Comparison of robotic-assisted minimally invasive esophagectomy versus minimally invasive esophagectomy: A propensity-matched study from a single high-volume institution.

OBJECTIVE: Robotic-assisted minimally invasive esophagectomy accounts for a growing proportion of esophagectomies, potentially due to improved technical capabilities simplifying the challenging aspects of standard minimally invasive esophagectomy. However, there is limited evidence directly comparing both operations. The objective is to evaluate the short-term and long-term outcomes of robotic-assisted minimally invasive esophagectomy in comparison with the minimally invasive esophagectomy approach for patients with esophageal cancer over a 7-year period at a high-volume center. The primary end points of this study were overall survival and disease-free survival. Secondary end points included operation-specific morbidity, lymph node yield, readmission status, and in-hospital, 30-day, and 90-day mortality. METHODS: Patients who underwent robotic-assisted minimally invasive esophagectomy or standard minimally invasive esophagectomy over a 7-year period were identified from a prospectively maintained database. Inclusion criteria were patients with stage I to III disease, operations performed past the learning curve, and no evidence of scleroderma or cirrhosis. A 1:3 propensity match (robotic-assisted minimally invasive esophagectomy:minimally invasive esophagectomy) for multiple clinical covariates was performed to identify the final study cohort. Perioperative outcomes were compared between the 2 operations. RESULTS: A total of 734 patients undergoing minimally invasive esophagectomy (n = 630) or robotic-assisted minimally invasive esophagectomy (n = 104) for esophageal cancer were identified. After exclusions and matching, a total cohort of 246 patients undergoing robotic-assisted minimally invasive esophagectomy (n = 65) or minimally invasive esophagectomy (n = 181) were identified. There was no difference in overall survival (P = .69) or disease-free survival (P = .70). There were no significant differences in rates of major morbidity: pneumonia (17% vs 17%, P = .34), chylothorax (8% vs 9%, P = .95), recurrent laryngeal nerve injury (0% vs 1.5%, P = 1), anastomotic leak (5% vs 4%, P = .49), intraoperative complications (9% vs 8%, P = .73), or complete resection rates (99% vs 96%, P = .68). There was no difference in in-hospital (P = .89), 30-day (P = .66) or 90-day mortality (P = .73) between both cohorts. The robotic-assisted minimally invasive esophagectomy cohort yielded a higher median lymph node harvest in comparison with the minimally invasive esophagectomy cohort (32 vs 29, P = .02). CONCLUSIONS: Robotic-assisted minimally invasive esophagectomy may improve lymphadenectomy in patients undergoing esophagectomy for cancer. Minimally invasive esophagectomy and robotic-assisted minimally invasive esophagectomy are otherwise associated with similar mortality, morbidity, and perioperative outcomes. Further prospective study is required to investigate whether improved lymph node resection may translate to improved oncologic outcomes.

Predicting benign, preinvasive, and invasive lung nodules on computed tomography scans using machine learning.

OBJECTIVE: The study objective was to investigate if machine learning algorithms can predict whether a lung nodule is benign, adenocarcinoma, or its preinvasive subtype from computed tomography images alone. METHODS: A dataset of chest computed tomography scans containing lung nodules was collected with their pathologic diagnosis from several sources. The dataset was split randomly into training (70%), internal validation (15%), and independent test sets (15%) at the patient level. Two machine learning algorithms were developed, trained, and validated. The first algorithm used the support vector machine model, and the second used deep learning technology: a convolutional neural network. Receiver operating characteristic analysis was used to evaluate the performance of the classification on the test dataset. RESULTS: The support vector machine/convolutional neural network-based models classified nodules into 6 categories resulting in an area under the curve of 0.59/0.65 when differentiating atypical adenomatous hyperplasia versus adenocarcinoma in situ, 0.87/0.86 with minimally invasive adenocarcinoma versus invasive adenocarcinoma, 0.76/0.72 atypical adenomatous hyperplasia + adenocarcinoma in situ versus minimally invasive adenocarcinoma, 0.89/0.87 atypical adenomatous hyperplasia + adenocarcinoma in situ versus minimally invasive adenocarcinoma + invasive adenocarcinoma, and 0.93/0.92 atypical adenomatous hyperplasia + adenocarcinoma in situ + minimally invasive adenocarcinoma versus invasive adenocarcinoma. Classifying benign versus atypical adenomatous hyperplasia + adenocarcinoma in situ + minimally invasive adenocarcinoma versus invasive adenocarcinoma resulted in a micro-average area under the curve of 0.93/0.94 for the support vector machine/convolutional neural network models, respectively. The convolutional neural network-based methods had higher sensitivities than the support vector machine-based methods but lower specificities and accuracies. CONCLUSIONS: The machine learning algorithms demonstrated reasonable performance in differentiating benign versus preinvasive versus invasive adenocarcinoma from computed tomography images alone. However, the prediction accuracy varies across its subtypes. This holds the potential for improved diagnostic capabilities with less-invasive means.

