JCO precision oncology最新文献

Purpose: Gastric cancer (GC) is a significant global health challenge, often diagnosed late, limiting treatment success. Early detection through biomarkers is critical for better outcomes. This study developed a noninvasive, cost-effective tool using circulating cell-free DNA (cfDNA) sequencing and advanced modeling for early GC detection.

Methods: We performed low-coverage whole-genome sequencing on plasma from 404 patients with GC and 428 non-GC participants (healthy donors and patients with colorectal, esophageal, or liver cancer). Participants were divided into a training set (579) and a validation set (253). A deep learning model, GastricAI, was trained on raw cfDNA reads using Mamba, an advanced sequence model enhancing the structured state space model (S4) with adaptive parameters. Performance was evaluated in a second validation set (73 GC patients, 94 controls).

Results: In the first validation set, GastricAI achieved an area under receiver operating characteristic curve (AUROC) of 0.886 (95% CI, 0.842 to 0.930), accuracy of 0.846, sensitivity of 0.789, and specificity of 0.900. Across TNM stages, it recorded sensitivity/specificity of 0.920/0.738 (stage I), 0.846/0.900 (stage II), 0.811/0.900 (stage III), and 1.000/0.623 (stage IV). In the second validation set, it showed an AUROC of 0.850 (95% CI, 0.793 to 0.907), accuracy of 0.772, sensitivity of 0.973, and specificity of 0.617. GastricAI consistently detected intestinal, diffuse, and mixed GC subtypes per Lauren's classification and outperformed conventional biomarkers carbohydrate antigen 724 (P < .001) and carcinoembryonic antigen (P = .037).

Conclusion: GastricAI provides a promising noninvasive method for GC detection via cfDNA sequencing, with high sensitivity and reasonable specificity. It holds potential for early screening to improve prognosis, although further prospective studies are needed to validate and refine its clinical utility across diverse populations.

Purpose: The atlas of immune microenvironment at single-cell level in BRCA1/2-mutated breast cancer is largely unknown and whether an immune signature on the basis of single-cell atlas is associated with response to neoadjuvant chemotherapy remains to be investigated.

Materials and methods: The immune microenvironment between BRCA1/2-mutated and BRCA wild-type breast tumors was explored using single-cell RNA sequencing (scRNA-seq) assay and was validated in an independent cohort of 40 BRCA1/2 carriers and 56 noncarriers via immunohistochemistry assay (IHC). Bulk RNA-seq was performed using RNA extracted from fresh-frozen pretreatment core-needle tumor tissues in 80 BRCA1/2 carriers with operable primary human epidermal growth factor receptor 2-negative tumors who received neoadjuvant chemotherapy, and the associations between immune cell subtypes defined by scRNA-seq and pathologic complete response (pCR) were investigated.

Results: BRCA1/2-mutated tumors exhibited an enriched immune microenvironment compared with the wild-type counterparts at single-cell level, particularly regulatory T cells and exhausted T cells, which were validated in the IHC cohort. Among the neoadjuvant chemotherapy cohort of 80 BRCA1/2 carriers, 36.2% achieved a pCR. We established an immune signature on the basis of the single-cell and bulk RNA-seq data, named BRCA-IM. The BRCA-IM model exhibited an excellent prediction of pCR in the neoadjuvant chemotherapy cohort, with AUC values of 0.81(95% CI, 0.69 to 0.92) in the training set and 0.91(95% CI, 0.79 to 1.00) in the test set, respectively; and the BRCA-IM model remained as an independent predictor for pCR after adjusting for other factors. Moreover, higher BRCA-IM scores were significantly associated with more favorable survival in the neoadjuvant chemotherapy cohort.

Conclusion: BRCA1/2-mutated breast cancer shows an enriched tumor immune microenvironment, and the BRCA-IM model exhibits a good performance in prediction of pCR in BRCA1/2-mutated tumors.

Purpose: We previously developed quality control materials (QCMs) to aid in the development of circulating tumor DNA (ctDNA) assays. In this study, we further characterize the performance of the QCMs relative to clinical samples.

Methods: QCMs were provided by three manufacturers. To functionally characterize the QCMs, we (1) evaluated EGFR L858R and ex19del (range, 0.5%-5.0% variant allele frequency [VAF]) in QCMs compared with clinical samples by droplet digital polymerase chain reaction (ddPCR), targeted-amplicon sequencing (Tag-seq), and hybrid capture next-generation sequencing (NGS); and (2) evaluated the QCMs and clinical samples near the Tag-seq limit of detection. A clinical pilot was also conducted in 11 clinical laboratories spanning four continents.

