Purpose: Artificial intelligence models like GPT-4 (OpenAI) have the potential to support clinical decision-making in oncology. This study aimed to assess the consistency between multidisciplinary tumor board (MTB) decisions and GPT-4 model predictions in cancer patient management. Patients and methods: A cross-sectional study was conducted involving patients aged ≥ 18 years with definite or suspicious cancer diagnoses presented at MTBs in Ankara University Hospitals, Türkiye, from February 2021 to June 2023. GPT-4 was utilized to generate treatment recommendations based on case summaries. Three independent raters evaluated the compatibility between MTB decisions and GPT-4 predictions using a 4-point Likert scale. Cases with mean compatibility scores equal to or below 2 were reviewed by two expert oncologists for appropriateness. Results: A total of 610 patients were included. The mean compatibility score across raters was 3.59 (SD = 0.81), indicating high agreement between GPT-4 predictions and MTB decisions. Cronbach’s alpha was 0.950 (95% CI 0.935–0.960), demonstrating excellent interrater reliability. Sixty-two cases (10.2%) had mean compatibility scores below the threshold of 2. The first expert oncologist deemed GPT-4's predictions inappropriate in 8 of these cases (12.9%), while the second deemed them inappropriate in 16 cases (25.8%). Cohen’s kappa showed moderate agreement (κ = 0.50, 95% CI 0.25–0.75, p < 0.001). Discrepancies were often due to rare cases lacking guideline information or misunderstandings of case presentations. Conclusion: GPT-4 exhibited high compatibility with MTB decisions in cancer patient management, suggesting its potential as a supportive tool in clinical oncology. However, limitations exist, especially in rare or complex cases.

Purpose: Immunotherapy efficacy in elderly patients with comorbidities and poor performance status is not well understood. More knowledge on this topic is needed to identify subgroups that will benefit from immunotherapy. We aimed to evaluate the effect of comorbidity burden in patients receiving immunotherapy. Methods/patients: Patients older than 18 years of age and diagnosed with various malignancies, followed up in our tertiary cancer center were screened. Patients treated with immunotherapy were included in this study. We used to Charlson Comorbidity Index (CCI) to evaluate patients’ comorbidity burden. The primary outcome was overall survival (OS). Hazard ratio (HR) with confidence interval (CI) was evaluated in multivariable analysis. Results: A total number of 197 patients were included. The median age was 62 years. Patients were grouped based on CCI scores: CCI-low (≤ 8) and CCI-high (> 8). One-hundred and seven patients (54.9%) had metastatic disease at the time of diagnosis. Most frequently used immunotherapy agent was nivolumab (n = 124, 62.9%), followed by pembrolizumab (n = 36, 18.3%). The median OS was shorter in the CCI-high group than in the CCI-low group (10.6 vs. 21.2 months, p = 0.002) In multivariable analysis, treatment with anti-CTLA4 (HR: 1.85, 95% CI 1.07–3.20, p = 0.028), ECOG performance status (2–4 vs. 0–1) (HR: 2.17; 95% CI 1.25–3.75; p = 0.005), and higher CCI scores (CCI-high vs. CCI-low) (HR: 1.97; 95% CI 1.3–3.0; p = 0.001) were independently associated with worse OS. Conclusions: Comorbidity burden and performance status independently predict survival outcomes in immunotherapy-treated cancer patients. Comprehensive comorbidity assessment is essential for optimizing treatment and improving patient outcomes.

Background: Medical practitioners are increasingly using artificial intelligence (AI) chatbots for easier and faster access to information. To our knowledge, the accuracy and availability of AI-gen...

