Category: Statistics

Front Oncol. 2026 Jun 26;16:1780642. doi: 10.3389/fonc.2026.1780642. eCollection 2026.

ABSTRACT

BACKGROUND: Bone cancer is a rare malignancy worldwide, with incidence patterns that vary by age, sex, and geographic region. In Saudi Arabia, however, comprehensive population-based evidence describing the national epidemiology of bone cancer remains limited. This study aimed to describe bone cancer incidence in Saudi Arabia according to age group, sex, calendar year, and administrative region, with particular emphasis on age-standardized incidence rates (ASIRs).

METHODS: A retrospective population-based descriptive study was conducted using data from the Saudi Cancer Registry. All primary bone cancer cases diagnosed between 2004 and 2020 were included. Incidence patterns were summarized using frequencies, age-specific incidence rates, crude incidence rates (CIRs), and ASIRs, stratified by sex, age group, year of diagnosis, and region. Statistical analyses were performed using SPSS software version 30.

RESULTS: Between 2004 and 2020, a total of 2,275 primary bone cancer cases were recorded in Saudi Arabia, including 1,318 males (57.9%) and 957 females (42.1%). Bone cancer accounted for approximately 2.0% of all cancers among males and 0.9% among females. Mean ASIRs were higher in males (≈1.0 per 100,000) than females (≈0.7 per 100,000), while CIRs remained below 2.0 per 100,000 throughout the study period. Age-specific incidence showed a clear adolescent peak, most prominent in the 15-19-year age group, followed by the 10-14-year group. Regional variation in ASIRs was observed, with higher rates in Al-Jouf and Najran and persistently lower rates in Jazan.

CONCLUSION: Bone cancer in Saudi Arabia is rare but demonstrates distinct variation by sex, age, and region. The observed male predominance and adolescent peak are consistent with international epidemiological patterns. Continued enhancement of population-based cancer surveillance is essential to support accurate epidemiological assessment and informed public health planning.

PMID:42434761 | PMC:PMC13349887 | DOI:10.3389/fonc.2026.1780642

Front Oncol. 2026 Jun 26;16:1764691. doi: 10.3389/fonc.2026.1764691. eCollection 2026.

ABSTRACT

PURPOSE: To assess the cross-software reproducibility of Computed Tomography (CT) radiomic features extracted using three widely adopted platforms (Siemens syngo.via Frontier, 3D Slicer/PyRadiomics, and mint Lesion) and to identify a subset of highly robust features suitable for multi-platform and multi-center radiomics applications.

METHODS: A retrospective cohort of 97 lesions (primary colorectal cancer, colorectal liver metastases, and hepatocellular carcinoma) who underwent contrast-enhanced Computed Tomography (CT) in the portal venous phase was analyzed. Semi-automatic 3D lesion segmentations were exported for radiomic extraction across the three platforms. Shared radiomic features among tools were harmonized and z-score normalized. Cross-platform similarity was assessed using distribution distance metrics, hierarchical clustering, and the Adjusted Rand Index (ARI). A novel Composite Robustness Index (CI) integrating Pearson correlation, Kolmogorov-Smirnov statistics, and mean fold-difference was developed to quantify feature-level reproducibility.

RESULTS: First-order intensity features and key GLCM descriptors (e.g., Correlation, Joint Average, Sum Entropy) demonstrated the highest cross-software stability, with nearly superimposable distributions and strong concordance in clustering structure. Siemens syngo.via Frontier and 3D Slicer/PyRadiomics showed the highest agreement (mean ARI >0.85), while mint Lesion™-which lacks higher-order texture families-showed moderate deviations (mean ARI ≈ 0.70-0.75). High-order features, particularly GLDM and GLRLM metrics, exhibited substantial variability across platforms. The CI ranking enabled identification of a reproducible set of “highly reproducible features, “ including glcm_Correlation, firstorder_Mean, firstorder_RMS, firstorder_90Percentile, and shape axis-length descriptors.

