Skip to main content
Download PDF
Download ePub

Estimating lung cancer risk from e-cigarettes and heated tobacco products: applications of a tool based on biomarkers of exposure and of potential harm

Harm Reduction Journal volume 22, Article number: 45 (2025) Cite this article

Abstract

Background

Reliable epidemiological data are limited on the lung cancer risk of groups using e-cigarettes (ECIGs) and groups using heated tobacco products (HTPs).

Aim

We describe a methodology to estimate the lung cancer risk of these groups according to their levels of biomarkers of exposure (BOEs) and of potential harm (BOPHs).

Methods

Using 28 search terms for BOEs and 82 for BOPHs we sought publications reporting biomarker-specific data from North America and Europe comparing individuals who smoke cigarettes and individuals who use other established products (ETPs; cigars, pipes, smokeless tobacco (ST) and/or snuff/snus). Publications were selected using defined inclusion/exclusion criteria. Additionally using lung cancer relative risk (RR) estimates for users of specific ETPs derived from recent meta-analyses of epidemiological studies in these regions, we derived a regression model predicting the lung cancer RR by level of each specific biomarker. Separately for groups using ECIGs and using HTPs the lung cancer risk was then estimated by combining RR estimates for selected biomarkers. Our primary estimates only considered biomarkers statistically significantly (p < 0.01) related to lung cancer risk which showed no significant (p < 0.01) misfit to the RR of 1.0 for non-users—those with no use of ETPs, ECIGs or HTPs.

Results

Based on 38 available publications, we extracted biomarker-specific data for ETPs for 56 BOEs within 21 of the 28 search terms considered and for 54 BOPHs within 29 of the 82. The regression slope fitted to the lung cancer risk was significant (p < 0.01) for 22 BOEs and six BOPHs. However, the predicted RR for non-users significantly (p < 0.01) differed from 1.0 for 16 of these biomarkers. We estimated the lung cancer RR for using ECIGs, derived from 30 estimates for 10 biomarkers, as 1.88 (95% CI 1.60–2.22), the excess risk (ER = RR − 1) being 6.8% of that for smokers of cigarettes. The RR estimate varied little in most sensitivity analyses conducted, but increased markedly after removing the restriction to significant model fit.

We estimated the lung cancer RR for using HTPs, combining estimates for four BOEs, as 1.44 (0.41–5.08), the ER being 3.4% of that for smokers of cigarettes.

Conclusions

Despite some methodological limitations, our approach estimates risk when reliable epidemiological data are unavailable. Using the biomarkers considered here, the model indicates that the lung cancer risk for individuals using ECIGs is much lower than for smokers of cigarettes, and suggests that the risk for those using HTPs is also low. Research using additional data could add precision to these findings.

Background

E-cigarettes (ECIGs) were developed in the early 2000’s as a new tobacco product with a potentially reduced disease risk compared to cigarettes, and were first marketed in the USA in 2007 [1,2,3]. Since then, the prevalence of ECIG use has increased dramatically, though few current users in 2025 will have used them for much more than 10 years. Heated tobacco products (HTPs) were not introduced until about 10 years ago, and most current users in 2025 will have used them for less than 5 years. Particularly for ECIGs, and especially for ischaemic heart disease and stroke where the decline following quitting in the smoking-associated relative risk (RR) is relatively rapid [4, 5], useful epidemiological data could have been collected from a suitably designed large prospective study. This would compare risk for those smoking cigarettes at baseline who (a) subsequently continued to smoke cigarettes, (b) at differing times during follow-up, switched partly or wholly to the new product or (c) quit completely. However, no such study has been conducted, and good epidemiological data on risk, especially for lung cancer and chronic obstructive lung disease, where the decline in the RR on quitting is much slower [6, 7], may not be available for some time after 2025.

Given the paucity of epidemiological data, a critical population health question that remains unanswered is whether or not the use of ECIGs and HTPs is, in fact, safer than for smoking of cigarettes, and how to quantify the extent of the risk reduction. An early paper [3], which compared the risk of 12 nicotine-containing products, estimated that use of ECIGs caused only 5% of the harm of smoking of cigarettes. However, this was based on the opinions of experts using what was described as a “multi-criteria decision analysis approach”, with a noted limitation of the study being “the lack of hard evidence for the harms of most products on most of the criteria”. A number of other reviews, such as those of ECIGs [8,9,10], of HTPs [11, 12], or both [13,14,15], have used data from a range of biomarkers of exposure (BOEs) and of biomarkers of potential harm (BOPHs) to give insight into whether ECIGs and HTPs might in fact reduce disease risk, but generally do not reach any quantitative conclusions specifically for lung cancer. Some [12,13,14,15] only compare risk from use of ECIGs or HTPs with risk from smoking of cigarettes, and do not consider how biomarker levels and lung cancer risk varies between users of various established tobacco products (ETPs) other than cigarettes, including cigars, pipes, smokeless tobacco (ST) and snuff/snus.

Here, we apply a new approach to try to estimate quantitatively the excess lung cancer risk for users of ECIGs and HTPs relative to that for smokers of cigarettes. The applications of our approach combine data on the risk of lung cancer among individuals who smoke/smoked cigarettes and of users of the various ETPs, expressed relative to that of individuals who have never used tobacco products, and data on the levels of a range of BOEs and BOPHs measured in groups never using tobacco, using the ETPs, and using ECIGs and/or HTPs. For each biomarker considered, we attempt to use the available data for groups who have never used tobacco products and that for users of the different ETPs to fit a regression model relating the reduction in biomarker level, compared to that for those smoking cigarettes, to the corresponding reduction in excess RR (ER = RR-1). Then, for those biomarkers with a good fit to the model (showing a statistically significant (p < 0.01) relationship to lung cancer risk and no significant (p < 0.01) misfit to the RR of 1.0 for non-users of any product), we use the model to estimate the ER for ECIGs and for HTPs based on their levels of that biomarker. These estimates are then combined over a range of biomarkers taking into account the differing uncertainties of the individual ER estimates for the biomarkers considered.

Below we describe the methodology used to develop the equations used to predict the excess lung cancer RR for each biomarker, and to combine the estimates over a range of biomarkers showing a good model fit. We then use the methodology to predict the excess lung cancer RR based on data from publications which give levels of different sets of biomarkers for users of ECIGs and/or HTPs as well as for smokers of cigarettes.

We also make available to those interested both the database we have collected for the work described herein, as well as the methodological applications reported here. These applications also allow the user to extend our estimates given additional biomarker data, and also to derive RR estimates relating to ECIG and HTP use for diseases other than lung cancer, given suitable input data on RRs related to use of the various ETPs.

Methods

RRs for ETPs

A recent meta-analysis [16] provides separate estimates of the lung cancer RR for individuals smoking cigarettes, cigars and pipes based on epidemiological studies conducted in the US and Europe, and published in English in 2010–2020. For cigarettes, it estimated the RRs for smokers relative to individuals who have never smoked as 15.15 (95% CI 12.77–17.96) for North America and as 12.30 (9.77–15.49) for Europe, and the work described here is based on a combined estimate of 13.86 (11.32–16.96). For cigars and pipes, where there are far fewer data, the cited RRs of 2.73 (2.36–3.15) for cigar smoking and of 4.93 (1.97–12.32) for pipe smoking are used, each RR being estimated relative to individuals never using the product. Estimates of 1.59 (1.06–2.39) for individuals using ST and of 0.80 (0.40–1.30) for snuff/snus use are taken from another recent meta-analysis [17], based on publications in 1990–2020.

