An improved transformer network for skin cancer classification.

BACKGROUND: Use of artificial intelligence to identify dermoscopic images has brought major breakthroughs in recent years to the early diagnosis and early treatment of skin cancer, the incidence of which is increasing year by year worldwide and poses a great threat to human health. Achievements have been made in the research of skin cancer image classification by using the deep backbone of the convolutional neural network (CNN). This approach, however, only extracts the features of small objects in the image, and cannot locate the important parts. OBJECTIVES: As a result, researchers of the paper turn to vision transformers (VIT) which has demonstrated powerful performance in traditional classification tasks. The self-attention is to improve the value of important features and suppress the features that cause noise. Specifically, an improved transformer network named SkinTrans is proposed. INNOVATIONS: To verify its efficiency, a three step procedure is followed. Firstly, a VIT network is established to verify the effectiveness of SkinTrans in skin cancer classification. Then multi-scale and overlapping sliding windows are used to serialize the image and multi-scale patch embedding is carried out which pay more attention to multi-scale features. Finally, contrastive learning is used which makes the similar data of skin cancer encode similarly so that the encoding results of different data are as different as possible. MAIN RESULTS: The experiment is carried out based on two datasets, namely (1) HAM10000: a large dataset of multi-source dermatoscopic images of common skin cancers; (2)A clinical dataset of skin cancer collected by dermoscopy. The model proposed has achieved 94.3% accuracy on HAM10000 and 94.1% accuracy on our datasets, which verifies the efficiency of SkinTrans. CONCLUSIONS: The transformer network has not only achieved good results in natural language but also achieved ideal results in the field of vision, which also lays a good foundation for skin cancer classification based on multimodal data. This paper is convinced that it will be of interest to dermatologists, clinical researchers, computer scientists and researchers in other related fields, and provide greater convenience for patients.

Value of a Signature of Immune-Related Genes in Predicting the Prognosis of Melanoma and Its Responses to Immune Checkpoint Blocker Therapies.

Melanoma is becoming increasingly common worldwide, with high rates of transformation into malignancy compared to other skin lesions. The prognosis of patients with melanoma at an advanced stage is highly unsatisfying despite the development of immunotherapy, target therapy, or combinative therapy. The major barrier to exploiting immune checkpoint therapies and achieving the best benefits clinically is resistance that can easily develop if regimens are not selected appropriately. In this study, we investigated the possibility of using immune-related genes to predict patient survival and their responses to immune checkpoint blocker therapies with the expression profiles available at The Cancer Genome Atlas (TCGA) Program plus expression data from the Gene Expression Omnibus (GEO) for validation. A five gene signature that is highly correlated with the local infiltration of cytotoxic lymphocytes in the tumor microenvironment was identified, and a scoring model was developed with stepwise regression after multivariate Cox analyses. The score calculated strongly correlates with Breslow depth, and this model effectively predicts the prognosis of patients with melanoma, whether primary or metastasized. It also depicts the heterogenous immune-related nature of melanoma by revealing different predicted responses to immune checkpoint blocker therapies through its correlation to tumor immune dysfunction and exclusion (TIDE) score.

The promoting effect of enteric materials on the oral absorption of larotaxel-loaded polymer-lipid hybrid nanoparticles.

Enteric polymers have been found with absorption promotion effect on nanoparticles. To study the role of enteric polymers played in the process of oral nanoparticle delivery, Eudragit L100-55 (EU) and sodium alginate (SA) were selected as model enteric polymers and larotaxel (LTX) as model drug. Suspensions composed of LTX-loaded nanoparticles, HPMC and different enteric polymers (EU and SA) were prepared (NP@EU, NP@SA). And aspects like precipitate morphology upon contact with acid, nanoparticle encapsulation capability, in vitro drug release, intestinal residence and in vivo oral bioavailability were studied. It was found that precipitates formed by EU could encapsulate more NP in acidic environment than those of SA (>95% of EU vs. approximately 70% of SA), and this difference in NP encapsulation was found correlated with the morphology of the precipitates formed: precipitates of EU appeared as three dimensional granules with dense inner structure, while SA precipitated into film-like porous structures. Results of pharmacokinetic study indicated that both EU and SA were capable in improving LTX absorption with absolute bioavailability of 77.1% and 42.5%, respectively. And the better absorption promoting effect of NP@EU was correlated with its longer intestinal residence shown by the results of ex vivo imaging study. In conclusion, both EU and SA could improve the oral bioavailability of LTX-loaded NP, and NP encapsulation capability and intestinal residence time are considered as key factors affecting the degree of absorption promotion.

