Search

Background: Few studies have examined the performance of artificial intelligence (AI) content detection in scientific writing. This study evaluates the performance of publicly available AI content detectors when applied to both human-written and AI-generated scientific articles. Methods: Articles published in Annals of Surgical Oncology (ASO) during the year 2022, as well as AI-generated articles using OpenAI's ChatGPT, were analyzed by three AI content detectors to assess the probability of AI-generated content. Full manuscripts and their individual sections were evaluated. Group comparisons and trend analyses were conducted by using ANOVA and linear regression. Classification performance was determined using area under the curve (AUC). Results: A total of 449 original articles met inclusion criteria and were evaluated to determine the likelihood of being generated by AI. Each detector also evaluated 47 AI-generated articles by using titles from ASO articles. Human-written articles had an average probability of being AI-generated of 9.4% with significant differences between the detectors. Only two (0.4%) human-written manuscripts were detected as having a 0% probability of being AI-generated by all three detectors. Completely AI-generated articles were evaluated to have a higher average probability of being AI-generated (43.5%) with a range from 12.0 to 99.9%. Conclusions: This study demonstrates differences in the performance of various AI content detectors with the potential to label human-written articles as AI-generated. Any effort toward implementing AI detectors must include a strategy for continuous evaluation and validation as AI models and detectors rapidly evolve. [ABSTRACT FROM AUTHOR]

Background: Cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) improves survival in select patients with peritoneal metastases (PM), but the impact of social determinants of health on CRS/HIPEC outcomes remains unclear. Patients and Methods: A retrospective review was conducted of a multi-institutional database of patients with PM who underwent CRS/HIPEC in the USA between 2000 and 2017. The area deprivation index (ADI) was linked to the patient's residential address. Patients were categorized as living in low (1–49) or high (50–100) ADI residences, with increasing scores indicating higher socioeconomic disadvantage. The primary outcome was overall survival (OS). Secondary outcomes included perioperative complications, hospital/intensive care unit (ICU) length of stay (LOS), and disease-free survival (DFS). Results: Among 1675 patients 1061 (63.3%) resided in low ADI areas and 614 (36.7%) high ADI areas. Appendiceal tumors (n = 1102, 65.8%) and colon cancer (n = 322, 19.2%) were the most common histologies. On multivariate analysis, high ADI was not associated with increased perioperative complications, hospital/ICU LOS, or DFS. High ADI was associated with worse OS (median not reached versus 49 months; 5 year OS 61.0% versus 28.2%, P < 0.0001). On multivariate Cox-regression analysis, high ADI (HR, 2.26; 95% CI 1.13–4.50; P < 0.001), cancer recurrence (HR, 2.26; 95% CI 1.61–3.20; P < 0.0001), increases in peritoneal carcinomatosis index (HR, 1.03; 95% CI 1.01–1.05; P < 0.001), and incomplete cytoreduction (HR, 4.48; 95% CI 3.01–6.53; P < 0.0001) were associated with worse OS. Conclusions: Even after controlling for cancer-specific variables, adverse outcomes persisted in association with neighborhood-level socioeconomic disadvantage. The individual and structural-level factors leading to these cancer disparities warrant further investigation to improve outcomes for all patients with peritoneal malignancies. [ABSTRACT FROM AUTHOR]

Background: The AJCC 8th edition stratifies stage IV disseminated appendiceal cancer (dAC) patients based on grade and pathology. This study was designed to externally validate the staging system and to identify predictors of long-term survival. Methods: A 12-institution cohort of dAC patients treated with CRS ± HIPEC was retrospectively analyzed. Overall survival (OS) and recurrence-free survival (RFS) were analyzed by using Kaplan-Meier and log-rank tests. Univariate and multivariate cox-regression was performed to assess factors associated with OS and RFS. Results: Among 1009 patients, 708 had stage IVA and 301 had stage IVB disease. Median OS (120.4 mo vs. 47.2 mo) and RFS (79.3 mo vs. 19.8 mo) was significantly higher in stage IVA compared with IVB patients (p < 0.0001). RFS was greater among IVA-M1a (acellular mucin only) than IV M1b/G1 (well-differentiated cellular dissemination) patients (NR vs. 64 mo, p = 0.0004). Survival significantly differed between mucinous and nonmucinous tumors (OS 106.1 mo vs. 41.0 mo; RFS 46.7 mo vs. 21.2 mo, p < 0.05), and OS differed between well, moderate, and poorly differentiated (120.4 mo vs. 56.3 mo vs. 32.9 mo, p < 0.05). Both stage and grade were independent predictors of OS and RFS on multivariate analysis. Acellular mucin and mucinous histology were associated with better OS and RFS on univariate analysis only. Conclusions: AJCC 8th edition performed well in predicting outcomes in this large cohort of dAC patients treated with CRS ± HIPEC. Separation of stage IVA patients based on the presence of acellular mucin improved prognostication, which may inform treatment and long-term, follow-up strategies. [ABSTRACT FROM AUTHOR]

