For decades the first stage in the discovery of drugs for cancer and many other diseases has been testing in human cell cultures, grown in a single layer. Cultures of human cancer cells have been fantastically useful as a simple model of the disease, but they do have limitations – they are flat and uniform, whereas tumours are anything but. That difference between model and reality limits the ability of cell cultures to predict how a tumour will respond to treatment, and is perhaps one reason why cancer drug discovery has such a high failure rate - with only about one in 10 drugs that enter preclinical development are finally approved for patients.
When you think about it, it makes sense – how can cells grown on plastic reflect what is going on in a tumour in a human body? Indeed, researchers know that the ‘wiring’ of the internal signalling pathways and cellular interactions in these highly reductionist models are very different from those of tumour and host cells within the body, where cells are in a dynamic, reciprocal relationship with their microenvironment. It’s not surprising that cells grown in a dish fail to predict in vivo responses given that they lack the heterogeneity and complexity of real tumours, the gradients of oxygen and nutrients that result in abnormal metabolism and can’t replicate the physical barriers to drug access that solid tumours present.
Professor Sue Eccles, leader of the Tumour Biology and Metastasis Team here at The Institute of Cancer Research, London, is attempting to address some of these limitations by taking more aspects of drug discovery into 3D. One focus of her team’s work has been to develop a suite of robust, reproducible and rapid assays using tumour spheroids – or ‘micro-cancers’ – which are a bridge between standard 2D monolayer cultures and animal tumour models reflecting several key aspects of tumour heterogeneity. These were initiated by a talented post doc, Mara Vinci, now in Chris Jones’ group.
"3D micro-cancer models could translate into a significant reduction in the number of preclinical drug tests required in animals and result in the most effective drugs being rapidly progressed to the clinic."
Professor Eccles and her team believe there is a need for a 3D system that can measure not only cell proliferation, but other ‘hallmarks’ of malignant cancers such as cell migration, invasion and tumour angiogenesis – where cancer cells can generate their own blood supply from surrounding tissue. It is these features of cancers that contribute to their life-threatening expansion and ability to spread around the body. They wanted the new system to be easy to use, robust, versatile, validated for target validation and cancer drug discovery, all in a high throughput format. “We grew 96 identical micro-cancers in a single plate the size of a postcard, where the cultures were able to develop close cell-cell interactions and gradients of cell division limited by access to oxygen and nutrients”, explains Professor Eccles. “We initially focused on brain cancers which, because of their invasive nature, are often fatal in both adults and children within one year, and novel therapies are urgently needed. We have also extended the assays to other common cancers such as breast, lung, colon and prostate. All the assays have been validated using anti-cancer agents that target the underlying molecular abnormalities that drive cancer cell proliferation, motility and invasion.”
The micro-cancers – whose responses are measured in a new imaging cytometer – give highly reproducible results and so far seem to be more predictive of responses in vivo than 2D cultures. The team is using these assays routinely in target validation and drug discovery projects and has so far tested around 40 human cancer cell lines with more than 50 compounds. “We have now expanded the technology to 384-well plates, allowing us to screen for new agents that could inhibit 3D growth and spread – extending the utility of the assays beyond compounds that are already in hand”, says Professor Eccles. “Future refinements will include increasing the complexity of the models by including host cells within the spheroids and adding appropriate tissue-matched extracellular matrix proteins.”
Professor Eccles hopes the 3D micro-cancer models will also translate into a significant reduction in the number of preclinical drug tests required in animals and result in the most effective drugs being rapidly progressed to the clinic. “The simplicity of the assays and the automated analysis mean that this technology will be of great use for many research groups in academia and industry,” says Professor Eccles. “We see an important future for this novel suite of assays both in more efficient and accurate cancer drug screening, and in identifying and validating new molecular targets for therapy, leading the way to personalised cancer medicine.”
Of course, there is still some way to go before these 3D models become used routinely, and there are still many questions to be answered. Do they really replicate the wiring within a real-life tumour? What are the indications and limitations for their use? Could these models eventually help to reduce the need for animal experiments?
But researchers like Professor Eccles are adamant that there is a need for in vitro models that more accurately predict the complex behaviour of malignant cancers – in all their manifestations – and the chances are that in future, many of these will operate in three dimensions rather than two.
If you would like to hear more about 3D models and the tumour microenvironment, there is a ‘Goodbye flat biology’ EACR conference in November 2014.