In this phase 3 trial, penpulimab combined with chemotherapy was assessed against a regimen of placebo plus chemotherapy for the first-line treatment of recurrent or metastatic nasopharyngeal carcinoma (R/M NPC). 291 patients were randomised and allocated in a 1:1 ratio in order to receive penpulimab (n = 144; 200 mg) or placebo (n = 147; 200 mg), plus chemotherapy (cisplatin/carboplatin and gemcitabine) every 3 weeks. Patients followed by maintenance therapy with penpulimab or placebo after 6 cycles. The primary endpoint of this study was progression-free survival (PFS) according to RECIST v1.1, and a significantly longer median PFS in the penpulimab arm versus the placebo arm (9.63 versus 7.00 months; hazard ratio 0.45, 95% CI: 0.33-0.62, P < 0.0001) was demonstrated in this prespecified interim analysis. The key secondary endpoint was the overall survival (OS). However, the OS data were still immature, and the median OS was not achieved (hazard ratio, 0.94; 95% CI: 0.63-1.40). The occurrence of treatment-related adverse events (grade ≥ 3) was 89.0% and 85.9% in two arms, with the most common being reduced the quantity of neutrophil (56.2% vs. 62.0%), reduced the quantity of white blood cell (54.1% vs. 54.9%), and anemia (45.2% vs. 38.7%). In the penpulimab arm, 6 patients (4.1%) experienced immune-related adverse events (grade ≥ 3). Adding penpulimab to chemotherapy led to a notable enhancement in PFS for the first-line R/M NPC treatment, alongside a safety profile that was both manageable and tolerable. ClinicalTrials.gov identifier NCT04974398.

Patients with EGFR-positive non-small cell lung cancer (NSCLC) do not respond as well to immunotherapy as patients without an EGFR mutation. Nevertheless, recent studies indicate that subsequent immunotherapy following EGFR-TKI treatment may be effective in EGFR-positive NSCLC. Given the critical role of the tumor immune microenvironment (TIME) in determining immunotherapy efficacy, understanding the influence of EGFR-TKI on the TIME in NSCLC remains an important yet unresolved area of research. Single-cell RNA sequencing (scRNA-seq) was performed to investigate the TIME of NSCLC patients. Post-treatment samples were analyzed from three EGFR-positive NSCLC patients treated with neoadjuvant EGFR-TKI (Neo-TKI) and two EGFR-negative NSCLC patients treated with neoadjuvant immunotherapy (Neo-IO). The scRNA-seq datasets of seven EGFR-positive NSCLC patients undergoing surgery only (TN-TKI) from the National Genomics Data Center (CRA001477, CRA001963) and six EGFR-negative NSCLC patients received surgery only (TN-IO) from the Gene Expression Omnibus (GSE131907) were analyzed to assess the TIME in treatment-naïve conditions. A total of ∼173, 000 single cells were sequenced, revealing 6 subpopulations. Compared to TN-TKI samples, Neo-TKI samples showed a higher proportion of T and B cells, whereas the proportion of myeloid cells significantly decreased. Further analysis of T cell composition in Neo-TKI samples revealed a trend towards a higher prevalence of GZMK+CD8+T cells and enrichment of positive regulation of T cell activation pathway. The expression of TLS-associated genes in CD4+ T cells was significantly elevated, along with the upregulation of immune-related pathways, such as B cell activation signaling and IL-17-related signaling, underscoring the immunomodulatory role of EGFR-TKI. Regarding B cell composition, Neo-TKI samples showed a higher proportion of germinal center B cells with an enrichment in positive regulation of B cell signaling pathway compared to TN-TKI samples. Notably, a shift in tumor-associated macrophages from an M2 phenotype to a neutral phenotype was observed in Neo-TKI samples, supporting the notion that EGFR-TKI impacts the immune landscape. Further, we conducted an exploratory analysis comparing the effects of immunotherapy and EGFR-TKI treatment on the TIME. Distinct differences were observed in cellular subpopulation distribution, enriched genes, and activated pathways. Overall, EGFR-TKI therapy appears to have a more pronounced effect on the TIME of patients with NSCLC. EGFR-TKI therapy alleviated the immunosuppressive TIME of EGFR-positive NSCLC, providing valuable insights for optimizing subsequent immunotherapy strategies in EGFR-positive NSCLC. Fang Wu, Dan Fan, Yingzhe Zhang, Wenliang Liu, Fenglei Yu, Wen Hu, Yue Zeng, Yurong Peng, Chunhong Hu, Xianling Liu, Fang Ma, Chao Deng, Zhenhua Qiu, Shijiao Yao, Shuxing Wang, Xiaoqiao Lin, Aoxi Liang, Xinzhu Lin, Qian Guan, Han Yang, Yiyang Li. Neoadjuvant EGFR-TKI therapy reshapes tumor immune microenvironment in resectable EGFR-positive non-small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3278.

