Using Next-Generation Sequencing to Target Tumor Genomics in Metastatic Prostate Cancer

When metastatic prostate cancer becomes resistant to androgen deprivation therapy, patients develop what is known as castration-resistant prostate cancer (mCRPC). In this late-stage disease scenario, physicians turn to the sequential use of different therapies with the goals of improving quality of life, reducing complications, and increasing progression-free and overall survival.

They may also experiment with next-generation sequencing (NGS) in an effort to identify new therapeutic targets for such patients. Using commercially available DNA sequencing panels, they can analyze tumor tissue or cell-free circulating tumor DNA. This allows them to target putative key tumor drivers and pathways in genetically defined subsets of patients.

Whether this strategy translates into the clinical benefit of mCRPC patients is unclear. Recently oncologists at Cleveland Clinic undertook a retrospective study to determine whether NGS-based testing results alter clinical treatment decisions and ultimately if such changes or interventions can prolong survival in mCRPC patients.

“We are in an era in oncology that we define as precision oncology, and it’s gaining a lot of momentum,” says Jorge A. Garcia, MD, an internationally renowned genitourinary medical oncologist and the Kerscher Family Chair for Clinical Prostate Cancer Research at Cleveland Clinic. He was senior author on the study, which appeared in Targeted Oncology. “We are using genomic abnormalities or aberrations in the tumor profile and seeing if we can use that information to identify novel therapeutic targets.”

Sequencing influenced treatment

Dr. Garcia and his colleagues looked at the baseline characteristics of 66 metastatic prostate patients. The most common alterations included the genes TP53 (42%), PTEN (35%), androgen receptor (AR) (30%), DNA repair (30%), the mTOR/PIK3CA signaling pathway (21%), cyclin-dependent kinases (15%), BRAF (9%) and MMR/MSI genes (6%). There were 45 tumors with at least one targetable genomic alteration (tGA+).

The researchers used the tumor DNA data to try on- and off-protocol approaches to treatment, such as giving metastatic prostate cancer patients with the HER2 mutation the treatments used for HER2+ breast cancer patients. Another approach was to enroll patients with specific mutations in clinical trials focused on treating patients with these and other known specific mutations.

Overall the sequencing results influenced treatment in 29% of the patients: PARP inhibitors (n = 7), mTOR inhibitors (n = 4), immune checkpoint inhibitors (n = 2) and an anti-HER2 antibody (n = 1). Only two patients received a targeted therapy in a clinical trial. Seven patients received PARP inhibitors off-label, one of them (14%) after prior platinum-based regimen. NGS was repeated in six after a median of 15.8 months.

Some of the reasons behind patients’ not receiving NGS-guided treatment included physician or patient’s choice (44%), the patient’s electing observation (19%), and the patient’s having ongoing response to treatment at the time of tNGS testing (16%).

Results and implications

The median PFS for tGA+ patients receiving specific therapy was 4.1 months compared with 4.3 months for patients not treated with genomic alteration-guided therapy. Such results are not surprising, says Dr. Garcia.

“I think the biggest challenge with genomics in prostate cancer is the lack of actionability of the data we obtain,” he says.

Researchers are still struggling to understand the best timing for obtaining a patient’s NGS information.

“We remain somewhat limited to what tumor tissue one can use, when that tissue was collected and how it may differ from the present state of your disease,” says Dr. Garcia. “If you have primary archived material from the ’90s when your tumor resection was done, its genomic profile may be very different from when you developed metastatic disease 15 years later.”

The next step is to perform a prospective study, he says. “We need to look at patient subsets within prostate cancer, ones that have been defined at risk of harboring a specific genomic change that we may be able to treat earlier. We also need to be persistent and continue to collect this data to really define the best genomic platform to use, best tissue material and best timing for collections. Acquiring this data will help us elucidate the true actionability of genomic testing in the context of men with prostate cancer.”