Results: For functional characterization, part 1, QCM VAFs for hybrid capture were similar to ddPCR for EGFR L858R but lower for ex19del. By contrast, hybrid capture results for EGFR L858R clinical samples showed a positive trend compared with ddPCR. For amplicon NGS, QCMs performed similarly to clinical samples for both variants. For part 2, observed hit rates approximated expected values. In the clinical pilot, median ex19del VAF was higher for Tag-seq than hybrid capture for both 1.0% and 0.5% QCM formulations. Median QCM L858R VAFs were similar for Tag-seq and hybrid capture, with greatest interlaboratory differences observed for Thermo Fisher Scientific QCMs. For non-EGFR variants, we observed assay and QCM-dependent trends, with no particular QCM or assay driving these trends.

Conclusion: This project revealed unexpected differences in performance of both assays and QCMs. These findings highlight the need for further validation across diverse alteration types and merit consideration by laboratories that rely on QCMs to develop and perform ctDNA assays for diagnostic applications.

In clinical trials, the initial step typically involves assessing a new drug's toxicity profile, aiming to identify a tolerable dose level for subsequent studies. In phase I trials, the primary objective is to determine the maximum tolerated dose, defined as the highest dose associated with an acceptable level of toxicity. Numerous methods have been developed to guide dose escalation and de-escalation decisions during trial conduct. Among these approaches, the calibration-free odds (CFO) design has demonstrated superior operating characteristics and has emerged as one of the most effective approaches for dose finding. To facilitate the application of the CFO design in clinical trial practice, an R package and a Shiny app have been released. This study presents CFO decision tables in Excel files to further remove the barrier of applying the CFO design to real trials. Anyone involved in the trial conduct can implement the CFO design with no difficulties. During the trial, dose movement decisions can be made simply by referring to the cumulative data (including numbers of patients treated and observed toxicities) and the pregenerated decision tables, without any additional statistical calculation. This approach significantly enhances the usability of the CFO design and reduces the operational complexity associated with its implementation in clinical trials. The Excel CFO decision tables can be downloaded from CFO Shiny App.

Purpose: We performed spatial transcriptomics of estrogen receptor (ER)-negative, ER-low, and ER-high tumor regions of breast cancers that were intermediate (10%-60%) ER-positive by immunohistochemistry to better understand the intratumor heterogeneity in ER expression and to elucidate whether cells of different molecular subtypes (ie, Luminal A [LumA], Luminal B [LumB], human epidermal growth factor receptor 2-enriched, or Basal-like) can coexist in the same tumor.

Methods: Digital spatial profiling was performed on 10 ER-heterogeneous (10%-60% ER+) and 10 ER-high (>60% ER+) primary breast cancers using the NanoString GeoMx platform with the Human Whole Transcriptome Atlas probe set.

Results: LumA and LumB molecular subtypes were intermixed, but there were no Basal-like populations in these ER-heterogeneous tumors. The ER-negative (ER-) regions were LumB-like and showed lower expression of ESR1 and endocrine therapy sensitivity gene signatures but higher expression of immune-related genes and higher recurrence scores, indicating a more endocrine-resistant but chemotherapy-sensitive phenotype. We also found that ESR1 strongly positive cells enriched after preoperative chemotherapy in clinical trial tissues.

Conclusion: This study demonstrates mixed Lum-A and Lum-B molecular subtypes within ER-intermediate primary breast cancers and reveals that ER- tumor cell populations have molecular features of endocrine resistance but chemotherapy sensitivity. These findings may explain the worse clinical outcomes of patients with ER-heterogeneous breast cancers and suggest benefit from combined endocrine and chemotherapy strategies.

Purpose: Delta-like ligand 3 (DLL3) is an emerging target across neuroendocrine cancers, but remains underexplored in gastroenteropancreatic neuroendocrine neoplasms (GEP NENs), including poorly differentiated gastroenteropancreatic neuroendocrine carcinomas (GEP NECs) and well-differentiated neuroendocrine tumors (NETs). We aimed to define the landscape of DLL3 expression and feasibility of DLL3-targeted imaging in this population.

Patients and methods: We completed DLL3 immunohistochemistry (IHC) on 379 tumor samples from patients with GEP NENs, analyzing associations between DLL3 IHC positivity, clinicopathologic features, and outcomes. [⁸⁹Zr]Zr-DFO-SC16.56 DLL3 immuno-positron emission tomography-computed tomography (immunoPET-CT) imaging was performed in six patients with DLL3 IHC-positive advanced GEP NENs.

