Comparison of Large Language Models in Answering Immuno-Oncology Questions: A Cross-Sectional Study.

BACKGROUND: The capability of large language models (LLMs) to understand and generate human-readable text has prompted the investigation of their potential as educational and management tools for patients with cancer and healthcare providers. MATERIALS AND METHODS: We conducted a cross-sectional study aimed at evaluating the ability of ChatGPT-4, ChatGPT-3.5, and Google Bard to answer questions related to 4 domains of immuno-oncology (Mechanisms, Indications, Toxicities, and Prognosis). We generated 60 open-ended questions (15 for each section). Questions were manually submitted to LLMs, and responses were collected on June 30, 2023. Two reviewers evaluated the answers independently. RESULTS: ChatGPT-4 and ChatGPT-3.5 answered all questions, whereas Google Bard answered only 53.3% (P < .0001). The number of questions with reproducible answers was higher for ChatGPT-4 (95%) and ChatGPT3.5 (88.3%) than for Google Bard (50%) (P < .0001). In terms of accuracy, the number of answers deemed fully correct were 75.4%, 58.5%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (P = .03). Furthermore, the number of responses deemed highly relevant was 71.9%, 77.4%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (P = .04). Regarding readability, the number of highly readable was higher for ChatGPT-4 and ChatGPT-3.5 (98.1%) and (100%) compared to Google Bard (87.5%) (P = .02). CONCLUSION: ChatGPT-4 and ChatGPT-3.5 are potentially powerful tools in immuno-oncology, whereas Google Bard demonstrated relatively poorer performance. However, the risk of inaccuracy or incompleteness in the responses was evident in all 3 LLMs, highlighting the importance of expert-driven verification of the outputs returned by these technologies.

Utility of Large Language Models for Health Care Professionals and Patients in Navigating Hematopoietic Stem Cell Transplantation: Comparison of the Performance of ChatGPT-3.5, ChatGPT-4, and Bard.

BACKGROUND: Artificial intelligence is increasingly being applied to many workflows. Large language models (LLMs) are publicly accessible platforms trained to understand, interact with, and produce human-readable text; their ability to deliver relevant and reliable information is also of particular interest for the health care providers and the patients. Hematopoietic stem cell transplantation (HSCT) is a complex medical field requiring extensive knowledge, background, and training to practice successfully and can be challenging for the nonspecialist audience to comprehend. OBJECTIVE: We aimed to test the applicability of 3 prominent LLMs, namely ChatGPT-3.5 (OpenAI), ChatGPT-4 (OpenAI), and Bard (Google AI), in guiding nonspecialist health care professionals and advising patients seeking information regarding HSCT. METHODS: We submitted 72 open-ended HSCT-related questions of variable difficulty to the LLMs and rated their responses based on consistency-defined as replicability of the response-response veracity, language comprehensibility, specificity to the topic, and the presence of hallucinations. We then rechallenged the 2 best performing chatbots by resubmitting the most difficult questions and prompting to respond as if communicating with either a health care professional or a patient and to provide verifiable sources of information. Responses were then rerated with the additional criterion of language appropriateness, defined as language adaptation for the intended audience. RESULTS: ChatGPT-4 outperformed both ChatGPT-3.5 and Bard in terms of response consistency (66/72, 92%; 54/72, 75%; and 63/69, 91%, respectively; P=.007), response veracity (58/66, 88%; 40/54, 74%; and 16/63, 25%, respectively; P<.001), and specificity to the topic (60/66, 91%; 43/54, 80%; and 27/63, 43%, respectively; P<.001). Both ChatGPT-4 and ChatGPT-3.5 outperformed Bard in terms of language comprehensibility (64/66, 97%; 53/54, 98%; and 52/63, 83%, respectively; P=.002). All displayed episodes of hallucinations. ChatGPT-3.5 and ChatGPT-4 were then rechallenged with a prompt to adapt their language to the audience and to provide source of information, and responses were rated. ChatGPT-3.5 showed better ability to adapt its language to nonmedical audience than ChatGPT-4 (17/21, 81% and 10/22, 46%, respectively; P=.03); however, both failed to consistently provide correct and up-to-date information resources, reporting either out-of-date materials, incorrect URLs, or unfocused references, making their output not verifiable by the reader. CONCLUSIONS: In conclusion, despite LLMs' potential capability in confronting challenging medical topics such as HSCT, the presence of mistakes and lack of clear references make them not yet appropriate for routine, unsupervised clinical use, or patient counseling. Implementation of LLMs' ability to access and to reference current and updated websites and research papers, as well as development of LLMs trained in specialized domain knowledge data sets, may offer potential solutions for their future clinical application.

