Deep Learning-based Modeling for Preclinical Drug Safety Assessment.

In drug development, assessing the toxicity of candidate compounds is crucial for successfully transitioning from preclinical research to early-stage clinical trials. Drug safety is typically assessed using animal models with a manual histopathological examination of tissue sections to characterize the dose-response relationship of the compound - a time-intensive process prone to inter-observer variability and predominantly involving tedious review of cases without abnormalities. Artificial intelligence (AI) methods in pathology hold promise to accelerate this assessment and enhance reproducibility and objectivity. Here, we introduce TRACE, a model designed for toxicologic liver histopathology assessment capable of tackling a range of diagnostic tasks across multiple scales, including situations where labeled data is limited. TRACE was trained on 15 million histopathology images extracted from 46,734 digitized tissue sections from 157 preclinical studies conducted on Rattus norvegicus. We show that TRACE can perform various downstream toxicology tasks spanning histopathological response assessment, lesion severity scoring, morphological retrieval, and automatic dose-response characterization. In an independent reader study, TRACE was evaluated alongside ten board-certified veterinary pathologists and achieved higher concordance with the consensus opinion than the average of the pathologists. Our study represents a substantial leap over existing computational models in toxicology by offering the first framework for accelerating and automating toxicological pathology assessment, promoting significant progress with faster, more consistent, and reliable diagnostic processes.

AI-driven Discovery of Morphomolecular Signatures in Toxicology.

Early identification of drug toxicity is essential yet challenging in drug development. At the preclinical stage, toxicity is assessed with histopathological examination of tissue sections from animal models to detect morphological lesions. To complement this analysis, toxicogenomics is increasingly employed to understand the mechanism of action of the compound and ultimately identify lesion-specific safety biomarkers for which in vitro assays can be designed. However, existing works that aim to identify morphological correlates of expression changes rely on qualitative or semi-quantitative morphological characterization and remain limited in scale or morphological diversity. Artificial intelligence (AI) offers a promising approach for quantitatively modeling this relationship at an unprecedented scale. Here, we introduce GEESE, an AI model designed to impute morphomolecular signatures in toxicology data. Our model was trained to predict 1,536 gene targets on a cohort of 8,231 hematoxylin and eosin-stained liver sections from Rattus norvegicus across 127 preclinical toxicity studies. The model, evaluated on 2,002 tissue sections from 29 held-out studies, can yield pseudo-spatially resolved gene expression maps, which we correlate with six key drug-induced liver injuries (DILI). From the resulting 25 million lesion-expression pairs, we established quantitative relations between up and downregulated genes and lesions. Validation of these signatures against toxicogenomic databases, pathway enrichment analyses, and human hepatocyte cell lines asserted their relevance. Overall, our study introduces new methods for characterizing toxicity at an unprecedented scale and granularity, paving the way for AI-driven discovery of toxicity biomarkers.

FGFR Alterations in Thyroid Carcinoma: A Novel Class of Primary Drivers with Significant Therapeutic Implications and Secondary Molecular Events Potentially Mediating Resistance in Thyroid Malignancy.

Background: Diagnostic classification of thyroid malignancy is primarily accomplished through examination of histomorphological features and may be substantiated and clarified by molecular data. Individual molecular drivers show relatively robust and specific associations with histological subtypes of thyroid malignancy, including BRAF sequence variants and kinase gene fusions in papillary thyroid carcinoma, predominantly RAS variants in follicular-patterned neoplasia, and additional "late" mutations affecting TERT promoter, TP53, and the PI3K/AKT/PTEN pathway in high-grade malignancies. Given the oncogenic role of FGFR, particularly FGFR1-3, the goal of this study was to explore the role of FGFR in thyroid carcinoma biology. Methods: We completed a multicenter retrospective observational study for thyroid carcinomas with pathogenic alterations in the FGFR gene family. We performed this study by querying the molecular data accumulated for thyroid carcinomas from each center. Results: Overall, 5030 sequenced thyroid malignancies were reviewed, yielding 17 tumors with FGFR alterations, including 11 where FGFR was the primary molecular driver and 6 where FGFR was a secondary pathogenic alteration, with a subset for which there was available clinical follow-up data. Of the 11 carcinomas with an FGFR driver, 9 were gene fusions involving FGFR2:VCL (4 tumors), TG::FGFR1 (3 tumors), FGFR2::CIT, and FGFR2::SHTN1, and the remaining 2 were driven by FGFR1 amplification. In the 6 tumors where a canonical driver of thyroid neoplasia was present (5 cases) or no clear primary driver was detected (1 case), sequencing detected secondary FGFR2 p.W290C, p.Y375C, and p.N549K, as well as FGFR1 p.N546K in the respective tyrosine kinase domains, some at subclonal variant allele frequencies. Conclusions: This study presents the first description of a collection of thyroid carcinomas grouped by primary driver alterations in FGFR, as well as a cohort of thyroid tumors with secondary alterations that potentially lead to tumor progression or resistance to targeted therapy. Given the availability of small molecular inhibitors targeting oncogenic FGFR, this study emphasizes the significant implications for patients from identification of FGFR alterations as they are currently under-recognized in the literature and, most importantly, have potential novel treatment options.