Identifying a candidate list of biomarkers for study

We identified (see Tables 1 and 2) BOEs and BOPHs based on recent publications [8, 11, 13, 15, 18,19,20,21,22] using 28 search terms for BOEs and 82 for BOPHs. We ignore biomarkers directly related to nicotine exposure, it being well known that nicotine levels in those smoking cigarettes and those using snus are quite similar despite their very different lung cancer risks [16, 17]. We also do not include carbon monoxide as, according to Braznell et al. (2024) [11], there is no association with lung cancer risk, or chemicals related to propylene glycol and glycol, as these predominately result from use of ECIGs and not from use of ETPs.

Table 1 Search results for BOE data for ETPs
Full size table
Table 2 Search results for BOPH data for ETPs
Full size table

Initial searches—for biomarker data for ETPs

For each search term listed in Tables 1 and 2, we carried out Medline searches for papers including (in their title or abstract) the search term and also one or more references to pipe, cigar, ST, snus or snuff, it being clear that studies only investigating cigarettes or smoking (unidentified) are useless for comparing biomarker levels between the products of interest. Thus, for example for BOE search term 1 (acrylamide), we used the following query:

(Acrylamide) AND (“tobacco, smokeless” [Mesh Terms] OR “Pipe Smoking” [Mesh Terms] OR “cigar” [All Fields] OR “snus” [All Fields] OR “pipe” [All Fields] OR “snuff” [All Fields] OR “smokeless” [All Fields])

 

Examination of the output from the initial searches

For each search with non-zero hits, PNL first examined the search output, his conclusions then being checked by KJC. This output consists, for each paper, of the full reference, the title, the authors and the abstract. From this information, we selected papers appearing likely to provide relevant data. We used the following exclusion criteria to reject papers:

  • no relevant results for any selected biomarker;

  • results only available for a single ETP (results for water pipes being ignored);

  • no results specifically for smokers of cigarettes (for example. the paper referred to smoking but did not clarify if this was smoking of cigarettes or of any product);

  • results only available for heavy (> 10 per day) smokers of cigarettes;

  • the paper was nothing to do with tobacco;

  • described only a study on cells or animals;

  • described only a study on diseased patients;

  • described only a study on a population with specific exposures;

  • concerned a study not in North America or Europe;

  • considered marker levels measured only in the product and not in the individual;

  • described results of a trial in which users were tested after use of a range of products; or

  • was a review, rather than a study reporting biomarker levels.

Examination of the papers on ETPs identified as possibly relevant

Each paper identified from the title and abstract as possibly having relevant data was then obtained and examined, initially by PNL and then by KJC. For each paper which they agreed had no relevant data, a reason for exclusion was given, the remaining papers then being passed on for possible data entry. The papers passed on were then examined in more detail, with further exclusions and reasons for exclusion agreed by PNL and KJC.

Entry of biomarker data for ETPs

For each paper meeting our inclusion criteria, we entered the following data: reference ID, study location (country), year of publication, sex(es) for which data were available, and for each biomarker with data available, the biomarker search number and name, biomarker short and long name, data source in the paper (for example the Table or Figure number), matrix (for example urine), units, and means (geometric or not). For smoking of cigarettes, the data we recorded for each biomarker included the number of subjects for which data were available, the definition of use, and the mean level and either the standard deviation or the 95% CI. We entered similar information, where available, for non-users of tobacco, for each other ETP, and for ECIGs and HTPs. We also recorded details of biomarker data not entered and why, and any relevant comments.

Predicting the lung cancer RR based on levels of a given biomarker

For each biomarker, we used the available data from each study on biomarker levels for cigarettes, other ETPs and non-users, and the data on lung cancer RRs by ETP to derive a formula predicting the lung cancer RR relative to non-users (and its standard error) corresponding to any given biomarker level.

The model used was:

$$\text{ln}\left(\frac{{RR}_{C}}{{RR}_{p}}\right)=\beta .\text{ln}\left(\frac{Bio{M}_{C}}{Bio{M}_{P}}\right)+ \varepsilon$$

where RRC is the RR for cigarette smoking,

RRP is the RR for product P (cigars, pipe, ST or snuff/snus),

BioMC is the mean value of the biomarker for individuals smoking cigarettes,

BioMP is the mean value of the biomarker for individuals using product P,

β is the estimate of the slope relating the log RR ratios to the log biomarker ratios, and

\(\varepsilon\) is the error term, assumed to be an independent and identically distributed normal variable with mean 0 and variance σ2.

It should be noted that we used:

  1. 1)

    ratios of the biomarkers, so that the units used for the measurement of the biomarker would be irrelevant;

  2. 2)

    a model with no intercept so that when the product was set to be cigarettes it would be expected that the log RR ratio would be zero;

  3. 3)

    RRs associated with use of the different products taken from the literature as described above in the section “RRs for ETPs”; and

  4. 4)

    a model fitted using weighted estimates of the log ratios of the biomarker means. First, we calculated the standard error for each biomarker mean from the data available. Next, we used the approximation V[log(X)]≈E[X]−2V[X] to estimate the variance of the log biomarker mean. Finally we set the weighting to be the inverse of the sum of variances for the two log biomarker means forming the ratio.

Combining predictions for a set of biomarkers

For the purposes of combining predictions for a set of biomarkers we mainly restricted attention to “good” biomarkers showing a significant slope (p < 0.01) in the fitted model (suggesting a real relationship between the biomarker level and the RRs), and where the model did not show a significant (p < 0.01) misfit to the RR of 1.0 for non-users.

We then derived the predicted values of log RR as a weighted average of the individual predictions, using the inverse of the square of the SEs as the weighting. This procedure also returned the standard error of the predicted value, a t-test on whether the log RR was 0 (equivalent to no excess RR), and lower and upper CIs. Finally, we obtained estimates of the RR with lower and upper 95% CIs by taking e to the power of the log values, and using an approximate standard error estimated from the difference between the high and low 95% CIs divided by 2*Normal Inverse (0.975).

We then used this methodology to produce combined RR estimates for ECIGs and for HTPs and to check the validity of the modelling for non-users.

Sensitivity analyses

We carried out five sets of sensitivity analyses, the first four used in relation to the combined ECIG and HTP estimates and all five being used in relation to the combined estimate for non-users.

Sensitivity analysis 1 restricted attention to biomarkers where there were either at least four, or at least six, separate estimates of the ratio of the biomarker values for use of another product (cigars, pipes, ST, or snuff/snus) compared to that for smoking cigarettes.

Sensitivity analysis 2 restricted attention to biomarkers where the fitted \(\beta\) value was statistically significant at least at p < 0.05 or p < 0.001, or did not restrict attention at all on the significance of the fitted \(\beta\) value.

Sensitivity analysis 3 restricted attention to data from studies with “well-defined groups”, where the definition of the smoking groups was clearly exclusive; so that those classified as currently using one of the five ETPs (cigarettes, cigars, pipes, ST, snuff/snus) were not currently using any of the others.

Sensitivity analysis 4 varied the RR for snuff/snus from the value of 0.8 reported above in the section “RRs for ETPs” to the value of 1.00 as it might be considered unlikely that use of snuff/snus actually reduces the risk of lung cancer.

Sensitivity analysis 5 gave separate estimates based either on BOEs or on BOPHs.

