Trade group’s dispute with FDA highlights need to watch for report revisions

Joseph Burns

About Joseph Burns

Joseph Burns (@jburns18), a Massachusetts-based independent journalist, is AHCJ’s topic leader on health insurance. He welcomes questions and suggestions on insurance resources and tip sheets at joseph@healthjournalism.org.

Photo: Lori Greig via Flickr

Photo: Lori Greig via Flickr

The nation’s clinical laboratories have pushed back hard against a proposal by U.S. Food and Drug Administration in October 2014 that would regulate laboratory developed tests. More than a year later, the disagreement became more heated when the Association of Molecular Pathology (AMP) accused the FDA of fabricating examples in a November report on LDTs.

In the report, “The Public Health Evidence for FDA Oversight of Laboratory Developed Tests: 20 Case Studies,” the agency cited 20 examples of problematic LDTs, saying, “…these products may have caused or have caused actual harm to patients.” We covered this report in a blog post earlier this month.

But that’s just the beginning of the story, as Allison Proffitt, editorial director of BioITWorld, found.

AMP on Dec. 13 countered with a report of its own, accusing the FDA of fabricating examples in its November report. When Proffitt wrote about the response, she led with the most serious charge: “The Association of Molecular Pathology (AMP) accused the Food and Drug Administration of fabricating some of the examples included in a recent FDA report outlining the public health evidence for FDA oversight of laboratory­developed tests.”

She also explained that the alleged fabrication involved a description of the Oncotype DX test from Genomic Health, for recurrence of breast cancer and to gauge how a patient will respond to treatment. The AMP said the FDA analysis had inappropriately considered the HER2 marker test alone and not the full Oncotype DX test. As a result, the AMP report said, “FDA is reporting a case study that does not exist.”

Allison Proffitt

Allison Proffitt

Proffitt posted her article on the morning of Dec. 17. Sometime later, the AMP revised the language in its report, and Proffitt added an update to the top of her story just before 2 p.m. the same day:

“Update 1:56: The language of AMP’s rebuttal was updated on Thursday afternoon. On page one, instead of stating that some examples were ‘fabricated by the FDA,’ the updated language says that some examples were, ‘highly speculative.’ In the Oncotype DX example, the updated conclusion begins, ‘FDA seems to presume an intended use that does not exist.’ Links below direct the reader to the updated rebuttal, but quotes refer to the original posted on Dec 13.”

Proffitt’s reporting caused AMP to revise the language in its report, and in doing so, association removed some of the more inflammatory charges against the FDA.

In response to questions about why AMP revised the report, a spokesperson said, “During the interview with Allison Proffitt of Bio-IT World, it became apparent that the last round of edits was not included in the version that was distributed the day before.” The correct final version was reposted.

The spokesperson also said, “AMP replaced ‘fabricated’ with ‘highly speculative’ and replaced ‘reporting a case study that does not exist’ with ‘FDA seems to presume an intended use that does not exist.’ Let’s be clear: AMP has significant concerns with the misleading information in the FDA report. The team decided the softer, less divisive words would help keep readers focused on the primary issue of gross misrepresentation of public health concerns. They didn’t want the divisive words to detract from the overall message. The basic premise of the AMP response remains the same.”

But AMP never announced that it had made changes to the original report, and initially did not respond to a request for comment. When Proffitt learned that the report was changed, she compared the original (PDF) with the revised version (PDF) word for word, she told me.

So, why didn’t AMP announce that it revised the report? The spokesperson said, “The team did not feel the changes necessitated another press release since the edits were made to the report and not the press release. The link to the report in the release remained the same. We apologize for any confusion and inconvenience.” AMP made other comments about the FDA report and its response, which we have included at the end of this post.

The problem for reporters is it would be impossible to know that the association revised the report without checking back and comparing the versions. But who does that?

To her credit, Proffitt did. She also focused on the issues involved with LDTs, as highlighted in both the FDA and AMP reports. “I didn’t want the inflammatory language to detract from the issues, which are important and deserve to be debated,” she said. In addition, she added this important point about LDTs: almost all genetic and genomic tests are laboratory developed.

When journalists revise an article after publication, they almost always note that the original was changed to reflect new information or to correct an error. But associations operate under no such rules – although they should. To identify potential solutions to this problem, I asked my AHCJ topic leader colleagues for recommendations. Here’s what they suggested:

  • Remember that nothing is final, even reports in PDF. Liz Seegert, AHCJ’s topic leader on aging issues, said to consider association reports to be like legislation, which can change quickly.
  • When you note a change as serious as the one involving AMP, ask for comment, advises medical studies topic leader Tara Haelle. “It’s ethical and responsible to ask why language was changed and to mention in your story that you got a response or not.”
  • Think about the big picture, Haelle adds. “Sometimes the story isn’t the story – it’s how the story is told and who’s telling it.”

Finally, while it may be time consuming to compare the latest version of a document with the original word for word, you can use an old page designer’s trick to do a quick scan of whether the paragraphs end in the same place on the page. A difference may hint that more checking is needed.

Here is the rest of the response from AMP:

“As you may know, the AMP report is the result of input from several members. I can assure you that this was a simple clerical error and the softer language in the updated report was solely based on feedback from AMP staff and members. This is not a case of AMP backing off of its positions regarding oversight of LDPs or to pressures from anyone. These edits do not change the premise of the AMP report and how AMP believes the FDA grossly misrepresented public health concerns of LDPs … This is a critical time for precision medicine and patient care; AMP is urging HHS and OMB to perform a thorough, scientifically unbiased analysis of all the potential harms and benefits of FDA’s proposed regulation before any widespread policy changes are made.”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.