select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to epigenetic changes in cells of smokers and vapers

A study published in Cancer Research looks at epigenetic changes in response to smoking.


Dr Mangesh Thorat, Honorary Reader at Queen Mary University of London, said:

“Epigenetic alterations, for example DNA methylation, have been associated with risk of various cancers even though a causal association has not been robustly established. This study utilised a very distinct set of samples from a range of different studies. While such range of samples offers opportunity to address multiple questions, such comparisons could be substantially confounded due to disparate nature of the samples. Indeed, as authors acknowledge, a possibility of gender bias also exists.

“This study observed DNA methylation changes in e-cigarette users that are similar to those in cigarette smokers. However, there is no clear dose-response relationship that one would expect and many of the similar changes occur early on. Therefore, it could be hypothesised that the changes observed here are a stress, wounding and healing response to the e-cigarette vapour and cigarette smoke alike.

“The changes observed may not be the long-term alterations that increase the risk of cancers in smokers. The study findings are interesting as such because these point to a potential for e-cigarette vapour to cause local tissue injury and cellular changes at least in the short-term. The data are not sufficient to comment on long-term harms associated with e-cigarette use, or any association such changes may have with the risk of developing cancer.”


Associate Professor George Laking, Director of the Centre for Cancer Research, University of Auckland, said:

“Framing is everything in discussions of risk. The authors of this study write that “despite widespread endorsement by Public Health England, who have advocated e-cigarettes as ‘95% less harmful’ than combustible cigarettes, recent studies have highlighted potential drawbacks”.  A finding of drawbacks would not in itself be a ‘despite’ – it can still be compatible with the ‘95% less harmful’ figure, that also means ‘5% harmful’.

“The paper finds that users of e-cigarettes had more hypermethylation in cells lining their cheeks, compared with never-smokers.  They say at least some of these changes were ‘similar’ to those seen in smokers.  One should beware of the word ‘similar’, as it can have a lot of different meanings. For example, a lethal snake is in some ways ‘similar’ to a harmless earthworm. It’s important to highlight there were other changes in the cells of smokers that weren’t seen in e-cigarette users or non-smokers.  

“The development of cancer is a multi-step process. In this study an early step appears to be at play in users of e-cigarettes. I don’t see this as especially new information, when viewed alongside reports about inflammatory changes in tissues of e-cigarette users. Cellular and tissue mechanisms of repair and cancer control exist, which can counter the effects of such changes. 

“Overall, this paper should not change the basic public health messages of ‘Vaping to Quit.’ Vaping remains an essential tool for harm reduction compared to cigarettes. The question with harm reduction is always ‘by how much is the harm reduced?’ Although that is a question a lot of people are working on (including all the people who vape!), it is not a question this study can answer.”


Prof Ross Lawrenson,  Professor in Population Health, University of Waikato, said:

“Does vaping cause harm? In particular, does it cause lung cancer? To show causation, we need to know the incidence of cancer in vapers compared with non-vapers, the strength of the association, whether the association is consistent across different populations, and whether it’s greater in those who vape a lot compared with those who vape less or who have been vaping for a shorter time. Lastly, we need to know whether there is a plausible explanation for why vaping could cause cancer.

“This new study suggests that vaping is associated with epigenetic changes in the lining of the cheek. This could be a plausible explanation for a causal association with cancer, but so far we do not have any evidence that vaping does cause cancer. It’s important to note that smoking has been associated not only with lung cancer but also cancer of the larynx, oral cavity, lip, bladder and cervix.

“These findings are from a group mainly concerned with Women’s Health, and provide support for those who believe vaping will prove harmful. However the paper does not provide sufficient evidence of harmful outcomes that will be needed before the public will be convinced to stop.”


Prof Andrew Beggs, Professor of Cancer Genetics & Surgery at the University of Birmingham and University Hospitals Birmingham, said:

“This interesting study shows e-cigarette use is associated with similar damaging changes in methylation in human cells as smoking. Although it doesn’t show a direct causal effect, this study shows that further research must be done to understand the effects of e-cigarettes on human health and whether they could be linked to an increased risk of cancer.”


Prof Lion Shahab, Co-Director of the Tobacco and Alcohol Research Group, University College London, said:

“This international study used a relatively large number of samples (N>3,500), collected from different sites (saliva, blood, cervix) and comprising different cell types (epithelial, immune and lymphoid), to assess epigenetic changes that may be associated with cancer development. The study included current, ex- and never smokers, and also included a subsample of vapers (N=116) who had never regularly smoked cigarettes.