Background: Cancer incidence is expected to increase in coming decades, disproportionately so among minoritized communities. Racially and ethnically concordant care is essential to addressing disparities in cancer outcomes within at-risk groups. Here, we assess trends in racial and ethnic representation of medical students (MS), general surgery (GS) residents, and complex general surgical oncology (CGSO) fellows. Methods: This is a retrospective review of data from the American Association of Medical Colleges and the Accreditation Council of Medical Education (ACGME) from 2015 to 2020. Self-reported race and ethnicity was obtained for MS, GS, and CGSO trainees. Race and ethnicity proportions were compared with respective representation in the 2020 US Census. Mann–Kendall, Wilcoxon rank sum, and linear regression were used to assess trends, as appropriate. Results: A total of 316,448 MS applicants, 128,729 MS matriculants, 27,574 GS applicants, 46,927 active GS residents, 710 CGSO applicants, and 659 active CGSO fellows were included. With every progressive stage in training, there was a smaller proportion of URM active trainees than applicants. Further, URM, Hispanic/Latino, and Black/African American trainees were significantly underrepresented compared with 2020 Census data. While the proportion of White CGSO fellows increased over time (54.5–69.2%, p = 0.009), the proportion of Black/African American and Hispanic/Latino (URM) CGSO fellows did not significantly change over the study period, though URM representation was lower in 2020 as compared with 2015. Discussion: From 2015 to 2020, minority representation decreased at every advancing stage in surgical oncology training. Efforts to address barriers for URM applicants to CGSO fellowships are needed. [ABSTRACT FROM AUTHOR]

Early postoperative mortality risk prediction is crucial for clinical management of gastric cancer. This study aims to predict 90-day mortality in gastric cancer patients undergoing gastrectomy using automated machine learning (AutoML), optimize models for preoperative prediction, and identify factors influential in prediction. National Cancer Database was used to identify stage I–III gastric cancer patients undergoing gastrectomy between 2004 and 2016. 26 features were used to train predictive models using H2O.ai AutoML. Performance on validation cohort was measured. In 39,108 patients, 90-day mortality rate was 8.8%. The highest performing model was an ensemble (AUC = 0.77); older age, nodal ratio, and length of inpatient stay (LOS) following surgery were most influential for prediction. Removing the latter two parameters decreased model performance (AUC 0.71). For optimizing models for preoperative use, models were developed to first predict node ratio or LOS, and these predicted values were inputted for 90-day mortality prediction (AUC of 0.73–0.74). AutoML performed well in predicting 90-day mortality in a larger cohort of gastric cancer patients that underwent gastrectomy. These models can be implemented preoperatively to inform prognostication and patient selection for surgery. Our study supports broader evaluation and application of AutoML to guide surgical oncologic care. [ABSTRACT FROM AUTHOR]

This paper explores the transformative potential of Large Language Models (LLMs) within the context of surgical oncology and outlines the foundational mechanisms behind these models. LLMs, such as GPT-4, have rapidly evolved in terms of scale and capabilities, with profound implications for their applications in healthcare. These models, rooted in the Generative Pretrained Transformer architecture, exhibit advanced natural language understanding and generation skills. Within surgical oncology, LLMs, when integrated into a Generalist Medical AI (GMAI) framework, hold great promise in offering real-time support throughout the cancer journey. However, alongside these opportunities, this paper underscores the importance of ethical, privacy, and efficacy considerations, especially in light of issues like data drift and potential biases. Collaborative efforts among healthcare providers, AI developers, and regulatory bodies are pivotal in ensuring responsible and effective use of LLMs in surgical oncology, thereby contributing to enhanced patient care and safety. As LLMs continue to advance, they are poised to become indispensable tools in the delivery of high-quality, efficient care in this specialized medical field. [ABSTRACT FROM AUTHOR]

