Biomarkers, PARP Inhibitors Vital for Personalized Treatment in Breast Cancer Subtypes

In Partnership With:

Partner | Cancer Centers | <b>The University of Texas MD Anderson Cancer Center</b>

Banu K. Arun, MD, discusses the challenge of treating patients with TNBC and BRCA1-related breast cancers, the value of genotyping, and what she hopes to see in future trial results.

Banu K. Arun, MD

Developing predictive biomarkers will be key to treating patients with triple-negative breast cancer (TNBC), especially when choosing a targeted therapy, said Banu K. Arun, MD.

In a presentation during the 2018 OncLive® State of the Science Summit™ on Breast Cancer, Arun said there is evidence that PARP inhibitors as well as immunotherapy in combination with various agents may be effective in women with TNBC and BRCA1-related breast cancers, but the science isn’t there yet.

“Biomarker assessment, overall, needs to be standardized, but we are headed in the right direction,” said Arun, professor of breast medical oncology at The University of Texas MD Anderson Cancer Center.

OncLive: What were some key points of your lecture?

In an interview with OncLive during the meeting, Arun discussed the challenge of treating these patients, the value of genotyping, and what she hopes to see in future trial results.Arun: As you know, women with germline BRCA mutations mostly have high-grade cancers, especially BRCA1-related breast cancers. They have more triple-negative disease and more aggressive cancers. Recently, new approaches have been evaluated in these patients, including PARP inhibitors. I reviewed 2 very recently reported phase III studies looking at singleagent PARP inhibitors versus physician’s choice of therapy.

I also discussed the results from a small but important pilot study where a PARP inhibitor was administered as a single agent in the neoadjuvant setting for 6 months, and I briefly touched on the use of platinums in BRCA-associated breast cancers.

The second part of my presentation looked at TNBC. As we all know, TNBC is a very heterogeneous disease. It has multiple subtypes; therefore, it can most probably be targeted with different agents that target these pathways. We think that, likely, 1 agent will not be enough to treat that disease unless we know a very specific pathway that is involved in that subtype. But right now, we’re not there yet.

Has genotyping become more important since LOTUS was reported?

We don’t have any predictive biomarkers of response and we are just now beginning to do targeted therapy with immunotherapy and immunomodulation targeting the PI3K/AKT/mTOR pathway. We targeting the androgen receptor in TNBC. I reviewed the principles of these approaches but none of them are standard of care yet, unfortunately. We are still treating TNBC with standard therapies.The LOTUS trial looked at an AKT inhibitor plus chemotherapy versus chemotherapy alone. The combination was significantly better than single-agent chemotherapy.

Can you discuss immunogenicity in TNBC?

Genotyping is important, but these are the initial trials. We need to have validation. Biomarker assessment, overall, needs to be standardized, but we are headed in the right direction. Genotyping, not only to assess whether we would use this combination but for many agents targeting various subtypes in TNBC, will be important. However, we need more studies and more positive results.Immunotherapy is very interesting and very complicated, and it has become very exciting. When we started doing studies with PD-1/PD-L1 inhibitors, initial trials such as the KEYNOTE-012 trial showed some very promising results with an 18% response rate. However, the subsequent KEYNOTE-086 study in TNBC, unfortunately, showed only a 4.7% response rate.

What trial results are you most looking forward to?

Immunotherapy is not the answer as a single agent. We need to find the right subgroup, and perhaps the right combinations. Multiple studies are ongoing with combinations, either with chemotherapy or other targeted agents, including PARP inhibitors.The trials that use biomarkers to help us identify which subgroup will benefit most from a given targeted intervention will be very important. Right now, we are still putting patients with a very heterogeneous group of cancers in the same bucket and trying to understand the results. When they don’t show promising results, we don’t know which subgroup might benefit more. Biomarker-driven, personalized, targeted therapy studies that show promising results—that’s what I’m looking for.