Comparison of deep learning models to traditional Cox regression in predicting survival of colon cancer: Based on the SEER database.

BACKGROUND AND AIM: In this study, a deep learning algorithm was used to predict the survival rate of colon cancer (CC) patients, and compared its performance with traditional Cox regression. METHODS: In this population-based cohort study, we used the characteristics of patients diagnosed with CC between 2010 and 2015 from the Surveillance, Epidemiology and End Results (SEER) database. The population was randomized into a training set (n = 10 596, 70%) and a test set (n = 4536, 30%). Brier scores, area under the (AUC) receiver operating characteristic curve and calibration curves were used to compare the performance of the three most popular deep learning models, namely, artificial neural networks (ANN), deep neural networks (DNN), and long-short term memory (LSTM) neural networks with Cox proportional hazard (CPH) model. RESULTS: In the independent test set, the Brier values of ANN, DNN, LSTM and CPH were 0.155, 0.149, 0.148, and 0.170, respectively. The AUC values were 0.906 (95% confidence interval [CI] 0.897-0.916), 0.908 (95% CI 0.899-0.918), 0.910 (95% CI 0.901-0.919), and 0.793 (95% CI 0.769-0.816), respectively. Deep learning showed superior promising results than CPH in predicting CC specific survival. CONCLUSIONS: Deep learning showed potential advantages over traditional CPH models in terms of prognostic assessment and treatment recommendations. LSTM exhibited optimal predictive accuracy and has the ability to provide reliable information on individual survival and treatment recommendations for CC patients.

First-principles prediction of ferroelectric Janus Si2XY (X/Y = S/Se/Te, X ≠ Y) monolayers with negative Poisson's ratios.

Nowadays, two-dimensional (2D) materials with Janus structures evoke much attention due to their unique mechanical and electronic properties. In this work, Janus Pma2-Si2XY (X/Y = S/Se/Te, X ≠ Y) ferroelectric monolayers are firstly proposed and systematically investigated by first-principles calculations. These monolayers exhibit remarkable mechanical properties, including small Young's modulus values, negative Poisson's ratios (NPRs) and large critical strains, reflecting their exceptional flexibility and stretchability. More strikingly, the novel structures of Si2STe and Si2SeTe also endow them with in-plane spontaneous polarization (Ps) and low energy barrier for phase transition, with Ps and energy barrier values being 1.632 × 10-10 C m-1 and 159 meV for Si2STe and 1.149 × 10-10 C m-1 and 196.6 meV for Si2SeTe. The ab initio molecular dynamics (AIMD) simulations reveal high Curie temperatures (Tc) for Si2STe and Si2SeTe, ranging between 1300 K and 1400 K. Additionally, Si2XY monolayers exhibit high anisotropic carrier mobility (∼103 cm2 V-1 s-1) and an extraordinary light absorption coefficient (∼105 cm-1). Our research not only broadens the family of 2D Janus ferroelectric materials, but also demonstrates their potential applications in nanomechanical, nanoelectronic and optoelectronic devices.

Janus monolayer PXC (X = As/Sb) for photocatalytic water splitting with a negative Poisson's ratio.

Two-dimensional (2D) Janus materials have attracted considerable attention in photocatalysis owing to their robust redox capability and efficient segregation. In this study, we propose a novel Janus monolayer structure, denoted as PXC (X = As/Sb), exhibiting favorable stability in terms of dynamics, thermal properties, and mechanical characteristics. The PXC monolayers demonstrate a relatively smaller Young's modulus (132.5/119.5 N m-1 for PAsC/PSbC) and large negative Poisson's ratios (-0.15/-0.101 for PAsC/PSbC). Moreover, the HSE06 + SOC functional results show that PAsC/PSbC are indirect semiconductors with a 2.33/1.43 eV band gap, exhibiting a suitable band alignment for photocatalytic water splitting. The calculated high carrier mobility (104 cm2 V-1 s-1), along with a significant discrepancy, determined by the deformation potential theory and the built-up field induced by the large intrinsic dipole, effectively suppresses the recombination of photogenerated carriers. Furthermore, PXC monolayers possess a strong absorption capacity in the visible and ultraviolet light region (105 cm-1). Therefore, our results indicate that PXC monolayers hold great potential for application in the field of photocatalytic water splitting.

Natural Product-Inspired Molecules for Covalent Inhibition of SHP2 Tyrosine Phosphatase.

