How Novartis’ data manipulation case is a cautionary tale for transparency

Oct. 1, 2019
“Seeing something and saying something” can be applied to the life sciences when a lack of transparency could result in a company facing stiff regulatory action.

In today’s world, the catchphrase for those who observe unusual or questionable activity in public that seems threatening is “if you see something, say something.” And it’s not only just a post-9/11 plea to air travelers and pro sports enthusiasts; the U.S. Department of Homeland’s Security’s trademarked slogan is intended to encourage Americans to report suspicious activity to thwart potential acts of terrorism. 

“Seeing something and saying something” can be equally applied to the life sciences when it comes to clinical outcome assessments in which a lack of transparency that can potentially lead to data mismanagement or even manipulation could result in a company facing stiff regulatory action.

On Aug. 6, the U.S. Food and Drug Administration (FDA) publicly rebuked  the Swiss pharmaceutical company Novartis for manipulating data involving its $2.1 million gene therapy. (1) The FDA said that Novartis’ gene unit AveXis failed to report the falsified data prior to its gene therapy treatment Zolgensma getting agency approval in May 2019 despite the fact that the company knew of the issues as early as March 2019.

Though both Novartis and the FDA said the falsified data did not affect the safety, quality or efficacy of the drug, a therapy that treats a rare genetic disease known as spinal muscular atrophy, the agency said Novartis may face civil or criminal charges. In an Aug. 9 letter, five U.S. Senate Democrats criticized the company for submitting and failing to quickly disclose the manipulated data, and called upon the FDA to aggressively pursue legal action against the company. (2) 

“Anything short of a forceful response would signal a green light to future pharmaceutical misbehavior,” stated the letter addressed to FDA acting commissioner Ned Sharpless.

Clinical data (mis)management 

Vas Narasimhan, chief executive of Novartis, claims the company waited to conduct its own investigation before reporting to the FDA what it had found and that the falsified data was limited to experiments on mice used in early phases of the research. However, an FDA Form 483, filed after the agency inspected the lab based on the allegations, stated the data manipulation began in early 2018, and could have been uncovered by AveXis managers during steps in the clinical outcome assessment. (3) Novartis said it launched an internal investigation when a scientist raised allegations in March, however the company didn’t inform the FDA about the allegations until June 28, more than a month after Zolgensma received FDA approval. The FDA inspection report found that AveXis didn’t initiate a noncomformance report into the potential manipulated test data until two months after it was first discovered by AveXis.

In a Aug. 27 article from FiercePharma, the FDA inspection team that met with AveXis concerning the mismanaged data reported that AveXis staff explained away the data discrepancies as errors that had since been corrected after the company implemented a new procedure and better recordkeeping practices. It wasn’t until a month after that an AveXis scientist reported the inaccurate data to the FDA.(4)

Novartis’ actions are doubly troubling in the fact that not only was the delayed reporting suspicious from a regulatory perspective, but the staff’s inaction on the inaccurate data was contrary to the company’s own clinical policies as outlined in its position document.(5) “We have processes in place to ensure the timely disclosure of results regardless of the study outcome … ,” the document states. The company has since fired a small group of scientists who knew or should have known about the data manipulation. U.S. Sen. Chuck Grassley, chair of the Senate Finance Committee, has also initiated his own investigation alongside the FDA investigation.

Dr. Robert Califf, a professor of medicine at Duke University’s School of Medicine and a former FDA commissioner for medicinal products and tobacco, said had Novartis alerted the FDA earlier about the data integrity issue, the accelerated review of the drug would have been delayed, but not denied.(6) 

“Mistakes, sometimes quite subtle ones, are common in study design and conduct, as well as in the analysis and reporting of findings,” Califf said. “… As in other industries, ensuring quality requires intelligent design of the oversight system so most problems can be detected in real time.”

Dr. Peter Marks, director of the FDA’s Center for Biologics Evaluation and Research, said there are three important issues in the data manipulation allegations against Novartis.(7) First, if data is mishandled, patients can be harmed. Second, if patients are seriously harmed, the result could set the entire field of gene therapy back years if not decades. And finally, even if patients aren’t harmed as the result of falsified data, such as in the Novartis case where other data demonstrated that the drug is a safe and effective product for the children it treats, transparency in data reporting is an industry cornerstone.

“It’s making sure that the whole ecosystem understands that when people are working on these things that are highly technically complex, that they have to work truthfully and accurately,” Marks said. “Because that’s the foundation upon which the trust that patients put in these products is built.”

The high degree of responsibility and trust placed in pharma manufacturers is why the FDA slammed Novartis in such a public manner after the allegation came to light, he said.

