Editor’s Note: Currently, neoadjuvant therapy has become an optimized treatment modality for locally advanced and some early-stage breast cancers. It can optimize local surgery by downsizing tumors and provide efficacy and prognostic information to guide adjuvant therapy, thus improving survival. Moreover, it serves as an excellent platform for personalized treatment efficacy prediction and new drug development. Translational research based on neoadjuvant platforms has significant value in selecting benefiting populations, predicting effective regimens, and achieving personalized and precise treatments. At the “2024 National Breast Cancer Congress,” Professor Yingying Xu from The First Affiliated Hospital of China Medical University delivered a speech titled “Reflections on Translational Research Based on Neoadjuvant Platforms,” which we will introduce in this article.
Professor Yingying Xu introduced that since the 1970s and 1980s, experts and scholars in Italy, the United States, and other countries have been exploring the possibility of neoadjuvant chemotherapy combined with radiotherapy for locally advanced breast cancer. With the development of related research, neoadjuvant platforms have assumed increasingly diverse functions, such as assessing the efficacy of new drugs and guiding subsequent intensified treatments. After 2010, with studies like ADAPT exploring de-escalation strategies, translational research based on neoadjuvant platforms has also been exploring optimized regimens and selecting target populations for neoadjuvant therapy.
01 Examples of Translational Research Based on Neoadjuvant Platforms
The I-SPY2 study, which began in 2009, is a continuous multicenter, randomized, non-blinded, Phase II neoadjuvant clinical trial for early-stage breast cancer, serving as the cornerstone of “platform studies.” Using an adaptive randomization design, it compares multiple experimental groups and control groups within one trial framework, evaluating multiple groups and populations simultaneously. Through subsequent translational analyses, effective populations and regimen information are obtained. The series of study results have been reported in international high-impact journals and academic conferences multiple times.
In 2016, the study reported the benefit of neratinib in HR-/HER2+ breast cancer and found that the HER/PI3K/AKT signaling pathway was associated with neratinib efficacy, while EGFR Y1173 and STMN1 were effective predictive biomarkers. Carboplatin combined with veliparib was shown to be an advantageous regimen for triple-negative breast cancer (TNBC), and the efficacy of PARP inhibitors was not limited to BRCA1/2 germline mutations. In 2021, the DOP regimen (durvalumab/olaparib/paclitaxel) was shown to significantly improve pCR in high-risk HER2-negative breast cancer, with exploratory analysis suggesting that patients with a Mammaprint high-risk profile benefited more significantly from DOP treatment, with a pCR rate as high as 64%.
In 2022, researchers included 990 patients from 10 treatment arms of the I-SPY 2 study. Based on genomic, proteomic, phosphorylated proteinomic, and other multi-omics data, they systematically revealed efficacy predictive biomarkers for each treatment arm, optimized the molecular subtypes of breast cancer based on HR or HER2 status, and proposed corresponding treatment strategies. In 2023, it was reported that ctDNA could predict the efficacy of neoadjuvant chemotherapy in high-risk HER2-negative breast cancer early, providing theoretical basis for the design of subsequent clinical trials.
TNBC has been primarily treated with chemotherapy due to the lack of effective treatment targets until 2018, when the PD-L1 inhibitor Atezolizumab opened the door for the clinical application of immune checkpoint inhibitors (ICIs) in breast cancer. Compared to the advanced stage, the neoadjuvant period has a higher tumor antigen load, which is conducive to the activation of effector T cells. Moreover, tumor-specific T cells can enhance the detection and killing of subclinical metastases. The KEYNOTE-522 study achieved dual endpoints of pCR and EFS for the first time by combining the PD-1 inhibitor Pembrolizumab with neoadjuvant chemotherapy, and it was included in guidelines, establishing a new standard for neoadjuvant treatment of TNBC. However, considering the differences in the enrolled populations, regimen design, drug combinations, and results of several neoadjuvant immunotherapy studies such as NeoTRIP and GeparNeuvo, we can see that there are still unclear issues under the current standards that need to be explored through translational research based on neoadjuvant platforms to find answers.
