A nomogram based on CT intratumoral and peritumoral radiomics features preoperatively predicts poorly differentiated invasive pulmonary adenocarcinoma manifesting as subsolid or solid lesions: a double-center study.

Background: The novel International Association for the Study of Lung Cancer (IASLC) grading system suggests that poorly differentiated invasive pulmonary adenocarcinoma (IPA) has a worse prognosis. Therefore, prediction of poorly differentiated IPA before treatment can provide an essential reference for therapeutic modality and personalized follow-up strategy. This study intended to train a nomogram based on CT intratumoral and peritumoral radiomics features combined with clinical semantic features, which predicted poorly differentiated IPA and was tested in independent data cohorts regarding models' generalization ability. Methods: We retrospectively recruited 480 patients with IPA appearing as subsolid or solid lesions, confirmed by surgical pathology from two medical centers and collected their CT images and clinical information. Patients from the first center (n =363) were randomly assigned to the development cohort (n = 254) and internal testing cohort (n = 109) in a 7:3 ratio; patients (n = 117) from the second center served as the external testing cohort. Feature selection was performed by univariate analysis, multivariate analysis, Spearman correlation analysis, minimum redundancy maximum relevance, and least absolute shrinkage and selection operator. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate the model performance. Results: The AUCs of the combined model based on intratumoral and peritumoral radiomics signatures in internal testing cohort and external testing cohort were 0.906 and 0.886, respectively. The AUCs of the nomogram that integrated clinical semantic features and combined radiomics signatures in internal testing cohort and external testing cohort were 0.921 and 0.887, respectively. The Delong test showed that the AUCs of the nomogram were significantly higher than that of the clinical semantic model in both the internal testing cohort(0.921 vs 0.789, p< 0.05) and external testing cohort(0.887 vs 0.829, p< 0.05). Conclusion: The nomogram based on CT intratumoral and peritumoral radiomics signatures with clinical semantic features has the potential to predict poorly differentiated IPA manifesting as subsolid or solid lesions preoperatively.

Risk assessment of pneumothorax in colorectal lung metastases treated by percutaneous thermal ablation: a multicenter retrospective cohort study.

PURPOSE: To evaluate the risk of pneumothorax in the percutaneous image-guided thermal ablation (IGTA) treatment of colorectal lung metastases (CRLM). METHODS: Data regarding patients with CRLM treated with IGTA from five medical institutions in China from 2016 to 2023 were reviewed retrospectively. Pneumothorax and non-pneumothorax were compared using the Student's t -test, χ 2 test and Fisher's exact test. Univariate logistic regression analysis was conducted to identify potential risk factors, followed by multivariate logistic regression analysis to evaluate the predictors of pneumothorax. Interactions between variables were examined and used for model construction. Receiver operating characteristic curves and nomograms were generated to assess the performance of the model. RESULTS: A total of 254 patients with 376 CRLM underwent 299 ablation sessions. The incidence of pneumothorax was 45.5%. The adjusted multivariate logistic regression model, incorporating interaction terms, revealed that tumour number [odds ratio (OR)=8.34 (95% CI: 1.37-50.64)], puncture depth [OR=0.53 (95% CI: 0.31-0.91)], pre-procedure radiotherapy [OR=3.66 (95% CI: 1.17-11.40)], peribronchial tumour [OR=2.32 (95% CI: 1.04-5.15)], and emphysema [OR=56.83 (95% CI: 8.42-383.57)] were significant predictive factors of pneumothorax (all P <0.05). The generated nomogram model demonstrated a significant prediction performance, with an area under the receiver operating characteristic curve of 0.800 (95% CI: 0.751-0.850). CONCLUSIONS: Pre-procedure radiotherapy, tumour number, peribronchial tumour, and emphysema were identified as risk factors for pneumothorax in the treatment of CRLM using percutaneous IGTA. Puncture depth was found to be a protective factor against pneumothorax.

Discrimination of ground-glass nodular lung adenocarcinoma pathological subtypes via transfer learning: A multicenter study.