Management of Locally Advanced and Metastatic Esophageal Cancer in the Older Population.

PURPOSE OF REVIEW: This review aims to synthesise the current literature on the management of early-stage and metastatic esophageal cancers, focusing on the older population. In particular, we aim to dissect out the elderly-specific data from the relevant trials and to discuss the issues unique to this population. RECENT FINDINGS: While surgery is the curative modality in esophageal malignancies, the CROSS, MAGIC and FLOT trials demonstrate a clear advantage to neoadjuvant therapy (chemotherapy and chemoradiotherapy). These trials, however, included few elderly patients. There is a similar lack of elderly-specific data in the metastatic setting. Esophageal malignancies remain highly lethal with increasing incidence with age. Despite the relative lack of elderly-specific data, the fit older population appear to similarly benefit from multimodal therapy in early-stage and palliative therapy in metastatic disease.

Analysing the Combined Effects of Radiotherapy and Chemokine Receptor 5 Antagonism: Complementary Approaches to Promote T Cell Function and Migration in Oesophageal Adenocarcinoma.

The presence of an immunosuppressive tumour microenvironment in oesophageal adenocarcinoma (OAC) is a major contributor to poor responses. Novel treatment strategies are required to supplement current regimens and improve patient survival. This study examined the immunomodulatory effects that radiation therapy and chemokine receptor antagonism impose on T cell phenotypes in OAC with a primary goal of identifying potential therapeutic targets to combine with radiation to improve anti-tumour responses. Compared with healthy controls, anti-tumour T cell function was impaired in OAC patients, demonstrated by lower IFN-γ production by CD4+ T helper cells and lower CD8+ T cell cytotoxic potential. Such diminished T cell effector functions were enhanced following treatment with clinically relevant doses of irradiation. Interestingly, CCR5+ T cells were significantly more abundant in OAC patient blood compared with healthy controls, and CCR5 surface expression by T cells was further enhanced by clinically relevant doses of irradiation. Moreover, irradiation enhanced T cell migration towards OAC patient-derived tumour-conditioned media (TCM). In vitro treatment with the CCR5 antagonist Maraviroc enhanced IFN-γ production by CD4+ T cells and increased the migration of irradiated CD8+ T cells towards irradiated TCM, suggesting its synergistic therapeutic potential in combination with irradiation. Overall, this study highlights the immunostimulatory properties of radiation in promoting anti-tumour T cell responses in OAC and increasing T cell migration towards chemotactic cues in the tumour. Importantly, the CCR5 antagonist Maraviroc holds promise to be repurposed in combination with radiotherapy to promote anti-tumour T cell responses in OAC.

Applications of large language models in cancer care: current evidence and future perspectives.

The development of large language models (LLMs) is a recent success in the field of generative artificial intelligence (AI). They are computer models able to perform a wide range of natural language processing tasks, including content generation, question answering, or language translation. In recent months, a growing number of studies aimed to assess their potential applications in the field of medicine, including cancer care. In this mini review, we described the present published evidence for using LLMs in oncology. All the available studies assessed ChatGPT, an advanced language model developed by OpenAI, alone or compared to other LLMs, such as Google Bard, Chatsonic, and Perplexity. Although ChatGPT could provide adequate information on the screening or the management of specific solid tumors, it also demonstrated a significant error rate and a tendency toward providing obsolete data. Therefore, an accurate, expert-driven verification process remains mandatory to avoid the potential for misinformation and incorrect evidence. Overall, although this new generative AI-based technology has the potential to revolutionize the field of medicine, including that of cancer care, it will be necessary to develop rules to guide the application of these tools to maximize benefits and minimize risks.

Comparison of Large Language Models in Answering Immuno-Oncology Questions: A Cross-Sectional Study.

