Yasuo Miyoshi

Blood-based metabolomic profiling has been widely investigated for breast cancer (BC) detection; however, clinical implementation remains limited due to variability in sample handling, analytical reproducibility, and overfitting during statistical analysis. We established a plasma GC/MS metabolomics workflow for discriminating BC from healthy controls (HC) using conventional machine-learning algorithms. Plasma samples (n = 360; BC = 180, HC = 180) were collected prospectively under standardized preanalytical conditions before surgery and the initiation of systematic anticancer therapy and analyzed using a quantitative GC/MS platform with automated derivatization. Feature selection and model development were conducted using three machine-learning (ML) algorithms (Lasso logistic regression (LR), random forest classifier (RFC), and support vector machine (SVM)). A total of 45 metabolite candidate biomarkers were identified, and the optimal number of metabolite features for each algorithm was estimated by a recursive feature elimination (RFE)-based strategy. The best-performing models achieved area under the ROC curve values (AUC) of 0.910 (LR), 0.893 (RFC), and 0.843 (SVM). We selected prioritizing candidate biomarkers consistently expressed across the multi-algorithm pipeline. A bagging ensemble model improved stability (AUC = 0.911) and reduced false-positive predictions in the independent HC dataset. In addition, model stability with respect to false-positive predictions was assessed using an independent HC cohort (n = 15) that was collected at a separate institution. These results indicate that a plasma metabolomics workflow combined with conventional multi-algorithm ML, algorithm-specific feature selection, and independent assessment provides stable discrimination between BC and HC in a moderately sized cohort.

Table S1. Representativeness of the Study Population

Table S3. Change from Baseline to Week 18 in Patient-Reported Outcomes

Figure S2. Kaplan-Meier analysis of TTD

Table S2. Summary of Drug Exposure During the Post-induction Period in the All-Patients-As-Treated Population

Figure S1. Change from baseline to week 18 in PROs

Background In the EMBRACE study, eribulin (ERI) monotherapy improved the overall survival (OS) of patients with HER2-negative advanced breast cancer (HER2-negative ABC). A post hoc analysis identified the baseline absolute lymphocyte count (ALC) as a predictive marker in the ERI arm; nevertheless, factors affecting post-ERI had not been well explored. Patients and methods We retrospectively analyzed clinical data of 370 patients receiving ERI for HER2-negative ABC between July 2011 and June 2024 across 3 institutions. Clinical data, including ALC, were extracted. Statistical analyses included the log-rank test and Cox hazard model. The OS was defined as survival from ERI initiation (OS1) and from ERI termination (OS2). Results Identical to the previous report, the OS1 was significantly improved in patients with ALC ≥ 1000/μL at ERI initiation and with ER-positive. Regarding the OS2, we identified independent factors which significantly improve the OS2 as follows: ALC ≥ 1000/μL at the termination of ERI, ≥ 120 days time-to-treatment-discontinuation of ERI, and first-line use of ERI. Furthermore, ALC was maintained during ERI therapy, whereas it significantly decreased in other regimens (P < 0.001). Conclusion An ALC ≥ 1000/μL at ERI completion was associated with an improved post-ERI OS (OS2). It is suggested that not only the factors at the initiation of ERI, but also the immunological status at the end of ERI, may have prognostic value after ERI.

The phase 3 KEYNOTE‐522 study in high‐risk early‐stage triple‐negative breast cancer (TNBC) showed significantly improved efficacy outcomes with neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab versus neoadjuvant chemotherapy alone. We present findings from the KEYNOTE‐522 Japan subgroup. Eligible participants (aged ≥ 18 years) with untreated locally advanced TNBC (stage T1c N1–2 or T2–4 N0–2) were randomized 2:1 to neoadjuvant pembrolizumab 200 mg or placebo plus chemotherapy every 3 weeks for 8 cycles followed by surgery and adjuvant pembrolizumab or placebo for ≤ 9 cycles. Primary endpoints were pathologic complete response (pCR; ypT0/Tis ypN0) at the time of surgery and event‐free survival (EFS). Of 76 participants enrolled in Japan, 45 were randomized to the pembrolizumab arm and 31 to the placebo arm. Median time from randomization to data cutoff (March 22, 2024) was 76.3 months. Twenty‐four participants (53%) in the pembrolizumab arm and 15 (48%) in the placebo arm achieved pCR (between‐treatment arm difference, 4.9%; 95% CI, −17.6% to 27.1%); findings were similar regardless of PD‐L1 expression. Rates of EFS at 60 months were 84% and 73%, respectively (HR, 0.54; 95% CI, 0.20–1.50). Grade 3 or 4 treatment‐related AEs occurred in 37 of 45 participants (82%) treated with pembrolizumab and 23 of 30 participants (77%) treated with placebo; there were no grade 5 AEs. In conclusion, neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab showed improved efficacy outcomes and manageable safety versus neoadjuvant chemotherapy alone in Japanese participants, supporting the use of this regimen in Japanese patients with high‐risk early‐stage TNBC. Trial Registration: The study ( ClinicalTrials.gov , NCT03036488) was conducted in compliance with local and/or national regulations and International Council for Harmonization Good Clinical Practice guidelines and in accordance with the ethical principles originating from the Declaration of Helsinki

