Medicine and Pharmacology

Abstract: Persistent therapeutic variability remains a central challenge in cancer therapy. Contemporary precision oncology has addressed this problem primarily through molecular targeting and patient stratification. Although these approaches have substantially improved therapeutic prediction and patient selection, they remain fundamentally limited in their ability to resolve therapeutic variability, to overcome resistance, and to convert nonresponders to responders.Based on a recently proposed state-space and trajectory-evolution framework for cancer therapy, which emphasizes therapeutic trajectory evolution as the central determinant of outcome, I extend this formulation by introducing a control-oriented extension termed state-space modulation, aimed at modulating state-space organization to regulate trajectory evolution and thereby establish a unified and extensible theoretical structure for cancer therapy. In that formulation, therapeutic outcomes are interpreted not as direct consequences of therapeutic inputs, but as emergent consequences of the therapeutic trajectory evolution initiated by those inputs in the accessible state-space organization. State-space organization refers to the global organization of the tumor–host system arising from biological and physicochemical variables and the composition, relationships, interactions, and constraints among them. Because therapeutic trajectory evolution occurs within and is constrained by this state-space organization, identical therapeutic inputs can generate different trajectory evolution under different state-space organizations, ultimately producing different therapeutic outcomes, including therapeutic response and nonresponse. Under this perspective, therapeutic variability, sensitivity, and resistance emerge as manifestations of the therapeutic trajectory evolution. Since therapeutic variability, sensitivity, and resistance arise from differences in state-space organization, modification of state-space organization represents a rational approach for therapeutic control. State-space modulation is defined as the intentional modification of tumor–host state-space organization to redirect therapeutic trajectory evolution toward more favorable therapeutic outcomes. To achieve this objective, state-space modulation operates through state-space modulating operators, which act primarily through system-level reorganization of the conditions under which therapeutic trajectory evolution occurs. Such reorganization could be achieved through modulation of fundamental organizing variables that serve as modulating interfaces for global state-space modification. Among currently identifiable candidates, pH, oxygen availability, and temperature appear particularly important because of their capacity to induce coordinated changes across multiple biological dimensions. Our previous experimental and clinical studies demonstrate that bicarbonate-mediated pH modulation in hepatocellular carcinoma markedly increases the therapeutic efficacy of TACE and anti-PD-1 therapy and support the state-space modulation concept. In summary, state-space modulation extends therapeutic intervention into a new dimension, providing a conceptual framework that complements precision oncology by rational modification of tumor–host state-space organization. This perspective may guide the development of future therapeutic strategies aimed at improving efficacy, reducing therapeutic variability, converting nonresponders to responders, overcoming resistance, and promoting convergence toward optimal clinical outcomes.

Abstract: Feature engineering remains a major challenge in metabolomics-based prediction, particularly when rich biochemical knowledge is available but underutilized. Conventional metabolomics models rely primarily on measured variables and statistically driven feature selection, often overlooking the molecular and pathway context encoded in curated metabolite knowledge bases. Here, we propose SEFA (Semantic Embedding–based Feature Augmentation), a model agnostic feature engineering framework that integrates metabolite level textual knowledge into structured metabolomics and clinical modeling for lung cancer. SEFA leverages semantic embeddings derived from external metabolite resources to enrich feature construction through embedding based semantic projection and knowledge-guided feature selection, yielding a compact and biologically interpretable feature representation. Using a targeted lung cancer metabolomics dataset, we evaluate SEFA across multiple learning algorithms, including linear, kernel based, tree based, and neural network models, and show that embedding informed feature construction consistently improves prediction performance over approaches that rely solely on tabular metabolite and clinical features. Analysis of the selected features revealed enrichment in pathways including arginine and proline metabolism and glycine, serine, and threonine metabolism, indicating that the embedding derived feature representations capture biologically relevant metabolic processes associated with lung cancer. These results support the use of semantic knowledge in metabolomics feature engineering and position SEFA as a practical, general machine learning framework for enhancing predictive modeling while preserving biological interpretability through pathway level analysis.

