Reducing Radiation Risk from Hospital Scans

Even though it is something carried out in hospitals hundreds of times a day, X-rays and CT (computed topography) scans are procedures that expose the patient to radiation. Yes, the radiation dosage for the majority of scans carried out is very little when compared to every day exposure; for example an X-ray of the arm is 0.001 mSv (millisievert), a dental X-ray is 0.005 mSv, a chest X-ray is 0.020 mSv – in comparison the average background radiation received over one day is 0.010 mSV, while someone flying across the continental USA would receive 0.040 mSV. However, other scans are higher, a mammogram is 0.400 mSv (equivalent of 40 days worth of exposure in one dose), while a head CT scan gives a dose of 2 mSv (equivalent to ~7 months) and a chest CT scan 7 mSv (equivalent to ~20 months) (see the below image from Randall Munroe for more examples).

Although the cells in our body are able to repair and restore DNA damage resulting from radiation, the greater the dose received in one go, and the greater received in the long-term, the more likely it is that damage won’t be repaired correctly. Thus, we should aim to reduce patients’ exposure to radiation where possible. A recent paper by Kitchen and colleagues may have an answer by using phase-contrast x-ray imaging.[1] Because soft tissue has similar X-ray absorption properties to bone, which results in poor image contrast the radiation dosage has to be increased in standard scans. This new technique combines CT scans with phase retrieval and an algorithm to define edges, densities, etc. and results in a reduction in dosage by a factor of 300 fold (with the potential for a reduction factor in the tens of thousands), while still retaining equivalent image quality. Although the study only tested this in an animal model it is an important first step.

— Peter Chilton, Research Fellow

Reference:

1. Kitchen MJ, Buckley GA, Gureyev TE, et al. CT dose reduction factors in the thousands using X-ray phase contrast. Sci Rep. 2017; 7: 15953.

How Many Mutations for Cancer?

During our lifetime our somatic cells (non-reproductive cells) constantly accumulate mutations – in some cases these mutations lead to uncontrolled proliferation and allow the cells to invade other tissues and spread to other organs – i.e. become cancerous. Most of the mutations in cancerous cells are unimportant – it is only a few that are ‘drivers’ of cancer and dictate the way the cell behaves. However, we do not know how many mutations are actually required to cause cancer, or whether this number varies across cancer types.

Researchers working for the Wellcome Trust looked at over 7,500 tumours of 29 cancer types using methods adapted from molecular evolution to see which mutations were more common in cancerous than in non-cancerous cells.[1] They found that, on average, cancerous cells have around four coding substitutions (where a DNA nucleobase is exchanged for another, such as switching from adenine to guanine) that are ‘driver mutations’. This ranged from around one mutation per tumour in thyroid and testicular cancer, four in breast and liver cancer, to more than ten in endometrial and colorectal cancer. Of these ‘driver mutations’ around half occur in cancer genes that have yet to be discovered.

In the long-term these findings could help advance the development of precision cancer treatment, allowing drugs to be specifically targeted at the appropriate mutation(s).

— Peter Chilton, Research Fellow

Reference:

1. Matrincorena I, Raine KM, Gerstung M, Dawson KJ, Haase K, Van Loo P, Davies H, Stratton MR, Campbell PJ. Universal Patterns of Selection in Cancer and Somatic Tissues. Cell. 2017.