Statistical Works
Clinical Works
In many trials and experiments, subjects are not only observed once but multiple times, resulting in a cluster of possibly correlated observations (e.g., brain regions per patient). Observations often do not fulfill model assumptions of mixed models and require the use of nonparametric methods. In this article, we develop and present a purely nonparametric rank-based procedure that flexibly allows the unbiased and consistent estimation of the Wilcoxon–Mann–Whitney effect P(X < Y) + 0.5 P(X = Y) in clustered data designs. Compared with existing methods, we allow flexible weights to be used in effect estimation. Additionally, we develop global and multiple contrast test procedures to test null hypotheses formulated regarding the generalized Mann–Whitney effects and for the computation of range-preserving simultaneous confidence intervals in a unified way. Extensive simulation studies show that these methods control the type-I error rate well and have reasonable power to detect alternatives in various situations.
The drastic growth of the cryptocurrencies market capitalization boosts investigation of their diversification benefits in portfolio construction. In this paper with a set of classical and modern measurement tools, we assess the out-of-sample performance of eight portfolio allocation strategies relative to the naive 1/N rule applied to traditional and crypto-assets investment universe.
Evaluated strategies include a range from classical Markowitz rule to the recently introduced LIBRO approach (Trimborn et al. in Journal of Financial Econometrics 1–27, 2019). Furthermore, we also compare three extensions for strategies with respect to input estimators applied.
The results show that in the presence of alternative assets, such as cryptocurrencies, mean–variance strategies underperform the benchmark portfolio. In contrast, CVaR optimization tends to outperform the benchmark as well as geometric optimization, although we find a strong dependence of the former's success on trading costs.
Furthermore, we find evidence that liquidity-bounded strategies tend to perform very well. Thus, our findings underscore the non-normal distribution of returns and the necessity to control for liquidity constraints at alternative asset markets.
Chronic subdural hematoma (cSDH) is a common condition in older adults that often requires neurosurgical evacuation. Postoperative delirium is a frequent and clinically relevant postoperative complication in this population. This study investigated whether burr hole craniotomy for cSDH performed under local anesthesia (LA) reduces the risk of postoperative delirium and complications compared to general anesthesia (GA).
The ABC-SDH trial was a single-center, open-label, phase 2, prospective randomized clinical trial conducted at a tertiary academic medical center between October 2023 and November 2024. Fifty consecutive patients with a confirmed diagnosis of cSDH underwent burr hole craniotomy performed under either LA or GA. Primary outcomes were postoperative delirium (Confusion Assessment Method, CAM) and complications rates until discharge, and the recruitment rate. Secondary outcomes included procedural times and clinical outcome measures at discharge and at 30-day follow-up.
LA was associated with a significantly lower rate of postoperative delirium (4% vs. 32%, OR 0.09; 95%CI, 0.01–0.79; p = 0.03; NNT = 4). Complications occurred in 8% (LA) versus 32% (GA) (OR 0.18, 95% CI 0.03–0.96; p = 0.08). LA was also associated with significantly reduced procedural times (139.3 ± 53.0 vs. 196.0 ± 52.1 min; p = 0.002). The severity of complications and functional outcomes were comparable between groups.
LA was associated with significantly lower rates of postoperative delirium and a shorter periprocedural duration compared to GA in patients undergoing cSDH evacuation. As the first prospective trial to systematically assess the incidence of delirium following burr hole craniotomy for cSDH, the ABC-SDH trial highlights the feasibility and safety of LA in a delirium-prone population.
Epidemiological data show that while age-standardized mortality rates of oral squamous cell carcinoma (OSCC) have slightly declined, the absolute number of deaths continues to rise, with a concerning increase among younger patients and persistently high mortality in the elderly. The aim of this study was to assess the influence of age on oncological prognosis and to classify it within a multivariable model alongside established histopathological risk factors.
In this retrospective single-center study, patients surgically treated for OSCC between 2012 and 2023 were included according to predefined eligibility criteria and stratified into three age groups (< 50, 50–69, ≥ 70 years). Primary endpoints were overall survival (OS) and disease-free survival (DFS). Prognostic factors were further evaluated with multivariable Cox regression models and competing-risk analyses, with age assessed both as a categorical and continuous variable using restricted cubic splines.
