The cover of the book conveys the gist: the cover bears a black and red stamp, clearly alluding to popular beliefs about the power of printed paper (“stamping is a ritual that transforms paper into truth”, “science doubts, bureaucracy stamps”).

Valentin Thomas is a social sciences researcher at the French National Center for Scientific Research (CNRS). In 2016, on behalf of the University of Paris-Dauphine, he participated as an observer in the final meetings of two working groups of the International Agency for Research on Cancer (IARC) in Lyon, which were evaluating the evidence of carcinogenicity of coffee, mate, and very hot beverages, as well as a number of industrially used substances (volumes 115 and 116 of IARC Monographs). More recently, he also was an observer at the meeting for volume 135, which evaluated certain per- and polyfluoroalkyl substances.

The book consists of an introduction and a conclusion, with six chapters in-between, entitled: “La classification comme rite d’institution”, “Mettre les agents à l’agenda”, “Des définitions concurrentes de la cancérogénicité”, “La fabrique des lobbys”, “Transformer la définition de la preuve”, and “Découper le monde toxique”. The chapters are grouped in pairs into sections of the work entitled “Saturation”, “Compétition” and “Standardisation”.

Valentin Thomas has pieced his human experience of those weeks with documents filed in the early decades of the IARC’s existence and with a detailed personal reflection on the “Monographs” programme. He (rightly) claims the ability of the social sciences to focus on the usefulness and limitations of the programme, largely entrusted to experts. He is not interested in the scientific criteria used for the recognition of carcinogens, but rather “la forme, les countours, les preésupposés, l’historicité, les prises en charge concrètes“ (p. 9) of the question that ‘obsesses’ scientific and political authorities: whether this or that agent can be labelled as carcinogenic. “Plutôt que d’émettre un jugément sur la qualité du discours scientifique [...], le regard sociologique porte son atténtion sur la manière dont cette rationalité est pensée et construite” (p. 25). 

The monograph programme was conceived following a 1971 resolution by the Governing Council of the IARC to prepare “monographs on the evaluation of the carcinogenic risk to humans of chemical substances”. At that time, the concept of ‘risk’ did not distinguish between the two stages of recognising an agent’s potential to cause harm and estimating the probability of individual harm. Later, ‘risk’ became ‘hazard’ and the focus was expanded to non-chemical carcinogenic hazards.  

In fact, the monographs have always concluded but with a weighted, evidence-based judgement on the strength of the scientific evidence that the agent in question has carcinogenic potential (on an ordinal scale, from “definitely carcinogenic” to “unclassifiable”) not with a quantitative risk assessment,

Valentin Thomas defines the IARC’s choice to focus on hazards while omitting a quantification of risk as an “institutional ritual”: “En acceptant de faire de l’évaluation du danger, le programme des monographies accepte d’être assigné à un rôle dans lequel d’autres institutions ont intérêt à le maintenir et dont il devient difficile de s’extraire” (p. 121).

These are strong words. Valentin Thomas does not say who these overbearing “other institutions” are, nor under what circumstances IARC would have accepted a subordinate position in the control of carcinogens. Above all, Thomas does not say what advantages a systematic quantification of risk would bring in terms of cancer prevention, which is the primary objective of the monographs programme, nor in how he would like to see such a gap systematically filled.

I do not share Valentin Thomas’ opinion that a highly prestigious WHO body such as IARC should routinely undertake the exercise of quantifying cancer risks. The mass of data needed to reliably reconstruct the dose-response relationship between exposure and risk measurement is not always available, especially for “low” exposures. Polluters who respect “scientific data” or say so, who want to have a clear conscience and feel protected in any court disputes, would like to be told what the maximum “acceptable” exposures to a carcinogen are. However, for carcinogens, no threshold exists, if only because of the possible interactive effects between exposures to different carcinogens. Furthermore, recognising such a threshold would pose major operational problems. Often, different studies on the same agent have led to conflicting results between studies on the same agent. This is partly due to the different co-occurrence of confounders and effect modifiers among studies, partly due to the range of accuracy in the estimation (and sometimes also in the measurement) of exposures, especially if they are past exposures. An irrefutable principle of epidemiological research is that the less precise the estimation of exposures, the more the risk is underestimated. In the case of the relationship between the intensity of asbestos exposure and cancer risk, it has been documented that more accurate studies recognise an excess of tumours at exposure levels considered safe in less accurate studies on asbestos and lung cancer.1 Finally, where carcinogens are recognised as such only in experimental systems, the extrapolation of quantitative data from laboratory animals or cell cultures to humans is problematic.