Natural products have been playing indispensable roles in the development of lifesaving drug molecules. They are also valuable sources for covalent protein modifiers. However, they often are scarce in nature and have complex chemical structures, which are limiting their further biomedical development. Thus, natural product-inspired small molecules which still contain the essence of the parent natural products but are readily available and amenable for structural modification, are important and desirable in searching for lead compounds for various disease treatment. Inspired by the complex and diverse ent-kaurene diterpenoids with significant biological activities, we have created a synthetically accessible and focused covalent library by incorporating the bicyclo[3.2.1]octane α-methylene ketone, which is considered as the pharmacophore of ent-kaurene diterpenoids, as half of the structure, and replacing the other half with much less complex but more druglike scaffolds. From this library, we have identified and characterized selective covalent inhibitors of protein tyrosine phosphatase SHP2, an important anti-cancer therapeutic target. The success of this study demonstrated the importance of creating and evaluating natural product-inspired library as well as their application in targeting challenging disease targets.

Multiple Gas Detection by Cavity-Enhanced Raman Spectroscopy with Sub-ppm Sensitivity.

Accurate and sensitive detection of multicomponent trace gases below the parts-per-million (ppm) level is needed in a variety of medical, industrial, and environmental applications. Raman spectroscopy can identify multiple molecules in the sample simultaneously and has excellent potential for fast diagnosis of various samples, but applications are often limited by its sensitivity. In this contribution, we report the development of a cavity-enhanced Raman spectroscopy instrument using a narrow-line width 532 nm laser locked with a high-finesse cavity through a Pound-Drever-Hall locking servo, which allows continuous measurement in a broad spectral range. An intracavity laser power of up to 1 kW was achieved with an incident laser power of about 240 mW, resulting in a significant enhancement of the Raman signal in the range of 200-5000 cm-1 and a sub-ppm sensitivity for various molecules. The technique is applied in the detection of different samples, including ambient air, natural gas, and reference gas of sulfur hexafluoride, demonstrating its capability for the quantitative measurement of various trace components.

Nano-DESI Mass Spectrometry Imaging of Proteoforms in Biological Tissues with High Spatial Resolution.

Mass spectrometry imaging (MSI) is a powerful tool for label-free mapping of the spatial distribution of proteins in biological tissues. We have previously demonstrated imaging of individual proteoforms in biological tissues using nanospray desorption electrospray ionization (nano-DESI), an ambient liquid extraction-based MSI technique. Nano-DESI MSI generates multiply charged protein ions, which is advantageous for their identification using top-down proteomics analysis. In this study, we demonstrate proteoform mapping in biological tissues with a spatial resolution down to 7 µm using nano-DESI MSI. A substantial decrease in protein signals observed in high-spatial-resolution MSI makes these experiments challenging. We have enhanced the sensitivity of nano-DESI MSI experiments by optimizing the design of the capillary-based probe and the thickness of the tissue section. In addition, we demonstrate that oversampling may be used to further improve spatial resolution at little or no expense to sensitivity. These developments represent a new step in MSI-based spatial proteomics, which complements targeted imaging modalities widely used for studying biological systems.

Association between anemia and depression: results from NHANES 2005-2018 and mendelian randomization analyses.

BACKGROUND: The relationship between anemia and depression remains controversial. OBJECTIVE: To explore the association between anemia/hemoglobin and depression. METHODS: The data for our cross-sectional study were obtained from the National Health and Nutrition Examination Survey (NHANES) 2005-2018. Weighted multivariate logistic regression was performed to examine the association between anemia/hemoglobin and depression. Inverse variance weighted (IVW), weighted-median, and MR-Egger were used in MR analyses to assess the causal relationship between anemia/hemoglobin and depression. Heterogeneity and directional pleiotropy were assessed using the Cochrane Q test and Egger-intercept test, respectively. Sensitivity analysis was conducted by the leave-one-out approach. All analyses were carried out using IBM SPSS 24.0 and R version 4.2.2. RESULTS: A total of 29,391 NHANES participants were included in this study. After adjusting for all covariates, the association between anemia/hemoglobin and depression was not significant (P < 0.05). IVW estimates revealed that broad anemia had no significant effect on the risk of depression (OR = 1.00, 95% CI = 0.99-1.01, P = 0.432). Findings of weighted median and MR-Egger were consistent with those from IVW (weighted median: OR = 1.00, 95% CI = 0.99-1.02; P = 0.547; MR-Egger: OR = 1.01, 95% CI = 0.98-1.03, P = 0.605). The results of three MR Analyses methods also showed no causal association between hemoglobin and depression. CONCLUSIONS: Our findings do not support a causal association between anemia and depression. The association between hemoglobin concentration and depression was not statistically significant either.