Background: Pathologic response to preoperative chemotherapy predicts survival in patients with colorectal liver metastases (CLMs) who undergo hepatectomy. In multiple CLMs, mixed pathologic response, wherein tumors exhibit different degrees of treatment response, is possible. We sought to evaluate survival outcomes of mixed response in patients with multiple CLMs. Methods: We conducted a retrospective cohort study using a single-institution database of patients with two or more CLMs who underwent preoperative chemotherapy and hepatectomy (2010–2018). Pathologic response of each tumor was measured on pathology. Patients were stratified by pathologic response as complete (pCR) = 0–1% viability; major (pMajR) = 2–49% viability; minor (pMinR) = 50–99% viability; or mixed (pMixR) = at least one pCR/MajR tumor and one pMinR. Recurrence-free survival (RFS) and overall survival (OS) were estimated using the Kaplan–Meier method, and adjusted risk of death was evaluated using Cox regression. Results: Among 444 patients, 6% had pCR, 34% had pMajR, 36% had pMinR, and 24% had pMixR. Median and 5-year RFS for patients with pMixR was 10.4 months and 16%, respectively, compared with pMajR (11.3 months and 18%, respectively), pMinR (7.7 months and 13%, respectively), and pCR (23.1 months and 38%, respectively) [log-rank p < 0.001]. Median and 5-year OS for patients with pMixR was 77.4 months and 60%, respectively, compared with pMajR (80.5 months and 63%, respectively), pMinR (49.9 months and 39%, respectively), and pCR (median OS not reached; median follow-up of 37.1 months and 5-year OS of 65%) [log-rank p = 0.002]. pMixR was associated with a 52% risk of death reduction (hazard ratio 0.48, 95% confidence interval 0.30–0.78 vs. pMinR). Conclusions: One-quarter of patients with multiple CLMs have pMixR following preoperative chemotherapy and hepatectomy. OS and RFS for patients with pMixR mirror those of pMajR rather than pMinR, suggesting the greatest response achieved in any metastasis best predicts survival. [ABSTRACT FROM AUTHOR]

Background: Socioeconomic- and demographic-based disparities exist in the treatment of pancreatic adenocarcinoma (PDAC). Medicaid expansion (ME) may have an impact on these disparities. Analyses of patients with PDAC from the National Cancer Database (NCDB) were performed to examine the impact of ME on access to treatment and outcomes. Methods: Patients with non-metastatic PDAC diagnosed between 2006 and 2016 were identified. Multiple logistic regression analyses were performed to evaluate factors associated with curative-intent surgical resection, multimodal therapy, treatment at a high-volume facility (HVF), and survival. Results: The study identified 41,876 patients who met the criteria. Medicaid expansion was independently associated with curative-intent resection (odds ratio [OR] 1.54; 95 % confidence interval [CI] 1.43–1.67; p < 0.001). In a multivariable analysis, ME was independently associated with multimodal therapy (OR 1.60; 95 % CI 1.44–1.76; p < 0.001) and treatment at an HVF (OR 1.57; 95 % CI 1.42–1.74; p < 0.001). Medicaid expansion was independently associated with improved 30-day mortality (OR 0.49; 95 % CI 0.34–0.79) and 90-day mortality (OR 0.48 95 % CI 0.35–0.59). Cox regression analysis demonstrated that after adjustment for other variables, ME status was associated with improved overall survival (hazard ratio [HR], 0.82; 95 % CI 0.73–0.90; p < 0.001). Conclusions: Medicaid expansion is associated with increased use of care processes that improve outcomes in PDAC, operative outcomes, and overall survival. The study data suggest that ME has helped to improve disparities in PDAC in ME states. [ABSTRACT FROM AUTHOR]

Background: An increasing number of patients with cancer diagnoses and prior SARS-CoV-2 infection will require surgical treatment. The objective of this study was to determine whether a history of SARS-CoV-2 infection increases the risk of adverse postoperative events following surgery in patients with cancer. Methods: This was a propensity-matched cohort study from April 6, 2020 to October 31, 2020 at the UT MD Anderson Cancer Center. Cancer patients were identified who underwent elective surgery after recovering from SARS-CoV-2 infection and matched to controls based on patient, disease, and surgical factors. Primary study outcome was a composite of the following adverse postoperative events that occurred within 30 days of surgery: death, unplanned readmission, pneumonia, cardiac injury, or thromboembolic event. Results: A total of 5682 patients were included for study, and 114 (2.0%) had a prior SARS-CoV-2 infection. The average time from infection to surgery was 52 (range 20–202) days. Compared with matched controls, there was no difference in the rate of adverse postoperative outcome (14.3% vs. 13.4%, p = 1.0). Patients with a SARS-CoV-2-related inpatient admission before surgery had increased odds of postoperative complication (adjusted odds ratio [aOR] 7.4 [1.6–34.3], p = 0.01). Conclusions: A minimal wait time of 20 days after recovering from minimally symptomatic SARS-CoV-2 infection appears to be safe for cancer patients undergoing low-risk elective surgery. Patients with SARS-CoV-2 infections requiring inpatient treatment were at increased risk for adverse events after surgery. Additional wait time may be required in those with more severe infections. [ABSTRACT FROM AUTHOR]