Antonella Martino

Background: Circulating tumor DNA (ctDNA), shed into bodily fluids by cancer cells through apoptosis, necrosis, or active secretion, is currently used in the field of genomic investigation in clinical settings, primarily for advanced stages of non-small-cell lung cancer (NSCLC). However, its potential role in guiding the multi-omic approach to early-stage NSCLC is emerging as a promising area of investigation. Efforts are being made to integrate the genomics not only in surgery, but also in the definition of long-term prognosis after surgical or radiotherapy and for the prediction of recurrence. Methods: An extensive literature search was conducted on PubMed, covering publications from 2000 to 2024. Using the advanced search tool, titles and abstracts were filtered based on the following keywords: ctDNA, early stage, NSCLC. From this search, 20 studies that fulfilled all inclusion criteria were selected for analysis in this review. Results: This review highlights the growing body of evidence supporting the potential clinical use of ctDNA as a genomic biomarker in managing early-stage NSCLC. Baseline ctDNA levels offer valuable information about tumor molecular biology and histological characteristics. Beyond its prognostic value before treatment, liquid biopsy has proven useful for tracking minimal residual disease and forecasting recurrence following curative interventions such as surgery or radiotherapy. Future adjuvant treatment decisions may increasingly rely on predictive models that incorporate liquid biopsy findings alongside other clinical factors. Conclusions: The potential use of this analyte introduces new opportunities for the integration of genomic data in treatment, as well as relapse monitoring with more accurate and innovative than traditional methods, particularly in patients with early-stage NSCLC

Background: Circulating tumor DNA (ctDNA) may be released from neoplastic cells into biological fluids through apoptosis, necrosis, or active release. In patients with non-small-cell lung cancer (NSCLC), ctDNA analysis is being introduced in clinical practice only for advanced disease management. Nevertheless, an interesting and promising field of application is the analysis of ctDNA in the management of early stage non-small-cell lung cancer, both for evaluation before treatment, such as diagnosis and screening, and for prediction of histology or pathological features. Methods: A thorough review of the literature published between 2000 and 2024 was performed on PubMed, utilizing the advanced search feature to narrow down titles and abstracts containing the following keywords: ctDNA, early stage, and NSCLC. A total of 20 studies that met all inclusion criteria were chosen for this review. Results: In this review, we summarize the increasing evidence suggesting that ctDNA has potential clinical applications in the management of patients with early stage NSCLC. ctDNA levels in early stage cancers are very low, posing many technical challenges in improving the detection rate and sensitivity, especially in clinical practice, if it is to be implemented for early detection. Presently, the main limitation of ctDNA experimental and clinical studies, especially in early stage settings, is the lack of definitive standardization and consensus regarding methodology, the absence of systematically validated analyses, and the lack of adoption of sensitive approaches. Conclusions: Possible applications of this analyte open up new fields of diagnosis, treatment, and follow up, which are less invasive and more precise than other approaches currently in use, especially in early stage NSCLC patients.

Artificial Intelligence (AI) has revolutionized the management of non-small-cell lung cancer (NSCLC) by enhancing different aspects, including staging, prognosis assessment, treatment prediction, response evaluation, recurrence/prognosis prediction, and personalized prognostic assessment. AI algorithms may accurately classify NSCLC stages using machine learning techniques and deep imaging data analysis. This could potentially improve precision and efficiency in staging, facilitating personalized treatment decisions. Furthermore, there are data suggesting the potential application of AI-based models in predicting prognosis in terms of survival rates and disease progression by integrating clinical, imaging and molecular data. In the present narrative review, we will analyze the preliminary studies reporting on how AI algorithms could predict responses to various treatment modalities, such as surgery, radiotherapy, chemotherapy, targeted therapy, and immunotherapy. There is robust evidence suggesting that AI also plays a crucial role in predicting the likelihood of tumor recurrence after surgery and the pattern of failure, which has significant implications for tailoring adjuvant treatments. The successful implementation of AI in personalized prognostic assessment requires the integration of different data sources, including clinical, molecular, and imaging data. Machine learning (ML) and deep learning (DL) techniques enable AI models to analyze these data and generate personalized prognostic predictions, allowing for a precise and individualized approach to patient care. However, challenges relating to data quality, interpretability, and the ability of AI models to generalize need to be addressed. Collaboration among clinicians, data scientists, and regulators is critical for the responsible implementation of AI and for maximizing its benefits in providing a more personalized prognostic assessment. Continued research, validation, and collaboration are essential to fully exploit the potential of AI in NSCLC management and improve patient outcomes. Herein, we have summarized the state of the art of applications of AI in lung cancer for predicting staging, prognosis, and pattern of recurrence after treatment in order to provide to the readers a large comprehensive overview of this challenging issue.

BACKGROUND: The aim of this study is to analyze the prognostic factors in patients that underwent induction therapy and surgery for clinical stage III NSCLC. METHODS: Clinical and pathological characteristics of stage III NSCLC patients for N2 involvement that underwent neoadjuvant treatment (NAD) and surgery from 1/01/1998 to 31/12/2017 were collected and retrospectively analyzed. Tumor characteristics, yClinical, yPathological stage and lymph node characteristics were correlated to Overall Survival (OS). RESULTS: The analysis was conducted on 180 patients. Five-year OS (5YOS) was 50.9%. Univariable analysis results revealed old age (p = 0.003), clinical N2 post-NAD (p = 0.01), pneumonectomy (0.005), persistent pathological N2 (p = 0.039, HR 1.9, 95% CI 1.09–2.68) and adjuvant therapy absence (p = 0.049) as significant negative prognostic factors. Multivariable analysis confirmed pN0N1 (p = 0.02, HR 0.29, 95% CI 0.13–0.62) as a favorable independent prognostic factor and adjuvant therapy absence (p = 0.012, HR 2.61, 95% CI 1.23–5.50) as a negative prognostic factor. Patients with persistent N2 presented a 5YOS of 35.3% vs. 55.8% in pN0N1 patients. Regarding lymph node parameters, the lymph node ratio (NR) significantly correlated with OS: 5YOS of 67.6% in patients with NR < 50% vs. 29.5% in NR > 50% (p = 0.029). CONCLUSION: Clinical response aided the stratification of prognosis in patients that underwent multimodal treatment for stage III NSCLC. Adjuvant therapy seemed to be an important option in these patients, while node ratio was a strong prognosticator in patients with persistent nodal involvement.

The Cox Proportional Hazards regression is among the most widely used models in clinical and epidemiological research for investigating the association between time-to-event outcomes and multiple predictors, that, in the modern perspective of personalized medicine, tend to belong to ever wider spheres relating to the patient and his medical condition. When the goal is to include a large number of variables in a prediction model, feature selection techniques are often required to ensure a certain level of interpretability of the results and federated learning is necessary to recruit in the study the sufficient number of patients for reliable model outcomes, overcoming the main problems of data privacy and ownership. In this regard, we here propose an adaptation for federated learning of the optimization algorithm of the Cox Proportional Hazards regression model with LASSO regularization as feature selector and we demonstrate the efficacy of our algorithm on real and simulated data sets in a simulated distributed environment with no patient-level data sharing by comparing its model parameter estimation performances with its centralised version.