Are No Two Tumor Cells Alike?

Author:  Jordan Hall
Institution:  Duke University
Date:  August 2012

As Charles Darwin famously stated, “[it] is not the strongest of the species that survives, nor the most intelligent, but the one most responsive to change.” This philosophy not only applies to animals in an ecosystem – but also to tumors. Extensive research conducted by scientists from the Cancer Research UK’s London Research Institute suggests that tumors evolve as they grow. Further, no two samples from the same tumor are identical – even those from adjacent regions. The findings of the study published in The New England Journal of Medicine in March may represent a major breakthrough in our understanding of tumor biology.

For a cancer to form, only a single cell needs to divide uncontrollably. This unrestrained growth is the result of mutations, or changes in the DNA sequence, in the tightly-controlled process of cell division. Once cells become cancerous, they rapidly accumulate further mutations as they grow and divide.

It is commonly believed that only the rate of cell division differs between cancer cells and normal cells. The actual logistics of the process, however, are thought to remain identical, with a single parent cell producing two genetically-identical daughter cells. The large team of researchers led by Dr. Charles Swanton wanted to validate this widely held notion.

“I think that intratumor heterogeneity may be the ‘elephant in the room’ for cancer research, and [it] may help explain some of the perplexing behaviors of tumors - such as drug resistance to novel compounds and tumor recurrence,” explains Dr. Swanton.

In this study, multiple tissue samples were collected from a patient enrolled in a clinical drug trial for metastatic kidney cancer. These samples were taken from spatially distinct regions of the primary kidney tumor and additional ‘metastatic’ sites elsewhere in the body where the cancer had spread. The researchers then used ‘next-generation’ DNA sequencing techniques to rapidly characterize the full DNA sequence in each sample.

After determining the DNA sequence, the team then used a single nucleotide polymorphism (or ‘SNP’) array to identify the mutations present in the tumor samples. By comparing the amount and types of mutations in each primary or metastatic kidney cancer sample, the researchers could determine the degree of similarity among the samples. To ensure that they were correct, they then validated their results by repeating this process in three other patients enrolled in the same clinical trial.

A total of 128 distinct mutations were identified across all tissue samples. Only 32% of mutations, however, were shared in all samples. The remaining two-thirds were either unique to a specific sample or shared in a particular region (e.g., primary tumor site or metastatic site). Each sample had a unique profile. In fact, an average sample contained only 55% of all mutations detected in the entire tumor, depicting the wide heterogeneity.

Dr. Swanton and his team also compared the mutations in the tissue samples to four recurrent genes known to mutated in kidney cancer. Only one of the genes, VHL, was mutated ubiquitously in all samples. The other three genes – KDM5C, SETD2, and MTOR – were mutated in different ways in different regions of the tumor, suggesting convergent evolution. “Despite genetic divergence during tumor progression, phenotypic convergent evolution occurs, indicating a high degree of mutational diversity, a substrate for Darwinian selection, and evolutionary adaptation,” writes primary author Dr. Marco Gerlinger in the empirical article.

The implications of this research are important to our understanding of tumor biology.

First, tumor evolution and adaption may explain why tumors become drug resistant and recurrent. As cancers form from just a single uncontrolled cell, only one tumor cell needs to be immune to a particular drug in order for the tumor to survive. Evidenced by the extensive regional heterogeneity, the probability of this cell existing is high in a widely diverse pool of tumor cells, each with a unique genetic profile.

Second, intratumor heterogeneity could account for the difficulties in creating robust diagnostic tools and for why the same tumor exhibits a wide-range of prognoses for different patients. “Genomic analyses from a single tumor-biopsy specimen may underestimate the mutational burden on hetereogenous tumors,” says Dr. Swanton. If a single biopsy sample contains only half of the tumor’s mutations, any decisions regarding treatment and therapy may be based on an inaccurate portrayal of the tumor’s true genetic profile.

Finally, the findings of this research provide a novel avenue for cancer research and drug development. As Dr. Swanton explains, “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.” Future research needs to identify those initial mutations that are shared among all samples and are instrumental in driving the tumor’s progression. Developing therapeutic compounds that target these early mutations could prevent the tumors from becoming aggressive or spreading. In essence, cancer research must evolve and adapt to the many tricks of the tumor.