Caffeine in pregnancy and leukaemia risk

This research discovered issues in the technique used to take blood samples and test them for DNA damage. This led to the researchers developing a better method that may also be suitable for older blood samples that have been archived

  • Topic: Blood cancers
  • Institution: University of Leicester
  • Country: United Kingdom
  • Status: Completed
Researcher: Marcus Cooke

Grant title

Maternal caffeine intake and the presence of chromosomal abnormalities in neonatal blood: implications for future cancer risk

Scientific abstract

(View plain language abstract)


Silent chromosomal translocations frequently occur in newborn children. However, the origin of these translocations remains poorly understood. Caffeine, a natural inhibitor of topoisomerase II, induces DNA double strand breaks and may be involved in the production of translocations.

Moreover, caffeine enhances error-prone repair by inhibiting ATM kinase, essential for maintaining DNA ends in repair complexes. The dose of caffeine reaching the fetus depends predominantly upon maternal intake and toxicokinetics.

The objective of the project was to improve the limited understanding of the factors affecting in utero chromosome translocation. The project aimed to demonstrate proof of principle that RNA can be extracted from Guthrie blood spots, and screened for translocations.


From a well-defined mother/baby cohort of 1,340 pregnancies, we identified 1,300 individuals and recovered 1,287 Guthrie cards (Department of Clinical Chemistry, Sheffield Children’s Hospital). These were punched and shipped to the University of Maastricht where, from a defined sub-set, RNA was isolated and amplified in preparation for mutation screening. RNA was isolated from Guthrie cards with the RNeasy Micro Kit (Qiagen) according to the manufacturer’s instructions. Mutation screening was performed using the Hemavision Screen kit (DNA Technology, Rosskov, Denmark).

In addition to the planned work we have evaluated the effect of sampling procedure and storage conditions of whole blood, upon levels of oxidatively damaged DNA. We have also developed methodology which allows for the comet assay (including the enzyme-modified variant) analysis of small volumes of whole blood.


The amplification step resulted in a low RNA yield, which was insufficient for the mutation assay. As a positive control for the mutation assay, we performed the assay on two cord blood samples for which we knew the RNA quality was good. For one sample, the kit’s control RNA band was slightly visible but was absent for the other sample. This indicated to us that the kit was not working poorly.

From our additional work, we successfully demonstrated that small volumes of blood can be simply integrated into the comet assay (including the enzyme-modified variant) for rapid analysis of DNA damage. This significantly increased the speed and throughput of the assay by removing the need for isolation of peripheral blood mononuclear cells prior to analysis.

The assay revealed that, surprisingly, whole blood appears to act as an effective cryopreservative, when small volumes of samples are stored at -80°C. We also demonstrated that small volumes of blood, such as those obtained via a Lancet/pin-prick (~250 µL) show long-term stability at -80°C, and are amenable to our modification of the comet assay.


We continued to check the RNA quality from blood spots, concluding that, contrary to earlier reports in the literature, RNA quality is too low for further downstream applications. We also continued to isolate DNA from the Guthrie cards, as originally proposed, and will then perform limit mutation analyses as originally proposed in the application.

We also reported conditions that allows the storage of whole blood at -80°C, without cryopreservative, for at least one month without artefactual formation of DNA damage, as measured by the comet assay.

Plain language abstract


Leukaemia is the most common form of childhood cancers (35% of all childhood malignancies). Chromosomal (DNA) alterations are frequently found in newborn children and have been linked to increased risk of childhood leukaemia. However, the origin of these alterations remains poorly understood.

The diet contains compounds, such as caffeine, which promote formation (and persistence) of these chromosomal alterations. There is a wide range of dietary sources of caffeine, and it freely crosses the placenta, leading to the baby’s blood levels being the same as the mother’s. We propose that mother’s caffeine intake increases the incidence of certain common chromosomal translocations, which are seen in childhood leukaemia.

From an existing mother/baby group of 1,340 pregnancies, we proposed to screen DNA from the newborns’ heel-prick test for common chromosomal alterations, evaluate mothers’ caffeine intake and ability to metabolise caffeine, using information and specimens already collected. Known risk factors, such as folate status, birth weight and DNA damage will be taken into account, plus other dietary sources of DNA-damaging compounds.

This is a unique opportunity to investigate the source of chromosomal alterations during pregnancy, which, after further testing, may be a biomarker of newborns with increased susceptibility to future disease.

Methods and key findings

This study continues to work to establish whether RNA can be extracted from Guthrie cards, that have been stored for up to seven years, as reported in the literature, and analyse this for the 28 main DNA modifications reported in chronic and acute leukaemia. To date, it would appear that the kit which analyses these modifications is not performing as it should. Our fallback position is to extract DNA from the Guthrie blood spots, and screen these for the DNA changes, as originally proposed in the application.

A total of 1,340 mother-baby pairs were potentially available for inclusion in the study. From these, the EuroKing system identified 1,300. We have now assembled a total of 1,287 Guthrie cards in Maastricht, awaiting analysis for the DNA changes most frequently associated with leukaemia.

In addition to the planned work, and in anticipation of an extension, we have evaluated the effect of sampling procedure and storage conditions of whole blood, upon levels of DNA damage. We showed that rather than isolating white blood cells prior to analysis, storage of whole blood protects the cells from damage induced by storage at frozen temperatures. We have also modified an existing methodology to analyse DNA damage in white blood cells in whole blood, without the need for their isolation prior to analysis.

This is a significant improvement upon present methods, which are labour intensive, significantly increasing the ease and speed of the assay, and giving the potential for retrospective analysis of previously stored samples.

Grant publications

> Simplified method for the collection, storage, and comet assay analysis of DNA damage in whole blood