Electrostatic gating dependent multiple band alignments in ferroelectric VS2/Ga2O3 van der Waals heterostructures.

Two-dimensional (2D) van der Waals (vdW) heterostructures with spontaneous intrinsic ferroelectrics play an essential role in ferroelectric memories. Also, the reversal of polarized directions induces band alignment transitions among different types to provide a new path for multifunctional devices. In this work, the structural and electronic properties of 2D VS2/Ga2O3 vdW heterostructures under different polarizations were investigated using first-principles calculations with the vdW correction of the DFT-D2 method. The results reveal that the polarized direction of a 2D Ga2O3 monolayer can cause a distinct band structure reversion from a metal to a semiconductor due to the shift of band alignment induced by the interlayer charge transfer. Moreover, the VS2/P↑ Ga2O3 heterostructures retain type-I and type-II band alignments in the majority and minority channel, respectively, under an external electric field. Interestingly, applying the external electric field for VS2/P↓ Ga2O3 heterostructures can lead to a transition from type-II to type-I in the majority channel, and from type-II to type-III in the minority channel. Our work provides a feasible way to realize 2D VS2/Ga2O3 vdW heterostructures for potential applications in ferroelectric memories and electrostatic gating dependent multiple band alignment devices.

Robust ferroelectricity in low-dimensional δ-SiX (X = S/Se): a first-principles study.

Low-dimensional ferroelectric materials hold great promise for application in nonvolatile memory devices. In this work, ferroelectricity in two-dimensional monolayers and one-dimensional nanowires based on δ-SiX (X = S and Se) materials with spontaneous polarization and ferroelectric switching energy barriers has been predicted using the first-principles method. The results show that the intrinsic ferroelectric values due to spontaneous polarization of 2D-SiS, 2D-SiSe, 1D-SiS and 1D-SiSe are 3.22 × 10-10 C m-1, 3.00 × 10-10 C m-1, 7.58 × 10-10 C m-1 and 6.81 × 10-10 C m-1, respectively. The Monte Carlo simulations and ab initio molecular dynamics (AIMD) simulations both indicate that 2D-SiX and 1D-SiX exhibit room-temperature ferroelectricity. Moreover, the polarization and ferroelectric switching energy barrier can be tuned by applying a strain. Notably, spontaneous spin polarization can be achieved by hole doping in one-dimensional nanowires. Our findings not only broaden the research field of low-dimensional ferroelectric materials, but also provide a promising platform for the application of novel nano-ferroelectric devices.

Diagnostic performance of magnetic resonance imaging-based machine learning in Alzheimer's disease detection: a meta-analysis.

PURPOSE: Advanced machine learning (ML) algorithms can assist rapid medical image recognition and realize automatic, efficient, noninvasive, and convenient diagnosis. We aim to further evaluate the diagnostic performance of ML to distinguish patients with probable Alzheimer's disease (AD) from normal older adults based on structural magnetic resonance imaging (MRI). METHODS: The Medline, Embase, and Cochrane Library databases were searched for relevant literature published up until July 2021. We used the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool and Checklist for Artificial Intelligence in Medical Imaging (CLAIM) to evaluate all included studies' quality and potential bias. Random-effects models were used to calculate pooled sensitivity and specificity, and the Deeks' test was used to assess publication bias. RESULTS: We included 24 models based on different brain features extracted by ML algorithms in 19 papers. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the summary receiver operating characteristic curve for ML in detecting AD were 0.85 (95%CI 0.81-0.89), 0.88 (95%CI 0.84-0.91), 7.15 (95%CI 5.40-9.47), 0.17 (95%CI 0.12-0.22), 43.34 (95%CI 26.89-69.84), and 0.93 (95%CI 0.91-0.95). CONCLUSION: ML using structural MRI data performed well in diagnosing probable AD patients and normal elderly. However, more high-quality, large-scale prospective studies are needed to further enhance the reliability and generalizability of ML for clinical applications before it can be introduced into clinical practice.

Identifying depression in the United States veterans using deep learning algorithms, NHANES 2005-2018.