“Why this is so serious is because we expect data that comes to us to be truthful, complete, and accurate,” Marks said. “And if it’s not, that’s a real big problem. It’s a problem for this product. It’s a problem for other products and it’s particularly a problem for products that are of this complex technical nature.” 

When pharma and other life sciences companies implement and exercise the practices, policies and company culture that values and guards the integrity of their product’s design, clinical and testing data, they are in a sense protecting patients and themselves. They’re also adhering to the compliance standards of good clinical practice (GCP) with regard to pre-clinical test data along with the quality accuracy of documentation that regulators expect.

Data integrity achieved through transparency

In 2018, after an unprecedented increase in the cases of data compliance violations, the FDA reiterated its guidance “Data Integrity and Compliance with CGMP.”(8) The agency advised life sciences manufacturers to implement specific plans, policies and procedures as part of an overall auditing strategy to prevent data management issues that can lead to data mismanagement and potential regulatory action. Among the FDA’s recommendations were that:

  • Data must be recorded — attributable, legible, contemporaneously recorded, original or true copy and accurate (ALCOA).
  • Metadata needs to be attributed with title, author, data/time stamp and structured to be retrievable.
  • Data must have an audit trail, a secure, computer-generated, time-stamped electronic record that allows for reconstruction of events relating to the creation, modification or deletion of an electronic record.
  • Electronic copies of records can be used as true copies of paper or electronic records.
  • Workflows, such as an electronic master production and control records (MPCR), are an intended use of a validated computer system.
  • Changes to MPCRs or similar records can only be made by authorized personnel. Technology to limit such access should be employed.

A digital quality management system (QMS) is designed specifically to perform all of the above functions outlined by the FDA data integrity guidance and many more. An automated QMS can help eliminate inaccurate metadata during a revision process while digitally tracking all changes to records in real time, simplifying the auditing process and reducing the errors common with paper-based systems. An electronic QMS can also be configured to keep records indefinitely or for a specified period of time, and if necessary, to restore any deleted data. Finally, the numerous levels of security in the digital QMS allow a manufacturer to automatically track and ensure the authenticity of each user in a system, registering every signature combination while preventing duplication or reassignment of a user ID and signature combination.

Similar to the cautionary tale of the health tech company Theranos, which became a global poster child for data manipulation after investors learned that the secretive company had fudged the results on 10,000 blood tests, the Novartis situation underscores the need for greater transparency in the life sciences.(9) Had Novartis’ clinical trial research entities provided more transparency with its test data both internally and externally, it may have avoided the stiff legal penalties it could potentially face. 

One avenue that provides companies with greater transparency is digitized documentation. A fully automated QMS gives life sciences companies the ability to create and manage accurate electronic records of product design data, clinical test data and production documentation throughout the entire life cycle of the product, helping companies ensure that their data integrity is sound and secure.


As it is still early in the FDA’s investigation of what went wrong regarding the weighty allegations against Novartis, more information and clues are likely to surface in the weeks and months to follow. However, life sciences manufacturers can be proactive about protecting their data integrity by applying principles outlined in relevant FDA guidance documents; seeking greater transparency in the documentation of their data; and employing digital solutions that support accurate documentation, preclude data manipulation, and preserve a detailed audit trail. 


  1. “Novartis C.E.O. Defends Company’s Decision to Withhold False Data from the F.D.A.,” by Katie Thomas. The New York Times. Aug. 7, 2019.
  2. “Sanders, Warren join team of senator blasting Novartis’ Zolgensma data rig,” by Angus Liu. Aug. 12, 2019.
  3. “Novartis inspection puts into question who knew what—and when—about Zolgensma data manipulation,” by Eric Palmer. Aug. 13, 2019.
  4. “Novartis earlier acknowledged Zolgensma data errors to FDA but blamed them on poor data,” by Eric Palmer. FiercePharma. Aug. 27, 2019.
  5. “Novartis Position on Responsible Clinical Trials,” Novartis Corporate Affairs, Public Affairs Policy. May 2019.
  6. “Novartis violated FDA’s sacred principle: In God we trust, all others must bring data,” by Robert M. Califf. Stat News. Aug. 14, 2019.
  7. “FDA on Novartis data manipulation controversy: ‘We happened to be lucky,’” by Matthew Herper. Aug. 21, 2019.
  8. “Data Integrity and Compliance With Drug CGMP Questions and Answers Guidance for Industry,” U.S. Food and Drug Administration. December 2018.
  9. “MasterControl software provides efficiency, transparency for life sciences companies,” by Alex Keown. BioSpace. May 8, 2019.
    This article was originally published on MasterControl's blog, GxP Lifeline
About the Author

Mike Rigert | Staff Writer