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 was to evaluate the Masaoka–Koga and the tumor node metastases (TNM) staging system in thymic epithelial tumors (TET) considering possible improvements. Methods: We reviewed the data of 379 patients who underwent surgical resection for TET from 1 January 1985 to 1 January 2018, collecting and classifying the pathological report according to the Masaoka–Koga and the TMN system. The number of involved organs was also considered as a possible prognostic factor and integrated in the two staging systems to verify its impact. Results: Considering the Masaoka–Koga system, 5- and 10-year overall survival (5–10YOS) was 96.4% and 88.9% in stage I, 95% and 89.5% in stage II and 85.4% and 72.8% in stage III (p = 0.01), with overlapping in stage I and stage II curves. Considering the TNM system, 5–10YOS was 95.5% and 88.8% in T1, 84.8% and 70.7% in T2 and 88% and 76.3% in T3 (p = 0.02), with overlapping T2–T3 curves. Including the number of involved structures, in Masaoka–Koga stage III, patients with singular involved organs had a 100% and 76.6% vs. 87.7% 5–10YOS, which was 76.6% in patients with multiple organ infiltration. Considering the TNM, T3 patients with singular involved structures presented a 5–10YOS of 100% vs. 62.5% and 37.5% in patients with multiple organ involvement (p = 0.07). Conclusion: The two staging systems present limitations due to overlapping curves in early Masaoka–Koga stages and in advanced T stages for TNM. The addition of the number of involved organs seems to be a promising factor for the prognosis stratification in these patients.

Chest Ultrasonography (chest US) has shown good sensibility in detecting pneumothorax, pleural effusions and peripheral consolidations and it can be performed bedside. The aim of the study was to analyze agreement between chest US and chest X-ray in patients who have undergone thoracic surgery and discuss cases of discordance. Patients undergoing thoracic surgery were retrospectively selected. Patients underwent routinely Chest X-ray (CXR) during the first 48 h after surgery. Chest US have been routinely performed in all selected patients in the same date of CXR. Chest US operators were blind to both reports and images of CXR. Ultrasonographic findings regarding pneumothorax (PNX), subcutaneous emphysema (SCE), lung consolidations (LC), pleural effusions (PE) and hemi-diaphragm position were collected and compared to corresponding CXR findings. Inter-rater agreement between two techniques was determined by Cohen’s kappa-coefficient. Twenty-four patients were selected. Inter-rater agreement showed a moderate magnitude for PNX (Cohen’s Kappa 0.5), a slight/fair magnitude for SCE (Cohen’s Kappa 0.21), a fair magnitude for PE (Cohen’s Kappa 0.39), no agreement for LCs (Cohen’s Kappa 0.06), high levels of agreement for position of hemi-diaphragm (Cohen’s Kappa 0.7). Analysis of agreement between chest X-ray and chest US showed that ultrasonography is able to detect important findings for surgeons. Limitations and advantages have been found for both chest X-ray and chest US. Knowing the limits of each one is important to really justify and optimize the use of ionizing radiations.

Background: The use of ¹⁸F FDG PET/CT scan in thymic epithelial tumours (TET) has been reported in the last two decades, but its application in different clinical settings has not been clearly defined. Methods: We performed a pictorial review of pertinent literature to describe different roles and applications of this imaging tool to manage TET patients. Finally, we summarized future prospects and potential innovative applications of PET in these neoplasms. Results: ¹⁸FFDG PET/CT scan may be of help to distinguish thymic hyperplasia from thymic epithelial tumours but evidences are almost weak. On the contrary, this imaging tool seems to be very performant to predict the grade of malignancy, to a lesser extent pathological response after induction therapy, Masaoka Koga stage of disease and long-term prognosis. Several other radiotracers have some application in TETs but results are limited and almost controversial. Finally, the future of PET/CT and theranostics in TETs is still to be defined but more detailed analysis of metabolic data (such as texture analysis applied on thymic neoplasms), along with promising preclinical and clinical results from new “stromal PET tracers”, leave us an increasingly optimistic outlook. Conclusions: PET plays different roles in the management of thymic epithelial tumours, and its applications may be of help for physicians in different clinical settings.

Background: The usefulness of ¹⁸FDG PET/CT scan in the evaluation of thymic epithelial tumours (TETs) has been reported by several authors, but data are still limited and its application in clinical practice is far from being defined. Methods: We performed a narrative review of pertinent literature in order to clarify the role of ¹⁸FDG PET/CT in the prediction of TET histology and to discuss clinical implications and future perspectives. Results: There is only little evidence that ¹⁸FDG PET/CT scan may distinguish thymic hyperplasia from thymic epithelial tumours. On the other hand, it seems to discriminate well thymomas from carcinomas and, even more, to predict the grade of malignancy (WHO classes). To this end, SUVmax and other PET variables (i.e., the ratio between SUVmax and tumour dimensions) have been adopted, with good results. Finally, however promising, the future of PET/CT and theranostics in TETs is far from being defined; more robust analysis of imaging texture on thymic neoplasms, as well as new exploratory studies with “stromal PET tracers,” are ongoing. Conclusions: PET may play a role in predicting histology in TETs and help physicians in the management of these insidious malignancies.

Background: Outcomes for locally advanced NSCLC with pathological complete response (pCR), i.e., pT0N0 after induction chemoradiotherapy (IT), have been seldom investigated. Herein, long-term results, in this highly selected group of patients, have been evaluated with the aim to identify prognostic predictive factors. Methods: Patients affected by locally advanced NSCLC (cT1-T4/N0-2/M0) who underwent IT, possibly following surgery, from January 1992 to December 2019, were considered for this retrospective analysis. Survival rates and prognostic factors have been studied with Kaplan-Meier analysis, log-rank and Cox regression analysis. Results: Three-hundred and forty-three consecutive patients underwent IT in the considered period. Out of them, 279 were addressed to surgery; among them, pCR has been observed in 62 patients (18% of the total and 22% of the operated patients). In the pCR-group, clinical staging was IIb in 3 (5%) patients, IIIa in 28 (45%) patients and IIIb in 31 (50%). Surgery consisted of (bi)lobectomy in the majority of cases (80.7%), followed by pneumonectomy (19.3%). Adjuvant therapy was administered in 33 (53.2%) patients. Five-year overall survival and disease-free survival have been respectively 56.18% and 48.84%. The relative risk of death, observed with the Cox regression analysis, was 4.4 times higher (95% confidence interval (CI): 1.632–11.695, p = 0.03) for patients with N2 multi-station disease, 2.6 times higher (95% CI: 1.066–6.407, p = 0.036) for patients treated with pneumonectomy and 3 times higher (95% CI: 1.302–6.809, p = 0.01) for patients who did not receive adjuvant therapy. Conclusions: Rewarding long-term results could be expected in locally advanced NSCLC patients with pCR after IT followed by surgery. Baseline N2 single-station disease and adjuvant therapy after surgery seem to be associated with better prognosis, while pneumonectomy is associated with poorer outcomes.