Importance:Comprehensive clinical data stratified by microsatellite instability (MSI) status in gastroesophageal adenocarcinoma (GEA) are limited. Objective:To characterize treatment pathways and clinical outcomes following immune checkpoint inhibitor (ICI) therapy in patients with deficient mismatch repair (dMMR) or MSI-high (MSI-H) GEA. Design, Setting, and Participants:This retrospective cohort study was conducted from June 2020 to August 2025 at a single tertiary cancer center in Texas. Of 3316 patients diagnosed with GEA, 1638 who underwent MSI testing were included. Exposures:ICI-based therapy with nivolumab plus ipilimumab or with pembrolizumab. Main Outcomes and Measures:Incidence of dMMR, overall survival, best objective response, clinical or pathological complete response (cCR/pCR), and clinicopathological factors associated with ICI response. Results:Among the 1638 patients included, the mean (SD) age was 62.5 (12.8) years, and 1204 (73.5%) were male. There were 83 individuals (5.1%) with dMMR tumors. The 3-year overall survival rate for patients with dMMR was 84.7% with and 68.5% without curative-intent surgery vs 72.3% with and 32.3% without curative-intent surgery for patients with proficient mismatch repair. Among the 83 patients with dMMR, 34 had metastatic and 49 had locoregional disease; 30 and 25 patients in each group received ICI-based therapy, respectively, with 16 of 26 evaluable patients (61.5%) achieving best objective response and 12 of 23 (52.2%) achieving pCR/cCR, respectively (both considered ICI responsive). Among the 25 locoregional cases treated with ICI, 10 proceeded to surgery, with all achieving residual tumor (R0) resection and 3 (30%) demonstrating pCR. Of the remaining patients, 9 achieved cCR and were transitioned to active surveillance. For patients who achieved cCR or pCR, response was reached after 2 to 4 cycles of nivolumab plus ipilimumab or 6 to 12 cycles of pembrolizumab, and none experienced recurrence or death (median [IQR] follow-up after achieving cCR or pCR, 26.0 [16.3-40.0] months). Individuals who were ICI responsive and nonresponsive did not significantly differ in human epidermal growth factor receptor 2 (ERBB2; formerly HER2 or HER2/neu) status, combined positive score, or claudin-18, isoform 2 (CLDN18.2), expression, whereas Lynch syndrome was more common in responders. Conclusions and Relevance:In this single-center cohort of dMMR GEA, ICI-based therapy demonstrated high response rates in both metastatic and locoregional settings. The durable responses in locoregional disease suggest the potential to consider nonoperative management in selected individuals. Further research is needed to improve prediction of ICI response and refine treatment strategies.

BACKGROUND:Gastroesophageal junction (GEJ) adenocarcinoma with signet ring cells (SRCs) is associated with resistance to therapy and poor outcomes. Standard neoadjuvant regimens for SRCs have been chemoradiation (nCRT) or chemotherapy (nCT), with little data specific to SRCs indicating which is optimal. We compared outcomes of resectable (stage I-III) GEJ SRC patients treated with nCRT or nCT followed by surgery. METHODS:Patients undergoing esophagectomy for GEJ SRCs after neoadjuvant therapy at 2 different institutions (2005-2022) were stratified by receipt of nCRT or nCT. Propensity scores, based on demographic and clinical characteristics, were used to match groups (2 nCRT:1 nCT). Outcomes were assessed using Kaplan-Meier and multivariable Cox regression models. RESULTS:In sum, 104 (66.7%) nCRT and 52 (33.3%) nCT patients were included. Most patients were male (nCRT, n = 88 of 104, 84.6%; nCT, n = 43 of 52, 82.7%, P = .76) with clinical stage III (nCRT, n = 92 of 104, 88.5%; nCT, n = 48 of 52, 92.3%, P = .33). Patients treated with nCRT achieved pathologic complete response 11.5% vs 7.7% for nCT patients. At median follow-up of 27 months, median survival for nCRT patients was 34.3 months (95% CI, 24.9-72.1) compared with 34.5 months (95% CI, 29.0-58.8; P = 0.28) for nCT patients. Multivariate Cox regression indicated that lymphovascular invasion was an independent predictor of progression-free and overall survival, but choice of neoadjuvant therapy was not. CONCLUSIONS:The treatment-resistant nature of SRCs pose consequential challenges that are not overcome either by nCRT or nCT. Novel therapeutic strategies are needed to improve outcomes for patients with GEJ adenocarcinoma with SRC.