Young People's Barriers and Facilitators of Engagement with Web-Based Mental Health Interventions for Anxiety and Depression: A Qualitative Study.

BACKGROUND: The prevalence of anxiety and depressive symptoms in young people have increased in many countries around the world. Web-based mental health interventions (or W-MHIs) have the potential to reduce anxiety and depression symptoms for young people. Although W-MHIs have become more widely used by young people since the coronavirus disease 2019 (COVID-19) pandemic, real-world engagement in these W-MHIs has remained low compared with engagement reported in research studies. Moreover, there are limited studies examining factors influencing engagement with W-MHIs in the post-COVID-19 pandemic years. OBJECTIVE: This study aims to explore barriers and facilitators of engagement with W-MHIs for anxiety and depression among young people. METHOD: Seventeen semi-structured interviews and one focus group with three participants were conducted online via Zoom between February and March 2023. Participants were young people aged 18-25 years who had self-reported experience of anxiety and/or depression in the past 6 months, lived in Australia, and considered using W-MHIs to manage their anxiety and/or depression symptoms. Inductive thematic analysis was performed to understand the key barriers and facilitators of young people's engagement with W-MHIs. RESULTS: Both individual- and intervention-related factors influenced young people's engagement with W-MHIs. Facilitators of engagement included personal trust and beliefs in W-MHIs, ability to contact a health professional, programme suitability (e.g., affordability, content aligning with user needs), programme usability (e.g., user interface), and accessibility of the online platform. Barriers included concerns about online security, lack of human interaction and immediate responses from health professionals (if any), and negative experience with mental health programmes. Participants expressed greater willingness to pay if they could contact health professionals during the programme. CONCLUSION: Better promotion strategies for mental health and W-MHI awareness are needed to increase the perceived importance and priority of mental health interventions among young people. Young people should be involved in the W-MHI co-design to enhance the programme suitability and usability for young people, fostering their engagement with W-MHIs.

Purification of Spherical Graphite as Anode for Li-Ion Battery: A Comparative Study on the Purifying Approaches.

Graphite is a versatile material used in various fields, particularly in the power source manufacturing industry. Nowadays, graphite holds a unique position in materials for anode electrodes in lithium-ion batteries. With a carbon content of over 99% being a requirement for graphite to serve as an electrode material, the graphite refinement process plays a pivotal role in the research and development of anode materials for lithium-ion batteries. This study used three different processes to purify spherical graphite through wet chemical methods. The spherical graphite after the purification processes was analysed for carbon content by using energy-dispersive X-ray (EDX) spectroscopy and was evaluated for structural and morphological characteristics through X-ray diffraction (XRD), scanning electron microscopy (SEM), and Brunauer-Emmett-Teller (BET) analyses. The analyses results indicate that the three-step process via H2SO4-NaOH-HCl cleaning can elevate the carbon content from 90% to above 99.9% while still maintaining the graphite structure and spherical morphology, thus enhancing the surface area of the material for anode application. Furthermore, the spherical graphite was studied for electrochemical properties when used as an anode for Li-ion batteries using cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD) measurements. The results demonstrated that the purification process significantly improves the material's capacity with a specific capacity of 350 mAh/g compared to the 280 mAh/g capacity of the anode made of spherical graphite without purification.

The adaptation of 3T3 cells to simulated microgravity by retrieving the major cell cycle-related protein expression during long-term in vitro proliferation.