In 525 included patients (mean age 63.5 years), age distribution was multimodal with three peaks in concordance with the predefined groups. Local recurrence, distant metastasis, and secondary cervical lymph node metastasis (CLNM) were weakly correlated overall, with a stronger association between CLNM and distant metastasis in younger patients. Age significantly predicted overall survival (HR per 10 years: 1.56, 95% CI: 1.29–1.90), but showed no independent association with disease-free survival.
Age presented with a multimodal distribution with distinct clinical patterns across the subgroups and was independently associated with worse OS. Nevertheless, no association was observed for DFS, suggesting that age-related differences in mortality may be influenced by non-oncologic factors. While correlations between recurrence events were generally weak, younger patients displayed a stronger link between regional and distant metastasis, emphasizing the importance of age-related follow-up.
Moyamoya angiopathy (MMA) is a cerebrovascular disease treated by surgical revascularization. This study aimed to present long-term outcomes of European MMA patients after surgical revascularization.
MMA patients treated surgically between 1998 to 2023 were included. Follow-up comprised clinical and radiological examinations and electronic questionnaires. Patient data were analyzed retrospectively, and logistic regressions were performed to identify predictors of unfavorable outcomes.
A total of 276 patients with 428 surgically treated hemispheres were included. The mean age was 35.0 ± 17.2 years, including 192 women and 84 men, predominantly of Caucasian ethnicity (88.0%). Ischemia (82.2%) and hemorrhage (10.1%) were the most common symptoms at onset. Most adults (80.9%) were treated with combined bypass surgery via mini-craniotomy. Within 30 days, complications occurred in 6.3% of surgeries, and the perioperative stroke risk was 2.6%. A total of 148 patients underwent in-house follow-up. No mRS scores ≥ 5 were observed, with 86.3% of patients achieving a score of 0–2, alongside a bypass patency rate of 95.7% and a stroke risk of 0.3% per patient-year. Preoperative stroke and silent ischemic MRI lesions were identified as predictors of an unfavorable outcome. Additionally, 201 patients were followed up electronically. No mRS scores ≥ 5 were observed, with 87.9% of patients achieving a score of 0–2. Moreover, 90.7% were independent in daily life, but only 50.5% returned to work or school.
Combined bypass surgery yielded favorable outcomes with low complication and stroke rates. Patients with preoperative stroke require closer follow-up, and individualized strategies are needed to address persistent functional impairments.
Evidence regarding the clinical course of unruptured giant intracranial aneurysms (GIAs) after conservative management (CON), microsurgical treatment (SURG) or endovascular treatment (EVT) remains limited. We aimed to assess mortality, functional outcome, symptom course and retreatment rates after different management strategies.
In this prospective, multinational registry, patients with unruptured GIAs treated at 37 neurovascular centres between 2008 and 2018 were included. Outcomes after CON, EVT or SURG were analysed with standardised 3-year follow-up. The study is registered at ClinicalTrials.gov.
We included 339 patients of whom 22.7% received CON, 42.8% EVT and 34.5% SURG. Three-year survival was 64.0% (95% CI 53.73% to 76.20%) in the CON group, 82.0% (95% CI 75.42% to 89.17%) in the EVT group and 93.5% (95% CI 88.91% to 98.28%) in the SURG group (p<0.001). Favourable neurological outcome (modified Rankin Scale 0–2) declined to 53.0%, 66.9% and 75.7% in the CON, EVT and SURG groups, respectively (p<0.01). Symptom improvement at 3 years occurred in 10.3% of CON, 22.1% of EVT and 11.1% of SURG patients, while deterioration was observed in 7.7%, 8.5% and 15.8%, respectively. Retreatment was required in 22.8% of EVT patients and 6.2% of SURG patients (p<0.01).
EVT and SURG were associated with significantly improved survival and functional outcomes compared with conservative management of unruptured GIAs. In patients eligible for intervention, both treatment modalities represent reasonable and effective options, with a high proportion of patients achieving favourable clinical outcomes.