Abstract: Colorectal cancer (CRC) ranks as the third most common cancer and second leading cause of cancer-related mortality worldwide with significant geographic disparities and rising burden in regions like Nepal. Chronic inflammation like inflammatory bowel disease (IBD) drives colorectal carcinogenesis through key pathways including NF-κB, COX-2/PGE2, IL-6/STAT3, and IL-23/Th17. They create a pro-tumorigenic microenvironment that promotes proliferation, angiogenesis, invasion, and metastasis.The purpose of this study is to address the protective function of dietary fibers, which are converted into short-chain fatty acids (SCFAs), mainly butyrate, acetate, and propionate, by the gut microbiota. By blocking HDACs, inhibiting NF-κB and STAT3 signaling, strengthening intestinal barrier integrity through tight junctions and mucin production, regulating immune responses, and influencing epigenetic processes like miRNA regulation, these metabolites have strong anti-inflammatory, anti-proliferative, and pro-apoptotic effects. Higher fiber intake is consistently linked to a lower risk of colorectal cancer (CRC), mostly through microbiota-dependent SCFA generation, according to preclinical models (such as AOM/DSS and Apc-mutant mice) and epidemiological data.The paper also discusses potential improvements to anti-inflammatory treatments, synergies with probiotics (synbiotics), and obstacles like the need for extensive, long-term clinical trials, inter-individual microbiome variations, and variety in fiber types and sources. Optimizing fiber-based therapies may be possible with precision nutrition strategies based on microbial profiles. In conclusion, by interfering with inflammation-driven carcinogenesis, dietary fibers offer a secure, practical, and economical approach to CRC prevention and treatment. To lessen the worldwide incidence of this cancer, future research should focus on tailored therapies, targeted prebiotics, and incorporation into all-encompassing preventive programs.

Abstract: Lysine β-hydroxybutyrylation (Kbhb) is a metabolite-derived post-translational modification of histone and non-histone proteins that couples β-hydroxybutyrate (BHB) availability to gene expression. Yet the transcription factors that govern the Kbhb substrate program in cancer remain unidentified. Existing studies have cataloged Kbhb-modified substrates or examined individual proteins, without identifying the transcriptional regulators of the program in a defined tumor context. Here, we performed network-based Master Regulator Analysis (MRA), implemented in the viper package, on a molecular signature restricted to experimentally validated Kbhb substrates, across two independent Basal-like breast cancer cohorts profiled on orthogonal platforms: TCGA-BRCA (RNA-seq; n = 195 tumor, 113 normal) and METABRIC (microarray; n = 209 tumor, 148 normal). Dataset-specific regulatory networks were inferred with ARACNe-AP and integrated by cross-platform Stouffer meta-analysis. Of 1,493 Kbhb substrates, 1,322 and 1,213 were expressed in the respective cohorts. The analysis identified seven concordant transcriptional master regulators (six activated, one repressed; cross-cohort NES correlation r = 0.64), with CENPA (meta-NES +4.55) and FOXM1 (meta-NES +4.27) as the dominant drivers. These findings nominate a BHB–Kbhb–FOXM1/CENPA axis linking ketone-body metabolism to mitotic transcription, with potentially protumoral implications for ketogenic regimens in triple-negative breast cancer.

Abstract: Glypican-4 (GPC4) is a member of the heparan sulfate proteoglycan family and regulates a wide range of growth factor signaling through interactions with heparan sulfate-binding ligands. Pathogenic variants of GPC4 have been identified in human congenital disorders, and dysregula-tion of GPC4 has been reported in human cancer. Therefore, specific anti-GPC4 monoclonal anti-bodies (mAbs) are needed for both basic research and clinical applications. In this study, we estab-lished novel anti-glypican-4 (GPC4) mAbs using a flow cytometry–based high-throughput screening approach. Among the isolated clones, G₄Mab-9 (IgG_2b, κ) showed specific binding to Chinese hamster ovary-K1 (CHO/GPC4) cells overexpressed GPC4, with no reactivity against pa-rental CHO-K1. Furthermore, G₄Mab-9 detected endogenous GPC4 expression in human embry-onic kidney 293FT cells. Specificity analyses demonstrated that G₄Mab-9 selectively recognized GPC4 without cross-reactivity with other glypicans in CHO-K1 cells overexpressing them. The dissociation constant (K_D) of G₄Mab-9 was determined to be 1.4 × 10⁻7 M for CHO/GPC4, indicat-ing low binding affinity. Moreover, G₄Mab-9 detected a 33-kDa cleaved fragment of GPC4 by western blotting and stained CHO/GPC4 but not CHO-K1 in immunohistochemistry. Collectively, these results indicate that G₄Mab-9 is a valuable tool for GPC4-related basic research and a poten-tial candidate for the development of diagnostic and therapeutic approaches.

