Carcinogenicity of Cell Phone Radiation: 2B or not 2B…

… this post is open for comments …

In May/June 2011, 30 experts (I was one of them) invited by the WHO International Agency for Research on Cancer (IARC), gathered in Lyon to discuss the scientific studies on radio-frequency electromagnetic fields (RF-EMF) and cancer.

After intense deliberations, to the great surprise of the world-at-large, experts decided to classify RF-EMF emitted by e.g. cell phones, cell towers and wi-fi networks, as a possible human carcinogen – in IARC scale “2B carcinogen”.

IARC has somewhat complicated but detailed set of rules that guide classification of carcinogenicity. The rules are in place to prevent “out of the blue sky” classifications. Protocols need to be followed and requirements fulfilled before carcinogen is classified.

Once the evidence from human studies is determined to be limited, and the same limited evidence is assigned to the evidence from experimental animal studies, classification of RF was automatically set as 2B possible carcinogen (for details see Preamble of IARC Monograph 102).

There are three possible scenarios that could change classification from 2B possible carcinogen to a higher group of carcinogenicity.

  • The first scenario: if the evidence from human studies would be changed from limited to sufficient then, automatically, no matter what other evidence is, the classification would be – group 1 – carcinogenic to humans.
  • The second scenario: if the evidence from experimental animal studies would be changed from limited to sufficient, and the evidence from human studies would remain as limited, the classification would change and become 2A – probable carcinogen.
  • The third scenario: if the mechanistic evidence shows that the agent (RF) clearly belongs to a class of agents for which one or more members have been classified in group 1 or group 2A. This means that, in practice, mechanistic evidence for RF effects can be fully ignored by the IARC classification system because RF does not belong to a group of agents classified as group 1 or 2A carcinogens.

Something is, however, wrong with the third scenario option because, it would be possible, a priori, to “forget” the review of mechanistic studies related to RF because, no matter what, outcome of this evaluation cannot affect classification stemming from human and animal studies, even if these are evaluated as limited evidence. At the same time, the IARC Preamble says, “…the body of evidence is considered as a whole, to reach an overall evaluation of the carcinogenicity of the agent to humans”. Something is indeed wrong…

However, in 2011 in Lyon at the IARC experts meeting it was clearly spoken that if mechanistic studies provide a plausible mechanism by which cell phone radiation induces biological effects, this evidence will be sufficient to strengthen and even upgrade carcinogenicity classification. More on this topic, later, below.

What is the current situation? New evidence was published since 2011. How it plays with the old evidence – does it make it stronger or weaker?

The 2011 IARC classification of cell phone radiation as a possible carcinogen was based on the limited evidence from human studies and limited evidence from animal studies.

The limited human evidence was based on the results of two sets of epidemiological studies – European Interphone group and Hardell group in Sweden. After the IARC evaluation, in 2014, was published a new epidemiological study – the French CERENAT. This new study reached similar conclusions as Interphone and Hardell previously – long term avid use of cell phone increases a risk of developing brain cancer.

It means that now there are three replications of the same epidemiological type of study, the case-control study, that all suggest cell phone radiation might increase a risk of brain cancer.

It is important to consider, in context of these three results, that brain cancer latency is long and takes several tens of years for the cancer to develop and be diagnosed. Epidemiological studies showing that already little over ten years use of cell phone leads to an increased risk of brain cancer should be taken as a serious warning sign. These studies, due to intrinsic design bias, cannot be considered as a proof of brain cancer being caused by cell phone radiation, but they are a warning sign that such option might be not only “possible” but even “probable”. These are three replicates. These are what the WHO and the telecom industry always demands to get – independently replicated results – they now got it and should consider them as a potentially serious warning. A big red flag.

In respect of the large animal studies, nothing significant happened after 2011. The ongoing in the US a large National Toxicology Program (NTP) study is still in progress, and we need to wait for the results to be published.

In the area of mechanistic studies, both positive and negative studies are being continuously published and the arguing over the existence of the non-thermal effects is still not resolved. Or is this argument indeed still not resolved? Anyone, reading carefully the IARC Monograph Volume 102, may find the following quotes, coming from the deliberations of IARC experts in 2011 (p. 414 of the IARC Monograph Volume 102):

The data to evaluate mechanisms by which RF radiation may cause or enhance carcinogenesis are extensive and diverse.”

Then there are comments concerning thermal and non-thermal effects:

Many studies were confounded by significant increases in the temperature of the cells, leading to thermal effects that could not be dissociated from non-thermal RF-induced changes.”

And in the next sentence is the most important statement:

The conclusions presented in this section [of the Monograph] for results in vivo and in vitro pertain only to those studies for which the Working Group concluded that thermal confounding did not occur.”