This study aimed to evaluate the transfer accuracy of virtually planned maxillary repositioning within a Surgery First protocol using 3D-printed guides and patient-specific implants. In addition, the feasibility of a fully digital interdisciplinary workflow was analyzed.
The patients in this study were treated with a completely virtually planned interdisciplinary treatment pathway according to the Surgery First concept using a 3D-printed maxillary guide and patient-specific implants as well as a 3D-printed final occlusion-splint. The osteosynthesis in the mandible was performed conventionally using miniplates. The postoperative cone beam computed tomography (CBCT) was overlaid with the planning data set and the transfer precision was analyzed with defined bony landmarks in the upper jaw using Onyx Ceph software (Image Instruments).
63 patients were included in the analysis. The gender distribution was almost 2:1. Included were 41 female and 22 male patients. Mean age was 37.9 ± 10.1 years, and 12 patients presented with OSAS. Postoperative complications were rare and managed conservatively. The transfer accuracy of maxillary positioning was within ±1–2 mm at all landmarks, with only occasional minimal exceedance of the 3-mm threshold, indicating high clinical accuracy. Regional differences were observed, with slightly larger deviations at L4 and L5 and smaller deviations at L9–L11 (global p < 0.01); however, these did not compromise overall clinical precision.
Virtual surgical planning combined with patient-specific cutting guides and customized osteosynthesis plates enables clinically accurate maxillary positioning in Surgery First patients, with transfer deviations predominantly within the accepted 1–2 mm threshold. The fully digital interdisciplinary workflow allows predictable intraoperative transfer despite the absence of a presurgical occlusal reference and can be safely applied across different indications for bimaxillary surgery.
Nightmares are a hallmark of post-traumatic stress disorder (PTSD), yet pharmacological treatments remain limited. Here we conducted a multicenter, double-blind, randomized, placebo-controlled trial to evaluate the efficacy and safety of dronabinol (BX-1; Δ9-tetrahydrocannabinol (THC)) for PTSD-related nightmares. Adults with PTSD and recurrent nightmares were randomly assigned to receive dronabinol (2.5–15 mg) or placebo, once daily before bedtime, for 10 weeks. The primary endpoint was change from baseline to week 10 in the Clinician-Administered PTSD Scale for DSM-IV (CAPS-IV) B2 item assessing nightmare frequency and intensity. A total of 171 patients underwent randomization (mean age, 37.9 ± 12.8 years; 79.3% female), with 87 assigned to dronabinol and 84 to placebo. At week 10, reduction in CAPS-IV B2 scores was significantly greater with dronabinol than with placebo (between-group difference −1.50; 95% confidence interval −2.28 to −0.71; Cohen’s d = 0.65; P < 0.001). Adverse events occurred in 78 patients (89.7%) receiving dronabinol and 64 (77.1%) receiving placebo. Treatment discontinuation due to adverse events occurred in five (5.7%) and six (7.2%) patients, respectively. Serious adverse events occurred in seven patients (8.0%) receiving dronabinol and in none receiving placebo. This trial provides evidence that dronabinol reduces the frequency and intensity of PTSD-related nightmares, although long-term efficacy and safety require further evaluation.
Moyamoya angiopathy (MMA) is characterized by the plasticity to develop endogenous collateral blood vessels to compensate for progressive steno-occlusion of proximal intracranial arteries. Bypass surgery has been anecdotally reported to induce regression of these collateral vessels, but a detailed analysis of their natural history is lacking. Here, the authors characterize these collaterals after bypass surgery.
A single-center retrospective analysis of the medical records of 81 predominantly Caucasian MMA patients (121 hemispheres) treated with a combined superficial temporal artery–middle cerebral artery bypass and encephalodurosynangiosis between January 2011 and December 2021 was performed. Clinical data and longitudinal angiographic images were investigated to compare the development of different collateral types and to analyze the dependency between collateral vessels and bypass quality.