Abstract: Disulfidptosis is a newly identified form of regulated cell death driven by high SLC7A11 expression under glucose starvation. Mechanistically, massive cystine influx via SLC7A11, coupled with NADPH depletion due to glucose deprivation, leads to cystine accumulation, disulfide stress, aberrant crosslinking of actin cytoskeleton (notably at Cys257, Cys284, and Cys373), and eventual membrane rupture. This process is independent of apoptosis, ferroptosis, and necroptosis. This review systematically outlines the molecular regulatory network of disulfidptosis, including upstream triggers, key execution molecules (glutathione and thioredoxin systems, RAC1/WRC/Arp2/3 axis), and downstream effectors. We further evaluate its therapeutic potential across various malignancies (glioblastoma, hepatocellular, gastric, colorectal, pancreatic, breast, lung, and prostate cancers) and other diseases (neurodegenerative, metabolic, orthopedic, cardiovascular). Recent advances in nanodelivery systems for selective disulfidptosis induction are summarized. Key translational challenges are critically analyzed, including targeting specificity, safety, delivery barriers, resistance mechanisms, and biomarker deficiency. By exploiting metabolic vulnerabilities (high SLC7A11 expression and glucose dependence) of pathological cells, disulfidptosis opens a new therapeutic window that can synergize with ferroptosis, cuproptosis, and immunotherapy. This review provides a comprehensive framework for basic research and clinical translation of disulfidptosis-targeted therapies.

Abstract: Background/Objectives: Ubiquitin-specific protease 7 (USP7) is a deubiquitinating enzyme that regulates multiple oncogenic and tumor-suppressive pathways and has emerged as a promising target for anticancer drug discovery. However, most reported USP7 inhibitors belong to a limited number of structural classes, and no USP7-targeted therapy has yet reached clinical application. This study aimed to identify structurally diverse USP7 inhibitors with anticancer potential using an integrated computational and experimental screening strategy. Methods: A drug discovery workflow combining pharmacophore modelling, structure-based virtual screening, molecular docking, and in silico pharmacokinetic assessment was employed to identify candidate USP7 inhibitors. Selected compounds were evaluated for inhibition of recombinant USP7 and for antiproliferative activity in a panel of human cancer cell lines. Molecular docking analyses were performed to investigate predicted binding modes. Results: The screening campaign identified 32 hit compounds with in vitro inhibitory activity against recombinant USP7, including structurally distinct approved drugs and compounds from an in-house chemical library. Four selected hits exhibited antiproliferative activity across cancer cell lines with different molecular backgrounds. Docking studies predicted binding within a pharmacologically relevant region of the USP7 catalytic cleft and revealed interactions with key residues involved in ligand recognition. Among the compounds evaluated in cells, rafoxanide demonstrated the most favorable combination of predicted USP7 engagement and antiproliferative activity. Conclusions: This integrated virtual screening and experimental validation approach enabled the identification of structurally diverse USP7 inhibitors with antiproliferative activity. The identified hits include approved drugs with previously unrecognized USP7 inhibitory activity, as well as underexplored scaffolds that expand the chemical space of USP7-targeting molecules. This provides a foundation for the future optimization and development of USP7 targeted anticancer agents.