“This comprehensive analysis confirms well-known epigenetic changes associated with smoking observed in blood, extending this to saliva and specific cell types, as well as identifying novel associations with cigarette smoking, which are also seen in separate cancer tissue and early lesion samples. Crucially, some of the epigenetic changes were also observed in saliva samples from vapers, albeit to a lesser extent and not across all cell types.

“While this study provides useful information about potential corollaries of e-cigarette use also seen with cigarettes, which need to be studied further, in my mind this does not provide proof that e-cigarettes cause cancer for several reasons:

  1. It is difficult to assess the extent to which the changes seen here translate into actual cancer development. Most cancers involve complex multicellular and multistep processes involving different biological systems, which can include epigenetic changes. However, these alone are unlikely to tell the whole story.
  2. The changes reported here are based on differences seen in smokers, some of which are also replicated in a smaller group of vapers. However, such similarities could be the result of confounding (e.g., if vapers are more likely to engage in a variety of unmeasured harmful behaviours that may drive epigenetic change seen in smokers). This study did not assess or control for this kind of confounding, which is compounded by the fact that the participants from the different datasets used here provided samples from different sites and likely varied in important characteristics.
  3. Because the group of vapers studied was relatively small, it was not possible to link epigenetic changes to vaping exposure in a clear dose-response manner. While the authors looked at vaping duration, there is only limited evidence that longer duration was associated with greater changes.
  4. The paper does not assess epigenetic changes unique to e-cigarettes but looked at those seen in smokers. This is of importance as we already know that e-cigarettes, compared with cigarettes, expose users to much lower levels of known tobacco-related carcinogens that drive cancer (in part because e-cigarettes do not involve combustion, which generates high levels of harmful chemicals seen in tobacco smoke). This means that the health consequences (and underlying causative processes) of e-cigarette use may be quite distinct from those of cigarettes use. These e-cigarette-specific changes need to be investigated in their own right.

“Notwithstanding these limitations, the findings in this paper highlight the continued need to elucidate the potential impact of vaping on disease, including assessing effects distinct from tobacco exposure. It is important to remember that e-cigarettes are a harm reduction product, not a risk-free product, aimed at those using uniquely harmful cigarettes. This paper does not change this risk calculus and serves to underscore the CMO’s advice: ‘if you smoke, vaping is much safer; if you don’t smoke, don’t vape’.”


Prof Peter Shields, Emeritus Professor of Medical Oncology at the Ohio State University, said:

“The journal Cancer Research is a really high quality journal, but epidemiologically this study does not lead to the conclusions the authors claim, with a substantial weakness for their smoking/vaping analysis.

“The authors report novel data but there are substantial limitations that are not mentioned in the manuscript.  They did their best to find suitable cohorts to study, but the problem is that they are combining very disparate types of study sets and so do not know what confounding they have.  To understand the overlap, they needed to show the results from each set for smokers, and then one for the vapers, smokers and never-smokers – without that, it is just impossible to visualize the overlap. 

“To make their claims of adverse effects, there are critical pieces of information that are missing.  They do not have good vaping/smoking data so mostly rely on each person saying yes/no, or self-reporting years of use.  This is very crude and without the dose-response assessment (e.g., packs per day, vapes per day, etc), they are still a far distance from being able to show causality.  For the vapers, there was no biochemical verification that they were actually not also smokers. 

“The crucial question is not whether vaping has epigenetic changes (which we cannot say establish a risky cell type or a physiological protective mechanism to prevent disease) – but how do vapers compare to smokers and never-smokers.  They really do not explore that well, but in fact the data looks like vapers are actually more like never-smokers – implying their risk of cancer is not increased by vaping!  This is not discussed.  The authors really need to be more cautious.”



‘DNA methylation changes in response to cigarette smoking are cell- and exposure-specific and indicate shared carcinogenic mechanisms with e-cigarette use’ by Chiara Herzog et al. was published in Cancer Research at 11:01pm UK time on Tuesday 19 March 2024, which is also when the embargo will lift.


DOI: 10.1158/0008-5472.CAN-23-2957



Declared interests

Andrew Beggs: no conflicts of interest relevant to this

Lion Shahab: no relevant COI

Peter Shields: in the past (>5 years ago) I have served as a consultant and expert witness in tobacco litigation on behalf of plaintiffs.

George Laking: “I Chair End Smoking New Zealand, an organisation that supports vaping to quit smoking. I’m also involved in research for vaping to quit smoking, and for nicotine replacement to quit vaping.”

Ross Lawrenson: No conflict of interest declared.

Mangesh Thorat: No conflicts

For all other experts, no reply to our request for DOIs was received.

in this section

filter RoundUps by year

search by tag