The present study aimed to assess the effects of simulated microgravity (SMG) on 3T3 cell proliferation and the expression of cell cycle regulators. 3T3 cells were induced to SMG by Gravite® for 8 days, while the control group was treated with 1G condition. The result showed that the SMG condition causes a decrease in proliferative activity in 3T3 cells. In the SMG group, the expression of cell cycle-related proteins was lower than the control on day 3. However, these proteins were upregulated in 3T3 cells of the SMG group on day 5, suggesting that these cells were rescued from the arrest and retrieved a higher proliferation. A down-regulation of cell cycle-related proteins was observed in 3T3 cells of both SMG and control groups on day 7. In conclusion, SMG results in the attenuation of cell proliferation during the initial exposure to SMG, but the cells will adapt to this condition and retrieve normal proliferation by increasing the expression of cell cycle regulators.

The Modeled Cost-Effectiveness of a Prevention Program Targeting Both Eating Disorders and High BMI.

OBJECTIVE: Prevention programs for eating disorders (EDs) and high body index mass (BMI) have the potential to reduce the onset of these interconnected public health concerns. However, it remains unclear whether routine implementation of such programs would be cost-effective. This study aims to determine the cost-effectiveness of an intervention that aims to prevent both ED and high BMI. METHOD: A Markov model was developed to evaluate the incremental cost-effectiveness of a targeted school-based program, Healthy Weight, that aims to prevent both EDs and high BMI among Australian adolescents with body image concerns (aged 15-18 years), versus a "no intervention" comparator. A cost-utility analysis was conducted from a "healthcare and education" sector perspective with costs (measured in 2019 Australian dollars) and health impacts modeled over the lifetime of the target population. An incremental cost-effectiveness ratio (ICER), expressed as cost per health-adjusted life year (HALY) gained, was calculated. Sensitivity analyses were done to test model assumptions. RESULTS: The mean intervention cost and HALYs gained were AUD$2.13 million (95% CI, AUD$1.83-2.43 million) and 146 (95% CI, 90-209), respectively. With healthcare cost-savings (AUD$3.97 million) included, the intervention was predicted to be cost-saving (AUD$1.83 million; 95% CI, AUD$0.51-3.21 million). Primary findings were robust to extensive sensitivity analyses. DISCUSSION: The Healthy Weight intervention is likely to represent good value-for-money. To ensure the successful implementation of this program at the population level, further research on its feasibility and acceptability among schools and the wider community is required.

ATM and 53BP1 regulate alternative end joining-mediated V(D)J recombination.

G0-G1 phase alternative end joining (A-EJ) is a recently defined mutagenic pathway characterized by resected deletion and translocation joints that are predominantly direct and are distinguished from A-EJ in cycling cells that rely much more on microhomology-mediated end joining (MMEJ). Using chemical and genetic approaches, we systematically evaluate potential A-EJ factors and DNA damage response (DDR) genes to support this mechanism by mapping the repair fates of RAG1/2-initiated double-strand breaks in the context of Igκ locus V-J recombination and chromosome translocation. Our findings highlight a polymerase theta-independent Parp1-XRCC1/LigIII axis as central A-EJ components, supported by 53BP1 in the context of an Ataxia-telangiectasia mutated (ATM)-activated DDR. Mechanistically, we demonstrate varied changes in short-range resection, MMEJ, and translocation, imposed by compromising specific DDR activities, which include polymerase alpha, Ataxia-telangiectasia and Rad3-related (ATR), DNA2, and Mre11. This study advances our understanding of DNA damage repair within the 53BP1 regulatory domain and the RAG1/2 postcleavage complex.

A multimodal generative AI copilot for human pathology.

Computational pathology1,2 has witnessed considerable progress in the development of both task-specific predictive models and task-agnostic self-supervised vision encoders3,4. However, despite the explosive growth of generative artificial intelligence (AI), there have been few studies on building general-purpose multimodal AI assistants and copilots5 tailored to pathology. Here we present PathChat, a vision-language generalist AI assistant for human pathology. We built PathChat by adapting a foundational vision encoder for pathology, combining it with a pretrained large language model and fine-tuning the whole system on over 456,000 diverse visual-language instructions consisting of 999,202 question and answer turns. We compare PathChat with several multimodal vision-language AI assistants and GPT-4V, which powers the commercially available multimodal general-purpose AI assistant ChatGPT-4 (ref. 6). PathChat achieved state-of-the-art performance on multiple-choice diagnostic questions from cases with diverse tissue origins and disease models. Furthermore, using open-ended questions and human expert evaluation, we found that overall PathChat produced more accurate and pathologist-preferable responses to diverse queries related to pathology. As an interactive vision-language AI copilot that can flexibly handle both visual and natural language inputs, PathChat may potentially find impactful applications in pathology education, research and human-in-the-loop clinical decision-making.

