Updated

Taking a sample or biopsy from just one part of a tumor might not give a full picture of its genetic diversity and may explain why doctors, despite using genetically targeted drugs, are often unable to save patients whose cancer has spread, scientists said.

A study by British researchers found there are more genetic differences than similarities between biopsies taken from separate areas of the same tumor, and yet further gene differences in samples taken from secondary tumors.

That might help explain why, despite recent development of a wave of highly targeted drugs designed to tackle cancers of specific genetic types, the prognosis remains poor for many patients with so-called solid-tumor disease like breast, lung, or kidney cancer that has spread to others parts of the body.

But the researchers, whose study was partly funded by charity Cancer Research UK and published in the New England Journal of Medicine, said it also pointed to a way forward.

The team carried out the first ever genome-wide analysis of the genetic changes or faults in different regions of the same tumor.

They looked at four patients with cancer in their kidneys, taking samples from different regions of the primary tumor and also from other organs where the tumor had spread.

They found that the majority of gene faults, around two-thirds, were not the same in one sample as in another, even when the biopsies were taken from the same tumor.

Samples taken from secondary tumors - which are a result of the disease spreading to other parts of the body - had yet more different genetic faults, suggesting that basing treatment decisions on just one primary tumor sample is not sufficient.

"We've known for some time that tumors are a patchwork of faults, but this is the first time we've been able to use cutting-edge genome sequencing technology to map out the genetic landscape of a tumor in such exquisite detail," said Charles Swanton, of University College London's cancer institute, who led the study and presented its results at a briefing in London on Tuesday.

He said they had uncovered "an extraordinary amount of diversity" at a genetic level both within tumors and within a single patient, with more differences between biopsies from the same tumor than similarities.

"The next step will be to understand what's driving this diversity in different cancers and identify key driver mutations that are common throughout all parts of a tumor," Swanton said.

Personalized medicine 

Genetic profiling of patients and their tumors has become more common in cancer treatment in wealthy countries as drug companies develop new generations of so-called "personalized medicines" that target cancers with specific genetic features.

Roche's blockbuster breast cancer drug Herceptin is designed to treat only women who make too much of the HER2 protein, for example, while Novartis's Afinitor targets mTOR, a protein that acts as an important regulator of tumor cell division, blood vessel growth and cell metabolism.

James Larkin, an oncologist at London's Royal Marsden Hospital who also worked on the study, said the findings suggest the reality of personalized cancer treatment is far more complex than previously thought.

"The molecular changes that drive the growth of the cancer once it has spread may be different from those that drive the growth of the primary tumor," he said.

The researchers compared genetic faults in various tumor samples taken from the four patients.

They found 118 different mutations - 40 of which were "ubiquitous mutations" found in all biopsies, 53 "shared mutations" that were found in most but not all biopsies, and 25 "private mutations" only found in a single sample.

By analyzing where the shared mutations were in relation to the whole tumor, the researchers were able to trace the origins of certain subtypes of cancer cells back to what they called key "driver mutations." This allowed them to create a map of how the pattern of faults might have evolved over time.

Swanton likened the findings to a tree, in which the trunk is the primary tumor and the branches the secondary tumors from the cancer's spread.

While he stressed the results would need to be replicated with larger numbers of patients and in different types of cancer, he said these early indications showed "the importance of targeting common mutations found in the trunk of the tree as opposed to those found in the branches."

"It may also explain why surgery to remove the primary kidney tumor can improve survival," he added, since cutting out a tumor reduces the risk that cells resistant to drug treatment could go on to re-grow the tumor or spread elsewhere.