Despite Advances, Unmet Need Remains in TNBC

Kevin Kalinsky, MD, MS

Despite the emergence of PARP inhibitors, antibody-drug conjugates (ADCs), and immunotherapy in the treatment of patients with triple-negative breast cancer (TNBC), more work is needed to propel further progress, according to Kevin Kalinsky, MD, MS.

“The ultimate take-home message is that TNBC is a subtype of breast cancer for which there remains an unmet need for advances,” said Kalinsky. We have seen approvals for PARP inhibitors in patients with germline BRCA mutations, and we have a lot of exciting data with immunotherapy and with ADCs. My hope is that, in the next couple of years, this landscape is really going to evolve.”

In an interview during the 2020 OncLive^® State of the Science Summit™ on Breast Cancer, Kalinsky, an associate professor of medicine at the NewYork-Presbyterian Hospital/Columbia University Medical Center, highlighted recent developments in TNBC, including the potential role of immunotherapy in the neoadjuvant setting.

OncLive: Could you share the impact of the IMpassion130 trial?

Kalinsky: The IMpassion130 data really are game-changing. [Data from this trial led to] the first approval of immunotherapy in patients with breast cancer, specifically those with PD-L1—positive metastatic TNBC who are being treated frontline. Specifically, for TNBC, PD-L1 positivity is [found in] about 40% of patients based upon this 1 assay: the Ventana PD-L1 (SP142) assay. I would say that for those patients, giving nab-paclitaxel (Abraxane) plus atezolizumab (Tecentriq) really is the standard of care.

What has your anecdotal experience been with the regimen thus far?

A subset of patients really can benefit from checkpoint inhibition. Right now, we give checkpoint inhibitors to anyone who has PD-L1 positivity that's defined as at least 1%. It’s also important to note that in IMpassion130, there must have been at least a 12-month delay from the time that patients had received and completed their operable therapy to the time that they received treatment in the metastatic setting.

With that caveat, when we treat patients with frontline triple-negative PD-L1—positive disease, they tolerate this regimen well, anecdotally. We have to look for adverse events (AEs) like neuropathy from nab-paclitaxel, as well as immune-related AEs. It's important to be mindful of those events. [With this regimen], we see an improvement in progression-free survival (PFS) as well as this trend toward overall survival.

What are some of the immune-related AEs to be aware of?

When you are [administering treatment] with immunotherapy, there are several things that you should do. For one, it's important to continue to monitor thyroid [function] because patients can experience some thyroid abnormalities and may not even be necessarily symptomatic from that. Also, there can be some subtle toxicities, including issues with adrenal insufficiency, so you should be looking for [symptoms], such as fatigue. Lastly, some very rare toxicities [can occur], including pneumonitis, inflammation of the lungs, or colitis, among others. You should always just be mindful of [those events, especially] if patients are starting to complain of some strange toxicities.

Are you informing patients that fatigue could potentially be life-threatening? What goes into that discussion?

Sometimes, being able to discriminate AEs from chemotherapy as opposed to immunotherapy can be somewhat difficult. For instance, with chemotherapy, patients can [also] complain of fatigue. [These toxicities should] always just be in the back of your mind. It is important, before we start immunotherapy, that we explain to the patients that [they] may [experience] potentially lifelong toxicities. This is particularly important, even when we're thinking about utilizing immunotherapy in patients in the operable setting.

What other immunotherapies are under investigation in the metastatic setting?

Right now, the approval [for atezolizumab] is specifically for those patients who have PD-L1—positive metastatic TNBC and are being treated in the frontline setting. One study has looked at giving, for instance, single-agent pembrolizumab (Keytruda) compared with standard-of-care chemotherapy. However, as a whole, that was a negative study.

Other studies include the KEYNOTE-355 trial, in which investigators are giving various chemotherapy regimens along with pembrolizumab. We’re waiting for those data. [KEYNOTE-355] is another frontline study that may also change the treatment landscape of this disease.

Is PD-L1 expression the sole biomarker of response to immunotherapy in this space?

