Improved green and red GRAB sensors for monitoring dopaminergic activity in vivo.

Dopamine (DA) plays multiple roles in a wide range of physiological and pathological processes via a large network of dopaminergic projections. To dissect the spatiotemporal dynamics of DA release in both dense and sparsely innervated brain regions, we developed a series of green and red fluorescent G-protein-coupled receptor activation-based DA (GRABDA) sensors using a variety of DA receptor subtypes. These sensors have high sensitivity, selectivity and signal-to-noise ratio with subsecond response kinetics and the ability to detect a wide range of DA concentrations. We then used these sensors in mice to measure both optogenetically evoked and behaviorally relevant DA release while measuring neurochemical signaling in the nucleus accumbens, amygdala and cortex. Using these sensors, we also detected spatially resolved heterogeneous cortical DA release in mice performing various behaviors. These next-generation GRABDA sensors provide a robust set of tools for imaging dopaminergic activity under a variety of physiological and pathological conditions.

Opioid Receptors Modulate Firing and Synaptic Transmission in the Paraventricular Nucleus of the Thalamus.

The paraventricular nucleus of the thalamus (PVT) is involved in drug addiction-related behaviors, and morphine is a widely used opioid for the relief of severe pain. Morphine acts via opioid receptors, but the function of opioid receptors in the PVT has not been fully elucidated. Here, we used in vitro electrophysiology to study neuronal activity and synaptic transmission in the PVT of male and female mice. Activation of opioid receptors suppresses the firing and inhibitory synaptic transmission of PVT neurons in brain slices. On the other hand, the involvement of opioid modulation is reduced after chronic morphine exposure, probably because of desensitization and internalization of opioid receptors in the PVT. Overall, the opioid system is essential for the modulation of PVT activities.SIGNIFICANCE STATEMENT Opioid receptors modulate the activities and synaptic transmission in the PVT by suppressing the firing rate and inhibitory synaptic inputs. These modulations were largely diminished after chronic morphine exposure.

A chemical probe targeting the PWWP domain alters NSD2 nucleolar localization.

Nuclear receptor-binding SET domain-containing 2 (NSD2) is the primary enzyme responsible for the dimethylation of lysine 36 of histone 3 (H3K36), a mark associated with active gene transcription and intergenic DNA methylation. In addition to a methyltransferase domain, NSD2 harbors two proline-tryptophan-tryptophan-proline (PWWP) domains and five plant homeodomains (PHDs) believed to serve as chromatin reading modules. Here, we report a chemical probe targeting the N-terminal PWWP (PWWP1) domain of NSD2. UNC6934 occupies the canonical H3K36me2-binding pocket of PWWP1, antagonizes PWWP1 interaction with nucleosomal H3K36me2 and selectively engages endogenous NSD2 in cells. UNC6934 induces accumulation of endogenous NSD2 in the nucleolus, phenocopying the localization defects of NSD2 protein isoforms lacking PWWP1 that result from translocations prevalent in multiple myeloma (MM). Mutations of other NSD2 chromatin reader domains also increase NSD2 nucleolar localization and enhance the effect of UNC6934. This chemical probe and the accompanying negative control UNC7145 will be useful tools in defining NSD2 biology.

Subpocket Similarity-Based Hit Identification for Challenging Targets: Application to the WDR Domain of LRRK2.

We herewith applied a priori a generic hit identification method (POEM) for difficult targets of known three-dimensional structure, relying on the simple knowledge of physicochemical and topological properties of a user-selected cavity. Searching for local similarity to a set of fragment-bound protein microenvironments of known structure, a point cloud registration algorithm is first applied to align known subpockets to the target cavity. The resulting alignment then permits us to directly pose the corresponding seed fragments in a target cavity space not typically amenable to classical docking approaches. Last, linking potentially connectable atoms by a deep generative linker enables full ligand enumeration. When applied to the WD40 repeat (WDR) central cavity of leucine-rich repeat kinase 2 (LRRK2), an unprecedented binding site, POEM was able to quickly propose 94 potential hits, five of which were subsequently confirmed to bind in vitro to LRRK2-WDR.

Prediction of coexisting invasive carcinoma on ductal carcinoma in situ (DCIS) lesions by mass spectrometry imaging.

