REPORTING FROM SABCS 2017
SAN ANTONIO (FRONTLINE MEDICAL NEWS) – The immune checkpoint inhibitor pembrolizumab overcomes trastuzumab resistance in HER2-positive advanced breast cancer provided that the tumor expresses programmed death ligand 1 (PD-L1), a trial reported at the San Antonio Breast Cancer Symposium suggests. But presence of immune cells in the tumor is a major additional determinant of benefit.
“HER2-positive breast cancers have been observed to contain high levels of T-cell infiltration. Tumor-infiltrating lymphocytes (TILs) are associated with improved prognosis and higher response rates to trastuzumab and chemotherapy, suggesting that modulating anti-tumor immunity could further improve survival outcomes,” according to first author Sherene Loi, MD, PhD , an associate professor at Peter MacCallum Cancer Centre in Melbourne, explaining the trial’s rationale. “Trastuzumab itself has been shown to have immune-mediated mechanisms of action, and preclinical studies have suggested that immune-mediated mechanisms of trastuzumab resistance can be overcome with checkpoint inhibition combinations.”
The single-arm phase 1b/2 trial, called PANACEA (also KEYNOTE-014), enrolled 58 patients with HER2-positive advanced breast cancer that had progressed on trastuzumab (Herceptin) or trastuzumab emtansine (Kadcyla). All were given pembrolizumab (Keytruda), which unleashes antitumor immunity by targeting the programmed death-1 receptor on immune cells, in combination with trastuzumab.
With a median follow-up of 13.6 months, the cohort of patients having tumors positive for PD-L1 achieved an overall response rate of 15.2% and a disease control rate of 24%,” Dr. Loi reported in a press briefing and session, on behalf of the International Breast Cancer Study Group and Breast International Group. In contrast, there were no responses in the PD-L1–negative cohort.
Within the PD-L1–positive cohort, stromal levels of TILs in the metastatic lesion – which were low overall – influenced likelihood of benefit. The response rate was almost eight times higher in patients who had at least 5% of the stromal area densely infiltrated with TILs.
“The PANACEA study met its primary endpoint in the PD-L1–positive cohort. For responders, this combination offers durable control without chemotherapy,” Dr. Loi summarized.
“Metastatic HER2-positive breast cancer in this [heavily pretreated] setting is poorly immunogenic, as evidenced by the majority of patients having low TILs in their metastatic lesions. Saying that, however, we did observe a higher response rate in this study as compared to the equivalent triple-negative breast cancer studied in KEYNOTE-086,” she noted. “Future directions in this disease space should focus on combinations with effective anti-HER2 therapy, particularly in low-TIL patients.”
Predicting benefit
The trial is noteworthy for its efforts to identify the subset of patients most likely to benefit from immune checkpoint inhibition, according to press briefing moderator Virginia Kaklamani, MD , a professor of medicine in the division of hematology/oncology at the University of Texas Health Science Center, San Antonio, and a leader of the Breast Cancer Program at the UT Health San Antonio Cancer Center.
“With the triple-negative patient population, we are still struggling to find that subset, and all of the studies that are now being done are looking at all triple-negative breast cancers,” she said. In contrast, “with FISH, we can identify the HER2-positives, and now we are starting to look at PD-L1 expression, we are starting to look at TILs, instead of all the HER2 positive patients.”
In similar studies among patients with HER2-negative breast cancer, PD-L1 did not pan out as a strong predictive biomarker. “What do you think the difference is between that subset and the HER2-positive subset?” Dr. Kaklamani asked.
“First off, I think that there are technical issues with the PD-L1 assay. And we find that patients with high TILs or immune infiltration usually have high levels of PD-L1 expression on their TILs,” Dr. Loi replied. “So I think that PD-L1 can be expressed on the tumor as well as the TIL, and it certainly seems to be the TIL infiltrate that probably enriches for responders to a PD-L1 checkpoint inhibitor on its own or in this case with trastuzumab.”
Study details
In the PANACEA trial (additionally known as IBCSG 45-13 and BIG 4-13), the most common adverse event of any grade and type with the pembrolizumab-trastuzumab combination was fatigue, seen in 21% of patients, Dr. Loi reported. For immune-related adverse events specifically, 19.0% of patients experienced an event, 10.3% experienced an event of grade 3 or worse, and 6.9% stopped treatment because of these events.
“These frequencies are consistent with what has been reported in other solid tumor types with pembrolizumab,” she commented. There were no cardiac events reported.
Efficacy analyses were restricted largely to the PD-L1–positive cohort, given the lack of any response in the negative cohort.
Median duration of response in the positive cohort was 3.5 months, and median duration of disease control was 11.1 months. Five patients (10.8%) remain on treatment with no progression; three of them have completed 2 years of pembrolizumab.
Median progression-free and overall survival were 2.7 and 16.1 months, respectively; corresponding 12-month rates were 13% and 65%. “There is a tantalizing suggestion of a tail on the curve. … Obviously, this requires further follow-up, and the numbers are small,” Dr. Loi commented.
The median baseline stromal TIL level in metastatic lesions was just 1%. “This is 20 times less than what we observe in primary HER2-positive breast cancers,” she pointed out.
Compared with the PD-L1–negative cohort, the PD-L1–positive cohort had higher TIL levels. Additionally, within that latter cohort, TIL level was higher among patients achieving response versus not (P = .006) and patients achieving disease control versus not (P = .0006).
“We then went on to try to identify a TIL cutoff that could enrich the population for responders. This has been done in other solid tumor types,” Dr. Loi explained.
Analyses in the PD-L1-positive cohort showed that TIL levels down to 5% predicted benefit. The 41% of patients having 5% or more TILs were dramatically more likely to have a response (39% vs. 5%) and disease control (47% vs. 5%).
TIL levels varied widely according to site of the metastasis, with higher levels seen in metastases from lung and lymph nodes, and lower levels seen in those from liver and skin.
“At this stage, we are not sure which is the chicken and the egg: Patients could have disease in their lung and their lymph nodes because their immune system is better controlling their disease,” Dr. Loi commented. “How we treat these patients is still an open question. In patients with liver metastases, perhaps we need to be more aggressive with the primary or tumor-control anti-HER2 therapy.”
Improving efficacy
Going forward, one strategy for improving pembrolizumab efficacy in this patient population might be priming the immune response, according to Dr. Loi.
“In HER2 disease, it’s very clear that oncogenic signaling is the driver, so targeting HER2 potently also will help relieve tumor-mediated immune suppression,” she elaborated. “In this particular context, targeting HER2 well is the key. Whether you need the addition of a little bit of chemo or some radiation, all this needs to be studied.”
Another strategy for improving pembrolizumab efficacy might be moving the drug to earlier disease settings, Dr. Loi proposed.
“By the time you get to advanced stage and have had multiple treatments, you actually have low levels of T-cell infiltration in your metastatic lesion, for whatever reasons – tumor burden, immunosuppression, multiple lines of treatment. That all reduces your chance of responding to pembrolizumab, for example, as monotherapy,” she elaborated. “We don’t know yet if chemotherapy in addition to pembrolizumab could change that tumor microenvironment. But still, I think the earlier in lines you go, the more chance you are going to have of preexisting effective antitumor immunity that can be reactivated with the addition of pembrolizumab.”
Dr. Loi disclosed that her institution receives research funding from Novartis, Pfizer, Merck, Genentech/Roche, and Puma. Merck provided study drug and support for PANACEA.
SOURCE: Loi S et al. SABCS 2017 Abstract GS2-06 .