This statement clearly suggests that any RF effects described in the mechanistic studies part of the IARC Monograph Volume 102 are non-thermal effects.

These non-thermal effects, effects not having thermal confounding, are as follows:

  • p. 415 “Overall, the Working Group concluded that there was weak evidence that RF radiation is genotoxic…”
  • p. 416 “…the Working Group concluded that data from studies of genes, proteins and changes in cellular signaling show weak evidence of effects from RF radiation…”
  • p. 417 “…Overall, the Working Group concluded that there was weak evidence that exposure to RF radiation affects oxidative stress and alters the levels of reactive oxygen species.”
  • p. 417 “The evidence that exposure to RF radiation alters the blood-brain barrier was considered weak.”
  • p. 417 “The evidence that exposure to RF radiation alters apoptosis was considered weak.
  • p. 417 “The evidence that RF radiation alters cellular replication was considered weak.”
  • p. 417 “There was weak evidence from in vitro studies that exposures to RF radiation alters ornithine decarboxylase activity.”
  • p. 417 “The evidence that exposure to RF radiation, at intensities below the level of thermal effects, may produce oxidative stress in brain tissue and may affect neural functions was considered weak.”

Consistently, the mechanistic group found all mechanistic evidence to be weak. This, however, does not fully reflect the opinions presented in Lyon. In most of the cases, when the consensus was impossible to reach, the group members had to vote, and… voting went along the certain, predetermined lines. Consistently, the same members voted down the opinions of the differently thinking minority.

The composition of the experts of the mechanistic group:

  • Igor Belyaev
  • Carl Blackman
  • Rene De-Seze
  • Jean-Francois Dore
  • Jukka Juutilainen
  • Dariusz Leszczynski
  • James McNamee,
  • Junji Miyakoshi,
  • Chris Portier (Chair)
  • Stanislaw Szmigielski
  • Luc Verschaeve
  • Vijaylakshmi

The weak evidence of mechanistic studies reflected the opinions of the majority. Democracy might be not the best way to deal with scientific evidence.

Was there, during Lyon’s deliberations, presented mechanistic evidence that could be considered as plausible mechanism for the cell phone radiation-induced effects? My answer is yes. There was evidence from six different research groups using six different experimental models and six different exposure conditions, but all of them arrived at the same result – cell phone radiation activated stress response in living cells. This means that the radiation is recognized by the living cells and an agent who might cause harm, and it is necessary to protect from this harm by the launch of the stress response. I have presented this option in my talk in London at the meeting on Childhood Cancer 2012. Slide from this talk is presented below.

Stress response

Already at the IARC meeting in 2011, using the mechanistic evidence, it was possible to classify cell phone radiation as a probable carcinogen (2A). However, in my observation, the sole “shock” of the probable carcinogen classification making it through, against all expectations, prevented any reasonable debate over the classification of this radiation as a probable carcinogen.

The third replicate of the epidemiological evidence, published in 2014, makes the evidence from human studies clearly stronger. Some might say that the human evidence is now sufficient, what automatically means that the cell phone radiation is a human carcinogen, category 1 in IARC scale.

Others, like myself, might be more moderate and consider that three replicates of epidemiological evidence are a strong indication of a probability that cell phone radiation is a probable carcinogen. This is not, however, a definite proof because a variety of bias is involved in the case.control studies. Furthermore, the epidemiological evidence is supported by the limited evidence from animal studies and by the mechanistic evidence showing activation of stress response – a major set of pathways regulating, among others, development of cancer.

For those dismissing scientific evidence of probable carcinogenicity – a memento. Thus far we focused on whether cell phone radiation alone causes cancer. We did not study whether cell phone radiation in combination with chemicals or other radiation could enhance carcinogenicity. Hints of such a possibility come from the animal studies used as evidence during IARC 2011 deliberations.

Any person, scientist or non-scientist, before claiming that there is no risk from cell phone radiation and dismissing the science as a bogus, should remember that it is necessary to study all possible options of effects before making such statements – option of co-effects of cell phone radiation and chemicals was not studied yet…

In conclusion, I consider that currently the scientific evidence is sufficient to classify cell phone radiation as a probable human carcinogen – 2A category in IARC scale. Time will show whether ‘the probable’ will change into’ the certain’. However, it will take tens of years before issue is really resolved. In the mean time we should implement the Precautionary Principle. There is a serious reason for doing so.


