Natural selection is thought to govern tumour growth, which means that the cells present in a full-grown tumour are the most evolutionarily fit, or likely to survive.

But the Big Bang model asserts that when a mutation occurs is more important than its evolutionary fitness in determining its prevalence in a full-grown tumour.

Christina Curtis, PhD, assistant professor of medicine and of genetics at the Stanford University School of Medicine, is the senior author of a paper describing the new model.

The lead author is Andrea Sottoriva, PhD, former postdoctoral scholar at the University of Southern California.

Curtis conducted most of the research for the paper as an assistant professor of preventive medicine at USC, in collaboration with Darryl Shibata, MD, professor of pathology there.

The paper was published online in Nature Genetics.

The Big Bang model could be used to identify growth-pattern differences at points very early in the life of a tumour, at the resolution of just a few thousand cells.

The researchers found that differences in early growth patterns were associated with whether the cells became a benign or malignant tumour: If growth was ordered, the cells would likely grow into an adenoma, or a benign tumour.

But if the growth trajectory was disordered, the cells were likely to grow into a malignant, or cancerous, tumour.

The researchers determined that early disordered growth patterns led to patterns of genetic variegation in mature, malignant tumours but not in benign adenomas.

Curtis added that these findings suggest that tumours are “born to be bad,” and malignant potential is determined early.

Cell mixing in the nascent tumour can give rise to detectable patterns of genetic variegation that could potentially serve as a biomarker, enabling early detection of cancerous growths, Curtis said.

A panoramic view of tumour differences

Tumours are made up of a diverse spectrum of mutated cells.

Cells within a single tumour vary, and tumours themselves differ from one another.

Differences among and within tumours present challenges for diagnosis and treatment of cancer.

It is thought that specific differences among tumours might determine which ones become malignant and which of these then go on to metastasise, or invade other parts of the body.

“Differences between tumour cells are pervasive,” Curtis said.

“Yet the origins of these differences are poorly understood.”

Curtis and her colleagues wanted to understand how and when genetic differences in tumours arise.

This wasn’t easy to do: Direct observation of tumour growth is impossible, and tumours are not routinely detected until composed of at least 1 million cells.

The researchers circumvented these obstacles by exploiting the fact that cells in full-grown tumours contain an “ancestry map” in their DNA.

They created tumour ancestry maps by analysing genomic data from mature tumours taken from the human colon during surgery and input the genomic data into a computational model.

Like the Big Bang model of the formation of the universe, their model deduced conditions in the past based on current data.

Instead of cosmic radiation, the Big Bang model of human colon cancer growth uses genomic data from a mature tumour to infer how that tumour grew, starting from when it was just a small number of cells.

The researchers studied both benign tumours and malignant tumours of the human colon to test the Big Bang model.

Currently, clinicians cannot accurately predict which benign growths will ultimately go on to become malignant growths in the colon.

Using an array of genomic techniques, Curtis and her team profiled colorectal tumours at multiple spatial scales, ranging from single cells to tumour glands consisting of fewer than 10,000 cells, as well as “bulk” tumour fragments that were taken from opposite sides of a full-grown tumour.

These methods provided a panoramic and high-resolution view of how cells within a tumour were different and how tumours from the same patient differed from one another.

From the genomic data, the researchers reconstructed a tumour growth history.

A tumour growth history can be thought of like a slide-show at a graduation party, which starts off with baby pictures and ends with images of the young adult.

The Big Bang model describes how a tumour evolves from a few thousand cells to a full-grown tumour composed of billions of cells.

The importance of time in tumour evolution

The researchers discovered that the timing of a mutation was more important than the relative evolutionarily fitness advantage it conferred in determining whether cells harbouring that mutation would be prevalent in a full-grown tumour.

Mutations continuously occur throughout the tumour lifespan, but only those that occur early have enough time to expand to a detectable size through cell division.

“What you see in the final cancer was there from the beginning,” Curtis said.

The Big Bang model provides a new way to interpret colon tumour growth dynamics and predicts that the earliest mutations will be pervasive throughout the final tumour, the study says.

Source: Stanford Medicine