Ovarian cancer analysis identifies potential treatment biomarkers

MD Anderson researchers have identified potential biomarkers for the response of high-grade serous ovarian cancers (HGSC) to surgical resection alone or to chemotherapy before surgery, providing new avenues to explore for improving ovarian cancer treatment.

Their ovarian cancer research findings, in collaboration with Inova Health System, were published today in Cell Reports.

HGSC accounts for nearly 90% of ovarian cancer deaths. Patients who undergo complete surgical resection (R0) have better outcomes, but it is unknown whether this survival benefit is due to the aggressive surgery or to the biology of resectable high-grade serous ovarian cancers.

“Our group was involved in the Cancer Genome Atlas (TCGA) efforts to discover molecular differences in ovarian cancers. However, these studies used only samples from patients undergoing surgical resection and did not account for the most aggressive tumors that need to be treated with chemotherapy first before surgery,” says Anil Sood, M.D., professor and vice chair for Translational Research in the departments of Gynecologic Oncology and Cancer Biology and senior author of the paper.

TCGA also did not conduct immune or cellular analyses and did not include samples from metastatic sites.

Search for biomarkers to guide ovarian cancer treatment

To understand the biology of these tumors in depth, Sood, first author Sanghoon Lee, Ph.D., assistant professor of Systems Biology, and colleagues, conducted extensive analyses, including high-pass whole-genome sequencing, targeted deep sequencing, RNA sequencing, reverse-phase protein array, mass spectrometry-based proteomics/phosphoproteomics, immune-profiling, and integrated data analyses. The team conducted this work through MD Anderson’s Ovarian Cancer Moon Shot.

“Most studies just perform RNA analysis or one or two types of analyses, but we wanted to carry out an all-out effort to find biomarkers that can be used for clinical decision-making,” Sood says.

This study used baseline primary and metastatic tumor samples from 30 patients triaged by a systematic algorithm to either surgical resection or neoadjuvant chemotherapy and surgery: 10 patients had no visible residual disease after primary surgery (R0), 10 underwent neoadjuvant therapy but had a poor response, and 10 underwent neoadjuvant therapy with an excellent response. The researchers found significant molecular and cellular differences between the R0 and neoadjuvant therapy groups.

Genetic differences may help determine high-grade serous ovarian cancer treatment

They found that tumor samples from the R0 group had significantly lower expression of the  NF1 gene, RNA, and protein than samples from the neoadjuvant therapy group. Previous studies have shown that NF1 helps in tumorigenesis and chemotherapy resistance in HGSC. Although the biological mechanisms are unknown, these findings suggest that NF1 may serve as a biomarker that could predict response to surgical resection versus neoadjuvant therapy for patients with high-grade serous ovarian cancer.

Compared to the neoadjuvant therapy samples, R0 samples also had fewer chromothripsis-like patterns, in which a chromosome shatters and is not reassembled correctly. R0 samples had fewer copy number switches, and the chromothripsis-like patterns were localized on chromosomes 6 and 19, whereas the chromothripsis-like patterns in the neoadjuvant tumors were enriched on chromosomes 8 and 17 (NF1 is located on chromosome 17).

The researchers did not find any significant genomic differences between primary tumors and metastatic tumors. This suggests that the genomic instability of HGSC, including copy number and structure variation, might occur at an early point in disease progression.

Immune response differences

R0 tumors and tumors with excellent response to neoadjuvant therapy had significantly higher levels of neoantigens than the tumors with poor response to neoadjuvant therapy, and the R0 tumors had significantly more strong-binding neoantigens. These high levels of neoantigens also were associated with a better immune profile, such as more infiltration of tumor immune cells and fewer macrophages. These tumor-associated neoantigens might be good immunotherapeutic targets that could spare healthy cells.

Chemotherapy response markers

Compared to tumors that had poor response to neoadjuvant therapy, those with excellent response had significantly lower expression levels of phosphosites of LCK and YES1, Src-family kinases. These phosphosites have been implicated in T cell development and migration signaling, and these differences could serve as biomarkers to predict chemotherapy response in HGSC.

Next steps for high-grade serous ovarian cancer research

Because this in-depth analysis of high-grade serous ovarian cancer used baseline tumor samples before treatment, the research team plans to study tumor samples from patients after treatment. They also will look at promising biomarkers identified in this study to see if these genetic differences could be used for clinical decision-making or as therapeutic targets.

“Patients with high-grade serous ovarian cancer that does not respond to neoadjuvant therapy have the worst outcomes; we’d like to identify molecular/cellular targets and design drugs that can improve these patients’ outcomes,” Sood says.  “Accurate biomarkers could help us tailor precise treatment strategies for all patients with HGSC.”