3 days popular7 days popular1 month popular3 months popular

Promising prognostic biomarker candidates for ovarian cancer uncovered by Roswell Park team

Cancer researchers at () have identified two independent classes of novel candidate prognostic markers for , advancing efforts to develop targeted therapies for the disease. The findings resulted from two separate studies published in the peer-reviewed journal PLoS ONE and based on data from The Cancer Genome Atlas (TCGA), the world’s largest public database on gene expression in different tumor types.

The first study establishes an association of this often-deadly cancer with the immune system and clarified the role of a class of immunogenic tumor antigens known as cancer testis (CT) antigens, and the second reports new evidence that certain molecular interactions influence ovarian cancer prognosis. In context with recent evidence that the immune system can potently inhibit the growth of cancer cells, these novel findings may enable development of a new strategy for identifying those patients most likely to benefit from particular targeted therapies.

“There is a lot of interest right now in what to do with the human genome,” says Kevin Eng, PhD, an Assistant Professor of Oncology in the Department of Biostatistics and Bioinformatics at RPCI who was first author on both studies. “We are focused on finding the gene or combination of genes that are going to predict how long a woman’s ovarian cancer is going to remain in remission or what treatment is best for her cancer.”

An earlier project based on the TCGA ovarian cancer study outlined a method for identifying a subset of advanced serous ovarian cancers that interact with the immune system, but that approach involved almost 200 genes, making it difficult to use in a clinical setting. In a computational study published in PLoS ONE, Dr. Eng and colleague Takemasa Tsuji, PhD, from the Center for Immunotherapy at RPCI, set out to refine the associations between prognosis and the expression of immune genes, finding that just five genes reproducibly predicted survival in three independent data sets, including the TCGA.

“Obviously, it is easier to measure five genes instead of two hundred,” Dr. Eng says. “Studies like TCGA were meant to be broad screens to uncover gene signatures. Now we have an opportunity to exploit the key findings from those public resources, honing in on those molecular interactions most likely to be clinically relevant.”

The team found preliminary evidence that some CT antigens are associated with survival when examined in concert with the five-gene signature.

“CT antigens like NY-ESO-1 are readily recognized by the immune system and are therefore prime targets for cancer immunotherapies, so this work unveiling interaction between cancer and immune cells has real implications for our development of efficient immunotherapies at the Center for Immunotherapy,” explained Dr. Tsuji, an Assistant Professor of Oncology at Roswell Park. “Determining patterns of both the immunological markers and CT antigens present in a women’s tumor can lead the way to personalized medicine, an approach that uses the best combination of therapeutic modality and target antigen.”

The next steps, Dr. Eng notes, will be to build a standard protocol for testing and calibrating the five-gene signature and to confirm through clinical research that it continues to make accurate predictions.

In a second study, published online in PLoS ONE in September, Dr. Eng and Christina Ruggeri, a graduate student and researcher with the Center for Personalized Medicine at RPCI, surveyed the same TCGA data from a different perspective.

“One important concept in cancer is how cancers signal and how a cancer cell knows what to do,” Dr. Eng explained. “With our new computational approach, we are starting to deduce what signaling mechanisms matter and which ones are related to prognosis and survival in a clinical population.”

All cells interact with their environment through a variety of receptors and their matching proteins. The researchers developed a way to computationally score a receptor and its matching protein and found that these scores could be correlated with favorable prognoses. Importantly, these matching pairs were organized into signaling clusters, which suggests that different mechanisms apply to different patients.

“In breast cancer, treatment decisions are made based on a patient’s status for three key hormone receptors. We don’t have anything like that yet in ovarian cancer, so having a set of candidate receptors is a suggestive way of thinking about them,” Dr. Eng points out. “We found that different sets of receptors influenced different ovarian cancers. Targeted therapies target receptors, and when we cross-reference receptors with drugs that we have in the arsenal, it seems reasonable to conclude that these kinds of scores might predict who needs what drug.”

Going forward, Dr. Eng and his colleagues plan to isolate the tumor and immune components from ovarian cancer tissue samples. They will then study both components using genomic technology to see how they differ and how they interact. “Hopefully, the results of that study will help to clarify and reinforce the scores and signatures that we’ve already developed,” Dr. Eng notes.

The study by Drs. Eng and Tsuji, “Differential Antigen Expression Profile Predicts Immunoreactive Subset of Advanced Ovarian Cancers,” used shared resources supported by RPCI’s Cancer Center Support Grant (CCSG) from the National Cancer Institute (NCI), grant P30CA016056, and was also supported by a grant from the Roswell Park Alliance Foundation.

The paper by Dr. Eng and Ruggeri is “Connecting Prognostic Ligand Receptor Signaling Loops in Advanced Ovarian Cancer.” That project also relied on shared resources from RPCI’s Cancer Center Support Grant from the NCI, P30CA016056, and was supported by additional grants from the Roswell Park Alliance Foundation and NCI (the RPCI-UPCI Ovarian SPORE award, grant P50CA159981).


Differential Antigen Expression Profile Predicts Immunoreactive Subset of Advanced Ovarian Cancers, Kevin H. Eng, Takemasa Tsuji, PLoS ONE, DOI: 10.1371/journal.pone.0111586, published 7 November 2014.

Connecting Prognostic Ligand Receptor Signaling Loops in Advanced Ovarian Cancer, Kevin H. Eng, Christina Ruggeri, PLoS ONE, DOI: 10.1371/journal.pone.0107193, published 22 September 2014.

Source: Roswell Park Cancer Institute (RPCI)