Background: The capability of large language models (LLMs) to understand and generate human-readable text has prompted the investigation of their potential as educational and management tools for cancer patients and healthcare providers. Materials and Methods: We conducted a cross-sectional study aimed at evaluating the ability of ChatGPT-4, ChatGPT-3.5, and Google Bard to answer questions related to four domains of immuno-oncology (Mechanisms, Indications, Toxicities, and Prognosis). We generated 60 open-ended questions (15 for each section). Questions were manually submitted to LLMs, and responses were collected on June 30th, 2023. Two reviewers evaluated the answers independently. Results: ChatGPT-4 and ChatGPT-3.5 answered all questions, whereas Google Bard answered only 53.3% (p <0.0001). The number of questions with reproducible answers was higher for ChatGPT-4 (95%) and ChatGPT3.5 (88.3%) than for Google Bard (50%) (p <0.0001). In terms of accuracy, the number of answers deemed fully correct were 75.4%, 58.5%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (p = 0.03). Furthermore, the number of responses deemed highly relevant was 71.9%, 77.4%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (p = 0.04). Regarding readability, the number of highly readable was higher for ChatGPT-4 and ChatGPT-3.5 (98.1%) and (100%) compared to Google Bard (87.5%) (p = 0.02). Conclusion: ChatGPT-4 and ChatGPT-3.5 are potentially powerful tools in immuno-oncology, whereas Google Bard demonstrated relatively poorer performance. However, the risk of inaccuracy or incompleteness in the responses was evident in all three LLMs, highlighting the importance of expert-driven verification of the outputs returned by these technologies.

Visceral adipose tissue secretome from early and late-stage oesophageal cancer patients differentially affects effector and regulatory T cells.

AIM: Visceral obesity is a key risk factor in the development of oesophagogastric junctional adenocarcinoma (OGJ), predominantly via generation of systemic low grade inflammation. Obesity-induced inflammation promotes resistance to current standards of care, enhancing tumour cell growth and survival. This study investigates the effect of the visceral adipose tissue secretome from OGJ patients with early versus advanced tumours on T-cell immunity and the role of immune checkpoint blockade in enhancing anti-tumour immunity. METHODS AND RESULTS: Visceral adipose conditioned media (ACM) from both early and late-stage OGJ patients significantly altered T cell activation status, upregulating co-stimulatory marker CD27 on T cells. ACM from both early and late-stage OGJ patients significantly altered immune checkpoint expression profiles downregulating immune checkpoints (ICs) on the surface of dual Th1/17-like and Th17-like cells and upregulating ICs on the surface of Th1-like cells and Treg cells. ACM derived from early-stage OGJ patients but not late-stage OGJ patients increased IFN-γ production by T cells. The addition of immune checkpoint blockers (ICBs) did not increase IFN-γ production by T cells in the presence of late-stage ACM, collectively highlighting the dichotomous immunostimulatory effect of early-stage ACM and immune-inhibitory effect of late-stage ACM. Interestingly, ACM from early-stage OGJ patients was more pro-inflammatory than ACM from late-stage patients, reflected by decreased levels of IL-17A/F, TNF-α, IL-1RA and IL-5. CONCLUSION: The ACM-induced upregulation of ICs on T cells highlights a therapeutic vulnerability that could be exploited by ICBs to harness anti-cancer immunity and improve clinical outcomes for OGJ patients. Schematic workflow - (A) visceral adipose tissue was taken from OAC patients at time of surgery and cultured for 72 h in media. (B) The harvested ACM was co-cultured with healthy donor PBMCs that were concurrently activated with anti-CD3/28 for 48 h and T cell immunophenotyping was carried out by flow cytometry. Key findings - (A) Early and late stage ACM enhanced a Th1-like phenotype and upregulated CTLA-4 on Th1-like cells. A Th17-like phenotype was also enhanced in addition with a Treg-like phenotype. CTLA-4 and PD-L1 were upregulated on the surface of Treg-like cells. (B) ICB-attenuated IL-17 production by T cells. However, ACM attenuated ICB-mediated reduction in IL-10 production by T cells. Higher levels of pro-inflammatory factors were found in early stage ACM compared with late stage ACM.

FLOT and CROSS chemotherapy regimens alter the frequency of CD27+ and CD69+ T cells in oesophagogastric adenocarcinomas: implications for combination with immunotherapy.