Intrapleural interleukin-2-expressing oncolytic virotherapy enhances acute antitumor effects and T-cell receptor diversity in malignant pleural disease.

OBJECTIVE: The mainstay of treatment for patients with malignant pleural disease is fluid drainage and systemic therapy. A tumor-specific oncolytic virus or T-cell-activating interleukin-2 immunotherapy may provide an opportunity for local control. We previously developed a vaccinia virus-expressing interleukin-2, an oncolytic virus that mediated tumor regression in preclinical peritoneal tumor models with expansion of tumor-infiltrating lymphocytes. We evaluated the antitumor efficacy and immune modulatory effects of vaccinia virus-expressing interleukin-2 in malignant pleural disease. METHODS: A murine model of malignant pleural disease was established with percutaneous intrapleural deposition of the Lewis lung carcinoma cell line and monitored with bioluminescent imaging. After intrapleural or systemic administration of vaccinia viruses (vaccinia virus yellow fluorescent protein control, vaccinia virus-expressing interleukin-2), systemic anti-programmed cell death-1 antibody, or combination therapy (vaccinia virus-expressing interleukin-2 and anti-programmed cell death-1), tumor mass, immune cell infiltration, T-cell receptor diversity, and survival were assessed. RESULTS: Intrapleural vaccinia virus resulted in significant tumor regression compared with phosphate-buffered saline control (P < .05). Inclusion of the interleukin-2 transgene further increased intratumoral CD8+ T cells (P < .01) and programmed cell death-1 expression on CD8+ tumor-infiltrating lymphocytes (P < .001). Intrapleural vaccinia virus-expressing interleukin-2 was superior to systemic vaccinia virus-expressing interleukin-2, with reduced tumor burden (P < .0001) and improved survival (P < .05). Intrapleural vaccinia virus-expressing interleukin-2 alone or combined treatment with systemic anti-programmed cell death-1 reduced tumor burden (P < .01), improved survival (P < .01), and increased intratumoral αß T-cell receptor diversity (P < .05) compared with systemic anti-programmed cell death-1 monotherapy. CONCLUSIONS: Intrapleural vaccinia virus-expressing interleukin-2 reduced tumor burden and enhanced survival in a murine malignant pleural disease model. Increased CD8+ tumor-infiltrating lymphocytes and αß T-cell receptor diversity are associated with enhanced response. Clinical trials will enable assessment of intrapleural vaccinia virus-expressing interleukin-2 therapy in patients with malignant pleural disease.

Adjuvant chemotherapy for pulmonary sarcomatoid carcinoma: A retrospective analysis of the National Cancer Database.