CONCLUSION: Despite intrinsic software differences, a consistent subset of radiomic features remains reproducible across heterogeneous extraction tools. The combined use of distribution analysis, hierarchical clustering, and the Composite Robustness Index offers a rigorous framework for evaluating cross-platform reliability. These findings support the feasibility of multi-tool radiomics and provide a validated feature set for harmonized quantitative imaging pipelines.

PMID:42434756 | PMC:PMC13349753 | DOI:10.3389/fonc.2026.1764691

Front Oncol. 2026 Jun 26;16:1852397. doi: 10.3389/fonc.2026.1852397. eCollection 2026.

ABSTRACT

OBJECTIVE: Walking is a simple and accessible form of aerobic exercise that may improve cardiorespiratory fitness (CRF) in cancer patients; however, its independent effects remain unclear. This systematic review and meta-analysis aimed to evaluate the effects of walking training on CRF, fatigue, dyspnea, safety, and adherence in adult cancer patients.

METHODS: A systematic search was conducted in 4 databases-Web of Science, Embase, PubMed, and Cochrane Library Database-covering the period from the inception of each database through November 28, 2025. The inclusion criteria were randomized controlled trials (RCTs) evaluating walking interventions for cancer patients. Priority was given to meta-analyses of outcome measures using a random-effects model. If the statistical heterogeneity is less than 40%, a fixed-effects model is used.

RESULTS: 11 randomized controlled trials involving 649 participants were included. Compared with standard care or no intervention, walking training was associated with statistically significant improvements in the two outcomes: peak oxygen uptake (VO₂ peak) (SMD = 0.56; 95% CI: [0.06, 1.06]; I² = 75%; τ²=0.28; P = 0.03) and fatigue (SMD = -0.45; 95% CI: [-0.71, -0.18]; I² = 32%; τ²=0; P = 0.001). In contrast, no statistically significant effects were observed for maximal oxygen uptake (VO₂ max) (SMD = 0.20;95% CI: [-0.15, 0.54]; I² = 0%; P = 0.26), 6-Minute Walk Distance (6MWD) (MD = 53.97;95% CI: [-23.00, 130.93]; I² = 80%; τ²=2538;P = 0.17), and dyspnea (SMD = -0.18, 95% CI: [-0.47, 0.11]; I² = 0%; P = 0.23). No serious intervention-related adverse events were reported, and adherence rates ranged from 67% to 94%.

CONCLUSION: Current evidence suggests that walking training may improve VO₂ peak and reduce fatigue in cancer patients, while demonstrating acceptable short-term safety and adherence. However, these findings should be interpreted cautiously because the certainty of evidence remains low. Larger, high-quality randomized controlled trials with standardized intervention protocols and longer follow-up periods are needed to confirm these findings.This study conducted a systematic literature search in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Prior to the search, the study protocol was prospectively registered on the international systematic review registry platform (PROSPERO, registration number: CRD420251140743) to ensure the scientific rigor and methodological soundness of this study.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/prospero/, identifier CRD420251140743.

PMID:42434753 | PMC:PMC13349785 | DOI:10.3389/fonc.2026.1852397

Front Oncol. 2026 Jun 26;16:1866001. doi: 10.3389/fonc.2026.1866001. eCollection 2026.

ABSTRACT

INTRODUCTION: Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy with a rising global incidence. Differentiating PDAC from non-malignant pancreatic conditions, particularly chronic pancreatitis (CP), remains challenging due to overlapping clinical and radiological features, highlighting the need for new biomarkers. The best-validated serum biomarker, carbohydrate antigen 19-9 (CA19-9), has limited clinical utility due to its suboptimal sensitivity and specificity. This study aimed to evaluate the diagnostic performance of serum macrophage inhibitory cytokine 1 (GDF15), syncollin (SYCN), and thrombospondin-2 (TSP-2), both alone and in multi-marker panels, for differentiating PDAC from CP and healthy controls (HCs). The selection of these markers was based on prior evidence linking GDF15 to PDAC diagnosis and prognosis, SYCN to pancreatic tissue damage, and TSP-2 to tumor microenvironment remodeling.

