Accuracy and consistency of publicly available Large Language Models as clinical decision support tools for the management of colon cancer.

BACKGROUND: Large Language Models (LLM; e.g., ChatGPT) may be used to assist clinicians and form the basis of future clinical decision support (CDS) for colon cancer. The objectives of this study were to (1) evaluate the response accuracy of two LLM-powered interfaces in identifying guideline-based care in simulated clinical scenarios and (2) define response variation between and within LLMs. METHODS: Clinical scenarios with "next steps in management" queries were developed based on National Comprehensive Cancer Network guidelines. Prompts were entered into OpenAI ChatGPT and Microsoft Copilot in independent sessions, yielding four responses per scenario. Responses were compared to clinician-developed responses and assessed for accuracy, consistency, and verbosity. RESULTS: Across 108 responses to 27 prompts, both platforms yielded completely correct responses to 36% of scenarios (n = 39). For ChatGPT, 39% (n = 21) were missing information and 24% (n = 14) contained inaccurate/misleading information. Copilot performed similarly, with 37% (n = 20) having missing information and 28% (n = 15) containing inaccurate/misleading information (p = 0.96). Clinician responses were significantly shorter (34 ± 15.5 words) than both ChatGPT (251 ± 86 words) and Copilot (271 ± 67 words; both p < 0.01). CONCLUSIONS: Publicly available LLM applications often provide verbose responses with vague or inaccurate information regarding colon cancer management. Significant optimization is required before use in formal CDS.

A second hereditary cancer predisposition syndrome in a patient with lynch syndrome and three primary cancers.

Current National Comprehensive Cancer Network ® (NCCN ®) guidelines for Colorectal Genetic/Familial High-Risk Assessment provide limited guidance for genetic testing for individuals with already diagnosed hereditary cancer conditions. We are presenting the case of a 36-year-old woman who was diagnosed with Lynch Syndrome at age 23 after genetic testing for a familial variant (c.283del) in the MLH1 gene. The patient had a previous history of Hodgkin Lymphoma at the time of familial variant testing, and she would later develop stage IIIa cecal adenocarcinoma at age 33 and metastatic papillary thyroid carcinoma at age 35. The patient's family history included a first-degree relative who was diagnosed with colorectal cancer at age 39, multiple second-degree relatives with colorectal, endometrial, and stomach cancer, and third and fourth-degree relatives with breast cancer. In light of her personal and family history, a comprehensive cancer panel was recommended. This panel found a second hereditary cancer predisposition syndrome: a likely pathogenic variant (c. 349 A > G) in the CHEK2 gene. This specific CHEK2 variant was recently reported to confer a moderately increased risk for breast cancer. The discovery of this second cancer predisposition syndrome had important implications for the patient's screening and risk management. While uncommon, the possibility of an individual having multiple cancer predisposition syndromes is important to consider when evaluating patients and families for hereditary cancer, even when a familial variant has been identified.

En bloc chest wall resection in locally advanced cT3N2 (stage IIIB) lung cancer involving the chest wall: Revisiting guidelines.

Objectives: Current National Comprehensive Cancer Network guidelines recommend definitive chemoradiation rather than surgery for patients with locally advanced clinical stage T3 and N2 (stage IIIB) lung cancer involving the chest wall. The data supporting this recommendation are controversial. We studied whether surgery confers a survival advantage over definitive chemoradiation in the National Cancer Database. Methods: We identified all patients with clinical stage T3 and N2 lung cancer in the National Cancer Database from 2004 to 2017 who underwent a lobectomy with en bloc chest wall resection and compared them with patients with clinical stage T3 and N2 lung cancer who had definitive chemoradiation. We used propensity score matching to minimize confounding by indication while excluding patients with tumors in the upper lobes to exclude Pancoast tumors. We used 1:1 propensity score matching and Kaplan-Meir survival analyses to estimate associations. Results: Of 4467 patients meeting all inclusion/exclusion criteria, 210 (4.49%) had an en bloc chest wall resection. Patients undergoing surgical resection were younger (mean age = 60.3 ± 10.3 years vs 67.5 ± 10.4 years; P < .001) and had more adenocarcinoma (59.0% vs 44.5%; P < .001) but were otherwise similar in terms of sex (37.1% female vs 42.0%; P = .167) and race (Whites 84.3% vs 84.0%; P = .276) compared with the definitive chemoradiation group. After resection, there was an unadjusted 30- and 90-day mortality rate of 3.3% and 9.5%, respectively. A substantial survival benefit with surgical resection persisted after propensity score matching (log-rank P < .001). Conclusions: In this large observational study, we found that in select patients, en bloc chest wall resection for locally advanced clinical stage T3 and N2 lung cancer was associated with improved survival compared with definitive chemoradiation. National Comprehensive Cancer Network guidelines should be revisited.

