Posted: March 9, 2017

Copy Number Analysis Reveals Link to Immunotherapy Outcomes

Amy E. Blum, M.A.

Image: Human chromosomes visualized by spectral karyotyping. Credit: Thomas Ried, NCI Center for Cancer Research

Immune checkpoint blockade immunotherapies, which prevent tumor cells from evading immune attack, are becoming an increasingly widespread and promising treatment option for many cancer types. However, these therapies are only effective for a subset of patients. By understanding the genetic attributes that mediate patients’ responses to immunotherapy, research may be able to help distinguish between patients who are likely to benefit and those who are unlikely to respond, thereby improving outcomes and minimizing side effects.

A recent study by researchers at Harvard Medical School mined data on 12 cancer types in The Cancer Genome Atlas (TCGA), seeking to understand the role of copy number alterations, (amplifications and deletions of parts of chromosomes), in cancer. Their findings, published in Science, revealed that cancers that harbor many copy number alterations in their genomes tend to have less immune involvement and worse response to immune checkpoint blockade immunotherapies. Combining this insight with other correlates of immunotherapy response may help scientists, and in the future, clinicians, predict patient outcomes and recommend tailored therapies.

Copy Number Alterations in TCGA

Previous studies have established that the number of mutations in cancer cells is associated with patients’ responses to immunotherapy. The more mutations present, the more opportunities there are for tumors to produce neoantigens, small peptides that immune cells recognize as abnormal. However, before this study, little was known about the impact that copy number alterations may have on immune involvement in cancer tissue.

The researchers began by searching for groups of genes with significantly different expression levels in tumors with many copy number changes versus those with few. They found that tumors with many copy number alterations also tend to have suppressed activity of the immune system.

Specifically, they found that in 10 out of 12 tumor types in TCGA, three groups of genes were expressed at lower levels in tumors harboring many copy number alterations. These sets of genes included genes that encode components of the adaptive immune system, such as genes for components of T-cells and B-cells, genes whose expression enables anti-tumor immune function, such as granzymes, and genes that indicate a strong and sustained immune response, such as proinflammatory molecules. The reduced expression of these genes in the tumor environment indicates that cells of the immune system are less involved in these high copy number tumors. The brain tumors glioblastoma multiforme and lower grade glioma were exceptions to this trend.

The analysis also identified a distinction between focal copy number alterations (small alterations of only a fraction of the chromosome) and broad copy number alterations (arm- to entire chromosome- level changes). While broad copy number alterations were associated with immune suppression, focal copy number alterations were more strongly associated with a cell growth signature. This highlights that focal and broad copy number alterations may represent distinct attributes of cancer cells with independent mechanisms.

A New Factor in Precision Immunotherapy

The investigators then applied their findings to an analysis of two clinical trials of anti-CTLA-4 blockade immunotherapy in melanoma. They hypothesized that tumors with many number of copy number changes would be less likely to respond to this type of immunotherapy: fewer immune cells appear to be acting on these cancers; therefore, empowering tumor-fighting cells to recognize the cancer might not be effective. Their analysis supported this hypothesis, revealing that copy number alterations were associated with worse outcomes in the immunotherapy trials.

Further, the researchers demonstrated that incorporating these findings into a predictive model of patient response to immunotherapy added significant value. In their model, the level of copy number changes in tumors was predictive of response independent of the number of predicted neoantigens, and when combined together, the two measures yielded increased predictive power. 

As immunotherapy becomes an integral part of cancer care, research that can help predict the likelihood of response for individual patients is paramount. These results, and further investigation into copy number alterations in cancer, provide key information that may ultimately help assign immunotherapy to those who are most likely to benefit.


Davoli, T., Uno, H., Wooten, E.C., and Elledge, S.J. (2017) Tumor Aneuploidy Correlates with Markers of Immune Evasion and with Reduced Response to Immunotherapy. Science. 355: 6322, doi: 10.1126/science.aaf8399