Venceslau Hespanhol

Introduction: Liquid biopsies based on plasma circulating tumour deoxyribonucleic acid (ctDNA) have shown promise in monitoring lung cancer evolution. The expression of ctDNA across time, its relationship with clinicopathological parameters and its association with lung cancer progression through imaging allow us to weigh how useful ctDNA could be in monitoring surgically resectable lung cancer. The aim of this study was to assess the impact of ctDNA analysis implementation in early-stage lung cancer.Methods: A cohort of 47 patients was sequentially recruited. Only 34 patients with early-stage lung cancer were included. All patients had a tissue specimen and five blood samples drawn: at the preoperative stage, from the pulmonary vein, at surgical discharge, at the first follow-up and at the last follow-up. All blood samples were evaluated for ctDNA expression. Results: On average, the maximum yield of ctDNA was obtained in liquid biopsies at the surgical discharge of patients when compared with PO, PV, and F1 (p < 0.0001, p < 0.0001, p < 0.0001 respectively). No statistically significant differences were found when comparing the last follow-up to surgical discharge ctDNA expression (p = 0.851). The correlation between ctDNA concentration according to five-time points and the four clinicopathological characteristics showed that patients younger than 70 years had a statistically significant reduction of the concentration of ctDNA at the preoperative and surgical discharge time point [β = -16 734 (-27 707; - 5760); p = 0.003; β = -21 785 (-38 447; -5123); p = 0.010], as opposed to an increase of the concentration of ctDNA at the pulmonary vein and last follow-up time points [β = 8369 (0.359; 16 378); p = 0.041; β = 34 402 (12 549; 56 254); p = 0.002] all with a confidence level of 95%. In the cases where actionable mutations were identified in tissue biopsies, the expected mutation was found in five out of six patients plasma samples at the pre-operatory time point and in two out of six patients plasma samples at the pulmonary vein time point. Two out of six patients with actionable mutations had disease progression.Conclusion: The results of this pilot study suggest that the maximum yield of ctDNA is obtained at the surgical discharge of the patients and that the pre-operatory timepoint is the one offering the highest sensitivity for the detection of actionable mutations in ctDNA in early-stage lung cancer.

Annual vaccination is fundamental for individual and group protection against seasonal influenza infection. International and Portuguese healthcare organizations have established influenza vaccination coverage rate (VCR) targets for risk groups, namely 75% in people ≥ 65 years old. The Vacinómetro® initiative has been monitoring influenza VCR among target risk groups in Portugal since 2009,: Group 1, ≥ 65 years old; Group 2, patients with chronic conditions; Group 3, healthcare workers in direct contact with patients; and Group 4, 60-64 years old. Besides VCR, social-demographic and health-related variables have been evaluated. During the study period (2009/2010 - 2019/2020), the VCR increased in the 4 target risk groups: from 58.6% to 76.0% in Group 1 (reaching the WHO target); 33.3% to 72.0% in Group 2; 25.0% to 58.9% in Group 3; and 36.6% to 42.8% in Group 4. "Physician recommendation" was the main driver for vaccination whereas "lack of habit" was the main barrier to vaccination. Vacinómetro® data demonstrate that free-of-charge vaccination has a positive impact on VCR. The observed positive trends in influenza VCR demonstrate that public health measures implemented in Portugal to facilitate access to influenza vaccine result in increased vaccine uptake. Strategies to promote population literacy and the physician's awareness should be continued and reinforced. Free-of-charge vaccination criteria extended to more risk groups would also contribute to higher influenza VCR in Portugal.

To evaluate the degree to which evidence from large clinical trials can be applied to patients treated in a local hospital cohort of COPD outpatients. The authors selected seventeen RCTs identified in a systematic way from GOLD 2019 consensus document, and applied their inclusion and exclusion criteria to a real-world cohort of a previous cross-sectional study of 303 COPD outpatients included consecutively. When the inclusion criteria of the 17 RCTs were applied to a real-world cohort of COPD outpatients, only a small portion of them were eligible to participate in the referred trials, from 4.29% to 60.07%. However, when both the inclusion and the exclusion criteria were applied, only as little as 3.63% to as much as 40.59% of patients were eligible to participate. Hence, only a small fraction of patients from this cohort could benefit from the findings of these RCTs. There is a need to complement the efficacy evidence provided by large RCTs according to the extent to which their results, designed to target significant patient populations, can be applied to typical patients treated in routine clinical practice.

