Engin Eren Kavak

Dr. Engin Eren Kavak

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 February 2025 in BMC Cancer
Verified
Journal Article
Open access
PDF file
EEEfe Cem Erdat
Efe Cem Erdat
EKEngin Eren Kavak
Engin Eren Kavak
Large language models (LLMs) have shown promise in various medical applications, including clinical decision-making and education. In oncology, the increasing complexity of patient care and the vast volume of medical literature require efficient tools to assist practitioners. However, the use of LLMs in oncology education and knowledge assessment remains underexplored. This study aims to evaluate and compare the oncological knowledge of four LLMs using standardized board examination questions. We assessed the performance of four LLMs—Claude 3.5 Sonnet (Anthropic), ChatGPT 4o (OpenAI), Llama-3 (Meta), and Gemini 1.5 (Google)—using the Turkish Society of Medical Oncology’s annual board examination questions from 2016 to 2024. A total of 790 valid multiple-choice questions covering various oncology topics were included. Each model was tested on its ability to answer these questions in Turkish. Performance was analyzed based on the number of correct answers, with statistical comparisons made using chi-square tests and one-way ANOVA. Claude 3.5 Sonnet outperformed the other models, passing all eight exams with an average score of 77.6%. ChatGPT 4o passed seven out of eight exams, with an average score of 67.8%. Llama-3 and Gemini 1.5 showed lower performance, passing four and three exams respectively, with average scores below 50%. Significant differences were observed among the models’ performances (F = 17.39, p < 0.001). Claude 3.5 and ChatGPT 4.0 demonstrated higher accuracy across most oncology topics. A decline in performance in recent years, particularly in the 2024 exam, suggests limitations due to outdated training data. Significant differences in oncological knowledge were observed among the four LLMs, with Claude 3.5 Sonnet and ChatGPT 4o demonstrating superior performance. These findings suggest that advanced LLMs have the potential to serve as valuable tools in oncology education and decision support. However, regular updates and enhancements are necessary to maintain their relevance and accuracy, especially to incorporate the latest medical advancements.

14 April 2025 in BMC Cancer
Verified
Journal Article
Open access
PDF file
EKEngin Eren Kavak
Engin Eren Kavak
YÜYüksel Ürün
Yüksel Ürün
Classical Kaposi's sarcoma (CKS) is a rare angioproliferative disease associated with HHV-8, usually seen in the Mediterranean and Middle East regions. Knowing the subtypes, affected regions, and factors influencing prognosis is important for disease management. To analyze the demographic characteristics, prognostic factors, treatment modalities, and survival of patients diagnosed with CKS. Our center's records of patients diagnosed with CKS between January 2010 and December 2021 were retrospectively analyzed. Thirty-eight patients with histopathologically proven CKS were included in the study. Demographic and clinical characteristics of the patients, macroscopic, histopathologic, and immunohistochemical features of the lesions, treatments, and responses to treatment were evaluated. Kaplan–Meier survival curves were used to estimate survival outcomes, and log-rank test analyses were performed for intergroup comparisons. The median age at diagnosis of the patients was 71.0(39.0–93.0) years. Ten patients were female, and 28 were male. At the time of diagnosis, 63.2% of the patients had localised disease, nine patients were locally advanced, and five patients were metastatic. The tumor was most commonly localised in the lower extremity (65.8%), followed by the upper extremity. The median follow-up period was 69 (49–77.6) months. Local recurrence was detected in 24 patients during the follow-up. Median overall survival was not reached (NR) in localised disease(95% CI: 70.5-NR). In locally advanced disease, it was 31.1 months (95% CI: 13.8–63.0). In metastatic disease, it was 16.3 (95% CI: 12.6–20.0) months (p = 0.005). This study emphasizes that CKS in our centre predominantly affects older males and typically manifests with nodular, early-stage lesions at the time of diagnosis. The majority of patients exhibited localised disease with no evidence of systemic involvement, while lymphedema was a frequent accompanying condition. Ulcerative manifestations were relatively uncommon, and survival outcomes varied significantly based on disease stage, with a marked decline in overall survival for patients with metastatic disease. The findings emphasize the importance of early diagnosis and the development of tailored treatment strategies to improve patient outcomes.

3 November 2023 in Biomolecules and Biomedicine
Verified
Journal Article
Open access
PDF file
SASabin Goktas Aydin
Sabin Goktas Aydin
EKEngin Eren Kavak
Engin Eren Kavak
ATAtakan Topcu
Atakan Topcu
ABAyberk Bayramgil
Ayberk Bayramgil
AAAkgül Akgül
Akgül Akgül
İÇİrfan Çiçin
İrfan Çiçin
Regorafenib, an oral multikinase inhibitor, has improved survival in metastatic colorectal cancer (mCRC) patients who have progressed on standard therapies. Our study aimed to evaluate prognostic factors influencing regorafenib treatment and assess the optimal dosing regimen in a real-life setting. We retrospectively analysed 263 patients with mCRC from multiple medical oncology clinics in Turkey. Treatment responses and prognostic factors for survival were evaluated using univariate and multivariate analysis. Of the patients, 120 were male, and 143 were female; 28.9% of tumors were located in the rectum. RAS mutations were present in 3.0% of tumors, while BRAF, K-RAS, and N-RAS mutations were found in 3.0%, 29.7%, and 25.9% of tumor tissues, respectively. Dose escalation was preferred in 105 (39.9%) patients. The median treatment duration was 3.0 months, with an objective response rate (ORR) of 4.9%. Grade ≥ 3 treatment-related toxicity occurred in 133 patients, leading to discontinuation, interruption, and modification rates of 50.6%, 43.7%, and 79.0%, respectively. Median progression-free survival (PFS) and overall survival (OS) were 3.0 and 8.1 months, respectively. RAS/RAF mutation (hazard ratio [HR] 1.5, 95% confidence interval [CI] 1.1-2.3; P = 0.01), pretreatment carcinoembryonic antigen (CEA) levels (HR 1.6, 95% CI 1.1-2.3; P = 0.008), and toxicity-related treatment interruption or dose adjustment (HR 1.6, 95% CI 1.1-2.4; P = 0.01) were identified as independent prognostic factors for PFS. Dose escalation had no significant effect on PFS but was associated with improved OS (P < 0.001). Independent prognostic factors for OS were the initial TNM stage (HR 1.3, 95% CI 1.0-1.9; P = 0.04) and dose interruption/adjustment (HR 0.4, 95% CI 0.2-0.9; P = 0.03). Our findings demonstrate the efficacy and safety of regorafenib. Treatment line influences the response, with dose escalation being more favorable than adjustment or interruption, thus impacting survival.

18 November 2025 in Discover Oncology
Verified
Journal Article
Open access
PDF file
MYMerih Yalçıner
Merih Yalçıner
EEEfe Cem Erdat
Efe Cem Erdat
EKEngin Eren Kavak
Engin Eren Kavak
GUGüngör Utkan
Güngör Utkan
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.

1 April 2024 in Turkish Journal of Osteoporosis
Verified
Journal Article
Closed access
SKsongül keskin kavak
songül keskin kavak
LALale Aktekin
Lale Aktekin
EKEngin Eren Kavak
Engin Eren Kavak
BSBurcu Sevingül
Burcu Sevingül
MÖMerve Önoğul
Merve Önoğul

14 August 2024 in Clinical and Translational Oncology
Verified
Journal Article
Closed access
PDF file
EKEngin Eren Kavak
Engin Eren Kavak
IDİsMail Dilli
İsMail Dilli
GYGüler Yavaş
Güler Yavaş