Chemical & Engineering News

September 17, 1990

Fluoride bioassay study under scrutiny

by Bette Hileman

A recent animal study by the National Toxicology Program that failed to give fluoride a totally clean bill of health is now being questioned as being maybe too lenient on the widely used tooth decay preventative. Also, a government panel appointed to assess the risks and benefits of human exposure to fluoride and to put the NTP study into broader perspective is apparently running into some problems and has delayed its final report. In addition, a Congressional committee is beginning to show concern about the NTP study.

In the two-year NTP bioassay, mice and rats were exposed to sodium fluoride in their drinking water at fluoride levels of 0, 11, 45, and 79 ppm. Last spring, the NTP staff and a technical review panel concluded that the study showed "equivocal" evidence that sodium fluoride caused osteosarcoma, a rare form of bone cancer, in male rats.

At the ACS national meeting in Washington, D.C., William L. Marcus, senior scientific advisor in the office of drinking water at the Environmental Protection Agency, stated that fluoride should perhaps have been placed in the next highest category, called "some evidence" of carcinogenic activity. (NTP has five categories for chemicals in tests for cancer in rodents: clear evidence, some evidence, equivocal evidence, no evidence, and inadequate study.)

Marcus gave a number of reasons for his doubts about NTP's conclusions. First, the level of fluoride the low- and mid-dose animals had in their drinking water was within an order of magnitude of what humans are exposed to when drinking water containing the EPA-established maximum level of 4 ppm fluoride. This is almost unheard of in animal bioassays. Usually, animal exposure is four to six orders of magnitude more than what humans receive.

Furthermore, at the end of the study, the high-dose animals had accumulated lower levels of fluoride in their bones (5300 to 6200 ppm) than do humans after drinking water with 4 ppm fluoride for at least 10 years (reported to be 7000 ppm). "This is the first time in my memory that [test] animals have lower concentrations of the carcinogen at the site of adverse effect than do humans," Marcus said.

"But the most disturbing part of the report was the repeated reference to the historical controls as having the same or higher numbers of cancers than the test groups," Marcus noted. In previous NTP studies of other chemicals both control and test animals received 0.7 to 1.2 mg fluoride per kg of body weight in their feed each day. (The fish meal that is put in standard lab chow for rodents contains high levels of fluoride.) In the fluoride study, the animals received daily doses of 0.2 mg of fluoride per kg of body weight from their feed. Consequently, the historical control animals used to evaluate the significance of tumors found in the fluoride study were actually a dosed group receiving fluoride sufficient to place them between the low- and mid-dose groups in the fluoride study. Thus the cancer incidences in historical controls, which often were used to downplay the significance of certain tumors in this study, should not have been employed at all for comparison purposes, Marcus said.

Furthermore, when the incidence of osteosarcomas found in historical controls (0.5%) and the incidence found in this study are plotted against fluoride levels, an approximate linear relationship is found. Because it is unusual to find such a clear dose-response curve, Marcus said, this too is a strong indication that fluoride caused the osteosarcomas.

Marcus was particularly concerned about what he calls the "systematic downgrading" of lesions found by the original contractor for the NTP study. The final report for the study was prepared by the NTP staff, but the testing itself was done by Battelle Columbus Laboratories under contract to NTP. A report prepared by Battelle was audited by a quality assurance contractor, and a separate group of pathologists reviewed the studies. In the process, a number of postive findings in the original Battelle report were downgraded. Slides first diagnosed as showing a rare form of liver cancer called hepatochlolangiocarcinoma were later said to indicate hepatoblastoma, another type of rare malignant lesion, and finally to show the far more common cancer hepatocarcinoma. These hepatocarcinomas were combined with the other hepatocarcinomas found in both treated and control animals, Marcus said.

In addition, dose-dependent oral lesions noted in the Battelle report were downgraded from dysplasia and metaplasia to degeneration. Some other liver carcinomas were eventually reclassified as n onmalignant lesions. Because of what he calls systematic downgrading of the slides, Marcus has written a memo to the director of the criteria and standards division in the office of drinking water asking that EPA assemble an independent board of pathologists to review the slides again.

John R. Bucher, who was in charge of the NTP study, says that it is very common for diagnoses determined by the original contractor to be changed during later review. The hepatochlolangiocarcinomas were not downgraded but, based on their predominant cell type, simply revised to hepatoblastomas, which are actually a more unusual lesion, Bucher says. Later, because the hepatoblastomas are hepatocarcinomas, they were included with the other hepatocarcinomas for purposes of statistical evaluation, he says.

Marcus agrees that it is not unusual for diagnoses of slides to be changed in NTP studies, but it is unusual for nearly all diagnoses that suggest a dose-response relationship to a chemical to be altered.

A number of other government officials who asked not to be identified also have told C&EN that they have concerns about the conclusions of the NTP study. They, too, believe that fluoride should have been placed in the "some evidence" category, in part because osteosarcoma is a very rare form of cancer in rodents.

A panel appointed by Frank E. Young, deputy assistant secretary for health, science, and environment at Health & Human Services, is currently in the process of preparing a report on the risks and benefits to humans from fluoride exposure. One or more of the panel members reportedly agree that fluoride should have been placed in a higher category in the NTP study. Originally, the report was scheduled for release in June, but it is now slated for November.

EPA is committed to reevaluating its 4-ppm standard for fluoride in drinking water. But it won't go ahead until after the HHS panel's report is complete. Before the end of the year, the Government operations Subcommittee on Human Resources & Intergovernmental Affairs chaired by Rep. Ted Weiss (D.-N.Y) will be reviewing the NTP study.