Data within the IMpassion130 trial have looked at other potential biomarkers, including CD8 positivity in tumor-infiltrating lymphocytes, and BRCA1/2 mutation status. Right now, PD-L1 positivity, as defined by the Ventana PD-L1 SP142 assay, remains the sole predictor of who will benefit from the addition of atezolizumab.

You mentioned BRCA1/2 mutations. What is the utility of this biomarker with regard to PARP inhibitors?

When we have a patient with metastatic TNBC, or a patient with hormone receptor (HR)—positive, HER2-negative disease, we do genetic testing to see whether or not a BRCA1/2 mutation is present. The reason [that testing is] important is because currently 2 PARP inhibitors are approved. [The studies that led to those approvals] looked at giving either a PARP inhibitor or chemotherapy. An improvement in PFS was observed [with the PARP inhibitors], as well a significant improvement in response rates. As such, we check for [BRCA1/2 mutations] in these patients; [these agents] could potentially benefit them.

If you order a test and it comes back that the patient has PD-L1 positivity of ≥1% and a BRCA1/2 mutation is also present, how would you approach treatment?

If we have a patient who is a BRCA1/2 mutation carrier and then also has a PD-L1—positive tumor, several studies are looking at combining [PARP inhibitors and immunotherapy agents] together. I would consider a clinical trial for any patients with metastatic TNBC. However, in the absence of a trial, I would commonly start with a checkpoint inhibitor in clinical practice because those agents have the potential to induce durable responses in a subset of patients. I would start with that agent and then sequence to a PARP inhibitor in the absence of putting someone on a clinical trial.

What are some ADCs that are under investigation?

Various ADCs are being examined right now, including, sacituzumab govitecan. We’ve seen the data from the phase I/II trial that have been published in the New England Journal of Medicine. Also, the ASCENT trial has compared that agent with standard-of-care chemotherapy; we don't have the data [from that study yet]. This drug is not yet commercially available; however, it appeared to be quite active in patients with metastatic TNBC. Right now, Trop-2 is not an integrated marker, meaning that a patient does not necessarily have to express that [in order to receive this agent]; [it doesn’t] define who would benefit. However, we're awaiting the official approval [of the drug] and what the exact label will say.

Switching to the neoadjuvant setting, could you discuss the data from the KEYNOTE-522 trial?

We originally saw the data from KEYNOTE-522 presented at the 2019 ESMO Congress, then some updates at the 2019 San Antonio Breast Cancer Symposium. In the trial, patients were randomized to [pembrolizumab or placebo, both given with paclitaxel plus carboplatin, then doxorubicin or epirubicin plus cyclophosphamide]. They went on to surgery and then patients who were randomized to pembrolizumab continued that for 1 year versus observation.

Ultimately, the initial data showed a significant improvement in pathologic complete response (pCR). It did seem that the patients who were higher risk, meaning those who were lymph node—positive or had stage II/III disease, for instance, had a higher benefit with the addition of pembrolizumab. This combination is not FDA approved at this point. It’s also worth noting that an initial early read-out of event-free survival [started to show] some improvement in the Kaplan–Meier curve for those patients. However, again, it's important to reiterate that at this moment, this is not an FDA-approved neoadjuvant regimen.

Could pCR be a surrogate marker for event-free survival (EFS)?

For patients with certain types of breast cancer—TNBC, HER2-positive disease, even potentially proliferative tumors, such as those that are HR-positive and HER2-negative—there's this notion [based on] pooled data [suggesting that] having an improvement of pCR may ultimately be a surrogate marker for EFS. This has been replicated with some studies—for instance, what we've seen in the HER2 space with targeted therapies. For patients who are receiving checkpoint inhibition or immunotherapy, the question of whether or not an improvement of pCR will ultimately translate into an improvement in EFS [remains unanswered]. Even the NeoTRIP trial, which did not achieve an improvement of pCR, was powered to look at an improvement in EFS. We’ll see, from longer-term follow-up, whether or not a patient could still [experience] an improvement in EFS in the absence of a pCR improvement.