e20505 Background: Immune checkpoint inhibitors (ICIs) significantly improved the outcome in patients with lung cancer. However, heterogeneity in clinical benefit and absence of a reliable biomarker still represent an unmet need. Metabolic reprogramming plays a critical role in cancer behaviour. In particular, heme-ossigenase-1 (HO-1) expression on monocytes may have a role in treatment resistance. Methods: “ImmunoEGA” is a prospective, observational study in patients with advanced tumors, treated with ICIs at the ‘Maggiore della Carità’ Hospital in Novara, IT. Overall, 150 patients have been recruited from 2018 to 2023; of these, 122 were affected by lung cancer. In the present study we analysed monocytes subpopulations and HO-1 expression on blood samples collected from 34 patients at the following time points: baseline, at time of disease evaluation and at disease progression (PD). Monocytes were classified into three different phenotypes based on CD14 and CD16 expression: classical (CD14+CD16–), non-classical (CD14-CD16+) and intermediate (CD14+CD16+). The primary endpoint was treatment outcome in terms of progression free survival (PFS) and/or overall survival (OS). Results: At baseline, the frequency of classical monocytes was higher as compared to non-classical and intermediate subgroups, independently from type of response (mean percentage: 48,4% vs 7,8 vs 23,9%; p < 0.0001). Compared with baseline, we observed a relative increase in the frequency of classical (54,8 vs. 56,9) and intermediate (19,5 vs. 21,1) monocytes in patients with stable disease (SD) at the time of best response. Conversely, patients with PD showed an increase in non-classical (6,3 vs.14,1) and intermediate monocytes (26,3 vs. 30,9). In patients with partial response (PR) all monocyte subgroups increased proportionally (classical: 40,6 vs. 45,5; intermediate 24,2 vs. 25,9; non-classical 7,8 vs. 13,6). A higher baseline frequency of classical monocyte positively correlated with PFS (14 vs 6.5 m, p = 0,0171); these data were not observed for OS. No differences were detectable in HO-1 expression across monocytes subtypes at baseline and at the time of best response (SD or PR). Conversely, HO-1 expression showed an increasing trend across monocytes subpopulations at PD. Lower on classical (mPFS 15 vs 7 m, p = 0,0186) and intermediate monocytes (mPFS 15.5 vs 6.5 m p = 0,0057) at baseline correlated with better PFS. Patients with lower HO-1 expression on intermediate monocytes at baseline had a better outcome (OS 18 vs 11.5 m, p = 0.0378). Conclusions: In patients with lung cancer, higher frequency of classical monocytes and lower HO-1 expression in intermediate and classical monocytes predict better outcomes with ICI. The expression of HO-1 in classical and intermediate monocytes significantly increases at PD, suggesting its implication in promoting cancer progression. HO-1 expression could be used as a biomarker of immunotherapy benefit.

559 Background: In BC, pCR is associated with substantial survival benefit. Hence, escalated neoadjuvant treatments (NATs) are being increasingly employed to improve pCR rates, warranting biomarkers for better patients’ selection. T-cell exhaustion and senescence are dysfunctional states that coexist in patients with cancer, impairing antitumor immunity and sustaining a suppressive immune microenvironment. Here, we assessed the impact of such states on pCR across different BC subtypes and NATs. Methods: This is a prospective study on patients undergoing NAT for BC at the Maggiore Hospital of Novara, Italy. Blood was collected at baseline and after NAT. Circulating T lymphocytes were characterized with exhausted (CD28, PD1, LAG3) and senescent (CD57, KLRG-1) biomarkers utilizing flow cytometry. Neutrophiles/lymphocytes ratio (NLR) and lymphocytes/monocytes ratio (LMR) were also retrieved. Subpopulations with frequency <1% were normalized for total CD3+ cells. Results: From October 2020 to January 2024, 134 patients were enrolled and 115 completed NAT (85.8%). Fifty-one (38,1%) had adequate samples for analysis, 16 (31,4%) with hormone receptor positive HER2 negative BC (HR+HER2-), 15 (29,4%) with HER2+, and 20 (39,2%) with triple negative BC (TN). At baseline, patients with TNBC had the lowest NLR (p=0.018) and lowest senescent CD4+CD28- (p=0.042). Associations with pCR are shown in the Table. Overall, LMR associated with increased pCR (OR 1.34 [95% CIs 1.00-1.89]), and NLR with decreased pCR (OR 0.67 [0.39-1.04]). Both associations were driven by HR+ status (interaction p = 0.061 and 0.060). Senescent CD4+ predicted lower pCR in both HR-HER2+ (OR 0.10 [0.01-0.36]) and TNBC (OR 0.10 [0.01-0.5]). In TNBC, exhausted CD8+ predicted higher pCR (OR 1.78 [1.08-3.79]). When assessed separately in patients who received NAT with or without pembrolizumab, exhausted CD8+ remained predictor of pCR in chemotherapy + pembrolizumab (OR 1.84 [1.03-5.00]), but not with chemotherapy alone (p=0.203). The dynamics of paired pre-NAT and post-NAT TNBC samples revealed significant differences. Senescent CD4+ increased after chemotherapy, but not with chemotherapy + pembrolizumab (p=0.007). Pembrolizumab was associated with depletion of PD1-positive CD8+, while chemotherapy alone was not (p=0.020). Conclusions: The characterization of circulating T lymphocytes could help identify patients who benefit more in terms of pCR from specific NATs, including chemoimmunotherapy. [Table: see text]

