JCO precision oncology最新文献

Purpose: Multidisciplinary molecular tumor boards (MTBs) have demonstrated improved clinical outcomes in various malignancies. Sequential next-generation sequencing (NGS) is widely performed at the time of recurrence, which may lead to multiple MTB reviews. This study assessed the clinical benefit of multiple MTB reviews for sequential NGS.

Methods: A retrospective cohort review was performed to compare patients reviewed once at a single-institution MTB and those reviewed multiple times for the same diagnosis with sequential NGS between January 2016 and December 2023. Demographics were compared, and regression and survival analysis was performed.

Results: In all, 3,702 patients were included, 3,446 (91.6%) were reviewed once, and 256 (8.4%) were reviewed multiple times. Patients reviewed once had higher proportion of actionable findings (52.7% v 44.5%, P = .01) compared with those reviewed multiple times at initial review. Approximately half of the patients at their second (47.3%) and third or more (62.5%) reviews had actionable findings. The mean time on recommended targeted therapy for patients reviewed multiple times was 8.24 months (standard deviation [SD], 9.34; range, 0.07-42.4), and the mean time on recommended immunotherapy was 7.02 months (SD, 9.47; range, 0.1-39.2). Patients with multiple MTB reviews had a 30% lower risk of death compared with those reviewed once, even after controlling for primary site, age, presence of actionable NGS findings, and stage (hazard ratio, 0.7, 95% CI, 0.55 to 0.89; P = .004).

Conclusion: Sequential NGS identified actionable findings in approximately 50% of those with multiple reviews, and many patients stay on recommended therapy for at least 6 months. Patients with sequential NGS and multiple reviews have a survival advantage, and although this may be due to therapy sensitivity, it reflects the importance of NGS testing and usefulness of MTB in interpreting those results.

The presence of circulating tumor DNA (ctDNA) in patients indicates post-treatment molecular residual disease (MRD). Given the complexity of ctDNA-based MRD detection tests, consensus on analytical validation (AV) criteria is needed. To address this, the Blood Profiling Atlas in Cancer (BLOODPAC) Consortium's MRD AV Working Group evaluated existing protocols to develop standardized guidance for tumor-informed assays. Protocols pertaining to blood collection tube types, quantitative output, tissue processing, tumor or matched normal sequencing, software, and clinical validation were considered out of scope. After alignment on objectives and assumptions, study designs on the basis of best practices in the field, available assay validation guidance documents, and unique performance challenges for tumor-informed MRD assays were authored. Each protocol contains introduction, experimental design, statistical analysis, and an example data presentation per the US Food and Drug Administration (FDA) Center for Devices and Radiological Health standard format. Biostatisticians were consulted to define minimal test requirements, sample size, and appropriate statistical analyses. The protocols were submitted to the FDA via the presubmission process for formal written feedback followed by a meeting. BLOODPAC's generic protocols for the AV of tumor-informed ctDNA assays for MRD are designed to provide test developers with a core baseline of standardized AV protocols such that methods described can be adapted and applied for any tumor-informed MRD assay irrespective of technology, panel design algorithm, or workflow component. These protocols aim to optimize test developers' presubmission reviews with the FDA, ensuring productive meetings while enabling reviewers to streamline feedback. As always, test developers are encouraged to communicate with FDA directly around their particular AV methods.

Purpose: RAS mutation is a key biomarker of anti-epidermal growth factor receptor (EGFR) antibody resistance in colorectal cancer (CRC). However, the clinical impact of RAS amplification on the efficacy of anti-EGFR therapy remains unclear. This study aimed to characterize RAS-amplified CRC and evaluate the sensitivity of these tumors to anti-EGFR antibodies.

Methods: We conducted a retrospective observational study using the Center for Cancer Genomics and Advanced Therapeutics database in Japan, which includes clinical and genomic data from patients who underwent comprehensive genomic profiling. We analyzed the data from 9,135 patients with unresectable colorectal adenocarcinoma (CRA) who underwent FoundationOne CDx testing.