METHODS: This study included 188 individuals: 78 diagnosed with PDAC, 79 with CP and 31 HCs. PDAC and CP were diagnosed based on clinical, imaging, histopathological, and laboratory findings, and classified according to the 8th TNM classification and the updated Cambridge system, respectively. Serum GDF15, SYCN, and TSP-2 levels were quantified using ELISA. Statistical analyses included group comparisons, correlation testing, and receiver operating characteristic (ROC) curve analysis with assessment of the area under the curve (AUC).

RESULTS: GDF15 and SYCN were significantly elevated in PDAC compared with both CP and HCs, whereas TSP-2 did not differ significantly between those groups. For PDAC vs HCs, GDF15 provided the strongest overall discrimination (AUC = 0.86; sensitivity 97%, specificity 71%), while SYCN showed a lower AUC (0.77) but very high specificity (94%). The combined GDF15 + SYCN panel increased AUC to 0.89. For PDAC vs CP, GDF15 achieved moderate diagnostic performance (AUC = 0.73), SYCN performed less well (AUC = 0.65), and TSP-2 performed near chance (AUC = 0.52). Incorporating age and bilirubin in the model improved discrimination between PDAC and CP, yielding a maximum AUC of 0.88. Additionally, positive correlations were observed between SYCN and diabetes, as well as between GDF15 and hyperbilirubinemia, in patients with PDAC.

CONCLUSIONS: These results suggest that GDF15 and SYCN are promising serum biomarkers for PDAC, whereas TSP-2 appears to have limited diagnostic utility.

PMID:42434748 | PMC:PMC13349920 | DOI:10.3389/fonc.2026.1866001

Front Oncol. 2026 Jun 26;16:1842330. doi: 10.3389/fonc.2026.1842330. eCollection 2026.

ABSTRACT

BACKGROUND: Primary small-bowel tumors are uncommon and frequently present with non-specific symptoms. Timely and accurate diagnosis relies on a range of radiologic and endoscopy-based modalities, yet their comparative diagnostic performance remains uncertain.

METHODS: We performed a systematic review and diagnostic test accuracy meta-analysis. PubMed, Embase, CNKI, and Wanfang databases were searched from inception to August 31, 2025, and the reference lists of included studies and relevant reviews were hand-searched. Two reviewers independently screened studies, extracted data, and assessed the risk of bias using the QUADAS-2 tool. To avoid unit-of-analysis errors, the primary analysis used a de-duplicated dataset with one representative arm per study cohort according to a prespecified hierarchy: the primary or original/consensus reading reported by the source article, then the arm with the largest analyzable 2×2 denominator, and finally the estimate closest to the within-study median diagnostic odds ratio if ties remained. Pooled sensitivity and specificity were estimated using hierarchical bivariate random-effects models, while alternative arms were retained for sensitivity analyses.

RESULTS: Twenty-one studies met the eligibility criteria and were included in the quantitative synthesis. The overall pooled sensitivity was 0.91 (95% CI: 0.87-0.93) and specificity was 0.88 (95% CI: 0.78-0.94), with an area under the summary receiver operating characteristic curve of 0.95 (95% CI: 0.93-0.97). Between-study heterogeneity was substantial. In modality-stratified analyses, dedicated enterography techniques (magnetic resonance enterography and computed tomography enterography) demonstrated the highest comprehensive diagnostic accuracy, outperforming conventional contrast-enhanced CT and functional imaging.

CONCLUSIONS: Contemporary imaging modalities exhibit high overall diagnostic performance for primary small-bowel tumors. Enterography-based cross-sectional imaging yielded the most favorable point estimates, although confidence intervals overlapped with those of conventional CT. These findings support CTE and MRE as strong diagnostic options within a multimodality pathway rather than proving clear statistical superiority.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420251144585.

PMID:42434747 | PMC:PMC13349763 | DOI:10.3389/fonc.2026.1842330

Front Oncol. 2026 Jun 26;16:1852617. doi: 10.3389/fonc.2026.1852617. eCollection 2026.