Abstract: Uterine leiomyosarcoma (uLMS) is an aggressive uterine cancer with limited durable benefit from current systemic therapy. Genomic profiling shows recurrent inactivation of TP53, RB1, and ATRX, leading to genomic instability, complex copy-number alterations, and telomere dysfunction via alternative lengthening of telomeres. Additional events in the PTEN and PI3K–AKT–mTOR pathways occur, and a minority of tumors harbor homologous recombination repair defects, most often involving BRCA2, with RAD51B being reported less frequently. These features create actionable liabilities rather than a single dominant driver. Opportunities include DNA damage response approaches such as PARP inhibition in HRR-deficient disease, and checkpoint targeting of replication stress through ATR, CHK1, or WEE1. Pathway-directed strategies targeting the PI3K-AKT-mTOR axis, as well as epigenetic or transcriptional interventions, and endocrine therapy for selected ER- or PR-positive, indolent, low-volume tumors broaden treatment options. Circulating tumor DNA is emerging as a tool to monitor response. We propose biomarker-enriched trial designs, such as basket or umbrella trials, that assign patients to genotype-defined cohorts and integrate endpoints, including HRD and ATRX or ALT, aiming to transition from empiric therapy toward molecular selection in uLMS. These studies should also incorporate patient-reported outcomes, health-related quality of life metrics, and time-to-next-treatment measures to better capture clinically meaningful benefit of precision oncology strategies.

Abstract:

Background: Breast cancer (BC) frequently disseminates to bone. Technetium 99m methylene diphosphonate bone scintigraphy (^99m Tc‐MDP bone scintigraphy) is utilized in the staging work‐up. In detecting lytic bone lesions, bone marrow and soft tissue metastases, 2‐deoxy‐2[fluorine 18] fluoro‐D‐glucose positron emission tomography/computed tomography scan (¹⁸F‐FDG PET/CT scans) is superior. We compared bone metastases (BM) detection in both imaging modalities, and its impact on patient management. Method: One hundred BC patients underwent ¹⁸F‐FDG PET/CT scan, and ^99m Tc‐MDP bone scintigraphy performed within a maximum of six weeks of each other, between January 2017 and September 2024. Comparison based on lesion‐by‐lesion analysis. Changes in stage and management recorded. Results: The median patient age was 55 years. ^99m Tc‐MDP bone scintigraphy detected 243 BM and ¹⁸F‐FDG PET/CT scans detected 421 BM. Stage and management upgrades, secondary to detected BM, were seen in 22 patients for each imaging modality. ¹⁸F‐FDG PET/CT scans additionally detected 5 patients with unsuspected, isolated soft tissue metastases, all resulting in stage and management change (p<0.001). Conclusion: Both ¹⁸F‐FDG PET/CT scan and ^99m Tc‐MDP bone scintigraphy demonstrated the ability to detect BM in an equivalent number of patients. However, ¹⁸F‐FDG PET/CT scan proved superior in overall metastatic assessment, particularly in identifying additional BM and concomitant soft tissue metastases. This broader disease characterization translated into significantly higher rate of management modification compared with ^99m Tc‐MDP bone scintigraphy. Contribution: A proposed revision of current BC guidelines reflecting ¹⁸F‐FDG PET/CT scan’s validated diagnostic superiority, will reduce financial burden whilst improving patient management and compliance.

Abstract: This study aimed to characterize claudin 18.2 (CLDN18.2) expression in HER2 negative gastric or gastroesophageal junction cancer (GC/GEJC) and to evaluate whether CLDN18.2 status is associated with clinicopathological features and outcomes after first line chemoimmunotherapy. We retrospectively analyzed 189 patients with HER2 negative GC/GEJC treated at our institution from October 2019 to September 2024. CLDN18.2 expression was assessed by immunohistochemistry using two prespecified positivity thresholds: moderate to strong membranous staining (2+) in ≥40% or ≥75% of tumor cells. CLDN18.2 positivity was observed in 92/189 patients (48.7%) using the ≥40% threshold and 69/189 (36.5%) using the ≥75% threshold. PD L1 CPS ≥5 was less frequent in CLDN18.2 positive than in CLDN18.2 negative tumors at both thresholds (≥40%: 8.7% vs. 23.7%, P = 0.003; ≥75%: 8.7% vs. 20.8%, P = 0.019). Among 87 patients receiving first line chemoimmunotherapy, CLDN18.2 status was not associated with significant differences in objective response rate, progression free survival, or overall survival. CLDN18.2 positive tumors showed lower PD L1 expression, but CLDN18.2 status did not identify a subgroup with differential benefit from first line chemoimmunotherapy. These findings support CLDN18.2 primarily as a therapeutic target rather than a predictive biomarker for immune checkpoint inhibitor-based treatment.

