.

Another important consideration, the Spark Therapeutics attorney pointed out, is that unlike Zolgensma—which is one of only three approved treatments for spinal muscular atrophy—Hemophilia A has several already approved products on the market, which changes the risk-benefit analysis that FDA or other regulatory bodies look at.

Case 3: bluebird bio additional CMC info

Another CMC issue involves bluebird bio and its LentiGlobin™ Sickle Cell Disease Gene Therapy product.

In November 2020, FDA asked bluebird bio to provide additional information on its manufacturing process as well as for additional analytical comparability strategy.  The company announced that this would delay its submission by about two years.

The FDA request would appear to emphasize that the agency is getting more sophisticated in its review and understanding of what data would be included in the CMC portion of a submission, Garabedian commented.

However, it is worth noting that LentiGlobin™ was granted priority medicines (PRIME) designation by the European Medicines Agency just over a month earlier.  The PRIME initiative, according to EMA, “is a scheme launched by EMA to enhance support for the development of medicines that target an unmet medical need.”

Emerging Trends and Data Integrity

Noting that the gene therapy product area is a “fast paced and complex field that poses many unique questions during product review,” Garabedian discussed points that she feels need to be considered and emerging trends that she is seeing.

With many gene therapy products there is the potential for one time administration, and the question of durability comes into play—whether the gene therapy product will continue to achieve a therapeutic effect in the body without causing adverse side effects over a long period of time.

As a result of these unique characteristics, “companies may want to consider when and how they engage FDA regarding questions of data integrity, manufacturing, and product comparability. It is important to engage with FDA and other regulatory bodies as soon as possible to talk about these issues, and always remember to be transparent.”

The company announced that this would delay its submission by about two years

It is important to understand that meeting agreed upon “prespecified criteria” such as clinical end points constitutes only one part of FDA’s approval decision, pointing to the importance of preclinical and clinical research and the design of development programs.  The agency is looking more closely at earlier clinical data given the small populations studied and looking for longer term efficacy.  This focus also brings greater scrutiny to whether the product was adequately characterized and controlled during scale up to commercial manufacturing.

In addition, understanding that additional safety and durability data may be relevant particularly in a setting where there is an established treatment option will be important for sponsor companies.  Also, if there is a comparability study that bridges Phase 1/2 to Phase 3, payers may be relying on such data to establish durability.

Regarding reimbursement, payers will struggle with the number of new high cost cell and gene therapies reaching the market. Also, parents will struggle with the high costs of the gene and cell therapies—for example, Zolgensma’s cost is $2.1 million per dose, the most expensive therapy in the world. Parents will also want to be assured that the safety and durability data can be established and is able to be relied upon.

Best Data Integrity Practices Discussed

Regarding best practices, Garabedian recommended familiarity with the 2018 FDA guidance, “Data Integrity and Compliance with CGMP.”  In addition, she recommended that companies:

• Ensure a robust culture of quality and sustainable GXP compliance
• Audit/assess the quality organization from a structural, effectiveness, and capability standpoint
• Ensure the presence of a cross functional quality counsel or hire a data integrity officer
• Perform routine gap assessments of key quality policies and procedures
• Ensure a clearly defined pathway for reporting concerns regarding potential misconduct or fraud
• Clearly communicate the importance of quality and data integrity
• Ensure project teams are not under pressure to cut timelines or punished for delays
• Review policies and ensure that potential quality issues, including issues relating to or potentially relating to the integrity of quality data, are escalated in a timely manner to senior quality leadership

Additional Resources

FDA Guidance on Data Integrity and Compliance with CGMP

FDA August 2019 Zolgensma Press Release

Novartis August 2019 Zolgensma Press Release

[Related: Author Jerry Chapman covers data integrity and other related issues in the latest Redica Systems report, “The Current Pharmaceutical Quality Landscape.”]