Andrei Alexandru Cosma

Background/Aim. We previously described four distinct breast cancer (BC)-associated adipose tissue (BCAAT) subtypes that have a significant impact on survival. In this exploratory study, we aimed to determine whether these BCAAT subtypes exhibit significant correlations with neoadjuvant and adjuvant therapies and BC molecular subtypes. Methods. Four BCAAT subtypes previously identified as fibroblast-rich (F^Rich_BCAAT), myofibroblast-rich (MyoF^Rich_BCAAT), vascular-rich (V^Rich_BCAAT), and mixed vascular- and inflammation-rich (VI^Rich_BCAAT) were analyzed according to their distribution in BC molecular subtypes, as well as in relation to neoadjuvant therapy and survival. Results. Triple-negative BC and Luminal B (LB) BC are related to V^Rich and VI^Rich BCAAT subtypes and were affected by Epirubicin/Cyclophosphamide (EC)+ paclitaxel (PTX) therapy, which significantly enhanced overall survival (OS) and disease-free survival (DFS). For the Luminal A (LA) BC subtype, EC + docetaxel (DTX) had significant impact on survival independent of BCAAT subtype. Conclusions. BCAAT subtypes strongly influenced neoadjuvant therapy response and survival depending on BC molecular subtype.

Background/Objectives: The fundamental classification based on white, brown, pink, and beige adipose tissue morphology together with fat vacuole content released into the tumor microenvironment incompletely defines breast cancer-associated adipose tissue (BCAAT) heterogeneity and does not sufficiently explain its controversial impact on invasion, recurrence, or survival in breast cancer (BC). We aim to expand BCAAT characterization by systematically evaluating stromal cellular elements within peritumoral adipose tissue, including CD34-positive fibroblasts, smooth muscle actin (SMA)-positive myofibroblasts, inflammatory cells, and microvascular structures to define distinct BCAAT subgroups. Methods: CD34 and smooth muscle actin (SMA) double immunohistochemistry was performed on 109 BC tissue specimens from patients aged 35 to 79 years old, followed by microscopic evaluation of cellular and vascular components inside peritumor adipose tissue. Microscopic findings were then correlated to age, body mass index (BMI), lymphovascular (LVI) and perineural invasion (PnI), recurrence (R), and tertiary lymphoid structures (TLSs). Results: Four BCAAT subtypes have been identified as fibroblast-rich (F^Rich_BCAAT), myofibroblast-rich (MyoF^Rich_BCAAT), vascular-rich (V^Rich_BCAAT), and mixed-vascular and inflammatory-rich (VI^Rich_BCAAT). The F^Rich_BCAAT subtype predominates for the age subgroup 35 to 49 years old and is a significantly worse prognostic factor for survival (p = 0.022). For the age subgroup of 50 to 69 years old, the VI^Rich_BCAAT subtype significantly influences PnI (p = 0.05) but not survival (Log-rank test, z = 0.57, p = 0.57). V^Rich_BCAAT was significantly impactful for BC patient survival aged 70 to 75 years old (p = 0.043). BMI did not correlate with any of the BCAAT subtypes but was strongly correlated with prognostic markers for each BCAAT subtype. Conclusions: Based on immunohistochemically detected cellular and vascular components, four microscopic BCAAT subtypes were identified. Three of four BCCAT subtypes specifically affect BC patient prognosis and survival depending on age.

Background/Objectives: Immunotherapy has improved outcomes for selected patients with advanced non-small-cell lung cancer (NSCLC), yet the predictive value of individual biomarkers such as PD-L1 remains limited. Systemic inflammatory indices derived from routine blood tests may complement molecular and immunohistochemical features, offering a broader view of host–tumor immunobiology. Methods: We conducted a retrospective study of 298 patients with stage IIIB–IV NSCLC treated with immune checkpoint inhibitors (ICIs) at a tertiary oncology center between 2022 and 2024. Baseline neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), and systemic immune–inflammation index (SII) were collected alongside PD-L1 expression and molecular alterations (EGFR, KRAS, ALK, TP53). Patients were stratified into inflammatory–molecular clusters integrating these parameters. Associations with objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) were evaluated using Kaplan–Meier and multivariate Cox analyses. Results: Four distinct inflammatory–molecular clusters demonstrated significantly different outcomes (p < 0.001). Patients with low NLR and high PD-L1 expression (Cluster A) showed the highest ORR (41%), longest median PFS (13.0 months), and OS (22.5 months). The EGFR/ALK-driven, inflammation-dominant cluster (Cluster C) exhibited poor response (ORR 7%) and shortest survival (PFS 4.3 months). High NLR (HR 2.12), PD-L1 < 1% (HR 1.91), and EGFR mutation (HR 2.36) independently predicted shorter PFS. A combined model incorporating NLR, PD-L1, and molecular status outperformed individual biomarkers (AUC 0.82). Conclusions: Integrating systemic inflammatory indices with PD-L1 expression and molecular alterations identifies clinically meaningful NSCLC subgroups with distinct immunotherapy outcomes. This multidimensional approach improves prediction of ICI response and may enhance real-world patient stratification, particularly in settings with limited access to extended molecular profiling.