OBJECTIVES: Pulmonary sarcomatoid carcinoma (PSC) is a rarely occurring variant of non-small cell lung cancer with sarcoma-like features. Compared with traditional non-small cell lung cancer, PSC patients typically present later and have poorer prognoses, irrespective of stage. The standard of care is resection, but guidelines for the use of adjuvant chemotherapy have not been established. To advance the development of evidence-based management algorithms for PSC after resection, a statistical analysis on a nationwide representative sample of patients was performed. METHODS: A retrospective cohort study was performed by querying the National Cancer Database for patients with a diagnosis of PSC between 2004 and 2015. Patients who received complete anatomical resection with or without adjuvant chemotherapy were included. Multivariable regression was used to detect factors associated with the receipt of adjuvant chemotherapy. Multivariable Cox regression of overall survival and Kaplan-Meier survival analysis on propensity-matched groups was conducted to study the association between adjuvant chemotherapy and prognosis. RESULTS: We included 1497 patients with PSC in the final analysis. Factors associated with receiving adjuvant chemotherapy were age, histology, and receipt of adjuvant radiation. The results of multivariable Cox analysis and Kaplan-Meier analysis on propensity matched groups yielded similar trends: adjuvant chemotherapy was associated with improved 5-year overall survival for stage II and III disease, but not for stage I disease. CONCLUSIONS: Multiple factors are associated with receipt of adjuvant chemotherapy for PSC, and this treatment appears to be associated with improved survival in stage II and stage III, but not stage I patients.

Racial Disparity in Surgical Therapy for Thymic Malignancies.

BACKGROUND: The primary curative treatment for thymic malignancies is surgery. For lung and esophageal cancer, substantive disparities in outcomes by race exist. Many of these disparities are attributed to the decreased use of surgery in non-White patients. Although thymic malignancies are treated by the same specialists as lung and esophageal cancer, it is unknown if there are racial disparities in the treatment of thymic malignancies. RESEARCH QUESTION: Do racial disparities exist in the surgical treatment of thymic malignancies? STUDY DESIGN AND METHODS: A retrospective cohort analysis was performed using the National Cancer Data Base of patients diagnosed with thymoma and thymic carcinoma between 2004 and 2016. Univariate comparisons of demographics were compared using χ 2 and rank-sum tests. Multivariable analysis was performed to determine if race was an independent variable associated with receiving surgical resection. Preoperative and postoperative care was compared between races. RESULTS: Seven thousand four hundred eighty-nine patients met inclusion criteria. Four thousand nine hundred sixty-two (66%) were White, 1,311 (18%) were Black, 487 (7%) were Hispanic, 580 (8%) were Asian or Pacific Islander, and 143 (2%) were other races. Black patients with thymic malignancies were more likely to have a median income < $38,000 and not received surgery. Black and Hispanic patients had the lowest median age (54.3 and 53.6 years, respectively) and were most likely to be uninsured (8.2% and 12.5%, respectively). White patients received surgical therapy 1 week sooner and had a postoperative length of stay 1.5 days shorter than Black patients. Multivariable analysis controlling for age, sex, tumor size, insurance status, comorbidity score, histology, and facility type showed that race remained independently associated with the receipt of surgical resection. White patients had the greatest likelihood of receiving surgery with Black patients being least likely to receive surgery (OR, 0.60). INTERPRETATION: A racial disparity exists in surgical therapy for thymic malignancies.

Early Distant Recurrence in Patients With Resected Stage I Lung Cancer: A Case Series of "Blast Metastasis".

BACKGROUND: The standard of care in the management of stage I non-small-cell lung cancer (NSCLC) has been anatomic lung resection with multistation lymph node sampling of ≥ 10 lymph nodes. The 5-year survival for NSCLC has ranged from 73% to 93% (for stage IB and stage IA, respectively) and will be more favorable for patients with fewer comorbidities and those with a higher state of premorbid functioning and who undergo surgical resection. Despite the positive prognosis for operable stage I NSCLC, a subset of patients will develop metastatic disease within as few as 12 months after resection. Using an institutional database, we have presented the data from 68 patients who had developed distant metastatic recurrence after resection of pathologic stage I NSCLC within 1 year after surgery. PATIENTS AND METHODS: A retrospective study was conducted of a prospectively maintained intuitional database. The final cohort included patients with pathologic stage I NSCLC who had undergone anatomic resection but had subsequently presented with multiple sites of distant recurrence within 1 year. The study period extended from 2003 to 2020. Patients with broad local recurrence or recurrence at a single distant site were excluded. Kaplan-Meier analysis was used to estimate the 5-year survival. RESULTS: A total of 2827 patients had undergone surgical resection for stage I NSCLC during the 17-year period and 68 met the criteria for inclusion. Most of the patients (n = 48) were smokers, and the dominant histologic type was adenocarcinoma (n = 37). After recurrence, 22 patients (33%) had undergone chemoradiotherapy and 19 (28%) had received chemotherapy alone. The mean and median overall survival were 23.7 and 14 months, respectively. The 5-year survival from recurrence and surgery were both 13.2%. CONCLUSIONS: Limited data are available on the risk factors for early metastasis after resected stage I NSCLC. The results from our cohort have demonstrated poor survival after recurrence. These data might be the basis for determining a phenotype for patients prone to early widespread metastasis despite seemingly curative surgical resection.