Due to limited biopsy samples, ~20% of DCIS lesions confirmed by biopsy are upgraded to invasive ductal carcinoma (IDC) upon surgical resection. Avoiding underestimation of IDC when diagnosing DCIS has become an urgent challenge in an era discouraging overtreatment of DCIS. In this study, the metabolic profiles of 284 fresh frozen breast samples, including tumor tissues and adjacent benign tissues (ABTs) and distant surrounding tissues (DSTs), were analyzed using desorption electrospray ionization-mass spectrometry (DESI-MS) imaging. Metabolomics analysis using DESI-MS data revealed significant differences in metabolite levels, including small-molecule antioxidants, long-chain polyunsaturated fatty acids (PUFAs) and phospholipids between pure DCIS and IDC. However, the metabolic profile in DCIS with invasive carcinoma components clearly shifts to be closer to adjacent IDC components. For instance, DCIS with invasive carcinoma components showed lower levels of antioxidants and higher levels of free fatty acids compared to pure DCIS. Furthermore, the accumulation of long-chain PUFAs and the phosphatidylinositols (PIs) containing PUFA residues may also be associated with the progression of DCIS. These distinctive metabolic characteristics may offer valuable indications for investigating the malignant potential of DCIS. By combining DESI-MS data with machine learning (ML) methods, various breast lesions were discriminated. Importantly, the pure DCIS components were successfully distinguished from the DCIS components in samples with invasion in postoperative specimens by a Lasso prediction model, achieving an AUC value of 0.851. In addition, pixel-level prediction based on DESI-MS data enabled automatic visualization of tissue properties across whole tissue sections. Summarily, DESI-MS imaging on histopathological sections can provide abundant metabolic information about breast lesions. By analyzing the spatial metabolic characteristics in tissue sections, this technology has the potential to facilitate accurate diagnosis and individualized treatment of DCIS by inferring the presence of IDC components surrounding DCIS lesions. © 2023 The Pathological Society of Great Britain and Ireland.

Physiologically based pharmacokinetic modeling for confirming the role of CYP3A1/2 and P-glycoprotein in detoxification mechanism between glycyrrhizic acid and aconitine in rats.

Fuzi, an effective common herb, is often combined with Gancao to treat disease in clinical practice with enhancing its efficacy and alleviating its toxicity. The major toxic and bioactive compounds in Fuzi and Gancao are aconitine (AC) and glycyrrhizic acid (GL), respectively. This study aims to elucidate detoxification mechanism between AC and GL from pharmacokinetic perspective using physiologically based pharmacokinetic (PBPK) model. In vitro experiments exhibited that AC was mainly metabolized by CYP3A1/2 in rat liver microsomes and transported by P-glycoprotein (P-gp) in Caco-2 cells. Kinetics assays showed that the Km and Vmax of AC towards CYP3A1/2 were 2.38 µM and 57.3 pmol/min/mg, respectively, whereas that of AC towards P-gp was 11.26 µM and 147.1 pmol/min/mg, respectively. GL markedly induced the mRNA expressions of CYP3A1/2 and MDR1a/b in rat primary hepatocytes. In vivo studies suggested that the intragastric and intravenous administration of GL significantly reduced systemic exposure of AC by 27% and 33%, respectively. Drug-drug interaction (DDI) model of PBPK predicted that co-administration of GL would decrease the exposure of AC by 39% and 45% in intragastric and intravenous dosing group, respectively. The consistency between predicted data and observed data confirmed that the upregulation of CYP3A1/2 and P-gp was the crucial detoxification mechanism between AC and GL. Thus, this study provides a demonstration for elucidating the compatibility mechanisms of herbal formula using PBPK modeling and gives support for the clinical co-medication of Fuzi and Gancao.

The effect of childbirth trauma on willingness to donate milk in women following perinatal loss: The mediating role of resilience.

