Akihito Tsuji
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (9)
Tumor response, OS, and PFS. A and B, A summary of ( A ) best reduction and ( B ) percent change from baseline in the sum of the diameters of target tumors in each patient with NSCLC treated with nivolumab. C and D, A summary of ( C ) best reduction and ( D ) percent change from baseline in the sum of the diameters of target tumors in each patient with gastric cancer (GC) treated with nivolumab. E and F, A summary of the probability of OS in patients with ( E ) NSCLC and ( F ) GC. G and H, A summary of the probability of PFS in patients with ( G ) NSCLC and ( H ) GC. The 95% CI was determined by the Brookmeyer–Crowley method with log–log transformation. Circles indicate censoring. CR, complete response; SD, stable disease.
Year:
2025
Evaluation of PD-1 + Treg cells and PD-1 + CD8 + T cells. A and B, The scatter plots that show PD-1 positivity (%) in CD8 + T cells and Treg cells in patients with ( A ) NSCLC and ( B ) gastric cancer (GC). Straight line indicates PD-1 positivity (%) in CD8 + T cells/PD-1 positivity (%) in Treg cells of ≥1.
Year:
2025
Purpose: It is challenging to identify the appropriate patients who benefit from anti–PD-1/PD-L1 monotherapy. For predicting effectiveness of anti–PD-1/PD-L1 monotherapy, this open-label phase II study (ONO-4538-88) evaluated the potential of the tumor-infiltrating lymphocyte (TIL) biomarker: the balance between cytotoxic T cells and regulatory T cells. Patients and Methods: Patients with advanced non–small cell lung cancer (NSCLC) or gastric cancer were screened between March 2021 and January 2022. Eligible patients who met the prespecified TIL biomarker criteria received nivolumab monotherapy. The primary endpoint was objective response rate (ORR). The secondary endpoints included overall survival and progression-free survival. Conventional biomarkers (tumor proportion score, combined positive score, tumor mutation burden, and microsatellite instability) were exploratorily analyzed, and safety was also assessed. Results: Thirty-seven patients with NSCLC and 127 patients with gastric cancer were eligible for TIL analysis: 6 (16.2%) and 15 patients (11.8%) met the TIL biomarker criteria, respectively; a part of them were assessed. For NSCLC and gastric cancer, the ORR was 80% (4/5 patients) and 36.4% (4/11 patients), respectively; all the five patients and 5/11 patients had a reduction in tumor size, respectively; the median overall survival was not reached and 25 months, respectively; and the median progression-free survival was not reached and 5.59 months, respectively. Treatment-related adverse events occurred in 13/19 patients overall: 5/6 patients for NSCLC and 8/13 patients for gastric cancer. Conclusions: Although the low positive rate of the TIL biomarker limits interpretation, the promising ORRs suggest signs of the TIL biomarker’s predictability for nivolumab monotherapy. Significance: In this phase II study, we examined the predictive utility of the TIL biomarker for nivolumab monotherapy. Although the positivity of the TIL biomarker was limited, the promising efficacy and safety profile in the TIL biomarker–positive patients may suggest the potential utility of the TIL biomarker.
Year:
2025
Purpose: It is challenging to identify the appropriate patients who benefit from anti–PD-1/PD-L1 monotherapy. For predicting effectiveness of anti–PD-1/PD-L1 monotherapy, this open-label phase II study (ONO-4538-88) evaluated the potential of the tumor-infiltrating lymphocyte (TIL) biomarker: the balance between cytotoxic T cells and regulatory T cells. Patients and Methods: Patients with advanced non–small cell lung cancer (NSCLC) or gastric cancer were screened between March 2021 and January 2022. Eligible patients who met the prespecified TIL biomarker criteria received nivolumab monotherapy. The primary endpoint was objective response rate (ORR). The secondary endpoints included overall survival and progression-free survival. Conventional biomarkers (tumor proportion score, combined positive score, tumor mutation burden, and microsatellite instability) were exploratorily analyzed, and safety was also assessed. Results: Thirty-seven patients with NSCLC and 127 patients with gastric cancer were eligible for TIL analysis: 6 (16.2%) and 15 patients (11.8%) met the TIL biomarker criteria, respectively; a part of them were assessed. For NSCLC and gastric cancer, the ORR was 80% (4/5 patients) and 36.4% (4/11 patients), respectively; all the five patients and 5/11 patients had a reduction in tumor size, respectively; the median overall survival was not reached and 25 months, respectively; and the median progression-free survival was not reached and 5.59 months, respectively. Treatment-related adverse events occurred in 13/19 patients overall: 5/6 patients for NSCLC and 8/13 patients for gastric cancer. Conclusions: Although the low positive rate of the TIL biomarker limits interpretation, the promising ORRs suggest signs of the TIL biomarker’s predictability for nivolumab monotherapy. Significance: In this phase II study, we examined the predictive utility of the TIL biomarker for nivolumab monotherapy. Although the positivity of the TIL biomarker was limited, the promising efficacy and safety profile in the TIL biomarker–positive patients may suggest the potential utility of the TIL biomarker.
