Lessons learned from research with chloroform.

New scientific evidence from the Chemical Industry Institute of Toxicology (CIIT) shows that the ability of chloroform to produce liver cancer in mice is secondary to a continual state of tissue damage and repair, produced when the chemical is given daily in large toxic doses over the lifetime of the animal. These findings suggest the need for significant changes in the design, conduct, and interpretation of animal cancer studies.

Chloroform is produced in trace amounts in the process of chlorinating drinking water [in the range of 1 to 50 parts per billion (ppb)]. It is also formed as a by-product in some industrial processes, such as the bleaching of paper. Thus, it is important to define the health risks associated with continual low-level exposures to this chemical.

Chloroform induced liver tumors in female B6C3F1 mice when given daily for two years by bolus gavage dosing (a single daily dose placed in the stomach) at the maximum tolerated dose (MTD) and one-half the MTD. A similar dose for a person would be about one ounce of pure chloroform every morning. However, when the chloroform was given in the drinking water, no increase in liver cancer was seen, even though the mice were consuming about the same daily intake of chloroform as in the gavage study.

To make sense of these results, CIIT scientists first determined that chlorofor did not produce cancer by directly inducing mutations in the DNA. Next, tissue toxicity and repair were examined in the liver under conditions similar to thos of the previous cancer studies. The results were clear cell damage, cell killing, and regenerative cell growth were seen in the liver at both doses that had induced cancer in the gavage study, but not at any dose when given in the drinking water. It is doubtful that there would be any carcinogenic response, i it were not for those toxic effects.

The reason for the difference in the gavage and drinking water responses is tha chloroform, given as a series of bolus doses by gavage, results in repeated massive doses to the liver that overwhelm defense mechanisms and kill liver cells. Chloroform was not toxic or carcinogenic to the liver in the drinking water study, because it was ingested in small sips throughout the day and, thus was delivered to the target tissue at rates low enough to be detoxified.

Using the mouse liver cancer incidence data from the gavage study, the linearized multistage risk model used by regulatory agencies to extrapolate fro high to low exposures suggests that a 1-in-100,000 lifetime increased risk of cancer would occur with 4 ppb chloroform in the drinking water. This model is very conservative because it assumes that a cancer risk is associated with even the most minute intake of chloroform. In fact, no cancer or increased cell proliferation was observed in the drinking water study, with 1,800,000 ppb of chloroform. Clearly, the conventional model dramatically overestimates risk, an more realistic risk assessment approaches, using more scientific information, are needed for chemicals such as chloroform.

Lessons learned from the CIIT research that impact carcinogen testing and risk assessment:

* Toxicity and tissue repair should be repaired in selecting doses for cancer studies;

* Bolus gavage dosing is suspect and may provide results that cannot be extrapolated to environmental exposures, such as drinking water;

* In interpreting cancer studies, cell damage, cell killing, and regenerative cell proliferation may be the driving force in tumor formation at both the MTD and one-half the MTD;

* For chemicals that do not react with DNA, it is not appropriate to use the linearated multistage model to extrapolate cancer risks from high toxic doses t lower nontoxic exposures.

Larson, J.L, C.S. Sprankle, and B.E. Butterworth (1994), Lack of chloroform-induced DNA repair in vitro and in vivo in hepatocytes of female B6C3F1 mice. Environ Mol Mutagen 23:132-136.

Larson, J.L., D.C. Wolf, and B.E. Butterworth (1994), Induced cytotoxicity and cell proliferation in the hepatocarcinogenicity of chloroform in female B6C3F1 mice: Comparison of administration by gavage in corn oil vs. ad libitum in drinking water. Fundam Appl Toxicol 22:90-102.