Results: RAS amplification was identified in 2.1% (188/9,135) of patients with CRA. Among 1,649 patients with RAS wild-type CRA who received first-line chemotherapy with anti-EGFR antibodies, those with RAS amplification had a lower overall response rate (ORR) and a shorter time to treatment failure (TTF) compared with those without RAS amplification (ORR, 37.5% [21/56] v 52.9% [843/1,593]; median TTF, 195 days [95% CI, 129 to 224] v 274 days [95% CI, 258 to 293]; P = .023 and P = .005, respectively). By contrast, among 4,858 patients treated with bevacizumab, no significant differences were observed in ORR (37.3% [31/83] v 37.1% [1,772/4,775]; P = .964) or TTF (median, 231 days [95% CI, 175 to 273] v 259 days [95% CI, 252 to 270]; P = .445).

Conclusion: RAS amplification is a rare alteration that may confer resistance to anti-EGFR antibodies. Assessing RAS amplification status may help guide the appropriate use of anti-EGFR antibodies in clinical practice.

Purpose: High levels of DNA replication stress and defects in the DNA damage response (DDR) pathways are vulnerabilities of many poor prognosis childhood malignancies. Ataxia telangiectasia and Rad3-related protein (ATR) is a key regulator of these pathways and constitutes an attractive target, especially in combination. However, the malignancies where ATR inhibitors have maximum benefit and synergistic combinations differ between adults and children.

Design: ACCELERATE convened a multistakeholder meeting and conducted review and analysis to propose the optimal pathway for the development of ATR inhibitors in pediatric malignancies.

Results: Considering the lack of identified biomarkers, the initial evaluation of ATR inhibitors should focus on Ewing sarcoma, rhabdomyosarcoma, and neuroblastoma in view of their high levels of DNA replication stress and defects in DDR pathways. Early phase trials of ATR inhibitors should be iterative, based on a clear hypothesis with responders and nonresponders undergoing detailed molecular analysis and a revised new hypothesis generated. Trial designs should restrict monotherapy evaluation to a brief exposure in a small number of patients and progress rapidly to combinations. Highlighted combination partners are poly(ADP-ribose) polymerase inhibitors and antibody drug conjugates with topoisomerase I inhibitor payloads. Combinations with ALK inhibitors (in ALK/MYCN-aberrant neuroblastoma) and aurora A kinase (in MYCN-amplified) are supported by robust mechanisms of action and preclinical data. Early interactions with regulators are crucial, and early phase clinical trials should be conducted in regulatory-approved, academic-sponsored, industry-supported, platform trials.

Conclusion: ATR inhibitors are a prototype for the development of medicinal products in a limited pediatric population. For the substantial potential of ATR inhibitors in children with malignancy to be realized, strategic planning between academia, industry, regulators, and patient advocates is vital.

Purpose: This study evaluates deep learning (DL) approaches, particularly long short-term memory (LSTM) networks, for analyzing dynamic changes in the hepatocellular carcinoma (HCC) tumor microenvironment (TME) to improve disease understanding and treatment prediction.

Materials and methods: Multimodal HCC TME data (high-throughput sequencing, protein expression, and time-series imaging) were integrated. Spatial features were extracted using convolutional neural networks (CNNs), while temporal patterns were modeled with LSTMs. Generative adversarial networks (GANs) augmented data for robust training. Model performance was assessed on the basis of dynamic TME characterization and outcome prediction accuracy.

Results: LSTMs demonstrated superior performance in analyzing TME time-series data, effectively capturing long-term dependencies and predicting cellular interactions (accuracy: 92.3% v 85.7% for CNNs alone). The integrated DL framework successfully characterized spatiotemporal TME evolution and treatment response patterns.

Conclusion: This study demonstrates that LSTMs are powerful tools for analyzing and understanding the dynamic changes in the HCC immune microenvironment. Their strong capability in time-series data analysis provides new opportunities for uncovering the mechanisms of HCC progression and optimizing therapeutic strategies. However, because of the lack of detailed etiological information, stratified validation across different etiologies such as hepatitis B virus, hepatitis C virus, and non-alcoholic steatohepatitis has not yet been conducted in this study and should be addressed in future work. The findings highlight the important role of DL technologies in future research and treatment of HCC.