PurposesThe presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients.MethodsThis retrospective study included a bi-parametric MRI of 1567 consecutive male patients from six centers (Centers 1-6) between Jan 2015 and Dec 2020. Deep learning models with double channel attention modules based on MRI (AttenNet) for predicting PCa and csPCa were constructed separately. Each model was first pretrained using 1144 PI-RADS 1-2 and 4-5 images and then retrained using 238 PI-RADS 3 images from three training centers (centers 1-3), and tested using 185 PI-RADS 3 images from the other three testing centers (centers 4-6).ResultsOur AttenNet models achieved excellent prediction performances in testing cohort of center 4-6 with the area under the receiver operating characteristic curves (AUC) of 0.795 (95% CI: [0.700, 0.891]), 0.963 (95% CI: [0.915, 1]) and 0.922 (95% CI: [0.810, 1]) in predicting PCa, and the corresponding AUCs were 0.827 (95% CI: [0.703, 0.952]) and 0.926 (95% CI: [0.846, 1]) in predicting csPCa in testing cohort of center 4 and center 5. In particular, 71.1% to 92.2% of non-csPCa patients were identified by our model in three testing cohorts, who can spare from invasive biopsy or RP procedure.ConclusionsOur model offers a noninvasive screening clinical tool to re-stratify risks in PI-RADS 3 patients, thereby reducing unnecessary invasive biopsies and improving the effectiveness of biopsies.Critical relevance statementThe deep learning model with MRI can help to screen out csPCa in PI-RADS category 3.Key PointsAttenNet models included channel attention and soft attention modules.71.1-92.2% of non-csPCa patients were identified by the AttenNet model.The AttenNet models can be a screen clinical tool to re-stratify risks in PI-RADS 3 patients.

BACKGROUND:GEMSTONE-302 was a phase 3 trial in patients with treatment-naive metastatic squamous or non-squamous non-small-cell lung cancer (NSCLC), showed significant improvement in progression-free survival and overall survival with sugemalimab, a PD-L1 inhibitor, plus chemotherapy versus placebo plus chemotherapy. We report the 4-year outcomes from this study. METHODS:This randomised, double-blind, phase 3 trial was conducted across 35 hospitals and academic research centres in China. Eligible patients were aged 18-75 years; had treatment-naive, histologically or cytologically confirmed stage IV NSCLC, irrespective of PD-L1 expression levels; and had an Eastern Cooperative Oncology Group performance status of 0 or 1. Patients were randomised (2:1) by investigators using an interactive web response or voice response system via permuted blocks (block sizes of three or six, randomised within each stratum). Patients received histology-specific platinum-based chemotherapy combined with either sugemalimab (1200 mg; sugemalimab group) or placebo (placebo group) for up to four cycles, followed by for up to 35 cycles of maintenance therapy with sugemalimab alone for patients with squamous NSCLC and sugemalimab plus pemetrexed for patients with non-squamous NSCLC in the sugemalimab group, or placebo for patients with squamous NSCLC and placebo plus pemetrexed for patients with non-squamous NSCLC in the placebo group, administered intravenously. Treatment beyond 35 cycles was permitted at the investigator's discretion. The primary endpoint was investigator-assessed progression-free survival in the intention-to-treat population. Here, we report post-hoc 4-year efficacy and safety outcomes from GEMSTONE-302. This study is registered with ClinicalTrials.gov (NCT03789604) and concluded on May 15, 2023, with all patients discontinued. FINDINGS:Between December 13, 2018, and May 15, 2020, 846 patients were assessed for eligibility. 479 patients were randomly assigned into the sugemalimab group (n=320) and placebo group (n=159). 254 (79%) patients were men and 66 (21%) were women in the sugemalimab group and 129 (81%) were men and 30 (19%) were women in the placebo group. All patients were Asian. As of the data cutoff on May 15, 2023, median follow-up durations were 43·5 months (IQR 41·2-46·9) in the sugemalimab group and 43·0 months (40·7-44·8) in the placebo group; median treatment durations were 7·2 months (4·2-18·8) with sugemalimab and 4·6 months (2·8-6·9) with placebo. Median progression-free survival was 9·0 months (95% CI 7·4-10·9) in the sugemalimab group versus 4·9 months (4·8-5·2) in the placebo group (hazard ratio [HR] 0·49 [95% CI 0·39-0·60]). Median overall survival was 25·2 months (20·1-30·2) in the sugemalimab group versus 16·9 months (12·8-20·7) in the placebo group (HR 0·68 [0·54-0·85]). The 4-year overall survival rates were 32·1% (95% CI 26·7-37·6) in the sugemalimab group versus 17·3% (11·1-24·7) in the placebo group. The most common grade 3-4 treatment related adverse events were decreased neutrophil count (105 [33%] with sugemalimab vs 52 [33%] with placebo), decreased white blood cell count (48 [15%] vs 27 [17%]), anaemia (44 [14%] vs 18 [11%]), and decreased platelet count (35 [11%] vs 15 [9%]). Treatment-related serious adverse events occurred in 82 (26%) patients with sugemalimab and 31 (20%) with placebo. No additional treatment-related deaths occurred since the previous overall survival interim analysis. No new safety signals were identified. INTERPRETATION:Sugemalimab with chemotherapy showed a superior long-term overall survival benefit compared with placebo with chemotherapy, as a first-line treatment for patients with NSCLC with no known sensitising EGFR, ALK, ROS1, or RET genomic alterations. These results underscore the efficacy of sugemalimab plus platinum-based chemotherapy as a standard first-line treatment option for both squamous and non-squamous metastatic NSCLC while maintaining a manageable safety profile. FUNDING:CStone Pharmaceuticals. TRANSLATION:For the Chinese translation of the abstract see Supplementary Materials section.