Background: Non-small-cell lung cancer (NSCLC) is often diagnosed at stage IV, when prognosis depends on metastatic spread. The impact of histopathology on the first metastatic site remains underexplored. Methods: We retrospectively analyzed 364 patients with stage IV NSCLC diagnosed at OncoHelp Medical Center, Timișoara, Romania (2020–2024). All underwent baseline CT chest–abdomen–pelvis, whole-body FDG PET-CT, and brain MRI within seven days of histological confirmation. Patients were stratified into adenocarcinoma (n = 164), squamous cell carcinoma (n = 112), and large-cell carcinoma (n = 88). The first metastatic site was defined as the earliest confirmed location. Associations were evaluated using Fisher’s exact test and multinomial logistic regression. Results: Histology was associated with the first metastatic site (global p = 0.013). Adenocarcinoma was more likely than squamous carcinoma to present with brain metastases (RRR 3.74, 95% CI 1.48–9.45; p = 0.005; p_FDR = 0.053) and showed directional signals toward bone and adrenal involvement. Squamous carcinoma more frequently spread to the pleura as the first site (adjusted p = 0.008). Large-cell carcinoma showed no consistent differences compared to squamous carcinoma. In adenocarcinoma subgroups, EGFR-mutant tumors most often metastasized to the brain (55.6%), KRAS-mutant tumors to the liver (44.4%), and ALK-rearranged tumors to bone (100%). Conclusions: The first metastatic site in NSCLC follows histology-specific patterns, with adenocarcinoma favoring hematogenous spread and squamous carcinoma showing locoregional involvement. Molecular status further refines these patterns in adenocarcinoma. Incorporating histology into baseline staging may improve diagnostic efficiency and prognostic accuracy.

Background and Objectives: Systemic inflammatory markers from an ordinary complete blood count (CBC) may foreshadow where non-small-cell lung cancer (NSCLC) will first spread, but organ-specific signatures remain poorly defined. Materials and Methods: We retrospectively reviewed 302 adults (mean age 60.7 ± 13.4 years; 80.8% men) with stage IV NSCLC managed at OncoHelp Medical Center, Timișoara, between January 2022 and December 2024. Eligibility demanded a single radiologically confirmed distant site at diagnosis and pre-treatment CBC. Neutrophil-to-lymphocyte (NLR), platelet-to-lymphocyte (PLR), and lymphocyte-to-monocyte (LMR) ratios were compared across pleural (n = 52), bone (n = 86), liver (n = 66), and brain (n = 98) metastases using Kruskal–Wallis tests with Bonferroni adjustment; z-standardized logistic models identified independent predictors. Results: Metastases clustered most often in brain (32.5%), followed by bone (28.5%), liver (21.9%), and pleura (17.2%). Median PLR rose selectively in pleural disease (274 vs. 217–253 in other sites; p = 0.006). LMR fell to 2.0 in bone but climbed to 2.8 in brain lesions (p = 0.032 and 0.008, respectively). NLR was globally elevated (6.7–7.6), yet differed significantly only for bone and liver deposits. Logistic modeling showed that each standard-deviation rise in absolute neutrophil count quadrupled the odds of hepatic involvement (Odd Ratio (OR) 4.26; 99% Confidence inerval (CI) 2.20–6.25), monocytosis nearly doubled bone risk (OR 1.83; 1.01–3.33), while higher erythrocytes, eosinophils, and lymphocytes independently protected against pleural seeding (all p < 0.01). Age-stratified analysis revealed that osseous and cerebral metastases predominated in patients ≤ 50 years, whereas inflammatory indices were age-invariant. Conclusions: Routine CBC ratios encode distinct “inflammatory fingerprints” that mirror the first metastatic destination in NSCLC: platelets herald pleural spread, neutrophils favor liver and bone, and divergent lymphocyte–monocyte balances separate bone from brain. Although no substitute for cross-sectional imaging, these low-cost markers could refine clinical suspicion, guide targeted work-up, and illuminate the biology of organ-selective dissemination, particularly in resource-limited settings.

