Integrating mutational signatures and immune circulating biomarkers to enhance immunotherapy efficacy in triple negative breast cancer

Background and Rationale: Immune checkpoint inhibitors (ICIs) have improved the outcomes of patients with triple-negative breast cancer (TNBC) in both early and metastatic settings. However, the risk-to-benefit ratio of ICIs remains suboptimal due to a scarcity of predictive biomarkers. Current research has identified two types of potential biomarkers for immunotherapy: those that track T-cell inflammation in the tumor microenvironment (TME) and those related to tumor neoantigen load. Tumor infiltrating lymphocytes (TILs) are the most validated biomarker of the first category. However, TILs vary significantly across metastatic sites, rendering them unsuitable for immune response evaluation in metastatic patients. Tumor mutational burden (TMB) and mutational signatures (specifically HRD, APOBEC, and MRD) serve as indirect measures of tumor antigenicity caused by somatic mutations. Compelling data suggest that these biomarkers could help predict the benefit of immunotherapy in TNBC. Nonetheless, the interplay between TME-specific and tumor-specific biomarkers in TNBC treated with ICIs has not yet been investigated. On these premises, we aim to develop a biomarker integrating TME and TMB/mutational signatures to improve patient selection for immunotherapy. Additionally, we aim to overcome the limitation of TILs variability in metastases by using circulating lymphocytes profiling as a non-invasive, indirect measure of the TME immune status. Aims: The project has two aims: 1)To improve the prediction of ICI efficacy in TNBC (both early and metastatic) by stratifying patients based on a combined evaluation of circulating lymphocytes and TMB/mutational signatures; 2)To enhance the assessment of the tumor-host immune axis in the metastatic setting using circulating lymphocytes as surrogates for TILs. Tasks: This project is a correlative analysis of the ongoing IRIS trial, which investigates tissue and blood biomarkers and their association with ICI benefit in TNBC. Peripheral blood and tumor biopsies collected from patients initiating either neoadjuvant or first-line chemo-immunotherapy within the IRIS study will be analyzed. Multi-parameter flow cytometry will be used to assess the percentage of circulating lymphocyte populations and their functional states. Primary tumors and metastatic biopsies will be used to determine TMB and mutational signatures in early-stage and metastatic cases, respectively. A panel of 553 key cancer genes will be employed. TILs and their populations will be assessed in the primary tumors of early-stage patients. The translational results will be correlated with pathologic complete response, progression-free survival, and overall survival, as appropriate. Impact on Patients: The Food and Drug Administration (FDA) has recently issued warnings about the risks for patients with curable cancers treated with immunotherapy, which is associated with infrequent, though potentially fatal, toxicities. In early TNBC, tumor stage alone currently dictates the eligibility for prescribing ICIs with neoadjuvant chemotherapy, placing patients with TNBC at higher risk of unnecessary exposure. Regarding metastatic TNBC, the current selection of patients based on PD-L1 assessment by IHC has shown modest improvements in outcomes. About 10-15% of patients still progress or die within three months of initiating ICIs, and only 50% survive beyond 24 months. The results generated in this project will advance our knowledge of how to better select these patients for immunotherapy both in the early and metastatic setting, improving the effectiveness of this treatment and the lives of patients with TNBC. BACKGROUND AND SCIENTIFIC RATIONALE Triple negative breast cancer (TNBC) is the most aggressive breast cancer subtype and a leading cause of cancer-related death among women worldwide1,2 . Approximately one in every three patients with early-stage TNBC experiences a distant recurrence within 2-3 years of diagnosis2 . For those whose disease has spread to distant sites, the median estimated life expectancy drops to only 18-24 months3,4 . In this challenging context, immune checkpoint inhibitors (ICIs) targeting the programmed cell death-1/ ligand-1 axis (PD-1/PD-L1) have demonstrated significant efficacy against both early-stage and metastatic disease. Pembrolizumab and atezolizumab – two ICIs targeting respectively PD1 and PD-L1 – have recently been approved in combination with chemotherapy for the treatment of PD-L1 positive (PD-L1+) metastatic TNBC3–6 . Shortly thereafter, the incorporation of pembrolizumab into multi-agent neoadjuvant chemotherapy was also approved for treating stage II-III early TNBC, irrespective of PD-L1 expression7,8 . Despite such therapeutic advances, the benefit of immunotherapy in TNBC is currently hampered by suboptimal patient selection. In the early setting, reliance on tumor stage alone fails to consider the biological heterogeneity of TNBC, leading to some patients facing unnecessary toxicities and others missing out on potentially beneficial treatments9–11 . In the metastatic setting, PD-L1 expression in the tumor sample is the sole validated biomarker for ICIs. However, its assessment suffers from substantial variability across assays and biopsy sites. Furthermore, even among metastatic patients with PD-L1+ tumors, 10-15% still experience disease progression within 3 months of initiating immunotherapy, and only about 50% survive beyond 2 years3,4 . Given these challenges, there is a need for novel biomarkers to better identify mechanisms of response and resistance to ICIs, improve patient selection, and rescue those at risk of rapid treatment failure. Ongoing research has identified several new immune biomarkers, which can be placed broadly into to categories: those tracking the immune activation in the tumor microenvironment (TME), such as tumor infiltrating lymphocytes (TILs)12; and those tracking the immunogenicity of tumors based on their neoantigen load, such as tumor mutational burden (TMB)13 . Preliminary data suggests that TME-specific biomarkers like TILs may add incremental predictive information beyond PD-L1 in patients treated immunotherapy. However, the use of TILs as a predictive biomarker has been limited by their low reliability in advanced disease stages. Like PD-L1, their frequency is lower in metastases compared to primary tumors overall; moreover, their expression varies deeply across metastatic sites, being highest in lymph nodes and lowest in bone, brain and liver14–17 . Because of such variability, both TILs and PD-L1 provide a suboptimal perspective of the tumor-host immune axis in the metastatic setting. In this context, the characterization of lymphocytes in the peripheral blood through flow cytometry could serve as a surrogate of TILs assessment. Similarly to TILs, circulating lymphocytes change their functional states during TNBC treatment and progression, indicating that their characterization could provide an indirect measure of changes occurring in the TME18,19 . Tumor mutational burden (TMB) and COMSIC-defined mutational signatures – namely homologous recombination repair deficiency (HRD), APOBEC, and mismatch repair deficiency (MRD) – represent another emerging biomarker for ICIs, being an indirect measure of the tumor neoantigen burden generated by somatic mutations20,21 . TMB and mutational signatures are tumor-specific biomarkers, thus having lower correlation with PD-L1 expression and the other TME-specific biomarkers22 . Compelling preliminary evidence suggest that TMB could predict higher response rates and longer survival in patients with metastatic breast cancer treated with single-agent immunotherapy13,23–26. Moreover, TMB and mutational signatures increase significantly from primary tumor to metastasis but, unlike PD-L1 and TILs, their assessment is more stable across metastatic sites13. To date, the interplay between TME-specific and tumor-specific biomarkers and its role in determining responsiveness to ICIs has not been elucidated yet. Building on this knowledge, we hypothesize that the joint evaluation of TMB/mutational signatures in the tumor tissue and that of lymphocytes in blood could improve the prediction of ICIs benefit in TNBC. Immunotherapy could elicit a more tumor-driven response when both TME inflammation and tumor neoantigen burden are present, while an inflamed TME alone could result in a less specific immune response. On these premises, we propose to assess the potential of circulating lymphocyte and TMB/mutational signatures to jointly predict ICIs benefit in TNBC, by utilizing tissue and blood samples from patients already enrolled in a prospective biomarker study at our Institution.