OBJECTIVES:Neoadjuvant immunotherapy has demonstrated favorable efficacy in patients with resectable non-small cell lung cancer (NSCLC). However, its clinical application remains limited by the lack of reliable and non-invasive biomarkers. Although existing histological biomarkers such as programmed death-ligand 1 (PD-L1) and tumor mutation burden (TMB) can be used for reference, they rely on invasive sampling and are susceptible to tumor heterogeneity. This study evaluated a series of peripheral blood inflammation-related indicators, including neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), systemic immune-inflammation index (SII), and interleukin-6 (IL-6), to explore their potential as non-invasive predictive and prognostic biomarkers for NSCLC. Furthermore, a prediction model based on the above indicators was constructed to provide a practical and feasible tool for optimizing individualized clinical management in patients with resectable NSCLC. METHODS:A retrospective analysis was conducted on 144 patients with resectable (stage IB-IIIB) NSCLC who underwent surgery after receiving neoadjuvant immunotherapy combined with chemotherapy at the Second Xiangya Hospital, Central South University, between 2019 and 2022. Peripheral blood-related indicators at baseline and before surgery were collected. Clinical data that might influence treatment efficacy were also recorded, including age, sex, body mass index, smoking history, pathological type, clinical stage, and use of immune checkpoint inhibitors. The relationships between peripheral blood inflammatory indicators (NLR, LMR, PLR, SII, and IL-6) and objective response rate (ORR), pathological complete response (pCR), major pathological response (MPR), and disease-free survival (DFS) were analyzed. Receiver operating characteristic (ROC) curves were used to determine optimal cutoff values for each indicator. A prediction model for the efficacy of neoadjuvant immunotherapy in NSCLC was constructed using least absolute shrinkage and selection operator (LASSO) regression combined with a multivariate Cox proportional hazards model. RESULTS:The median age of included patients was 58 years, and 91.0% (131/144) were male. Among pathological types, squamous cell carcinoma accounted for 74.3% (107/144), adenocarcinoma for 22.9% (33/144), and other types for 4 cases. The overall ORR, pCR, and MPR rates were 69.2%, 42.4%, and 61.5%, respectively. Univariate analysis showed that patients with squamous cell carcinoma had significantly higher ORR (P=0.007), pCR (P=0.027), and MPR (P=0.019). Lower baseline LMR was associated with a higher ORR. Elevated baseline PLR was significantly associated with pCR (P=0.014) and MPR (P=0.043). Increased baseline SII (P=0.015) and IL-6 (P=0.043) were associated with higher MPR rates. Multivariate analysis showed that squamous cell carcinoma was an independent predictor of MPR (OR=7.34, 95% CI 1.02 to 52.51, P=0.047), and lower baseline LMR was an independent predictor of ORR in NSCLC (OR=0.21, 95% CI 0.05 to 0.92, cutoff value 3.12; P=0.04). Further survival analysis indicated that low baseline NLR (HR=0.363, P=0.014), low preoperative LMR (HR=0.260, P=0.018), and high preoperative SII (HR=0.278, P=0.003) significantly reduced the risk of DFS. A prediction model including 9 factors (age, pathological type, baseline NLR, baseline neutrophils, baseline IL-6, baseline monocytes, preoperative lymphocytes, preoperative SII, and preoperative LMR) was established for predicting the efficacy of neoadjuvant immunotherapy in NSCLC, with an AUC of 0.818. CONCLUSIONS:Neoadjuvant immunotherapy demonstrates favorable clinical efficacy in patients with NSCLC, particularly in those with squamous cell carcinoma. Meanwhile, peripheral blood inflammation-related indicators may serve as important biomarkers for predicting the efficacy and prognosis of neoadjuvant immunotherapy in NSCLC.

