Big pharma admits data manipulation in FDA application for multi-million-dollar gene therapy

By: Braeden Lichti

Aug. 14, 2019 The Food and Drug Administration (FDA) said data manipulation took place during the approval of Novartis’ studies of Zolgensma, the world’s most expensive drug. The medicine, costing around $2.1 million for a one-time infusion, treats children with an especially devastating, sometimes fatal form of spinal muscular atrophy (SMA). Novartis knew of the data irregularities for two months before the gene therapy’s approval by the FDA in May 2019.

But Novartis did not inform regulators until June 2019, a delay that led the FDA to issue a very rare public warning of potential civil or criminal penalties for AveXis. Novartis bought the biotech startup in early 2018 for $8.7 billion mainly due to promising data for the then-experimental Zolgensma.

Phase 1 and Phase 3 data manipulation

As explained in our previous articles, FDA approved medicines go through long, expensive and gruelling clinical trials before being approved. AveXis manipulated results in Phase 1 clinical trials as well as those from some nonclinical studies included in Novartis’ approval application. Fifteen infants with the most severe form of SMA received Zolgensma, and all remained alive and off permanent ventilation at two years, a milestone seldom achieved in untreated patients.

According to Wilson Bryan, head of the FDA’s Office of Tissues and Advanced Therapies both the Phase 1 and Phase 3 versions of Zolgensma use the same vector and therapeutic gene, giving him confidence the clinical results from Phase 1 confirm the effectiveness of the Phase 3 product.

After learning of the data manipulation, the FDA inspected AveXis’ San Diego, CA facility from 24 July-2 August, handing the company an inspection report finding failings to thoroughly review unexplained data discrepancies, incomplete laboratory records and failure to follow laboratory test procedures. The FDA intends to continue its investigations and could require AveXis submit “one or more” supplemental applications, a process that could take several months. Novartis and the FDA have assured the public that the falsified data did not affect the safety, quality or efficacy of Zolgensma and will remain on the market

Impact on future gene therapy approvals

Zolgensma is only the second gene therapy for an inherited disease to win FDA approval, marking a significant milestone for the growing field. At $2.1 million per patient, Zolgensma is also the most expensive drug ever brought to market. The FDA investigations are relevant for the whole sector as gene therapy is a promising emerging field and many active, well-funded biotech companies are working to develop much-needed therapies.

Data fiddling is more common than you think

Manipulating data to make them more meaningful is a well know problem in statistics and is known as p-hacking. The probability value or p-value measures whether the data would be at least as extreme compared to no real difference between the groups or phenomena being compared. The term p-hacking describes the conscious or subconscious manipulation of data in a way that produces a desired p-value. Researchers collect or select data or statistical analyses until nonsignificant results become significant. A p-value of 0.05 or 95% probability is often the de-facto standard to get published in academic literature. You can check out the following visualization to find out how easy it is to manipulate data.

Should you trust the data shared by companies?

Whether you are reading scientific journals or the latest article in the Wall Street Journal, always seek additional guidance when picking a biopharma investment. Connect with an experienced team of biologists, investors, and statisticians to help you to spot troubling signs earlier in your investment journey.

Our views are based on experience and for educational purposes only. We encourage inquiries, suggestions, and comments.

Regards,

Braeden Lichti
Managing Partner

DISCLAIMER: All insights, suggestions, and advice provided herein are for educational purposes only. Nothing contained in this article or within this web site should be interpreted as a recommendation to buy or sell any securities, nor make an offer, solicitation or recommendation of another kind. All readers should always do further research before making a final investment decision.

The author is not a United States Securities Dealer nor Broker nor US Investment Adviser.

Leave a Reply

Your email address will not be published. Required fields are marked *