As for the prevention of occupational cancers, it is impossible to imagine a strategy other than one pointing to a “zero exposure” and replacement with a safer substitute (in this regard, I would like to recall once again Luigi Mara’s words: “Primary prevention of workers’ health begins with plant engineering and with replacement of harmful with safer chemicals”). The history of asbestos has taught us that “controlled” use, already recommended in industrial settings as “safe”, is not only unsafe, but also requires monitoring tools that are difficult to implement, especially in less technologically advanced countries where most workers currently exposed to asbestos live. 

I believe that, if IARC shifted its focus to quantitative risk estimates for any agent considered carcinogenic for humans, disputes among experts would increase dramatically. This would create an ideal situation for confusing the courts and, above all, for distracting public health from its primary objective of eliminating or minimising the exposure of workers and the general population to harmful exogenous agents.

That said, over the past fifty years, there have been countless instances in international institutions and courts in many countries where the IARC’s exclusively qualitative assessments have helped to create fair regulations for the protection of populations and to ensure justice for victims of undue environmental exposure.2

Valentin Thomas also criticises the programme for spending “un temps considérable à réévaluer d’anciens dossiers, alors que le nombre de substances mises sur le marché ne cesse de croître” (p. 193). It is true that, due to resource issues, the IARC programme regularly lags behind the need for assessments (it would lag even further if it had to produce quantitative risk estimates), but it is also true, as the preamble to the monographs points out, that for many substances recently placed on the market, there are no studies (including experimental studies) for a carcinogenicity assessment.

Valentin Thomas overlooks the fact that updates are necessary – and useful – when new scientific studies become available with results that may lead the IARC to change its classification, providing scientific support for interventions of varying severity in cancer prevention. As David Michaels3 said, “scientists do not possess the truth, they aim for the truth”.

As one reads on, the feeling that Valentin Thomas is guided by a degree of prejudice becomes stronger – at least, it did in my case. I will mention two excerpts that strengthened my impression. 

The first concerns Lorenzo Tomatis and industrial power (p. 124). Valentin Thomas notes Tomatis’s conversion to a “rapport conflictuel avec les firmes privées et avec les autorités publiques dont il pouvait fustiger l’inaction sanitaire”. However, he considers it too late and insufficient to free Tomatis from the “règle [...] de conserver les avantages économiques générés par la production et la commercialisation des substances chimiques. L’intervention sur ce marché doit constituer l’exception”.

What proof? In 1974 poor Tomatis, and Ruggero Montesano, when the identification of substances harmful to humans based on carcinogenesis experiments was in its infancy, had mentioned4 the possibility of ‘false positive’ results in experimental models and the consequent inappropriate elimination of ‘useful’ substances in the modern world. Perhaps Lorenzo’s highly critical approach was still in its infancy, but he did not fail to point out that the experimental systems available at the time – albeit limited – had confirmed the documented carcinogenicity of certain substances for humans in laboratory animals. 

The other example concerns the predictability of the results of IARC evaluations (p. 75). Valentin Thomas refers to a conversation during which members of the IARC secretariat whom he had asked hinted – privately and confidentially – that they were able to anticipate the outcome of such evaluations. Valentin was astonished: he uses strong words such as “sensitive information”, “unacceptable truths”, “rigged dice”, and “open secret”. The exact terms of the dialogues are not given. In the form in which they are reported, they seem like unpleasant episodes. Observers at meetings should not approach the secretariat, let alone ask inappropriate questions. And members of the secretariat should not engage in indiscretions. That said, Thomas’s comment is specious and disingenuous. It is natural that those whose job is to prepare IARC monographs and who are familiar with scientific work can predict the working group’s response with a high degree of accuracy and a high probability of being correct. On the eve of the meeting of the working group that evaluated tobacco smoke, would Thomas have liked the IARC secretariat not to expect this agent to end up in the “definitely carcinogenic to humans” category?

