University of Maryland Medical Center

The proliferation of administrative databases, electronic health records, and large observational cohorts has profoundly expanded the scope of surgical research. At the same time, it has intensified a long-standing interpretive error: the routine conflation of statistical significance with clinical importance. In large data sets, small and often unstable numerical differences readily achieve conventional significance thresholds, driven by sample size, analytic flexibility, and selective reporting rather than by clinically meaningful or biologically coherent effects. Epidemiologic evidence demonstrates that null hypothesis significance testing (NHST) and dichotomous P -value thresholds are poorly suited for many observational, prognostic, and exploratory analyses common in surgical research, where effect size, precision, robustness, and plausibility are the true determinants of evidentiary value. This editorial synthesizes methodological insights from epidemiology, biostatistics, and surgical outcomes research to explain how selection processes, analytic latitude, publication bias, and large-sample artifacts systematically inflate the appearance of statistically "positive" findings. These concepts are translated into clinician-focused language, with illustrative examples drawn from surgery, where low-frequency but high-impact outcomes are particularly vulnerable to statistical fragility. To address these challenges, we introduce the CLARITY-SURG Statement (Clinical Linking of Association, Robustness, and Interpretability), a practical interpretive framework designed to help clinicians correctly evaluate observational study results. The framework provides a structured approach for assessing effect magnitude and precision, interrogating biological and surgical plausibility, examining analytic robustness, and determining the appropriate level of inference. By prioritizing clinical interpretability over threshold-based significance testing, the CLARITY-SURG Statement offers a reproducible, clinician-centered method for interpreting observational surgical research and for improving the quality, credibility, and clinical relevance of reported statistical findings.

Conditional survival (CS) provides estimates of how survival changes over time. The primary objective of this secondary analysis is to report CS estimates for patients with malignant gliomas enrolled on RTOG 9006. A post hoc analysis of RTOG 9006 was performed using data obtained from the NCI NCTN Data Archive. Eligible patients included those enrolled on RTOG 9006 (n=632). CS estimates from the time of diagnosis and following 1, 3, and 5 years of survival were calculated using Kaplan-Meier. Multivariable Cox proportional hazards modeling multivariate analysis (MVA) was performed to evaluate the prognostic significance of age, KPS, treatment arm, histology, and extent of surgical resection following 1, 3, and 5 years of survival. The RPA class was evaluated on univariate analysis following the same initial survival periods. Among 632 patients, OS at 1 and 3 years was 50.9% and 17.3%, respectively. CS improved over time, with patients who survived the first year having a 46.4% chance of surviving an additional year and 82.9% chance for those who had already survived 3 years. Age, KPS, and extent of surgical resection were significant predictors of OS at diagnosis but prognostic value declined over time. Glioblastoma multiforme (GBM) histology was significant at all time points. Patients who survived 3 years had significantly improved additional survival with conventional fractionation compared with hyperfractionation. In addition, RPA lost prognostic significance over time. Given the diminishing influence of KPS and surgical resection, treatment decisions should be based on individualized approaches considering evolving survival probabilities, not solely initial prognosis.

In surgical research, statistical sophistication is too often mistaken for scientific rigor. Across a growing body of plastic surgery literature, adjusted odds ratios, hazard ratios, and regression coef- ficients are frequently presented without the crude event rates or absolute measures of effect that give findings clinical meaning. We describe this phenomenon as "runic statistics": results that are statistically valid yet clinically opaque. Through examples drawn from contemporary plastic surgery studies, we highlight three recurrent interpretive flaws: reliance on statistical significance without consideration of clinical relevance, reporting of relative measures without baseline risks or absolute differences, and conflation of association with causation. We further demonstrate how case-mix imbalances can create apparent contradictions in results (Simpson's paradox), and how identical odds ratios can translate into very different clinical implications depending on the base- line risk. To address these challenges, we propose a thirteen-step reporting framework designed to promote transparency, interpretability, and clinical applicability. Key elements include explicit definition of the estimand, presentation of both crude and adjusted data, translation of relative effects into absolute risks and patient-facing numbers, assessment of minimal clinically important differences, careful handling of confounding, and restraint in the use of causal language. By an- choring statistical reporting in clinical realities, surgical research can remain both methodologically rigorous and directly relevant to patient care. Our goal is not to simplify science, but to ensure that its communication is clear, transparent, and ultimately useful at the bedside.