Year:
2025
Details of microbiome according to cancer types and PPI use in cohort 1. A, α diversity based on the observed number of ASVs according to cancer type, which was based on the analysis of at least five specimens obtained from patients in cohort 1 ( n = 817). B, α diversity based on the observed number of KO indices according to cancer type, which was based on the analysis of at least five specimens obtained from patients in cohort 1 ( n = 817). C, Heatmaps depicting relative increases and decreases in bacterial species in patients with each cancer type compared with all patients as measured using 16S rRNA analysis and presented as ALDEx2 effect size. The distance between cancer types was used to determine effect sizes using the ALDEx2 algorithm, and hierarchical clustering was performed using the Ward method. Red and blue indicate bacterial species that were abundant or scarce, respectively, compared with the reference (all samples). The area boxed in red indicates increased levels of oral bacteria in patients with upper GI cancer. Only the genera with the top 1% of the largest absolute effect size are listed on the right-hand side of the figure. D, Violin plot of the proportion of oral bacteria in the feces of patients with upper GI cancer ( n = 296) vs. those with non-upper GI cancer ( n = 512). NET, neuroendocrine tumor. E, Differences in fecal bacterial flora between PPI users ( n = 199) and non-users ( n = 593) based on ALDEx2. An adjusted P value 0.2 were used to define abundant species. Species defined as oral bacteria are indicated by red boxes.
Year:
2025
α diversity of gut microbiome according to clinical features and cancer types. A, Diagram depicting the selection of the MONSTAR-SCREEN cohorts. ECOG, Eastern Cooperative Oncology Group; PS, performance status. B, Definition of oral bacteria. A total of 10 genera were defined as oral bacteria in this study from the data of the article by Atarashi and colleagues. C, Violin plot of α diversity based on ASV counts in feces according to concomitant medication use. A two-sided P value was calculated with the Welch t test. D, Violin plot of α diversity by ASV count in feces according to lifestyle habit. A two-sided P value was calculated with the Welch t test. For the number of defecations per day, the Kruskal–Wallis test was applied. LABB, lactic acid bacteria beverage.
Year:
2025
Relationship between ICI efficacy and bacterial flora/concomitant medication use. A, Kaplan–Meier plots of the PFS of patients treated with ICIs in the high ASV group and the low ASV group. The ASV cutoff was 206.0 (median). B, Forest plot analysis depicting the association between clinical features and PFS in patients with ICIs. Dots and error bars represent point estimates of the HR and 95% CIs. Antibio, antibiotics; MSI-H, MSI high; MSS, microsatellite stable; TMB-H, TMB high; TMB-L, TMB low; Probio, probiotics. C, Forest plot of the impact of concomitant medication use on the PFS of patients treated with ICIs. “Before” was defined as use within 3 months before the initiation of ICI treatment. “During” was defined as use during ICI treatment. D, Kaplan–Meier plots of PFS of patients treated with ICIs based on the proportion of oral bacteria in feces. “Oral bacteria high” was defined as patients with the proportion of oral bacteria above median. E, Kaplan–Meier plots of the PFS of patients treated with ICIs based on PPI use and the abundance of oral bacteria in the feces. “Oral bacteria high” was defined as patients with the proportion of oral bacteria above median. F, Kaplan–Meier plots of the PFS according to the proportion of oral bacteria in patients with PPI use after propensity score matching. G, Kaplan–Meier plots of the PFS according to the proportion of oral bacteria in patients without PPI use after propensity score matching. mPFS, median PFS; Ref, reference.
