Glossary

ADI acceptable daily intake

An estimate of the amount of a food additive, per kilogram of body weight, that can be eaten every day over a lifetime without appreciable health risk. It is a result of risk assessment, not hazard identification; for aspartame JECFA set it at 0–40 mg/kg. Read more →

Advisory Group

An international, interdisciplinary panel convened by IARC, roughly every five years, to recommend which agents the Monographs should evaluate next. The group that met in March 2024 set priorities for 2025–2029. Read more →

Agent

In the IARC Monographs, anything whose ability to cause cancer is evaluated: a chemical, a physical agent such as radiation, a biological agent such as a virus, or an exposure circumstance such as working as a painter. Read more →

Autochrome

The Lumières’ colour photograph, whose colours come from a microscopic mosaic of dyed potato starch grains. Early Autochromes are on display at the Institut Lumière. Read more →

Bias

In the Preamble’s sense, the effect of factors in the design or conduct of a study that make an association appear stronger or weaker than it really is. Together with chance and confounding, it is one of the explanations that must be ruled out before an association is accepted as causal. Read more →

Cancer hazard

An agent that is capable of causing cancer under some circumstances. It is a property of the agent itself and says nothing about how much exposure is needed or how likely harm is. Identifying hazards is the task of the IARC Monographs. Read more →

Cancer risk

An estimate of the probability that cancer will occur at a given level of exposure to a hazard. It depends on how much, how often and for how long someone is exposed. IARC does not assess risk; other bodies, such as JECFA for food additives, do. Read more →

Cinématographe

The apparatus patented by the Lumière brothers of Lyon on 13 February 1895 that could both film and project moving pictures onto a screen. Its first paying screening took place at the Grand Café in Paris on 28 December 1895. Read more →

Confounding

A mixing of the effect of another factor with the effect being studied. Early studies linking coffee to bladder cancer did not adequately account for tobacco smoking, a strong cause of bladder cancer that was common among heavy coffee drinkers. Read more →

Dose–response relationship

A pattern in which risk changes with the amount of exposure. The Preamble treats an observation that cancer risk rises with increasing exposure as a strong indication of causality, as for lung cancer and outdoor air pollution. The relationship need not be a straight line. Read more →

Epidemiology

The study of how health-related states and events are distributed in populations and what determines them. In the Monographs, evidence of cancer in humans usually comes from epidemiological studies that compare people who were and were not exposed to an agent. Read more →

GIOŚ Chief Inspectorate of Environmental Protection (Poland)

The Polish government body that publishes an annual assessment of air quality in the country’s 46 assessment zones. In its assessment for 2024 it named benzo[a]pyrene as Poland’s biggest air-pollution problem. Read more →

Group 1 carcinogenic to humans

The IARC category used when the evidence that an agent causes cancer is convincing, usually because there is sufficient evidence from studies in people. Tobacco smoking, processed meat and outdoor air pollution are all in Group 1, although the size of the risk they carry differs greatly. Read more →

Group 2A probably carcinogenic to humans

The IARC category generally used when at least two of three findings are made: limited evidence in people, sufficient evidence in experimental animals and strong mechanistic evidence. Red meat and drinking very hot beverages (above 65 °C) are examples. “Probably” here has no quantitative meaning. Read more →

Group 2B possibly carcinogenic to humans

The IARC category generally used when only one of the three findings listed for Group 2A is made. Aspartame and radiofrequency electromagnetic fields are examples. The label signals weaker evidence than Group 2A, not a measured level of risk. Read more →

Group 3 not classifiable as to its carcinogenicity to humans

The IARC category for agents that fit none of the other groups, most often because the evidence is inadequate or limited. It is the largest group. A Group 3 result is not a finding of safety; coffee, for example, was moved from Group 2B to Group 3 in 2016. Read more →

IARC International Agency for Research on Cancer

The specialised cancer agency of the World Health Organization, created by the World Health Assembly in 1965, with its headquarters in Lyon. It is a research agency, not a regulator. Read more →

IARC Monographs

IARC’s programme, running since 1971, that identifies cancer hazards. Each Monograph is a review of already published studies, ending in a classification of the agent in one of the groups. The Monographs make no recommendations on regulation or policy. Read more →

Institut Lumière

A museum and film archive on the site of the Lumière factories in Lyon’s Monplaisir district, housed in the family villa. Next to it stands the Hangar du Premier-Film, the last surviving part of the factories, now a cinema. Read more →

JECFA Joint FAO/WHO Expert Committee on Food Additives

A committee of the WHO and the Food and Agriculture Organization that carries out risk assessments of food additives, taking into account the conditions and levels of exposure. In 2023 it evaluated aspartame in parallel with IARC. Read more →

Key characteristics of carcinogens

Ten properties often shown by agents that cause cancer in people, listed in the 2019 Preamble to organise mechanistic evidence. Examples are being genotoxic (damaging DNA), inducing chronic inflammation and being immunosuppressive. Showing such a property does not by itself settle a classification. Read more →

Limited evidence

In studies of people, the level at which a causal explanation of an observed association is credible, but chance, bias or confounding cannot be ruled out with reasonable confidence. The 2011 evaluation of radiofrequency fields rested on limited evidence for glioma. Read more →

Mechanistic evidence

The third stream of evidence in the Monographs, alongside studies in people and in animals: data on whether an agent acts in ways known to lead to cancer. The Preamble rates it as strong, limited or inadequate; it is strong when results in several experimental systems are consistent and coherent. Read more →

Musée des Confluences

Lyon’s museum of natural sciences, human sciences and technology, opened on 20 December 2014 where the Rhône and the Saône meet. Its building was designed by the Austrian architects Coop Himmelb(l)au. Read more →

Nouveau Centre IARC headquarters building

IARC’s headquarters at 25 avenue Tony Garnier in Lyon’s Gerland district. Staff moved in on 20 January 2023 after about 50 years at Grange Blanche, and the building was inaugurated on 12 May 2023. Read more →

Observational study

A study in which researchers record people’s exposure and health without assigning the exposure themselves. In a cohort study they follow exposed and unexposed people to see who develops the disease; in a case-control study they start from people with the disease and compare them with people without it. Read more →

Preamble Preamble to the IARC Monographs

The document that sets out the rules of every Monographs evaluation: the scientific principles, the levels of evidence and the criteria for each group. Its current version was amended in January 2019. Read more →

Relative risk risk ratio

The risk of a disease in one group, usually the exposed, divided by the risk in another group. A value of 1.0 means equal risk; above 1.0, higher risk in the exposed. A relative figure such as “18% higher” does not say how large the starting risk is. Read more →

Sufficient evidence

The highest level of evidence in the Monographs: a causal relationship between exposure to the agent and cancer has been established, in people or in experimental animals. Sufficient evidence in people is enough on its own for Group 1. Read more →

Working Group

The panel of independent international experts who carry out a Monographs evaluation, meeting at IARC in Lyon for about eight days. Members take part as individual scientists, not as representatives of a government, organisation or industry, and only they make the evaluations. Read more →

Sources checked on: 3 November 2026