BACKGROUND: The surgical approach and prognosis for invasive adenocarcinoma (IAC) and minimally invasive adenocarcinoma (MIA) of the lung differ. However, they both manifest as identical ground-glass nodules (GGNs) in computed tomography images, and no effective method exists to discriminate them. METHODS: We developed and validated a three-dimensional (3D) deep transfer learning model to discriminate IAC from MIA based on CT images of GGNs. This model uses a 3D medical image pre-training model (MedicalNet) and a fusion model to build a classification network. Transfer learning was utilized for end-to-end predictive modeling of the cohort data of the first center, and the cohort data of the other two centers were used as independent external validation data. This study included 999 lung GGN images of 921 patients pathologically diagnosed with IAC or MIA at three cohort centers. RESULTS: The predictive performance of the model was assessed using the area under the receiver operating characteristic curve (AUC). The model had high diagnostic efficacy for the training and validation groups (accuracy: 89%, sensitivity: 95%, specificity: 84%, and AUC: 95% in the training group; accuracy: 88%, sensitivity: 84%, specificity: 93%, and AUC: 92% in the internal validation group; accuracy: 83%, sensitivity: 83%, specificity: 83%, and AUC: 89% in one external validation group; accuracy: 78%, sensitivity: 80%, specificity: 77%, and AUC: 82% in the other external validation group). CONCLUSIONS: Our 3D deep transfer learning model provides a noninvasive, low-cost, rapid, and reproducible method for preoperative prediction of IAC and MIA in lung cancer patients with GGNs. It can help clinicians to choose the optimal surgical strategy and improve the prognosis of patients.

A Sequential Stein's Method for Faster Training of Additive Index Models.

The additive index models (AIMs) can be viewed as a kind of artificial neural networks based on nonparametric activation or so-called ridge functions. Recently, they are shown to achieve enhanced explainability after incorporating various interpretability constraints. However, the training of AIMs by either the backfitting algorithm or the joint stochastic optimization is known to be very slow for especially high dimensional inputs. In this article, we propose a novel sequential approach based on the celebrated Stein's lemma. The proposed SeqStein method can successfully decouple the training of AIMs into two separable steps, namely, the following: 1) Stein's estimation of the projection indices and 2) nonparametric estimation of ridge functions using the smoothing splines. We show through numerical experiments that the SeqStein algorithm is not only more efficient for training AIMs, but also inclined to produce more interpretable models that have smooth ridge functions with sparse and nearly orthogonal projection indices.

Preoperative prediction of tumor deposits in rectal cancer with clinical-magnetic resonance deep learning-based radiomic models.

Background: This study aimed to establish an effective model for preoperative prediction of tumor deposits (TDs) in patients with rectal cancer (RC). Methods: In 500 patients, radiomic features were extracted from magnetic resonance imaging (MRI) using modalities such as high-resolution T2-weighted (HRT2) imaging and diffusion-weighted imaging (DWI). Machine learning (ML)-based and deep learning (DL)-based radiomic models were developed and integrated with clinical characteristics for TD prediction. The performance of the models was assessed using the area under the curve (AUC) over five-fold cross-validation. Results: A total of 564 radiomic features that quantified the intensity, shape, orientation, and texture of the tumor were extracted for each patient. The HRT2-ML, DWI-ML, Merged-ML, HRT2-DL, DWI-DL, and Merged-DL models demonstrated AUCs of 0.62 ± 0.02, 0.64 ± 0.08, 0.69 ± 0.04, 0.57 ± 0.06, 0.68 ± 0.03, and 0.59 ± 0.04, respectively. The clinical-ML, clinical-HRT2-ML, clinical-DWI-ML, clinical-Merged-ML, clinical-DL, clinical-HRT2-DL, clinical-DWI-DL, and clinical-Merged-DL models demonstrated AUCs of 0.81 ± 0.06, 0.79 ± 0.02, 0.81 ± 0.02, 0.83 ± 0.01, 0.81 ± 0.04, 0.83 ± 0.04, 0.90 ± 0.04, and 0.83 ± 0.05, respectively. The clinical-DWI-DL model achieved the best predictive performance (accuracy 0.84 ± 0.05, sensitivity 0.94 ± 0. 13, specificity 0.79 ± 0.04). Conclusions: A comprehensive model combining MRI radiomic features and clinical characteristics achieved promising performance in TD prediction for RC patients. This approach has the potential to assist clinicians in preoperative stage evaluation and personalized treatment of RC patients.

Proteomic analysis of small extracellular vesicles from the plasma of patients with hepatocellular carcinoma.