Real-world utilization, patient characteristics, and treatment patterns among men with localized prostate cancer tested with the 17-gene genomic prostate score® (GPSTM) assay.

OBJECTIVES: Descriptive study focusing on real-world utilization and characteristics of men with prostate cancer tested with the 17-gene Genomic Prostate Score® (GPS™) assay by linking administrative claims and electronic health record (EHR) data with GPS results. METHODS: This retrospective, observational cohort study (January 1, 2013 to December 31, 2020) included men aged 40-80 years with localized prostate cancer claims, continuous enrollment in Optum's Integrated Claims data set, ≥1 day of EHR clinical activity, and a GPS result. Men were classified as undergoing definitive therapy (DT) (prostatectomy, radiation, or focal therapy) or active surveillance (AS). AS and DT distribution were analyzed across GPS results, National Comprehensive Cancer Network® (NCCN®) risk, and race. Costs were assessed 6 months after the first GPS result (index); clinical outcomes and AS persistence were assessed during the variable follow-up. All variables were analyzed descriptively. RESULTS: Of 834 men, 650 (77.9%) underwent AS and 184 (22.1%) DT. Most men had Quan-Charlson comorbidity scores of 1-2 and a tumor stage of T1c (index). The most common Gleason patterns were 3 + 3 (79.6%) (AS cohort) and 3 + 4 (55.9%) (DT cohort). The mean (standard deviation) GPS results at index were 23.2 (11.3) (AS) and 30.9 (12.9) (DT). AS decreased with increasing GPS result and NCCN risk. Differences between races were minimal. Total costs were substantially higher in the DT cohort. CONCLUSIONS: Most men with GPS-tested localized prostate cancer underwent AS, indicating the GPS result can inform clinical management. Decreasing AS with increasing GPS result and NCCN risk suggests the GPS complements NCCN risk stratification.

Exploring the landscape of AI-assisted decision-making in head and neck cancer treatment: a comparative analysis of NCCN guidelines and ChatGPT responses.

PURPOSE: Recent breakthroughs in natural language processing and machine learning, exemplified by ChatGPT, have spurred a paradigm shift in healthcare. Released by OpenAI in November 2022, ChatGPT rapidly gained global attention. Trained on massive text datasets, this large language model holds immense potential to revolutionize healthcare. However, existing literature often overlooks the need for rigorous validation and real-world applicability. METHODS: This head-to-head comparative study assesses ChatGPT's capabilities in providing therapeutic recommendations for head and neck cancers. Simulating every NCCN Guidelines scenarios. ChatGPT is queried on primary treatments, adjuvant treatment, and follow-up, with responses compared to the NCCN Guidelines. Performance metrics, including sensitivity, specificity, and F1 score, are employed for assessment. RESULTS: The study includes 68 hypothetical cases and 204 clinical scenarios. ChatGPT exhibits promising capabilities in addressing NCCN-related queries, achieving high sensitivity and overall accuracy across primary treatment, adjuvant treatment, and follow-up. The study's metrics showcase robustness in providing relevant suggestions. However, a few inaccuracies are noted, especially in primary treatment scenarios. CONCLUSION: Our study highlights the proficiency of ChatGPT in providing treatment suggestions. The model's alignment with the NCCN Guidelines sets the stage for a nuanced exploration of AI's evolving role in oncological decision support. However, challenges related to the interpretability of AI in clinical decision-making and the importance of clinicians understanding the underlying principles of AI models remain unexplored. As AI continues to advance, collaborative efforts between models and medical experts are deemed essential for unlocking new frontiers in personalized cancer care.