BACKGROUND: Depression is a common mental health problem among veterans, with high mortality. Despite the numerous conducted investigations, the prediction and identification of risk factors for depression are still severely limited. This study used a deep learning algorithm to identify depression in veterans and its factors associated with clinical manifestations. METHODS: Our data originated from the National Health and Nutrition Examination Survey (2005-2018). A dataset of 2,546 veterans was identified using deep learning and five traditional machine learning algorithms with 10-fold cross-validation. Model performance was assessed by examining the area under the subject operating characteristic curve (AUC), accuracy, recall, specificity, precision, and F1 score. RESULTS: Deep learning had the highest AUC (0.891, 95%CI 0.869-0.914) and specificity (0.906) in identifying depression in veterans. Further study on depression among veterans of different ages showed that the AUC values for deep learning were 0.929 (95%CI 0.904-0.955) in the middle-aged group and 0.924(95%CI 0.900-0.948) in the older age group. In addition to general health conditions, sleep difficulties, memory impairment, work incapacity, income, BMI, and chronic diseases, factors such as vitamins E and C, and palmitic acid were also identified as important influencing factors. CONCLUSIONS: Compared with traditional machine learning methods, deep learning algorithms achieved optimal performance, making it conducive for identifying depression and its risk factors among veterans.

Discovery of a SHP2 Degrader with In Vivo Anti-Tumor Activity.

Src homology 2 domain-containing phosphatase 2 (SHP2) is an attractive target for cancer therapy due to its multifaceted roles in both tumor and immune cells. Herein, we designed and synthesized a novel series of proteolysis targeting chimeras (PROTACs) using a SHP2 allosteric inhibitor as warhead, with the goal of achieving SHP2 degradation both inside the cell and in vivo. Among these molecules, compound P9 induces efficient degradation of SHP2 (DC50 = 35.2 ± 1.5 nM) in a concentration- and time-dependent manner. Mechanistic investigation illustrates that the P9-mediated SHP2 degradation requires the recruitment of the E3 ligase and is ubiquitination- and proteasome-dependent. P9 shows improved anti-tumor activity in a number of cancer cell lines over its parent allosteric inhibitor. Importantly, administration of P9 leads to a nearly complete tumor regression in a xenograft mouse model, as a result of robust SHP2 depletion and suppression of phospho-ERK1/2 in the tumor. Hence, P9 represents the first SHP2 PROTAC molecule with excellent in vivo efficacy. It is anticipated that P9 could serve not only as a new chemical tool to interrogate SHP2 biology but also as a starting point for the development of novel therapeutics targeting SHP2.

Small Molecule Degraders of Protein Tyrosine Phosphatase 1B and T-Cell Protein Tyrosine Phosphatase for Cancer Immunotherapy.

Protein tyrosine phosphatase 1B (PTP1B) and T-cell protein tyrosine phosphatase (TC-PTP) play non-redundant negative regulatory roles in T-cell activation, tumor antigen presentation, insulin and leptin signaling, and are potential targets for several therapeutic applications. Here, we report the development of a highly potent and selective small molecule degrader DU-14 for both PTP1B and TC-PTP. DU-14 mediated PTP1B and TC-PTP degradation requires both target protein(s) and VHL E3 ligase engagement and is also ubiquitination- and proteasome-dependent. DU-14 enhances IFN-γ induced JAK1/2-STAT1 pathway activation and promotes MHC-I expression in tumor cells. DU-14 also activates CD8+ T-cells and augments STAT1 and STAT5 phosphorylation. Importantly, DU-14 induces PTP1B and TC-PTP degradation in vivo and suppresses MC38 syngeneic tumor growth. The results indicate that DU-14, as the first PTP1B and TC-PTP dual degrader, merits further development for treating cancer and other indications.

Accuracy of clustered regularly interspaced short palindromic repeats (CRISPR) to diagnose COVID-19, a meta-analysis.

OBJECTIVE: To estimate the accuracy of clustered regularly interspaced short palindromic repeats (CRISPR) in determining coronavirus disease-19 (COVID-19). METHODS: As of January 31, 2022, PubMed, Web of Science, Embase, Science Direct, Wiley and Springer Link were searched. Sensitivity, specificity, likelihood ratio (LR), diagnostic odds ratio (DOR) and area under the summary receiver-operating characteristic (AUC) curve were used to assess the accuracy of CRISPR. RESULTS: According to the inclusion criteria, 5857 patients from 54 studies were included in this meta-analysis. The pooled sensitivity, specificity and AUC were 0.98, 1.00 and 1.00, respectively. For CRISPR-associated (Cas) proteins-12, the sensitivity, specificity was 0.96, 1.00, respectively. For Cas-13, the sensitivity and specificity were 0.99 and 0.99. CONCLUSION: This meta-analysis showed that the diagnostic performance of CRISPR is close to the gold standard, and it is expected to meet the Point of care requirements in resource poor areas.