158 Background: We previously reported that olaparib significantly improves radiographic progression-free survival (rPFS) when added to radium-223 for bone-dominant mCRPC in the COMRADE study (NCT03317392). Here, we investigate ctDNA and PSA dynamics by arm from patients on the COMRADE study. Methods: Cell free DNA from banked plasma was sequenced on FoundationOne Monitor as a research use test. ctDNA was quantified as ctDNA tumor fraction (TF). rPFS and overall survival (OS) were assessed by Kaplan-Meier analysis and univariate Cox proportional hazard models, landmarked from the relevant sample collection date; patients with progression before the landmark were excluded. Molecular response (MR) and PSA response (PSA50) were defined as ≥50% decrease from C1D1. Results: Of 113 patients receiving any study treatment, ctDNA results were successfully generated for 102 patients at cycle 1 day 1 (C1D1), of whom 53 were randomized to radium-223+olaparib (Arm A) and 49 to radium-223 alone (Arm B). Across arms, the median ctDNA TF at C1D1 was 16% [Interquartile Range: 1.3-43%]. Across arms, 99 patients had ctDNA results at C2D1: 5 with progression prior to C2D1, 9 patients with no result at C1D1, and 8 with unquantifiable ctDNA change were excluded. ctDNA TF clearance from C1D1 to C2D1 was achieved in 13/77 patients (17%; Arm A: 17%, Arm B: 17%), and 50-99% reduction in 14/77 patients (18%; Arm A: 22%, Arm B: 15%). 15/77 patients (19%; Arm A: 22%; Arm B: 17%) were negative at both timepoints and not included in assessment of MR. MR was associated with significant, favorable PFS (MR: 4.6 months vs No MR: 2.0 months, p=0.01) and OS (MR: 21.2 months vs No MR: 12.2 months, p=0.005) with similar trends within both treatment arms. A greater improvement in rPFS with the addition of olaparib was observed for patients who achieved MR (MR: HR = 0.35 [95% confidence interval (CI): 0.14-0.91]; No MR: HR = 0.65 [95% CI 0.29-1.43]). Similar to TF, baseline PSA was prognostic: when stratified at the median (51.6 ng/mL), patients with lower PSA demonstrated significantly improved rPFS (11.0 vs 3.8 months for low vs high PSA, p<0.0001) and OS (26.6 vs 13.2 months, p<0.0001). Of the 13 patients that ever-achieved PSA50, including after crossover, 69% achieved MR compared to 37% in 49 patients without PSA50. Conclusions: Early MR (by C2D1) was achieved in 35% of patients and strongly associated with improved rPFS and OS while PSA50 responses by C2D1 were not observed in any patients. The rPFS benefit of adding olaparib to radium-223 appeared more evident in those who achieved a MR. Early on treatment ctDNA dynamics in bone-dominant mCRPC may be a valuable tool for treatment decisions with radium-223 therapy. Clinical trial information: NCT03317392 .