Neoadjuvant chemoimmunotherapy (NCIO) is a new and effective treatment for cancer, but its efficacy in treating certain patients is unclear. We previously found that comorbidity was an independent factor associated with the pathological complete response (pCR) of non-small cell lung cancer (NSCLC) to NCIO. However, we did not address which comorbidities or comorbidity indices were associated with pCR. The present study retrospectively collected the data for NSCLC patients who underwent NCIO after surgery at The Second Xiangya Hospital of Central South University (Hunan, China) between January 2019 and July 2022. The associations between comorbidities/comorbidity indices and pCR rates/clinicopathological factors were analyzed. In total, 101 eligible patients with stage IIB-IIIC NSCLC were enrolled. Comorbid hypertension [odds ratio (OR)=0.321(0.110-0.937)], vascular disease [OR=0.275 (0.111-0.677)] and cardiovascular disease [OR=0.272 (0.114-0.646)] were all significantly associated with pCR (all P<0.05). The comorbidity indices Cumulative Illness Rating Scale-Geriatric (CIRS-G) ≥2 [OR=0.360 (0.154-0.840)], CIRS-G ≥3 [OR=0.404 (0.179-0.912)], CIRS-G ≥4 [OR=0.293 (0.105-0.817)] and Adult Comorbidity Evaluation-27 (ACE-27) ≥2 [OR=0.427 (0.192-0.950)] were all significantly associated with pCR (all P<0.05). Cardiovascular disease was the only independent risk factor for pCR [adjusted OR=0.272 (0.114-0.646); P=0.003] according to multivariate logistic analysis. In conclusion, cardiovascular comorbidities and the CIRS-G and ACE-27 indices were associated with the effectiveness of NCIO and clinicopathological factors. These results could help to screen for the most suitable NSCLC patients for NCIO.

Studies have shown that the combination of PD-1 inhibitors with chemotherapy exhibits promising efficacy as a first-line treatment for Asian patients with recurrent or metastatic nasopharyngeal carcinoma (R/M NPC). This presentation reports the results of a global phase 3 clinical trial with ethnically diverse patients treated with penpulimab plus chemotherapy vs. placebo plus chemotherapy as the first-line therapy for R/M NPC (NCT04974398). AK105-304 trial was conducted across 46 sites worldwide. Participants aged 18-75 years with previously non-systemically treated R/M NPC, stratified by disease stages (de novo metastases vs. recurrent), ECOG (0 vs. 1), liver metastasis (present vs. absent), were randomized (1:1) to receive penpulimab or placebo (200mg, Day1) in combination with gemcitabine (1000mg/m2, Day 1 and 8) and cisplatin (80mg/m2, Day1) or carboplatin (AUC5, Day1) every 3 weeks (Q3W) for up to 6 cycles, followed by maintenance therapy with penpulimab or placebo (200mg, Q3W). Placebo-arm patients were allowed to crossover to receive penpulimab monotherapy (200 mg, Q3W) upon confirmed disease progression by blinded independent center review (BICR). Primary endpoint was PFS assessed by BICR, and key secondary endpoint was OS. Other secondary endpoints included ORR, DoR and safety. 291 patients were randomized to penpulimab arm (n=144) or placebo arm (n=147). Baseline characteristics were generally balanced between treatment arms. By April 29, 2024, median follow-up time was 19.1 months. Per BICR assessment, median PFS was 9.6 months (95% CI: 7.1, 12.5) and 7.0 months (95% CI: 6.9, 7.3), respectively, for penpulimab+chemo and placebo+chemo (HR=0.45, 95% CI: 0.33, 0.62, two-sided P < 0.0001). Confirmed ORR was 68.1% vs. 63.9%, and median DoR was 9.8 months (95% CI: 7.0, 17.5) vs. 5.7 months (95% CI: 5.5, 6.7) (HR=0.4, 95%CI: 0.27, 0.59). OS was not mature, with 48 deaths in penpulimab arm and 49 in placebo arm (HR=0.94, 95% CI: 0.63, 1.40). After adjusting for crossover of patients from placebo to penpulimab upon PD, OS benefit became more evident, with HR of 0.62 (0.41, 0.94) by Rank Preserving Structural Failure Time (RPSFT) model, 0.75 (0.43, 1.30) by Inverse Probability of Censoring Weighting (IPCW) method, and 0.78 (0.52, 1.17) by two-stage Accelerated Failure Time (AFT) model. The incidence of Grade ≥3 TRAEs was 89.0% vs. 85.9%; SAEs was 50.7% vs. 48.6%; irAEs was 30.8% vs. 8.5%; Grade ≥3 irAEs was 4.1% vs. 0. Penpulimab combined with gemcitabine and cisplatin or carboplatin demonstrated statistically significant and clinically meaningful benefit with a manageable safety profile, and provides a new beneficial treatment option in the first-line treatment for R/M NPC patients globally. Chaosu Hu, Xiaozhong Chen, Tingting Xu, Shuang Huang, Feng Liu, Song Qu, Lisha Chen, Ping Zhou, Shenhong Qu, Xiaohong Ai, Yong Chen, Meilian Liu, Rensheng Wang, Kelvin KW Chan, Peng Zhang, Chunhong Hu, Jiyu Wen, Jian Zhang, Qin Lin, Xiaojiang Li, Kangsheng Gu, Li Xiang, Dongxia Wang, Jingao Li, Daren Lin, Desheng Hu, Jianwu Ding, Siyang Wang, Xiaoming Huang, Lin Wang, Feng Jin, David G. Pfister, Milena Perez Mak, Pedro Rafael Martins de Marchi, Yi Jiang, Haihua Yang, Xiaoye Hu, Tianrun Liu, Dehua Wu, Aditya Shreenivas, Thiago Bueno de Oliveira, Carlos Eduardo Baston Silva, Gustavo Vasconcelos Alves, Xianming Li, Zhifang Yao, Dongmei Lu, Mingxiu Hu, Zhongmin Maxwell Wang, Baiyong Li, Michelle Xia. Penpulimab versus placebo in combination with chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal carcinoma: A global, multicenter, randomized, double-blind, phase 3 trial (AK105-304) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr CT011.