19 thoughts on “Carcinogenicity of Cell Phone Radiation: 2B or not 2B…

  1. Pingback: Odźwierny globalistów prof. dr hab. inż. Andrzej Krawczyk | Piotr Bein's blog = blog Piotra Beina

  2. Pingback: Google Moves to Blanket 95% of US Land Mass with Microwave Radiation | Dublinmick's Breaking News

  3. I find it remarkable that in spite of the growth of cell phone use in the United States from near zero in 1985 to near ubiquitous in in 2010 the incidence of new brain and nervous system cancers in the U.S. continues to decline by approximately 0.6% per year according to the latest SEER data. See for example the recent paper by Little et al. which shows that if recent Swedish studies claiming up to a 4 times higher risk of brain cancer in heavy cell phone users are correct, then that should be observable in the SEER data.

    While it is true that brain cancers can have a long latency period, the Swedish studies claim that 10 years of use is enough to show an increase in the brain cancer rate. Why then is this increase not showing up in the SEER data?

  4. Pingback: IARC Mobile Phone Radiation Classification Question | Pong Pulse Blog

  5. Pingback: Cupertino, We Have A Problem | Andrew Geller Online

  6. I have to wonder: If RF/MW electromagnetic fields are reclassified as a Class 2A– or even a Class 1 Carcinogen, what will that mean in the real world today? Do you think we will dismantle our wireless infrastructures and go back to wires?

    Which classification would be necessary for the schools to stop microwaving our children everyday–is that number defined?

    Is there a classification system for other hazards such as neurological and immune system dysregulation? Do we need one? I can’t remember why I wanted to know.

    If most people’s bodies are capable of repairing the harm–I mean “certain biological changes”– done by RF exposure, does that mean RF is harmless? Does this extend to other living things?

    Is there an acceptable percentage of people who can be expected to succumb over a certain period of time to RF induced illness? Can we measure this? If not, does it exist?

  7. I see that Devra Davis tweeted this post and that you reciprocated by tweeting a YouTube video.

    About 5-10 minutes or so, she indicated that ICNIRP was dominated by physicists and engineers who were not qualified to make decisions on public health. I have a few issues with this:

    1) ICNIRP has biologists and MDs on its committee.

    2) Dr. Davis, who posits herself as a responsible steward on EMF radiation, received her PhD. from the Philosophy department at the University of Chicago. Her thesis — “Conceptualizations of religion and science in some writings of Immanuel Kant and Auguste Comte” — scholarly as it may be does not seem at all relevant. What we have here is a philosopher-in-a-lab-coat critiquing a committee of genuine scientists.

    I am at a loss on why you associate with this self-promoting pseudo-scientist.

  8. Dariusz and Tom,
    The mobile phone can monitor more than transmitted power and duration. If the user operates a Bluetooth headset or a speaker, the mobile phone knows about it, it activates those devices. In addition the current smartphones can estimate the proximity of a human head by using the built in camera. So exposure can be estimated quite well.
    I think that since all this is well known, the industry should not to use the “we do not have enough information about exposure” argument too easily since this lack of information could have been corrected years ago.
    Michael Peleg

  9. For aabbcc1122 –

    “How come this knowledge couldn’t be considered to cause cancer?” – Good question!

    The answer can be found in conclusion section of the reference you provided a link to:

    “It is often difficult to evaluate the reported results, however, because details of the exposure in terms of frequency, duration, and intensity are quite variable, and sometimes poorly reported. This coupled with problems of measurement encountered in such studies, creates a rather confusing body of data from which to draw objective and absolute conclusions regarding the significance of the research.”

    Go back and read the review again.

    This time don’t just skip through it looking for snippets that confirm to your preconceived notions.

  10. Absolutely, there are many pieces to this equation… However, such app is a progress from the minutes/day “disinformation”…

  11. I agree that software modified handsets, as mentioned by Michael – and which have already been used in small scale validation studies, would certainly be a big step forward for exposure assessment. However, this would not be a silver bullet cure. Because the power output of the device is only one half of the equation.

    The use of texting, tethering, Bluetooth devices, etc., all have confounding effects on the determination of where, and how much of the emitted energy is actually deposited on the user of a software modified wireless handset.

  12. Professor:

    Are the French, Hardell and Interphone studies really replicates? How similar are the findings among them in the highest risk categories?

    I’m also concerned about a letter with your signature:

    This letter contains pseudoscience that includes conflating the toxicity of DDT with its Group 2B classification as well as unsubstantiated claims of the risk of carrying cell phones in bras.

    The letter was not worthy of scholarly endorsement and I was disappointed by your signature. You have better things to do.

  13. In 2011 IARC experts evaluated only peer-reviewed studies and reviews of studies to a smaller degree. We did not evaluate the pre-existing large body of various reports available from different organizations. We looked only at studies that were immediately and directly relevant to cancer. All other health hazards will be evaluated by ICNIRP and by the ongoing WHO evaluation.