Limited studies have been conducted on Chinese women's willingness to donate milk following perinatal loss. In this study, we explore the relationship among childbirth trauma, willingness to donate milk, and resilience in women following perinatal loss, and the mediating effect of resilience between childbirth trauma and willingness to donate milk. A cross-sectional study was carried out throughout 4 months. We used convenience sampling methods and recruited 241 women following a perinatal loss from eight tertiary hospitals in Sichuan Province, China. Participants completed four questionnaires during a face-to-face individual interview: the general information questionnaire, the Willingness to Donate Milk Scale (WMDS), the City Birth Trauma Scale, and the Brief Resilience Scale. SPSS 20.0 was used to analyze the collected data. In our study, childbirth trauma was negatively correlated with the total and each dimension score of WMDS (p < 0.001). Resilience was positively correlated with the total and each dimension score of WMDS (p < 0.001). Resilience partially mediated the relationship between childbirth-related symptoms and willingness to donate milk (ß = -0.38, 95% confidence interval [CI]: -0.50 to -0.26), which accounted for 69.03% of the total effect. Resilience partially mediated the relationship between general symptoms and willingness to donate milk (ß = -0.31, 95% CI: -0.40 to -0.21), which accounted for 66.89% of the total effect. Resilience partially mediated the relationship between childbirth trauma and willingness to donate milk in women following perinatal loss. Our findings suggest that resilience can play a significant role in mediating the relationship between childbirth trauma and willingness to donate milk in women following perinatal loss. These results could help healthcare professionals design interventions for physical and mental recovery after perinatal loss.

Discovery of a Potent and Selective Targeted NSD2 Degrader for the Reduction of H3K36me2.

Nuclear receptor-binding SET domain-containing 2 (NSD2) plays important roles in gene regulation, largely through its ability to dimethylate lysine 36 of histone 3 (H3K36me2). Despite aberrant activity of NSD2 reported in numerous cancers, efforts to selectively inhibit the catalytic activity of this protein with small molecules have been unsuccessful to date. Here, we report the development of UNC8153, a novel NSD2-targeted degrader that potently and selectively reduces the cellular levels of both NSD2 protein and the H3K36me2 chromatin mark. UNC8153 contains a simple warhead that confers proteasome-dependent degradation of NSD2 through a novel mechanism. Importantly, UNC8153-mediated reduction of H3K36me2 through the degradation of NSD2 results in the downregulation of pathological phenotypes in multiple myeloma cells including mild antiproliferative effects in MM1.S cells containing an activating point mutation and antiadhesive effects in KMS11 cells harboring the t(4;14) translocation that upregulates NSD2 expression.

Transcription factor ETV1-induced lncRNA MAFG-AS1 promotes migration, invasion, and epithelial-mesenchymal transition of pancreatic cancer cells by recruiting IGF2BP2 to stabilize ETV1 expression.

We investigated the mechanism of ETS-translocation variant 1 (ETV1)/lncRNA-MAFG-AS1 in pancreatic cancer (PC). MAFG-AS1 and ETV1 levels in PC cell lines and HPNE cells were determined using reverse transcription quantitative polymerase chain reaction (RT-qPCR) and Western blotting (WB). After transfection with sh-MAFG-AS1, PC cell invasion, migration, proliferation, and epithelial-mesenchymal transition (EMT)-related proteins were measured by 5-ethynyl-2'-deoxyuridine (EdU), Transwell assay, and WB. The binding between ETV1 and MAFG-AS1 was studied using dual-luciferase assay and chromatin immunoprecipitation. The interactions between MAFG-AS1, IGF2BP2, and ETV1 were tested. Combined experiments were further performed using sh-MAFG-AS1 and pcDNA-ETV1 simultaneously. ETV1/MAFG-AS1 was highly expressed in PC cells. Blocking MAFG-AS1 inhibited the malignant behaviors of PC cells. ETV1 induced MAFG-AS1 transcription in PC cells. MAFG-AS1 stabilized ETV1 mRNA by recruiting IGF2BP2. ETV1 overexpression partially antagonized the suppression of silencing MAFG-AS1 on PC cells. ETV1-induced MAFG-AS1 stabilized the ETV1 expression by recruiting IGF2BP2 and promoted PC cell migration, invasion, proliferation, and EMT.

Combining the tumor-stroma ratio with tumor-infiltrating lymphocytes improves the prediction of pathological complete response in breast cancer patients.