Background: Prone trans-psoas (PTP) interbody fusion is a minimally invasive technique for sagittal correction in adult spinal deformity. Advantages include enhanced lordosis due to prone positioning and avoidance of patient repositioning. The same approach allows for anterior longitudinal ligament release, which may improve lordotic correction. However, limited data exist on the combined technique, prompting this investigation into postoperative complications and radiographic outcomes of PTP interbody fusion versus combined PTP with anterior longitudinal ligament (ALL) release (PTP-ALL). Methods: A retrospective cohort study was conducted at a tertiary academic center from November 2020 to October 2023. Patients were grouped into two cohorts: PTP and PTP-ALL. Radiographic outcomes included changes in segmental lordosis, lumbar lordosis, L4-S1 lordosis, pelvic incidence-lumbar lordosis (PI-LL) mismatch, T1-pelvic angle, and sagittal vertical axis. Complications assessed were wound infections, venous thromboses, transfusions, and mortalities. Welch’s modified two-sample t-tests and chi-squared tests were used for analysis. Results: Twenty-two patients underwent PTP and six underwent PTP-ALL. There were no significant differences in operative time ( P =0.443), length of stay ( P =0.169), or estimated blood loss ( P =0.795). No complications occurred in the PTP-ALL group. Compared to PTP, PTP-ALL achieved greater improvements in segmental lordosis (9.3° vs. 1.6°, P =0.134), lumbar lordosis (30.6° vs. 8.5°, P =0.003), L4-S1 lordosis (10.2° vs. 6°, P =0.25), and PI-LL mismatch correction (25.8° vs. 8.9°, P =0.036). Conclusions: PTP with ALL release may be a safe and effective option to enhance sagittal correction in adult spinal deformity without increasing perioperative complications. However, the small sample size limits the generalizability of these findings and underscores the need for further study. Level of Evidence: III

The expansion of treatment options for prostate cancer (PC) has improved disease-specific and overall survival outcomes but has also raised questions about the optimal level of treatment needed for patients based on their individual prognosis and accounting for potential toxicity, incorporating quality of life considerations. A panel of experts met to discuss current controversies in the care of patients with PC across the disease continuum. Multidisciplinary experts review advances and persistent uncertainties in biomarker-guided assessment, imaging, and systemic therapy for prostate cancer. The discussion outlines priority gaps in evidence that must be addressed to optimize individualized patient care. Workshop topics included use of genomic biomarkers and artificial intelligence-guided tools to identify and manage high-risk and very-high risk localized disease, management of biochemical recurrence, identification of patients with metastatic hormone-sensitive PC who warrant treatment escalation, radiopharmaceutical therapy for metastatic castration-resistant PC including optimal sequencing of approved therapies, role of imaging in identification and management of extraprostatic disease, and lifestyle interventions to optimize survivorship. Many questions remain about management of PC related to biomarker-based risk stratification to guide treatment selection, use of prostate-specific membrane antigen-positron emission tomography, and balancing the risk for PC-related death with risks for treatment-related toxicity. Ongoing research efforts are needed to optimize risk-based treatment, sequence of therapies throughout the disease continuum, and survivorship care.

ABSTRACT Mesenchymal stem cell extracellular vesicles (MSC EVs) have been widely studied for regenerative medicine and tissue repair applications. However, clinical translation of EV therapeutics has been hampered by low potency and a lack of scalable production strategies. This work aims to develop a novel approach that exploits the mechanosensitivity of MSCs to enhance EV potency in the context of enhanced production via bioreactor culture. MSCs are well known to respond to mechanical stimuli such as substrate stiffness and shear stress, and here it is shown that exposing MSCs to another mechanoregulatory parameter, confinement, enhances the pro‐angiogenic bioactivity of their EVs. This is consistent across both donor‐derived primary MSCs and induced pluripotent stem cell‐derived MSCs (iMSCs). The development of a 3D‐printed perfusion bioreactor system that enables culture of confined MSCs under flow is also detailed here, resulting in a 67‐fold increase in EV production compared to flask culture. iMSC EVs obtained downstream of this confinement‐bioreactor culture induce greater vascularization and generally improve wound healing in a diabetic mouse model compared to iMSC EVs from conventional tissue culture. Overall, this work establishes the development of a scalable, bioreactor‐based iMSC EV production platform that provides a solution to major translational bottlenecks of therapeutic EVs.

