Can We Live with NDMA?

Related articles

A series of class action cases claim that Zantac can cause cancer. It is not Zantac per se, but a degradation-produced, N-nitrosodimethylamine (NDMA), which may cause cancer at high doses. What do we know about NDMA?

My colleague, Dr. Billauer, focuses on the legal aspect of the Zantac court case; as a toxicologist, I want to focus on the suspected carcinogen NDMA. The judge in the federal Zantac case was faced with the question of whether the NDMA found in Zantac (ranitidine) was responsible for the cancer seen in the plaintiffs. To answer that, it would be vital for her to know what level of NDMA causes human cancers. And to answer that question, we need to explore NDMA’s

  • Health effects
  • Dose-response (how much is needed to cause an adverse effect)
  • How regulators choose their “safe” levels for chemicals. 

NDMA is everywhere

NDMA is one of many chemicals of the nitrosamine chemical group which are manufactured only for research applications. However, they are found everywhere because they are produced by the reaction of a nitrogen-containing compound (such as nitrous acid, nitrate, and nitrite) with other chemical compounds (amines) that occur naturally in foods, food processing and preservation, as well as manufacturing processes for consumer products and pharmaceuticals, in drinking water, and even within our bodies.    

These compounds are being found everywhere because scientists can now measure chemical levels at the nanogram level; a nanogram is a billionth of a gram, about the weight of one human cell! (For the visually inclined, a gram contains 1,000,000,000 nanograms) 

Health Effects

NDMA and other nitrosamine compounds are considered by the EPA, International Agency for Research on Cancer (IARC), and the National Toxicology Program (NTP) as a probable human carcinogen.  Why probable and not “known?” Because there is no direct evidence that it causes cancer in humans, but there are animal studies showing cancer after NDMA exposure. 

Animal studies have shown liver, esophageal, and lung cancer in rats, mice, and hamsters after exposure to high levels of NDMA in the diet. However, due to differences between animal and human physiology, behavior, genetics, and pharmacokinetics (the transformation of chemicals or drugs within the body), a chemical causing cancer in animals does not mean it will necessarily cause cancer in humans.

Some human studies have reported an association between eating foods containing higher levels of NDMA and an increased risk of stomach, colorectal, and lung cancer. Most of these studies were case-control: people were separated into two groups – one with cancer and one without cancer, involving food recall questionnaires. The problem with the findings of these studies should be obvious – how many people remember what they ate a year ago; most of us can’t even remember what we ate yesterday! 


Dose-response examines the level of a chemical that causes an adverse health effect in humans. This discussion concerns the dose response for cancer, the harmful concern for NDMA. When considering short-term or non-cancerous impact, such as liver damage, regulators use a different approach to assess risk, using other or no models. 

Cancer develops over time; it is a long-term health effect; even a potent carcinogen, such as tobacco, requires years for cancer to form. To answer the question of the level of a chemical that causes cancer, scientists use models where high-dose data from animal studies are extrapolated to low levels of chemicals in the environment to determine cancer risk in humans.  

There are three main models used, and the difference between them is that two of them, the threshold and the radiation hormesis models, have a dose called a threshold, below which no adverse health effects are noted. The linear non-threshold model does not exhibit a threshold, which means that any quantity, no matter how small, could cause an adverse effect.

EPA and most other agencies generally usually use the linear non-threshold model because it is the most conservative; i.e., it shows effects at lower doses than either of the other two models. EPA acknowledges that these are upper-bound estimates of risk based on unquantifiable assumptions about effects at low doses.

These models do not determine how much of a chemical is necessary to cause cancer; instead, they determine cancer risk. Cancer risk measures how many excess cancer cases, beyond those that might naturally occur, develop in a population after exposure to a chemical over their entire lifetime (estimated at 70 years).

For NDMA, the EPA modeled a study showing liver cancer in rats from NDMA in their drinking water. The EPA estimated that there would be one excess death from cancer per 100,000 individuals in the presence of 6.0 nanograms of NDMA per liter of water. Health Canada used the same model but a different animal study and calculated the cancer risk at 40 ng of NDMA per liter of water. [1]

How Regulators Choose “Safe” Levels of Chemicals

These cancer risks are not set as regulatory limits; however, they are often used by regulatory agencies to establish acceptable levels of chemicals in water, air, or in the case of the FDA, pharmaceuticals and consumer products. Regulatory agencies often consider an excess cancer risk of one-in-ten thousand to one in a million as an “acceptable” level of risk (even though this has no scientific basis) and use the concentration of a chemical in this risk range to set a regulatory number.

To determine the acceptable daily intake of NDMA, the FDA,

  • Used Health Canada’s one-in-a-hundred thousand risk level of 40 ng/L,  
  • Multiplied it by two because the average adult consumes about 2 liters of water per day
  • Added in an extra 16 nanograms

The result: 96 nanograms per day of NDMA over a lifetime of exposure; it is a cumulative exposure – not the short-term exposure one might experience taking Zantac


The precise answer to the judge’s question about the level of NDMA that causes human cancer is not readily available. The reason is the significant variability based on the studies and models used. Because science can’t readily answer this question, it leaves the average person subject to unnecessary fear because scientists and the media fail to articulate the difference between short-term exposure that can cause life-threatening effects and a situation involving a cumulative risk.

NDMA comes from a wide variety of sources, including food, and trying to pin down the risk of cancer from a single source, such as Zantac, is problematic at best. What gives me hope is that the judge in the Zantac case understood this and did not allow unreliable, inconsistent, and irrelevant scientific data to rule the day.          

[1] A typical rat drinks anywhere from 20 to 50 ml of water daily  

[2] FDA documentation fell short and did not say the source for the 16 nanograms. My best judgment, which I could not verify, is that it was probably an estimate of NDMA from other sources, such as food and water.  

Sources: EPA Six-Year Review 3 Technical Support Document for Nitrosamines

Guidelines for Canadian Drinking Water Quality