A total of 58 female and 23 male patients with a mean age of 41 ± 13.1 years at the time of first surgery were included. The majority of patients (92.6%) were European Caucasian. Ischemic events were the most common onset symptom (88.9%), followed by hemorrhage (11.1%). The mean follow-up time of digital subtraction angiography examinations was 19.8 ± 20.4 (range 0–108) months. Postoperatively, the majority of collateral vessels showed no changes over time. If changed, deep basal MMA collaterals as well as anterior leptomeningeal collaterals showed a consistent reduction over time, whereas posterior leptomeningeal collaterals, callosal collaterals, and extracranial-intracranial collaterals showed an increase more frequently (p < 0.006). Endogenous collateral vessels developed irrespective of bypass quality, while direct and indirect bypasses showed a synergistic development.
This study represents the first longitudinal angiographic characterization of endogenous collateral vessels in Caucasian MMA patients after combined bypass surgery. Collaterals within the region of the anterior circulation supplied by the bypass showed a consistent reduction over time. The development of collaterals depending on the presence and location of the bypass but not its quality indicates the individual endogenous need of moyamoya hemispheres as the determining factor and highlights the enduring plasticity and dynamic nature of the MMA collateral system over time.
The T classification of oral squamous cell carcinoma (OSCC) is linear according to the tumor size, excluding T4a by its criteria of invasion into adjacent structures, such as cortical bone. This may lead to the upstaging of otherwise small tumors. The objective was to analyze patients with OSCC and negative nodal staging to assess the impact of T-staging with tumor size on the incidence of occult cervical lymph node metastasis (CLNM) and regional neck failure.
This retrospective cohort study included patients with OSCC and clinically negative necks (cN0), treated surgically between 2010 and 2024. All T4a OSCC classified due to bone invasion were additionally reclassified into T1–T3 based on size and depth of invasion according to the current staging manual. The primary endpoint of this study was the association between OSCC stratified by T-stage and tumor size as well as the presence of occult CLNM.
A total of 642 patients were included, with an overall occult CLNM rate of 20.2%. Bone invasion in T1-sized tumors was significantly associated with occult CLNM (OR 6.38, 95% CI: 1.48–27.42), whereas no such association was observed in T2 or T3 tumors (OR 0.80, 95% CI: 0.37–1.73; and OR 0.77, 95% CI: 0.37–1.62, respectively). Additionally, in T1–T2 tumors, bone invasion did not correlate with worse survival outcomes.
Bone invasion was not significantly associated with occult CLNM in T2-3 sized OSCC, suggesting that the prognostic relevance is size-dependent. These findings question the uniform upstaging to T4a and support a more differentiated approach, potentially enabling neck management de-escalation in selected early-stage cases.
Oral squamous cell carcinoma (OSCC) with bone invasion are staged as pT4a, potentially upstaging smaller tumors. This study aimed to evaluate the oncological benefit of postoperative radiotherapy (PORT) in pT4aN0 OSCC with respect to tumor size and without other risk factors.
This retrospective matched cohort study included pT4aN0 OSCC patients with bone invasion treated surgically (R0) between 2010 and 2022. Each case was 1:1 matched to a pT1–3 N0 OSCC patient based on tumor size, but without bone invasion. The primary endpoint was overall survival (OS), secondary endpoints included the recurrence-free survival and outcome predictors.
A total of 156 patients were analyzed (78 per group). There were no statistically significant differences in 3-year OS between both groups in general (78.2%, 95%CI: 68.6–87.8 vs. 80.0%, 95%CI: 68.4–91.6). After stratification for pT2 criteria, there was also no significant difference between both groups if PORT was omitted (63.9%, 95%CI: 44.2–92.4 vs. 70.5%, 95%CI: 55.0–90.0). Multivariate analysis identified age and poor differentiation (grade III) as significant predictors of worse OS, while PORT showed no independent survival benefit.
In small OSCC staged pT4a due to bone invasion and lacking other risk factors, PORT demonstrated no statistically significant improvement in OS when matched for tumor size. Further prospective trials and larger cohorts are warranted to confirm these findings.
