SAN ANTONIO — Researchers at the San Antonio Breast Cancer Symposium presented data from a number of studies demonstrating that immune-related biomarkers may be useful in differentiating between patients who will likely benefit from certain kinds of oncologics and those who won't.
"Breast cancer has not typically been felt to be a disease that we should target with immune modulating therapy," Edith Perez, the deputy director at large at Mayo Clinic Cancer Center, said at the meeting. As such, studies in this space are preliminary, and the immune-related biomarker data presented at the meeting represent early findings that need further exploration, study investigators reminded.
In one study, researchers identified tumor-infiltrating lymphocytes (TILs) as a possible biomarker that may identify which HER2-positive patients can respond well to just chemotherapy and potentially forego HER2-targeted treatment. In another small, early-phase study funded by Merck, researchers saw signs that a small subset of triple-negative breast cancer (TNBC) patients with PD-L1-positive tumors responded to the PD-1 inhibiting immunotherapy Keytruda (pembrolizumab).
Drugmakers who are advancing targeted drugs and immunotherapies are interested in identifying biomarkers that can shed light on the ability of patients' own immune systems to fight cancer. In particular, if biomarker strategies can help elucidate which patients' immune cells aren't attacking the tumor and why, that can inform efforts to develop treatments than can enhance immune responses in poor responders or advance drugs that target specific immune response-inhibiting signals.
For example, Nature recently published several papers investigating why some cancer patients respond to PD-1 inhibitors, and others don't. PD-1, or programmed cell death protein 1, when it interacts with a ligand expressed on tumor cells called PD-L1, has a negative impact on the ability of T cells to attack tumors, essentially enabling the tumor to evade an immune attack. The Nature studies suggested, however, that patients who are PD-1 and PD-L1 positive respond better to PD-1-inhibiting drugs when they have substantial TILs compared to those who don't.
A number of companies are developing diagnostics that can identify which patients have these characteristics. In collaboration with the University of California, Los Angeles, drug developer OncoSec recently said it would test patient samples using PerkinElmer's quantitative imaging platform to see if it can differentiate responders to PD-1-targeting drugs from non-responders.
Meantime, this week scientists from the Institute of Cancer published a paper in the Journal of the Royal Society Interface describing a test that combines imaging technology and computerized analysis of tumor data that they believe can be used to assess the five-year survival rate in women with TNBC. The test employs an "intra-tumor lymphocyte" algorithm developed by ICR researchers that measures TILs based on their location in the tumor in relation to the amount of cancer cells. Women who fell below a certain threshold based on this algorithm had a worse survival prognosis, they said.
In this vein, at SABCS, researchers from Genentech also discussed data from animal models on its investigational HER2 T cell-dependent bispecific antibody for the treatment for HER2-positive breast cancer patients who have stopped responding to Herceptin and other HER2 targeted drugs.
Similarly, OncoSec is developing ImmunoPulse, which is designed to promote interleukin 12 (IL-12) – a protein that upregulates the immune system – directly in the tumor and boost tumor immunogenicity by increasing TILs and cancer-killing T cells. The company is studying the drug in combination with Keytruda to assess if ImmunoPulse can enhance the response in patients who otherwise would not benefit from anti-PD-1 drugs.
TILs in chemo response
In the nearly 1,000-patient Phase III study, dubbed N9831, researchers led by Perez had previously reported that the combination of chemotherapy and Genentech's HER2-targeted drug Herceptin improved recurrence-free and overall survival in women with HER2-positive breast cancer compared to those receiving just chemotherapy. However, the data also suggested that some patients were responding well to just chemo due to a greater number of TILs, and so Perez and her team wanted to further probe this possibility.
They found that at the median 4.4 years follow up, among the approximately 10 percent of study subjects who had tumors highly infiltrated with lymphocytes, those treated with just chemotherapy had an 80 percent decrease in the risk of their breast cancer returning.
"The patients who had lymphocyte predominant breast cancer had much better outcomes to chemotherapy alone compared to those who did not have lymphocyte predominant breast cancer and this was statistically significant," Perez said. Patients whose tumor samples comprised 60 percent or more immune cells were considered to considered to have high levels of TILs.
The greater benefit in recurrence-free survival observed in the chemo-treated lymphocyte predominant group was not a statistically significant finding, Perez emphasized. But in the subset of patients whose tumors had lower levels of TILs, "there was a clear benefit" for using chemotherapy plus Herceptin compared to just chemo, she noted.
Still, the fact that patients with high levels of TILs who received the Herceptin/chemo combination didn't have better outcomes than those with low levels surprised Perez and colleagues. An earlier study had hinted that patients with high TIL levels might benefit from Herceptin, but that trial was smaller and had lower statistical power than the latest analysis.
Ultimately, Perez and her team concluded that for the 10 percent of study participants with high levels of TILs, they could not prove that there is a benefit with chemotherapy plus Herceptin over just chemo. As such, the standard of care should not be altered, Perez stressed. Although the difference in outcomes was "quite suggestive" there were only 94 patients with high levels of TILs. "This is something we feel is potentially relevant," Perez noted, "but we need to do additional studies to corroborate our findings."
Perez and her team plan to research whether certain subtypes of immune cells – such as helper and suppressor TILs – differentially impact patients' disease recurrence, as well as correlate TILs with immune gene profiles. "Ultimately we would like to determine whether changing the amount and type of TILs will help improve patient outcomes," Perez said.