Esophageal cancer (EC) is a highly lethal malignancy with geographic, histologic, and biologic heterogeneity. Advances in endoscopic therapy, multimodality treatment, biomarker-driven systemic therapy, and supportive care have reshaped contemporary management. This review summarizes current approaches to diagnosis, staging, and treatment of EC, with emphasis on stage-, histology-, and biomarker-directed decision making within a multidisciplinary framework. This is a narrative review of clinical guidelines, landmark trials, and recent evidence informing the management of esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC) across disease stages, including evolving systemic therapies and supportive care strategies. Early-stage EC is increasingly managed with endoscopic resection and ablative therapies, enabling esophageal preservation in carefully selected patients. Locally advanced disease requires multimodality treatment tailored by resectability and histology. Neoadjuvant chemoradiation followed by surgery remains a standard approach for resectable ESCC, whereas perioperative systemic therapy is favored for many patients with EAC due to its higher risk of early distant dissemination. Adjuvant immunotherapy improves disease-free survival in patients with residual disease after neoadjuvant chemoradiation and resection. In unresectable or metastatic EC, biomarker-driven strategies incorporating immune checkpoint inhibitors and targeted agents have improved outcomes compared with chemotherapy alone. Supportive care interventions addressing nutrition, symptom control, and psychosocial needs are essential throughout the disease course. Optimal management of EC requires individualized, multidisciplinary care integrating tumor stage, histology, molecular characteristics, and patient fitness. Continued advances in systemic therapy, immunotherapy, and supportive care are refining treatment paradigms with the goal of improving survival while preserving quality of life.

BACKGROUND:Recent trials demonstrate comparable outcomes between sublobar resection (SLR) and lobar resection (LR) for stage IA1-2 (≤2 cm) non-small cell lung cancer (NSCLC). Conversely, historical studies reported more recurrences with SLR than with LR for stage IA3 NSCLC (>2 cm, ≤3 cm), although these investigations predated diagnostic and therapeutic advancements. We aimed to compare SLR and LR outcomes in these patients in the contemporary era. METHODS:Patients assigned to clinical stage IA3 NSCLC undergoing resection at a single institution (2015-2024) were stratified by wedge resection, anatomic segmentectomy, or LR. Patients with neoadjuvant therapy receipt and neuroendocrine tumors were excluded. Outcomes were evaluated by Kaplan-Meier and multivariable Cox regression models. RESULTS:We identified 260 individuals: 40 (15.4%) underwent wedge resection, 67 (25.8%) segmentectomy, and 153 (58.8%) LR. Most patients were male (wedge, n = 14/40 [35.0%]; segmentectomy, n = 30/67 [45.0%]; LR, n = 65/153 [42.5%]; P = .59). Groups differed in median number of lymph nodes harvested (wedge, 6 [4-8]; segmentectomy, 11 [7-15]; LR, 14 [9.5-18.5]; P < .01). Recurrences were more frequent after wedge resection (wedge, n = 11/40 [27.5%]; segmentectomy, n = 11/67 [16.4%]; LR, n = 18/153 [11.8%]; P = .04). At median follow-up (42.8 months), 3-year disease-free survival for wedge resection patients was 76.8% (95% CI, 63.7-92.7) compared with 82.7% (95% CI, 73.4-93.1) for segmentectomy patients and 90.9% (95% CI, 85.8-96.2; P = .03) for LR patients. CONCLUSIONS:LR and segmentectomy achieved comparable outcomes in stage IA3 NSCLC, whereas wedge resection was associated with higher recurrence and inferior disease-free survival. These data contribute to discussions about the role of SLR in treating this subgroup of patients.

