Background Lung adenocarcinoma (ADC) is the main cause of death related to lung cancer. The aim of this study was to identify poor prognostic factors for overall survival (OS) in patients with stage IV lung ADC in real‐world clinical practice. Methods Patients were selected from the Surveillance Epidemiology and End Results (SEER) database. Chi‐square bivariate analysis was used for the association of binary qualitative variables. A multivariate Cox regression analysis was performed to determine the impact of these prognostic factors on OS. Results A total of 46 030 patients were included (51.3% men, mean age 67.03 ± 11.6), of whom 41.3% presented with metastases in bone, 28.9% in brain, 17.1% in liver and 31.8% in lung. Patients with liver metastases presented with two or more metastatic sites more frequently than patients without liver metastases ( P < 0.001). Male sex (HR 0.78, 95% CI: 0.76–0.80), age ≥ 65 years (HR 1.37, 95% CI: 1.33–1.40), lack of family support (HR 0.80, 95% CI: 0.78–0.81) and presence of liver (HR 1.45, 95% CI: 1.40–1.50), bone (HR 1.21, 95% CI: 1.18–1.24) or brain metastases (HR 1.18, 95% CI: 1.15–1.21) were identified as poor prognostic factors for OS. Patients with liver metastasis showed the highest hazard ratio value ( P < 0.001). Conclusions The presence of liver metastases was the worst prognostic factor for patients with metastatic lung ADC. This factor should be considered as a stratification factor for future studies evaluating new cancer treatments including immunotherapy. Key points Significant findings of the study Regression analysis identified poor prognostic factors for overall survival. Factors were male sex, age ≥ 65 years, lack of family support and presence of liver, bone and brain metastases. Patients with liver metastasis showed the highest HR (HR = 1.45 95% CI: 1.40–1.50). This study included the highest number of adenocarcinoma patients analyzed so far ( N = 46 030). What this study adds The presence of liver metastases should be considered as a stratification factor for future studies evaluating new cancer treatments including immunotherapy.

This work demonstrates that clinical and inherent tumor characteristics define a subset of patients with gastrointestinal stromal tumor (GIST) with increased likelihood to achieve durable response to first-line imatinib therapy. Patients reaching ≥5 years on imatinib have a greater chance of remaining progression free over time, although the disease is unlikely to be cured. Imatinib is well tolerated for >5 years, and emergent toxicities are overall manageable. Resistance to imatinib emerging in patients with GISTs after long-term imatinib treatment does not involve polyclonal expansion of KIT secondary mutations.

No abstract

characteristics and treatment. All cases were re-reviewed by an expert adjudicator. Accuracy and concordance between automated and manual methods are reported. Results: Automated extraction required significantly less time (<72 hours) than manual extraction (225 person-hours). Collection of demographic data (age, sex, diagnosis) was highly accurate and concordant with both methods, (96-100%). Accuracy and concordance were lower for unstructured data elements in EHR, such as ECOG performance status, date of stage IV diagnosis and smoking status (automated accuracy: 94%, 93%, 88% respectively; manual accuracy: 83%, 78% and 94%). Detection of biomarker testing was highly accurate and concordant (96-98%), although detection of final results was more variable (accuracy 84-100%, concordance 84-99%). Automated extraction identified metastatic sites more accurately than manual (concordance 70-99%), with the exception of lymph node metastasis (automated 66%, manual 92%, concordance 58%), due to use of analogous terms in radiology reports not included in the gold standard definition. Concurrent medications (86-100%) and comorbid conditions (96-100%), were reported with high accuracy and concordance. Treatment details were also accurately captured with both methods (84-100%) and highly concordant (83-99%). Conclusions: Automated data abstraction from unstructured EHR is highly accurate and faster than manual abstraction. Key challenges include poorly structured EHR and use of analogous terms beyond the gold standard definition. Legal entity responsible for the study: The authors.

No abstract