Temperature dependence of the dielectric function and critical points of monolayer WSe2.

Monolayer materials typically display intriguing temperature-dependent dielectric and optical properties, which are crucial for improving the structure and functionality of associated devices. Due to its unique photoelectric capabilities, monolayer WSe2 has recently received a lot of attention in the fields of atomically thin electronics and optoelectronics. In this work, we focus on the evolution of the temperature-dependent dielectric function (ε = ε1 + i ε2) of monolayer WSe2 over energies from 0.74 to 6.40 eV and temperatures from 40 to 350 K. We analyze the second derivatives of ε with respect to energy to accurately locate the critical points (CP). The dependence of the observed CP energies on temperature is consistent with the alternative domination of the declining exciton binding energy as the temperature increases.

Spatial and temporal-trend assessment of desertification-sensitive land using the desertification sensitivity index in the provincial Ninh Thuan, Vietnam.

Desertification is a specific land-degrading process, reducing soil productivity and potentially threatening global food security. Therefore, spatially and temporally identifying and mapping desertification-sensitive areas is essential for better management. The current study aimed to (1) assess spatial areas sensitive to desertification and (2) examine the changing tendency of the desertification-sensitive areas over the past 25 years in the provincial Ninh Thuan. The desertification sensitivity index (DSI) was computed based on the Medalus model using 10 quantitative parameters, grouped into the soil, climate, and vegetation quality indexes, computed for the years 1996, 2005, 2010, and 2016. GIS was used to map desertification-sensitive areas associated with five DSI classes. Results showed that classes II and III had the highest area percentage, followed by classes IV and V, and class I. The classes most sensitive to desertification (classes IV and V) covered around 13 to 17%, and classes II and III were 25 to 32% of the total study area, respectively. The coastal areas located in the southeastern parts were more sensitive to desertification than the other parts. Over the four examined periods, the areas of classes IV and V increased while those of classes II and I decreased. These indicated that the study province tended to increase in its desertification sensitivity with a severe increase in the coastal areas over the past 25 years. The key factors involved in these changes could be related the human activities and climate variation, which could be more serious in southeastern areas than in the other areas.

A model approach to show that monocytes can enter microporous ß-TCP ceramics.

ß-TCP ceramics are versatile bone substitute materials and show many interactions with cells of the monocyte-macrophage-lineage. The possibility of monocytes entering microporous ß-TCP ceramics has however not yet been researched. In this study, we used a model approach to investigate whether monocytes might enter ß-TCP, providing a possible explanation for the origin of CD68-positive osteoclast-like giant cells found in earlier works.We used flow chambers to unidirectionally load BC, PRP, or PPP into slice models of either 2 mm or 6 mm ß-TCP. Immunofluorescence for CD68 and live/dead staining was performed after the loading process.Our results show that monocytes were present in a relevant number of PRP and BC slices representing the inside of our 2 mm slice model and also present on the actual inside of our 6 mm model. For PPP, monocytes were not found beyond the surface in either model.Our results indicate the possibility of a new and so far neglected constituent in ß-TCP degradation, perhaps causing the process of ceramic degradation also starting from inside the ceramics as opposed to the current understanding. We also demonstrated flow chambers as a possible new in vitro model for interactions between blood and ß-TCP.

Cost effectiveness of interventions to prevent the occurrence and the associated economic impacts of child maltreatment: A systematic review.

BACKGROUND: Child maltreatment is a pressing public health concern that poses long-lasting health and economic impacts on children and society. While several preventive interventions have demonstrated their effectiveness in reducing the occurrence of child maltreatment and its associated economic impacts, the cost-effectiveness of such interventions remains unclear. OBJECTIVE: This study aims to provide a comprehensive overview and a narrative synthesis of the available economic evidence on child maltreatment preventive interventions in both high-income and low-middle-income countries. METHOD: Systematic searches were conducted in MEDLINE, PsycINFO, Embase, CINAHL, Web of Science, and Econlit to identify full economic evaluations and return-on-investment studies on child maltreatment preventive interventions. The methodological quality of eligible studies was assessed using Drummond's 10-point checklist. This review adhered to the PRISMA guidelines, and summarized findings in a narrative synthesis. RESULTS: Twenty-six studies met the inclusion criteria. Of these, eight evaluated home visiting, four evaluated early childhood education (ECE), four assessed multi-component (MC), and three examined group-based parent education (GPE) interventions. The remaining studies assessed interventions to prevent abusive head trauma (AHT; n = 2), child sexual abuse (n = 2), physical abuse at school (n = 1), as well as individualized intensive parenting (IIP; n = 2), and counseling (n = 1) interventions. Two studies were conducted in low-middle-income countries, while the others were all in high-income countries. CONCLUSIONS: The included studies generally exhibited high methodological quality. Only AHT, ECE, IIP, and MC interventions demonstrated promising cost-effectiveness credentials in preventing child maltreatment. More economic evaluations are needed for interventions with mixed findings (e.g. GPE) and in low-middle-income countries.

