Percorrer por data de Publicação, começado por "2026-03-16"
A mostrar 1 - 1 de 1
Resultados por página
Opções de ordenação
- Liquid Biopsies of Endobronchial Ultrasound Transbronchial Needle Aspiration Supernatant for the Molecular Characterisation of Non Small Cell Lung CancerPublication . Rodrigues, Luís Miguel Vaz; Sousa, Vitor Manuel Leitão de; Barata, Luís Manuel Taborda; Cordovilla Pérez, RosaIntroduction The diagnosis and molecular characterisation of non-small cell lung cancer (NSCLC) has undergone a paradigm shift over the last decades, driven by the need for timely and comprehensive biomarker profiling to guide precision therapies. Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has become a cornerstone technique, providing a minimally invasive approach that allows simultaneous diagnosis and mediastinal staging in a single procedure with high yield. However, challenges remain in optimising tissue acquisition, balancing morphological, immunohistochemical, and molecular needs, and reducing turnaround times to avoid delays in treatment initiation. The primary objective of this thesis was to evaluate the role of EBUS-TBNA in the diagnostic and molecular workflow of advanced NSCLC, focusing on its ability to provide adequate material for biomarker testing and the potential of its supernatant as a tumourproximal liquid biopsy source. This aim was explored through four complementary areas, aligned with secondary objectives that collectively address the primary question: (i) the positioning of EBUS-TBNA within real-world diagnostic pathways; (ii) the adequacy and limitations of sequential testing strategies; (iii) the comparative performance of sequential versus massively parallel next-generation sequencing (MPNGS); and (iv) the feasibility, concordance, and clinical utility of the supernatant phase of EBUS-TBNA derived samples. Methods This work integrates five studies: three retrospective observational analyses, one systematic review and meta-analysis, and one prospective comparative study. All studies were conducted primarily at the Pulmonology Department of the Francisco Gentil - Portuguese Oncology Institute of Coimbra (Instituto Português de Oncologia de Coimbra, Francisco Gentil – IPOC-FG), in collaboration with the Institute's Molecular Pathology Laboratory, and, for the prospective study, the Institute of Anatomical Pathology and the Molecular Pathology Laboratory of the Faculty of Medicine, University of Coimbra. Ethical approval (ref. 23-2022) was obtained, and written informed consent was secured from all participants, or from their legal representatives when applicable. Data collection included demographic, clinical, histological, procedural, and molecular variables, as well as timelines from diagnostic suspicion to molecular results and treatment initiation. Retrospective studies analysed electronic medical records and pathology databases, while the prospective study collected paired cell pellet (CP) and supernatant (SP) samples during routine EBUS-TBNA procedures, processed according to standard protocols. DNA and RNA were extracted, quantified, and analysed by NGS in Studies I, III, and V. Studies II and III also applied a sequential workflow, with EGFR testing performed by real-time polymerase chain reaction (RT-PCR) and ALK/ROS1 rearrangements assessed by fluorescence in situ hybridisation (FISH). The systematic review and meta-analysis (Study IV) followed PRISMA 2020 guidelines with a prospectively registered protocol (PROSPERO CRD42024600046). Statistical analyses were performed under the supervision of a biostatistician from the IPOC-FG Research Support Unit, using IBM SPSS v.27.0 (IBM Corp., NY, USA); a two-sided pvalue <0.05 was considered statistically significant. Detailed methodology and specific statistical approaches are provided in each study. Results Study I involved a real-world cohort of 205 patients with stage IV NSCLC. Median age was 68 years (range, 38–89), with 59.5% male. EBUS-TBNA and EUS-B accounted for over half of the initial diagnostic procedures (51.7%), enabling significantly shorter times from clinical evaluation to biopsy compared with transthoracic biopsy or surgical approaches (median 8 and 5 vs. 20.5 and 24.5 days, respectively; p < 0.001). Molecular profiling was attempted in all cases, achieving high adequacy across modalities (97.6%). Actionable alterations identified included EGFR mutations (26.3%), KRAS (14.1%), ALK (6.8%), and less frequent alterations such as