The expression and function of podoplanin (PDPN) in the normal human placenta has been debated in placental evaluation. This study emphasizes the importance of a multimodal approach of PDPN expression in normal human placentas. A complete examination is performed using immunohistochemistry, RNAscope and automated Digital Image examination (DIA) interpretation. QuPath DIA-based analysis automatically generated the stromal and histological scores of PDPN expression for immunohistochemistry and RNAscope stains. The umbilical cord’s isolated fibroblasts and luminal structures expressed PDPN protein and PDPN_mRNA. RNAscope detected PDPN_mRNA upregulation in syncytial placental knots trophoblastic cells, but immunohistochemistry did not certify this at the protein level. The study found a significant correlation between the IHC and RNAscope H-Score (p = 0.033) and Allred Score (p = 0.05). A successful multimodal strategy for PDPN assessment in human placentas confirmed PDPN expression heterogeneity in the full-term human normal placenta and umbilical cord at the protein and mRNA level. In placental syncytial knots trophoblastic cells, PDPN showed mRNA overexpression, suggesting a potential role in placenta maturation.

Chloride intracellular channel 1 (CLIC1) is involved in cell migration and metastasis. The histological growth patterns of liver metastasis are as follows: desmoplastic (d-HGP), replacement (r-HGP), pushing (p-HGP), and mixed. The aim of this study was to evaluate the relation between HGP, angiogenesis, and CLIC1 expression. Materials and Methods: A total of 40 cases of primary tumors and their LM: d-HGP (12 cases), r-HGP (13 cases), and p-HGP (15 cases), were evaluated through simple and double immunostaining. CLIC1 assessment was conducted as follows: scores of 0 (less than 10% of positive cells), 1 (10–30%), 2 (30–50%), or 3 (more than 50%) were assigned. Heterogeneous CLIC1 expression was found. CLIC1 in primary tumors correlated with grade G for all cases of LM with a p-HGP (p = 0.004). The CLIC1 score for LMs with an r-HGP correlated with grade G of the corresponding primary tumor (p = 0.027). CLIC1 and CD34+/Ki67+ vessels (p = 0.006) correlated in primary tumors. CLIC1 in primary tumors correlated with CD34+/Ki67+ vessels of LMs with a d HGP (p = 0.024). Conclusions: The CLIC1 score may have prognostic value, mainly for LMs with a p-HGP and r-HGP, and therapeutic value for LMs with a d-HGP.

Background: Overexpression of chloride intracellular channel protein 1 (CLIC1) in tumor cells has been confirmed, but it has received less attention in the tumor blood vessel endothelium. Aim: The assessment of CLIC1 expression in ccRCC tumor blood vessels and its relationship with TNM parameters and tumor cell CLIC1 expression. Methods: CLIC1 immunostaining in ccRCC was evaluated in 50 cases in both malignant cells and tumor blood vessels (CLIC1 microvessel density-CLIC1-MVD) and was correlated with TNM staging parameters. Results: CLIC1-MVD was observed in approximately 65% of cases, and CLIC1 co-localization in both tumor and endothelial cells was observed in 59% of cases. ccRCC was classified into four groups (Classes 0–3) based on the percentage of positive tumor cells, with each group including sub-groups defined by CLIC1 expression in the endothelium. Class 3 (60–100% positive tumor cells) had the highest CLIC1-MVD, with an impact on T and M parameters (p value = 0.007 for T, and p value = 0.006 for M). For cases with CLIC1 intracellular translocation, there was a strong correlation between CLIC1-MVD and M (p value < 0.001). Conclusions: Co-expression of ccRCC tumor and endothelial cells promotes tumor progression and metastasis and should be investigated further as a potential therapeutic target for ccRCC and other human malignancies.

We aimed to analyze the involvement of adipose-sourced mast cells in nerve repair. Sixteen Wistar rats underwent complete transection of the sciatic nerves followed by either direct neurorrhaphy or neurorrhaphy and processed abdominal fat. Four animals were used as controls. Specimens were obtained at 4 and 10 weeks and analyzed using luxol fast blue stain, mast cell tryptase and CD34 (for angiogenesis) per microscopic field ×200. When assessed by luxol fast blue, normal nerves showed an average of 2–3 mast cells/field. At 4 weeks, there were 9.25 for the simple nerve sutures and 16 for the augmented repairs. At 10 weeks, there were 23 and 27.6. When assessed by mast cell tryptase, there were no positives in the controls. At 4 weeks, we found an average of 4 in the simple sutures and 2.5 in the augmented repairs. At 10 weeks, there were 4.5 and 0.2. In controls, there were 1–2 CD34+ blood vessels per field. At 4 weeks, simple repairs showed an average of 4 and, in those with adipose addition, 5.5. At 10 weeks, there were 7 and 12. Mechanically processed adipose tissue augmented nerve repair does not seem to increase mast cell expression but may support angiogenesis in an experimental model.