The benefiting population is the most concerning issue for clinicians. Mechanistically, tumor mutational burden (TMB), T cell infiltration (TILs), and PD-L1 expression levels can all affect the efficacy of immunotherapy. The IMpassion130 and KEYNOTE-355 studies of first-line treatment in advanced stages confirmed that only PD-L1-positive TNBC patients benefited from immunotherapy. However, in neoadjuvant treatment studies such as IMpassion031 and KEYNOTE-522, there was no statistically significant correlation between PD-L1 status and efficacy.
Subgroup analyses of IMpassion031 and GeparNeuvo showed that although higher stromal TILs were associated with better efficacy at baseline, changes in intratumoral TILs before and after treatment were more significantly correlated with pCR benefit, indicating that while TILs have predictive value, they lack specificity. Trials like GeparNeuvo have used methods like RNA-seq to depict mutational differences between pCR and non-pCR populations, showing that TMB cannot be an independent predictive factor and is often used in conjunction with TILs and immune gene characteristic factors. Therefore, in neoadjuvant immunotherapy for breast cancer, none of these three factors can independently predict the advantageous population. Since the immune response involves multiple steps such as immune activation, infiltration, and killing, establishing a multi-factor predictive model seems more reasonable. The GeparNeuvo trial once built a five-gene pCR prediction model with good early predictive efficacy; the NeoTRIP study used mass spectrometry imaging to analyze multidimensional spatial information, confirming that cell phenotype, activation, and spatial tumor microenvironment collectively determine the efficacy of immunotherapy.
A Phase II single-arm study (TREND) evaluating the efficacy and safety of neoadjuvant trastuzumab deruxtecan combined with P-EC chemotherapy in early TNBC, in which Professor Yingying Xu’s team participated, collected tumor tissue and peripheral blood samples at baseline, after one cycle, and after surgery, and conducted multi-omics detection using methods such as scRNA-seq, scTCR-seq, CyTOF, RNA-seq, WES, and multiplex immunofluorescence. As the immune microenvironment is dynamic, we not only screened genes related to baseline and pCR to build predictive models but also added dimensions. Clustering analysis was performed based on the trend characteristics of gene changes and pCR status after one cycle and after surgery, revealing significant upregulation of T cell activation, lymphocyte proliferation, and related immune functions in the pCR population after one cycle.
In the future, it is possible to integrate genomic, transcriptomic, proteomic, and other molecular-level omics data with clinical and imaging data, achieving a true “multi-dimensional omics” approach to more comprehensively present patients’ response to immunotherapy. Since predictive models developed based on a single study have limitations, they can be optimized through data integration and validated with large-sample multicenter data before being truly generalized to clinical practice.
02 Evaluation of Neoadjuvant Immunotherapy Efficacy
Functionally intact tumor-draining lymph nodes are rich in tumor-specific PD-1+ CD8+ T cells, crucial for tumor antigen presentation and T cell activation, and closely related to the efficacy of immunotherapy. The commonly used neoadjuvant treatment evaluation system is the MP grading system, which has the drawback of only assessing the primary lesion and not including lymph nodes. Subgroup analysis of KEYNOTE-522 showed that patients with lymph node positivity achieved higher pCR rates with chemotherapy combined with immunotherapy, but the EFS benefit was similar to patients with lymph node negativity, failing to provide efficacy value information for lymph nodes in immunotherapy.
In 2023, the SABCS reported two studies, KEYNOTE-756 and Checkmate 7FL, which targeted high-risk HR+/HER2- breast cancer with neoadjuvant chemotherapy combined with immunotherapy, both achieving a clinical benefit of approximately 10% increase in pCR. Although survival outcomes are still pending, these two studies provide new insights into treatment strategies for traditionally considered “cold tumor” subtypes. Interestingly, in the Checkmate 7FL study, 79.6% of patients were lymph node positive, and in the KEYNOTE-756 trial, the proportion of patients with lymph node positivity was as high as 90.1%. Although the data currently released did not separately show the remission status of primary lesions and lymph nodes, based on the direct targeting effect of immunotherapy on lymph nodes, it is reasonable to speculate that the remission of lymph nodes may increase the overall pCR rate.