Note that while sensitivity analyses 1, 2 and 5 use the fitted regression coefficients from the main model, sensitivity analyses 3 and 4 use alternative regression coefficients, either those fitted to the data from the studies with “well-defined groups” or those using the alternative RR estimate for individuals using snuff/snus.

Further searches—for biomarker data for ECIGs and HTPs

For each broad search term with relevant data identified for ETPs, we carried out further Medline searches for publications that included (in their title or abstract) the search term and one or more references to ECIGs or HTPs. Thus, for example for BOE search term 1 (acrylamide), the search used the following query:

(Acrylamide) AND ((((((((((((("Vaping"[Mesh]) OR (vaping)) OR (ECig)) OR (E-cig)) OR (Ecigarette)) OR (E-cigarette)) OR (Vape)) OR (vapes)) OR (ENDS)) OR ("electronic nicotine delivery system")) OR (electronic nicotine delivery system)) OR ("Electronic Nicotine Delivery Systems"[Mesh])) OR ((((((((((((Heated tobacco product) OR (heated tobacco products)) OR (HTP)) OR ("heat not burn")) OR (tobacco heating system)) OR ("tobacco heating system")) OR (electrically heated cigarette smoking system)) OR ("electrically heated cigarette smoking system")) OR (carbon heated tobacco product)) OR ("carbon heated tobacco product")) OR (carbon-heated tobacco product)) OR ("carbon-heated tobacco product")))

 

Examination of the output from the further searches

For each search with non-zero hits, KJC examined the output first, and PNL then checked it, this output, for each paper, consisting of the full reference, title, authors and abstract. From this information, papers appearing likely to provide relevant data were selected. A paper was excluded if:

  • it contained no relevant results for any specific biomarker where regression equations had been derived above;

  • it provided no results for individuals smoking cigarettes;

  • no results were available for individuals using either ECIGs or HTPs; or

  • for any other reason listed above under “Examination of the output from the initial searches”.

Examination of the papers on ECIGs and HTPs and entering the data

The procedures were as described above in the corresponding sections for ETPs.

Estimating the lung cancer RR for use of ECIGs and HTPs and conducting sensitivity analyses

The available data for ECIGs and NTPs were then used to estimate the lung cancer RR and carry out sensitivity analyses using the regression models developed from the ETP data as described above.

Results

Initial examination of the searches for biomarker data for ETPs

As shown in Tables 1 and 2 there were 657 hits from the 28 searches for data on BOEs, and 3,741 hits from the 82 searches for data on BOPHs. Initially PNL identified 442 papers as possibly relevant from examination of the abstracts, but on checking by KJC this reduced to 405. 183 of these were duplicates identified in multiple searches, leaving 222 papers to be obtained.

Initial examination of the papers on ETPs

Among the 222 papers, full texts for two studies [23, 24] could not be accessed, and initial examination of the 220 obtained suggested that 74 could be considered for data entry, the excluded papers being listed in Additional File 1 Table 1, with reasons for exclusion given, the most common reason being that the paper contained no relevant results.

Data entry from the initial searches

As shown in Additional File 1 Table 2, we excluded about half the papers considered after initial examination, common reasons for exclusion being that smoking was undefined or included multiple products. Finally, as shown in Table 3, data were entered from 38 of the papers identified and also from two additional papers [20, 21] obtained from other sources which provided data for ECIGs and HTPs. As can be seen in Table 3, all the papers provided biomarker results for groups smoking cigarettes, with all but five providing results for non-users of tobacco. Each paper provided results for at least one ETP of interest, with eight giving results for cigars, four for pipes, 18 for ST and 14 for snus or snuff. One paper identified in the searches [19] provided data for ECIGs and HTPs, as well as for cigarettes, ST and non-use. As shown in Tables 1 and 2, relevant biomarker data proved to be available for 56 BOEs, falling within 21 of the 28 groups considered, and for 54 BOPHs, falling within 29 of the 82 groups considered. The data recorded for each study and each biomarker are given in Additional File 2.

Table 3 Papers accepted for data entry from initial searches
Full size table

Main regression equations for each biomarker fitted to the ETP data

Additional File 2 also gives the results of the regression equations for each biomarker. Tables 4 (BOE) and 5 (BOPH), with the summary for each biomarker giving its search number, the biomarker name, the number of observations on which the regression analysis was based, and the fitted slope, together with its standard error, t value and p value, and finally the maximum significance for the non-users. For each table the results are shown in increasing order of the p value, so the biomarkers with the strongest relationship between the biomarker level and the RR appear first. Note that results for the BOE CYMA, identified under search term 2, and the BOE CEMA, identified under search term 16, were combined as CYMA under search term 2 as CYMA and CEMA are alternative abbreviations for the same chemical, and the majority of the results were found under search term 2. Note also that results for one particular BOPH, nitrotyrosine (search term 55), do not appear in Table 5 as the means were the same in each tobacco group for which data were available, so a slope could not be fitted.

Table 4 Summary of fitted regression slopes for BOEs shown in increasing order of p value
Full size table
Table 5 Summary of fitted regression slopes for BOPHs shown in increasing order of p value
Full size table

As shown in Table 4, 22 of the 56 BOEs considered have a fitted slope significant at p < 0.01, including five of the seven markers for phenanthrene and all three of the markers for naphthalene. Further restricting attention to those that did not show a significant (p < 0.01) misfit from an RR of 1.0 for non-users, leaves the nine BOEs underlined in Table 4:—two markers of acrylamide (AAMA and GAMA), two of acrylonitrile (CYMA and HEMA), two of phenanthrene (2/3-OH-Phe and 4-OH-Phe), and one each of anabasine (NAB), anatabine (NAT) and toluene (Toluene).

As seen in Table 5, six of the 53 BOPHs have a fitted slope significant at p < 0.01, each relating to a different search term. Three of these showed no significant (p < 0.01) misfit for non-user:—markers of adhesion molecule (SICAM-1), interleukin (IL-6) and isoprostane (8,12-iso-iPF2α-VI).

Further details of the regressions are given in Additional File 3, which presents separate plots for each biomarker studied. In the plots the x-axis is the ratio of the mean biomarker value for those smoking cigarettes to that for those using the other ETP, while the y­axis is the lung cancer RR (relative to never smokers) for those smoking cigarettes relative to those using the other ETP. The scales for the axes are both in logarithmic form as the regression is based on the logs of those values. The plots show the fitted regression line and show the individual points to which the line is fitted, with different colours distinguishing the different other products (pipe, snuff, cigar, ST and none). On the regression line the default point for those smoking cigarettes is marked as a black circle.

Figures 1 and 2 give examples of the plots. The first is for the acrylonitrile BOE CYMA where the mean values for those using cigarettes, cigars, ST and never tobacco reduce in correspondence with the lung cancer RRs for the products and the regression slope is highly significant (p = 0.0000). The second is for the BOE anabasine where the fit to the model is poor, because levels for users of ST and snuff are higher than for smokers of cigarettes, despite the much higher lung cancer risk for smokers of cigarettes.

Fig. 1
figure 1

Plot for biomarker CYMA

Full size image
Fig. 2
figure 2

Plot for Biomarker Anabasine

Full size image

Initial examination of the searches for biomarker data for ECIGs and HTPs

As shown in Tables 6 and 7 there were 1,451 hits from the 21 searches for data on BOE, and 5,440 hits from the 29 searches for data on BOPH. Initially PNL identified 209 papers as possibly relevant from examination of the abstracts, but on checking by KJC this reduced to 182, 84 of which were duplicates identified in multiple searches, leaving 98 papers to be obtained.