PURPOSE: The tumor-stroma ratio (TSR) is a common histological parameter that measures stromal abundance and is prognostic in breast cancer (BC). However, more evidence is needed on the predictive value of the TSR for the pathological complete response (pCR) to neoadjuvant chemotherapy (NAC). The purpose of this study was to determine the importance of the TSR in predicting pCR in NAC settings. METHOD: We evaluated the TSR on pretreatment biopsies of 912 BC patients from four independent Chinese hospitals and investigated the potential value of the TSR for predicting pCR. Meanwhile, stromal tumor-infiltrating lymphocytes (sTILs) were assessed, and we evaluated the predictive value of the combination of sTILs and TSR (TSRILs). RESULTS: Patients with low stroma showed a higher pCR rate than those with high stroma among the four independent hospitals, and in multivariate analysis, the TSR was proven to be an independent predictor for pCR to NAC with an odds ratio of 1.945 (95% CI 1.230-3.075, P = 0.004). Moreover, we found that TSRILs could improve the area under the curve (AUC) for predicting pCR from 0.750 to 0.785 (P = 0.039); especially in HER2-negative BCs, the inclusion of TSRILs increased the AUC from 0.801 to 0.835 in the discovery dataset (P = 0.048) and 0.734 to 0.801 in the validation dataset (P = 0.003). CONCLUSION: TSR and sTILs can be easily measured in pathological routines and provide predictive information without additional cost; with more evidence from clinical trials, TSRILs could be a candidate to better stratify patients in NAC settings.

A circuit from lateral septum neurotensin neurons to tuberal nucleus controls hedonic feeding.

Feeding behavior is regulated by both the homeostatic needs of the body and hedonic values of the food. Easy access to palatable energy-dense foods and the consequent obesity epidemic stress the urgent need for a better understanding of neural circuits that regulate hedonic feeding. Here, we report that neurotensin-positive neurons in the lateral septum (LSNts) play a crucial role in regulating hedonic feeding. Silencing LSNts specifically promotes feeding of palatable food, whereas activation of LSNts suppresses overall feeding. LSNts neurons project to the tuberal nucleus (TU) via GABA signaling to regulate hedonic feeding, while the neurotensin signal from LSNts→the supramammillary nucleus (SUM) is sufficient to suppress overall feeding. In vivo calcium imaging and optogenetic manipulation reveal two populations of LSNts neurons that are activated and inhibited during feeding, which contribute to food seeking and consumption, respectively. Chronic activation of LSNts or LSNts→TU is sufficient to reduce high-fat diet-induced obesity. Our findings suggest that LSNts→TU is a key pathway in regulating hedonic feeding.

Development of LM-41 and AF-2112, two flufenamic acid-derived TEAD inhibitors obtained through the replacement of the trifluoromethyl group by aryl rings.

The Hippo pathway regulates organ size and tissue homeostasis by controlling cell proliferation and apoptosis. The YAP-TEAD transcription factor, the downstream effector of the Hippo pathway, regulates the expression of genes such as CTGF, Cyr61, Axl and NF2. Aberrant Hippo activity has been identified in multiple types of cancers. Flufenamic acid (FA) was reported to bind in a liphophilic TEAD palmitic acid (PA) pocket, leading to reduction of the expression of Axl and NF2. Here, we show that the replacement of the trifluoromethyl moiety in FA by aromatic groups, directly connected to the scaffold or separated by a linker, leads to compounds with better affinity to TEAD. Co-crystallization studies show that these compounds bind similarly to FA, but deeper within the PA pocket. Our studies identified LM-41 and AF-2112 as two TEAD binders that strongly reduce the expression of CTGF, Cyr61, Axl and NF2. LM-41 gave the strongest reduction of migration of human MDA-MB-231 breast cancer cells.

Discovery of a Novel DCAF1 Ligand Using a Drug-Target Interaction Prediction Model: Generalizing Machine Learning to New Drug Targets.

DCAF1 functions as a substrate recruitment subunit for the RING-type CRL4DCAF1 and the HECT family EDVPDCAF1 E3 ubiquitin ligases. The WDR domain of DCAF1 serves as a binding platform for substrate proteins and is also targeted by HIV and SIV lentiviral adaptors to induce the ubiquitination and proteasomal degradation of antiviral host factors. It is therefore attractive both as a potential therapeutic target for the development of chemical inhibitors and as an E3 ligase that could be recruited by novel PROTACs for targeted protein degradation. In this study, we used a proteome-scale drug-target interaction prediction model, MatchMaker, combined with cheminformatics filtering and docking to identify ligands for the DCAF1 WDR domain. Biophysical screening and X-ray crystallographic studies of the predicted binders confirmed a selective ligand occupying the central cavity of the WDR domain. This study shows that artificial intelligence-enabled virtual screening methods can successfully be applied in the absence of previously known ligands.

Effects of luxS gene on growth characteristics, biofilm formation, and antimicrobial resistance of multi-antimicrobial-resistant Vibrio parahaemolyticus Vp2015094 isolated from shellfish.