PURPOSE: Liver cancer is one of the most common tumors with the seventh-highest incidence and the third-highest mortality. Many studies have shown that small extracellular vesicles (sEVs) play an important role in liver cancer. Here, we report comprehensive signatures for sEV proteins from plasma obtained from patients with hepatocellular carcinoma (HCC), which might be valuable for the evaluation and diagnosis of HCC. METHODS: We extracted sEVs from the plasma of controls and patients with HCC. Differentially expressed proteins in the sEVs were analyzed using label-free quantification and bioinformatic analyses. Western blotting (WB) was used to validate the abovementioned sEV proteins. RESULTS: Proteomic analysis was performed for plasma sEVs from 21 patients with HCC and 15 controls. Among the 335 identified proteins in our study, 27 were significantly dysregulated, including 13 upregulated proteins that were involved predominantly in the complement cascade (complement C1Q subcomponent subunit B (C1QB), complement C1Q subcomponent subunit C (C1QC), C4B-binding protein alpha chain (C4BPA), and C4B-binding protein beta chain (C4BPB)) and the coagulation cascade (F13B, fibrinogen alpha chain (FGA), fibrinogen beta chain (FGB), and fibrinogen gamma chain (FGG)). We verified increased levels of the C1QB, C1QC, C4BPA, and C4BPB proteins in the plasma sEVs from patients with HCC in both the discovery cohort and validation cohort. CONCLUSIONS: The complement cascade in sEVs was significantly involved in HCC progression. C1QB, C1QC, C4BPA, and C4BPB were highly abundant in the plasma sEVs from patients with HCC and might represent molecular signatures.

Development characteristics of the rock fracture field in strata overlying a mined coal seam group.

The fracture development of the overlying strata after coal mining is an important guarantee of efficient gas drainage. In order to explore the fracture evolution characteristics close to a mined coal seam group, the F15.16-24130 working face in the Pingdingshan No. 10 coal mine was taken as the research background. The FLAC3D numerical simulation software was used to study the migration and failure characteristics of the overlying strata during mining of a coal seam group, and the fracture evolution process of the stope was investigated. The results show that as the advancing distance increased, the fracture density and fracture height increased continuously due to deformation and failure of the overlying rock. The displacement of the overlying rock initially increased and then decreased, and the displacement of the floor rock initially decreased and then increased. When working face F15.16 of the coal seam advanced to 75 m, a saddle-shaped plastic zone gradually formed in the upper part of the goaf and the floor of the goaf was formed. The pressure relief depth was proportional to the advancement distance. As the advancement distance of the working face increased, the pressure relief depth gradually extended to the F17 coal seam, which was conducive to the development and penetration of the fractures in the coal floor and rock mass and was convenient for pressure relief gas drainage from the F17 coal seam.

S1PR1 induces metabolic reprogramming of ceramide in vascular endothelial cells, affecting hepatocellular carcinoma angiogenesis and progression.

Angiogenesis is a fundamental process underlying the occurrence, growth and metastasis of hepatocellular carcinoma (HCC), a prevalent tumour type with an extremely poor prognosis due to abundant vasculature. However, the underlying mechanism of angiogenesis in HCC remains largely unknown. Herein, we found that sphingosine-1-phosphate receptor 1 (S1PR1) plays an important role in HCC angiogenesis. S1PR1 was found to be selectively and highly expressed in the blood vessels of HCC tissues compared with those of paratumour tissues. Functionally, high expression of S1PR1 in endothelial cells (ECs) promoted angiogenesis and progression of HCC in vitro and in vivo. Mechanistically, proangiogenic factors (S1P, IL-6, VEGFA) in conditioned medium from HCC cells induced the upregulation of S1PR1 in ECs via the phosphorylation of STAT3 at Y705. Further study also revealed that S1PR1 promotes angiogenesis by decreasing ceramide levels via CerS3 downregulation. Interestingly, we demonstrated that S1PR1 downregulates CerS3 by inducing CerS6 translocation into the nucleus to inhibit CerS3 at the transcriptional level in ECs. In addition, we found that a high concentration of Lenvatinib significantly downregulated the expression of S1PR1 and obviously enhanced S1PR1 knockdown-mediated angiogenesis inhibition, indicating that S1PR1 may be a target by which Lenvatinib combats angiogenesis in HCC. Thus, S1PR1 may be an important target for suppressing angiogenesis in HCC, and inhibiting S1PR1 is a promising approach to antitumor therapy in HCC.