Combining immunostimulatory chemotherapies with immunotherapy is an attractive strategy to enhance treatment responses in oesophagogastric junctional adenocarcinoma (OGJ). This study investigates the immunostimulatory properties of FLOT, CROSS and MAGIC chemotherapy regimens in the context of OGJ using in vitro and ex vivo models of the treatment-naïve and post-chemotherapy treated tumour microenvironment. FLOT and CROSS chemotherapy regimens increased surrogate markers of immunogenic cell death (HMGB1 and HLA-DR), whereas the MAGIC treatment regimen decreased HMGB1 and HLA-DR on OGJ cells (markedly for epirubicin). Tumour-infiltrating and circulating T cells had significantly lower CD27 expression and significantly higher CD69 expression post-FLOT and post-CROSS treatment. Similarly, the supernatant from FLOT- and CROSS-treated OGJ cell lines and from FLOT- and CROSS-treated OGJ biopsies cultured ex vivo also decreased CD27 and increased CD69 expression on T cells. Following 48 h treatment with post-FLOT and post-CROSS tumour conditioned media the frequency of CD69+ T cells in culture negatively correlated with the levels of soluble immunosuppressive pro-angiogenic factors in the conditioned media from ex vivo explants. Supernatant from FLOT- and CROSS-treated OGJ cell lines also increased the cytotoxic potential of healthy donor T cells ex vivo and enhanced OGJ patient-derived lymphocyte mediated-killing of OE33 cells ex vivo. Collectively, this data demonstrate that FLOT and CROSS chemotherapy regimens possess immunostimulatory properties, identifying these chemotherapy regimens as rational synergistic partners to test in combination with immunotherapy and determine if this combinatorial approach could boost anti-tumour immunity in OGJ patients and improve clinical outcomes.

Cooperation between chemotherapy and immune checkpoint blockade to enhance anti-tumour T cell immunity in oesophageal adenocarcinoma.

Response rates to immune checkpoint blockade (ICB) remain low in oesophageal adenocarcinoma (OAC). Combining ICB with immunostimulatory chemotherapies to boost response rates is an attractive approach for converting 'cold' tumours into 'hot' tumours. This study profiled immune checkpoint (IC) expression on circulating and tumour-infiltrating T cells in OAC patients and correlated these findings with clinical characteristics. The effect of first-line chemotherapy regimens (FLOT and CROSS) on anti-tumour T cell immunity was assessed to help guide design of ICB and chemotherapy combinations in the first-line setting. The ability of ICB to enhance lymphocyte-mediated cytolysis of OAC cells in the absence and presence of post-FLOT and post-CROSS chemotherapy tumour cell secretome was assessed by a CCK-8 assay. Expression of ICs on T cells positively correlated with higher grade tumours and a subsequent poor response to neoadjuvant treatment. First-line chemotherapy regimens substantially altered IC expression profiles of T cells increasing PD-1, A2aR, KLRG-1, PD-L1, PD-L2 and CD160 and decreasing TIM-3 and LAG-3. In addition, pro-inflammatory T cell cytokine profiles were enhanced by first-line chemotherapy regimens. T cell activation status was significantly altered; both chemotherapy regimens upregulated co-stimulatory markers ICOS and CD69 yet downregulated co-stimulatory marker CD27. However, ICB attenuated chemotherapy-induced downregulation of CD27 on T cells and promoted differentiation of effector memory T cells into a terminally differentiated state. Importantly, dual nivolumab-ipilimumab treatment increased lymphocyte-mediated cytolysis of OAC cells, an effect further enhanced in the presence of post-FLOT tumour cell secretome. These findings justify a rationale to administer ICBs concurrently with first-line chemotherapies.

The Liquid Biopsy for Lung Cancer: State of the Art, Limitations and Future Developments.

Lung cancer is a leading cause of cancer-related deaths, contributing to 18.4% of cancer deaths globally. Treatment of non-small cell lung carcinoma has seen rapid progression with targeted therapies tailored to specific genetic drivers. However, identifying genetic alterations can be difficult due to lack of tissue, inaccessible tumors and the risk of complications for the patient with serial tissue sampling. The liquid biopsy provides a minimally invasive method which can obtain circulating biomarkers shed from the tumor and could be a safer alternative to tissue biopsy. While tissue biopsy remains the gold standard, liquid biopsies could be very beneficial where serial sampling is required, such as monitoring disease progression or development of resistance mutations to current targeted therapies. Liquid biopsies also have a potential role in identifying patients at risk of relapse post treatment and as a component of future lung cancer screening protocols. Rapid developments have led to multiple platforms for isolating circulating tumor cells (CTCs) and detecting circulating tumor DNA (ctDNA); however, standardization is lacking, especially in lung carcinoma. Additionally, clonal hematopoiesis of uncertain clinical significance must be taken into consideration in genetic sequencing, as it introduces the potential for false positives. Various biomarkers have been investigated in liquid biopsies; however, in this review, we will concentrate on the current use of ctDNA and CTCs, focusing on the clinical relevance, current and possible future applications and limitations of each.