ABSTRACT

BACKGROUND: The PACIFIC regimen (consolidation durvalumab following chemoradiotherapy) is the standard of care for unresectable stage III non-small cell lung cancer (NSCLC). With the publication of data from the phase III LAURA trial and the emergence of real-world evidence regarding sequential toxicity, concurrent chemoradiotherapy followed by sequential targeted therapy with EGFR tyrosine kinase inhibitors (TKIs) is recommended for patients with EGFR mutations. However, the optimal combination regimen remains to be determined.

METHODS: We systematically searched the PubMed, Embase, Cochrane Library, and Web of Science databases to identify randomized controlled trials (RCTs) and high-quality retrospective studies comparing various therapeutic strategies for unresectable stage III EGFR-mutated NSCLC. The primary endpoints were progression-free survival (PFS) and overall survival (OS), while secondary endpoints included the objective response rate (ORR) and safety profiles. A network meta-analysis (NMA) was performed using a Bayesian random-effects model. Hazard ratios (HRs), odds ratios (ORs), and their corresponding 95% credible intervals (CrIs) were calculated.

RESULTS: A total of 12 studies involving 1,529 patients were analyzed to compare six therapeutic strategies: consolidation durvalumab following chemoradiotherapy (CRT+Durva), CRT alone, consolidation EGFR-TKIs after CRT (CRT+EGFR-TKI), EGFR-TKI monotherapy, EGFR-TKI in combination with chemotherapy (EGFR-TKI+Chemo), and EGFR-TKI integrated with radiotherapy (EGFR-TKI+RT) via induction, concurrent, or consolidation sequencing. NMA revealed that CRT+EGFR-TKI was the only strategy to demonstrate a statistically significant improvement in OS compared to CRT alone (HR = 0.63, 95% CrI: 0.41-0.94), while also achieving the highest ORR. EGFR-TKI+RT (chemotherapy-free regimen) ranked first for PFS (HR = 0.14, 95% CrI: 0.06-0.33) and exhibited a favorable safety profile, associated with the lowest risk of severe radiation pneumonitis (RP). Notably, CRT+Durva failed to yield a survival benefit (PFS: HR = 0.75; OS: HR = 0.82) and was characterized by higher toxicity. An RCT-only sensitivity analysis demonstrated consistent PFS benefits and a comparable OS trend (HR = 0.68, 95% CrI: 0.33-1.4), validating the integration of real-world data to maintain adequate statistical power.

CONCLUSIONS: For unresectable stage III EGFR-mutated NSCLC, CRT+EGFR-TKI represents the optimal strategy for extending OS. Conversely, the EGFR-TKI+RT (chemotherapy-free regimen) approach provides a superior balance between prolonged PFS and clinical tolerability.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/prospero/, identifier CRD420261285935.

PMID:42434745 | PMC:PMC13349829 | DOI:10.3389/fonc.2026.1852617

Front Oncol. 2026 Jun 26;16:1863609. doi: 10.3389/fonc.2026.1863609. eCollection 2026.

ABSTRACT

BACKGROUND: Fibroblastic/myofibroblastic tumors are among the most common soft-tissue tumors (STTs) encountered clinically. Several magnetic resonance imaging (MRI) features associated with malignant tumors overlap with benign tumors, making differential diagnosis challenging. Intravoxel incoherent motion (IVIM) is a valuable MRI technique for differentiating various tumors. This study aims to evaluate the abilities of conventional MRI and IVIM in differentiating benign, intermediate, and malignant fibrous STTs.

METHODS: Fifty-five patients with fibrous STTs were prospectively enrolled, comprising 18 benign, 18 intermediate, and 19 malignant cases. All the patients underwent MRI examinations including IVIM. Conventional MRI signs and standard-apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D^*), and perfusion fraction (f) were recorded. Statistical analyses were performed using Kruskal-Wallis H test, Chi-square test,post hoc test with Bonferroni correction, receiver operating characteristic (ROC) curves, and DeLong test. p < 0.05 indicated statistical significance.