Year:
2025
Although the gut microbiome is associated with cancer development and progression, little is known about the effects of the gut microbiome landscape and the efficacy of immune checkpoint inhibitors (ICI) across cancer types. We investigated the association between the microbiome, clinical features, and ICI efficacy across cancer types in a large nationwide screening project for solid tumors. Among 2,180 patients with advanced solid tumors enrolled in the SCRUM-Japan MONSTAR-SCREEN between October 2019 and September 2021, in the chemotherapy-naïve cohort ( n = 817), a high prevalence of oral bacteria was observed in patients using proton pump inhibitors (PPI) and those with upper gastrointestinal cancers, particularly postoperative patients with gastric or pancreatic cancer. Among patients treated with ICIs ( n = 333), a high abundance of sequence variants in the gut microbiome was not significantly associated with ICI efficacy across cancer types (HR = 0.94; 95% confidence interval, 0.73–1.21). However, high oral bacteria in feces significantly correlated with a shorter progression-free survival compared with low oral bacteria (median, 4.34 vs. 6.97 months; HR = 1.38; 95% confidence interval, 1.07–1.78). Notably, in patients using PPIs, a higher proportion of oral bacteria influenced progression-free survival outcomes of ICI treatment (median, 3.15 vs. 2.04 months; P = 0.08), unlike in PPI nonusers (median, 7.13 vs. 5.55 months; P = 0.74). This study of the gut microbiome has unveiled significant insights into its landscape and potential impact on ICI efficacy. It highlights that the abundance of oral bacteria in feces may play a critical role in diminishing ICI efficacy among patients using PPIs. Significance: As part of the MONSTAR-SCREEN, a prospective nationwide project for patients with solid tumors, we found that although gut microbiome diversity does not consistently predict ICI efficacy across cancer types, a high level of oral bacteria in the gut is linked to reduced ICI effectiveness, especially in patients using PPIs. These findings highlight the potential clinical impact of microbiome variations on cancer treatment outcomes.
Year:
2025
Background: This randomized controlled trial is aimed at clarifying the benefit of an extract of ninjin’yoeito (NYT, traditional Japanese herbal medicine, Kampo) when combined with capecitabine plus oxaliplatin (CapeOX) plus bevacizumab (Bmab) in 150 patients with metastatic or recurrent colorectal cancer with special reference to nutritional status. Methods/Design: Patients with unresectable, metastatic, or recurrent colorectal cancer who plan to receive CapeOX + Bmab will be recruited. Of these patients, those who provide written informed consent will be randomly allocated to the NYT group (NYT + CapeOX + Bmab) or the control group (CapeOX + Bmab). The co-primary endpoints will be nutritional status, evaluated with CONtrolling NUTritional Status scores, and severe non-hematological adverse events, assessed using the Common Terminology Criteria for Adverse Events v5.0. The secondary endpoints will include the interaction between NYT and capecitabine; quality of life evaluated using the European Organization for Research and Treatment of Cancer-QLQ-C30; relative dose intensity of oxaliplatin, capecitabine, and bevacizumab; and plasma levels of phosphorylated neurofilament heavy chain. The two groups will include 75 patients each, and the patient recruitment period is January 1 st , 2020 to December 31 st , 2022. Discussion: This trial evaluates the efficacy and safety of NYT in the maintenance of a better nutritional status and fewer adverse events during the standard chemotherapeutic regimen of CapeOX + Bmab administered to patients with advanced or recurrent colorectal cancer. The potential implications include that NYT could be a strong supportive therapy in patients with colorectal cancer receiving chemotherapy. Trial registration number: This study was registered in the Japan Registry of Clinical Trials (registration number: jRCTs041190071) on October 2, 2019.
Year:
2022
Collaborators (6)
Taroh Satoh
Professor
Osaka University Hospital
Tadamichi Denda
Head
Kanagawa Cancer Center
Hiroyuki Takeda
Assistant professor
St. Marianna University
Tatsuyuki Chiyoda
Keio University School of Medicine
Yoshito Komatsu
Hokkaido University Hospital
Eiji Oki
Kyushu University
Make your next move.
Prepare your application
Review drafts before you send them.