Advancements in the development of computer-aided decision (CAD) systems for clinical routines provide unquestionable benefits in connecting human medical expertise with machine intelligence, to achieve better quality healthcare. Considering the large number of incidences and mortality numbers associated with lung cancer, there is a need for the most accurate clinical procedures; thus, the possibility of using artificial intelligence (AI) tools for decision support is becoming a closer reality. At any stage of the lung cancer clinical pathway, specific obstacles are identified and “motivate” the application of innovative AI solutions. This work provides a comprehensive review of the most recent research dedicated toward the development of CAD tools using computed tomography images for lung cancer-related tasks. We discuss the major challenges and provide critical perspectives on future directions. Although we focus on lung cancer in this review, we also provide a more clear definition of the path used to integrate AI in healthcare, emphasizing fundamental research points that are crucial for overcoming current barriers.

Introduction: Atezolizumab represents a therapeutic option in patients with NSCLC previously treated and anti-PD-1/PD-L1-naive. We aimed to assess which patients benefit most from atezolizumab. Methods: We retrospectively analyzed patients with NSCLC, who had received ≥1 previous cytotoxic chemotherapy regimen. Patients with EGFR or ALK genetic alterations had previously received TKI therapy. Disease control was considered if there was stability or objective response. Results: We studied 39 cases, 74.4% were male, with median age of 66 (57-70) years, 71.8% had smoking exposure and 74.4% had an ECOG 0-1; 64.1% were adenocarcinoma and 35.9% squamous cell carcinoma. Concerning staging, when anti-PD-L1 was started, 64.1% had stage IV disease, 33.3% stage III and 2.6% stage IIB. Twenty-five (67.6%) patients expressed PD-L1 of ≥ 1% and 45.8% had driver mutations. After diagnosis, the median time until atezolizumab was 12 (7-21) months, and median duration of atezolizumab was 2 (1-5) months. Disease control was obtained in 15.4% of the cases, 69.2% had disease progression and in 15.4% of patients the response was not assessed. The median overall survival was 18 (10-26) months and was longer in cases with disease control (p=0.001), for patients not-stage IV at diagnosis (p=0.023), and lower for patients with multiple metastatic lesions (p=0.024). Adjusting survival for age, histology, smoking, constitutional symptoms, ECOG and disease control; disease control was the only factor whose association remained statistically significant. Conclusion: In our series, only 15% of patients had a clear benefit with atezolizumab as second-line therapy. Stabilization and disease response were the only markers of survival benefit.

Introduction: Cell-free DNA (cfDNA) analysis offers a non-invasive method to identify sensitising and resistance mutations in advanced Non-Small Cell Lung Cancer (NSCLC) patients. Next-generation sequencing (NGS) of circulating free DNA (cfDNA) is a valuable tool for mutations detection and disease′s clonal monitoring. Material and methods: An amplicon-based targeted gene NGS panel was used to analyse 101 plasma samples of advanced non-small cell lung cancer (NSCLC) patients with known oncogenic mutations, mostly EGFR mutations, serially collected at different clinically relevant time points of the disease. Results: The variant allelic frequency (VAF) monitoring in consecutive plasma samples demonstrated different molecular response and progression patterns. The decrease in or the clearance of the mutant alleles was associated with response and the increase in or the emergence of novel alterations with progression. At the best response, the median VAF was 0% (0.0% to 3.62%), lower than that at baseline, with a median of 0.53% (0.0% to 9.9%) (p = 0.004). At progression, the VAF was significantly higher (median 4.67; range: 0.0–36.9%) than that observed at the best response (p = 0.001) and baseline (p = 0.006). These variations anticipated radiographic changes in most cases, with a median time of 0.86 months. Overall, the VAF evolution of different oncogenic mutations predicts clinical outcomes. Conclusion: The targeted NGS of circulating tumour DNA (ctDNA) has clinical utility to monitor treatment response in patients with advanced lung adenocarcinoma.

Artificial intelligence (AI)-based solutions have revolutionized our world, using extensive datasets and computational resources to create automatic tools for complex tasks that, until now, have been performed by humans. Massive data is a fundamental aspect of the most powerful AI-based algorithms. However, for AI-based healthcare solutions, there are several socioeconomic, technical/infrastructural, and most importantly, legal restrictions, which limit the large collection and access of biomedical data, especially medical imaging. To overcome this important limitation, several alternative solutions have been suggested, including transfer learning approaches, generation of artificial data, adoption of blockchain technology, and creation of an infrastructure composed of anonymous and abstract data. However, none of these strategies is currently able to completely solve this challenge. The need to build large datasets that can be used to develop healthcare solutions deserves special attention from the scientific community, clinicians, all the healthcare players, engineers, ethicists, legislators, and society in general. This paper offers an overview of the data limitation in medical predictive models; its impact on the development of healthcare solutions; benefits and barriers of sharing data; and finally, suggests future directions to overcome data limitations in the medical field and enable AI to enhance healthcare. This perspective is dedicated to the technical requirements of the learning models, and it explains the limitation that comes from poor and small datasets in the medical domain and the technical options that try or can solve the problem related to the lack of massive healthcare data.