Abstract P5-13-12: Immune signatures and MammaPrint (ultra) high risk class (MP2) as predictors of response to pembrolizumab combined with the TLR9 agonist SD101 in the neoadjuvant I-SPY 2 TRIAL
Cancer Research(2022)
Abstract
Abstract Background: Pembrolizumab, an anti-PD-1 immune checkpoint inhibitor, is approved for treatment in multiple cancers and has been shown to increase pathologic complete response (pCR) and survival in the neoadjuvant setting in breast cancer. Pembrolizumab combined with paclitaxel followed by doxorubicin/cyclophosphamide (P+T->AC) was evaluated in HER2- patients in the neoadjuvant I-SPY 2 TRIAL and graduated in the HER2-, HR+HER2- and triple negative (TN) signatures. Our biomarker analysis revealed that immune cell abundance and MP2 class predicts response in HR+HER2- patients whereas tumor-immune proximity scores (multiplex-IF) and signaling signatures (mRNA) predict response in TN patients. In an effort to further improve response, the TLR9 agonist SD101 was added to Pembro (P+S+T -> AC) for testing in I-SPY 2. While P+S increased estimated pCR rates relative to control (T->AC), it did not graduate for efficacy. To better understand the biology underlying response to P+S, we evaluated 31 expression based biomarkers relating to immune, ER and proliferation as predictors of response to P+S overall and within subtypes. Methods: Data from 72 patients (HR+HER2-: 45; TN: 27) treated with P+S were available. Pre-treatment FFPE biopsies were assayed using Agilent gene expression arrays. We evaluated genes/signatures representing 6 immune checkpoint/targets (CD274, PDCD1, TLR9, TIGIT, LAG3, and TIM3), 14 immune cell types (e.g., TILs, T cells, CD8 T cells, Tregs, cytotoxic cells, dendritic cells, mast cells, B cells, macrophages, and neutrophils), 3 T/B-cell prognostic (e.g., ICS5), 5 Tumor-immune signaling (e.g., STAT1, Chemokine12, TIS, and Geparsixto), and 1 TGFB signaling signatures as predictors of response to P+S. We also assess ESR1/PGR and proliferation, and the prognostic marker MP2 class. Signature scores were calculated as previously published. We used logistic modeling to assess biomarker performance (likelihood ratio test, p<0.05). This analysis was also performed in a model adjusting for HR status, and within receptor subsets. For the dichotomous MP1/2, we used Bayesian modeling to estimate the pCR rates of patients in each class. Multiple hypothesis testing adjustment was performed using the Benjamini-Hochberg method. Our statistics are descriptive rather than inferential and do not adjust for multiplicities of other biomarkers outside this study. Results: Higher levels of most (24/29) immune biomarkers associate with pCR in the population as a whole (BH LR p<0.05). Among target genes, CD274 and PDCD1 strongly associate with pCR; however, TLR9 did not associate with response. As seen in previous biomarker analyses of IO agents including Pembro, there are major differences in predictive biology between receptor subsets. Immune cell subpopulation abundance signatures (13/14) and T/B-cell prognostic signatures (3/3) associate with pCR in HR+HER2- but not TN subsets. Whereas tumor-immune signaling signatures (4/5) dominated by chemokines and cytokines associate with pCR in both HR+HER2- and TN subsets. In addition, high ESR1/PGR and low proliferation signature levels associate with response in HR+HER2-, as does MP1/2 class, with an estimated 45% pCR in MP2 versus 17% pCR in MP1. Analysis of multiplex-IF immune markers is pending. Conclusion: Though TN patients are more responsive to Pembro+SD101 and other immunotherapies than HR+HER2- patients, many more immune biomarkers associate with pCR in the latter group. Only tumor-immune signaling signatures associate with pCR in both HR+HER2- and TN subsets. Response in the HR+HER2- subset is higher in MP2 class, high-proliferation, lower-ER tumors. Lack of predictive signal for TLR9 may help explain why the addition of SD101 to Pembro failed to further boost response. Citation Format: Denise M Wolf, Christina Yau, Michael Campbell, Hatem Soliman, Mark Magbanua, Ruixiao Lu, Nicholas O'Grady, Lamorna Brown-Swigart, Gillian Hirst, Laura Sit, Yvonne M Florence, I-SPY 2 TRIAL Investigators, Smita Asare, Doug Yee, Angie DeMichele, Don Berry, Laura Esserman, Jo Chien, Laura van 't Veer. Immune signatures and MammaPrint (ultra) high risk class (MP2) as predictors of response to pembrolizumab combined with the TLR9 agonist SD101 in the neoadjuvant I-SPY 2 TRIAL [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-13-12.
MoreTranslated text
Key words
tlr9 agonist sd101,pembrolizumab,immune signatures,mp2,i-spy
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined