Covariance Model Research Articles

Mode-wise principal subspace pursuit and matrix spiked covariance model

Abstract This paper introduces a novel framework called Mode-wise Principal Subspace Pursuit (MOP-UP) to extract hidden variations in both the row and column dimensions for matrix data. To enhance the understanding of the framework, we introduce a class of matrix-variate spiked covariance models that serve as inspiration for the development of the MOP-UP algorithm. The MOP-UP algorithm consists of two steps: Average Subspace Capture (ASC) and Alternating Projection. These steps are specifically designed to capture the row-wise and column-wise dimension-reduced subspaces which contain the most informative features of the data. ASC utilizes a novel average projection operator as initialization and achieves exact recovery in the noiseless setting. We analyse the convergence and non-asymptotic error bounds of MOP-UP, introducing a blockwise matrix eigenvalue perturbation bound that proves the desired bound, where classic perturbation bounds fail. The effectiveness and practical merits of the proposed framework are demonstrated through experiments on both simulated and real datasets. Lastly, we discuss generalizations of our approach to higher-order data.

Study of B(s) meson decays to D0*(2300) , Ds0*(2317) , Ds1(2460) and Ds1(2536) within the covariant light-front approach

In this work, we investigate the form factors of the transitions B(s)→D0*(2300),Ds0*(2317), Ds1(2460), and Ds1(2536) in the covariant light-front quark model (CLFQM), where these final states are considered as P-wave excited charmed mesons. In order to obtain the form factors for the physical transition processes, we extend these form factors from the spacelike region to the timelike region. The q2-dependence for each transition form factor is also plotted. Then, combined with those form factors, the branching ratios of the two-body nonleptonic decays B(s)→D(s)0*(2300,2317)M, Ds1(2460,2536)M with M being a light pseudoscalar (vector) meson or a charmed meson are calculated by considering the quantum chromodynamics (QCD) radiative corrections to the hadronic matrix elements within the QCD factorization approach. Most of our predictions are comparable to the results given by other theoretical approaches and the present available data. Published by the American Physical Society 2024

Probing the internal structures of pΩ and ΩΩ with their production at the Large Hadron Collider

The strange dibaryons pΩ (S25) and ΩΩ (S01) are likely bound, existing either in molecular states like the deuteron or as more exotic compact six-quark states. Here, we investigate the production of these two dibaryons in Pb+Pb collisions at sNN=2.76 TeV at the CERN Large Hadron Collider within a covariant coalescence model, which employs a blast-wave-like parametrization for the phase-space configurations of constituent particles at freeze-out. For the molecular states, the pΩ and ΩΩ are produced via p−Ω and Ω−Ω coalescence, respectively, while for the six-quark states, they are formed through uudsss and ssssss coalescence. We find that the yield ratios NpΩ/NΩ and NΩΩ/NΩ have a distinct centrality dependence between the molecular and multiquark states, thus offering a promising way for distinguishing the two states. Our results suggest that the measurements of pΩ and ΩΩ production in relativistic heavy-ion collisions can shed light on their internal structures. Published by the American Physical Society 2024

A flexible and interpretable spatial covariance model for data on graphs

A flexible and interpretable spatial covariance model for data on graphs

Covariate-dependent spatio-temporal covariance models

Geostatistical regression models are widely used in environmental and geophysical sciences to characterize the mean and dependence structures for spatio-temporal data. Traditionally, these models account for covariates solely in the mean structure, neglecting their potential impact on the spatio-temporal covariance structure. This paper addresses a significant gap in the literature by proposing a novel covariate-dependent covariance model within the spatio-temporal random-effects model framework. Our approach integrates covariates into the covariance function through a Cholesky-type decomposition, ensuring compliance with the positive-definite condition. We employ maximum likelihood for parameter estimation, complemented by an efficient expectation conditional maximization algorithm. Simulation studies demonstrate the superior performance of our method compared to conventional techniques that ignore covariates in spatial covariances. We further apply our model to a PM2.5 dataset from Taiwan, highlighting wind speed’s pivotal role in influencing the spatio-temporal covariance structure. Additionally, we incorporate wind speed and sunshine duration into the covariance function for analyzing Taiwan ozone data, revealing a more intricate relationship between covariance and these meteorological variables.

Protocol for a randomised trial of a self-directed digital pain management intervention (Empowered Relief) tailored to adults with chronic pain and prescription opioid misuse/disorder: the MOBILE Relief study

IntroductionChronic pain increases the risk of prescription opioid misuse or opioid use disorder (OUD). Non-pharmacological treatments are needed to dually address pain and opioid risks. The purpose of the Mobile...

Association of caffeine consumption with cerebrospinal fluid biomarkers in mild cognitive impairment and Alzheimer's disease: A BALTAZAR cohort study.

We investigated the link between habitual caffeine intake with memory impairments and cerebrospinal fluid (CSF) biomarkers in mild cognitive impairment (MCI) and Alzheimer's disease (AD) patients. MCI (N=147) and AD (N=116) patients of the Biomarker of AmyLoid pepTide and AlZheimer's diseAse Risk (BALTAZAR) cohort reported their caffeine intake at inclusion using a dedicated survey. Associations of caffeine consumption with memory impairments and CSF biomarkers (tau, p-tau181, amyloid beta 1-42 [Aβ1-42], Aβ1-40) were analyzed using logistic and analysis of covariance models. Adjusted on Apolipoprotein E (APOE ε4), age, sex, education level, and tobacco, lower caffeine consumption was associated with higher risk to be amnestic (OR: 2.49 [95% CI: 1.13 to 5.46]; p=0.023) and lower CSF Aβ1-42 (p=0.047), Aβ1-42/Aβ1-40 (p=0.040), and Aβ1-42/p-tau181 (p=0.020) in the whole cohort. Data support the beneficial effect of caffeine consumption to memory impairments and CSF amyloid markers in MCI and AD patients. We studied the impact of caffeine consumption in the BALTAZAR cohort. Low caffeine intake is associated with higher risk of being amnestic in MCI/AD patients. Caffeine intake is associated with CSF biomarkers in AD patients.

Randomized Controlled Trial of a Nurse-Led Brief Behavioral Intervention for Dyspnea in Patients With Advanced Lung Cancer.

In patients with lung cancer, dyspnea is one of the most prevalent and disabling symptoms, for which effective treatments are lacking. We examined the efficacy of a nurse-led brief behavioral intervention to improve dyspnea in patients with advanced lung cancer. Patients with advanced lung cancer reporting at least moderate breathlessness (n = 247) were enrolled in a randomized trial of a nurse-led two-session intervention (focused on breathing techniques, postural positions, and fan therapy) versus usual care. At baseline and weeks 8 (primary end point), 16, and 24, participants completed measures of dyspnea (Modified Medical Research Council Dyspnea Scale [mMRCDS]; Cancer Dyspnoea Scale [CDS]), quality of life (Functional Assessment of Cancer Therapy-Lung [FACT-L]), psychological symptoms (Hospital Anxiety and Depression Scale), and activity level (Godin-Shephard Leisure Time Physical Activity Questionnaire). To examine intervention effects, we conducted analysis of covariance and longitudinal mixed effects models. The sample (Agemean = 66.15 years; 55.9% female) primarily included patients with advanced non-small cell lung cancer (85.4%). Compared with usual care, the intervention improved the primary outcome of patient-reported dyspnea on the mMRCDS (difference = -0.33 [95% CI, -0.61 to -0.05]) but not the CDS total score at 8 weeks. Intervention patients also reported less dyspnea on the CDS sense of discomfort subscale (difference = -0.59 [95% CI, -1.16 to -0.01]) and better functional well-being per the FACT-L (difference = 1.39 [95% CI, 0.18 to 2.59]) versus the control group. Study groups did not differ in overall quality of life, psychological symptoms, or activity level at 8 weeks or longitudinally over 24 weeks. For patients with advanced lung cancer, a scalable behavioral intervention alleviated the intractable symptom of dyspnea. Further research is needed on ways to enhance intervention effects over the long-term and across additional outcomes.

A Sparse Approximate Factor Model for High-Dimensional Covariance Matrix Estimation and Portfolio Selection

Abstract We propose a novel estimation approach for the covariance matrix based on the l1-regularized approximate factor model (AFM). Our sparse approximate factor (SAF) covariance estimator allows for the existence of weak factors and hence relaxes the pervasiveness assumption generally adopted for the standard AFM. We prove the consistency of the covariance matrix estimator under the Frobenius norm as well as the consistency of the factor loadings and the factors. Our Monte Carlo simulations reveal that the SAF covariance estimator has superior properties in finite samples for low and high dimensions and different designs of the covariance matrix. Moreover, in an out-of-sample portfolio forecasting application, the estimator uniformly outperforms alternative portfolio strategies based on alternative covariance estimation approaches and modeling strategies including the 1/N-strategy.

Long-term grassland diversity-productivity relationship regulated by management regimes in northern China

Grasslands are the most extensively distributed terrestrial ecosystems on Earth, providing a range of ecosystem services that are vital for sustaining human life and critical for sustainable development at the global scale. However, the relationship between the two most important attributes of grassland, plant diversity, and productivity, remains controversial even after many years of research. Here, we develop an analysis of covariance (ANCOVA) model based on decadal-scale experimental data from a degraded meadow steppe in northeastern Inner Mongolia, China to quantify the response of aboveground biomass (AGB) to plant species diversity under varying management regimes. We report that AGB responds negatively to the plant diversity in fallow grasslands and positively in grazing grasslands, transiting from negative to positive in mowing grasslands as mowing became more frequent. We show that the changing diversity-productivity relationships are driven by changes in species composition of the plant community, given the significant productivity gap between rare and non-rare species. This highlights the role of management in regulating the diversity-productivity relationships in grasslands. These results not only provide provocative insights into the relationships between plant diversity and productivity but also support more sustainable use and management of grassland resources.

The Effects of Volume-Matched 1- and 2-Day Repeated Backward Sprint Training Formats on Physical Performance in Youth Male Soccer Players.

Bouguezzi, R, Negra, Y, Sammoud, S, Uthoff, A, Moran, J, Behrens, M, and Chaabene, H. The effects of volume-matched 1- and 2-day repeated backward sprint training formats on physical performance in youth male soccer players. J Strength Cond Res XX(X): 000-000, 2024-This study examined the effects of a 6-week, 1 vs. 2 sessions of volume-matched weekly repeated backward sprint training (RBST) on measures of physical fitness in youth male soccer players. Thirty male youth soccer players from a regional soccer team were randomly assigned to a 2-day group (n = 15; age = 16.40 ± 0.64 years; Maturity-offset = 2.19 ± 0.65 years) or a 1-day group (n = 15; age = 16.27 ± 0.51 years; Maturity-offset = 1.91 ± 0.40 years). Measures of jumping ability, linear sprint speed, change-of-direction (CoD) speed, aerobic endurance (AE), and repeated sprint ability (RSA) were measured before and after 6 weeks of training. The training interventions involved 1 to 2 sets, each comprising 7 repetitions over a 20-m distance of RBST in the 2-day group, and 2 to 4 sets, also with 7 repetitions each over the same distance, in the 1-day group. Statistical analyses were conducted using the analysis of covariance model with baseline measurements entered as covariates. Results indicated that RBST over 2 days generated greater benefits compared with a single day for improving CoD speed (∆4.91 vs. 0.04%; effect size [d] = 0.78 vs. 0.00, respectively), linear sprint speed (10-m: ∆4.74 vs. 0.36%; d = 0.96 vs. 0.00, respectively; 20 m: ∆3.34 vs. 0.82%; d = 1.00 vs. 0.49, respectively), and RSA performances (RSAbest: ∆2.61 vs. 0.29%; d = 0.71 vs. 0.16, respectively; RSAmean: ∆2.86 vs. 0.89%; d = 0.59 vs. 0.16, respectively; RSAtotal: ∆2.86 vs. 0.89%; d = 0.62 vs. 0.41, respectively). However, similar improvements in the 2 RBST formats were observed on jumping (2-day group: ∆6.9%; d = 0.75; 1-day group: ∆10.6%, d = 1.26) and AE performance (2-day group: ∆17.24%, d = 1.04; 1-day group: ∆27.25%, d = 2.25). In summary, the findings suggest that, when volume is matched, spreading the RBST regimen over 2 days may result in greater improvements compared with a single-day approach for enhancing CoD speed, linear sprint speed, and RSA performance.