Results: Among GEP NECs, DLL3 expression was identified in 55/78 (71%) tumors, was enriched for small cell histology, and did not demonstrate prognostic significance. Among well-differentiated gastroenteropancreatic neuroendocrine tumors, DLL3 expression was identified in 5/235 (2%) of grade 1-2 and 25/66 (40%) grade 3 (G3) tumors, most commonly G3 pancreatic NETs (PanNETs; 22/52, 43%), with univariate analysis revealing increased mortality risk among patients with DLL3-positive advanced G3 PanNETs (hazard ratio 3.27 [95% CI, 1.09 to 9.78]). Between May 28, 2024, and February 10, 2025, six patients with DLL3 IHC-positive GEP NENs underwent [⁸⁹Zr]Zr-DFO-SC16.56 immunoPET-CT imaging, which delineated DLL3-avid tumor lesions in five of six patients (two of two GEP NECs, three of four G3 PanNETs). Tumor-specific uptake of [⁸⁹Zr]Zr-DFO-SC16.56 varied between patients, with maximum standard uptake values ranging from 7.4 to 36.7, with four of six cases demonstrating DLL3 avidity in ≥50% of tumor lesions.

Conclusion: DLL3 is expressed on a majority of GEP NECs and on a subset of high-grade PanNETs marked by poor outcomes. Functional imaging suggests DLL3 as a promising therapeutic target in both GEP NECs and high-grade PanNETs.

Purpose: Renal cell carcinoma (RCC) that metastasizes to the pancreas (RCC-PM) is a rare but known phenomenon. These patients have been described to have indolent disease and longer overall survival than patients with metastasis to other sites. We investigated the genomic landscape and outcomes for patients with RCC-PM.

Methods: We used two cohorts for this study: (1) a Tempus cohort (TC) to investigate the genomic landscape and (2) a Stanford cohort (SC) where we include disease course and outcomes in addition to the genomic landscape. All patients included underwent testing with commercial next-generation sequencing (NGS) platform (Tempus AI, Inc, Chicago, IL.). For the TC, we compared the genomic landscape based on tissue samples that underwent molecular analysis (DNA and RNA sequencing and immune cell subtypes) from RCC tumors metastatic to the pancreas, liver, lung, or brain. For SC, patients from 2000 to 2024 with RCC-PM had a tumor tissue sample undergo NGS testing, and we report baseline characteristics, NGS, treatment history, and outcomes.

Results: Between the TC and SC, we identified 83 patients with RCC-PM. Compared with other sites of metastasis, RCC-PM had enrichment in PBRM1 mutations, similar tumor mutational burden but lower rates of infiltrating B cells and PD-L1 positivity compared with other metastatic sites. In the SC, patients demonstrated long and indolent disease courses, but without clear genomic predictors of benefit to tyrosine kinase inhibitors or immunotherapies.

Conclusion: Our study furthers RCC with pancreatic metastasis as a good clinical prognostic marker. We also identified enrichment of angiogenic signatures, such as PBRM1 mutations and potential suppression of an immunogenic environment. However, despite these findings, no systemic treatment strategy had significantly better outcomes.

Purpose: Neoadjuvant immune checkpoint inhibitors (nICIs) have demonstrated high response rates in deficient mismatch repair/microsatellite instability-high (dMMR/MSI-H) gastroesophageal adenocarcinoma (GEA). The NEONIPIGA and INFINITY trials demonstrated high rates of pathologic complete response (pCR) in this patient population. Furthermore, the INFINITY trial explored the feasibility of managing these patients nonoperatively, demonstrating promising results. This study aimed to evaluate clinical outcomes of nICIs in resectable dMMR/MSI-H GEA, with a focus on the feasibility of nonoperative management (NOM).

Materials and methods: This retrospective cohort study included patients with resectable dMMR/MSI-H GEA and treated with nICIs ± surgery at the Mayo Clinic. Patients were identified from institutional records, and clinical data were retrospectively reviewed. Primary outcomes were clinical complete response (cCR) and pCR. Secondary outcomes included event-free survival (EFS), radiologic complete response (rCR), and immune-related adverse events (irAEs).

Results: A total of 26 patients treated between April 1, 2017, and July 30, 2025, were identified. Nine patients (34.6%) underwent surgery, of whom six (66.7%) achieved pCR. Seventeen patients (65.4%) pursued NOM, with 10 (71.4%) of 14 evaluable patients achieving cCR and 14 (82.4%) of 17 evaluable achieving rCR. One patient who initially achieved cCR had a local recurrence on surveillance endoscopy and underwent salvage endoscopic resection. At a median follow-up of 19.3 months, 15 (88.2%) of 17 patients in the NOM cohort were alive and metastasis-free, with EFS rates of 87.3% at 12 and 24 months for all patients. irAEs occurred in nine patients (34.6%), with no grade ≥3 toxicities.

Conclusion: In this retrospective cohort study, nICIs led to high cCR and pCR rates in resectable dMMR/MSI-H GEA, supporting the use of immune checkpoint inhibitors in this setting and the feasibility of NOM in select patients.