Risk prediction of pleural effusion in lung malignancy patients treated with CT-guided percutaneous microwave ablation: a nomogram and artificial neural network model

Applications of artificial intelligence in interventional oncology: An up-to-date review of the literature

Interventional oncology provides image-guided therapies, including transarterial tumor embolization and percutaneous tumor ablation, for malignant tumors in a minimally invasive manner. As in other medical fields, the application of artificial intelligence (AI) in interventional oncology has garnered significant attention. This narrative review describes the current state of AI applications in interventional oncology based on recent literature. A literature search revealed a rapid increase in the number of studies relevant to this topic recently. Investigators have attempted to use AI for various tasks, including automatic segmentation of organs, tumors, and treatment areas; treatment simulation; improvement of intraprocedural image quality; prediction of treatment outcomes; and detection of post-treatment recurrence. Among these, the AI-based prediction of treatment outcomes has been the most studied. Various deep and conventional machine learning algorithms have been proposed for these tasks. Radiomics has often been incorporated into prediction and detection models. Current literature suggests that AI is potentially useful in various aspects of interventional oncology, from treatment planning to post-treatment follow-up. However, most AI-based methods discussed in this review are still at the research stage, and few have been implemented in clinical practice. To achieve widespread adoption of AI technologies in interventional oncology procedures, further research on their reliability and clinical utility is necessary. Nevertheless, considering the rapid research progress in this field, various AI technologies will be integrated into interventional oncology practices in the near future.

Safety and Effectiveness of Track Cauterization for Lung Cryoablation

PURPOSE

To evaluate the safety and effectiveness of track cauterization for lung cryoablation through comparison of postprocedural adverse event (AE) rates.

MATERIALS AND METHODS

Fifty-nine patients who underwent 164 percutaneous lung cryoablation procedures between 2013 and 2018 were included in this retrospective study. The study cohort was subdivided by whether track cauterization was conducted or not at the end of the procedure. The study cohort was also subdivided by the number of probes (1-2 probes vs 3-4 probes). Postablation AE rates were assessed by immediate and delayed (at 1 month or later) AEs, pneumothorax, hemothorax, pleural effusion, and whether intervention was required. Univariate and multivariate logistic regression analyses were used to compare differences in AE rates.

RESULTS

Patients who underwent procedures with track cautery were 2.6 times less likely to exhibit pleural effusion (P = .017). Patients who underwent procedures conducted with a higher number of probes were 3.8 times more likely to receive interventions (P < .001), 1.6 times more likely to experience pneumothorax (P = .037), and 2.1 times more likely to experience pleural effusion (P = .003). History of lung surgery, increased number of probes, size of the probe, and absence of track cautery were noted to be significant predictors of AEs and need for interventions (all P < .05).

CONCLUSIONS

Track cauterization in lung cryoablation was proven to reduce pleural effusion, but no difference in pneumothorax or delayed AEs was noted. The use of fewer probes was associated with a lower rate of AEs.

Stereotactic ablative brachytherapy versus percutaneous microwave ablation as salvage treatments for lung oligometastasis from colorectal cancer

BACKGROUND

The treatment for lung oligometastasis from colorectal cancer (CRC) remains challenging. This retrospective study aimed to compare the local tumor control, survival and procedure-related complications in CRC patients undergoing low-dose rate stereotactic ablative brachytherapy (L-SABT) versus percutaneous microwave ablation (MWA) for lung oligometastasis.

METHODS

Patients between November 2017 and December 2020 were retrospectively analyzed. Local tumor progression-free survival (LTPFS) and overall survival (OS) were analyzed in the entire cohort as well as by stratified analysis based on the minimal ablation margin (MAM) around the tumor.

RESULTS

The final analysis included 122 patients: 74 and 48 in the brachytherapy and MWA groups, respectively, with a median follow-up of 30.5 and 35.3 months. The 1- and 3-year LTPFS rate was 54.1% and 40.5% in the brachytherapy group versus 58.3% and 41.7% in the MWA group (P = 0.524 and 0.889, respectively). The 1- and 3-year OS rate was 75.7% and 48.6% versus 75.0% and 50.0% (P = 0.775 and 0.918, respectively). Neither LTPFS nor OS differed significantly between the patients with MAM of 5-10 mm versus > 10 mm. Pulmonary complication rate did not differ in the overall analysis, but was significantly higher in the MWA group in the subgroup analysis that only included patients with lesion within 10 mm from the key structures (P = 0.005). The increased complications was primarily bronchopleural fistula.

CONCLUSIONS

Considering the caveats associated with radioisotope use in L-SABT, MWA is generally preferable. In patients with lesion within 10 mm from the key pulmonary structures, however, L-SABT could be considered as an alternative due to lower risk of bronchopleural fistula.

Computed tomography-guided microwave ablation for right middle lobe pulmonary nodules: a retrospective, single-center, case-control study

PURPOSE

This retrospective, single-center, case-control study evaluated the safety and efficacy of Computed tomography (CT)-guided microwave ablation (MWA) for pulmonary nodules located in the right middle lobe (RML), a challenging location associated with a high frequency of complications.

METHODS

Between May 2020 and April 2022, 71 patients with 71 RML pulmonary nodules underwent 71 MWA sessions. To comparison, 142 patients with 142 pulmonary nodules in non-RML were selected using propensity score matching. The technical success, technique efficacy, complications, and associated factors were analyzed. The duration of the procedure and post-ablation hospital stay were also recorded.

