Explore the vast range of titles available.

Background Lymphomas are ranked as the fifth most common cancer in Lebanon. There is concern about the need of information regarding the prevalence of lymphoid neoplasm particularly Non-Hodgkin lymphoma (NHL) subtypes in the Lebanese population. This study intended to establish a descriptive status of NHL histological subtypes distribution in Lebanon thus identifying the most common types, knowing that the literature is poor regarding the distribution of lymphoid malignancies particularly NHLs in Lebanon. Methods A bicenter retrospective descriptive study was performed. Patients aged above 18, diagnosed with NHL between January 1984 and March 2019 and registered in two Lebanese Medical centers were included in this study; 699 medical files were reviewed and the baseline characteristics of the disease were collected. Histological classification was based on the Working Formulation (WF) and World Health Organization (WHO) classification systems, whereas staging was based on the Ann Arbor system. Disease status was monitored with imaging studies. Results The mean age at diagnosis was 53.52 ± 17.46 years in the studied population, with 380 (54.4%) males and 319 (45.6%) females. B-cell lymphoma (BCL) accounted for 86.3% while T-cell neoplasms accounted for 13.7%. The most common subtype was diffuse large B-cell lymphoma (DLBCL) (54%) followed by follicular lymphoma (FL) (17.2%). Mantle cell lymphoma (MCL) represented 3% of all BCL and small lymphocytic lymphoma (SLL) comprised less than 2%. Mucosa-associated lymphoid tissue (MALT) and Burkitt’s lymphomas represented 3 and 1.7% respectively. 36.5% of the patients had extranodal disease at diagnosis. High-grade tumor represented 80.1% with 33.1% stage IV disease. Conclusion These observations indicate that the epidemiological patterns of NHLs in Lebanon were comparable to Western countries. Aggressive lymphomas account for the majority of NHLs in Lebanon.

Vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic (VEXAS) syndrome is a rare, adult-onset autoinflammatory disease that has only been described since late 2020. Given the rarity of the disease and the absence of established treatment guidelines, management remains challenging and largely based on clinical experience and case reports. This paper reports the first documented case of VEXAS syndrome in Lebanon. A 73-year-old man presented with fever, severe asthenia, erythematous skin lesions, and anemia, initially diagnosed as giant cell arteritis. Following a diagnosis of VEXAS syndrome, he was treated with corticosteroids and methotrexate but developed refractory anemia and required erythropoietin therapy. Tocilizumab was introduced to manage inflammation, but the patient's condition remained challenging due to corticosteroid dependence and myelodysplastic syndrome (MDS). Given the patient's comorbidities and intermediate-risk MDS, azacitidine was initiated as a therapeutic option. Despite initial neutropenia and infections, adjustments to the azacitidine regimen led to significant clinical and hematologic improvements. The patient achieved complete remission, became transfusion-independent, and maintained stable hemoglobin levels. This case highlights the efficacy of azacitidine in managing VEXAS syndrome with MDS, particularly among patients ineligible for hematopoietic stem cell transplantation, offering a potential pathway to sustained remission and reduced corticosteroid dependence.

Background: For metastatic colorectal cancer a series of novel therapies has emerged during the last decade but their use in routine clinical practice and their costs are not well documented. Aims: This study evaluated the clinical effectiveness of metastatic colorectal cancer patients in Lebanese oncologic units and estimated the costs. Methods: A prospective cohort study was conducted on metastatic colorectal cancer patients during 2008-2013. The type of medical management, overall survival and total costs from diagnosis to death were described. Cost analysis was performed using tariffs from 2013 in US dollars. Results: One hundred and seventy-nine metastatic patients were selected among which 84.9% had colorectal cancer involvement. The average follow-up from diagnosis until death or the latest news was 34.8 months. Around 49.7% were still alive at last follow-up date. Three lines of treatment accounted for 4.5%, 39.6% and 55.9% with an average duration of 14.5, 11.4 and 14.6 months respectively. A 73.2% of patients benefited from targeted therapy. The median overall survival was 20.8 months. The mean total costs of drugs was $22 256 in patients with standard therapy alone whereas the cost increased to $80 396 after the addition of targeted therapy. The mean global total cost was estimated at $64 805 per patient (min $3703; max $304 086). Conclusions: Targeted therapy associated to standard therapy is highly prevalent in Lebanon in metastatic disease and the associated medical cost substantial. This study is the first to show the clinical effectiveness and costs of targeted therapy in patients with metastatic colorectal cancer. Keywords: metastatic colorectal cancer, cytotoxic agents, targeted therapies, effectiveness, cost.

Adopting a healthy lifestyle during cancer treatment enhances patients’ outcomes. The purpose of this study is to evaluate the current practices and attitudes adopted by hematologists/oncologists on providing lifestyle recommendations to cancer patients. The secondary objective includes a correlation between the hematologists/oncologists’ sociodemographic with their provision of lifestyle recommendations to their patients. This prospective, cross-sectional study surveyed Lebanese hematologists/oncologists in five major Lebanese governorates. The questionnaire collected information on participants’ demographics and personal lifestyle choices as well as practices and perceptions related to lifestyle recommendations provided to cancer patients. A total of 40 hematologists/oncologists practicing in Lebanon completed the questionnaire with a response rate of 33.3%. The top three recommendations that hematologists/oncologists provided to their patients include quit smoking (95%), increase your physical activity (92.5%), and improve your nutrition/diet (85%). The mean number of recommendations provided per hematologist/oncologist was 3.88 (± 1.067). These discussions were consistently provided during clinical encounters with patients. The most frequently reported barriers hindering patient education on lifestyle recommendations included practitioners’ lack of time and knowledge and patients’ advanced stages of cancer and lack of interest in the topic. Hematologists/oncologists perceived these recommendations to be beneficial to patients’ mental health and performance status, while few of them identified other benefits. Only one statistically significant correlation was identified between hematologists/oncologists’ sociodemographic and providing the lifestyle recommendations to their patients. To achieve higher quality patient-centered care, communication gaps between hematologists/oncologists and cancer patients should be addressed, and solutions to identified barriers should be implemented.