Abstract: Background: Reliable biomarkers to predict response to immune checkpoint inhibitors (ICIs) in non–small cell lung cancer (NSCLC) remains limited.ABO blood group antigens may influence tumor immunity and response to immunotherapy. Methods: We retrospectively analyzed patients with advanced NSCLC treated with ICIs between 2019 and 2024. Patients were classified according to ABO blood group. Overall survival (OS), progression-free survival (PFS), and treatment response were evaluated. Results: Patients with blood group O demonstrated significantly longer OS (18.6 vs 11.0 months, p< 0.001) and PFS (9.4 vs 4.2 months, p=0.001). Conclusion: ABO blood group, particularly blood group O, may serve as a clinically relevant host-related prognostic biomarker in NSCLC immunotherapy.

Abstract: Background/Objectives: Improved preoperative prediction of nodal metastasis in prostate cancer could refine selection for extended pelvic lymph node dissection, a high-morbidity procedure. Sperm-associated antigen 1 (SPAG1) is a candidate marker of nodal status, but its incremental value beyond clinical staging and the associated transcriptional state remain unevaluated. Methods: In The Cancer Genome Atlas prostate adenocarcinoma (TCGA-PRAD) cohort (497 patients with matched clinical and RNA-sequencing data), we evaluated the association between SPAG1 expression and pathological N stage by logistic regression with 2000-resample bootstrap optimism correction and sensitivity analyses for missing nodal data and batch effects. Hallmark enrichment analysis compared SPAG1 expression extremes (quartile 4 vs 1) and, separately, N1 versus N0 tumours adjusted for T stage, Gleason grade, and tissue source site; directional concordance was assessed. Results: N1 rates rose across SPAG1 quartiles from 7.6% to 39.0% (per-quartile odds ratio [OR], 1.83; p = 2.55×10⁻⁵). After adjusting for T stage and Gleason grade, SPAG1 remained an independent predictor (adjusted OR, 2.14; 95% confidence interval [CI], 1.50–3.13; p = 4.8×10⁻⁵), stable across both sensitivity analyses. Adding SPAG1 improved discrimination (area under the receiver-operating characteristic curve, 0.783 to 0.838; ΔAUC, 0.056; paired DeLong p = 3.03×10⁻⁵). The SPAG1 transcriptional programme showed cell-cycle, immune-inflammatory, and mTORC1/TGF-β signalling activation with suppressed differentiation and metabolism; all 15 overlapping Hallmark pathways were directionally concordant with the adjusted N1 signature. Conclusions: SPAG1 expression in primary prostate tumours is a candidate predictor of pathological lymph node metastasis with statistically robust incremental discrimination beyond clinical staging. Independent external validation and biopsy-based feasibility studies are required before clinical application.

Abstract: Background/Objectives: Cancer vaccines represent the next frontier in immunotherapy, aiming to elicit long-lasting protective anti-tumor protective immune responses. Human epidermal growth factor receptor 2 (HER2) is a well-established therapeutic target in breast cancer. Active immunization with HER2-displaying M13 bacteriophages can induce a therapeutic immune response against HER2-positive breast cancer, offering a promising alternative to trastuzumab. However, the duration of anticancer immune protection triggered by anti-HER2 phage-based vaccines is limited by tumor-immune suppressive mechanisms. Methods: In this study, two vaccination cycles with ECTM phages displaying the extracellular (EC) and transmembrane (TM) domains of human HER2 were combined with palbociclib, a CDK4/6 inhibitor, to enhance antitumor im-munity in the clinically relevant Δ16HER2 transgenic preclinical model of breast cancer. Results: The proposed combination treatment resulted in a better and long-lasting control of tumor growth rate and multiplicity than either palbociclib or phage vaccination alone, correlating with a significantly stronger anti-HER2 humoral response (IgG2a isotype). Analysis of the tumor immune infiltrate revealed an increased presence of CD8⁺ T cells concomitant with a reduction in FoxP3⁺ regulatory T cells (Tregs) in tumors explanted from mice receiving the combination therapy. Conclusions: These preclinical results provide a rationale for the clinical translation of CDK4/6 inhibitors combined with anti-HER2 active immunotherapies in breast cancer, as they may yield sustained antitumor responses by reverting the immunosuppressive tumor environment.

