Entropy production in stochastic thermodynamics is defined as the average over all the possible system trajectories of a quantity (observable) which measures the divergence between a trajectory and its time-reversal. In this way a quantitative measure of the irreversible evolution of a system out of equilibrium can be given. In this note we consider a Markov system and substitute the linear average with a more general notion of average over the trajectories (Kolmogorov-Nagumo mean) which corresponds to substituting Shannon entropy with Rény entropy. We show that, unlike the linear average case, the resulting notion of entropy production depends on how the observable is defined.

Antibodies approved by the FDA are failing to treat neurological disorders, and recently, it was reported that these trials may have violated patient’s rights and subjected them to high, likely lethal risk. The challenge with developing antibodies to treat neurological orders is their almost negligible bioavailability, requiring high dosing that can be toxic. The high potency of these drugs should also be viewed considering the placebo effects since all antibodies have shown severe side effects that are not prevented by the placebo responses. In this critical and urgent advice to the FDA, I am suggesting a guideline amendment to all clinical trials requiring proof of sufficient bioavailability at the site of action, where it is known. For antibodies to cross the blood-brain barrier, there are proven solutions such as conjugating with transferrin protein, making clinical trials in its absence more questionable and unethical.

The growing number of algorithmic decision-making environments, which blend machine and bounded human rationality, strengthen the need for a holistic performance assessment of such systems. Indeed, this combination amplifies the risk of local rationality, necessitating a robust evaluation framework. We propose a novel simulation-based model to quantify algorithmic interventions within organisational contexts, combining causal modelling and data science algorithms. To test our framework's viability, we present a case study based on a bike-share system focusing on inventory balancing through crowdsourced user actions. Utilising New York's Citi Bike service data, we highlight the frequent misalignment between incentives and their necessity. Our model examines the interaction dynamics between user and service provider rule-driven responses and algorithms predicting flow rates. This examination demonstrates why these dynamics are necessary for devising effective incentive policies. The study showcases how sophisticated machine learning models, with the ability to forecast underlying market demands unconstrained by historical supply issues, can cause imbalances that induce user behaviour, potentially spoiling plans without timely interventions. Our approach allows problems to surface during the design phase, potentially avoiding costly deployment errors in the joint performance of human and AI decision-makers.

: Ovarian cancer has a dismal prognosis. Standard treatment following surgery relies on platinum-based chemotherapy. However, sizeable percentages of patients are unresponsive. Identification of markers predicting response to chemotherapy might help select eligible patients while sparing unresponsive ones treatment-associated toxicity. Cancer/testis antigens (CTA) are expressed by healthy germ cells and malignant cells of diverse histological origin. This expression profile identifies them as attractive targets of cancer immunotherapies. We analyzed correlations between expression of MAGE-A10 and New York esophageal-1 cancer (NY-ESO-1) CTAs at protein level and effectiveness of platinum-based chemotherapy in patients with advanced-stage high-grade serous ovarian carcinoma (HGSOC). MAGE-A10 and NY-ESO-1 protein expression was analyzed by immunohistochemistry (IHC) in formalin-fixed, paraffin-embedded samples from 93 patients with advanced-stage HGSOC treated at our institutions between January 1996 and December 2013. Correlation between expression of these markers and response to platinum-based chemotherapy, evaluated according to RECIST 1.1 criteria, platinum sensitivity, measured as platinum free interval (PFI), progression free (PFS) and overall survival (OS) was explored. MAGE-A10 protein expression predicts unresponsiveness to platinum-based chemotherapy (p=0.005), poor platinum sensitivity (p<0.001), and poor PFS (p<0.001) and OS (p<0.001). Multiva-riate analysis identifies MAGE-A10 protein expression as independent predictor of poor platinum sensitivity (p=0.005) and shorter OS (p<0.001). Instead, no correlation was observed between NY-ESO-1 protein expression and response to platinum-based chemotherapy (p=0.832), platinum sensitivity (p= 0.168), PFS (p=0.126) and OS (p=0.335). MAGE-A10 protein expression reliably identifies advanced-stage HGSOC unresponsive to platinum-based chemotherapy. Targeted im-munotherapy could represent an important alternative therapeutic option in these cancers.

Background: PRAME, NY-ESO-1 and SSX2 are cancer testis antigens (CTAs), which in normal tissues are expressed in testicular germ cells with re-expression in numerous cancer types. Their ability to elicit humoral and cellular immune responses have rendered them promising targets for cancer immunotherapy, but they have never been studied in a large and well-characterized cohort soft tissue sarcomas (STS). Methods: On protein level, we examined PRAME, NY-ESO-1 and SSX2 expression in tumour tissues of 249 STS using immunohistochemistry. We correlated expression levels with clinicopathological parameters including Tumour-infiltrating lymphocyte (TIL) counts, grading and long- term survival. Results: Expression of PRAME, NY-ESO-1 and SSX2 was observed in 25 (10%), 19 (8%), and 11 (4%) of 249 specimens with distinct patterns for histo subtypes. Expression of PRAME was associated with shorter patient survival (p=0.005) and higher grade (G2 vs G3, p=0.001) while NY-ESO-1 expression was correlated with more favourable survival (p=0.037) and low grade (G2 vs G3, p=0.029). Both PRAME and NY-ESO-1 expression was more frequent in STS with low TILs counts. Conclusions: CTAs PRAME, NY-ESO-1 and SSX2 show distinct expression patterns in different STS subtypes. These results may guide future immunotherapeutic approaches in STS.