A Systematic Study of the Temperature Dependence of the Dielectric Function of GaSe Uniaxial Crystals from 27 to 300 K.

We report the temperature dependences of the dielectric function ε = ε1 + iε2 and critical point (CP) energies of the uniaxial crystal GaSe in the spectral energy region from 0.74 to 6.42 eV and at temperatures from 27 to 300 K using spectroscopic ellipsometry. The fundamental bandgap and strong exciton effect near 2.1 eV are detected only in the c-direction, which is perpendicular to the cleavage plane of the crystal. The temperature dependences of the CP energies were determined by fitting the data to the phenomenological expression that incorporates the Bose-Einstein statistical factor and the temperature coefficient to describe the electron-phonon interaction. To determine the origin of this anisotropy, we perform first-principles calculations using the mBJ method for bandgap correction. The results clearly demonstrate that the anisotropic dielectric characteristics can be directly attributed to the inherent anisotropy of p orbitals. More specifically, this prominent excitonic feature and fundamental bandgap are derived from the band-to-band transition between s and pz orbitals at the Γ-point.

Cost-utility analysis of adjunct repetitive transcranial magnetic stimulation for treatment resistant bipolar depression.

OBJECTIVE: To evaluate the cost-effectiveness of repetitive transcranial magnetic stimulation (rTMS) as an adjunct to standard care from an Australian health sector perspective, compared to standard care alone for adults with treatment-resistant bipolar depression (TRBD). METHODS: An economic model was developed to estimate the cost per disability-adjusted life-year (DALY) averted and quality-adjusted life-year (QALY) gained for rTMS added to standard care compared to standard care alone, for adults with TRBD. The model simulated the time in three health states (mania, depression, residual) over one year. Response to rTMS was sourced from a meta-analysis, converted to a relative risk and used to modify the time in the depressed state. Uncertainty and sensitivity tested the robustness of results. RESULTS: Base-case incremental cost-effectiveness ratios (ICERs) were $72,299 per DALY averted (95 % Uncertainty Interval (UI): $60,915 to $86,668) and $46,623 per QALY gained (95 % UI: $39,676 - $55,161). At a willingness to pay (WTP) threshold of $96,000 per DALY averted, the base-case had a 100 % probability of being marginally cost-effective. At a WTP threshold of $64,000 per QALY gained, the base-case had a 100 % probability of being cost-effective. Sensitivity analyses decreasing the number of sessions provided, increasing the disability weight or the time spent in the depression state for standard care improved the ICERs for rTMS. CONCLUSIONS: Dependent on the outcome measure utilised and assumptions, rTMS would be considered a very cost-effective or marginally cost-effective adjunct to standard care for TRBD compared to standard care alone.

Towards a general-purpose foundation model for computational pathology.

Quantitative evaluation of tissue images is crucial for computational pathology (CPath) tasks, requiring the objective characterization of histopathological entities from whole-slide images (WSIs). The high resolution of WSIs and the variability of morphological features present significant challenges, complicating the large-scale annotation of data for high-performance applications. To address this challenge, current efforts have proposed the use of pretrained image encoders through transfer learning from natural image datasets or self-supervised learning on publicly available histopathology datasets, but have not been extensively developed and evaluated across diverse tissue types at scale. We introduce UNI, a general-purpose self-supervised model for pathology, pretrained using more than 100 million images from over 100,000 diagnostic H&E-stained WSIs (>77 TB of data) across 20 major tissue types. The model was evaluated on 34 representative CPath tasks of varying diagnostic difficulty. In addition to outperforming previous state-of-the-art models, we demonstrate new modeling capabilities in CPath such as resolution-agnostic tissue classification, slide classification using few-shot class prototypes, and disease subtyping generalization in classifying up to 108 cancer types in the OncoTree classification system. UNI advances unsupervised representation learning at scale in CPath in terms of both pretraining data and downstream evaluation, enabling data-efficient artificial intelligence models that can generalize and transfer to a wide range of diagnostically challenging tasks and clinical workflows in anatomic pathology.