Zanu, a potent, specific next-generation BTK inhibitor with a favorable safety profile, is approved in over 70 countries globally for the treatment of multiple B-cell malignancies. Pts with R/R DLBCL face a poor prognosis, and the CD79B mutation is an unfavorable prognostic factor for survival, especially following immunochemotherapy. Zanu has demonstrated modest antitumor activity in R/R non-germinal center B-cell (GCB) DLBCL in clinical trials, and retrospective biomarker analyses have indicated that pts with mutated CD79Bshow an enhanced response to zanutreatment(Yang et al Blood Adv 2022; Liu et al Leuk Lymphoma 2024). Currently, there is no established standard of care for CD79B-mutated R/R DLBCL, highlighting an unmet clinical need. Thus, this study assessed antitumor activity and related biomarkers of zanu for CD79B-mutated R/R DLBCL. In this phase 2 trial (NCT05068440), pts with centrally confirmed CD79B-mutated R/R DLBCL who were ineligible for high-dose therapy/stem cell transplant and had received ≥1 prior line of systemic therapy were enrolled across 20 sites in China. Treatment included zanu 160 mg BID po continuously in 28-day cycles until disease progression, unacceptable toxicity, loss to follow-up, or end of the study. Evaluations occurred at baseline, then every 12 weeks for 24 months [mo], and then every 24 weeks thereafter. Primary endpoint was overall response rate (ORR) per 2014 Lugano criteria. Secondary endpoints included complete response (CR) rate, duration of response (DOR), time to response (TTR), progression-free survival (PFS), overall survival (OS), and safety per NCI-CTCAE v5.0. Immunohistochemistry (IHC) DLBCL subtyping data were collected from local sites. For pts without local IHC data, central laboratory staining of CD10, BCL-6, and MUM1 with FFPE tissue specimens were used to classify phenotypes using the Han's Algorithm. Molecular profiling was conducted in biomarker-evaluable pts using DNA (custom Oncolym-413 panel [Gene+]) and RNA sequencing data. Between August 2021 and March 2025, a total of 65 pts were enrolled. All pts were Asian; 52.3% (n=34) of pts were male. Median age was 66 years (range: 42-92 years), 86.2% had non-GCB DLBCL, 63.1% had relapsed disease, 81.5% had Ann Arbor stage III/IV disease, and 84.6% had Eastern Cooperative Oncology Group performance status of ≤1. Median number of prior therapies was 1. All pts were included in the safety and efficacy analysis set. With a median follow-up of 13.9 mo (range: 0.5-36.4 mo), the ORR was 46.2%, CR was 29.2%, and PR was 16.9%. In responders, the median DOR was 22.7 mo (2.8-NE). Median TTR was 2.8 mo. Median PFS and OS were 4.3 mo (95% CI 2.7-5.5) and 18.1 mo (95% CI 11.4-NE), respectively. Pts with non-GCB DLBCL achieved encouraging response rates (ORR 51.8%). Zanu demonstrated a tolerable and manageable safety profile consistent with the established safety profile. The reported incidence of grade ≥3 treatment-related treatment-emergent adverse events (TRAEs) was 16 pts (24.6%). The most frequently reported TRAEs of grade ≥3 were pneumonia (6.2%), decreased neutrophil count (6.2%), decreased platelet count (3.1%), anemia (3.1%), and decreased lymphocyte count (3.1%). With a median exposure time to zanu of 4.0 mo (range: 0.3-36.4 mo), no atrial fibrillation or flutter, hypertension, opportunistic infections, second primary malignancies, or tumor lysis syndrome events were reported. Retrospective biomarker analyses demonstrated that 32/64 pts with low levels of ctDNA (<275.898 hGE/mL) at baseline had favorable clinical outcomes including higher response rates (ORR 62.5% vs 28.1%; P<.0001) and longer PFS (P<.01). Further, pts who achieved CR had a high rate of undetectable ctDNA at first response following zanu treatment (17/18; 94.4%). Pts with co-occurring CD79B and MYD88L265P mutations demonstrated better ORR and CR compared with pts without (ORR 14/23 vs 15/41, P=.070; CR 10/23 vs 8/41, P<.05). Acquired BTK and PLCG2 mutations were observed in 17/34 pts who progressed following zanu treatment.Conclusion: Zanu demonstrated encouraging antitumor activity and a tolerable safety profile in pretreated CD79B-mutated R/R DLBCL. Retrospective biomarker analyses demonstrated improved zanu response in pts with non-GCB subtype, low baseline ctDNA levels, or co-occurring CD79B and MYD88L265P mutations. These data suggest zanu may provide clinical benefit for these pts with limited therapeutic options.