Radiation-Associated Angiosarcoma as a Presentation of Disease Progression in a Patient on Immunotherapy for Metastatic Non-small Cell Lung Cancer.

Non-small cell lung cancer (NSCLC) is the dominant form of lung cancer, comprising around 85% of cases. Stage 4 NSCLC has a grim prognosis; however, immunotherapy and radiation therapy have become vital treatments for advanced-stage NSCLC, despite the risk of inducing a second primary malignancy. This case report focuses on a 45-year-old female diagnosed with NSCLC and metastasis to the 11th thoracic vertebral body. After various treatments, including radiation, a potential radiation-associated secondary malignancy, epithelial angiosarcoma, was discovered. Following treatment modification, the patient achieved complete metabolic remission, highlighting the importance of clinicians being cautious about secondary primary cancers in NSCLC patients with a history of radiation therapy. Accurate diagnosis through biopsy and continuous surveillance are essential in managing NSCLC patients effectively.

Overview of glottic laryngeal cancer treatment recommendation changes in the NCCN guidelines from 2011 to 2022.

BACKGROUND: The treatment of glottic cancer remains challenging, especially with regard to morbidity reduction and larynx preservation rates. The National Comprehensive Cancer Network (NCCN) has published guidelines to aid decision-making about this treatment according to the tumor site, clinical stage, and patient medical status. AIM: The present review was conducted to identify changes in the NCCN guidelines for glottic cancer treatment made between 2011 and 2022 and to describe the published evidence concerning glottic cancer treatment and oncological outcomes in the same time period. METHODS AND RESULTS: Clinical practice guidelines for head and neck cancer published from 2011 up to 2022 were obtained from the NCCN website (www.NCCN.org). Data on glottic cancer treatment recommendations were extracted, and descriptive analysis was performed. In addition, a review of literature registered in the PubMed database was performed to obtain data on glottic cancer management protocols and treatment outcomes from randomized controlled trials, systematic reviews, and meta-analyses published from 2011 to 2022. In total, 24 NCCN guidelines and updates and 68 relevant studies included in the PubMed database were identified. The main guideline changes made pertained to surgical and systemic therapies, the consideration of adverse features, and new options for the treatment of metastatic disease at initial presentation. Early-stage glottic cancer received the most research attention, with transoral endoscopic laser surgery and radiotherapy assessed and compared as the main treatment modalities. Reported associations between treatment types and survival rates for this stage of glottic cancer appear to be similar, but functional outcomes can be highly compromised. CONCLUSION: NCCN panel members provide updated recommendations based on currently accepted treatment approaches for glottic cancer, constantly reviewing new surgical and non-surgical techniques. The guidelines support decision-making about glottic cancer treatment that should be individualized and prioritize patients' quality of life, functionality, and preferences.

The Value and Process of Inclusion: Using Sensitive, Respectful, and Inclusive Language and Images in NCCN Content.

A core component of NCCN's mission is to improve and facilitate equitable cancer care. Inclusion and representation of diverse populations are essential toward this goal of equity. Within NCCN's professional content, inclusivity increases the likelihood that clinicians are prepared to provide optimal oncology care to all patients; within NCCN's patient-facing content, it helps ensure that cancer information is relevant and accessible for all individuals. This article describes changes that have been made in the language and images used in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) and the NCCN Guidelines for Patients to promote justice, respect, and inclusion for all patients with cancer. The goals are to use language that is person-first, nonstigmatizing, inclusive of individuals of all sexual orientations and gender identities, and anti-racist, anti-classist, anti-misogynist, anti-ageist, anti-ableist, and anti-fat-biased. NCCN also seeks to incorporate multifaceted diversity in images and illustrations. NCCN is committed to continued and expanding efforts to ensure its publications are inclusive, respectful, and trustworthy, and that they advance just, equitable, high-quality, and effective cancer care for all.