Abstract: Human T-cell leukemia virus-1 (HTLV-1) is the etiological agent of a series of chronic inflammatory diseases such as HTLV-associated myelopathy/Tropical spastic paraparesis (HAM/TSP), uveitis, dermatitis and pneumonitis, and, importantly, of a T-cell lymphoproliferative neoplasm designed Adult T-cell leukemia/lymphoma (ATL). Two viral proteins, Tax-1 and HBZ are crucially involved in HTLV-1 infectivity and in ATL by altering key pathways of cell homeostasis. A fundamental distinction between the expression of the two oncoproteins exists, witnessed by the fact that Tax-1 is expressed in early phases of HTLV-1 infectivity and ATL onset but may be lost in a substantial number of established ATL, whereas HBZ is always expressed in all phases of HTLV-1 infection and in all ATL. Additionally, while Tax-1 can be localized both in the cytoplasm and nucleus in all cases of disease, recent evidence indicate that HBZ is localized solely in the cytoplasm in cells of HTLV-1 infected individuals, asymptomatic carriers (AC) and patients suffering from HAM/TSP. Importantly, ATL instead marks a progressive dislocation of HBZ in the nucleus. Thus both the expression and the subcellular localization of HBZ represent distinctive elements in the process of HTLV-1-associated pathology. Within this frame, recent studies point to a very important involvement of HBZ in disarranging the homeostasis of the cell not only at transcriptional but most importantly at post-transcriptional level as a result of the interaction with crucial factors regulating RNA splicing and stability. These recent aspects of the HBZ biology will be discussed for their implication in HTLV-1 mediated oncogenesis.

Abstract: Background/Objectives: Chronic ulcers are common among older and multimorbid hospitalized patients and may reflect systemic vulnerability beyond the local wound condition. Malignancy recorded among secondary diagnoses may identify patients with reduced physiological reserve and increased inpatient risk, but its prognostic significance in hospitalized chronic ulcer populations remains insufficiently characterized. This study aimed to evaluate whether malignancy coded among secondary diagnoses was associated with in-hospital mortality among adults hospitalized with chronic ulcers. Methods: This nationwide retrospective cohort study used anonymized Romanian public-hospital discharge data for adults aged ≥18 years hospitalized with chronic ulcers between 1 January 2017 and 31 December 2022. The index-episode cohort included 69,349 patients generating 116,264 hospitalizations. Exposure was defined as at least one ICD-10 C00–C97 malignant neoplasm code recorded among secondary diagnoses in the relevant analytical hospitalization. The primary outcome was in-hospital mortality. Crude and adjusted odds ratios were estimated using logistic regression models. Results: Overall, 1,837 patients had C00–C97 codes recorded among secondary diagnoses, with 73 deaths. In-hospital mortality was 3.97% among exposed patients and 1.78% among unexposed patients, corresponding to a crude odds ratio of 2.28 (95% CI 1.79–2.90). After adjustment for age group, sex, admission type, chronic ulcer category, and hospitalization pattern, malignancy recorded among secondary diagnoses remained associated with mortality (adjusted OR 1.87, 95% CI 1.42–2.45; p < 0.001). Additional adjustment for the number of non-malignant secondary diagnoses yielded similar results (adjusted OR 1.88, 95% CI 1.42–2.47; p < 0.001). Conclusions: Malignancy coded among secondary diagnoses may serve as a pragmatic administrative marker of increased in-hospital mortality risk among patients hospitalized with chronic ulcers. However, residual confounding and the absence of cancer-stage information limit causal interpretation.