ObjectivesTo develop and validate radiomics and deep learning models based on contrast-enhanced MRI (CE-MRI) for differentiating dual-phenotype hepatocellular carcinoma (DPHCC) from HCC and intrahepatic cholangiocarcinoma (ICC).MethodsOur study consisted of 381 patients from four centers with 138 HCCs, 122 DPHCCs, and 121 ICCs (244 for training and 62 for internal tests, centers 1 and 2; 75 for external tests, centers 3 and 4). Radiomics, deep transfer learning (DTL), and fusion models based on CE-MRI were established for differential diagnosis, respectively, and their diagnostic performances were compared using the confusion matrix and area under the receiver operating characteristic (ROC) curve (AUC).ResultsThe radiomics model demonstrated competent diagnostic performance, with a macro-AUC exceeding 0.9, and both accuracy and F1-score above 0.75 in the internal and external validation sets. Notably, the vgg19-combined model outperformed the radiomics and other DTL models. The fusion model based on vgg19 further improved diagnostic performance, achieving a macro-AUC of 0.990 (95% CI: 0.965-1.000), an accuracy of 0.935, and an F1-score of 0.937 in the internal test set. In the external test set, it similarly performed well, with a macro-AUC of 0.988 (95% CI: 0.964-1.000), accuracy of 0.875, and an F1-score of 0.885.ConclusionsBoth the radiomics and the DTL models were able to differentiate DPHCC from HCC and ICC before surgery. The fusion models showed better diagnostic accuracy, which has important value in clinical application.Critical relevance statementMRI-based deep learning radiomics were able to differentiate DPHCC from HCC and ICC preoperatively, aiding clinicians in the identification and targeted treatment of these malignant hepatic tumors.Key PointsFusion models may yield an incremental value over radiomics models in differential diagnosis.Radiomics and deep learning effectively differentiate the three types of malignant hepatic tumors.The fusion models may enhance clinical decision-making for malignant hepatic tumors.

To develop and validate radiomics and deep learning models based on Gd-EOB-DTPA enhanced MRI for differentiation between hepatocellular carcinoma (HCC) and focal nodular hyperplasia (FNH) showing iso- or hyperintensity in the hepatobiliary phase (HBP). 112 patients from three hospitals were collected totally. 84 patients from hospital a and b with 54 HCCs and 30 FNHs randomly divided into a training cohort (n = 59: 38 HCC; 21 FNH) and an internal validation cohort (n = 25: 16 HCC; 9 FNH). A total of 28 patients from hospital c (n = 28: 20 HCC; 8 FNH) acted as an external test cohort. 1781 radiomics features were extracted from tumor volumes of interest (VOIs) in the pre-contrast phase (Pre), arterial phase (AP), portal venous phase (PP) and HBP images. 512 deep learning features were extracted from VOIs in the AP, PP and HBP images. Pearson correlation coefficient (PCC) and analysis of variance (ANOVA) were used to select the useful features. Conventional, delta radiomics and deep learning models were established using machine learning algorithms (support vector machine [SVM] and logistic regression [LR]) and their discriminatory efficacy assessed and compared. The combined deep learning models demonstrated the highest diagnostic performance in both the internal validation and external test cohorts, with area under the curve (AUC) values of 0.965 (95

Objective:: In this study, a radiomics model was created based on High-Resolution Computed Tomography (HRCT) images to noninvasively predict whether the sub-centimeter pure Ground Glass Nodule (pGGN) is benign or malignant. Methods:: A total of 235 patients (251 sub-centimeter pGGNs) who underwent preoperative HRCT scans and had postoperative pathology results were retrospectively evaluated. The nodules were randomized in a 7:3 ratio to the training (n=175) and the validation cohort (n=76). The volume of interest was delineated in the thin-slice lung window, from which 1316 radiomics features were extracted. The Least Absolute Shrinkage and Selection Operator (LASSO) was used to select the radiomics features. Univariate and multivariable logistic regression were used to evaluate the independent risk variables. The performance was assessed by obtaining Receiver Operating Characteristic (ROC) curves for the clinical, radiomics, and combined models, and then the Decision Curve Analysis (DCA) assessed the clinical applicability of each model. Results:: Sex, volume, shape, and intensity mean were chosen by univariate analysis to establish the clinical model. Two radiomics features were retained by LASSO regression to build the radiomics model. In the training cohort, the Area Under the Curve (AUC) of the radiomics (AUC=0.844) and combined model (AUC=0.871) was higher than the clinical model (AUC=0.773). In evaluating whether or not the sub-centimeter pGGN is benign, the DCA demonstrated that the radiomics and combined model had a greater overall net benefit than the clinical model. Conclusion:: The radiomics model may be useful in predicting the benign and malignant sub-centimeter pGGN before surgery.

