Heat shock proteins (HSPs) are highly conserved molecular chaperones that play a key role in maintaining protein homeostasis, or proteostasis, especially under stressful environmental conditions such as hyperthermia, hypoxia, or the presence of reactive oxygen species. In pancreatic cancer, the expression of many HSP isoforms is dysregulated, contributing to the activation of mechanisms that promote tumor development, including proliferation, invasion, angiogenesis, treatment resistance, and cancer cachexia syndrome. HSPs are significant diagnostic and prognostic biomarkers. Some of them, such as HSP27, HSP70, and HSP90, have been shown to correlate with treatment response and patient survival. Others, including HSPA2 and HSPB6, may indicate an increased risk of disease recurrence. These proteins also represent promising therapeutic targets. Preclinical and clinical studies suggest that inhibiting HSP activity and associated signaling pathways may inhibit tumor growth and increase treatment efficacy. These therapeutic effects include inducing apoptosis, autophagy, and ferroptosis, as well as sensitizing cancer cells to chemotherapy and immunotherapy. This article summarizes the current knowledge about the role of HSPs in pancreatic cancer biology, their significance as biomarkers, and their potential therapeutic applications in treating pancreatic ductal adenocarcinoma (PDAC). Most studies conducted so far have been preclinical, and due to the promising results, further clinical investigation is warranted.

Pancreatic neuroendocrine neoplasms (pNENs) are rare tumors with significant biological diversity. Despite significant improvements in diagnostics and a growing range of available therapies, long-term disease control remains difficult in advanced cases. The tumor microenvironment, which in pNENs adopts a predominantly immunosuppressive profile and promotes tumor development, is attracting increasing attention. A complex network of interactions dominates the tumor tissue, including M2 macrophages, regulatory T cells, and numerous pathways that inhibit effector lymphocyte activity. M2 macrophages, through the secretion of anti-inflammatory cytokines and exosome-mediated signaling, support angiogenesis while simultaneously attenuating the cytotoxic response. Simultaneously, receptors and ligands associated with immune checkpoints are overexpressed. In addition to classic molecules such as PD-1/PD-L1 and CTLA-4, the role of B7x and CD276 is increasingly being emphasized, as their presence correlates with rapid disease progression and poor prognosis. To date, attempts to use checkpoint inhibitors as monotherapy have yielded modest clinical benefits. However, approaches based on combination strategies—both in the form of dual immune blockade and in combination with chemotherapy or angiogenesis-targeted therapy—have shown significantly greater activity. Therapies using tyrosine kinase inhibitors, such as sunitinib and newer drugs (lenvatinib, surufatinib, cabozantinib), may partially normalize the tumor’s disrupted vascular architecture and thus increase its susceptibility to immunological interventions. In the coming years, it will be crucial not only to overcome the immunosuppressive nature of the TME but also to identify predictive biomarkers that will allow for more precise patient selection. This approach may open the way to more effective, personalized therapies for pNENs.

Heat shock proteins (HSPs) are highly conserved molecular chaperones that play a key role in maintaining protein homeostasis, or proteostasis, especially under stressful envi-ronmental conditions such as hyperthermia, hypoxia, or the presence of reactive oxygen species. In pancreatic cancer, the expression of many HSP isoforms is dysregulated, con-tributing to the activation of mechanisms that promote tumor development, including proliferation, invasion, angiogenesis, treatment resistance, and cancer cachexia syndrome. HSPs are significant diagnostic and prognostic biomarkers. Some of them, such as HSP27, HSP70, and HSP90, have been shown to correlate with treatment response and patient survival. Others, including HSPA2 and HSPB6, may indicate an increased risk of disease recurrence. These proteins also represent promising therapeutic targets. Preclinical and clinical studies suggest that inhibiting HSP activity and associated signaling pathways may inhibit tumor growth and increase treatment efficacy. These therapeutic effects include inducing apoptosis, autophagy, and ferroptosis, as well as sensitizing cancer cells to chemotherapy and immunotherapy. This article summarizes the current knowledge about the role of HSPs in pancreatic cancer biology, their significance as biomarkers, and their potential therapeutic applications in treating pancreatic ductal adenocarcinoma (PDAC). Most studies conducted so far have been preclinical, and due to the promising results, further clinical investigation is warranted.