Numerical analysis of hydrogen storage in lined rock cavern under high pressure and its implications to hydrogen embrittlement

Underground hydrogen storage could provide buffer capacity to store the excess energy from renewable resources. A lined rock cavern (LRC) is one of the hydrogen geologic storage site options. It offers great flexibility because it does not rely on the existence of salt caverns, aquifer formations or depleted hydrocarbon reservoir fields. However, many important aspects have to be investigated before its full-scale implementation, among them (1) the effect of spatial variation of rock mass properties and (2) hydrogen embrittlement on the steel lining. We present a numerical study on these two aspects by simulating field tests at a pilot LRC project. For the effect of rock heterogeneity, we assign a spatial variation of elastic properties based on the Gaussian covariance model. The results show a good agreement with the field measurements. The spatial correlation length of rock mass elastic properties only has a minor impact on the cavern surface deformation. Regarding hydrogen embrittlement, our simulations predict a significant hydrogen diffusion in the low alloy steel (S355) lining after ∼2 months of operation. The fracture simulations show little fracture propagation even after 10 years of contact with hydrogen gas. However, the product loss may be too high for the low alloy steel (S355). We would recommend the use of stainless steel to inhibit the hydrogen diffusion. We show that advanced numerical models are powerful tools to ensure the safety of hydrogen storage in LRCs.

Phase III Randomized, Placebo-Controlled Clinical Trial of Donepezil for Treatment of Cognitive Impairment in Breast Cancer Survivors After Adjuvant Chemotherapy (WF-97116).

To test efficacy of donepezil, a cognitive enhancer, to improve memory in breast cancer survivors who report cancer-related cognitive impairment 1-5 years postchemotherapy. Adult female BCS exposed to ≥4 cycles of adjuvant chemotherapy 1-5 years before enrollment who reported cancer-related cognitive impairment were eligible. Participants, enrolled at sites affiliated with the Wake Forest NCI Community Oncology Research Program (NCORP) Research Base, were randomly assigned to receive 5 mg of donepezil once daily for 6 weeks titrated to 10 mg once daily for 18 weeks or placebo. Cognition and self-report cognitive functioning was assessed at baseline, 12, 24 (end of intervention), and 36 (washout) weeks postrandomization. Mixed-effects repeated measures analysis of covariance models were used to assess treatment differences in immediate recall (primary outcome) on the Hopkins Verbal Learning Test-Revised (HVLT-R) and other cognitive domains (secondary outcomes) with covariates of treatment, time, time by treatment interaction, baseline outcome level, age stratification, and an unstructured covariance matrix to account for within participant correlation over time. Two hundred seventy-six BCS from 87 NCORP practices (mean age, 57.1, standard deviation [SD], 10.5) who were at a mean of 29.6 months (SD, 14.2) postchemotherapy were randomly assigned to donepezil (n = 140) or placebo (n = 136). At 24 weeks, treatment groups did not differ on HVLT-R scores (donepezil mean = 25.98, placebo = 26.50, P = .32). There were no statistically significant differences between treatments at 12, 24, or 36 weeks for attention, executive function, verbal fluency, processing speed, or self-reported cognitive functioning. Endocrine therapy and menopausal status did not affect results. BCS 1-5 years after completing chemotherapy with documented memory problems, randomly assigned to 24 weeks of 5-10 mg of donepezil once daily, did not perform differently at the end of treatment on tests of memory, other cognitive functions, or subjective functioning than those randomly assigned to placebo.