AIMS: Vibrio parahaemolyticus is an important foodborne pathogen worldwide, which can cause gastroenteritis. This study aimed to investigate the effect of quorum sensing system LuxS/AI-2-related gene luxS on the biological characteristics and antimicrobial resistance of V. parahaemolyticus Vp2015094 from shellfish, which carried a multi-antimicrobial-resistant plasmid. METHODS AND RESULTS: The critical gene luxS related to the synthesis of AI-2 in V. parahaemolyticus Vp2015094 was knocked out by homologous recombination with suicide plasmid. The effect of luxS on the biological characteristics of V. parahaemolyticus was determined by comparing the growth, AI-2 activity, motility, biofilm formation ability, and antibiotic resistance between the wildtype strain and the luxS deletion mutant. Compared with wildtype strain, the production of AI-2, the motility and biofilm formation ability, antimicrobial resistance, and conjugation frequency of luxS deletion mutant strain were decreased. The transcriptome sequencing showed that the transcriptional levels of many genes related to motility, biofilm formation, antimicrobial resistance, and conjugation were significantly downregulated after luxS deletion. CONCLUSIONS: Quorum sensing system LuxS/AI-2-related gene luxS in V. parahaemolyticus Vp2015094 played an important role in growth characteristics, biofilm formation, antimicrobial resistance, and resistance genes' transfer.

Tetrahedral DNA framework assisted rotational paper-based analytical device for differential detection of SARS-CoV-2 and influenza A H1N1 virus.

Coronavirus disease 2019 (COVID-19) and influenza A are two respiratory infectious diseases with similar clinical manifestations. Because of the complex global epidemic situation of COVID-19, the distinction and diagnosis of COVID-19 and influenza A infected persons is crucial for epidemic prevention and control. In this study, tetrahedral DNA framework (TDF) was combined with a rotational paper-based analytical device, and the color change generated by the reaction between horseradish peroxidase (HRP) and 3,3'5,5'-tetramethylbenzidine (TMB)-H2O2 was used for grayscale signal analysis by ImageJ software. The quantitative detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza A H1N1 virus were realized simultaneously. Under the optimal conditions, the paper-based analytical device showed a good linear relationship between the two viruses in the range of 10-14-10-8g/mL, and the two viruses were not affected by cross reaction. This sensor provides a convenient and reliable method for clinical rapid differentiation and diagnosis of COVID-19 and influenza A.

Sarcoeleganolides C-G, Five New Cembranes from the South China Sea Soft Coral Sarcophyton elegans.

Five new cembranes, named sarcoeleganolides C-G (1-5), along with three known analogs (6-8) were isolated from soft coral Sarcophyton elegans collected from the Yagong Island, South China Sea. Their structures and absolute configurations were determined by extensive spectroscopic analysis, QM-NMR, and TDDFT-ECD calculations. In addition, compound 3 exhibited better anti-inflammation activity compared to the indomethacin as a positive control in zebrafish at 20 µM.

Green photoreduction synthesis of dispersible gold nanoparticles and their direct in situ assembling in multidimensional substrates for SERS detection.

Gold nanoparticles (AuNPs) and their composites have been applied in surface-enhanced Raman scattering (SERS) detection methods, owing to their stable and excellent surface plasmon resonance. Unfortunately, methods for synthesizing AuNPs often require harsh conditions and complicated external steps. Additionally, removing residual surfactants or unreacted reductants is critical for improving the sensitivity of SERS detection, especially when employing AuNPs-assembled multidimensional substrates. In this study, we propose a simple and green method for AuNPs synthesis via photoreduction, which does not require external surfactant additives or stabilizers. All the processes were completed within 20 min. Along this way, only methanol was employed as the electron acceptor. Based on this photoreduction synthesis strategy, AuNPs can be directly and circularly assembled in situ in multidimensional substrates for SERS detection. The removal of residual methanol was easy because of its low boiling point. This strategy was employed for the preparation of three different dimensional SERS substrates: filter paper@AuNPs, g-C3N4@AuNPs, and MIL-101(Cr)@AuNPs. The limit of detection of filter paper@AuNPs for thiabendazole SERS detection was 1.0 × 10-7 mol/L, while the limits of detection of g-C3N4@AuNPs and MIL-101(Cr)@AuNPs for malachite green SERS detection were both 5.0 × 10-11 mol/L. This strategy presents potential in AuNP doping materials and SERS detection.