Table 6 Search results for BOE data for use of ECIGs and HTPs
Full size table
Table 7 Search results for BOPH data for use of ECIGs and HTPs
Full size table

Initial examination of the papers on ECIGs and HTPs

Initial examination of the papers suggested that 50 could be considered for data entry, the excluded papers being listed in Additional File 4 Table 1, with reasons for exclusion given. This additional file also cites seven exclusions of comments on these papers not considered elsewhere.

Data entry from the further searches

As shown in Additional File 4 Table 2, 21 papers considered after initial examination were excluded for various reasons. Finally, as shown in Table 8, data were available for ECIG and/or HTP users from 29 papers, including some considered in our earlier searches. Each paper provided biomarker results for those smoking cigarettes, with 24 providing results for non-users of tobacco, 27 for ECIGs and 7 for HTPs. The actual data recorded for each study and for each biomarker are given in Additional File 5.

Table 8 Papers accepted for data entry from further searches
Full size table

Estimated lung cancer RRs for ECIGs

As shown in Table 9, the estimated RR (95% CI) for ECIG use, based on 30 individual estimates for 10 different biomarkers, is 1.88 (1.60–2.22), with the estimate very similar to this based on the 23 estimates for BOHs (1.88, 1.55–2.28) or on the 7 estimates for BOPHs (1.89, 1.23–2.91). The 29 estimates vary between 0.18 and 3.21, each indicating a much lower RR than the combined estimate of 13.86 for smoking of cigarettes. Indeed, the estimated excess risk (ER) of 0.88 for use of ECIGs is only about 6.8% of the ER of 12.86 for smoking of cigarettes.

Table 9 Estimated RRs for use of ECIGs for individual studies and biomarkers and combined
Full size table

Sensitivity analyses for ECIGs

Table 10 summarizes the results of the sensitivity analyses. While most results indicate little variation in the overall estimate for those using ECIGs, the estimates are substantially increased if the restriction to biomarkers with a slope that is significant at p < 0.01 is lifted. Indeed, with no restriction, the estimate is as high as 13.03 (95% CI 11.91–14.25). This estimate is based partly on five extremely high individual RR estimates; 340.62 for the BOPH marker deoxyguanosine, derived from data for one study [25], 155.44, 87.46 and 66.92 for the BOPH markers haematocrit, platelets and haemoglobin, based on data for another study [26] and 52.99 for a marker of the BOE carbon disulfide from data for a third study [27]. The implausibility of these five estimates, and the need to insist on the model fitting the data very well, is emphasised by various considerations.

Table 10 Estimated RRs for use of ECIGs. Results of sensitivity analyses
Full size table

First, as can be seen in Additional file 5, all these five estimates were derived using data only from a single study showing relatively small differences between biomarker levels for smokers of cigarettes, non-users and users of either cigars [28], chewing tobacco [29] or snuff [30, 31]. Second, when the fitted models were used to predict RRs for non-users based on the ratio of biomarker levels for non-users to those smoking cigarettes from the ECIG studies providing data for these five biomarkers, the 95% CI for the predicted RR for non-users never included 1.0 and in many cases was similar to or greater than that for those smoking cigarettes. For example non-user RR estimates were 24.39 (5.47–108.78) from data for one study of platelets [26], 18.87 (8.34–42.66) from another study of platelets [32], 13.86 (13.86–13.86) from a study of haemoglobin [33] and 10.03 (6.69–15.03) from a study of deoxyguanosine [34]. Interestingly, when there was no restriction to biomarkers based on the significance of their fitted slope, but there was a restriction to using data for studies with “well-defined groups” the ECIG RR estimate of 12.89 (11.73–14.16) dropped dramatically to 2.06 (1.74–2.43), mainly because the first four of the five very high RR estimates cited above occurred in studies where the study was not classified as having “well-defined groups” as the definition of the smoking groups was not clearly exclusive.

Estimated lung cancer RRs for HTPs

As shown in Table 11 the estimated lung cancer RR for use of HTPs was 1.44 (95% CI 0.41–5.08). This estimate is based only on data identified in our initial searches, as no additional studies were found in our further searches that included biomarkers satisfying the requirements for inclusion in the combined estimates. Two studies [35, 36] not considered in our initial searches did provide data for blood pressure, body weight and cholesterol, while one study [37] provided data for blood pressure and body weight, and one [18] data for biomarkers of aminobiphenyl, naphthalene and toluene, but none of these biomarkers were considered adequate predictors to be considered in a combined analysis.

Table 11 Estimated RRs for use of HTPs for individual studies and biomarkers and combined
Full size table

Availability of data and software

Statistical analyses were carried out using R Version 4.2.2 (2022-10-31) for linear modelling, Available on request are the data files we have used, and an R program which enables users to enter biomarker data from their own studies and estimate the lung cancer RR of the product(s) considered. This is also available as a Shiny App at https://roelee.shinyapps.io/Biomarkers/

Discussion

We have described a method to estimate the lung cancer risk for those using new tobacco products such as ECIGs and HTPs, based on North American and European data on biomarker levels for those smoking cigarettes and those using other ETPs, combined with recent estimates of the lung cancer RR of these products relative to individuals who do not use tobacco products. When combining data from multiple biomarkers, we have restricted attention to biomarkers which significantly (p < 0.01) fit the regression model we used, and which did not significantly (p < 0.01) misfit the estimate of 1.0 for non-users.

Limitations

Below, we discuss some limitations of the work we have described. It should be made clear that many are not limitations of the methodology itself, but of our application of it, and could be overcome by additional research using our methods.

One limitation is the list of biomarkers considered, which some may regard as incomplete. The set of data considered could also be extended, perhaps by searching on detailed biomarker names as well as the broad search terms we considered (as listed in Tables 1 and 2). We did not consider variation in biomarker levels or lung cancer risk by sex, age, exposure to other hazards (such as those resulting from environmental and occupational exposure, including second-hand smoke and radon gas) or sub-types of tobacco product (for example, menthol versus non-menthol cigarettes, small versus large cigars, or different types of ST). We only considered risk of lung cancer. However, given up-to-date RR estimates for other common smoking-related diseases and data for additional biomarkers possibly related to these diseases, our software can readily be used to generate predicted RRs for ECIGs and HTPs.

Levels of some biomarkers we considered might result partly from exposures nothing to do with cigarettes, other ETPs, ECIGs or HTPs. In principle this does not matter, as if the data available for the biomarker for users of ETPs and non-users adequately fits the regression model, it can still be used to estimate risk in users of new tobacco products. Perhaps more of a concern is where the set of biomarkers used to estimate risk includes multiple correlated biomarkers of the same exposure. Here analyses could be run limiting attention to at most one biomarker for any given exposure. We did not attempt such an approach, but the results in Table 9 clearly show that analyses based on any subset of the biomarkers we considered would still have predicted a lung cancer risk in ECIG users much lower than that for smokers of cigarettes.

An important limitation of our approach is the failure to include biomarkers for chemicals present in ECIGs or HTPs that, in the ETPs we considered, are either not present, are present in very small amounts, or which have not been measured. Our method will not, however, help if use of ECIGs or HTPs happens to cause a disease not known to be related to smoking.

Further considerations

Our combined lung cancer RR estimates of 1.88 (95% CI 1.66–2.22) for ECIG use and 1.44 (95% CI 0.41–5.08) for HTP use clearly suggest that the RR for use of these two products is substantially lower than that for smoking of cigarettes, though the data for HTPs are very limited, with the RR having very wide CI. Our analyses for ECIGs clearly indicate that attention should be restricted to those biomarkers showing an adequate fit to the model.