Attentional Bias, Pupillometry, and Spontaneous Blink Rate: Eye Characteristic Assessment Within a Translatable Nicotine Cue Virtual Reality Paradigm.

Incentive salience processes are important for the development and maintenance of addiction. Eye characteristics such as gaze fixation time, pupil diameter, and spontaneous eyeblink rate (EBR) are theorized to reflect incentive salience and may serve as useful biomarkers. However, conventional cue exposure paradigms have limitations that may impede accurate assessment of these markers. This study sought to evaluate the validity of these eye-tracking metrics as indicators of incentive salience within a virtual reality (VR) environment replicating real-world situations of nicotine and tobacco product (NTP) use. NTP users from the community were recruited and grouped by NTP use patterns: nondaily (n=33) and daily (n=75) use. Participants underwent the NTP cue VR paradigm and completed measures of nicotine craving, NTP use history, and VR-related assessments. Eye-gaze fixation time (attentional bias) and pupillometry in response to NTP versus control cues and EBR during the active and neutral VR scenes were recorded and analyzed using ANOVA and analysis of covariance models. Greater subjective craving, as measured by the Tobacco Craving Questionnaire-Short Form, following active versus neutral scenes was observed (F1,106=47.95; P<.001). Greater mean eye-gaze fixation time (F1,106=48.34; P<.001) and pupil diameter (F1,102=5.99; P=.02) in response to NTP versus control cues were also detected. Evidence of NTP use group effects was observed in fixation time and pupillometry analyses, as well as correlations between these metrics, NTP use history, and nicotine craving. No significant associations were observed with EBR. This study provides additional evidence for attentional bias, as measured via eye-gaze fixation time, and pupillometry as useful biomarkers of incentive salience, and partially supports theories suggesting that incentive salience diminishes as nicotine dependence severity increases.

Neural Networks for Geospatial Data

Analysis of geospatial data has traditionally been model-based, with a mean model, customarily specified as a linear regression on the covariates, and a Gaussian process covariance model, encoding the spatial dependence. While nonlinear machine learning algorithms like neural networks are increasingly being used for spatial analysis, current approaches depart from the model-based setup and cannot explicitly incorporate spatial covariance. We propose NN-GLS, embedding neural networks directly within the traditional Gaussian process (GP) geostatistical model to accommodate nonlinear mean functions while retaining all other advantages of GP, like explicit modeling of the spatial covariance and predicting at new locations via kriging. In NN-GLS, estimation of the neural network parameters for the nonlinear mean of the Gaussian Process explicitly accounts for the spatial covariance through use of the generalized least squares (GLS) loss, thus, extending the linear case. We show that NN-GLS admits a representation as a special type of graph neural network (GNN). This connection facilitates the use of standard neural network computational techniques for irregular geospatial data, enabling novel and scalable mini-batching, backpropagation, and kriging schemes. We provide methodology to obtain uncertainty bounds for estimation and predictions from NN-GLS. Theoretically, we show that NN-GLS will be consistent for irregularly observed spatially correlated data processes. We also provide a finite sample concentration rate, which quantifies the need to accurately model the spatial covariance in neural networks for dependent data. To our knowledge, these are the first large-sample results for any neural network algorithm for irregular spatial data. We demonstrate the methodology through numerous simulations and an application to air pollution modeling. We develop a software implementation of NN-GLS in the Python package geospaNN. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Simultaneous enhancement of multiple functional properties using evolution-informed protein design

A major challenge in protein design is to augment existing functional proteins with multiple property enhancements. Altering several properties likely necessitates numerous primary sequence changes, and novel methods are needed to accurately predict combinations of mutations that maintain or enhance function. Models of sequence co-variation (e.g., EVcouplings), which leverage extensive information about various protein properties and activities from homologous protein sequences, have proven effective for many applications including structure determination and mutation effect prediction. We apply EVcouplings to computationally design variants of the model protein TEM-1 β-lactamase. Nearly all the 14 experimentally characterized designs were functional, including one with 84 mutations from the nearest natural homolog. The designs also had large increases in thermostability, increased activity on multiple substrates, and nearly identical structure to the wild type enzyme. This study highlights the efficacy of evolutionary models in guiding large sequence alterations to generate functional diversity for protein design applications.