Introduction: Conservative or upfront radical management for high- and very high-risk non-muscle-invasive bladder cancer continues to be debated, particularly for cases with adverse pathological features. We aimed to compare survival outcomes among NMIBC patients treated with transurethral resection of bladder tumour (TURBT) followed by either Bacillus Calmette-Guérin (BCG), sequential BCG plus electromotive administration of mitomycin C (EMDA-MMC), or upfront radical cystectomy (RC). Materials and Methods: High- and- very high-risk NMIBC cases undergoing TURBT followed by BCG, BCG plus EMDA-MMC, or RC at two international tertiary referral centres between 2009 and 2024 were retrospectively reviewed. Recurrence-free survival (RFS), progression-free survival (PFS), and overall survival (OS) were estimated using Kaplan-Meier methods. Multivariable Cox regression models were applied to identify factors independently associated with survival outcomes. Results: A total of 1178 patients were included: 852 received BCG, 249 received BCG/EMDA-MMC, and 77 underwent upfront RC. Kaplan-Meier analysis revealed no significant differences in RFS or PFS between the BCG and BCG/EMDA-MMC groups, nor in OS between the three treatment strategies. According to multivariable analysis, concomitant carcinoma in situ (CIS) and increasing T stage at TURBT were independently associated with poorer RFS (HR 1.39; 95% CI 1.05-1.85), PFS (HR 1.95; 95% CI 1.36-2.82), and OS (HR 2.28; 95% CI 1.60-3.25). A second resection conferred a protective effect on PFS (HR 0.72; 95% CI 0.54-0.95). Treatment modality (BCG, BCG/EMDA-MMC, or upfront RC) was not significantly associated with any survival endpoint. Conclusions: In this large multicentre series of patients with high- and very high-risk NMIBC undergoing TURBT, survival outcomes were primarily influenced by clinical-pathological characteristics rather than the adjuvant treatment of choice.

Background and Objectives: Endovascular thrombectomy (EVT) is effective for select patients with large vessel occlusion (LVO) stroke. While baseline dementia is a known negative predictor for good post-EVT outcomes, comparative efficacy and safety data of EVT versus best medical management (BMM) for LVO patients with dementia is limited. Herein, we compared hospitalization outcomes of EVT versus BMM for LVO patients with underlying dementia. Methods: Using the Nationwide Readmissions Database from 2016-2022, we identified adult patients with dementia who presented with acute ischemic stroke due to LVO. The primary outcome was discharge to home (as opposed to a rehabilitation or nursing home facility). Secondary outcomes included in-hospital mortality, length of stay, and complications. Patients who underwent EVT were compared to those who received BMM, after propensity score matching (PSM) to adjust for confounders. Interaction and subgroup analyses assessed treatment effect heterogeneity associated with age and dementia type. Results: Among 37,298 patients included, 6,622 (17.8%) received EVT. After PSM, 6,551 EVT and 13,029 BMM patients remained. Discharge to home rates were similar between groups (27.0% vs 26.1%, p=0.41). EVT was associated with longer length of stay (median 7 vs 5 days, p<0.001) and increased intracranial hemorrhage (23.9% vs 11.4%, p<0.001), with no significant difference in mortality (15.1% vs. 14.9%, p=0.79). Patient age significantly moderated EVT effectiveness (interaction p=0.001), and EVT among patients <75 years old was significantly associated with higher rates of home discharge (36.4% vs 29.3%, p=0.003). Additionally, EVT appeared more effective for patients with vascular dementia (as compared to Alzheimer’s disease, interaction p=0.016), among whom EVT was associated with significantly higher rates of home discharge (32.7% vs. 25.2%, p=0.008) Conclusion: Clinical response to EVT treatment is heterogeneous among LVO stroke patients with underlying dementia. Younger patients (<75 years old) and those with vascular dementia significantly benefited from treatment. Thus, dementia alone should not be protocolized as a general exclusion criteria for EVT, and future studies are needed to further refine patient selection strategies.