Until reliable epidemiological results become available, future research could usefully extend our results by including further data for those biomarkers we used in our estimates, taking account of additional biomarkers thought relevant, and considering other smoking-related diseases. The fact that our database and software are available to carry out further analyses should facilitate this future research. We emphasise, however, that our methodology will not help if using ECIGs or HTPs causes some disease not known to be related to smoking.

Conclusions

We describe a methodology to derive estimates of lung cancer risk for users of new tobacco products such as ECIGs and HTPs, from biomarker data for these products. Our methodology has limitations, but seems useful for estimating disease risk in the absence of epidemiological data. Applying it, and limiting attention to biomarkers satisfactorily fitting the model, indicates that the lung cancer risk from using ECIGs is much less than that from smoking cigarettes. Limited data also suggests that the risk from using HTPs is relatively low. Future research can extend these results based on data for additional biomarkers and smoking-related diseases.

Availability of data and materials

See final results section of paper "Availability of data and software".

Abbreviations

BOE:

Biomarker of exposure

BOPH:

Biomarker of potential harm

CI:

Confidence interval

ECIG:

E-cigarettes

ER:

Excess risk

ETP:

Established tobacco product

HTP:

Heated tobacco product

RR:

Relative risk

ST:

Smokeless tobacco

References

  1. Bittoni MA, Carbone DP, Harris RE. Vaping, smoking and lung cancer risk. J Oncol Res Ther. 2024. https://doiorg.publicaciones.saludcastillayleon.es/10.29011/2574-710x.10229.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Jackson SE, Brown J, Beard E. Associations of prevalence of e-cigarette use with quit attempts, quit success, use of smoking cessation medication, and the overall quit rate among smokers in England: a time-series analysis of population trends 2007–2022. Nicotine Tob Res. 2024;26(7):826–34. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntae007.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Nutt DJ, Phillips LD, Balfour D, Curran HV, Dockrell M, Foulds J, et al. Estimating the harms of nicotine-containing products using the MCDA approach. Eur Addict Res. 2014;20(5):218–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1159/000360220.

    Article  PubMed  Google Scholar 

  4. Lee PN, Fry JS, Hamling JS. Using the negative exponential distribution to quantitatively review the evidence on how rapidly the excess risk of ischaemic heart disease declines following quitting smoking. Regul Toxicol Pharmacol. 2012;64:51–67. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yrtph.2012.06.009.

    Article  PubMed  Google Scholar 

  5. Lee PN, Fry JS, Thornton A. Estimating the decline in excess risk of cerebrovascular disease following quitting smoking: a systematic review based on the negative exponential model. Regul Toxicol Pharmacol. 2014;68(1):85–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yrtph.2013.11.013.

    Article  PubMed  Google Scholar 

  6. Fry JS, Lee PN, Forey BA, Coombs KJ. How rapidly does the excess risk of lung cancer decline following quitting smoking? A quantitative review using the negative exponential model. Regul Toxicol Pharmacol. 2013;67:13–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yrtph.2013.06.001.

    Article  PubMed  Google Scholar 

  7. Lee PN, Fry JS, Forey BA. Estimating the decline in excess risk of chronic obstructive pulmonary disease following quitting smoking - a systematic review based on the negative exponential model. Regul Toxicol Pharmacol. 2014;68(2):231–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yrtph.2013.12.006.

    Article  PubMed  Google Scholar 

  8. Song MA, Freudenheim JL, Brasky TM, Mathe EA, McElroy JP, Nickerson QA, et al. Biomarkers of exposure and effect in the lungs of smokers, nonsmokers, and electronic cigarette users. Cancer Epidemiol Biomarkers Prev. 2020;29(2):443–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.Epi-19-1245.

    Article  PubMed  CAS  Google Scholar 

  9. Wilson N, Summers JA, Ait Ouakrim D, Hoek J, Edwards R, Blakely T. Improving on estimates of the potential relative harm to health from using modern ENDS (vaping) compared to tobacco smoking. BMC Public Health. 2021;21(1):2038. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-021-12103-x.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  10. Yayan J, Franke KJ, Biancosino C, Rasche K. Comparative systematic review on the safety of e-cigarettes and conventional cigarettes. Food Chem Toxicol. 2024;185:114507. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.fct.2024.114507.

    Article  PubMed  CAS  Google Scholar 

  11. Braznell S, Campbell J, Gilmore AB. What can current biomarker data tell us about the risks of lung cancer posed by heated tobacco products? Nicotine Tob Res. 2024;26(3):270–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntad081.

    Article  PubMed  CAS  Google Scholar 

  12. Znyk M, Jurewicz J, Kaleta D. Exposure to heated tobacco products and adverse health effects, a systematic review. Int J Environ Res Public Health. 2021. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijerph18126651.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Akiyama Y, Sherwood N. Systematic review of biomarker findings from clinical studies of electronic cigarettes and heated tobacco products. Toxicol Rep. 2021;8:282–94. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.toxrep.2021.01.014.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  14. Scherer G. Suitability of biomarkers of biological effects (BOBEs) for assessing the likelihood of reducing the tobacco related disease risk by new and innovative tobacco products: A literature review. Regul Toxicol Pharmacol. 2018;94:203–33. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yrtph.2018.02.002.

    Article  PubMed  CAS  Google Scholar 

  15. Scherer G, Pluym N, Scherer M. Intake and uptake of chemicals upon use of various tobacco/nicotine products: Can users be differentiated by single or combinations of biomarkers? Contrib Tobacco Nicot Res. 2021;30(4):167–98. https://doiorg.publicaciones.saludcastillayleon.es/10.2478/cttr-2021-0014.

    Article  CAS  Google Scholar 

  16. Lee PN, Coombs KJ, Hamling JS. Evidence relating cigarette, cigar and pipe smoking to lung cancer and COPD. Meta-analysing recent data from three regions. World J Metaanal. 2023;11(5):228–52. https://doiorg.publicaciones.saludcastillayleon.es/10.13105/wjma.v11.i5.228.

    Article  Google Scholar 

  17. Lee PN, Coombs KJ, Hamling JS. Review with meta-analysis relating North American, European and Japanese snus or smokeless tobacco use to major smoking-related diseases. World J Metaanal. 2022;10(3):130–42. https://doiorg.publicaciones.saludcastillayleon.es/10.13105/wjma.v10.i3.130.

    Article  CAS  Google Scholar 

  18. Scherer G, Riedel K, Pluym N, Scherer M. Assessment of the exposure to aromatic amines in users of various tobacco/nicotine products. ACS Omega. 2022;7(45):41775–82. https://doiorg.publicaciones.saludcastillayleon.es/10.1021/acsomega.2c06130.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Scherer G, Scherer M, Mütze J, Hauke T, Pluym N. Assessment of the exposure to tobacco-specific nitrosamines and minor tobacco alkaloids in users of various tobacco/nicotine products. Chem Res Toxicol. 2022;35(4):684–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1021/acs.chemrestox.2c00020.

    Article  PubMed  CAS  Google Scholar 

  20. Scherer G, Scherer M, Rögner N, Pluym N. Assessment of the exposure to polycyclic aromatic hydrocarbons in users of various tobacco/nicotine products by suitable urinary biomarkers. Arch Toxicol. 2022;96(11):3113–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00204-022-03349-4.