PURPOSE:Somatic-type malignancy (SM) of germ cell tumor (GCT) is rare but demonstrates aggressive behavior. Differences by primary site, time to transformation, and mutational status remain poorly characterized. PATIENTS AND METHODS:We reviewed all SM patients between June 2016 and May 2025 at a tertiary referral center. Time to somatic transformation (TST) was defined from the initial GCT diagnosis to SM detection. SM were classified based on time of detection- at initial diagnosis (de novo), at consolidative surgery, at relapse within 5 years, and evolved SM at relapse after 5 years. Descriptive statistics, Kaplan-Meier estimates, and Cox regression analyses were used. RESULTS:72 patients were identified: 40 (56%) with testicular, 24 (33%) with mediastinal, and 8 (11%) with other primaries. 22/24 (92%) mediastinal tumors with SM had sarcomatous transformation. Sarcoma was seen in 25/42 (60%) of de novo SM while adenocarcinoma was detected as an evolved entity in 5/6 (83.3%) cases. Embryonic type neuroectodermal tumor (ENET) histology carried poor prognosis compared to non-ENET histology (median, OS 1.8 vs 8 years; HR, 1.94; P = 0.1). Genomic data available showed PTEN-AKT-mTOR pathway mutations (33%) and TP53 mutations (33%), enriched in extra-gonadal sarcomatous SM. CONCLUSIONS:SM exhibits temporal, histologic, and molecular distinctions across primary sites. Response to frontline therapy and survival outcomes are poor. Future direction includes the exploration of underlying predictors of transformation and identification of targetable alterations in the relapsed setting.

BACKGROUND:Understanding patients' symptom recovery after upper gastrointestinal (UGI) cancer surgery is essential for patient-centered care, yet detailed longitudinal patient-reported outcome data remain limited. We conducted a prospective study using the novel MD Anderson Symptom Inventory for UGI Surgery (MDASI-UGI-Surg) tool. METHODS:Patients undergoing esophageal (n = 42), gastric (n = 27), or pancreatic (n = 74) cancer surgery from February to September 2024 were enrolled. The MDASI-UGI-Surg comprises 22 symptoms and 6 interference items. The five most severe symptoms and three most severe interference items on postoperative day (POD) 3 were identified. Recovery was defined as achieving mild severity for both symptom and interference composites. Multivariable analysis was performed to identify factors associated with recovery at postoperative month (POM) 1. RESULTS:Most symptoms peaked at POD3. The five most severe symptoms were pain, fatigue, sleep disturbance, drowsiness, and dry mouth; the top three interference items were general activity, working, and enjoyment of life. Symptom recovery followed three phases: an acute improvement phase (POD3-14), a plateau phase (POD14-POM1), and a persistent recovery phase extending to POM6. Symptom profiles were similar across organ groups, and fatigue remained prolonged. Cumulative recovery rates were 64.8% at POM1, 78.9% at POM3, and 90.8% at POM6. Postoperative complications and readmissions were associated with delayed recovery at POM1, and the type of surgery predicted recovery in multivariable models. CONCLUSIONS:This study provides a detailed characterization of symptom recovery after UGI cancer surgery. These findings support improved preoperative counseling and postoperative care planning.

