The ongoing data integrity scandal at Sequenom is very bad news for anyone seeking FDA clearance or insurer coverage for an innovative medical technology or pharmaceutical. For all the safeguards, system controls and audits that are part of a well-controlled and maintained clinical data system, the entire edifice still relies upon faith – the faith of government agencies, insurers, service providers, physicians and patients – in the essential integrity of the clinical data reported by companies and clinical trial sponsors.
Here is the unfolding story. Sequenom is a publicly-traded San Diego based genetic diagnostics company that had been a stock market darling on account of its late first trimester maternal blood test for Trisomy (Downs Syndrome). The appeal of the test rests in its ability to provide a definitive diagnosis of Downs Syndrome earlier in the pregnancy than is otherwise possible and without the need for amniocentesis (with its attendant risks). Sequenom’s shares were being watched closely by the analyst community, and it share price peaked at @ $25 on a January 28, 2009 Press Release and Investors’ Teleconference that trumpeted new clinical data demonstrating a 100% positive predictive value and a 99.9% negative predictive value for the Downs Syndrome test.
An April 29, 2009 Sequenom press release announced a delay in the expected June launch of the test, on account of “discovery by company officials of employee mishandling of R&D test data and results.” The share price plummeted to < $3.00. On September 28, the company released the results of an investigation by independent members of the Board, which concluded that none of the previously released data on the test could be relied upon, and announced the firing of the CEO/President, the Senior VP for Research and Development, the CFO, one other officer, and three other employees. An October 5 regulatory filing disclosed that Sequenom’s management had met with representatives of the U.S. Attorney's Office, the Securities and Exchange Commission, and the Federal Bureau of Investigations. Shares are trading at @ $3.37 as I write.
We don’t as yet know exactly what happened inside Sequenom, whether there was intent to deceive or simply a breakdown in controls, and if there was intent whether it originated from the top, or how and by whom the data errors were discovered. We do know that corrupt data were disseminated to the public and to regulatory agencies, that a lot of money was lost by investors, and that the test was close to launch. Had the company not stepped forward (and there is as yet no information as to how the problem was discovered and by whom) when it did, the test would have been covered by most insurers and widely and rapidly adopted – all on the basis of flawed or falsified data.
The medical device industry has for years urged CMS to consider “external data” – data from sources other than the agency’s own data files, in calculating payment rates and/or in making coverage determinations for various services. After all, some new technologies are used earlier in non-Medicare patients, and there can be substantial and relevant data from external sources that can illuminate service costs and impacts before there is a large body of Medicare data. Openness to all data sources would, the argument goes, yield better and faster decisions than reliance solely on Medicare data. CMS has generally responded that it is not sure it can rely upon data that are externally generated and controlled – it simply doesn’t trust such data. In recent years there has been a bit more openness – but the Sequenom debacle can only heighten distrust and set back recent progress.