    Article  PubMed  CAS  Google Scholar 

  21. Scherer G, Pluym N, Scherer M. Comparison of urinary mercapturic acid excretions in users of various tobacco/nicotine products. Drug Test Anal. 2023;15(10):1107–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/dta.3372.

    Article  PubMed  CAS  Google Scholar 

  22. Yach D, Scherer G. Applications of biomarkers of exposure and biological effects in users of new generation tobacco and nicotine products: Tentative proposals. Drug Test Anal. 2023;15(10):1127–32. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/dta.3567.

    Article  PubMed  CAS  Google Scholar 

  23. Kershbaum A, Bellet S. Cigarette, cigar, and pipe smoking. Some differences in biochemical effects. Geriatrics. 1968;23(3):126–34.

    PubMed  CAS  Google Scholar 

  24. Kershbaum A. A comparative study of cigarette, cigar and pipe smoking effects on blood lipids, catechlolamine excretion and nicotine content of the urine. Acta Cardiol. 1968;23(4):317–29.

    PubMed  CAS  Google Scholar 

  25. Sakamaki-Ching S, Williams M, Hua M, Li J, Bates SM, Robinson AN, et al. Correlation between biomarkers of exposure, effect and potential harm in the urine of electronic cigarette users. BMJ Open Respir Res. 2020. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bmjresp-2019-000452.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Wang Y, Zhu Y, Chen Z, Chen S, Fu G, Fu J. Association between electronic cigarettes use and whole blood cell among adults in the USA-a cross-sectional study of National Health and Nutrition Examination Survey analysis. Environ Sci Pollut Res Int. 2022;29(59):88531–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11356-022-21973-6.

    Article  PubMed  CAS  Google Scholar 

  27. Anic GM, Rostron BL, Hammad HT, van Bemmel DM, Del Valle-Pinero AY, Christensen CH, et al. Changes in biomarkers of tobacco exposure among cigarette smokers transitioning to ENDS use: the Population Assessment of Tobacco and Health Study, 2013–2015. Int J Environ Res Public Health. 2022. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijerph19031462.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Chang CM, Rostron BL, Chang JT, Corey CG, Kimmel HL, Sosnoff CS, et al. Biomarkers of exposure among US adult cigar smokers: Population Assessment of Tobacco and Health (PATH) Study Wave 1 (2013–2014). Cancer Epidemiol Biomarkers Prev. 2019;28(5):943–53. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.Epi-18-0539.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Giraud DW, Martin HD, Driskell JA. Erythrocyte and plasma B-6 vitamer concentrations of long-term tobacco smokers, chewers, and nonusers. Am J Clin Nutr. 1995;62(1):104–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ajcn/62.1.104.

    Article  PubMed  CAS  Google Scholar 

  30. Eliasson M, Lundblad D, Hägg E. Cardiovascular risk factors in young snuff-users and cigarette smokers. J Intern Med. 1991;230(1):17–22. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1365-2796.1991.tb00401.x.Available.

    Article  PubMed  CAS  Google Scholar 

  31. Prasad GL, Jones BA, Chen P, Gregg EO. A cross-sectional study of biomarkers of exposure and effect in smokers and moist snuff consumers. Clin Chem Lab Med. 2016;54(4):633–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1515/cclm-2015-0594.

    Article  PubMed  CAS  Google Scholar 

  32. Badea M, Gaman L, Delia C, Ilea A, Leasu F, Henríquez-Hernández LA, et al. Trends of lipophilic, antioxidant and hematological parameters associated with conventional and electronic smoking habits in middle-age Romanians. J Clin Med. 2019. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/jcm8050665.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Tattersall MC, Hughey CM, Piasecki TM, Korcarz CE, Hansen KM, Ott NR, et al. Cardiovascular and pulmonary responses to acute use of electronic nicotine delivery systems and combustible cigarettes in long-term users. Chest. 2023;164(3):757–69. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chest.2023.03.047.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  34. Goniewicz ML, Smith DM, Edwards KC, Blount BC, Caldwell KL, Feng J, et al. Comparison of nicotine and toxicant exposure in users of electronic cigarettes and combustible cigarettes. JAMA Netw Open. 2018;1(8):e185937. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jamanetworkopen.2018.5937.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Loffredo L, Carnevale R, Battaglia S, Marti R, Pizzolo S, Bartimoccia S, et al. Impact of chronic use of heat-not-burn cigarettes on oxidative stress, endothelial dysfunction and platelet activation: the SUR-VAPES Chronic Study. Thorax. 2021;76(6):618–20. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/thoraxjnl-2020-215900.

    Article  PubMed  Google Scholar 

  36. Schirone L, Loffredo L, Carnevale R, Battaglia S, Marti R, Pizzolo S, et al. Sex-related differences in oxidative, platelet, and vascular function in chronic users of heat-not-burn vs traditional combustion cigarettes. Antioxidants (Basel). 2022. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/antiox11071237.

    Article  PubMed  Google Scholar 

  37. Majek P, Jankowski M, Brożek GM. Acute health effects of heated tobacco products: comparative analysis with traditional cigarettes and electronic cigarettes in young adults. ERJ Open Res. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.1183/23120541.00595-2022.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Andersen A, Reimer R, Dawes K, Becker A, Hutchens N, Miller S, et al. DNA methylation differentiates smoking from vaping and non-combustible tobacco use. Epigenetics. 2022;17(2):178–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/15592294.2021.1890875.

    Article  PubMed  Google Scholar 

  39. Benowitz NL, Flanagan CA, Thomas TK, Koller KR, Wolfe AW, Renner CC, et al. Urine 4-(methylnitrosamino)-1-(3) pyridyl-1-butanol and cotinine in Alaska native postpartum women and neonates comparing smokers and smokeless tobacco users. Int J Circumpolar Health. 2018;77(1):1528125. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/22423982.2018.1528125.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Byhamre ML, Eliasson M, Söderberg S, Wennberg P, Oskarsson V. Association between snus use and lipid status in Swedish men. Scand J Clin Lab Invest. 2023;83(4):241–50. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/00365513.2023.2209915.

    Article  PubMed  CAS  Google Scholar 

  41. Campbell LR, Brown BG, Jones BA, Marano KM, Borgerding MF. Study of cardiovascular disease biomarkers among tobacco consumers, part 1: biomarkers of exposure. Inhal Toxicol. 2015;27(3):149–56. https://doiorg.publicaciones.saludcastillayleon.es/10.3109/08958378.2015.1013228.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Chang JT, Vivar JC, Tam J, Hammad HT, Christensen CH, van Bemmel DM, et al. Biomarkers of potential harm among adult cigarette and smokeless tobacco users in the PATH Study Wave 1 (2013–2014): A Cross-sectional Analysis. Cancer Epidemiol Biomarkers Prev. 2021;30(7):1320–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.Epi-20-1544.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  43. Chen J, Kettermann A, Rostron BL, Day HR. Biomarkers of exposure among US cigar smokers: an analysis of 1999–2012 National Health and Nutrition Examination Survey (NHANES) data. Cancer Epidemiol Biomarkers Prev. 2014;23(12):2906–15. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.Epi-14-0849.