Late detection and lack of standard treatment strategies in larynx cancer patients result in high levels of mortality and poor prognosis. Prognostic stratification of larynx cancer patients based on molecular prognostic tumor biomarkers may lead to more efficient clinical management. Krüppel-like factor 4 (KLF4) and Heat Shock Protein 27 (HSP27) have an important role in tumorigenesis and are considered promising candidate biomarkers for various types of cancer. However, their role in larynx carcinoma remains to be elucidated. The present study aimed to determine KLF4 and HSP27 expression profiles in laryngeal tumors. The protein and mRNA expression levels of KLF4 and HSP27 were evaluated by immunohistochemical and reverse transcription-polymerase chain reaction analyses in 44 larynx carcinoma samples and 21 normal tissue samples, and then correlated with clinical characteristics. A differential expression of KLF4 and HSP27 was observed between normal and tumor tissues. The protein and mRNA expression levels of KLF4 were significantly decreased in larynx squamous cell carcinoma (LSCC) compared with normal tissue, whereas HSP27 was significantly overexpressed in tumor tissues compared with normal tissues, at the protein and mRNA levels. KLF4 expression decreased gradually with tumor progression whereas HSP27 expression increased. A significant difference was observed between stages I and IV. KLF4 and HSP27 exhibit opposite functions and roles in the carcinogenic process of LSCC. Their role in laryngeal cancer initiation and progression emphasizes their use as potential future targets for prognosis and treatment. KLF4 and HSP27 expression levels may act as potential biomarkers in patients with cancer of the larynx.

Score-function based methods for policy learning, such as REINFORCE and PPO, have delivered strong results in game-playing and robotics, yet their high variance often undermines training stability. Using pathwise policy gradients, i.e. computing a derivative by differentiating the objective function, alleviates the variance issues. However, they require an accurate action-conditioned value function, which is notoriously hard to learn without relying on replay buffers for reusing past off-policy data. We present an on-policy algorithm that trains Q-value models purely from on-policy trajectories, unlocking the possibility of using pathwise policy updates in the context of on-policy learning. We show how to combine stochastic policies for exploration with constrained updates for stable training, and evaluate important architectural components that stabilize value function learning. The result, Relative Entropy Pathwise Policy Optimization (REPPO), is an efficient on-policy algorithm that combines the stability of pathwise policy gradients with the simplicity and minimal memory footprint of standard on-policy learning. Compared to state-of-the-art on two standard GPU-parallelized benchmarks, REPPO provides strong empirical performance at superior sample efficiency, wall-clock time, memory footprint, and hyperparameter robustness.

Building deep reinforcement learning (RL) agents that find a good policy with few samples has proven notoriously challenging. To achieve sample efficiency, recent work has explored updating neural networks with large numbers of gradient steps for every new sample. While such high update-to-data (UTD) ratios have shown strong empirical performance, they also introduce instability to the training process. Previous approaches need to rely on periodic neural network parameter resets to address this instability, but restarting the training process is infeasible in many real-world applications and requires tuning the resetting interval. In this paper, we focus on one of the core difficulties of stable training with limited samples: the inability of learned value functions to generalize to unobserved on-policy actions. We mitigate this issue directly by augmenting the off-policy RL training process with a small amount of data generated from a learned world model. Our method, Model-Augmented Data for TD Learning (MAD-TD), uses small amounts of generated data to stabilize high UTD training and achieve competitive performance on the most challenging tasks in the DeepMind control suite. Our experiments further highlight the importance of employing a good model to generate data, MAD-TD's ability to combat value overestimation, and its practical stability gains for continued learning.

Score-function based methods for policy learning, such as REINFORCE and PPO, have delivered strong results in game-playing and robotics, yet their high variance often undermines training stability. Using pathwise policy gradients, i.e. computing a derivative by differentiating the objective function, alleviates the variance issues. However, they require an accurate action-conditioned value function, which is notoriously hard to learn without relying on replay buffers for reusing past off-policy data. We present an on-policy algorithm that trains Q-value models purely from on-policy trajectories, unlocking the possibility of using pathwise policy updates in the context of on-policy learning. We show how to combine stochastic policies for exploration with constrained updates for stable training, and evaluate important architectural components that stabilize value function learning. The result, Relative Entropy Pathwise Policy Optimization (REPPO), is an efficient on-policy algorithm that combines the stability of pathwise policy gradients with the simplicity and minimal memory footprint of standard on-policy learning. Compared to state-of-the-art on two standard GPU-parallelized benchmarks, REPPO provides strong empirical performance at superior sample efficiency, wall-clock time, memory footprint, and hyperparameter robustness.