  14. Michael, you are absolutely right. It is a wonder why epidemiological studies have such poor, in fact nonexistent, exposure data. Lack of the exposure data hampers evaluation of human impact of the exposures.
    That is why I joined the Advisory Board of a small start-up company in Finland – Cellraid ( The company has just developed a software, an app for the cell phone, that continuously measures real time exposures of cell phone users. Information is then stored and can be used in human studies, either epidemiological or experimental volunteer studies.
    I have such app on my phone and must honestly say that any epidemiological study using minutes/day spoken on the phone as a measure of exposure misses completely the point.

  15. Hi Dariusz,
    Thanks for your article, it really annoys me when I see authors of articles on EMR and EMF state such things as this: “While the science on the health impacts of this form of radiation is inconclusive, many people are concerned about how long-term exposure to excessive EMR may impact human health and nature”, and then they go onto say: “the health impacts of this form of radiation is inconclusive”, which is a complete distortion of the truth, and a blatant lie.

    If the United States Defence Intelligence Agency Medical Section released a classified document (DST-1810S-076-76 March 1976) on the “Biological Effect of Electromagnetic Radiation-USSR” that states very clearly the dangers and health risks by exposure of military personnel to microwave non-ionised radiation, which is the very same radiation that is emitted from mobile phones and towers, DECT phones, Wi Fi hotspots, Wi Max towers, wireless computers and modems, smart metres, including the local backyard HAM Operators, then how come this WHO committee didn’t consider any of the medical research in their findings.

    I obtained my full copy of the declassified DST-1810S-076-76 March 1976 document from this site:

    How could these two extracts from this document that is conclusive of the health risks from long-term and prolonged exposure to microwave non-ionised radiation not considered, because the document was available before 2011:

    (1) “If strict enforcement of stringent exposure standards, there could be unfavourable effects on industrial output (electronic companies) and military functions (wireless communications and electronic warfare)…personnel (military communicators) exposed to microwave radiation below thermal levels experience more neurological, cardiovascular, and haemodynamic disturbances than do their unexposed counterparts (military non-communicators).”

    (2) “Some of the…effects attributed to exposure include bradycardia, hypotension, and changes in EKG indices…subjects exposed to microwave exhibited a variety of neurasthenic (neurasthenia) disorders against background of angiodystonia (abnormal changes in tonicity of blood vessels). The most common subjective complaints were headaches, fatigue, perspiring, dizziness, menstrual disorders, irritability, agitation, tension, drowsiness, sleeplessness, depression, anxiety, forgetfulness, and lack of concentration.”

    How come this knowledge couldn’t be considered to cause cancer?

  16. As an engineer, I am not very proud of our ignorance about the exposures of the individual users. The exposure data can be collected easily by the mobile phone device since it controls and knows the transmitted power and is aware of each call duration. It can also record the efforts of the user to protect himself/herself by using a Bluetooth headset or a speaker. Such an application is easy to implement and should be present in every smartphone to provide essential information to the user and to the epidemiologist. It seems to me that such an information would render the epidemiological results even stronger than the very significant ones we see already today.

  17. Tom, if you read my blogs posted over the years since 2009, you will clearly see that I am no enthusiast of epidemiological studies. I frequently criticized Interphone and Hardell studies. However, the same result seems to repeat in recent French study. It is necessary to consider it. And, finally, the evidence that we have at hand we need to use in risk predictions… and Interphone, Hardell and French study are what we have… Unlike many, I do not say that these studies prove something. However, three replicates should be taken as clear warning sign. It is wrong to just ignore it…

  18. I think many people, including you, put far too much faith in the results of retrospective case control studies. They all share the same crippling weakness – inadequate exposure assessment.

    Variable and unpredictable power output is a characteristic of wireless handsets, as this parameter is automatically adjusted by the different base stations which the handset is associated with from time to time – based on the signal quality of the radio link. This, and the other features of wireless technology, means that a handset in talk mode, with a peak power output of 2 Watts, can actually have an average power output as low as .0000625 Watts; a dynamic range of 45 dB – a factor of 32,000!

    The uncertainty of random and unmeasured power output variations makes it very difficult to determine reasonably accurate individual exposure characteristics, even using personal dosimeters. Using the number of calls or cumulative call time as a surrogate of exposure adds further uncertainty to the determination of dose. In fact, it is impossible to assess the actual exposure to the agent studied with any confidence at all. Without a clear understanding of the dose received by individual study participants, the data collected is not informative.

    I think that prospective cohort studies and time trend analyses will provide much more valuable information with respect to the potential health effects of RF emissions. Of course they do take a lot more time.

    By the way – regarding an earlier article; when a base station adjusts the power levels downward, it also adjusts the modulation scheme and data rate to compensate for a lower quality link. A slower data rate will increase message length and transmit time. Extra airtime increases the average power output for handsets and base stations, while peak power remains the same.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.