An Endogenous H2S-Activated Nanoplatform for Triple Synergistic Therapy of Colorectal Cancer.

Overproduced hydrogen sulfide (H2S) is a highly potential target for precise colorectal cancer (CRC) therapy; herein, a novel 5-Fu/Cur-P@HMPB nanomedicine is developed by coencapsulation of the natural anticancer drug curcumin (Cur) and the clinical chemotherapeutic drug 5-fluorouracil (5-Fu) into hollow mesoporous Prussian blue (HMPB). HMPB with low Fenton-catalytic activity can react with endogenous H2S and convert into high Fenton-catalytic Prussian white (PW), which can generate in situ a high level of •OH to activate chemodynamic therapy (CDT) and meanwhile trigger autophagy. Importantly, the autophagy can be amplified by Cur to induce autophagic cell death; moreover, Cur also acted as a specific chemosensitizer of the chemotherapy drug 5-Fu, achieving a good synergistic antitumor effect. Such a triple synergistic therapy based on a novel nanomedicine has been verified both in vitro and in vivo to have high efficacy in CRC treatment, showing promising potential in translational medicine.

Predictive Efficacy of a Radiomics Random Forest Model for Identifying Pathological Subtypes of Lung Adenocarcinoma Presenting as Ground-Glass Nodules.

Purpose: To establish and verify the ability of a radiomics prediction model to distinguish invasive adenocarcinoma (IAC) and minimal invasive adenocarcinoma (MIA) presenting as ground-glass nodules (GGNs). Methods: We retrospectively analyzed 118 lung GGN images and clinical data from 106 patients in our hospital from March 2016 to April 2019. All pathological classifications of lung GGN were confirmed as IAC or MIA by two pathologists. R language software (version 3.5.1) was used for the statistical analysis of the general clinical data. ITK-SNAP (version 3.6) and A.K. software (Analysis Kit, American GE Company) were used to manually outline the regions of interest of lung GGNs and collect three-dimensional radiomics features. Patients were randomly divided into training and verification groups (ratio, 7:3). Random forest combined with hyperparameter tuning was used for feature selection and prediction modeling. The receiver operating characteristic curve and the area under the curve (AUC) were used to evaluate model prediction efficacy. The calibration curve was used to evaluate the calibration effect. Results: There was no significant difference between IAC and MIA in terms of age, gender, smoking history, tumor history, and lung GGN location in both the training and verification groups (P>0.05). For each lung GGN, the collected data included 396 three-dimensional radiomics features in six categories. Based on the training cohort, nine optimal radiomics features in three categories were finally screened out, and a prediction model was established. We found that the training group had a high diagnostic efficacy [accuracy, sensitivity, specificity, and AUC of the training group were 0.89 (95%CI, 0.73 - 0.99), 0.98 (95%CI, 0.78 - 1.00), 0.81 (95%CI, 0.59 - 1.00), and 0.97 (95%CI, 0.92-1.00), respectively; those of the validation group were 0.80 (95%CI, 0.58 - 0.93), 0.82 (95%CI, 0.55 - 1.00), 0.78 (95%CI, 0.57 - 1.00), and 0.92 (95%CI, 0.83 - 1.00), respectively]. The model calibration curve showed good consistency between the predicted and actual probabilities. Conclusions: The radiomics prediction model established by combining random forest with hyperparameter tuning effectively distinguished IAC from MIA presenting as GGNs and represents a noninvasive, low-cost, rapid, and reproducible preoperative prediction method for clinical application.

GBA1-dependent membrane glucosylceramide reprogramming promotes liver cancer metastasis via activation of the Wnt/ß-catenin signalling pathway.

