A study can show an effect of aluminium on cells without demonstrating that a deodorant causes disease in its user. To read it, identify the model, the substance, the exposure, the outcome measured and the limitations. Sappino and Mandriota’s work raises important experimental questions; it is not enough to establish a link between antiperspirants and breast cancer in humans.
First question: what was actually exposed?
A scientific title may mention “human” cells and give the impression that the study followed people. These are two different situations. Cells of human origin can be grown in a laboratory, in a medium to which a substance is added directly. This makes it possible to study a mechanism without reproducing an organism’s entire exposure.
Look for the “methods” section, even if you start with the abstract. It tells you whether researchers observed users, grew cells or worked with animals. The title and conclusion alone do not always describe that experimental sequence.
What does Sappino’s 2012 paper show?
The study published in 2012 concerns human mammary cells exposed to aluminium chloride in culture. The researchers observed effects including DNA breaks and changes in growth behaviour. André-Pascal Sappino, an oncologist, is among the authors, alongside Stefano Mandriota, a biology researcher.
These observations provide a biological signal to investigate. They are neither a follow-up of women using an antiperspirant nor a measure of the number of cancers among its users. The authors also state that their results do not formally identify aluminium as a breast carcinogen. Keeping this limitation does not diminish the value of their research: it defines its scope.
What does Mandriota’s 2016 study add?
In the 2016 paper, mouse mammary cells were first exposed to aluminium chloride in culture. They were then transplanted into mice, where the researchers assessed tumour and metastasis formation. The results support the potential for transformation in this model.
The crucial detail is the transition from cultured cells to their transplantation. This was not a study in which antiperspirant was simply applied to mice. Summarising it as “deodorant gives mice cancer” therefore changes the experiment that was carried out.
| What you read | What to check |
|---|---|
| “Human cells” | Cells in culture or people followed? |
| “Test in mice” | Product applied or cells transplanted? |
| “Significant effect” | Which specific outcome was measured? |
| “Aluminium” | Which substance, route and exposure duration? |
Why does exposure change the interpretation?
The concentration in a product, the concentration added to a culture medium and the dose that reaches a tissue are not interchangeable. The route of exposure, duration and experimental conditions matter. Comparing concentrations can help formulate a hypothesis; on its own, it does not recreate use on the skin.
Also ask what the exposed group was compared with and which outcome was observed. DNA damage, altered growth and a cancer diagnosis are not three names for the same measurement. The outcome should retain its specific name when it moves from a scientific paper to a press article.
How can you place a paper in the wider body of evidence?
A study answers a limited question. To understand what can be concluded about a use, you also need to look at assessments examining several types of data and their exposure conditions.
In 2024, the European SCCS concluded that aluminium from cosmetics alone was safe under the conditions studied, and distinguished that scope from combined exposure involving other sources. The National Cancer Institute does not report an established link between antiperspirant or deodorant use and breast cancer.
A careful reading keeps these findings alongside the experimental questions. It avoids choosing a paper simply because it confirms what we already wanted to believe. To move on to a practical decision, our deodorant and antiperspirant guide distinguishes needs relating to odour, moisture and skin tolerance.
Frequently asked questions
Does a study on human cells prove a risk in humans?
No. It can shed light on a mechanism, but a cell culture does not reproduce a person’s entire exposure or how their body works.
Can studies on cells and mice be ignored?
No. They help us understand mechanisms and guide research. Their model and limitations must be retained when discussing them.
Does a doctor’s name make a conclusion indisputable?
No. The author’s qualifications are useful, but they do not replace the methods, results and their consistency with the other available data.
Sources and further reading
- Sappino et al., 2012 — Aluminium chloride promotes anchorage-independent growth in human mammary epithelial cells
- Mandriota et al., 2016 — Aluminium chloride promotes tumorigenesis and metastasis in normal murine mammary gland epithelial cells
- SCCS, 2024 — Safety of aluminium in cosmetic products, Submission IV
- National Cancer Institute — Antiperspirants/Deodorants and Breast Cancer