Credits to
Carlo Zocchetti

Some reflections on carcinogenicity assessments by the IARC (International Agency for Research on Cancer)

Issue 5/2019

AbstractThe work examines the programme of the IARC (International Agency for Research on Cancer) monographs on the assessment of the carcinogenic risk of chemical substances to humans, from its origin (1971) to the present day, in order to identify the criteria that have led the Agency to conclude that a given substance (job, processing, type of industry) causes cancer in humans.

After recalling that for IARC the proposed evaluations, whose methodological path is always explicitly described in a document (preamble) at the beginning of each volume of the monographs, are intended to assist national and international authorities in formulating decisions regarding prevention measures, the work indicates that, strictly speaking, the Agency has never formally defined a concept of cause, limiting itself to specifying, over time, that an agent is defined as carcinogenic if it is able to increase the incidence of malignant neoplasms, reduce their latency, or increase their severity or multiplicity.

The evaluation criteria of the IARC have changed and expanded over the years, and now provides three criteria for the causal interpretation of an association: the presence of case, bias, and confusionmust be excluded. It also suggests considering five elements whose recurrence may increase the confidence in inferring a causal association: strength of the association, concordance of several studies, existence of a dose-response relationship, reduction of the incidence following the reduction of exposure, specificity of the association. The entire evaluation process is discussed, and the specific elements of the criteria are commented on.

The Agency’s classification scheme provides for five categories: group 1 (the agent is carcinogenic to humans), group 2A (the agent is likely to be carcinogenic to humans), group 2B (the agent is likely to be carcinogenic to humans), group 3 (the agent is not classifiable as regards its carcinogenicity to humans), group 4 (the agent is likely to be non-carcinogenic to humans).

The work concludes by discussing a practical example of an assessment, proposed rather recently and concerning carcinogenicity for the lung of the rubber industry, highlighting in this case, on the one hand, the difficulties encountered in applying the criteria set out and, on the other hand, the distance between the statements of principle (the criteria) and their practical implementation.

The clarification of the IARC criteria and its difficulties in application are intended to highlight the differences with the criteria used usually in the field of law.

This article was submitted anonymously for evaluation by two expert reviewers, with a positive outcome.


SUMMARY 1. Introduction. – 2. The first 16 volumes of the monographs: 1972-1978. – 3. From 1978 to 1982. – 4. From 1982 to the present day. – 5. So? – 5.1. Objectives of the programme. –5.2. Concept of cause. – 5.3. Criteria for causality. – 6. An example of assessment. –6.1. The case. – 6.2. Bias. – 6.3. Confounding. – 6.4. Strength of the association. – 6.5. Concordance of several studies. – 6.6. Existence of a dose-response relationship. – 6.7. Reduction in incidence as a result of exposure reduction. – 6.8. Specificity of the association. – 7. Conclusions.


To read the Article, click on “open file”.


A meeting of knowledge on individual and society
to bring out the unexpected and the unspoken in criminal law


ISSN 2612-677X (website)
ISSN 2704-6516 (journal)


The Journal does not impose any article processing charges (APC) or submission charges