The effect of glucosylceramide (GlcCer) reprogramming on liver cancer metastasis remains poorly understood. In this study, we demonstrated that the protein expression of GBA1, which catalyses the conversion of GlcCer to ceramide, was downregulated in liver cancer tissue. A clinical relevance analysis revealed that low expression of GBA1 was associated with the metastatic potential of liver cancer cells. Furthermore, loss- and gain-of-function studies confirmed that low expression of GBA1 promoted metastasis of liver cancer both in vitro and in vivo. Mechanistic studies indicated that low expression of GBA1 enhanced the metastatic ability of liver cancer by promoting the epithelial-mesenchymal transition (EMT), in which Wnt signalling pathway is involved. In the plasma membrane (PM), GBA1-dependent GlcCer reprogramming increased LRP6 location in the PM leading to an interaction between GlcCer and LRP6, subsequently promoting LRP6 phosphorylation at Ser1490, and finally activating the Wnt/ß-catenin signalling pathway. To our knowledge, this is the first time to be found that GlcCer interacted with a protein. In addition, the results of mass spectrometry indicated that GlcCer d18:1/18:0 was the most notably increased studied species in the PM when GBA1 was downregulated, suggesting that GlcCer d18:1/18:0 may be the major functional lipid that promotes GBA1-dependent liver cancer metastasis. Thus, GBA1-mediated GlcCer reprogramming in the PM promotes metastasis of liver cancer via activation of the Wnt/ß-catenin signalling pathway, upregulation of GBA1 may be a potential therapeutic strategy to combat liver cancer metastasis.

Yolk Sac Tumor in an Infant with Androgen Insensitivity Syndrome: A Case Report and Review of the Literature.

Background: Androgen insensitivity syndrome (AIS) is a disorder of sexual differentiation caused by complete or partial resistance to the biological action of androgens. The common malignant tumors associated with this syndrome are seminomas. However, the risk of malignancy in childhood remains low. Case Report: A 8-month-old child with a female phenotype and a 46, XY karyotype, presented with bilateral inguinal hernia. The patient underwent right radical inguinal orchiectomy with high ligation of the spermatic cord and laparoscopic percutaneous extra-peritoneal herniorrhaphy. Final pathology confirmed a pure yolk sac tumor (YST) from the right testis. Androgen receptor (AR) gene mutation was found in the children. The follow-up ultrasonography shown no recurrence, with serum AFP returned to normal within 3 months. Conclusion: The case we presented is relatively infrequent in the literature with yolk sac tumor in a AIS children presented with a palpable lump inguinal region.

Speed sensorless control of a bearingless induction motor based on sliding mode observer and phase-locked loop.

To realize the low-cost and high-performance of a bearingless induction motor (BIM), a speed sensorless control strategy combined with the improved sliding mode observer (SMO) and phase locked-loop (PLL) is proposed. Based on analyzing the principle of the traditional SMO, the improved SMO adopts a switching function that has a double boundary layer structure to reduce the high-frequency jitter. Firstly, the observation equation of the rotor flux is established where the stator current, as well as the rotor flux are two state variables, and then the system stability is discussed with the Lyapunov theory. Secondly, to reduce the rotor chatter and enhance the speed tracking performance a flux electrical angle detecting module is constructed based on the PLL structure that extracts the rotor flux electrical angle and speed from the observed flux component. Finally, the BIM speed sensorless vector control system is established and the simulation results and experimental results are respectively presented by the MATLAB/Simulink simulation platform and a laboratory prototype. Both results prove that this strategy can not only enhance the speed self-detection capability and tracking performance, but can guarantee the excellent self-suspension function as well.

Reshaping the Tumor Immune Microenvironment Based on a Light-Activated Nanoplatform for Efficient Cancer Therapy.

The immunosuppressive tumor microenvironment (TME) always causes poor antitumor immune efficacy, prone to relapse and metastasis. Herein, novel poly(vinylpyrrolidone) (PVP) modified BiFeO3 /Bi2 WO6 (BFO/BWO) with a p-n type heterojunction is constructed for reshaping the immunosuppressive TME. Reactive oxygen species can be generated under light activation by the well-separated hole (h+ )-electron (e- ) pairs owing to the heterojunction in BFO/BWO-PVP NPs. Interestingly, h+ can trigger the decomposition of H2 O2 to generate O2 for alleviating tumor hypoxia, which not only sensitizes photodynamic therapy (PDT) and radiotherapy (RT), but also promotes tumor-associated macrophages (TAMs) polarization from M2 to M1 phenotype, which is beneficial to decrease the expression of HIF-1α. Importantly, such a light-activated nanoplatform, combining with RT can efficiently activate and recruit cytotoxic T lymphocytes to infiltrate in tumor tissues, as well as stimulate TAMs to M1 phenotype, dramatically reverse the immunosuppressive TME into an immunoactive one, and further boost immune memory responses. Moreover, BFO/BWO-PVP NPs also present high performance for computed tomography imaging contrast. Taken together, this work offers a novel paradigm for achieving O2 self-supply of inorganic nanoagents and reshaping of the tumor immune microenvironment for effective inhibition of cancer as well as metastasis and recurrence.

An effective SteinGLM initialization scheme for training multi-layer feedforward sigmoidal neural networks.

Network initialization is the first and critical step for training neural networks. In this paper, we propose a novel network initialization scheme based on the celebrated Stein's identity. By viewing multi-layer feedforward sigmoidal neural networks as cascades of multi-index models, the projection weights to the first hidden layer are initialized using eigenvectors of the cross-moment matrix between the input's second-order score function and the response. The input data is then forward propagated to the next layer and such a procedure can be repeated until all the hidden layers are initialized. Finally, the weights for the output layer are initialized by generalized linear modeling. Such a proposed SteinGLM method is shown through extensive numerical results to be much faster and more accurate than other popular methods commonly used for training neural networks.

Enhancing Explainability of Neural Networks Through Architecture Constraints.

Prediction accuracy and model explainability are the two most important objectives when developing machine learning algorithms to solve real-world problems. Neural networks are known to possess good prediction performance but suffer from a lack of model interpretability. In this article, we propose to enhance the explainability of neural networks through the following architecture constraints: 1) sparse additive subnetworks; 2) projection pursuit with orthogonality constraint; and 3) smooth function approximation. It leads to an enhanced explainable neural network (ExNN) with a superior balance between prediction performance and model interpretability. We derive sufficient identifiability conditions for the proposed ExNN model. The multiple parameters are simultaneously estimated by a modified minibatch gradient descent method based on the backpropagation algorithm for calculating the derivatives and the Cayley transform for preserving the projection orthogonality. Through simulation study under six different scenarios, we compare the proposed method to several benchmarks, including least absolute shrinkage and selection operator, support vector machine, random forest, extreme learning machine, and multilayer perceptron. It is shown that the proposed ExNN model keeps the flexibility of pursuing high prediction accuracy while attaining improved interpretability. Finally, a real data example is employed as a showcase application.

Framework Fe-Doped Mobil Five (MFI) Zeolites as Highly Active and Stable Fenton-Like Catalysts for Basic Dyes Degradation.

Highly active and stable framework Fe-doped ZSM-5 (f-Fe-ZSM-5) zeolites with different Fe contents, which were synthesized using a facile one-pot hydrothermal method, could effectively resolve the loss of iron element during the catalytic degradation of basic dyes. The successful introduction of Fe species into the framework of ZSM-5 was confirmed by elemental mappings, Fourier transform infrared (FT-IR) spectroscopy and Ultraviolet-visible spectroscopy (UV-vis spectra). The operational parameters, such as Fe content, H2O2 concentration, reaction temperature, types of dyes as well as the stability of the synthesized samples were extensively evaluated. It was demonstrated that the f-Fe(0.10)-ZSM-5 exhibited an efficient catalytic ability and excellent stability even after seven consecutive runs. The degradation efficiency of f-Fe-ZSM-5 for basic dyes was higher than that for acid dye. Therefore, f-Fe-ZSM-5 zeolites may present major potential for the treatment of basic dyes waste water without adjusting the initial pH value.

Organic Semiconducting Pro-nanostimulants for Near-Infrared Photoactivatable Cancer Immunotherapy.

In this study, an organic semiconducting pro-nanostimulant (OSPS) with a near-infrared (NIR) photoactivatable immunotherapeutic action for synergetic cancer therapy is presented. OSPS comprises a semiconducting polymer nanoparticle (SPN) core and an immunostimulant conjugated through a singlet oxygen (1 O2 ) cleavable linkers. Upon NIR laser irradiation, OSPS generates both heat and 1 O2 to exert combinational phototherapy not only to ablate tumors but also to produce tumor-associated antigens. More importantly, NIR irradiation triggers the cleavage of 1 O2 -cleavable linkers, triggering the remote release of the immunostimulants from OSPS to modulate the immunosuppressive tumor microenvironment. Thus, the released tumor-associated antigens in conjunction with activated immunostimulants induce a synergistic antitumor immune response after OSPS-mediated phototherapy, resulting in the inhibited growth of both primary/distant tumors and lung metastasis in a mouse xenograft model, which is not observed for sole phototherapy.