    Article  PubMed  CAS  Google Scholar 

  44. Dai HD, Benowitz NL, Rogan E, Degarege A, Buckley J, Khan AS. Biomarkers of toxic exposure and oxidative stress among US adult users of premium cigar versus other cigar subtypes: 2013–2019. Nicotine Tob Res. 2023;25(Suppl1):S84–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntad054.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Eliasson M, Asplund K, Evrin P-E, Lundblad D. Relationship of cigarette smoking and snuff dipping to plasma fibrinogen, fibrinolytic variables and serum insulin. The Northern Sweden MONICA study. Atherosclerosis. 1995;113(1):41–53. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0021-9150(94)05425-i.

    Article  PubMed  CAS  Google Scholar 

  46. Eliasson M, Asplund K, Nasic S, Rodu B. Influence of smoking and snus on the prevalence and incidence of type 2 diabetes amongst men: the northern Sweden MONICA study. J Intern Med. 2004;256:101–10. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1365-2796.2004.01344.x.

    Article  PubMed  CAS  Google Scholar 

  47. England LJ, Levine RJ, Mills JL, Klebanoff MA, Yu KF, Cnattingius S. Adverse pregnancy outcomes in snuff users. Am J Obstet Gynecol. 2003;189(4):939–43. https://doiorg.publicaciones.saludcastillayleon.es/10.1067/s0002-9378(03)00661-6.

    Article  PubMed  Google Scholar 

  48. Giraud DW, Martin D, Driskell JA. Plasma and dietary vitamin C and E levels of tobacco chewers, smokers, and nonusers. J Am Diet Assoc. 1995;95:798–800. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/S0002-8223(95)00220-0Available.

    Article  PubMed  CAS  Google Scholar 

  49. Hecht SS, Carmella SG, Murphy SE. Hemoglobin adducts as biomarkers of exposure to and metabolic activation of carcinogenic tobacco-specific nitrosamines. Biomed Environ Sci. 1991;4(1–2):93–103.

    PubMed  CAS  Google Scholar 

  50. Jacob P 3rd, Yu L, Liang G, Shulgin AT, Benowitz NL. Gas chromatographic-mass spectrometric method for determination of anabasine, anatabine and other tobacco alkaloids in urine of smokers and smokeless tobacco users. J Chromatogr. 1993;619(1):49–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0378-4347(93)80445-a.

    Article  PubMed  CAS  Google Scholar 

  51. Jacob P 3rd, Yu L, Shulgin AT, Benowitz NL. Minor tobacco alkaloids as biomarkers for tobacco use: comparison of users of cigarettes, smokeless tobacco, cigars, and pipes. Am J Public Health. 1999;5:731–6. https://doiorg.publicaciones.saludcastillayleon.es/10.2105/ajph.89.5.731.

    Article  Google Scholar 

  52. Krall EA, Garvey AJ, Garcia RI. Alveolar bone loss and tooth loss in male cigar and pipe smokers. J Am Dent Assoc. 1999;130(1):57–64. https://doiorg.publicaciones.saludcastillayleon.es/10.14219/jada.archive.1999.0029.

    Article  PubMed  CAS  Google Scholar 

  53. Lange P, Groth S, Nyboe J, Mortensen J, Appleyard M, Jensen G, et al. Decline of the lung function related to the type of tobacco smoked and inhalation. Thorax. 1990;45:22–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/thx.45.1.22ErratumappearsinThorax1990Mar;45(3):240.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  54. Marano KM, Kathman SJ, Jones BA, Nordskog BK, Brown BG, Borgerding MF. Study of cardiovascular disease biomarkers among tobacco consumers. Part 3: evaluation and comparison with the US National Health and Nutrition Examination Survey. Inhal Toxicol. 2015;27(3):167–73. https://doiorg.publicaciones.saludcastillayleon.es/10.3109/08958378.2015.1009196.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  55. Mushtaq N, Sarwar Z, Kouplen K, Ahmed R, Beebe LA. Association of cardiovascular disease risk factors with exclusive smokeless tobacco use among US males: cross-sectional analysis of NHANES data 2003–2018. Am J Health Promot. 2023;37(5):614–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/08901171221141980.

    Article  PubMed  Google Scholar 

  56. Naufal ZS, Marano KM, Kathman SJ, Wilson CL. Differential exposure biomarker levels among cigarette smokers and smokeless tobacco consumers in the National Health and Nutrition Examination Survey 1999–2008. Biomarkers. 2011;16(3):222–35. https://doiorg.publicaciones.saludcastillayleon.es/10.3109/1354750x.2010.546013.

    Article  PubMed  CAS  Google Scholar 

  57. Rezk-Hanna M, Warda US, Stokes AC, Fetterman J, Li J, Macey PM, et al. Associations of smokeless tobacco use with cardiovascular disease risk: Insights from the Population Assessment of Tobacco and Health study. Nicotine Tob Res. 2022;15(7):1063–70. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntab258.

    Article  CAS  Google Scholar 

  58. Rodu B, Stegmayr B, Nasic S, Cole P, Asplund K. The influence of smoking and smokeless tobacco use on weight amongst men. J Intern Med. 2004;255(1):102–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1046/j.0954-6820.2003.01244.x.

    Article  PubMed  CAS  Google Scholar 

  59. Rostron BL, Chang CM, van Bemmel DM, Xia Y, Blount BC. Nicotine and toxicant exposure among US smokeless tobacco users: results from 1999 to 2012 National Health and Nutrition Examination Survey data. Cancer Epidemiol Biomarkers Prev. 2015;24:1829–37. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.EPI-15-0376.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  60. Sharma E, Tang Z, Lauten K, Silveira ML, Delnevo CD, Edwards KC, et al. Cardiovascular disease outcomes among established cigar users 40 years and older: findings from the population assessment of tobacco and health (PATH) study, waves 1–5 (2013–2019). Prev Med Rep. 2024;37:102569. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.pmedr.2023.102569.

    Article  PubMed  Google Scholar 

  61. Söderström E, Nilsson TK, Schneede J, Ueland PM, Midttun Ø, Gylling B, et al. Plasma cotinine is positively associated with homocysteine in smokers but not in users of smokeless tobacco. Int J Environ Res Public Health. 2021. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijerph182111365.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Wei B, Blount BC, Xia B, Wang L. Assessing exposure to tobacco-specific carcinogen NNK using its urinary metabolite NNAL measured in US population: 2011–2012. J Expo Sci Environ Epidemiol. 2016;26(3):249–56. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/jes.2014.88.

    Article  PubMed  CAS  Google Scholar 

  63. Wennmalm Å, Benthin G, Granström EF, Persson L, Petersson A-S, Winell S. Relation between tobacco use and urinary excretion of thromboxane A2 and prostacyclin metabolites in young men. Circulation. 1991;83(5):1698–704. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/01.cir.83.5.1698.

    Article  PubMed  CAS  Google Scholar 

  64. Wickholm S, Söder P-Ö, Galanti MR, Söder B, Klinge B. Periodontal disease in a group of Swedish adult snuff and cigarette users. Acta Odontol Scand. 2004;62:333–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/00016350410001801.

    Article  PubMed  Google Scholar 

  65. Wulf HC, Husum B, Niebuhr E. Sister chromatid exchanges in smokers of high-tar cigarettes, low-tar cigarettes, cheroots and pipe tobacco. Hereditas. 1983;98(2):225–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1601-5223.1983.tb00598.x.

    Article  PubMed  CAS  Google Scholar 

  66. Xia B, Blount BC, Guillot T, Brosius C, Li Y, Van Bemmel DM, et al. Tobacco-specific nitrosamines (NNAL, NNN, NAT, and NAB) Exposures in the US Population Assessment of Tobacco and Health (PATH) Study Wave 1 (2013–2014). Nicotine Tob Res. 2021;23(3):573–83. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntaa110.