RESULTS: Malignant tumors had higher heterogeneity on T2WI (p = 0.020 and 0.009) and contrast enhancement T1WI (p = 0.013 and 0.029), and were more prone to necrosis (p < 0.001 andp = 0.001) compared with benign and intermediate tumors, respectively. Tail-like pattern (p = 0.034 and 0.009) and invasiveness (p = 0.018 and 0.033) were more frequently observed in intermediate and malignant tumors than in benign tumors, respectively. Standard-ADC_mean, standard-ADC_min, D_mean, and D_min values decreased from benign to intermediate and malignant fibrous STTs. Malignant STTs displayed higher f_mean and f_min values than benign tumors (p = 0.002 and 0.013, respectively). Standard-ADC_mean showed the highest AUC (0.894) in differentiating intermediate from benign STTs. D_mean showed the highest AUC (0.961 and 0.905) in differentiating malignancies from benign and intermediate STTs, respectively. For discriminating between benign and non-benign fibrous STTs, the combination of conventional MRI signs and IVIM parameters yielded the highest AUC of 0.971.

CONCLUSION: IVIM diffusion parameters differentiated benign, intermediate, and malignant fibrous STTs and can complement conventional MRI signs.

PMID:42434743 | PMC:PMC13349916 | DOI:10.3389/fonc.2026.1863609

Front Oncol. 2026 Jun 26;16:1852055. doi: 10.3389/fonc.2026.1852055. eCollection 2026.

ABSTRACT

PURPOSE: To compare the effectiveness, safety, and tolerability of adjuvant trastuzumab plus pertuzumab (HP) versus trastuzumab emtansine (T-DM1) in patients with HER2-positive breast cancer with residual invasive disease after neoadjuvant therapy, and to perform exploratory analyses of outcomes in clinically favorable subgroups.

MATERIALS AND METHODS: Patients with HER2-positive breast cancer and residual invasive disease after NAT, enrolled between 2020 and 2024, were included. Propensity score matching (1:1) was applied to adjust for baseline characteristic differences. Kaplan-Meier survival analysis and Cox proportional hazards models were used to compare survival outcomes (iDFS, RFS, OS) between the two groups. Additionally, the incidence of adverse events and treatment adherence were compared.

RESULTS: A total of 272 patients were analyzed, with 134 remaining after propensity score matching. After a median follow-up of 37.7 months, no statistically significant differences in short-term survival outcomes were detected between the two groups. Grade 3 or higher adverse events occurred more frequently in the T-DM1 group, particularly thrombocytopenia. Treatment interruption or regimen modification occurred in 22.7% of patients in the T-DM1 group and 2.2% in the HP group.

CONCLUSION: In HER2-positive breast cancer patients with residual invasive disease after NAT, no statistically significant difference in short-term recurrence or survival outcomes was detected between adjuvant HP and T-DM1, while HP was associated with a more favorable safety and tolerability profile. These findings should be interpreted as complementary real-world evidence in the contemporary dual-blockade era and as hypothesis-generating support for future risk-adapted adjuvant strategies.

PMID:42434737 | PMC:PMC13349879 | DOI:10.3389/fonc.2026.1852055

Front Ophthalmol (Lausanne). 2026 Jun 26;6:1830822. doi: 10.3389/fopht.2026.1830822. eCollection 2026.

ABSTRACT

BACKGROUND: Primary open-angle glaucoma (POAG) is the most common form of glaucoma and a leading cause of irreversible blindness worldwide. Treatment focuses on lowering intraocular pressure (IOP), often through cataract extraction with intraocular lens implantation (CE/IOL) alone or combined with minimally invasive glaucoma surgery (MIGS). However, comparative postoperative complication risks between these approaches remain unclear.