Achieving Fast and Efficient K+ Intercalation on Ultrathin Graphene Electrodes Modified by a Li+ Based Solid-Electrolyte Interphase.

Advancing beyond Li-ion batteries requires translating the beneficial characteristics of Li+ electrodes to attractive, yet incipient, candidates such as those based on K+ intercalation. Here, we use ultrathin few-layer graphene (FLG) electrodes as a model interface to show a dramatic enhancement of K+ intercalation performance through a simple conditioning of the solid-electrolyte interphase (SEI) in a Li+ containing electrolyte. Unlike the substantial plating occurring in K+ containing electrolytes, we found that a Li+ based SEI enabled efficient K+ intercalation with discrete staging-type phase transitions observed via cyclic voltammetry at scan rates up to 100 mVs-1 and confirmed as ion-intercalation processes through in situ Raman spectroscopy. The resulting interface yielded fast charge-discharge rates up to ∼360C (1C is fully discharge in 1 h) and remarkable long-term cycling stability at 10C for 1000 cycles. This SEI promoted the transport of K+ as verified via mass spectrometric depth profiling. This work introduces a convenient strategy for improving the performance of ion intercalation electrodes toward a practical K-ion battery and FLG electrodes as a powerful analytical platform for evaluating fundamental aspects of ion intercalation.

High-Throughput Discovery and Characterization of Covalent Inhibitors for Protein Tyrosine Phosphatases.

Covalent inhibition has gained increasing interest in targeting the undruggable protein tyrosine phosphatases (PTPs). However, a systematic method for discovering and characterizing covalent PTP inhibitors has yet to be established. Here, we describe a workflow involving high-throughput screening of covalent fragment libraries and a novel biochemical assay that enables the acquisition of kinetics parameters of PTP inhibition by covalent inhibitors with higher throughput.

Application of exercise therapy in patients with chronic kidney disease-induced muscle atrophy: a scoping review.

BACKGROUND: The prevalence of muscle atrophy in patients suffering from chronic kidney disease (CKD) presents a significant challenge to healthcare providers, necessitating innovative approaches to management and care. Against this backdrop, this study embarks on a comprehensive review of literature concerning the application of exercise interventions in the nursing care of these patients. Such interventions are critical in addressing the debilitating effects of the condition, which include progressive loss of muscle mass and strength, adversely affecting patient mobility, quality of life, and overall survival. This review aims to identify the specific exercise modalities, contents, outcome indicators, and application effects associated with this intervention, in the context of the complex interplay of metabolic, inflammatory, and hormonal factors contributing to muscle wasting in CKD patients. By examining the efficacy of various exercise interventions, this study seeks to elucidate optimal strategies for mitigating the impact of CKD-induced muscle atrophy, thereby informing clinical practices and improving patient outcomes. METHODS: According to the method of a scoping review, nine databases (Cochrane, PubMed, EMBASE, Web of Science, ProQuest, Ovid, CNKI, Wanfang Data, and VIP) were searched until September 28, 2023. The included literature was screened, summarized, and analyzed. RESULTS: A total of 20 pieces of literature were included. Some types include aerobic exercise, resistance exercise, and aerobic combined resistance exercise. The exercise intensity primarily falls within the mild to moderate range, with a recommended frequency of 2 - 3 times a week, lasting 30 - 60 min each time. The types of outcomes encompassed in this study include body composition, functional testing, strength measurements, laboratory examinations, cardiopulmonary function assessments, and patient-reported outcomes. To varying degrees, exercise intervention positively impacts the subjects' physical activity ability, body composition, and skeletal muscle status. Currently, resistance training is the primary type of intervention used for preventing and treating CKD patients induced by muscle atrophy. CONCLUSION: Exercise intervention can improve muscle strength, physical function, and quality of life in patients with CKD muscle atrophy. Therefore, patients should be fully informed of the effect of exercise intervention in the treatment of chronic kidney disease-induced muscle atrophy in future, so as to promote the standardized implementation of exercise intervention.

