The results come from a personalised medicine trial which set out to treat patients, not on the basis of their cancer site or type, but by the types of mutations found in their tumours.
'This study is the first deliverable of precision medicine. We have proven that histology-independent, biomarker-selected basket studies are feasible and can serve as a tool for developing molecularly targeted cancer therapy,' said Dr José Baselga, the study's senior author from Memorial Sloan Kettering Cancer Center in New York. 'While we can - and should - be cautiously optimistic, this is what the future of precision medicine looks like.'
The researchers treated 122 patients with the metastatic melanoma drug vemurafenib. All the patients tumours tested positive for the BRAF V600 mutation the drug targets - one that is common in skin cancer but rarer in other cancers.
In the 20 patients with non-small-cell lung cancer, 42 percent responded to the drug and the median progression-free survival time was 7.3 months, the researchers report in the New England Journal of Medicine.
The strategy also showed efficacy in two rare conditions involving white blood cell overproliferation, Erdheim-Chester disease and Langherhans cell histiocytosis, where the response rate was 43 percent. And, despite a median treatment duration of 5.9 months, no patients progressed during therapy.
However, results for other cancer types were more modest with smaller or no effects seen in patients with colorectal cancer, brain tumours, multiple myeloma, anaplastic thyroid cancer or bile duct cancer.
Dr Otis Brawley, chief medical officer of the American Cancer Society, who was not involved in the study, told Reuters the research provided 'proof-of-concept': 'We are going to see more and more things like this - someone with colon cancer who didn't do well with the normal treatment may get their tumour analysed and (one) of these small molecule drugs the FDA has approved for some other cancer might be tried.'
Studies that take such an approach, like this one, are sometimes called 'basket trials' and are becoming more common.
However, in an editorial accompanying the study, Professors David Hunter and Ralph D'Agostino argue that they should be seen only as a first step. Basket trials are hindered by factors such as small sample sizes, no comparator groups and the difficulty of identifying which mutations in a particular tumour are actually responsible for driving the disease, they say.
They argue that establishing open-access databases of off-label drug use may yield more useful and significant data.
'Developing such observational databases is far from trivial, but the costs per patient would be small in relation to the monthly costs of many of the targeted therapies,' they write. 'Perhaps the plural of anecdote could be data after all.'