The World Health Organization classifies cancer-causing substances through its International Agency for Research on Cancer (IARC), which has evaluated over 1,000 agents and placed them into four groups based on the strength of evidence linking them to cancer. As of early 2026, 135 agents are confirmed carcinogens, 97 are probable carcinogens, and 324 are possible carcinogens. These classifications shape public health policy worldwide, but understanding what they actually mean requires knowing how the system works.
How the Classification Groups Work
IARC sorts agents into four groups. The grouping reflects how strong the evidence is that something can cause cancer, not how dangerous it is in practice. This is a critical distinction that trips up many people.
- Group 1: Carcinogenic to humans. There is sufficient evidence from studies in humans, or a combination of strong mechanistic evidence and animal data. This group includes 135 agents, from tobacco smoke and asbestos to processed meat and certain industrial chemicals.
- Group 2A: Probably carcinogenic to humans. The evidence is strong but falls short of certainty, typically combining limited human evidence with sufficient animal evidence or strong mechanistic data. This group contains 97 agents.
- Group 2B: Possibly carcinogenic to humans. Only one line of supporting evidence exists, whether limited human data, sufficient animal data, or strong mechanistic evidence. This group is the largest, with 324 agents.
- Group 3: Not classifiable. This does not mean safe. It means the available evidence is too limited or inconsistent to draw a conclusion. Significant research gaps often explain a Group 3 placement.
Hazard vs. Risk: A Key Distinction
IARC identifies hazards, not risks. The difference matters enormously. A hazard assessment asks: “Can this thing cause cancer under any circumstances?” A risk assessment asks: “How likely is it to cause cancer given how people actually encounter it?” Sunlight and plutonium are both Group 1 carcinogens, but no one would argue they pose the same level of danger in daily life.
This is why a Group 1 classification for something like processed meat can sound alarming without context. It means the evidence that processed meat can cause cancer is as solid as the evidence for tobacco. It does not mean eating a hot dog is as dangerous as smoking a cigarette. The classification tells you about the quality of the science, not the magnitude of the threat.
Group 1: Confirmed Carcinogens
The 135 agents in Group 1 span a wide range. Some are well-known: tobacco, alcohol, asbestos, ultraviolet radiation, and certain viruses like hepatitis B and human papillomavirus. Others are less obvious.
Outdoor air pollution earned a Group 1 classification after IARC found sufficient evidence linking it to lung cancer, with a positive association for bladder cancer as well. Fine particulate matter, evaluated separately, also received a Group 1 classification. Processed meat joined the list based on data showing that every 50-gram daily portion (roughly two slices of bacon) increases the risk of colorectal cancer by 18%.
In 2023, IARC classified PFOA, one of the “forever chemicals” found in nonstick coatings and firefighting foam, as a Group 1 carcinogen. The decision rested on sufficient evidence from animal studies combined with strong mechanistic evidence observed in exposed humans, including changes to gene expression and immune suppression.
Recent High-Profile Classifications
Several recent evaluations have drawn public attention. In July 2023, IARC classified the artificial sweetener aspartame as Group 2B (possibly carcinogenic) based on limited evidence linking it to liver cancer in humans. The WHO’s food safety body simultaneously noted that typical consumption levels remain within safe guidelines, illustrating the gap between hazard identification and practical risk.
PFOS, a related forever chemical often found alongside PFOA, received a Group 2B classification in the same year. While both chemicals share similar properties, the evidence for PFOS was limited to mechanistic data rather than direct human cancer studies.
In 2024, talc was reclassified as Group 2A (probably carcinogenic). The evaluation combined limited evidence of cancer in humans, sufficient evidence in animals, and strong mechanistic evidence from human cell studies. This replaced earlier, narrower assessments that had evaluated talc separately depending on whether it contained asbestos. Talc contaminated with asbestos remains classified as Group 1 under the broader asbestos classification.
Night shift work also holds a Group 2A classification. The primary mechanism is disruption of the body’s normal circadian rhythms, the internal clock that regulates sleep, hormone release, and cell repair. Long-term disruption of these cycles appears to increase cancer susceptibility.
How Agents Get Evaluated
IARC doesn’t test substances itself. Instead, independent working groups of international scientists review all available published research on a given agent. They examine three streams of evidence: studies in humans, studies in animals, and mechanistic data (how a substance interacts with cells and biological processes at a molecular level).
Not everything gets evaluated. An advisory group recommends priorities based on two main criteria: evidence that people are currently exposed to the agent (whether broadly or in specific occupations) and whether enough published research exists to support a meaningful evaluation. Any single evidence stream, whether human, animal, or mechanistic, can be enough to justify prioritizing an agent for review. The full evaluations are published as IARC Monographs, a series that now spans 140 volumes.
What the Classifications Mean for You
The IARC list is a scientific reference point, not a personal action plan. A Group 1 classification for alcohol doesn’t come with a specific threshold below which risk disappears. A Group 2B classification for something like aloe vera extract doesn’t mean you need to throw out your moisturizer.
What the classifications do offer is a reliable signal about where the science stands. When an agent moves from Group 2B to Group 2A, or from 2A to Group 1, it typically means new research has strengthened the case. Regulatory agencies around the world use these classifications as a starting point for setting exposure limits, product safety standards, and workplace protections. The science of hazard identification feeds into the separate, equally important process of figuring out how much exposure actually matters in real-world conditions.