BackgroundEvaluation of extracellular vescicles (EVs) has been identified as a potential predictive biomarker in patients treated with IO. In particular, Glypican-1 (GPC-1), was reported to be abnormally expressed in BC tissues and GPC-1+ EVs resulted significantly higher in BC patients than in healthy controls. Triple negative BC (TNBC) is characterized by poor prognosis, due to the lack of effective treatments; in this context, Immune Checkpoint Inhibitors (ICIs) are regarded as promising agents; however, their impact is not the same as in other cancers, such as melanoma and lung. With these premises, the aim of the IRIS project is to evaluate the prognostic and predictive role of tumor and leukocyte-derived EVs in patients with early and advanced TNBC, treated with ICIs.MethodsTwenty consecutive patients have been enrolled so far: 10 with TN MBC PD-L1+ treated with Ist line Nab-Paclitaxel + Atezolizumab and 10 with high-risk TN early BC treated with neo-adjuvant pembrolizumab (KEYNOTE-522 schedule). In all patients, 5 ml of peripheral blood in Na-Citrate were drawn at baseline. An optimized flow cytometry method was used to identify and subtype circulating EVs. EVs were identified as LCD+/phalloidin−events, falling in the scatter area with physical parameters lower than platelets. CD45, GPC-1, EpCAM and PD-L1 were chosen as markers for EVs characterization. Data acquisition and analyses were performed with flow cytometry (FACS Symphony) and FACSDiva v9.0.2. Volumetric count was applied to calculate EV concentrations. Statistical analysis was performed with GraphPad software, and a non-parametric t-test was utilized for EVs characterization, with a p-value < 0.05.ResultsOverall, a significantly lower concentration of circulating tumor-derived EVs (CD45-EpCAM+GPC-1+) was detected as compared to leukocyte-derived EVs (CD45+): Mean±SD = 95.16 ± 219.3 vs 1623 ± 1582; p<0.0001. Similar results were obtained when PD-L1 expression on EVs was considered, with a lower concentration of CD45-EpCAM+GCP-1+PD-L1+ compared to CD45+PD-L1: Mean ± SD: 12.63 ± 26.97 vs 656.8 ± 1012; p<0.0001. When taking into account the two patient subsets, ie MBC TN and early high risk TN, no significant difference was detected in concentration for circulating leukocytes- and tumor-derived PD-L1+ EVs: Mean ± SD cancer-derived EVs in MBC vs early BC: 6.4 ± 11.19 vs 19.56 ± 37.22, p=0.68; leukocytes-derived EVs in MBC vs early BC: 924.8 ± 1316 vs 359 ± 405, p=0.97. ConclusionThese preliminary data demonstrate that tumor-derived PD-L1+ EVs can be detected in advanced as well as in early TNBC and may represent a novel biomarker to assess response to ICIs. Patient recruitment and Flow Cytometry analyses are ongoing to confirm these data and to assess their predictive/prognostic role in TNBC. Citation Format: Veronica Martini, Arianna Stella, Rahma Ben Ayed, Carmen Branni, Andrea Tassone, Nazanin Keivan, Chiara Saggia, Francesca D'Avanzo, Valentina Rossi, Francesca Platini, Incoronata Romaniello, Elena Giacobino, Valentina Guarneri, Renzo Luciano Boldorini, Francesca Mercalli, Ivan Dodaro, Anna Gambaro, David James Pinato, Alessandra Gennari. Evaluation of extracellular vesicles (EVs) as potential predictors of response to immunotherapy (IO) in triple negative early and advanced breast cancer (BC): preliminary evidence from the IRIS project [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2174.

Emerging evidence highlights the potential prognostic relevance of circulating lipids in metastatic castration-resistant prostate cancer (mCRPC), with a proposed 3-lipid signature. This study aims to analyze the lipidomic profiles of individuals with mCRPC to identify lipid species that could serve as predictive indicators of prognosis and therapeutic response. Plasma samples were collected from mCRPC patients initiating first-line treatment (1 L) (n = 29) and those previously treated with at least two lines of therapy (> 2 L) (n = 19), including an androgen-receptor signaling inhibitor and a taxane. Employing an untargeted lipidomic approach, lipids were extracted from the plasma samples and subjected to analysis. A comprehensive identification and quantification of 789 plasma lipids was achieved. Notably, 75 species displayed significant dysregulation in > 2 L patients in comparison to the 1 L group. Among these, 63 species exhibited elevated levels, while 12 were reduced. Patients included in > 2 L cohort showed elevated levels of acylcarnitines (CAR), diacylglycerols (DG), phosphatidylethanolamines (PE), triacylglycerols (TG), and ceramides (Cer). Notably, some upregulated lipids, including CAR 14:0, CAR 24:1, Cer d18:1/16:0, Cer d18:1/18:0 (C18 Cer), Cer d18:2/18:0, Cer d18:1/24:1, and Cer d20:1/24:1, showed significant associations with overall survival (OS) in univariate models. Specifically, increased levels of C18 Cer remained significantly associated with poorer OS in the multivariate model, even after adjusting for treatment line and PSA levels (Hazard Ratio: 3.59 [95% Confidence Interval 1.51–8.52], p = 0.004). Employing quantitative mass spectrometry, our findings underscore the independent prognostic significance of C18 Cer in individuals with mCRPC. This discovery opens avenues for further studies within this field.