Background Accurate prediction of pathological complete response (pCR) after preoperative chemoradiation therapy, followed by surgery (trimodality therapy) in esophageal adenocarcinoma (EAC) and gastroesophageal junction adenocarcinoma (GEJAC) may improve clinical decision-making and patient counseling before esophagectomy. This study aimed to develop predictive models for pCR after trimodality using machine learning (ML) approaches. Patients and Methods A total of 569 patients with EAC and GEJAC who received trimodality therapy at MD Anderson Cancer Center between 2002 and 2022 were included. Clinicopathological characteristics and survival benefit of patients who achieved a pCR were reviewed via descriptive and survival analyses. Subsequently, ML models based on clinical variables were employed to predict pCR, including BART, random forest, and XGBoost, logistic regression, and LASSO. Results pCR was achieved in 132 patients (23.2%). Poorly differentiated tumors, tumors with signet ring cell component, higher T stage, higher clinical stage, residual tumor on biopsy after chemoradiation, and higher SUVmax on positron emission tomography-contract tomography (PET-CT) after chemoradiation were significantly associated with non-pCR. pCR patients had significantly longer overall survival (OS) and relapse free survival (RFS) compared with non-pCR patients (median OS, 10.40 versus 4.42 years, log-rank p = 0.0041; median RFS, 10.40 versus 2.35 years, log-rank p < 0.0001). The random forest model showed the highest predictive ability for pCR with an AUC value of 0.702 among the employed models. Conclusions This first exploratory study supports the validity and potential utility of ML-based models for predicting pCR after trimodality therapy in EAC and GEJAC. Further validation is warranted before clinical application.

Objective:The relationship between imaging and pathologic findings after systemic therapy in stage IV non-small cell lung cancer with pleural involvement remains unclear. Morphologic and metabolic imaging changes may predict pleural clearance and inform further treatment decisions. Methods:We reviewed 24 patients with stage IV non-small cell lung cancer and initial pleural involvement who underwent subsequent pulmonary resection between 2016 and 2025. Baseline and preoperative computed tomography and positron emission tomography-computed tomography scans were assessed for primary, nodal, metastatic, and pleural disease. Residual pleural disease (RPD) was defined by pathologically or cytologically confirmed tumor; undetectable pleural disease (UPD) was defined by absence of gross disease or negative tissue/cytology when available. Major pathologic response (≤10% viable tumor) was assessed in the primary tumor. Results:Seventeen patients (70.8%) had UPD and 7 (29.2%) had RPD. Patients with UPD had greater rates of major pathologic response in the primary tumor (52.9% vs 0%), complete radiologic response of the primary tumor (64.7% vs 0%, P = .014), complete nodal radiologic response (94.1% vs 5.9%, P = .001), and complete metabolic response in both primary (88.2% vs 20.0%, P = .009) and nodal sites (88.2% vs 11.8%, P = .021). Radiologic improvement in pleural disease did not differ between patients with UPD or RPD (52.9% vs 42.9%, P = .652); however, complete metabolic response was more frequent in patients with UPD (64.7% vs 16.7%, P = .069). Conclusions:Complete radiologic or metabolic response of the primary and nodal sites strongly correlates with UPD at surgery. Residual pleural disease on imaging should not preclude consideration for surgical consolidation. These findings underscore the potential utility of refined, evidence-based imaging criteria to guide multidisciplinary decision-making.

Background Marker-negative germ cell tumors (GCT) account for up to 40% of metastatic disease, making treatment monitoring and surveillance challenging. The utility of circulating tumor DNA (ctDNA) in teratoma and somatic-type malignancies (SM) remains undefined. Methods In this retrospective study, the GCT database at a tertiary referral center was queried to identify patients with available ctDNA reports and evidence of radiological progression without tumor marker elevation, or normal pre-orchiectomy tumor markers. ctDNA, stratified as either positive or negative was studied in conjunction with radiology and histology status to determine concordance. Serial trends of mean tumor molecules/ml across various therapeutic interventions and relapse were also investigated. Results Between December 2024 and February 2026, 34 patients with marker-negative germ cell tumors (GCTs) and 70 ctDNA assessments were identified. The most common histology was seminoma (15, 44%), followed by somatic-type malignancy (SM) (6, 18%). Pure teratoma was present in 3 patients (9%). ctDNA was undetectable in 35/38 (92.1%) assessments of response and detectable in 23/25 (92.0%) assessments of progression. One patient each with neuroendocrine somatic transformation and growing teratoma syndrome progressed with undetectable ctDNA. Conclusion ctDNA demonstrates high concordance with disease status in marker-negative non-teratomatous GCT, with potential utility in SM.