Abstract: Modern cancer therapy has achieved major advances through molecularly targeted therapies, immune checkpoint blockade, and antibody–drug conjugates, guided by the principles of precision oncology. Despite increasingly precise molecular interventions guided by actionable molecular targets and predictive biomarkers, therapeutic outcomes remain highly variable, making it a major bottleneck. Such variability suggests that therapeutic outcomes emerge not directly from molecular targeting, but through therapy-driven evolution of the tumor–host system. This intrinsically dynamical nature of cancer therapy motivates the introduction of concepts from dynamic systems theory.Here, a conceptual state-space formulation of therapeutic trajectory evolution, variability, and convergence is proposed. Within this framework, the tumor–host system is represented as an accessible state-space organization (θ) that constrains therapeutic trajectory evolution (T) and the attractor (A) from which therapeutic outcome (y) emerge. Therapeutic input (μ) drives system trajectory evolution, whereas state-space modulating operator (υ) modifies the accessible state-space organization θ, generating a modified accessible state-space organization θ_υ.The conceptual formulations y ~ A[T(θ, μ)] and y’ ~ A’[T’(θ_υ, μ)] represent therapeutic trajectory evolution within original and modified accessible state-space organization, respectively. Under modified accessible state-space organization, trajectory evolution may exhibit reduced diversity and increased convergence toward restricted attractors, thereby promoting therapeutic convergence.Bicarbonate-mediated tumor alkalization and intracellular lactic acidosis are discussed as representative examples of external and internal state-space modulating operators. In hepatocellular carcinoma, bicarbonate-enhanced transarterial chemoembolization and bicarbonate-augmented anti–PD-1 therapy demonstrated unusually high and convergent therapeutic responses, reflecting biased trajectory evolution within a modified accessible state-space organization. In contrast, intracellular lactic acidosis is represented as an endogenous state-space modulating operator arising from tumor metabolic adaptation.Taken together, the present framework provides a conceptual explanation for therapeutic variability as a major limitation of molecular-targeting-based cancer therapy and suggests that combining molecular targeting with state-space modulation may represent a promising direction for overcoming this limitation and improving therapeutic efficacy.

Abstract: Manual Pap smear analysis for cervical cancer screening is limited by inter-observer variability, time constraints, and restricted expert availability. Although convolutional neural networks (CNNs) have automated cervical cell classification, they remain limited in modeling long-range spatial dependencies and often lack clinical interpretability. In this study, Vision Transformer (ViT) architectures were systematically optimized to enhance automated cervical cancer screening, which resulted in improved interpretability. The Herlev dataset (917 images: 242 normal, 675 abnormal) was utilized to optimize ViT-Tiny, a lightweight Vision Transformer architecture designed for reduced computational complexity, through a comprehensive evaluation of augmentation strategies, class weighting, and hyperparameters. The optimal configuration achieved a cross-validation accuracy of approximately 95% (95.15% for the best replicated configuration), in which random horizontal flipping and class weighting (0.7 × 1.3) were identified as most effective. Gradient weighted Class Activation Mapping (Grad CAM) analysis confirmed that model attention corresponded to clinically relevant morphological features, which include nuclear regions, cell boundaries, and chromatin texture, which align with cytopathological criteria. These findings indicate that Vision Transformers can deliver accurate and interpretable decision support for cervical cancer screening, and that they combine competitive classification performance with the attention-based transparency relevant to medical AI. Further validation on larger, multi-center datasets remains necessary before clinical deployment.

Abstract: Background/Objectives: The homeostatic imbalance between tumor metastasis-promoting genes and -suppressing genes determines the metastatic potential of lung adenocarcinoma (LUAD) cells. However, the post-transcriptional regulation mechanism mediated by RNA-binding proteins (RBPs) in maintaining the expression balance of tumor metastasis-related genes remains unclear. Methods: The expression of MCPIP3 and IGF2BP3 in human lung cancer tissues was analyzed using bioinformatics methods and validated qRT–PCR, immunohistochemistry, and immunoblotting. In vitro cellular experiments and in vivo animal examinations were used to evaluate the effects of MCPIP3 and IGF2BP3 on metastasis. RNA-sequencing, PCRarray, RNA pull-down and mass spectrometry, RNA-EMSA, RNA immunoprecipitation (RIP), mRNA stability, luciferase assay, were performed to elucidate the mechanism. Results: MCPIP3 is expressed at low levels, while IGF2BP3 is highly expressed in LUAD tissues. Lower MCPIP3 expression and higher IGF2BP3 expression in lung cancer tissues were significantly associated with poor prognosis in LUAD patients. MCPIP3 significantly inhibited the metastasis of lung cancer cells both in vitro and in vivo, whereas IGF2BP3 promoted metastasis. Mechanistically, MCPIP3 specifically bound to the stem‒loop structure in the 3'UTR (untranslated region) of METAP2 transcript, and degraded its mRNA via its RNase domain. In contrast, by interacting with the common stem‒loop structure, IGF2BP3 could stabilize METAP2 mRNA and promote its expression. Notably, inverse correlations were observed between the expression of MCPIP3 and IGF2BP3 and the expression of METAP2 in LUAD tissues. Conclusions: MCPIP3 and IGF2BP3 antagonistically control lung cancer cell metastasis by balancing the expression of metastasis-promoting genes through mediating mRNA stability, thereby providing potential targets for lung cancer treatment.