In the concluding chapter, I find the overall assessment of progress in recognising the exogenous causes of cancer (pp. 193-194) overestimated. «Derrière quelques causes bien identifiées, prédispositions genétiques ou facteurs de risque comme la consommation d’alcool, du tabac, ou l’exposition à une poignée de toxiques amplement documentés, comme l’amiante, le benzène ou certains pesticides, l’émergence de cet espace (scientifique et réglementaire transnationale – ndr) a surtout accompagné la construction d’une machine technocratique de production d’ignorance, de controverses sans fin et d’inaction politique institutionalisée”. As recently as 70-80 years ago, the prevailing opinion was that cancer was caused by “bad luck”. Up to volume 139 of the monographs, the “handful” of toxic substances includes 135 agents that are indisputably carcinogenic to humans and another 97 that are probably carcinogenic. These account for a large proportion of the cancers that occur worldwide. I would not call the effort to produce and verify the reliability of this knowledge “the production of ignorance”. Nor would I speak of “controversy” without mentioning the underlying industrial interests (it is no coincidence that, over the decades, most expressions of dissent on IARC evaluations coincided with the interests of producers) and also the reluctance of health policy authorities in many countries to implement environmental remediation measures (for example, as recently as 2020, Indonesia imported 100,000 tonnes of asbestos, used mainly in construction, in the absence of rules for the use of this hazardous material.5 

Valentin Thomas is right to mention political inaction, linked to government decisions on the use of public funds. Dwindling resources are being allocated to research, including research on environmental carcinogens, while industry-funded studies are on the rise, especially reviews of the literature that are antagonistic to those of the IARC.

The closing words of the book emphasise the importance of “encourager toute velléité d’action collective à ne pas s’en remettre tout entière à l’unique soutien des experts patentés, mais de le considérer pour ce qu’il est: une ressource parmi d’autres d’un répertoire d’actions nécessairement plus large”. On this point, one cannot but agree. The inevitable handing over of the interpretation of scientific evidence to “experts” requires that the latter be subject to some form of accountability. Transparency is one of the key ingredients of accountability, as stated in the monographs preamble6 and it is conveyed through the disclosure of the mechanisms used in selecting both the agents to be evaluated and the identification of the members of the working groups, as well as through a detailed description of the multidisciplinary inferential reasoning, in a manner understandable by non-experts. This approach has been recognized by dozens of scientists from all over the world who have participated in the programme over the past half a century.7,8

Conflicts of interest: none declared.

References

  1. Burdorf A, Heederik D. Applying quality criteria to exposure in asbestos epidemiology increases the estimated risk. Ann Occup Hyg 2011;55(6):565-68. doi: 10.1093/annhyg/mer042
  2. Schubauer-Berigan MK. Invited Perspective: Good Measure-Assessing the Impact of Cancer Hazard Identification on Policies for Cancer Prevention. Environ Health Perspect 2023;131(12):121302. doi: 10.1289/EHP14099
  3. Michaels D. Doubt Is Their Product: How Industry’s Assault on Science Threatens Your Health. New York: Oxford University Press; 2008.
  4. Montesano R, Tomatis L. Introduction in Chemical carcinogenesis essays. IARC Scientific Publications 10. Lyon: IARC; 1974; pp. i-x.
  5. Kazan-Allen L. Asbestos Profile: Indonesia, International ban asbestos secretariat. August 2020. Available from: https://www.ibasecretariat.org/prof_indonesia.php
  6. Preamble to the IARC Monographs on the Identification of Carcinogenic Hazards to Humans.

    Lyon:IARC; 2019.

  7. Pearce N, Blair A, Vineis P, et al. IARC monographs: 40 years of evaluating carcinogenic hazards to humans. Environ Health Perspect 2015;123(6):507-14. doi: 10.1289/ehp.1409149
  8. Samet JM, Berrington de Gonzalez A, Lunn RM, Schubauer-Berigan MK. Commentary: Role and communications of cancer hazard determinations. Carcinogenesis 2022;43(2):79-81. doi: 10.1093/carcin/bgac001

 

       Visite