IARC 2B: what "possibly carcinogenic" means for cell phones

By the EMFCALM team · Updated November 9, 2026

In 2011 IARC, the World Health Organization's cancer agency, put radiofrequency (RF) electromagnetic fields, the radio energy phones use, in Group 2B: "possibly carcinogenic to humans." The panel found limited evidence that RF can cause cancer, mostly from studies of glioma and acoustic neuroma in the heaviest phone users. The label grades that evidence. How likely you are to get cancer from your phone is a separate question, and the panel didn't try to put a number on it.

What did IARC decide in 2011?

From May 24 to 31, 2011, a working group of 31 scientists from 14 countries met at IARC in Lyon, France. On the last day IARC announced the result in press release No. 208: RF fields are possibly carcinogenic to humans, "based on an increased risk for glioma, a malignant type of brain cancer, associated with wireless phone use."

The group reviewed exposure at work (radar and microwaves), from broadcast and wireless signals in the environment, and from personal use of wireless phones. Only the phone studies reached "limited," for glioma and for acoustic neuroma, a nerve sheath tumor. The evidence for other cancers, and for the work and environmental exposures, was judged inadequate.

The press release says the working group "did not quantitate the risk." It did mention one study of phone use up to 2004 that found a 40% higher glioma risk in the heaviest users, who reported about 30 minutes a day over 10 years.

A short summary by Baan and colleagues ran in The Lancet Oncology in July 2011. The full evaluation came out in 2013 as IARC Monographs Volume 102.

How do IARC's groups work?

IARC's rulebook is the Preamble to the Monographs. The current version, amended in January 2019, sorts agents into four groups by weighing cancer studies in humans, cancer studies in experimental animals, and mechanistic evidence (signs that an agent behaves the way known carcinogens do).

Source: IARC Monographs Preamble, amended January 2019, Part B, Section 6(d). Group 1 can also rest on strong mechanistic evidence in exposed humans plus sufficient evidence in animals.

If you read one line of the Preamble, we'd pick this one: the groups "refer to the strength of the evidence that an exposure is carcinogenic and not to the risk of cancer from particular exposures." The words "probably" and "possibly" have "no quantitative significance." They rank how solid the evidence is.

The Preamble also keeps two terms apart. A hazard is an agent capable of causing cancer. A risk is an estimate of how likely cancer is at a given level of exposure. The Monographs deal with hazards, and the Preamble puts extrapolating exposure-response estimates to lower exposures outside a working group's scope.

Group 3 gets misread too, so the Preamble says outright that it "is not a determination of non-carcinogenicity or overall safety." The 2011 press release listed one more group, Group 4, "probably not carcinogenic to humans." The 2019 Preamble has no Group 4.

What does "limited evidence" mean?

In the current Preamble, limited evidence in humans means a causal reading of the link seen in studies "is credible, but chance, bias, or confounding could not be ruled out with reasonable confidence." So studies saw an association, a cause is believable, and the studies couldn't rule out other explanations. Sufficient evidence, the Group 1 bar, means those explanations were ruled out with reasonable confidence.

The 2011 working group used nearly the same definition, printed as a footnote in the press release. Under the rules then, Group 2B covered agents with "limited evidence of carcinogenicity in humans and less than sufficient evidence of carcinogenicity in experimental animals." RF fit that description.

What evidence did the working group weigh?

Volume 102 covers RF from 30 kHz to 300 GHz. Its opening remarks say the evaluation asks whether this type of radiation causes cancer, from phones and other sources alike, and that it "does not provide a quantitative assessment of any cancer risk." The group considered hundreds of articles.

Source: IARC Monographs Volume 102 (2013), general remarks and sections 5 and 6; IARC press release No. 208 (May 31, 2011).