Factors Associated With Survival in Complete Pathologic Response Non-Small Cell Lung Cancer.

OBJECTIVES: There is a strong association with improved survival for patients with non-small cell lung cancer (NSCLC) who have developed a pathological complete response (pCR) after neoadjuvant therapy. A national database was used to investigate factors associated with long-term survival in this cohort of patients. PATIENTS: Retrospective review was completed of the National Cancer Database of patients who obtained pCR and had neoadjuvant therapy for stage I to stage III NSCLC between 2004 and 2014. All patients had neoadjuvant chemotherapy with or without radiation therapy. METHODS: Univariate and multivariable analysis was performed on factors associated with overall survival (OS), including gender, clinical stage, and nodal count. Patients were divided into 2 groups based on the Commission on Cancer-recommended median number of lymph nodes (LNs) examined: 0 to 9 LNs and ≥10 LNs. Chi-square and Wilcoxon rank-sum tests were used to compare patient, hospital, and clinical variables between groups. RESULTS: Increased age (hazard ratio [HR] 1.02, 95% confidence interval [CI], 1.00-1.03), neoadjuvant radiation therapy (HR 1.48, 95% CI, 1.10-2.00), and pneumonectomy (HR 1.64, 95% CI, 1.22-2.22) were associated with worse survival in the 759-patient cohort. Multivariable regression demonstrated having ≥10 nodes harvested (HR 0.71, 95% CI, 0.56-0.89) was associated with improved survival as was every increase in LN harvest up to 17 LNs. No significant differences in 5-year OS were found between clinical stage I, II, and III, respectively (66.1% vs. 60.9% vs. 58.6%, P = .288). CONCLUSION: This study shows that younger age, increasing LN harvest, female sex, the absence of neoadjuvant radiation therapy and non-pneumonectomy resections are all associated with improved OS in patients with NSCLC who have developed pCR.

Node-Positive Segmentectomy for Non-Small-Cell Lung Cancer: Risk Factors and Outcomes.

BACKGROUND: Segmentectomy for well-selected early stage non-small-cell lung carcinoma (NSCLC) has been shown to have similar oncologic outcomes and survival to lobectomy. However, these data are based on the presumption that the disease is node negative. Few data exist regarding the risk factors for and the outcomes of patients with disease treated with segmentectomy that is found to be node positive. We sought to determine the risk factors for and outcomes of clinical stage I NSCLC patients who are treated with segmentectomy but are determined to be node positive. PATIENTS AND METHODS: We queried patients with clinical stage I NSCLC ≤ 3 cm within the National Cancer Data Base between 2004 and 2014 who were treated with segmentectomy or lobectomy and found to have positive nodes. Kaplan-Meier curves with log-rank tests were used to compare overall survival (OS) between segmentectomy and lobectomy. For comparison only, segmentectomy patients with pathologically node-negative disease were identified to determine predictors of node positivity after segmentectomy via multivariable logistic regression. RESULTS: A total of 4556 patients with node-positive disease were identified, comprising 115 segmentectomy patients and 4441 lobectomy patients. Multivariable analysis identified increasing tumor size, squamous-cell histology, and increasing number lymph nodes sampled as significant predictors of node positivity after segmentectomy. There was no difference in OS between segmentectomy and lobectomy, with 3-year OS rates of 66.3% and 68.1%, respectively (P = .723). CONCLUSION: There are discrete risk factors for discovering positive nodes after segmentectomy. Segmentectomy is associated with similar OS compared to lobectomy for clinical stage I NSCLC found to be node positive.