Neuroendocrine tumors (NETs) are a diverse group of rare but clinically important neoplasms with increasing incidence and high biological heterogeneity. Heat shock proteins (HSPs) play a key role in the cellular response to stress, participating in the maintenance of proteome homeostasis and in the regulation of tumorigenesis processes such as proliferation, migration, apoptosis and development of resistance to therapy. This review analyzes the importance of HSPs in the diagnosis, prognosis and therapy of neuroendocrine tumors, as potential prognostic markers and targets of molecularly targeted therapy. The possibility of using HSP activity modulation to increase the efficacy of treatment, especially in cases refractory to standard therapeutic regimens, is emphasized. Due to the increasing clinical importance of NETs and limited therapeutic options in advanced forms, further exploration of the role of HSPs as biomarkers and therapeutic targets in this group of tumors is necessary.

Jaskra pierwotna zamkniętego kąta (primary angle-closure glaucoma – PACG) dotyka ponad 17 milionów ludzi na świecie. Płytka komora przednia i krótka długość osiowa gałki ocznej zwiększają ryzyko zamknięcia kąta przesączania, a w rezultacie rozwoju neuropatii jaskrowej. Rozwój PACG jest także związany z predyspozycją genetyczną. Do poradni okulistycznej zgłosiła się 48-letnia pacjentka. Siostra pacjentki została dwukrotnie poddana witrektomii centralnej odbarczającej z powodu zaawansowanej PACG. Pacjentce zlecono badania dodatkowe: tonometrię, optyczną koherentną tomografię przedniego odcinka oka (anterior segment optical coherence tomography – AS-OCT), gonioskopię i perymetrię statyczną. Pacjentka została zakwalifikowana do profilaktycznej laserowej irydotomii obwodowej (laser peripheral iridotomy – LPI) obu oczu. W badaniach AS-OCT i gonioskopii stwierdzono zamknięty kąt przesączania w oku prawym (oculus dexter – OD), szczelinowaty w oku lewym (oculus sinister – OS) i spłyconą komorę przednią w obu oczach. Ciśnienie wewnątrzgałkowe (intraocular pressure – IOP) wynosiło 20 mmHg w OD i 15 mmHg w OS. Stwierdzono obwodowy ubytek względny pola widzenia w OD. Tonometria kontrolna po 14 miesiącach od wykonania LPI wykazała, że IOP OD wynosi 27 mmHg. Zalecono tymolol 0,5%. Mimo to po 3 miesiącach IOP OD wynosiło 22 mmHg. Pacjentkę zakwalifikowano do powtórnego zabiegu LPI OD. W badaniu kontrolnym IOP OD wynosiło 21 mmHg. Ostrość wzroku wynosiła 1,0 w obu oczach. LPI jest metodą pierwszego rzutu w leczeniu PACG. Mimo wysokiej skuteczności zabieg nie gwarantuje długotrwałego efektu. W razie nawrotu należy rozważyć kwalifikację do zabiegu przeciwjaskrowego.

Background: Managing type 1 diabetes requires continuous adaptation to metabolic fluctuations. Standard automated systems use static rules rather than dynamic, patient-specific models. AI-driven Digital Twins (DTs) overcome this by pairing continuous multimodal data with virtual patient representations for proactive therapy optimization. Aim: This systematic review characterizes AI-driven DTs for T1D glycemic control, evaluating their underlying architectures (data-driven, mechanistic, hybrid), multimodal data integration, decision-support translation, and clinical validation levels. Methods: Following PRISMA guidelines, a PubMed search yielded 443 unique records after duplicate removal. Five independent reviewers screened articles using PICOS criteria. Ultimately, 42 studies (including 11 randomized trials) evaluating AI-driven DTs or predictive models in T1D were included for qualitative synthesis. Results: Most systems rely on continuous glucose monitoring (97.6%), while 40.5% incorporate the complete "metabolic triad" (CGM, insulin, carbohydrates). Architectures are shifting toward closed-loop paradigms combining patient-state estimation and adaptive feedback. Clinically, DT integration increased Time in Range by 7–26 percentage points (adding 3.0–6.2 hours daily), reduced Time Below Range (<4%), and lowered HbA1c by 0.32%–1.05% without severe adverse events. Capabilities range from short-term forecasting (30–120 minutes) to counterfactual simulation and autonomous closed-loop delivery. Hybrid models combining physiological mechanisms with machine learning represent the emerging standard. Conclusions: AI-driven DTs advance T1D care by enabling dynamic, personalized decision-making and automated control. However, widespread clinical implementation requires addressing technical complexity, establishing standardized architectures, and conducting robust prospective validation under real-world conditions.