    Article  PubMed  CAS  Google Scholar 

  67. Zhu AZ, Binnington MJ, Renner CC, Lanier AP, Hatsukami DK, Stepanov I, et al. Alaska Native smokers and smokeless tobacco users with slower CYP2A6 activity have lower tobacco consumption, lower tobacco-specific nitrosamine exposure and lower tobacco-specific nitrosamine bioactivation. Carcinogenesis. 2013;34(1):93–101. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/carcin/bgs306.

    Article  PubMed  CAS  Google Scholar 

  68. Caliri AW, Caceres A, Tommasi S, Besaratinia A. Hypomethylation of LINE-1 repeat elements and global loss of DNA hydroxymethylation in vapers and smokers. Epigenetics. 2020;15(8):816–29. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/15592294.2020.1724401.

    Article  PubMed  PubMed Central  Google Scholar 

  69. Carroll DM, Wagener TL, Peck JD, Brame LS, Thompson DM, Stephens LD, et al. Biomarkers of exposure in ENDS users, smokers, and dual users of American Indian descent. Tob Regul Sci. 2018;4(2):3–15. https://doiorg.publicaciones.saludcastillayleon.es/10.18001/trs.4.2.1.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Christensen CH, Chang JT, Rostron BL, Hammad HT, van Bemmel DM, Del Valle-Pinero AY, et al. Biomarkers of inflammation and oxidative stress among adult former smoker, current e-cigarette users-results from Wave 1 PATH Study. Cancer Epidemiol Biomarkers Prev. 2021;30(10):1947–55. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/1055-9965.Epi-21-0140.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  71. Cook S, Hirschtick JL, Barnes G, Arenberg D, Bondarenko I, Patel A, et al. Time-varying association between cigarette and ENDS use on incident hypertension among US adults: a prospective longitudinal study. BMJ Open. 2023;13(4):e062297. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bmjopen-2022-062297.

    Article  PubMed  PubMed Central  Google Scholar 

  72. Fetterman JL, Keith RJ, Palmisano JN, McGlasson KL, Weisbrod RM, Majid S, et al. Alterations in vascular function associated with the use of combustible and electronic cigarettes. J Am Heart Assoc. 2020;9(9):e014570. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/jaha.119.014570.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  73. Gupta R, Lin Y, Luna K, Logue A, Yoon AJ, Haptonstall KP, et al. Electronic and tobacco cigarettes alter polyunsaturated fatty acids and oxidative biomarkers. Circ Res. 2021;129(5):514–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/circresaha.120.317828.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  74. Hickman E, Payton A, Duffney P, Wells H, Ceppe AS, Brocke S, et al. Biomarkers of airway immune homeostasis differ significantly with generation of e-cigarettes. Am J Respir Crit Care Med. 2022;206(10):1248–58. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/rccm.202202-0373OC.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  75. Majid S, Keith RJ, Fetterman JL, Weisbrod RM, Nystoriak J, Wilson T, et al. Lipid profiles in users of combustible and electronic cigarettes. Vasc Med. 2021;26(5):483–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/1358863x211009313.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  76. Metzen D, M’Pembele R, Zako S, Mourikis P, Helten C, Zikeli D, et al. Platelet reactivity is higher in e-cigarette vaping as compared to traditional smoking. Int J Cardiol. 2021;343:146–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ijcard.2021.09.005.

    Article  PubMed  Google Scholar 

  77. Oliveri D, Liang Q, Sarkar M. Real-world evidence of differences in biomarkers of exposure to select harmful and potentially harmful constituents and biomarkers of potential harm between adult e-vapor users and adult cigarette smokers. Nicotine Tob Res. 2020;22(7):1114–22. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ntr/ntz185.

    Article  PubMed  CAS  Google Scholar 

  78. Perez MF, Mead EL, Atuegwu NC, Mortensen EM, Goniewicz M, Oncken C. Biomarkers of toxicant exposure and inflammation among women of reproductive age who use electronic or conventional cigarettes. J Womens Health (Larchmt). 2021;30(4):539–50. https://doiorg.publicaciones.saludcastillayleon.es/10.1089/jwh.2019.8075.

    Article  PubMed  Google Scholar 

  79. Podzolkov VI, Bragina AE, Druzhinina NA, Vasil’eva LV, Osadchiy KK, Dubchak AE, et al. Relation between tobacco smoking/electronic smoking and albuminuria/vascular stiffness in young people without cardiovascular diseases. Kidney Blood Press Res. 2020;45(3):467–76. https://doiorg.publicaciones.saludcastillayleon.es/10.1159/000507510.

    Article  PubMed  CAS  Google Scholar 

  80. Russo C, Cibella F, Mondati E, Caponnetto P, Frazzetto E, Caruso M, et al. Lack of substantial post-cessation weight increase in electronic cigarettes users. Int J Environ Res Public Health. 2018. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijerph15040581.

    Article  PubMed  PubMed Central  Google Scholar 

  81. Stokes AC, Xie W, Wilson AE, Yang H, Orimoloye OA, Harlow AF, et al. Association of cigarette and electronic cigarette use patterns with levels of inflammatory and oxidative stress biomarkers among US adults: Population Assessment of Tobacco and Health Study. Circulation. 2021;143(8):869–71. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/circulationaha.120.051551.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

We thank Japan Tobacco International S.A. for financial support. We also thank Yvonne Cooper for secretarial assistance and for obtaining the required references, Jan Hamling for assistance in checking the text, and John Hamling for creating the plots and the Shiny App.

Funding

Japan Tobacco International S.A. supported the time spent by the authors in the publication of this paper.

Author information

Authors and Affiliations

  1. P.N. Lee Statistics and Computing Ltd., 17 Cedar Road, Sutton, SM2 5DA, UK

    Peter N. Lee & Katharine J. Coombs

  2. RoeLee Statistics Ltd., 17 Cedar Road,, Sutton, SM2 5DA, UK

    John S. Fry

Authors
  1. Peter N. Lee
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Katharine J. Coombs
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. John S. Fry
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Lee PN planned the study; Literature searches and data entry were carried out by Coombs KJ and by Lee PN; Statistical analyses were carried out by Fry JS and checked by Lee PN; Lee PN drafted the text, which was checked by Coombs KJ and Hamling JS.

Corresponding author

Correspondence to Peter N. Lee.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Note that P.N. Lee Statistics and Computing Ltd. has now closed.

Supplementary Information

12954_2025_1188_MOESM1_ESM.docx

Additional file 1 Details of the papers excluded in the initial searches and the reasons for exclusion, both based on the first look at the papers and at the data entry stage.

Additional file 2 The data recorded for each study and for each biomarker from the initial searches.

Additional file 3 Presents plots for each biomarker based on the data obtained from the initial searches.

12954_2025_1188_MOESM4_ESM.docx

Additional file 4 Details of the papers excluded in the further searches and the reasons for exclusion, both based on the first look at the papers and at the data entry stage.

Additional file 5 The data recorded for each study and for each biomarker from the further searches.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lee, P.N., Coombs, K.J. & Fry, J.S. Estimating lung cancer risk from e-cigarettes and heated tobacco products: applications of a tool based on biomarkers of exposure and of potential harm. Harm Reduct J 22, 45 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12954-025-01188-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12954-025-01188-x

Keywords

Harm Reduction Journal

ISSN: 1477-7517