Acute ischemic stroke (AIS) is a leading cause of permanent disability in adults and one of the most time-sensitive emergencies in modern medicine. Rapid diagnosis and initiation of thrombolytic therapy, as well as immediate access to mechanical thrombectomy for patients with large-vessel occlusion (LVO), are critical determinants of favorable outcomes. While Mobile Stroke Units (MSUs) – ambulances equipped with computed tomography (CT) imaging – have demonstrated efficacy in improving outcomes, their deployment is often constrained to urban environments due to cost-efficiency considerations. Blood biomarkers offer a potentially cost-effective alternative for stroke diagnosis and subtyping. Especially for patients with LVO, a prehospital biomarker-based identification could enable a streamlined transport strategy directly to thrombectomy-capable stroke centers. In this study, we evaluate the diagnostic accuracy and feasibility of a novel point-of-care test (POCT) in predicting LVO stroke, along with the levels of vascular biomarkers—NT-proBNP, D-Dimer, and H-FABP—from ultra-early whole-blood samples of patients with suspected stroke in a prehospital setting. The test integrates a blood-based multiplex lateral flow assay (LFA) with clinical decision support (CDS) software, accessible through a Mobile Application (App). The study was registered in the German Clinical Trials Register (DRKS-ID: DRKS00037840).
This multicenter prospective observational study will include 800 patients with suspected stroke, enrolled within 24 h after symptom onset. Participants will be recruited at three Mobile Stroke Unit (MSU) sites in Berlin, Germany. Prehospital blood samples will be analyzed directly in the ambulance using the LVOCheck device. The test achieves its performance by quantifying the blood levels of the vascular biomarkers (NT-proBNP, D-Dimer, and H-FABP), and inputting these concentrations, along with neurological assessment score and patient-specific medical information such as age and blood pressure, to output predictive information on the probability of LVO stroke, together with all the input data. Additional clinical data, including final diagnoses, will be documented in electronic case report forms (eCRFs). The diagnostic performance of LVOCheck will be evaluated through comprehensive statistical analysis of the combined biomarker and clinical data, including assessment of sensitivity, specificity, and predictive models.
This real-world study aims to evaluate the diagnostic accuracy and feasibility (including human factors and usability) of LVOCheck, a portable, non-invasive multiplex POCT at prehospital settings, such as an ambulance, in the prediction of acute LVO diagnosis and triage of suspected acute LVO stroke patients. Utilizing Mobile Stroke Units (MSUs) as recruiting sites enables the analysis of ultra-early biomarker levels from prehospital whole-blood samples combined with patient-specific medical information to provide a predictive risk of LVO stroke. A cost-effective and practical POCT could provide additional biochemical information to support prehospital LVO identification in the future, thereby potentially assisting emergency medical services in transport decisions regarding direct transfer to thrombectomy-capable centers. This may improve triage efficiency, treatment metrics and outcomes in both urban and rural settings.
In advanced oral squamous cell carcinoma (OSCC), adjuvant therapy (AT) is an important part of the treatment to ensure extended locoregional control after primary surgical resection. The impact of the time interval between surgery and AT on the oncological prognosis remains unclear, particularly in high-risk constellations. The aim of this study is to categorize treatment delays and to determine their impact on the oncological prognosis within the context of the histopathological risk parameters of patients with advanced OSCC.
In this single-institutional retrospective cohort study, all patients treated for OSCC between 2016 and 2021 and who received postoperative chemoradiation (POCRT) were included. Patients were divided into two groups: Group I: ≤ 6 weeks between surgery and POCRT; and Group II: > 6 weeks between surgery and POCRT.