BACKGROUND & AIMS:Endoscopic eradication therapy is standard of care for T1a esophageal adenocarcinoma. There are limited data regarding long-term outcomes of endoscopic eradication therapy for T1b esophageal adenocarcinoma. The aim of this study was to assess clinical outcomes after endoscopic management of low-risk and high-risk T1b esophageal adenocarcinoma via a pooled analysis of individual patient-level data. METHODS:A pooled analysis of studies reporting endoscopic management of T1b esophageal adenocarcinoma was conducted. Inclusion required endoscopic resection revealing T1b esophageal adenocarcinoma with negative deep margins. High-risk T1b esophageal adenocarcinoma was defined by presence of: submucosal 2/3 or deep (>500 μm) submucosal invasion, lymphovascular invasion, or poor grade of differentiation. Clinical outcomes included all-cause and esophageal adenocarcinoma-related mortality, intraluminal recurrence, and extraluminal metastases. RESULTS:Of 418 studies identified, 7 with 216 patients were included (84% men; 94% White). The median follow-up was 48.1 months (interquartile range, 25.5-68.1 months). High-risk T1b esophageal adenocarcinoma was observed in 114 patients (53%). Baseline clinical characteristics were mostly similar between groups. Although the rates of intraluminal recurrence in the low-risk and high-risk groups were comparable (15% vs 22%; P = .30), the high-risk group had a 5-fold significantly higher risk of extraluminal metastases (11% vs 2%; P = .01). Both all-cause and esophageal adenocarcinoma-related mortality were significantly higher in the high-risk group (log rank P = .04). CONCLUSIONS:In this international pooled analysis, rates of extraluminal metastases and esophageal adenocarcinoma-related mortality were very low in patients with low-risk T1b esophageal adenocarcinoma, supporting the role of endoscopic eradication therapy in this group. However, outcomes were significantly worse for the high-risk T1b group, suggesting careful patient selection and shared decision-making in this population.

BACKGROUND:Financial toxicity (FT) refers to the financial stress and detrimental impact on quality of life experienced by patients due to treatment cost. In patients with resected lung cancer (LC), we sought to identify those at risk of developing moderate or severe ("major") FT using machine learning (ML) techniques based on preoperative characteristics. STUDY DESIGN:Patients who underwent LC resection at a single center between January 2016 and December 2021 were surveyed to ascertain demographic information, financial data, and presence of major FT. Clinicopathologic variables were extracted from a prospective database. Patients were randomly divided into training and test sets. First, we identified the most informative features. Then, 4 ML algorithms (decision tree, random forest, gradient boosting, and extreme gradient boosting) were trained. We ensembled the 4 models' predictions to optimize the model. RESULTS:There were 1,477 patients identified, of whom 462 (31.3%) completed the survey. Forty-six patients (10.0%) experienced major FT. The variables most influential in our models included age, race and ethnicity, smoking status, household income, credit score, marital and employment status, size of residence, BMI, histology, extent of resection, and preoperative forced expiratory volume in 1 second. The ensemble model yielded an accuracy of 0.86, precision of 0.93, and sensitivity of 0.86, leading to an F1 score of 0.88, indicative of a reliable algorithm. CONCLUSIONS:ML algorithms can accurately identify patients at risk of experiencing major FT after LC surgery. Preoperatively identifying patients with cancer vulnerable to financial stress may allow an opportunity for intervention to address downstream cost considerations.