SAN ANTONIO – In a sense, triple-negative breast cancer (TNBC) is quintuple negative.

The TNBC subtype is defined by what it doesn't have: There is no overexpression of the estrogen receptor (ER), the progesterone receptor (PR) or the HER2/Erbb2 receptor.

There are also, Melinda Telli told the audience at the San Antonio Breast Cancer Symposium (SABCS) earlier this month, no approved targeted therapies and no clear standard of care in metastatic disease.

Advances in understanding what TNBC is, rather than just what it is not, though, may be changing that.

"We've made a lot of progress in terms of understanding biology," Telli said, though she acknowledged that "dissecting this heterogeneity in the clinic has certainly proven complex... we have identified clinically relevant populations but lack a roadmap in terms of biomarkers to guide treatment selection."

Telli is assistant professor of medicine at Stanford University Medical Center and was speaking at a session symposium on "Molecular Subtypes and Clinical Ramifications of TNBC," where she gave an overview of new clinical approaches for TNBC.

'First branch point': DNA damage

In her overview of promising clinical trials for TNBC, Telli began by predicting that TNBC treatment is coming to a point "where we're going to reach our first node or branch point."

That branch point is to stratify patients on the basis of their germline BRCA status. Several trials, including Olympiad and Abrazo, have shown that BRCA-mutated tumors, because they are already deficient in DNA repair, are especially vulnerable to treatment with DNA-damaging agents, both platinum chemotherapy and PARP inhibitors.

At the meeting itself, phase III data from the EMBRACA trial comparing PARP inhibitor talazoparib (Medivation Inc.) to physician's choice chemotherapy in women with HER2-negative breast cancer added to the successes of PARP inhibitors in TNBC. Talazoparib-treated women in both the hormone receptor (HR)-positive and the TNBC subgroups had better progression-free survival than those treated with chemotherapy.

And while the overall survival data (OS) are not yet mature, there is a trend toward benefit in OS as well. If OS is indeed improved, that would be a welcome contrast to the Olympiad and Abrazo trials, which showed improvements in PFS but not OS with PARP inhibitor treatment.

The reason TNBCs are often particularly vulnerable to PARP inhibitors and platinum-based chemotherapy is that those agents cause DNA damage, which the cells are hard-pressed to repair. Several other experimental compounds that are showing promise in TNBC also cause DNA damage. Pharmamar SA's lurbinectedin (PM-01183), which made quite a splash at ESMO 2016 when a trial reported a 40.7 objective response rate and another 40 percent stable disease in patients with TNBC, works by binding to DNA and interfering with replication.

And Telli's talk at SABCS was directly preceded by one from Samuel Aparicio, the Nan & Lorraine Robertson chair in breast cancer research at the University of British Columbia, who spoke on targeting DNA repair deficiency in TNBC with G-quadruplex stabilizers. That approach is also being tested in phase I.

Many roads

There are many other targeted approaches being tested in TNBC beyond PARP inhibitors and DNA damage.

A group that is defined by drug type rather than mechanism of action is the antibody-drug-conjugates (ADCs). Despite the fact that the reported data were from a single-arm phase II trial, Immunomedics Inc.'s sacituzumab govitecan (IMMU-132) garnered a lot of attention at the meeting.

In a trial of 110 women with metastatic TNBC, sacituzumab govitecan used as a single agent in the third-line setting was able to achieve a 31 percent response rate and a median duration of response of around nine months. A randomized phase III trial is underway, and in the meantime, given the dearth of options for metastatic TNBC, Immunomedics plans to use the phase II data to file for accelerated approval.

Also in a registration trial is glembatumumab vedotin (CDX-011, Celldex Inc.), while ladiratuzumab vedotin (SGN-LIV1A, Seattle Genetics) is in phase I.

An improved understanding of the mutational landscape of TNBC has revealed that one frequent issue is the activation of PI3K/AKT pathway, which can occur through mechanisms including loss of the tumor suppressor PTEN and mutations or amplifications of PIK3CA or Akt.

Data from the phase II LOTUS trial of Akt inhibitor ipatasertib significantly improved PFS when added to chemotherapy for the first-line treatment of metastatic TNBC. A phase III trial is in the works

And in a basket trial of protein kinase B inhibitor AZD-5363 (Astrazeneca plc) in patients with the AKT 317K mutation, all six patients with TNBC in the trial had either stable disease or partial responses.

Attempts to harness immune oncology or androgen receptor blockers have been less promising.

Slightly less than 10 percent of TNBCs overexpress the androgen receptor, but trials of AR blockers have either led to no objective responses, or to response rates in the single digits. Nevertheless, Telli said, some patients also benefited in the form of prolonged stable disease. She said that despite the so-far anemic trial results, "there remains enthusiasm" for using AR blockers, and predicted that "other agents will move forward."

One such agent is Calithera Biosciences Inc.'s CB-839, a first-in-class glutaminase inhibitor. A poster with phase I results was presented in a spotlight session at the meeting. Although the numbers were so small as to be little more than case studies, the data showed that all three patients with luminal AR subtype – which is characterized by the expression of AR and several metabolic enzymes – had partial responses or stable disease, which was enough for the authors to highlight that patient group as having the strongest clinical benefit.

It is impossible to find an indication in oncology where PD-1/PD-L1 inhibitors are not being tested – with good reason, given how especially PD-1/PD-L1 blockers have transformed oncology.

In TNBC, though, those attempts are not yielding quite the spectacular results seen in some other indications. Proof of concept has been demonstrated in metastatic TNBC with several of the PD-1/PD-L1 targeting agents.

Telli described overall response rates as "modest" overall when the agents are used as monotherapies. Those response rates have been around 25 percent in the first-line setting in several clinical trials, and in the single digits when checkpoint inhibitors are used after other treatment options have failed. Combination approaches showing higher response rates have been seen in single-arm trials to date.

As in other tumor types, there have also been conflicting reports about the importance of PD-L1 expression levels as predictive biomarker in TNBC.

Telli summarized that "based on the modest response rate seen in later-line, there is some concern about ... whether that will ultimately prove to be a viable strategy." Given the number of trials addressing the question, investigators are sure to find out.