FDA Rejects Plea to Correct Smokeless Tobacco Warnings; A Closer Look at Flawed Interpretations

After two and a half years’ review, the FDA Center for Tobacco Products rejected Swedish Match’s request to eliminate or revise the 30-year-old, egregiously inaccurate warnings that are required to appear on snus products sold in the United States. 

In 2014, Swedish Match sought to:

·         Remove the warning, “This product can cause gum disease and tooth loss”;

·         Remove the warning, “This product can cause mouth cancer”; and,

·         Replace the warning, “This product is not a safe alternative to cigarettes” with this text: “No tobacco product is safe, but this product presents substantially lower risks to health than cigarettes.”

I explored this issue earlier (here).

The FDA justified its rejections in a six-page letter to the company (here) in a 115-page supporting document (here), and subsequently in an announcement proclaiming, “FDA Issues Science-Based Decisions on First MRTP Applications.” 

Science-based?  Following is my review of the FDA’s tortured interpretation of the scientific evidence in its decisions on the gum disease/tooth loss and mouth cancer warnings. 

Gum Disease/Tooth Loss Warning Decision

The agency interpreted removal of a warning as a marketing claim, obligating Swedish Match to prove that snus was entirely without risk – a virtually impossible task.  The FDA advised, for example, that “Omission of [the gum disease tooth loss warning] from a subset of smokeless tobacco products indicates that unlike other smokeless tobacco products, the eight General Snus products cannot cause gum disease or tooth loss.” (emphasis in original).

For this decision, the agency reviewed published studies of “dental conditions (e.g., plaque, caries, tooth wear or tooth loss), gingivitis, gingival recession and periodontal disease” that it claimed were related.  Here are the conclusions for each of these, with the FDA findings in bold:

·         Dental conditions: “Overall, the dental conditions data included no studies that evaluated tooth loss over time…The results on caries were mixed, and the only study to examine the association between Swedish snus and tooth wear found an association. No association was seen between Swedish snus and plaque…”

·         Gingivitis: “Overall the results of the studies on gingivitis were mixed.”

·         Gum recession: “Overall, the only adjusted study of gingival recession to include non-users found a significant positive association between Swedish snus and gingival recession, and several unadjusted studies found significant associations between Swedish snus and gingival recession, although the direction of the association was mixed.”

·         Periodontal disease: “Overall, nearly all of the studies which examine the association between Swedish snus and indicators of periodontal disease (plaque, pocket depth, attachment loss, bone loss) found no association…”

Note the repeated finding of “mixed,” suggesting that some studies showed a positive association while others showed no association or a negative one.  For periodontal, or gum, disease, there was no association. 

Despite these equivocal or nil findings, the FDA concluded: “Overall, the totality of the evidence demonstrates that the eight General snus products can cause gum disease and tooth loss, and, correspondingly, does not support the removal of the warning that these products can cause gum disease and tooth loss.”  The FDA's decision is entirely unsupported by the "totality of evidence."

Mouth Cancer Warning Decision

The FDA decision on mouth cancer is based on an improper analysis of six published studies.  Peter Lee in 2010 published a formal meta-analysis of these studies and one more (abstract here).  Here are his findings for mouth cancer:

“No overall association is seen for oropharyngeal cancer, the most studied cancer type.  For the whole population, an increase (RR 3.1, 95% CI 1.5–6.6) seen in the Uppsala county study (Roosaar et al., 2008), based on 11 cases, contrasts with six studies showing no increase, the overall estimate being 0.97 (0.68–1.37). The never smoker estimate, 1.01 (0.71–1.45), based on four studies, is also null. These results are supported by long-term follow-up of 1115 individuals with ‘‘snuff-dippers lesion’’ (Axéll et al., 1976), which observed no oral cancers at the sites of lesions seen initially (Roosaar et al., 2006).”

When Lee included Roosaar with the other studies in his analysis, there was no association of snus and mouth cancer.  The FDA basically agreed with Lee that Roosaar was the lone work that was positive for snus and mouth cancer, but it cited this study as the sole reason for maintaining the warning.  In fact, the FDA is saying it will ignore the broad consensus of scientific research (i.e. the "totality of evidence") if any one study reports a positive finding.

Having set this impossibly high bar for safer tobacco products, the FDA went further by focusing on tobacco-specific nitrosamines in snus.  Research documents that TSNAs exist in vanishingly small concentrations in snus – about two parts per million or lower (here), and there is no scientific evidence directly linking TSNAs to mouth cancer.  Still, the FDA cited “the presence of nitrosamines in the products that are the subject of these applications, the lack of a threshold dose for mouth cancer” as additional reasons to sustain the warning.  The FDA is effectively saying that TSNAs must be reduced to zero for the warning to be removed.

The FDA closed the door on the gum disease tooth loss warning, but it gave Swedish Match the option of submitting a revised application for the other warnings.  It appears that the agency’s revision/amendment pathway is designed to defeat all the but wealthiest and most determined applicants, leaving millions of smokers and future smokers with demonstrably false warnings against the use of safer smoke-free products.