Simulation sessions can produce high-fidelity emergency situations that facilitate the learning process. These sessions may also generate a complex stress response in the learners. This prospective observational study assessed psychological, physiological, immunological, and humoral levels of stress during high-fidelity simulation training. Fifty-six undergraduate medicine students who took part in a medical simulation session were assigned team roles (physician, nurse or assistant). Subsequently, each participant was assessed before the scenario (T0), after the procedure (T1), and two hours later (T2). Psychological stress and anxiety were measured at T0 and T1, using the State-Trait Anxiety Inventory (STAI) and Dundee Stress State Questionnaire (DSSQ). Cortisol, testosterone, secretory immunoglobulin class A (sIgA), alpha-amylase, and oxygen saturation level were measured at T0, T1, and T2, as was the physiological response indicated by heart rate (HR) and blood pressure (BP). It was found that the onset of task performance was related to increased anticipatory worry and higher oxygen saturation. The participants reported decreased worry, followed by increased emotional distress after the simulation training (T1). Participants trait anxiety predicted the intensity of worry, distress and task engagement. In contrast, no clear relationships were found between trait anxiety and biological stress markers. Testosterone levels were growing significantly in each phase of measurement, while physiological responses (BP, HR) increased at T1 and declined at T2. The levels of stress markers varied depending on the assigned roles; however, the trajectories of responses were similar among all team members. No evidence for prolonged cortisol response (T1, T2) was found based on psychological stress at the onset of simulation (T0). Regression analysis followed by receiver operating characteristics analyses showed uncertain evidence that initial state anxiety and worry predicted the levels of sIgA. Medical students are relatively resilient in terms of stress responses to medical simulation. The observed stress patterns and interrelationships between its psychological, physiological, hormonal, and immunological markers are discussed in accordance with theoretical concepts, previous research work, and further recommendations.