ERBB2 (3.9%), BRAF (1%), and MET (1%). Delays in diagnostic initiation independently predicted mortality, with delays ≥10 days from clinical suspicion to biopsy associated with higher risk of death (HR 1.66; 95% CI, 1.10–2.50; p = 0.016). Study II analysed a retrospective cohort of 59 patients diagnosed with NSCLC who underwent EBUS-TBNA as part of their initial diagnostic work-up. Among these, molecular testing was performed sequentially on EBUS-TBNA samples in 64.4% of cases. Adequacy rates were high, with successful analysis in 89.5% of samples for EGFR and 81.3% for ALK, confirming the feasibility of this approach. However, a progressive decline in yield across different biomarkers was observed, suggesting gradual sample exhaustion with multiple sequential tests. Additionally, in approximately one-third of cases, molecular testing was performed on alternative specimens despite the availability of adequate material obtained through EBUS-TBNA. Study III compared sequential molecular profiling (SMP) with MP-NGS in 106 patients with stage IV NSCLC. MP-NGS achieved universal adequacy (100%) and detected a significantly higher rate of actionable mutations compared with SMP (40.9% vs. 22.2%; p = 0.042), leading to a greater allocation to targeted therapies (44.3% vs. 22.2%; p = 0.038). This observation may have contributed to the trend toward improved overall survival in the MP-NGS group (median 672 vs. 138 days), although the difference did not reach statistical significance (p = 0.053). Turnaround times were prolonged in both strategies, exceeding recommended benchmarks, but were slightly shorter with SMP compared with externally outsourced MP-NGS (median 17 vs. 23 days; p = 0.076). Study IV, a systematic review and meta-analysis, synthesised evidence from seven studies including 506 patients. Molecular profiling from the supernatant phase demonstrated high feasibility (87–100%), excellent concordance with matched tissue samples (pooled κ=0.947), and consistently shorter turnaround times (up to seven days faster), despite heterogeneity in pre-analytical processing protocols. Subgroup analyses indicated no significant effects of storage medium or temperature on DNA yield, although protocol variability limited definitive conclusions on optimal processing workflows. Study V, a prospective study, evaluated paired, EBUS-TBNA derived, cell pellet and supernatant samples from 20 patients. Nucleic acid yields were significantly higher in supernatant compared to cell pellet (DNA: 28.95 ng/μL vs. 9.84 ng/μL, p = 0.025; RNA: 37.6 ng/μL vs. 15.95 ng/μL, p = 0.007). Molecular concordance between fractions was 85%, with three discordant cases: two due to insufficient supernatant material and one true molecular discrepancy, where a KRAS G12C mutation was detected exclusively in supernatant. Simulating a parallel diagnostic workflow, in which molecular testing on supernatant was initiated at the time of biopsy while cell pellet underwent standard histological processing, reduced the overall diagnostic time from a median of 21.5 days to 13 days (p < 0.001). Discussion and Conclusion This thesis demonstrates that EBUS-TBNA is a pivotal tool in the diagnostic and molecular workflow of advanced NSCLC, providing high-quality material for comprehensive characterisation and enabling shorter timelines to diagnosis and treatment initiation, a factor independently associated with survival. While sequential molecular strategies proved feasible, they were limited by sample depletion and reduced mutation detection. MP-NGS demonstrated superior performance, improving the identification of actionable mutations and access to targeted therapies. The integration of the supernatant phase, offering valuable nucleic acid yields, strong molecular concordance with conventional tissue based molecular profiling and the potential to accelerate turnaround times, represents a possible key advancement in the diagnostic workflow. Taken together, these findings support an integrated diagnostic framework where the cellular fraction is used for histology and immunohistochemistry while molecular profiling is performed in parallel from the supernatant phase. This approach optimises tissue use, ensures diagnostic accuracy, and accelerates access to molecular profiling results, aligning workflows with the demands of precision oncology. Validation through multicentre prospective studies and standardised protocols will be crucial to drive widespread clinical implementation.