Capacity of ChatGPT to Identify Guideline-Based Treatments for Advanced Solid Tumors.

BACKGROUND:  ChatGPT, created by OpenAI, is a large language model which has become the fastest growing consumer application in history, recognized for its expansive knowledge of varied subjects. The field of oncology is highly specialized and requires nuanced understanding of medications and conditions. Herein, we sought to better qualify the ability of ChatGPT to name applicable treatments for patients with advanced solid cancers. METHODS:  This observational study was conducted utilizing ChatGPT. The capacity of ChatGPT to tabulate appropriate systemic therapies for new diagnoses of advanced solid malignancies was ascertained through standardized prompts. A ratio of those medications listed by ChatGPT to those suggested in the National Comprehensive Cancer Network (NCCN) guidelines was produced and called the valid therapy quotient (VTQ). Additional descriptive analyses of the VTQ and its association with incidence and type of treatment were performed. RESULTS:  Some 51 distinct diagnoses were utilized within this experiment. ChatGPT was able to identify 91 distinct medications in response to prompts related to advanced solid tumors. The overall VTQ is 0.77. In all cases, ChatGPT was able to provide at least one example of systemic therapy suggested by the NCCN. There was a weak association between incidence of each malignancy and the VTQ. CONCLUSION:  The capacity of ChatGPT to identify medications used to treat advanced solid tumors indicates a level of concordance with the NCCN guidelines. As it stands, the role of ChatGPT to assist oncologists and patients in treatment decision making remains unknown. Nonetheless, in future iterations, it may be anticipated that accuracy and consistency in this domain will improve, and further studies will be needed to better quantify its capabilities.

Compliance with the Current NCCN Guidelines and Its Critical Role in Pancreatic Adenocarcinoma.

OBJECTIVES: Since 2019, the National Comprehensive Cancer Network (NCCN) has recommended genetic testing for patients diagnosed with pancreatic adenocarcinoma that includes universal germline testing and tumor gene profiling for metastatic, locally advanced, or recurrent disease. However, testing compliance with this guideline has not yet been published in the English literature. METHODS: A quality assurance/quality improvement retrospective review was done to identify patients diagnosed with pancreatic adenocarcinoma from January 2019 to February 2021 to include the patient's clinical status and genetic test results. RESULTS: There were 20 patient cases identified with pancreatic adenocarcinoma. A total of 11 cases had molecular tumor gene profiling and microsatellite instability/mismatch repair (MSI/MMR) testing performed and 1 case had only MSI/MMR testing by immunohistochemistry performed. Only 3 patients of the 20 in total received germline testing. CONCLUSION: There was a significant number of patients for whom tumor gene profiling or germline testing had never been attempted as per recommended NCCN guidelines.

Interpretation of NCCN guidelines version 1. 2023 thymomas and thymic carcinomas / 中国胸心血管外科临床杂志

@#National Comprehensive Cancer Network (NCCN) has updated and released the latest content of NCCN guidelines version 1. 2023 thymomas and thymic carcinomas (known as "guidelines"). The guideline sets standards for the diagnosis and treatment of thymoma and thymic carcinoma based on high quality clinical evidence and the latest advances in research. There have been some updates and revisions in the latest two versions of the guidelines, mainly focusing on the principles of radiotherapy, the principles of systematic therapy, multidisciplinary participation and the improvement of some footnotes, compared with the first version of the guidelines in 2022. In this paper, the contents of the new guideline will be interpreted in order to provide reference for the work of thymoma and thymic carcinoma in our country at the present stage.

