Main Menu

Why we need smarter clinical trials for cancer treatments

Posted on 21 November, 2013 by henryfrenchicr
Cancer is a very diverse disease. If we think of a ‘type’ of cancer – lung cancer, breast cancer, colon cancer – we are actually thinking of many different diseases that seem similar partly because they grow in a particular place. They have some things in common, but can be very different from each other, and vary from patient to patient.

One consequence of cancer’s diversity is that we need smarter ways of designing clinical trials for new treatments. It can be very difficult to divide patients into groups who are likely to respond to a treatment in the same way – and all too easy to put two patients whose cancers are very different into the same group, despite the fact a new treatment could work in one but not the other.


A review published recently in the journal Future Oncology - a joint effort between the ICR and the Seattle Cancer Centre (SCC) in the US - focuses on recent failures in the search for new treatments for sarcoma, a rare cancer that gives rise to tumours with a very distinctive appearance, often including recognisable tissue structures such as muscle, vessels, nerves and bone. The review was initiated by the SCC's Robin Jones - formerly of the ICR - and co-authored by ICR-Royal Marsden researchers Anastasia Constantinidou and Aisha Miah.

One key message in the review is that clinical trials of some new potential sarcoma treatments might have failed because they were actually tested against several very different diseases, rather than one. Each disease falls under the definition of sarcoma, but actually each sarcoma type might be so different that each potential treatment failed to show benefit against them all.

The authors discuss several potential drugs that showed promise in early sarcoma trials – including palifosfamide, which looked especially promising – but have ultimately failed to emerge as a new option for patients.

But would some of those drugs have made it into the clinic to treat sub-types of sarcoma, if only they had been tested more selectively?

Subdividing patients into smaller groups poses new challenges, as the authors say. The more patients are split into subgroups based on the specific features of their disease, the fewer patients there are to enrol into clinical trials.

This is a particular challenge in rare cancers like sarcoma, because you need relatively large numbers of patients to prove a treatment works. It has been a enormous challenge for researchers to subdivide even common cancers – breast cancer, for example – into a few groups based on their likely response to drugs, and then to develop treatments for different types.

The review discusses the importance of worldwide collaborations in future trials of potential treatments for sarcoma. They list four types of sarcoma – called osteosarcoma, Ewing’s sarcoma, embryonal rhabdomyosarcoma and gastrointestinal stromal tumours – which have been success stories in cancer treatment because of the use of combination treatments (first three) or treatment selection based on biology features (gastrointestinal stromal tumours). Maybe trialling treatments in each of these types could be a place to start?

The authors also state the vital importance of understanding the exact mechanism of potential drugs before testing them in large, expensive trials. To revisit an analogy used in my recent blog on breast cancer research – the more you know about the ‘bullet’ you’re using, as well as the target you are shooting at, the more chance you have of hitting that target.

Researchers at the ICR and Royal Marsden are already pioneering new ways of understanding drugs more effectively before large clinical trials – for example, by developing new imaging technology for evaluating drugs at the pre-clinical stage, and putting greater emphasis on confirming the mechanism of action of a potential drug in phase I clinical trials.

In the future, these approaches will hugely benefit the cancer research community by giving them new information that makes clinical trials more likely to succeed. But in the future, for some cancers, those running later stage trials will need to move away from traditional views of how to design these trials in order to make the most of all the time, effort and money that goes into creating new potential drugs.


comments powered by Disqus