INTERPHONE, a large international study run from 2000 to 2004, found no higher glioma risk in people who had ever been regular phone users. It did find hints of higher risk at the highest cumulative call time, on the side of the head where people held the phone, and for tumors in the temporal lobe, which the monograph calls the part of the brain most exposed when a phone is held at the ear. The monograph adds that chance or bias may explain that increase. Weighing INTERPHONE with the Swedish studies, the working group concluded the findings "could not be dismissed as reflecting bias alone" and that a causal interpretation was possible.

The decision wasn't unanimous. Volume 102 records that a majority chose "limited," and a minority thought the human evidence was inadequate. They pointed to a nationwide Danish cohort study that found no rise in glioma or acoustic neuroma, and to glioma rates that hadn't tracked the growth in phone use. We think that dissent is worth knowing about.

The monograph is also candid about the age of its data. The key studies were done in the late 1990s and early 2000s on GSM networks, and 3G and 4G phones with adaptive power control emit considerably less RF than those older handsets. A hands-free kit lowers brain exposure to under 10% of the exposure with the phone at your ear (it may raise exposure elsewhere on the body), and a Bluetooth hands-free kit puts out roughly 100 times less power on average than a phone. IARC's director then, Christopher Wild, suggested "pragmatic measures to reduce exposure such as hands-free devices or texting." Our guides on reducing exposure from your phone and on AirPods go into those habits.

How many agents are in each IARC group?

On November 9, 2026, IARC's list of agents classified by the Monographs covered Volumes 1 to 142, last updated October 30, 2026. It counted 1,060 agents. Group 2B, with 329, is second only to Group 3's 498, about 31% of everything IARC has classified.

It's a mixed group. Other 2B entries on that list include extremely low-frequency magnetic fields (evaluated 2001), gasoline engine exhaust (2012), Aloe vera whole leaf extract (2013), traditional Asian pickled vegetables (1992) and aspartame (2023). They sit together because the evidence for each reached a similar strength. That tells you nothing about whether they carry similar risks. IARC's own page tells readers to consult the full monographs and warns that "significant new information might support a different classification."

Has IARC scheduled RF for a new review?

IARC has flagged RF for a fresh look twice. As of November 9, 2026, it hasn't put RF on its meeting calendar.

The advisory group that set priorities for 2020 to 2024 met in March 2019 and gave RF a "High priority (and ready for evaluation within 5 years)." It cited new animal evidence, including the large US National Toxicology Program study in rats, which our NTP article covers. No new evaluation followed in that window: IARC's October 2026 list still shows RF under Volume 102, evaluated in 2011.

The next advisory group met March 19 to 22, 2024, and its report for 2025 to 2029 went online on November 4, 2024. It kept RF at high priority but advised waiting until the latter half of that five-year period, for ongoing animal studies, including ones in Japan and South Korea meant to check the NTP results. It calls the human evidence "mixed," notes new evidence in experimental animals, and says a change in the classification "is uncertain." Among the newer studies it describes, the European COSMOS cohort followed 264,574 people for a median of 7.12 years and found a hazard ratio of 1.00 for glioma per 100 hours of cumulative call time, the value that means no difference.

The same report lists 5G among today's RF sources (see our 5G guide) and notes that WHO is running its own RF health risk assessment, apart from IARC. Our overview of what health agencies say covers that side.

As for dates, IARC's July 2026 newsletter lists Meeting 143 on cannabis smoking (November 3 to 10, 2026) and Meeting 144 on lead and cadmium (October 19 to 26, 2027). IARC says it posts calls for data and experts about a year before each meeting. We found no announced RF meeting on IARC's site as of November 9, 2026.

This article explains a hazard classification. It isn't medical advice.

What to check when you see "2B" in a headline

Look the agent up on IARC's list and note when it was evaluated and which line of evidence put it there. For RF that's 2011, and limited human evidence on glioma and acoustic neuroma from phones used in the late 1990s and early 2000s. The next review is penciled in for the second half of 2025 to 2029. Going by IARC's own timing, a call for data on RF should appear on its website about a year before that panel meets.

General information about radio-frequency exposure from everyday devices. Not medical advice.