BACKGROUND:Treating relapsed or refractory classical Hodgkin lymphoma (R/R cHL) remains challenging. This report extends the three-year follow-up period for the phase Ⅱ YH-S001-04 trial, expanding upon the initial 15.8-month analysis. METHODS:Zimberelimab 240 mg was administered every two weeks for two years or until disease progression or death. The endpoint was the objective response rate (ORR), progression-free survival (PFS), overall survival (OS), and safety. RESULTS:The median follow-up was 38.0 months (3.5-42.8 months). The ORR was 91.6 % (95 % CI, 83.8-95.9). Median PFS was 23.6 months, with a longer PFS in responders (28.5 months) compared to non-responders (9.2 months) (P=0.0098). Complete responders had longer mPFS than partial responders (Not reached vs. 28.5 months, P=0.3469). Relapsed patients had improved mPFS compared to refractory cHL (23.6 vs. 10.6 months, P=0.0061). Patients with ＜3 lines of therapy showed longer mPFS compared to ≥3 lines (not reached vs. 23.6 months, P=0.0095). The 3-year OS rate was 94.0 % (95 % CI, 85.9-97.4). No serious adverse events with incidence >5 %. CONCLUSIONS:With encouraging data on both PFS and OS, zimberelimab demonstrates ongoing efficacy and safety in treating R/R cHL, supporting zimberelimab as an effective treatment alternative for R/R cHL (NCT03655483).

Background Prior MRI studies on vestibular migraine (VM) have revealed abnormalities in static regional intrinsic brain activity (iBA) and dynamic functional connectivity between brain regions or networks. However, the temporal variation and concordance of regional iBA measures remain to be explored. Methods 57 VM patients during the interictal period were compared to 88 healthy controls (HC) in this resting-state functional magnetic resonance imaging (fMRI) study. The dynamics and concordance of regional iBA indices, including amplitude of low-frequency fluctuations (ALFF) and regional homogeneity (ReHo), were examined by utilizing sliding time-window analysis. Partial correlation analyses were performed between clinical parameters and resting-state fMRI indices in brain areas showing significant group differences. Results The VM group showed increased ALFF and ReHo dynamics, as well as increased temporal concordance between ALFF and ReHo in the bilateral paracentral lobule and supplementary motor area relative to the HC group. We also found decreased ReHo dynamics in the right temporal pole, and decreased ALFF dynamics in the right cerebellum posterior lobe, bilateral angular gyrus and middle occipital gyrus (MOG) in the VM group compared with the HC group. Moreover, a positive correlation was observed between ALFF dynamics in the left MOG and vertigo disease duration across all VM patients. Conclusion Temporal dynamics and concordance of regional iBA indices were altered in the motor cortex, cerebellum, occipital and temporoparietal cortex, which may contribute to disrupted multisensory processing and vestibular control in patients with VM. ALFF dynamics in the left MOG may be useful biomarker for evaluating vertigo burden in this disorder.

OBJECTIVES:Peripheral whole blood cell counts have been used as prognostic indicators for various cancers, but their predictive value in nasopharyngeal carcinoma remain unclear. This study aims to evaluate the prognostic significance of the pretreatment hemoglobin×lymphocyte/monocyte ratio (HLMR) in non-recurrent, non-metastatic NPC patients undergoing definitive radiotherapy. METHODS:Clinical and follow-up data from 805 NPC patients who completed definitive radiotherapy or chemoradiotherapy were retrospectively analyzed. Pretreatment hemoglobin, lymphocyte count, and monocyte count were collected to calculate HLMR. Receiver operating characteristic (ROC) curves were used to determine the optimal cut-off value of HLMR. Patients were then classified into high and low HLMR groups. The association between HLMR and clinicopathological characteristic was assessed using chi-square tests. Independent prognostic factors for overall survival (OS) and progression-free survival (PFS) were identified using Cox proportional hazards models. A nomogram was constructed based on the independent predictors to estimate patient survival rates, and internal validation was performed using a validation cohort. RESULTS:The ROC curve identified 605.5 as the optimal HLMR cut-off value for predicting 5-year survival. Multivariate Cox regression analysis revealed that T stage (HR=1.886, 95% CI 1.331 to 2.673, P<0.001), N stage (HR=2.021, 95% CI 1.267 to 3.225, P=0.003), Eastern Cooperative Oncology Group (ECOG) score (HR=3.991, 95% CI 1.257 to 12.677, P=0.019), concurrent chemoradiotherapy regimen (HR=0.338, 95% CI 0.156 to 0.731, P=0.006), and HLMR (HR=0.648, 95% CI 0.460 to 0.912, P=0.013) were independent prognostic factors for OS. A nomogram including T stage, N stage, and HLMR in the training cohort was constructed to predict 3-, 5-, and 7-year OS, with a C-index of 0.713. The area under the curves for predicting 3-, 5-, and 7-year OS were 0.744, 0.665, and 0.682, respectively. Calibration curves showed good agreement between predicted and observed survival rates. The above results were further confirmed in the validation cohort. CONCLUSIONS:Pretreatment HLMR may serve as a promising prognostic biomarker for patients with nasopharyngeal carcinoma.