METHODS: This retrospective cohort study utilized the TriNetX US Collaborative Network to identify all adults (ages ≥18 years) with a diagnosis of POAG who underwent CE/IOL with or without concurrent MIGS between 2006 and 2026. The cumulative postoperative incidence of hyphema, cystoid macular edema (CME), retinal detachment (RD), and endophthalmitis were evaluated at four time intervals up to 90 days postoperatively. Propensity score matching was used to balance baseline characteristics and reduce confounding. Outcomes were compared using relative risks with 95% confidence intervals, and P-values were calculated using chi-square tests.

RESULTS: After propensity score matching, each routine cataract surgery cohort (with and without MIGS) comprised 7, 998 patients. Patients undergoing CE/IOL with MIGS had a significantly higher rate of hyphema compared with patients undergoing CE/IOL alone at all reported postoperative time points, with a cumulative incidence of 1.19% versus 0.15% at 1-90 days after surgery, respectively (P<0.0001). Conversely, the cumulative incidence of CME (2.585% vs 2.376%, P = 0.3993), RD (0.215% vs 0.139%, P = 0.2566) and endophthalmitis (0.276% vs 0.15%, P = 0.0862) at 1-90 days postoperatively at 1-90 days postoperatively were not statistically different between groups. Similar associations were observed among patients undergoing routine or complex CE/IOL combined with MIGS compared with those undergoing routine or complex CE/IOL alone with respect to hyphema, CME, and RD.

CONCLUSIONS: In this large retrospective cohort study, combined CE/IOL with MIGS was associated with a significantly increased risk of postoperative hyphema, while rates of CME, RD, and endophthalmitis remained comparable to CE/IOL alone.

PMID:42434709 | PMC:PMC13349877 | DOI:10.3389/fopht.2026.1830822

Front Ophthalmol (Lausanne). 2026 Jun 26;6:1776570. doi: 10.3389/fopht.2026.1776570. eCollection 2026.

ABSTRACT

PURPOSE: This study aimed to investigate how age affects the genetic and clinical characteristics of retinitis pigmentosa (RP), focusing on the detectability of causative genes and the age at disease onset.

METHODS: We conducted a single-center study of 506 patients with RP who underwent comprehensive genetic testing through targeted resequencing of 83 known RP-associated genes using next-generation sequencing. Patients were stratified by age at study entry into six groups: <40 years (20-39), 40s (40-49), 50s (50-59), 60s (60-69), 70s (70-79), and ≥80 years. Detection rates of causative genes were calculated and compared across age groups using the Cochran-Armitage trend test. Genetically solved cases included 42 with EYS, 19 with USH2A, 9 with RP1, 14 with RHO, and 7 with RPGR. Clinical data were collected retrospectively. Age at onset was defined as the age when the patient first noticed night blindness, visual field constriction, or decreased best corrected visual acuity. Age at onset was compared across causative genes using an one-way analysis of variance (ANOVA). For pairwise comparisons, the Wilcoxon rank-sum test was applied with Bonferroni correction to adjust for multiple testing.

RESULTS: The mean age of participants was 58.8 years, and our sample included 235 males and 271 females. Case numbers by age group were as follows: <40 years, 58; 40s, 92; 50s, 94; 60s, 125; 70s, 104; and ≥80 years, 33. Detection rates of causative genes declined steadily with age: 39.7% (<40), 41.3% (40s), 36.2% (50s), 27.2% (60s), 19.2% (70s), and 3.0% (≥80), showing a statistically significant trend (p = 8.22 × 10^-7, Cochran-Armitage trend test). In subset analysis, mean onset ages were RPGR (5.2 years), EYS (19.5 years), RHO (24.3 years), RP1 (25.2 years), and USH2A (34.1 years), indicating a significant difference among genes (p < 0.001). Pairwise comparisons showed significantly earlier onset in the RPGR group relative to USH2A (p = 0.004).

CONCLUSIONS: The detection rate of known causative genes of RP was lower in the elderly patients, potentially reflecting factors associated with a late-onset phenotype.

PMID:42434706 | PMC:PMC13349749 | DOI:10.3389/fopht.2026.1776570