A visual-language foundation model for computational pathology.

The accelerated adoption of digital pathology and advances in deep learning have enabled the development of robust models for various pathology tasks across a diverse array of diseases and patient cohorts. However, model training is often difficult due to label scarcity in the medical domain, and a model's usage is limited by the specific task and disease for which it is trained. Additionally, most models in histopathology leverage only image data, a stark contrast to how humans teach each other and reason about histopathologic entities. We introduce CONtrastive learning from Captions for Histopathology (CONCH), a visual-language foundation model developed using diverse sources of histopathology images, biomedical text and, notably, over 1.17 million image-caption pairs through task-agnostic pretraining. Evaluated on a suite of 14 diverse benchmarks, CONCH can be transferred to a wide range of downstream tasks involving histopathology images and/or text, achieving state-of-the-art performance on histology image classification, segmentation, captioning, and text-to-image and image-to-text retrieval. CONCH represents a substantial leap over concurrent visual-language pretrained systems for histopathology, with the potential to directly facilitate a wide array of machine learning-based workflows requiring minimal or no further supervised fine-tuning.

Barriers to and facilitators of user engagement with web-based mental health interventions in young people: a systematic review.

Many young people (YP) are diagnosed with mental illnesses and require support. Web-based mental health interventions (W-MHIs) have been increasingly utilized by YP, healthcare providers, and parents due to reasons including convenience and anonymity. W-MHIs are effective in improving mental health in YP. However, real-world engagement with W-MHIs remains low. Therefore, understanding barriers/facilitators of user engagement with W-MHIs is necessary to promote W-MHIs and help users gain optimal benefits through higher engagement. This review aims to identify barriers/facilitators of user engagement with W-MHIs in YP aged 10-24 years. A systematic search of five databases for English language, peer-reviewed publications was conducted between January 2010 and February 2023. Studies examining factors influencing user engagement with W-MHIs, described as barriers or facilitators, were included. Study quality was assessed using the Mixed Methods Appraisal Tool. A narrative synthesis was performed. Of 4088 articles identified, 69 studies were included. Barriers/facilitators were reported by young people (63 studies), providers (17 studies), and parents/caregivers (8 studies). YP perceived that usefulness and connectedness were the most common facilitators, whereas low-perceived need was the most reported barrier. Both providers and parents reported that perceived usefulness for YP was the most common facilitator, whereas concerns about program effectiveness and privacy were noted as barriers. This review found that program- and individual-related factors were important determinants of engagement with W-MHIs. This review provides guidance on the future design and development of new interventions, narrowing the gap between existing W-MHIs and unmet needs of users.

Age at first drink and its influence on alcohol use behaviours in young adulthood: Evidence from an Australian household-based panel study.

BACKGROUND: Public health guidelines recommend delaying the initiation age for alcohol. However, the causal link between age-at-first-drink (AFD) and future alcohol use in young adulthood is uncertain. This study examined the association between AFD and alcohol-related outcomes at age 20 years using an Australian sample. METHODS: Data were obtained from Waves 1-19 (years 2001-2019) of the Household, Income and Labour Dynamics in Australia Survey on 20-year-olds with responses across ≥3 consecutive waves (n = 2278). The AFD for each respondent (between 15 and 20 years) was analysed relative to Australian legal drinking age (18 years). Inverse probability treatment weighting was used to evaluate associations between AFD and four outcomes at age 20 years: risk of current alcohol use; quantity of weekly alcohol consumption; risk of binge drinking; and frequency of binge drinking. Adjustments were made for confounders (e.g., heavy drinking by parents). Robustness of study findings was evaluated using several diagnostic tests/sensitivity analyses. RESULTS: Among 20-year-olds, those with an AFD of 15-16 years consumed significantly more alcohol per week compared to an AFD of 18 years. Additionally, 20-year-old drinkers with an AFD of 16 years were significantly more likely to binge drink (though this association was likely confounded). An inverse dose-response relationship was observed between AFD and weekly alcohol consumption at 20 years, where a higher AFD led to lower alcohol consumption. CONCLUSION: Study findings indicate an association between a higher AFD and consuming less alcohol in young adulthood, which could potentially support the scale-up of prevention programs to delay AFD among Australian adolescents.