Vaping and lung cancer - A review of current data and recommendations.

OBJECTIVES: Lung cancer is the most common cause of cancer mortality worldwide and, while tobacco smoke remains the primary cause, there is increasing concern that vaping and E-cigarette use may also increase lung cancer risk. This review concentrates on the current data, scholarship and active foci of research regarding potential cancer risk and oncogenic mechanisms of vaping and lung cancer. MATERIALS AND METHODS: We performed a literature review of current and historical publications on lung cancer oncogenesis, vaping device/e-liquid contents and daughter products, molecular oncogenic mechanisms and the fundamental, potentially oncogenic, effects of electronic cigarette smoke/e-liquid products. RESULTS: E-cigarette devices and vaping fluids demonstrably contain a series of both definite and probable oncogens including nicotine derivatives (e.g. nitrosnornicotine, nitrosamine ketone), polycyclic aromatic hydrocarbons, heavy metals (including organometal compounds) and aldehydes/other complex organic compounds. These arise both as constituents of the e-liquid (with many aldehydes and other complex organics used as flavourings) and as a result of pyrolysis/complex organic reactions in the electronic cigarette device (including unequivocal carcinogens such as formaldehyde - formed from pyrolysis of glycerol). Various studies demonstrate in vitro transforming and cytotoxic activity of these derivatives. E-cigarette device use has been significantly increasing - particularly amongst the younger cohort and non-smokers; thus, this is an area of significant concern for the future. CONCLUSION: Although research remains somewhat equivocal, there is clear reason for concern regarding the potential oncogenicity of E-Cigarettes/E-Liquids with a strong basic and molecular science basis. Given lag times (extrapolating from tobacco smoke data) of perhaps 20 years, this may have significant future public health implications. Thus, the authors feel further study in this field is strongly warranted and consideration should be made for tighter control and regulation of these products.

The challenge of treating older patients with pancreaticobiliary malignancies.

Pancreatic and biliary tract cancers are aggressive malignancies. They commonly present with metastatic or unresectable disease. Those that do present with resectable cancer have high rates of recurrence. Despite recent advances in surgical technique, chemotherapy, and radiotherapy regimens, they are associated with poor survival outcomes. These cancers represent an exception to the trend of improved overall survival evident in most malignancies in recent decades. Depending on the goal of treatment, active management of pancreatic and biliary cancers involves surgery, chemotherapy, and radiation therapy, either alone or in combination. Both pancreatic and biliary tract cancers have a preponderance in the older population. Older patients are a heterogeneous group; although tolerability of multimodality treatment may be a challenge for some, many fit older patients may be undertreated based on their age alone. The growing field of geriatric oncology has highlighted the importance of a comprehensive assessment of these patients, and not relying on age alone as a discriminating factor for treatment. Management of older patients with pancreaticobiliary cancers is particularly challenging owing to limited prospective data in this population. As such, there is uncertainty with regard to optimal treatment approaches for these patients. In this article, we outline the therapeutic options available to patients with localized or advanced pancreatic and biliary tract cancers, and the evidence for specified treatment options in the elderly. We examine the inclusion and outcomes of elderly patients in relevant clinical trials; the morbidity that may be encountered by elderly patients receiving specified treatments and the tools that may assist the physician in selecting elderly patients for particular treatments.

Extended Adjuvant Endocrine Therapy in Hormone Receptor-Positive Early Breast Cancer.

Despite meaningful, incremental improvements in detection, local treatment and adjuvant systemic treatments for breast cancer, there remains a significant risk of late relapse in hormone receptor (HR)-positive disease. 5 years of tamoxifen or an aromatase inhibitor for all patients with HR-positive early breast cancer is considered standard; however, there are now data to support an extended approach using up to 10 years of treatment. This review will provide some historical background on endocrine therapy and summarize the key clinical trials that demonstrate the small absolute benefit of extended adjuvant therapy. We provide suggested treatment algorithms for both premenopausal and postmenopausal patients and an overview of ongoing adjuvant trials.