In the TREND study, Professor Yingying Xu’s team found that the lymph node pCR rate of patients receiving neoadjuvant immunotherapy was significantly higher than that of patients in the neoadjuvant chemotherapy cohort (non-head-to-head), and the pCR rate of primary lesions was lower than that of lymph nodes. Further scRNA and scTCR analysis suggested that precursor cell depletion was significantly higher in metastatic lymph nodes than in primary lesions, and the TCF7 and T cell activation pathways were significantly enriched in metastatic lymph nodes, indicating that even in the presence of metastasis, there are still many T cells in early depletion states in lymph nodes, which have the potential to be reversed and activated by immunotherapy. The TREND study suggests that metastatic lymph nodes can respond more quickly and effectively to immunotherapy, and they should have greater value in the immunotherapy efficacy evaluation system. For the local purpose of neoadjuvant treatment, the remission of lymph nodes also provides more opportunities for axillary downstaging surgery.
03 Timing of Neoadjuvant Immunotherapy Efficacy Evaluation
Unlike previous treatment modalities, immune-related adverse events may accompany patients for life, and serious adverse events such as immune storms and hyperprogression will result in permanent discontinuation of medication or death. Therefore, identifying non-sensitive patients for treatment regimen adjustment in the neoadjuvant stage can effectively reduce the collateral damage caused by ICIs. The translational research of the NeoTRIPaPDL1 trial sequenced biopsy tissues at baseline and on the first day of the second cycle, and jointly analyzed their intrinsic gene characteristics and the correlation between 27-gene immune scores and efficacy. The results showed that the baseline immune score could predict pCR in the immune combination group, and super responders with no residual tumors after the first cycle of treatment had higher baseline immune scores, and the dynamic changes before and after the first cycle of treatment could effectively reflect treatment efficacy, and the 27-gene score had early predictive value.
In the TREND study, Professor Yingying Xu’s team also discovered a non-typical response pattern different from hyperprogression, pseudo-progression, etc., called “early response,” which means achieving clinical CR early in treatment, and patients who experienced early response were more likely to achieve pCR. To explore the local tumor microenvironment characteristics of early response patients, further analysis of transcriptome and immunome data before treatment, after one cycle of treatment, and after completing the full course of neoadjuvant treatment revealed that in pCR patients, immune-related pathways were significantly activated after one cycle of treatment, consistent with those enriched after completing treatment; whereas in non-pCR patients, the activation of various pathways after one cycle of treatment was similar to that before treatment, and the activation of immune-related pathways occurred later. Immunome evidence showed that in the early response population, CD4+ T cells and B cells were enriched in the local tumor microenvironment, and the proportion of exhausted mid-term T cells with reversal potential was higher. The results of the above translational research indicate that early assessment can be made after 1-2 cycles of neoadjuvant immunotherapy to screen out immunosensitive patients for continued treatment, while considering early treatment adjustments for patients with poor response and poor local tumor immune status.
In summary, translational research based on neoadjuvant platforms focuses on unmet clinical needs and key scientific phenomena, utilizes high-quality clinical cohort data, combines biological samples and high-throughput sequencing to conduct in-depth molecular analysis of tumors and microenvironments, further defines the advantageous population, identifies effective regimens, and obtains resistance information to guide personalized treatment and improve patient survival. With clinical practice as the framework and translational research as the cornerstone, it is hoped to truly integrate the theory and practice of neoadjuvant treatment, building a bridge to precision treatment for breast cancer.
Professor Yingying Xu
Associate Chief Physician, Breast Surgery
Professor/Chief Physician/Doctoral Supervisor
Specially Appointed Professor of Liaoning Province
Member of the Oncology Branch of the Chinese Medical Association
Member of the Breast Cancer Group of the Oncology Branch of the Chinese Medical Association
Executive Director of the Youth Council of the Chinese Anti-Cancer Association
Deputy Director of the Breast Cancer Professional Committee of the Chinese Anti-Cancer Association
Deputy Director of the Youth Committee of the Breast Cancer Professional Committee of the Chinese Anti-Cancer Association
Member of the Metastasis Professional Committee of the Chinese Anti-Cancer Association
Deputy Director of the Oncology Branch of the Liaoning Medical Association
Visiting Scholar, MD Anderson Cancer Center
Principal Investigator of 4 National Natural Science Foundation projects, and more than ten provincial and ministerial projects. Winner of 7 provincial and ministerial science and technology progress awards, 9th Liaoning Youth Science and Technology Award. Selected as a member of the “Hundred, Thousand, Ten Thousand Talents Project” of Liaoning Province and awarded the title of Leading Talent in Science and Technology Innovation in Liaoning Province.