Sleep, brain systems, and persistent stress in early adolescents during COVID-19: Insights from the ABCD study.

PURPOSE: The first year of the COVID-19 pandemic constituted a major life stress event for many adolescents, associated with disrupted school, behaviors, social networks, and health concerns. However, pandemic-related stress was not equivalent for everyone and could have been influenced by pre-pandemic factors including brain structure and sleep, which both undergo substantial development during adolescence. Here, we analyzed clusters of perceived stress levels across the pandemic and determined developmentally relevant pre-pandemic risk factors in brain structure and sleep of persistently high stress during the first year of the COVID-19 pandemic. METHODS: We investigated longitudinal changes in perceived stress at six timepoints across the first year of the pandemic (May 2020-March 2021) in 5559 adolescents (50 % female; age range: 11-14 years) in the United States (U.S.) participating in the Adolescent Brain Cognitive Development (ABCD) study. In 3141 of these adolescents, we fitted machine learning models to identify the most important pre-pandemic predictors from structural MRI brain measures and self-reported sleep data that were associated with persistently high stress across the first year of the pandemic. RESULTS: Patterns of perceived stress levels varied across the pandemic, with 5 % reporting persistently high stress. Our classifiers accurately detected persistently high stress (AUC > 0.7). Pre-pandemic brain structure, specifically cortical volume in temporal regions, and cortical thickness in multiple parietal and occipital regions, predicted persistent stress. Pre-pandemic sleep difficulties and short sleep duration were also strong predictors of persistent stress, along with more advanced pubertal stage. CONCLUSIONS: Adolescents showed variable stress responses during the first year of the COVID-19 pandemic, and some reported persistently high stress across the whole first year. Vulnerability to persistent stress was evident in several brain structural and self-reported sleep measures, collected before the pandemic, suggesting the relevance of other pre-existing individual factors beyond pandemic-related factors, for persistently high stress responses.

Mild steam treatment: Enhancing the rehydration performance of instant rice noodles by changing the physicochemical properties and gel structure of rice starch.

The "instant" quality of instant rice noodles is significantly affected by slow rehydration during cooking. This happens as a result of the native rice starch's low ability to gelatinize as well as the high shear and pressure utilized in industries during the widely used extrusion molding process. In order to address this issue, the rice flour was pretreated with mild steam (MS) technology. The results revealed that the rehydration qualities of the rice noodles that were extruded from the steam-treated flour significantly improved. There was a reduction of 25.5% in the rehydration time, from 443 to 330 s. The MS-treated rice starch's peak viscosity increased to 4503 from 4044 mPa/s. Decreases in gelatinization enthalpy (ΔH) and short-range ordering also suggest a reduction in difficulty in accomplishing starch gelatinization. Scanning electron microscopy studies showed particle aggregation increased as the treatment duration lasted longer. In conclusion, our findings indicate that we successfully addressed the issue of slow rehydration in instant rice noodles while presenting a novel approach for their manufacturing in the manufacturing sector.

Off-target autophagy inhibition by SHP2 allosteric inhibitors contributes to their anti-tumor activity in RAS-driven cancers.

Aberrant activation of RAS-MAPK signaling is common in cancer, and efforts to inhibit pathway components have yielded drugs with promising clinical activities. Unfortunately, treatment-provoked adaptive resistance mechanisms inevitably develop, limiting their therapeutic potential. As a central node essential for receptor tyrosine kinase mediated RAS activation, SHP2 has emerged as an attractive cancer target. Consequently, many SHP2 allosteric inhibitors are now in clinical testing. Here we discovered a previously unrecognized off-target effect associated with SHP2 allosteric inhibitors. We found that these inhibitors accumulate in the lysosome and block autophagic flux in a SHP2-independent manner. We showed that off-target autophagy inhibition by SHP2 allosteric inhibitors contributes to their anti-tumor activity. We also demonstrated that SHP2 allosteric inhibitors harboring this off-target activity not only suppress oncogenic RAS signaling but also overcome drug resistance such as MAPK rebound and protective autophagy in response to RAS-MAPK pathway blockage. Finally, we exemplified a therapeutic framework that harnesses both the on- and off-target activities of SHP2 allosteric inhibitors for improved treatment of mutant RAS driven and drug resistant malignancies such as pancreatic and colorectal cancers. Brief Summary: SHP2 allosteric inhibitors elicit off-target autophagy blockade that can be exploited for improved treatment of RAS-driven and drug-resistant cancers.