RESULTS

Technical success was achieved in 100% of all patients. There were no significant differences in technique efficacy rates between the RML and non-RML groups (97.2% vs. 95.1%, p = 0.721). However, both major (47.9% vs. 19.7%, p < 0.001) and minor (26.8% vs. 11.3%, p = 0.004) pneumothorax were more common in the RML group than non-RML group. MWA for RML pulmonary nodules was identified as an independent risk factor for pneumothorax (p < 0.001). The duration of procedures (51.7 min vs. 35.3 min, p < 0.001) and post-ablation hospital stays (4.7 days vs. 2.8 days, p < 0.001) were longer in the RML group than non-RML group.

CONCLUSIONS

CT-guided MWA for RML pulmonary nodules showed comparable efficacy compared with other lobes, but posed a higher risk of pneumothorax complications, necessitating longer MWA procedure times and extended hospital stays.

Innovative Invasive Loco-Regional Techniques for the Treatment of Lung Cancer

Surgical resection is still the standard treatment for early-stage lung cancer. A multimodal treatment consisting of chemotherapy, radiotherapy and/or immunotherapy is advised for more advanced disease stages (stages IIb, III and IV). The role of surgery in these stages is limited to very specific indications. Regional treatment techniques are being introduced at a high speed because of improved technology and their possible advantages over traditional surgery. This review includes an overview of established and promising innovative invasive loco-regional techniques stratified based on the route of administration, including endobronchial, endovascular and transthoracic routes, a discussion of the results for each method, and an overview of their implementation and effectiveness.

Development and validation a simple scoring system to identify malignant pericardial effusion

Background

Malignant pericardial effusion (MPE) is a serious complication in patients with advanced malignant tumors, which indicates a poor prognosis. However, its clinical manifestations lack specificity, making it challenging to distinguish MPE from benign pericardial effusion (BPE). The aim of this study was to develop and validate a scoring system based on a nomogram to discriminate MPE from BPE through easy-to-obtain clinical parameters.

Methods

In this study, the patients with pericardial effusion who underwent diagnostic pericardiocentesis in Taizhou Hospital of Zhejiang Province from February 2013 to December 2021 were retrospectively analyzed. The eligible patients were divided into a training group (n = 161) and a validation group (n = 66) according to the admission time. The nomogram model was established using the meaningful indicators screened by the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression. Then, a new scoring system was constructed based on this nomogram model.

Results

The new scoring system included loss of weight (3 points), no fever (4 points), mediastinal lymph node enlargement (2 points), pleural effusion (6 points), effusion adenosine deaminase (ADA≦18U/L) (5 points), effusion lactate dehydrogenase (LDH&gt;1033U/L) (7 points), and effusion carcinoembryonic antigen (CEA&gt;4.9g/mL) (10 points). With the optimal cut-off value was 16 points, the area under the curve (AUC), specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), negative likelihood ratio (NLR) for identifying MPE were 0.974, 95.1%, 91.0%, 85.6%, 96.8%, 10.56 and 0.05, respectively, in the training set and 0.950, 83.3%, 95.2%, 90.9%, 90.9%, 17.50, and 0.18, respectively, in the validation set. The scoring system also showed good diagnostic accuracy in differentiating MPE caused by lung cancer from tuberculous pericardial effusion (TPE) and MPE including atypical cell from BPE.

Conclusion

The new scoring system based on seven easily available variables has good diagnostic value in distinguishing MPE from BPE.

Local progression after computed tomography-guided microwave ablation in non-small cell lung cancer patients: prediction using a nomogram model

OBJECTIVES

To develop an effective nomogram model for predicting the local progression after computed tomography-guided microwave ablation (MWA) in non-small cell lung cancer (NSCLC) patients.

METHODS

NSCLC patients treated with MWA were randomly allocated to either the training cohort or the validation cohort (4:1). The predictors of local progression identified by univariable and multivariable analyses in the training cohort were used to develop a nomogram model. The C-statistic was used to evaluate the predictive accuracy in both the training and validation cohorts.

RESULTS

A total of 304 patients (training cohort: n = 250; validation cohort: n = 54) were included in this study. The predictors selected into the nomogram for local progression included the tumor subtypes (odds ratio [OR], 2.494; 95% confidence interval [CI], 1.415-4.396, p = 0.002), vessels ≥3 mm in direct contact with tumor (OR, 2.750; 95% CI, 1.263-5.988; p = 0.011), tumor diameter (OR, 2.252; 95% CI, 1.034-4.903; p = 0.041) and location (OR, 2.442; 95% CI, 1.201-4.965; p = 0.014). The C-statistic showed good predictive performance in both cohorts, with a C-statistic of 0.777 (95% CI, 0.707-0.848) internally and 0.712 (95% CI, 0.570-0.855) externally (training cohort and validation cohort, respectively). The optimal cutoff value for the risk of local progression was 0.39.

CONCLUSIONS

Tumor subtypes, vessels ≥3 mm in direct contact with the tumor, tumor diameter and location were predictors of local progression after MWA in NSCLC patients. The nomogram model could effectively predict the risk of local progression after MWA. Patients showing a high risk (>0.39) on the nomogram should be monitored for local progression.