With Immune Cells in Breast Tumors, the More the Merrier

Neil Osterweil

December 11, 2013

SAN ANTONIO, Texas — Researchers are increasingly shedding old notions that breast cancers lack immunogenicity or the ability to produce an immune response, in light of new evidence from 3 clinical trials presented here today at the San Antonio Breast Cancer Symposium.

Each of the trials showed that the presence of tumor-infiltrating lymphocytes (TILs), which are a type of immune cell and were found primarily in breast stromal tissue, correlated positively with good outcomes.

The findings also suggest new immunotherapy options for patients with notoriously treatment-refractory tumors such as triple-negative breast cancers, investigators said.

Every 10% increase in stromal TILs was associated with a 14% reduction of risk for recurrence or death (P = .02), 18% reduction in risk for distant recurrence (P = .04), and 19% reduction in the risk for death (P = .01) in women with triple-negative breast cancer treated in two US randomized clinical trials from the Eastern Cooperative Oncology Group (ECOG), reported Sylvia Adams, MD, of the NYU Langone Medical Center in New York City, and colleagues.

Similarly, among women with operable or locally advanced HER2-positive breast cancer treated with trastuzumab in the GeparQuattro trial from Germany, every 10% increase in TILs was associated with a 16% increase in pathologic complete responses (pCR), reported Sherene Loi, MD, PhD, from the Peter MacCallum Cancer Centre in Melbourne, Australia.

And in another German trial, the GeparSixto study of women with triple-negative or HER2-positive breast cancers, patients with lymphocyte predominant breast cancers (LPBC) had a 75% pathologic complete response (pCR) rate with carboplatin, compared with only 34% for women without LPBC, said Carsten Denkert, MD, from the Institute of Pathology at Charité University in Berlin, Germany.

The findings of these 3 studies suggest that adding agents targeted at enhancing immune responses may enhance the efficacy of existing regimens.

"The lymphocytic infiltrate has a very prognostic impact because it can actually mediate an anti-tumor response," Dr. Adams said.

TILs in HER2

"Breast cancers have not traditionally been considered an immunogenic disease when compared to other solid tumor types like melanoma and renal cell carcinoma where immunotherapies have been used with some success," Dr. Loi said.

Her team previously demonstrated that the presence of TILs in breast tumor samples from the FinHer trial was significantly associated with greater response to trastuzumab in patients treated with trastuzumab and chemotherapy.

In the current GeparQuattro study, they sought to confirm the positive association between TILs and higher responses to trastuzumab and chemotherapy in tumor samples from GeparQuattro.

They found in a multivariable analysis stratified by chemotherapy type that the odds ratio (OR) for a pathologic complete response with every 10% increase in stromal TILs was 1.19 (P = .038). Neither age, nodal status, histologic grade or tumor stage was significantly associated with a likelihood of a pCR. The only other significantly predictive factor was estrogen-receptor status OR = 2.14, P = .03).

They then looked for clues to the composition of TILS and how trastuzumab might mediate their effect in samples from the FinHer trial. They found that trastuzumab appears to act both directly on tumor cells, and indirectlly to counteract suppression of anti-tumor effector cells.

They also found in a HER2-driven mammary tumor mouse model that combining trastuzumab with a T-cell checkpoint inhibiting agent, such as anti-CTLA-4 monoclonal antibodies (such as ipilumumab), or PD-1 inhibitors (such as nivolumab), have synergistic anti-tumor activity.

"These data provide the rationale to evaluate if a T-cell checkpoint inhibitor added to trastuzumab can further improve clinical outcomes in HER2-positive disease," Dr. Loi said.

Prognostic Value

Dr. Adams and colleagues also sought to confirm evidence of clinical validity of TILs as prognostic biomarkers, this time in patients with triple-negative cancers (ie, lacking the HER2, estrogen and progesterone receptors).

They reviewed data on patients enrolled in the ECOG trials E2197., and E1199, both of which compare various chemotherapy regimens plus tamoxifen in women with cancer metastatic to lymph nodes.

They found that among 481 evaluable samples, 80% of tumors had stromal TILs, but only 15% had intraepithelial TIL's. An additional 4.4% of the samples had LPBC, defined in this study as 50% or greater intraepithelial or stromal TILs in the tumor.

As noted before, the higher the TIL count, the better the prognosis. In a multivariate model, every 10% increase in stromal TILs was associated with a hazard ratio (HR) of 0.84 (P = .005) for disease-free survival, 0.81 (P = .02) distant recurrence free survival and 0.79 (P = .003) for overall survival, respectively.

Only stromal intraepithelial TIL scores proved to have prognostic value in a binary variable model, with a HR of 0.62 (P = .007). Neither intraepithelial TIL score nor LPBC were significant as predictors of outcome.

"There is a hope that by targeting the immune system, we can improve cure rates [for triple-negative breast cancers], " Dr. Adams said.

Pretreatment Analysis

The GeparSixto investigators prospectively evaluated TILs before randomization in the trials, with each core biopsy slide assessed by 2 of 5 pathologists according to predefined parameters.

Patients with centrally confirmed triple-negative or HER2-positive breast cancer were randomized to receive either paclitaxel and nonpegylated liposomal doxorubicin (Myocet) alone, or the same combination plus carboplatin, plus trastuzumab and lapatinib for women with HER2-positive tumors, and bevacizumab for women with triple-negative disease.

In contrast to Dr. Adams' group, they used a cutoff of 60% of greater TILs per tumor to define LPBC.

They found among all patients that stromal TILs as a continuous parameter were predictive of response to carboplatin, and that in patients with triple-negative tumors and HER2 –positive tumors considered separately, only the addition stromal TIL was predictive of prognosis only in those randomized to receive carboplatin.

"5000 Targets at Once"

A breast cancer expert who is currently engaged in clinical trials with anti-CTLA-4 and anti-PD-1 agents in triple-negative breast cancer says that studies have consistently shown that patients with triple-negative disease who have lymphocytic tumor infiltrates fare better.

"It isn't just triple-negative, though: we're seeing it in every tumor type," said. Jennifer Litton, MD, associate professor of Breast Medical Oncology at the University of Texas MD Anderson Cancer Center in Houston, Texas.

She said, however, that TIL levels are highly dynamic, with evidence of infiltration varying on pathologic sample staining from one day to the next.

Nonetheless, she says she is "excited' about the potential for harnessing the immune system in difficult-to-treat breast cancers.

"When you look at a triple-negative tumor especially, every time you subject it to chemotherapy, you're constantly causing more and more aberrations. So do we think one targeted therapy is going to fix it? No! It will just find another way," she said.

"The problem with targeted therapy is that you need to hit 5000 targets at once, but the immune system, the more irregular the cell looks like, if we can harness it correctly, it will find those really irregular cells. So putting it together with chemo, with possibly a targeted agent, upregulating the immune system could potentially be a game changer," she said in an interview with Medscape Medical News.

Dr. Litton was not involved in the studies described, but moderated a media briefing where Dr. Loi's data were reported.

Dr. Loi's study was funded by the European Union Seventh Framework Program, RESPONSIFY project, Breast Cancer Research Foundation, and the Fonds JC Heuson. She declared no conflicts of interest.

Dr. Adams' study was funded by the National Cancer Institute, with supporting grants from Sanofi-Aventis and the Breast Cancer Research Foundation. She is supported in part by NCI grants.

Dr. Denker's study was supported by RESPONSIFY. He declared ownership interest in Sividon Diagnostics.

36th Annual San Antonio Breast Cancer Symposium: Abstracts S1-05, S1-06, and S1-07, presented December 11, 2013.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.