3D Tumour Printing


Having for decades thought of ‘printing’ as an operation that produces an image in two dimensions (I know: physicists please don’t write in, you know what I mean) I’ve had a lot of difficulty grasping the idea of three-dimensional printing. It’s presumably an example of ‘old fogey set in ways’ because in fact the idea’s not that difficult.

How’s it done?

A computer-controlled printer deposits successive layers of material to build the 3D shape specified by the computer. A number of methods have been developed but so precise has the technology become that, using computer-generated models, 3D geometries of almost unlimited complexity can be produced.

The methods, referred to as additive manufacturing, have developed to the extent that they are now used in heavy engineering and architectural design, for model railways and prosthetic implants. In the medical field you can now obtain patient-specific implants, based on anatomical data from the patient acquired by making 3D pictures of the target (e.g., by CT scans), for pretty well any item from jaws to ankles. The product is a bespoke implant, tailored for the individual patient, and they must be a fantastic boon for surgeons.

Now for the really amazing bit …

That’s a stunning example of different areas of science converging to produce completely novel strategies but what got me thinking about 3D printing was a recent paper that had the extraordinary juxtaposition of bioprinting, human tumours and chips in its title! The stars who’d done this remarkable work were Dong-Woo Cho, Sun Ha Paek and colleagues in South Korea. The problem they’d tackled was to come up with a model system for the most common brain cancer, glioblastoma, for which the five-year survival rate is less than 5%. Model systems in which cells from a patient’s tumour are implanted into mice have been developed for brain tumours but so far these have been poor predictors of treatment response.

It’s been recognized for some time that the physiological locale of tumours (the microenvironment) is critical to their behaviour (see Trouble With the Neighbours, Mosaic Masterpieces) and the Korean groups tried to reproduce that more closely by using 3D printing. They added to the cancer cells an extract of animal brain tissue together with endothelial cells (that line blood vessels) and gas-permeable silicone. Deposited on glass, that creates a more realistic glioblastoma microenvironment that includes being able to mimic the variations in oxygen levels that occur in tumours.

The result was that not only did the tumours grow but within a couple of weeks it was possible to test their responses to various drug combinations.

3D tumour printing on a chip to test drug responses in vitro. The first step is to remove cells from a patient’s brain tumour (a glioblastoma — GBM) and to mix them with a ‘bioink’. Several other ‘inks’ are added to mimic the natural environment of the tumour. The mixture is printed onto a glass slide and grown for two weeks before testing combinations of candidate drugs to inform the treatment plan for the patient. From Yi et al., 2019.

A major step

This tumour-on-a-chip method promises to be a significant advance in customizing treatments for glioblastoma. What’s more, its use will not be limited to brain tumours. However, as always with scientific progress, it’s not the final deal. For one thing this system cannot reproduce an immune response and we know that is a critical modulator of tumour progression.

Even so, it represents a quite astonishing marriage of scientific approaches to the problem of cancer treatment.


Yi, H.-G. et al. (2019). A bioprinted human-glioblastoma-on-a-chip for the identification of patient-specific responses to chemoradiotherapy. Nature Biomedical Engineering, 18 March.

Gomez-Roman, N. and Chalmers, A.J.   Patient-specific 3D-printed glioblastomas. Nature Biomedical Engineering (2019).