Interpretations of the NCCN guidelines for breast cancer risk reduction (version 2023) / 中国胸心血管外科临床杂志

@#Breast cancer, the most common malignancy in the world, also causes the most death cases of women among malignancies. Breast cancer risk reduction guidelines (version 2023) was updated by National Comprehensive Cancer Network (NCCN). Based on high-level evidences from evidence-based medicine and the latest research progress, the guidelines provided standardized guidance for breast cancer risk assessment and risk reduction strategies for individuals without a history of invasive breast cancer or ductal carcinoma in situ, which has attracted widespread attention from clinicians worldwide. Breast cancer is also the most common malignancy in Chinese women, and the number of newly diagnosed breast cancer cases each year in China ranks first in the world due to the large population, so the breast cancer prevention has become a major public health challenge in China. Aimed to provide reference for breast cancer prevention in China, this article interpreted the guidelines (the new version) based on the characteristics of breast structure in Asian women and the epidemiological characteristics of breast cancer in China.

Best practices in surgical and nonsurgical management of head and neck Merkel cell carcinoma: An update.

Merkel cell carcinoma (MCC) is a rare, highly aggressive cutaneous neuroendocrine carcinoma. Controversy exists regarding optimal management of MCC as high-quality randomized studies and clinical trials are limited, and physicians are bound to interpret highly heterogeneous, retrospective literature in their clinical practice. Furthermore, the rising incidence and notably poor prognosis of MCC urges the establishment of best practices for optimal management of the primary tumor and its metastases. Herein, we summarized the relevant evidence and provided an algorithm for decision-making in MCC management based on the latest 2021 National Comprehensive Cancer Network guidelines. Additionally, we report current active MCC clinical trials in the United States. The initial management of MCC is dependent upon the pathology of the primary tumor and presence of metastatic disease. Patients with no clinical evidence of regional lymph node involvement generally require sentinel node biopsy (SLNB) while clinically node-positive patients should undergo fine needle aspiration (FNA) or core biopsy and full imaging workup. If SLNB or FNA/core biopsy are positive, a multidisciplinary team should be assembled to discuss if additional node dissection or adjuvant therapy is necessary. Wide local excision is optimal for primary tumor management and SLNB remains the preferred staging and predictive tool in MCC. The management of MCC has progressively improved in the last decade, particularly due to the establishment of immunotherapy as a new treatment option in advanced MCC. Ongoing trials and prospective studies are needed to further establish the best practices for MCC management.

Treatment heterogeneity and overall survival in patients with advanced/metastatic gastric or gastroesophageal junction adenocarcinoma in the United States.

Background: Gastric or gastroesophageal junction (GEJ) adenocarcinoma is the most common form of gastric cancer diagnosed in the United States (US) each year. Diagnosis typically is in later stages of disease when it has advanced. Patients have been treated with a variety of regimens. Methods: The goal of this retrospective study was to understand if treatment patterns were becoming more homogeneous or remaining heterogeneous using the Herfindahl-Hirschman index (HHI) and if treatments were becoming more concordant to treatment guidelines published by the National Comprehensive Cancer Network (NCCN). HHI scores were calculated for each site by 2-year increments. Trend analyses were conducted for HHI scores over time using a linear regression model. Concordance to Category 1 and any category NCCN guidelines was determined based on the date treatment was initiated with the version of the NCCN guidelines at that time. Time trend analyses were conducted using linear regression models. This study utilized data from the Flatiron Advanced Gastric/Esophageal cohort. This study also examined overall survival (OS) rates estimated by the Kaplan-Meier method by line of therapy. Results: There were no statistically significant differences in HHI scores in the first-line setting over time, suggesting heterogeneity has not improved. Concordance to NCCN treatment guidelines for any category significantly increased over time, however Category 1 regimen concordance remained low in the first-line setting. Concordance over time improved in second-line treatment. Median OS from the start of first-line therapy was 13.57 months. There was no relationship between OS time from initiation of first-line therapy and HHI score, concordance with NCCN guidelines, or concordance with NCCN Category 1 guidelines in the first-line setting. Conclusions: Treatment heterogeneity persists in gastric cancer care, though there is a significant association between heterogeneity and concordance with both Category 1 and any category in the NCCN treatment guidelines, and that concordance has increased over time.

