Cancer is not one disease – it’s hundreds of different diseases that manifest in different ways in different parts of the body. But there is one common theme across all of these unique diseases: genes gone wrong. At its core, cancer is the result of abnormalities in one or more genes that lead to the uncontrolled growth of cells. If we’re going to find new ways to stop cancer, we need to better understand its root genetic causes.

Until recently, sequencing and analyzing cancer genomes were very expensive and time-consuming endeavors. But advances in next-generation sequencing technologies have made it possible to do this work more effectively and efficiently than ever before. Using these tools, a team of Genentech scientists launched the Cancer Genome Project with the goal of hunting down new genetic targets around which to build new therapeutics. The team's work has already led to important discoveries across multiple types of cancer, and has also laid the foundation for future discoveries.

Faulty Fusion and Aberrant Amplification

One of the first discoveries of the Cancer Genome Project was in a gene family called R-spondin (RSPO). The team found that in a subset of people with colon cancer, two genes in the RSPO family fuse with other genes, creating new fusion proteins. These abnormal proteins are enticing potential targets for medicines because they are only found in cancer cells and are important to driving tumor growth. The fusion increases the expression levels of RSPO leading to activation of cancer signaling pathways. As a result of these discoveries, Genentech scientists are now working on developing investigational medicines that target RSPOs.

Mutations and gene fusions aren’t the only ways that genes can go wrong in cancer. The team identified extra copies (amplifications) of a gene called SOX2 in about 27 percent of small cell lung cancer cases. This discovery paved the way for further research into the way SOX2 amplification contributes to tumor growth, which may ultimately lead to the development of new targeted medicines for small cell lung cancer, a particularly aggressive form of the disease with poor prognosis and very limited treatment options.

Casting a Wider Net

To capture insights across a broad range of cancers, Genentech’s Cancer Genome Project team screened hundreds of human tumor samples from dozens of types of cancer, including gastric (stomach), non-small cell lung, kidney, skin, and ovarian cancers.

One key discovery was mutations in a gene called ERBB3 in about 11 percent of gastric and colon cancers. ERBB3 is a member of the HER family of receptors known to be essential for normal cellular growth and development. Previous studies focused on a related protein, ERBB2, leading to the development of small-molecule and antibody-based therapeutics for multiple cancer types. After identifying the ERBB3 mutations in gastric and colon cancers, the team at Genentech asked if some of the existing anti-ERBB antibodies and small molecule inhibitors could be repurposed for this new target. Their research showed that indeed, these drugs could block the ability of mutant ERBB3 to cause cancerous growth.

This discovery of cancer-activating ERBB3 mutations increases the importance of ERBB3 in cancer, and supports further research into therapeutics that specifically target ERBB3.

A Community Resource

In addition to finding new targets and diagnostic tests, another goal of Genentech’s Cancer Genome Project is to provide valuable resources for the broader cancer research community. Recently they applied their state-of-the-art methods to one of the most important tools in cancer research: tumor-derived cell lines. Although cell lines are widely used, a deeper knowledge of how they differ at the genomic level is needed to understand what exactly is driving their growth (and in turn, which therapeutic approaches may have the best chance for success).

To that end, the team at Genentech performed a comprehensive analysis of 675 human cancer cell lines for features including gene expression, mutations, gene fusions and expression of non-human sequences. One of the most powerful examples for the utility of this type of study is the ability to make predictive models. If a researcher uses a cancer cell line that has mutations in a target pathway that a therapy is modulating, the cell line itself may interfere with the therapy’s overall success and provide inaccurate data. The new data could be used in a predictive model for selecting the appropriate cancer cell line for a given therapy.

The Cancer Genome Project isn’t the end of the scientific journey but rather the start of many more. It’s helping uncover the hidden genetic drivers of cancer, revealing staggering diversity as well as surprising recurring themes. These insights are being explored at Genentech and throughout the research community, and may guide researchers safely towards new frontiers for cancer treatments.

Read more about these important findings in the following papers from Genentech scientists:

• Nature. "Recurrent R-spondin fusions in colon cancer"
• Nature Genetics. "Comprehensive genomic analysis identifies SOX2 as a frequently amplified gene in small-cell lung cancer"
• Cancer Cell. "Oncogenic ERBB3 mutations in human cancers"
• Nature Biotechnology. "A comprehensive transcriptional portrait of human cancer cell lines"