Tests under simple order in one‐way ANCOVA

AbstractAnalysis of covariance (ANCOVA) models are used when apart from the main treatments, some covariates also affect the response variable. In this article, the problem of testing the homogeneity of treatment effects against ordered alternatives is addressed in a fixed effects one‐way ANCOVA model when error variances are heterogeneous. The likelihood ratio test (LRT) and two approximate tests based on the asymptotic distributions of some union‐intersection type test statistics are proposed. A parametric bootstrap technique has been used to implement the LRT and its asymptotic validity is proved. A method to construct simultaneous confidence intervals is proposed. All the test procedures are further extended to the case of more than one covariate. The robustness of tests is also studied under departure from normality. Extensive simulation studies show that the proposed test procedures perform well in terms of achieving the nominal size value and good power values.

A Global Paleosecular Variation Database for the Paleogene: Stationary Secular Variation Behavior Since the Triassic?

AbstractPaleosecular variation analysis is a primary tool for characterizing ancient geomagnetic behavior and its evolution through time. This study presents a new high‐quality directional data set, paleosecular variation of the Paleogene (PSVP), with and without correction for serial correlation, compiled from 1,667 sites from 45 different localities from the Paleogene and late Cretaceous (84–23 Ma). The data set is used to study the variability, structure, and latitude dependence of the geomagnetic field during that period by varying selection criteria and PSV models. Modeled values for the equatorial virtual geomagnetic pole (VGP) dispersion have over‐lapping uncertainty intervals within their uncertainty bounds between 8.3° and 18.6° for the past 250 Ma. We investigate the suitability of two descriptive models of PSV, Model G‐style quadratic fits and covariant Giant Gaussian Process models, and find that both styles of model fail to satisfactorily reproduce the latitude dependent morphology of PSV, but suggest that estimates of the equatorial VGP dispersion may still robustly characterize aspects of Earth's long‐term field morphology. During this time where the PSV behavior has not changed substantially, the reversal frequency has varied widely. The lack of a clear relationship between PSV behavior and reversal frequency is not trivially explained in the context of published findings regarding numerical geodynamo simulations.

A randomized, double-blind, study to evaluate the efficacy, pharmacodynamics, safety and immunogenicity of FKS518 proposed biosimilar to denosumab with the originator in postmenopausal women with osteoporosis (LUMIADE-3 study).

3155 Background: Biosimilars can increase patient access to treatment. FKS518 is a candidate biosimilar of denosumab, a RANKL inhibitor used in patients with bone metastases from certain solid tumors and in multiple myeloma. To establish biosimilarity, comparative studies are conducted in the most sensitive patient population to detect if there are clinically meaningful differences. LUMIADE-3 was designed to assess the safety and efficacy of FKS518 compared to the originator in postmenopausal women with osteoporosis (PMO). Methods: A randomized, double-blind, multicenter, 2-arm study recruited female patients aged 55 to 85 years with confirmed postmenopausal status and lumbar spine bone mineral density (LS-BMD) T-score ≤ -2.5 and ≥ -4.0, as measured by central dual-energy X-ray absorptiometry (DXA) assessment. Subjects with prior osteoporosis treatment that could add risk for cumulative effect or affect the interpretation of results were excluded. Patients were randomized and received three 60 mg administrations. At week 52, patients receiving denosumab originator were re-randomized to continue their treatment or switch to FKS518 for the third dose. Those receiving FKS518 continued receiving FKS518. The primary efficacy endpoint was the percent change from baseline in LS-BMD DXA at week 52. An analysis of the covariance model was used to compare the two treatments and two separate one-sided tests at alpha=0.05 were performed to assess equivalence. Secondary efficacy endpoints included the percent change from baseline in BMD at the femoral neck and total hip. Results: A total of 553 patients were randomized to FKS518 (n= 276) or the originator (n=277). Clinically relevant increases in LS-BMD were evident at week 52 in both the FKS518 and originator product groups. The results demonstrated therapeutic equivalence: the lower bound of the 90% CI for non-inferiority (-0.05) was above -1.45, and the upper bound of the 90% CI for non-superiority (1.20) was below 1.45. All secondary objectives were met. At week 52, a similar percent change from baseline in BMD at the femoral neck and total hip was observed between the two groups. 374 patients (67.6%) experienced treatment-emergent adverse events: 185 (66.8%) in the FKS518 group and 189 (68.5%) in the originator product group. Safety evaluation did not point to notable differences between FKS518 and the originator product groups. Conclusions: We demonstrated the therapeutic equivalence of FKS518 and originator product. The safety data showed similar safety profiles for the FKS518 and the originator product groups. Results from this study add to the totality of evidence supporting the similarity of FKS518 as a proposed biosimilar to denosumab originator product, a RANKL inhibitor used in patients with bone metastases from solid tumors. Clinical trial information: NCT04934072 .