Understanding the Uptake and Challenges of Genetic Testing Guidelines for Prostate Cancer Patients.

BACKGROUND: Multiple studies have confirmed a high prevalence of prognostic germline mutations in prostate cancer. In recognition, the NCCN guidelines and recommendations for genetic counselling (GC) in prostate cancer patients were expanded. METHODS: Data on prostate cancer patients at a single tertiary cancer center from January 2019 - June 2019 were queried. The cohort of patients from the queried list were evaluated for their eligibility for genetic testing. From the patients that were eligible for testing, the rate of referrals was ascertained. A 10-item questionnaire was concurrently sent to providers to understand germline genetic testing patterns and potential barriers. RESULTS: Only 39% of the eligible prostate cancer patients were referred, with testing completed in 11% with indications. 30% of providers reported they would be comfortable completing genetic counseling themselves. The identified barriers to provide genetic testing themselves were lack of time and expertise (50%). Other barriers included: lack of genetic counselor workforce (70%), lack of knowledge of genetic testing and the inadequate co-ordination of referrals (60%). CONCLUSION: In this retrospective study, many patients met the criteria for GC, however, the referrals for this patient population are inconsistent, and only a handful of the eligible patients completed testing. Identified barriers were provider's knowledge and comfort with guidelines and testing, systemic bottlenecks such as limited capacity of genetic counsellors, and the creation of improved workflows.

Guideline discordant care in patients with metastatic germ cell tumors.

INTRODUCTION: Despite high curability, patients with metastatic germ cell tumors (GCT) in the United States general population persistently face inferior outcomes compared with those treated in specialty referral centers. We characterized guideline discordant management in patients with metastatic GCT who experienced relapse after first-line chemotherapy and compared those who were initially treated in community practices vs. academic referral centers. PATIENTS/METHODS: Retrospective analysis of 53 patients with relapsed GCT between 2005 and 2018. First-line GCT management was assessed against the National Comprehensive Cancer Network guidelines. Guideline discordant management, predictors of discordance, and associations with outcomes were assessed. RESULTS: Of 53 patients with relapsed GCT, 34% received guideline discordant care in the first-line setting. Guideline discordant care was more prevalent in patients initially treated in community practices (12/30, 40%) vs. those initially treated in academic centers (3/22, 14%), though in multivariate logistic regression, this difference was not statistically significant (odds ratio: 4.07, P = 0.08). Most patients in community settings who received guideline discordant care were undertreated (10/12, 83%). There were 3 major reasons for guideline discordant care: (1) failure to resect residual masses after chemotherapy (27%, 4/15), (2) mismanagement of chemotherapy-related adverse events (27%, 4/15), and (3) under staging at diagnosis, resulting either insufficient chemotherapy regimen intensity (13%, 2/15) and/or inappropriately receiving primary surgical resection for metastatic disease (20%, 3/15). CONCLUSION: Under treatment was identified in nearly half of patients initially treated in a community setting who later developed relapsed GCT. Referral to specialized centers for a second opinion should be considered for all metastatic GCT patients in the first-line setting and all patients with post-chemotherapy residual disease. More effective methods should be developed to facilitate second opinions from expert centers in the United States.

Interpretation of updated NCCN guidelines for non-small cell lung cancer (version 1, 2022) / 中国胸心血管外科临床杂志

@#The National Comprehensive Cancer Network (NCCN) has updated and released the latest content of the NCCN guidelines for the clinical diagnosis and treatment of non-small cell lung cancer (NSCLC) in the version 1, 2022. Based on high-quality clinical evidence and the latest research progress of the diagnosis and treatment of NSCLC, the guidelines have been widely recognized and welcomed by clinicians around the world. Compared with the version 7, 2021, the new version has been updated and revised in some parts of chapters and sections, mainly focusing on targeted therapies and molecular testing. This article will interpret the updated therapy content of the new version.