After FDA's asbestos finding, company then got lab results that were complicated
By Peter Loftus
Johnson & Johnson rushed to test its famous baby powder last month after the U.S. Food and Drug Administration found asbestos in a bottle and triggered a recall. The company announced 11 days later that independent testing had found no trace of the contaminant.
But J&J's push for a rapid turnaround contributed to results that were more complicated, a review of lab reports released by the company shows.
A Pennsylvania lab hired by J&J deviated from its normal process to meet J&J's timetable, adding a testing room normally used to analyze gunshot residue in crime investigations, according to letters J&J received from the testing company and released. Initially, the lab found trace amounts of asbestos in some samples of Johnson's Baby Powder but later determined that the second room was contaminated by an air-conditioning unit, according to the letters. Repeat tests in the lab's normal testing room found no asbestos.
A second lab that J&J hired, in Georgia, told the company its preliminary testing found no asbestos. It hadn't completed its tests when J&J announced on Oct. 29 that no traces of asbestos were found, a letter from that lab shows.
J&J said it is continuing to investigate, including with ongoing sample testing, and will provide an update in due course.
J&J's rapid response challenging the FDA's findings reflects how important Johnson's Baby Powder is to the New Brunswick, N.J., health-care conglomerate. Baby powder generates less than 1% of total sales, but because the product is so well-known, the FDA's asbestos finding represents a threat to J&J's reputation for quality.
At the same time, J&J is defending lawsuits from nearly 100,000 plaintiffs over product safety and marketing, including 16,000 suits claiming its powder caused ovarian cancer in women who used it for feminine hygiene, and a rare cancer known as mesothelioma in people who inhaled it. J&J says its talcum powder is safe, asbestos-free and doesn't cause cancer.
Plaintiffs' lawyers have sent the FDA's powder findings to a federal judge, who is expected to decide in coming weeks whether there is enough evidence for a majority of the lawsuits to proceed.
"Given the serious questions raised by the FDA's test results, our investigation of course started and moved with speed and diligence," J&J said in a statement. "We asked the labs to work as quickly as possible to retest those samples so we could better understand what FDA reported."
This account is based on a review of lab reports and letters released by both the FDA and J&J, court documents in the talc litigation and interviews with people involved in the testing and with independent testing experts.
Talc -- the basis of talcum powder -- and asbestos are naturally occurring minerals that can be found near each other in the earth. The FDA says talc can be contaminated by asbestos if mines aren't selected carefully and talc isn't sufficiently purified. Asbestos was widely used in construction but is now known to cause cancer.
Last year, in response to reports of cosmetics containing asbestos, the FDA started a round of product tests, contracting with AMA Analytical Services Inc., in Lanham, Md., for many.
In early September, the AMA lab analyzed a bottle of Johnson's Baby Powder provided by FDA. It used an optical microscope and a microscope that beams electrons to generate images with higher resolution but in smaller sample sizes.
AMA analyzed three subsamples, each a fraction of one 22-ounce bottle. The electron microscope detected asbestos in two subsamples, in amounts well under 0.1% by weight, and didn't detect it in the third. The optical-microscope analysis didn't detect asbestos in any of the subsamples.
AMA provided a report of results to the FDA dated Oct. 11. J&J said it received the report from FDA Oct. 17 and announced the next day it would voluntarily recall 33,000 bottles of powder from the same lot.
At J&J's request, the FDA retrieved the remainder of the positive sample bottle from AMA and provided it to J&J, which in turn provided it to a Monroeville, Pa., lab-testing firm, RJ Lee Group.
J&J also flew a separate sample of powder from the recalled lot and bags of milled talc that hadn't yet been made into a finished product to the Pittsburgh airport, where an RJ Lee employee picked them up, according to an Oct. 28 letter from the firm, which J&J posted online.
J&J didn't set a deadline but wanted "some very rapid turnaround for obvious reasons," Drew Van Orden, senior consulting scientist at RJ Lee, said in an interview. To prepare samples simultaneously, workers moved some of them from the standard preparation room to the room normally used for gunshot-residue testing "in order to meet the deliverable timeline," the RJ Lee letter said.
Using electron microscopes, RJ Lee technicians found traces of asbestos in some samples from the recalled lot and the milled talc that had been prepared in the second room. In contrast, samples prepared in the standard room tested negative, leading workers to suspect the second room was contaminated.
The second room was warm, and workers had brought in an air-conditioning unit. Lab workers investigated and found traces of asbestos in the air-conditioning unit. RJ Lee concluded that some samples had been contaminated by the environment.
The lab repeated testing on the recalled lot, this time preparing samples in the standard room, and didn't detect asbestos.
In Oct. 28 letters to J&J, RJ Lee said that in samples from the bottle tested by the FDA and from the same lot, "no asbestos contamination was found."
Mr. Van Orden said he read the report provided to the FDA from the Maryland firm and couldn't explain the different results. AMA's technical director, Andreas Saldivar declined to comment and referred questions to the FDA.
The FDA, in a statement, said the distribution of contaminants in a powder might not be uniform. It also said testing of different samples, even from the same bottle, could yield different results.
Frank Ehrenfeld, lab director of International Asbestos Testing Laboratories in Mount Laurel, N.J., who wasn't involved in the powder testing, said it is difficult to come to a conclusion about whether J&J's baby powder contains asbestos, because the FDA-hired lab found a small amount and the finding hasn't been replicated.
"You need more at-bats to really determine what the batting average is for this sample," he said.
J&J also sent samples to another lab, Bureau Veritas North America Inc., in Kennesaw, Ga. That lab sent a letter to J&J on Oct. 27 saying no asbestos was detected in some of the samples using electron-microscopy. The letter said testing was still under way and additional results were expected.
J&J announced that its baby powder tested asbestos-free on Oct. 29. The company said it released results because it had enough tests on the single bottle to have "full confidence it did not contain asbestos."
Bureau Veritas's director didn't respond to requests for comment.
Write to Peter Loftus at firstname.lastname@example.org