Inhaled fentanyl aerosol through a drug-device combination for breakthrough cancer pain: A multicenter, randomized, double-blind, placebo-controlled trial.

e24058 Background: Breakthrough cancer pain (BTcP) can significantly impact the life quality of the cancer patient. Wide-used rapid-onset opioids (ROOs) for BTcP are transmucosal fentanyl-based preparations. Inhaled fentanyl demonstrated similar pharmacokinetic properties as intravenous administration in healthy volunteers. Fentanyl aerosol can be generated via hand-held drug-device delivery system, providing rapid, consistent multi-dose administration. Fentanyl inhalant was explored as alternative treatment of BTcP in cancer patients in this study to evaluate its efficacy, safety, and tolerability in BTcP patients with well-controlled background cancer pain. Methods: Adults ≥18 years with cancer are eligible if they were experiencing persistent cancer- or treatment-related pain with above mild severity and 1-4 episode occurrences daily. Each patient was treated and observed for 6 BTcP episodes (4 with fentanyl inhalant, 2 with placebo). During each episode, patients were allowed to inhale up to six shots within 1 hour. Pain intensity was recorded using 11-point Numerical Rating Scale (NRS) at predefined time points. The primary endpoint was the time-weighted Sum of Pain Intensity Difference scores at 30 minutes (SPID30) after the first inhalation. Results: 335 BTcP episodes in 59 patients were treated. SPID30 was analyzed using mixed model of repeated measures with treatment (fentanyl inhalant or placebo), BTcP episode, baseline pain intensity score as fixed factors, and subject as random factor. Mean SPID30 ± SD was -97.4 ± 48.43 for fentanyl-treated episodes, and -64.6 ± 40.25 for placebo-treated episodes. Based on treatment adjustment in the mixed covariance model, significant difference (p &lt; 0.001) was reported with mean treatment difference of -27.7 (LS-mean ± SD, -94.2 ± 4.49 for fentanyl and -66.5 ± 4.86 for placebo). It enlarged at 60 minutes (SPID60) with a mean treatment difference of -57.9 (LS-mean ± SD, -231.4 ±105.14 for fentanyl and -162.2 ± 91.21 for placebo, p &lt; 0.001). Significant differences in PID for episodes treated with fentanyl versus placebo were seen 4 minutes after the first dosing and maintained up to 60 minutes. Other endpoints showed statistical significance favoring fentanyl treatment, including the proportion of NRS ≤ 3, NRS pain scores decreased ≥ 33% or ≥ 50% from baseline after completion of BTcP treatment within 30 minutes of first dose, and Pain Relief Score Within 30 Minutes of the First Dose (PR30). Only 4.4% of BTcP episodes required rescue medication in fentanyl group. Major adverse events were mild or moderate in severity and were typical for opioid drugs. Conclusions: Fentanyl inhalant is safe, effective, and well-tolerated for managing BTcP. It provides a non-invasive route for systemic drug delivery and can be self-administered or given by family caregivers as personalized and regular pain management. Clinical trial information: NCT05531422 .