The development of genus-specific and species-specific real-time PCR assays for the authentication of Patagonian toothfish and Antarctic toothfish in commercial seafood products.

BACKGROUND: The substitution or mislabeling of toothfish is an issue of significant concern for seafood authorities; it also reduces the effectiveness of marine conservation and management programs for its over-exploitation and illegal trafficking, boosting the need for identification methods. RESULTS: Two species-specific real-time polymerase chain reaction (PCR) assays for the identification of Patagonian toothfish (Dissostichus eleginoides) and Antarctic toothfish (Dissostichus mawsoni) and a genus-specific real-time PCR assay for Dissostichus spp. identification were developed based on fragments of the 16S rRNA and COI (cytochrome c oxidase subunit I) genes. These methods were confirmed to be rapid, simple, and sensitive (absolute sensitivity of 0.0002 ng µL-1 and relative sensitivity of 0.1 g kg-1 with good specificity). These methods can be applied to processed and commercial fish products. CONCLUSIONS: These approaches can be beneficial for protecting both consumers and producers from economic fraud and might also help protect toothfish from over-exploitation as well as combat illegal, unreported, and unregulated (IUU) fisheries. © 2021 Society of Chemical Industry.

Inhibition of the SUV4-20 H1 histone methyltransferase increases frataxin expression in Friedreich's ataxia patient cells.

The molecular mechanisms of reduced frataxin (FXN) expression in Friedreich's ataxia (FRDA) are linked to epigenetic modification of the FXN locus caused by the disease-associated GAA expansion. Here, we identify that SUV4-20 histone methyltransferases, specifically SUV4-20 H1, play an important role in the regulation of FXN expression and represent a novel therapeutic target. Using a human FXN-GAA-Luciferase repeat expansion genomic DNA reporter model of FRDA, we screened the Structural Genomics Consortium epigenetic probe collection. We found that pharmacological inhibition of the SUV4-20 methyltransferases by the tool compound A-196 increased the expression of FXN by ∼1.5-fold in the reporter cell line. In several FRDA cell lines and patient-derived primary peripheral blood mononuclear cells, A-196 increased FXN expression by up to 2-fold, an effect not seen in WT cells. SUV4-20 inhibition was accompanied by a reduction in H4K20me2 and H4K20me3 and an increase in H4K20me1, but only modest (1.4-7.8%) perturbation in genome-wide expression was observed. Finally, based on the structural activity relationship and crystal structure of A-196, novel small molecule A-196 analogs were synthesized and shown to give a 20-fold increase in potency for increasing FXN expression. Overall, our results suggest that histone methylation is important in the regulation of FXN expression and highlight SUV4-20 H1 as a potential novel therapeutic target for FRDA.

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer.

BACKGROUND: Pathological complete response (pCR) is considered a surrogate endpoint for favorable survival in breast cancer patients treated with neoadjuvant chemotherapy (NAC). Predictive biomarkers of treatment response are crucial for guiding treatment decisions. With the hypothesis that histological information on tumor biopsy images could predict NAC response in breast cancer, we proposed a novel deep learning (DL)-based biomarker that predicts pCR from images of hematoxylin and eosin (H&E)-stained tissue and evaluated its predictive performance. METHODS: In total, 540 breast cancer patients receiving standard NAC were enrolled. Based on H&E-stained images, DL methods were employed to automatically identify tumor epithelium and predict pCR by scoring the identified tumor epithelium to produce a histopathological biomarker, the pCR-score. The predictive performance of the pCR-score was assessed and compared with that of conventional biomarkers including stromal tumor-infiltrating lymphocytes (sTILs) and subtype. RESULTS: The pCR-score derived from H&E staining achieved an area under the curve (AUC) of 0.847 in predicting pCR directly, and achieved accuracy, F1 score, and AUC of 0.853, 0.503, and 0.822 processed by the logistic regression method, respectively, higher than either sTILs or subtype; a prediction model of pCR constructed by integrating sTILs, subtype and pCR-score yielded a mean AUC of 0.890, outperforming the baseline sTIL-subtype model by 0.051 (0.839, P = 0.001). CONCLUSION: The DL-based pCR-score from histological images is predictive of pCR better than sTILs and subtype, and holds the great potentials for a more accurate stratification of patients for NAC.