6039 Background: In October 2023, the US FDA approved toripalimab in combination with Gemcitabine-Cisplatin (GP) as first-line treatment for recurrent or metastatic (r/m) NPC based on the results of JUPITER-02 study (NCT03581786). Here we report the results of four-year overall survival (OS) follow-up and dynamic EBV DNA copy number and its correlation with clinical outcome. Methods: Patients with r/m NPC (n=289) were randomized (1:1) to receive toripalimab 240 mg (n=146) or placebo (n=143) in combination with GP once every 3 weeks (Q3W) for up to 6 cycles, followed by monotherapy with toripalimab or placebo until disease progression, intolerable toxicity, or completion of 2 years of treatment. Stratification factors were ECOG performance score (0 vs. 1) and extent of disease (recurrent vs.primary metastatic). The primary endpoint was progression free survival by an independent review committee. Secondary endpoints included OS and safety. Exploratory endpoints included dynamic blood EBV DNA copy number and its correlation with clinical efficacy. Results: By the cutoff date of January 9, 2024, 50 months after the last patient was enrolled, 150 deaths were recorded. Compared with the results from the final OS analysis, consistent survival improvement was observed for toripalimab over placebo: HR=0.61 (95% CI: 0.44-0.85), nominal p=0.0027. The median OS was not yet reached in the toripalimab arm and was 33.7 months in the placebo arm. The 5-year OS rates were 52.0% in the toripalimab arm, and 33.9% in the placebo arm. Among patients with detectable baseline EBV DNA copy number and at least one EBV result after study treatment, significantly more patients from the toripalimab arm had EBV DNA copy number decreased to undetectable level than those in the placebo arm, 96.3% vs. 84.5%, p= 0.004. In addition, significantly less patients experienced EBV DNA copy number rebound in the toripalimab arm than in the placebo arm after the initial reduction, 36.5% vs, 57.4%, p=0.002. The median time from the lowest EBV DNA copy number to the rebound was 20.5 vs. 6.0 months in the toripalimab and placebo arms respectively. The rebound also preceded investigator-assessed disease progression by a median of 1.9 months in the toripalimab arm. Conclusions: The combination of toripalimab and chemotherapy showed long term survival benefit than chemotherapy alone with a 5-year OS rate at 52%. EBV DNA copy number might be used to monitor clinical response and predict disease progression. Clinical trial information: NCT03581786 .

To investigate the value of radiomics analysis of dual-layer spectral-detector computed tomography (DLSCT)-derived iodine maps for predicting tumor deposits (TDs) preoperatively in patients with colorectal cancer (CRC). A total of 264 pathologically confirmed CRC patients (TDs + (n = 80); TDs − (n = 184)) who underwent preoperative DLSCT from two hospitals were retrospectively enrolled, and divided into training (n = 124), testing (n = 54), and external validation cohort (n = 86). Conventional CT features and iodine concentration (IC) were analyzed and measured. Radiomics features were derived from venous phase iodine maps from DLSCT. The least absolute shrinkage and selection operator (LASSO) was performed for feature selection. Finally, a support vector machine (SVM) algorithm was employed to develop clinical, radiomics, and combined models based on the most valuable clinical parameters and radiomics features. Area under receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis were used to evaluate the model’s efficacy. The combined model incorporating the valuable clinical parameters and radiomics features demonstrated excellent performance in predicting TDs in CRC (AUCs of 0.926, 0.881, and 0.887 in the training, testing, and external validation cohorts, respectively), which outperformed the clinical model in the training cohort and external validation cohorts (AUC: 0.839 and 0.695; p: 0.003 and 0.014) and the radiomics model in two cohorts (AUC: 0.922 and 0.792; p: 0.014 and 0.035). Radiomics analysis of DLSCT-derived iodine maps showed excellent predictive efficiency for preoperatively diagnosing TDs in CRC, and could guide clinicians in making individualized treatment strategies. The radiomics model based on DLSCT iodine maps has the potential to aid in the accurate preoperative prediction of TDs in CRC patients, offering valuable guidance for clinical decision-making.