Purpose: Pembrolizumab plus olaparib versus pembrolizumab plus chemotherapy was evaluated as postinduction therapy for patients with PD-L1–unselected locally recurrent inoperable/metastatic triple-negative breast cancer (TNBC) who derived clinical benefit from first-line pembrolizumab plus platinum-based chemotherapy induction therapy. Patients and Methods: In this phase 2 study (NCT04191135), participants with previously untreated locally recurrent inoperable/metastatic TNBC received first-line pembrolizumab 200 mg once every 3 weeks plus platinum-based chemotherapy. Participants with complete/partial response or stable disease (per RECIST v1.1) were randomly assigned 1:1 to pembrolizumab 200 mg once every 3 weeks plus olaparib 300 mg twice a day or pembrolizumab plus chemotherapy. Progression-free survival (PFS) and overall survival (OS) were primary endpoints. Results: Of 460 participants receiving induction treatment, 271 were randomly assigned to postinduction therapy. The median PFS was 5.5 months in the pembrolizumab plus olaparib group versus 5.6 months in the pembrolizumab plus chemotherapy group [hazard ratio (HR), 0.98; 95% confidence interval (CI), 0.72–1.33; P = 0.4556]. The median OS was 25.1 versus 23.4 months, respectively (HR, 0.95; 95% CI, 0.64–1.40). In participants with tumor BRCA1/BRCA2 mutations (tBRCAm), HRs for PFS (HR, 0.70; 95% CI, 0.33–1.48) and OS (0.81; 95% CI, 0.28–2.37) favored pembrolizumab plus olaparib. Treatment-related adverse events occurred in 84.4% and 96.2% of participants receiving pembrolizumab plus olaparib and pembrolizumab plus chemotherapy, respectively. Conclusions: Although the primary endpoint was not met, postinduction pembrolizumab plus olaparib therapy resulted in similar PFS and OS compared with pembrolizumab plus chemotherapy in this setting. The positive trend for PFS and OS in participants with tBRCAm suggests a potential nonchemotherapy strategy for maintaining clinical benefit attained with first-line pembrolizumab plus chemotherapy induction treatment. No new safety signals were identified.

Pharmacokinetics of pro-FTY and its effect on lymphocytes and tumor suppression in mice. A, BALB/c and control (Ct) mice implanted with the 4T1 breast cancer cell line were treated three times with FTY720 and pro-FTY, and tumor volumes were measured. B, pStat3 expression in representative tumors from BALB/c mice administered the 4T1 breast cancer cell line after treatment with FTY720 or pro-FTY. Scale bars, 25 μm. C and D, Pro-FTY, FTY720, and FTY-P levels in tumor and plasma samples from BALB/c mice were quantified by mass spectrometry on d1 (C) and d3 (D) after the last pro-FTY drug administration. E–N, Blood counts were performed the day after the last dose of drug was administered: white blood cells (WBC; E), red blood cells (RBC; F), hemoglobin (HGB; G), hematocrit (HCT; H), platelets (PLT; I), lymphocytes (LYM; J), neutrophils (NEU; K), eosinophils (EOS; L), basophils (BAS; M), and monocytes (MON; N). O, CD4+ and CD8+ lymphocyte abundances were measured by flow cytometry. P, Serum IL-6 levels were determined by performing ELISAs. Q and R, With the PDX mouse model, the effect of pro-FTY was evaluated by measuring tumor volumes (Q) over time and tumor weights (R) when the mice were euthanized. *, P < 0.05 vs. control. S, Body weights of the mice in each group during the treatment period.