Overall, 202 patients were included (Group I: 156 (77.2%) vs. Group II: 46 (22.8%)). There were no statistically significant differences in epidemiological aspects and histopathological risk factors between the two groups. The maximum time to initiation of POCRT was 11 weeks. Delayed POCRT initiation had no statistically significant influence on the 5-year OS (61.6% vs. 57.3%, p = 0.89), locoregional control rate (38.6% vs. 43.3%, p = 0.57), and RFS (32.3% vs. 30.4%, p = 0.21). On multivariate analysis, extracapsular spread (HR: 2.21, 95% CI: 1.21 – 4.04, p = 0.01) and incomplete surgical resection (HR: 2.01, 95% CI: 1.10 – 3.69, p = 0.02) were significantly correlated with OS. For RFS, ECS (HR: 1.82, 95% CI: 1.15 – 2.86, p = 0.01), incomplete resection (HR: 1.67, 95% CI: 1.04 – 2.71, p = 0.04), and vascular infiltration of the tumor (V-stage; HR: 2.15, 95% CI: 1.08 – 4.27, p = 0.03) were significant risk predictors.
Delays in POCRT initiation up to 11 weeks after surgical resection for advanced OSCC were not statistically significantly associated with impaired survival. In cases of prolonged surgical treatment due to management of complications, a small delay in AT beyond the recommended time limit may be justified and AT should still be pursued.
Targeted therapies for metastatic uveal melanoma have shown limited benefit in biomarker-unselected populations. The Treat20 Plus study prospectively evaluated the feasibility of a precision oncology strategy in routine clinical practice.
Fresh biopsies were analyzed by high-throughput genomics (whole-genome, whole-exome, and RNA sequencing). A multidisciplinary molecular and immunologic tumor board (MiTB) made individualized treatment recommendations based on identified molecular aberrations, patient situation, drug, and clinical trial availability. Therapy selection was at the discretion of the treating physician. The primary endpoint was the feasibility of the precision oncology clinical program.
Molecular analyses were available for 39/45 patients (87%). The MiTB provided treatment recommendations for 40/45 patients (89%), of whom 27/45 (60%) received ≥1 matched therapy. First-line matched therapies included MEK inhibitors (n = 15), MET inhibitors (n = 10), sorafenib (n = 1), and nivolumab (n = 1). The best response to first-line matched therapy was partial response in one patient (nivolumab based on tumor mutational burden), mixed response in two patients, and stable disease in 12 patients for a clinical benefit of 56%. The matched therapy population had a median progression-free survival and overall survival of 3.3 and 13.9 months, respectively. The growth modulation index with matched therapy was >1.33 in 6/17 patients (35%) with prior systemic therapy, suggesting clinical benefit.
A precision oncology approach was feasible for patients with metastatic uveal melanoma, with 60% receiving a therapy matched to identify molecular aberrations. The clinical benefit after checkpoint inhibitors highlights the value of tumor mutational burden testing.
The Berlin Grading System assesses clinical severity of moyamoya angiopathy (MMA) by combining MRI, DSA, and cerebrovascular reserve capacity (CVRC). Our aim was to validate this grading system using [15O]H2O PET for CVRC. We retrospectively identified bilateral MMA patients who underwent [15O]H2O PET examination and were treated surgically at our department. Each hemisphere was classified using the Suzuki and Berlin Grading System. Preoperative symptoms and perioperative ischemias were collected, and a logistic regression analysis was performed. A total of 100 hemispheres in 50 MMA patients (36 women, 14 men) were included. Using the Berlin Grading System, 2 (2.8%) of 71 symptomatic hemispheres were categorized as grade I, 14 (19.7%) as grade II, and 55 (77.5%) as grade III. The 29 asymptomatic hemispheres were characterized as grade I in 7 (24.1%) hemispheres, grade II in 12 (41.4%), and grade III in 10 (34.5%) hemispheres. Berlin grades were independent factors for identifying hemispheres as symptomatic and higher grades correlated with increasing proportion of symptomatic hemispheres (p < 0.01). The Suzuki grading did not correlate with preoperative symptoms (p = 0.26). Perioperative ischemic complications occurred in 8 of 88 operated hemispheres. Overall, complications did not occur in any of the grade I hemispheres, but in 9.1% (n = 2 of 22) and 9.8% (n = 6 of 61) of grade II and III hemispheres, respectively. In this study, we validated the Berlin Grading System with the use of [15O]H2O PET for CVRC as it could stratify preoperative symptomatology. Furthermore, we highlighted its relevance for predicting perioperative ischemic complications.