Objective: Reactivation of the hepatitis B virus (HBV) refers to an increase in HBV replication in a patient with inactive or resolved HBV. In this retrospective study, our aim is to present and compare HBV reactivation in multiple myeloma (MM) patients who received lenalidomide and/or bortezomib at any time during treatment, evaluate the factors associated with reactivation, and demonstrate the outcome of patients. Materials and Methods: We evaluated 178 MM patients who received lenalidomide (n=102) and/or bortezomib (n=174) during their treatment schedules. The HBsAg, anti-HBc, anti-HBs, HBeAg, and anti-HBe were detected by chemiluminescence by ARCHITECT lab analyzers using commercially available kits (Abbott, USA). HBV-DNA titers were determined by quantitative PCR. The results were evaluated by IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp., Armonk, NY, USA). Results: HBV reactivation was diagnosed in 6 patients (3%) after bortezomib and in 8 patients (8%) after bortezomib and lenalidomide. Three of the patients in each group had HBsAg+, HBeAg+, AntiHBeAg-, AntiHBc-, and AntiHBS+ status, whereas 5 patients in the bortezomib- and lenalidomide-treated group and 3 patients in the bortezomib-treated group had HBsAg-, HBeAg-, AntiHBeAg-, AntiHBc-, and AntiHBS+ status prior to treatment. There were no statistical differences observed between HBV reactivation in the bortezomib-treated or bortezomib- and lenalidomide-treated groups in terms of age at diagnosis, sex, International Staging System subtype, frequency of extramedullary disease, dialysis requirement, or receiving of autologous stem cell transplantation. In patients who received antiviral prophylaxis, a higher incidence of HBV reactivation was detected in HBsAg-positive patients compared to HBsAg-negative patients (4/4, 100% vs. 2/7, 29%; p=0.045). The 3-year and 5-year overall survival rates were similar in patients with or without HBV reactivation (83% vs. 84%, 73% vs. 74%, p=0.84). Conclusion: Close follow-up is recommended for not only HBsAg-positive but also HBsAg-negative patients.

Purpose Accurate prognostication in metastatic colorectal cancer (mCRC) remains challenging due to disease heterogeneity. This study aimed to develop and validate an artificial intelligence-augmented deep learning model for risk stratification in patients receiving first-line treatment. Methods/patients We developed a deep neural network with artificial intelligence augmentation, using data from patients with de novo mCRC treated at two major reference centers between 2010 and 2024. Patients with BRAF-mutated and MSI-high tumors were excluded. The model incorporated clinical characteristics, laboratory parameters, and treatment data. The primary outcome was progression-free survival (PFS). Results A total of 214 patients were included in the study, with 127 patients in the training and internal validation cohort and 87 patients in the external validation cohort. The model stratified patients into three distinct risk groups with significantly different PFS (log-rank p < 0.001). The low-risk group (n = 34) achieved a median PFS of 16.8 months with a 29% event rate, the medium-risk group (n = 33) showed a median PFS of 9.3 months with a 58% event rate, and the high-risk group (n = 34) demonstrated a median PFS of 7.5 months with a 76% event rate. Feature importance analysis identified carcinoembryonic antigen, neutrophil/lymphocyte ratio, and liver function tests as the strongest predictors of PFS. The model’s performance was consistent across both internal and external validation cohorts. Conclusions This deep learning model demonstrates robust prognostic capabilities in mCRC, effectively stratifying patients into distinct risk groups. The model could aid in clinical decision-making and treatment planning for patients receiving first-line therapy.

Objectives: Thrombotic thrombocytopenic purpura (TTP) is a rare thrombotic microangiopathy with no standardized prognostic model to predict mortality. The aim of the study is to determine parameters associated with TTP-related mortality. Methods: In this cross-sectional, retrospective study, 77 TTP patients, treated with therapeutic plasma exchange between 2001 and 2019 in Ankara University Faculty of Medicine were included. Results: There was no significant relationship between ADAMTS13 inhibitor levels, activity and mortality. Median number of plasmapheresis was 10 (2–32), higher in patients with complete response. Anemia, kidney injury and LDH levels were associated with survival; there was no significant relationship between platelet counts at the time of diagnosis and mortality. Mortality was lower in patients with platelet counts above 100 × 10 ⁹ /L and normalized LDH after treatment. Hemoglobin, albumin, LDH and creatinine levels at the time platelet counts exceeded 50 × 10 ⁹ /L were associated with survival. Conclusions: We determined several clinical and laboratory parameters associated with mortality. Fewer numbers of plasmapheresis was associated with mortality; thus other treatments as rituximab and caplacizumab should be considered early in non-responders. Including changes in laboratory parameters may be considered in prognostic scoring systems to be developed in the future.