Jennifer Litton, director of the Breast Medical Oncology Education Program at the University of Texas MD Anderson Cancer Center, characterized this study as yet another analysis showing that women with high immune infiltrates in their tumors just do better with their therapies. "This study also confirms, as we see in the triple-negative subtypes, that this is a small portion of patients," Litton observed. "Future directions will be to enhance the immune infiltrates."
PD-L1 expression
In the Merck-funded trial, researchers led by Rita Nanda from the University of Chicago enrolled 32 metastatic TNBC patients between the ages of 29 and 72 years who had early-stage relapsed or advanced disease that had progressed on treatment. The study population was highly pretreated.
All participants were PD-L1 positive and received 10 mg/kg infusions of Keytruda every two weeks. Of the patients screened for PD-L1 positivity using an assay developed by Merck, less than 60 percent were found to meet the criteria, which Nanda described as "liberal."
Out of 27 patients with measurable disease for response, one experienced a complete response, four had a partial response, and seven had stable disease. Meanwhile, 12 patients' breast cancer progressed on the drug.
In terms of drug-related toxicities, although most reactions were low grade and manageable, Nanda reported that four patients had a grade 3 event and one experienced a grade 4 adverse event. This patient with grade 4 toxicity had decreased blood fibrinogen levels and went on to develop a condition called disseminated intravascular coagulation, where blood clotting proteins start behaving abnormally resulting in clots or serious bleeding. This patient died.
Additionally, three patients discontinued treatment due to disease progression. At the time of the SABCS presentation, there was one patient with a complete response to Keytruda and two with partial responses were still receiving the drug.
Nanda concluded that the drug, in this early phase of study, demonstrated "an acceptable safety profile" and promising efficacy in PD-L1 positive patients, and was worthy of further study.
Currently, there are no targeted treatment options in TNBC, a particularly aggressive form of breast cancer that lacks expression of estrogen receptor, progesterone receptor, and HER2. Median survival is one year from the time of diagnosis. Even though one patient died in the study, Keytruda still has a better safety profile than chemotherapy in the heavily pretreated TNBC population, according to Nanda.
Eric Winer from the Dana-Farber Cancer Institute noted that in studies of investigational therapies, deaths are unfortunately a reality. "This one death in a study is tragic and it's tragic for that woman," Winer said. "We sometimes forget that people are volunteering to participate in new drug trials and there are real risks with this." He said that the fact that even a handful of highly pretreated TNBC patients were doing well on Keytruda for 40 or more weeks "gives you a signal that says you have to look at this a little more."
One weakness of the study was that archival tissues samples were utilized. Nanda noted that future studies may benefit from fresh tissue biopsies. "Also there didn't appear to be a correlation with the degree of PD-L1 positivity to [Keytruda] response," Nanda said. "So, PD-L1 alone may not be the appropriate biomarker."
Keytruda has shown better efficacy in other malignancies where there is greater interaction between the immune system interaction with cancer cells, for example melanoma. The US Food and Drug Administration in September approved Keytruda – which Merck has priced at $12,000 per month – for advanced or unresectable melanoma patients who are no longer responding to other drugs. Still, the data the agency used to approve the drug – that 24 percent of approximately 80 patients treated with Keytruda experienced tumor shrinkage – suggests the need for predictive strategies to identify responders and non-responders. Merck is also studying how PD-L1 expression in advanced lung cancer patients impacts their ability to respond to Keytruda.
"Harnessing the immune system, we are just scratching the surface in breast cancer and we really haven't harnessed it yet to the best of our ability," Litton said. With this Phase I study, researchers aren't trying to decide whether to change the standard of care, she added, but are trying to decide "whether to move on to the Phase II trial," which Litton supported based on the early findings.
Merck is planning to launch a Phase II trial of Keytruda in TNBC patients in the first half of 2015.
T cell-dependent HER2 antibody
While Perez and colleagues presented "provocative" biomarker data at the conference that may be able to identify which patients may do well without Herceptin, Genentech researchers unveiled data on their investigational HER2 T cell-dependent bispecific antibody (HER2-TDB) that the company hopes will destroy tumor cells in patients who have become refractory to currently approved HER2 targeted drugs (i.e. Herceptin).
Researchers led by Teemu Junttila of Genentech showed in animal models that the antibody had anti-tumor activity. Specifically, they demonstrated that PD-L1 expression inhibited the ability of antibodies to recruit T cells. This sign of cancer resistance was reversed, researchers reported in their abstract, with anti-PD-L1 therapy and the combination of HER2-TDB with anti-PD-L1 immunotherapy resulting "in enhanced inhibition of tumor growth, increased response rates, and durable responses," in animal models.
"We characterize a critical resistance mechanism, a potential diagnostic marker, [and] a novel transgenic efficacy model, and significantly improve the drug-like properties by using technology based on full-length antibodies with natural architecture," Junttila and colleagues said in their abstract. "Finally, we demonstrate the benefit of combining two immune therapies — direct polyclonal recruitment of T cell activity together with inhibiting the T cell suppressive PD-1/PD-L1 signaling — results in enhanced and durable long term responses."
In a presentation at SABCS, another Genentech researcher noted that the preclinical study did not show any HER2 amplified cells that did not respond to the antibody. In particular, the fact that HER2-TDB appeared to increase T cell proliferation in tumors suggests that it might be useful in boosting the immune response of patients who have stopped responding to HER2-targeted drugs.