Abstract: Background: The clinical outcomes of lung cancer brain metastases (LCBMs) are highly variable. Traditional pathology relies on bulk cell densities. These static measures fail to capture the spatial architecture of the tumor immune microenvironment (TIME). B7-H3 (CD276) represents a key immune checkpoint in LCBMs. We investigated whether the spatial orchestration of B7-H3-expressing cell populations predicts patient prognosis. Methods: We performed multiplex immunohistochemistry (mIHC) for B7-H3 and Iba1 (a macrophage marker) in surgically resected tissues from 22 patients. We used QuPath for single-cell segmentation and classification. We performed spatial point pattern and spatial autocorrelation analyses to evaluate the relative positioning of single cells. We computed spatial interaction metrics, which included cross-Moran’s I and the cross-K function, within a 35 μm radius. We correlated these metrics with postoperative overall survival (OS) and determined prognostic thresholds via time-dependent ROC curve analysis. Results: Standard cell densities generally did not correlate with OS, although B7-H3+ tumor-associated macrophage (TAM) density showed a positive correlation. Conversely, specific spatial metrics served as significant prognostic factors. High spatial mixing and clustering of B7-H3+ and B7-H3- TAMs, as shown by high cross-Moran’s I (p = 0.011) and the cross-K functions (p = 0.033), correlated with significantly shorter OS. This pattern suggests a coordinated local immunosuppressive network. Conversely, high spatial integration between B7-H3+ and B7-H3- tumor cells correlated with prolonged OS (p = 0.016), whereas spatial segregation of B7-H3+ tumor cells predicted poor outcomes. Conclusions: Decoding the spatial architecture of B7-H3-expressing cell subpopulations provides superior prognostic stratification compared with standard density-based metrics. These localized spatial niches represent potential biomarkers and therapeutic targets for personalized LCBM management.

Abstract: Theranostic nanoparticles, which integrate diagnostic imaging and therapeutic delivery within a single nanoplatform, represent a transformative paradigm in oncological nanomedicine. Despite substantial preclinical progress, the field faces persistent gaps in rational nanoparticle design informed by tumor biology, preclinical model fidelity, and clinical translation. This review critically synthesizes theranostic nanoparticle research across three underexplored domains. First, we examine tumor microenvironment features - reactive oxygen species dynamics, glutathione gradients, hypoxia, and proteasomal dysregulation - as mechanistic drivers of nanoparticle responsiveness. Second, we evaluate redox-responsive and proteasome-targeted nanoplatforms that exploit these cues for stimuli-triggered drug release and simultaneous imaging readout. Third, we address the unmet need for three-dimensional organoid and microfluidic tumor models as predictive preclinical testing environments, given the well-documented limitations of conventional two-dimensional cultures. Cancer subtype-specific applications are discussed for breast cancer, HPV-associated malignancies, colorectal cancer, and prostate cancer. Clinical translation barriers - including pharmacokinetic constraints, protein corona formation, immune clearance, anti-PEG antibodies, complement activation-related pseudoallergy, and FDA/EMA regulatory pathways - are addressed from a clinical oncology perspective. The review concludes with a research roadmap integrating proteomics-guided nanoparticle engineering, patient-derived organoid biobanks, and artificial intelligence-assisted design as priority areas for next-generation oncological theranostics.