Biopharmaceutical characters and bioavailability improving strategies of ginsenosides.

Deglycosylation is the most important gastrointestinal metabolism in which ginsenosides are split off from glycosyl moieties by the enzymes secreted from intestinal microflora, and two possible metabolic pathways of protopanaxdiol-type ginsenosides (PPD-type ginsenosides) and protopanaxtriol-type ginsenosides (PPT-type ginsenosides) have been concluded. The former is deglycosylated at C-3 and/or C-20, and transformed to protopanaxdiol (PPD). By comparison, the latter is deglycosylated at C-6 and/or C-20, and eventually transformed to protopanaxtriol (PPT) instead. The pharmacokinetic behavior of PPD-type ginsenosides and PPT-type ginsenosides is different, mainly in a faster absorption and elimination rate of PPT-type ginsenosides, but almost all of ginsenosides have a low oral bioavailability, which is relevant to the properties, the stability in the gastrointestinal tract, membrane permeability and the intestinal and hepatic first-pass effect of ginsenosides. Fortunately, its bioavailability can be improved by means of pharmaceutical strategies, including nanoparticles, liposomes, emulsions, micelles, etc. These drug delivery systems can significantly increase the bioavailability of ginsenosides, as well as controlling or targeting drug release. Ginsenosides are widely used in the treatment of various diseases, the most famous one is the Shen Yi capsule, which is the world's first clinical application of tumor neovascularization inhibitors. Hence, this article aims to draw people's attention on ocotillol-type ginsenosides, which have prominent anti-Alzheimer's disease activity, but have been overlooked previously, such as its representative compound-Pseudoginsenoside F11(PF11), and then provide a reference for the druggability and further developments of ocotillol-type ginsenosides by utilizing the homogeneous structure between dammarane-type ginsenosides and ocotillol-type ginsenosides.

Disulfiram-loaded mixed nanoparticles with high drug-loading and plasma stability by reducing the core crystallinity for intravenous delivery.

To develop an injectable formulation and improve the stability of disulfiram (DSF), DSF was encapsulated into mixed nanoparticles (DSF-NPs) through a high-pressure homogenization method. The Flory-Huggins interaction parameters (χFH) were calculated to predict the miscibility between DSF and the hydrophobic core, resulting in PCL5000 selected as the hydrophobic block to encapsulate the DSF, as PCL5000 had a lower χFH 3.39 and the drug loading of the nanoparticles prepared by mPEG5000-PCL5000 was relatively higher. mPEG5000-PCL5000 and PCL5000 were blended to reduce the leakage of DSF during preparation, as well as increase the stability of the nanoparticles. The cargo-loading capacity of the nanoparticles was improved from 3.35% to 5.50% by reducing the crystallinity of the PCL nanoparticle core, and the crystallinity decreased from 51.13% to 25.15% after adding medium chain triglyceride (MCT). The DSF-NPs prepared by the above method had a small particle size of 98.1 ±â€¯10.54 nm, with a polydispersity index (PDI) of 0.036, as well as drug loading of 5.50%. Furthermore, DSF-NPs containing MCT showed higher stability than DSF-NPs without MCT and DSF-sol (DSF dissolved in Cremophor EL and ethanol) in water and 90% plasma-containing PBS. The pharmacokinetics proved that DSF-NPs containing MCT enhanced the DSF concentration in the blood. Finally, DSF-NPs effectively inhibited H22 xenograft tumor growth in vivo.

mPEG5k- b-PLGA2k/PCL3.4k/MCT Mixed Micelles as Carriers of Disulfiram for Improving Plasma Stability and Antitumor Effect in Vivo.

The clinical application of disulfiram (DSF) in cancer treatments is hindered by its rapid degradation in the blood circulation. In this study, methoxy poly(ethylene glycol)- b-poly(lactide- co-glycolide)/poly(ε-caprolactone) (mPEG5k- b-PLGA2k/PCL3.4k) micelles were developed for encapsulation of DSF by using the emulsification-solvent diffusion method. Medium chain triglyceride (MCT) was incorporated into the mixed polymeric micelles to improve drug loading by reducing the core crystallinity. Differential scanning calorimetry (DSC) results implied that DSF is likely present in an amorphous form within the micelles, and is well dispersed. DSF is encapsulated within the core and the reservoir is stabilized by the hydrophilic shell to prevent rapid diffusion of DSF from the core. The DSF mixed micelles (DSF-MMs) showed good drug loading (5.90%) and a well-controlled particle size (86.4 ± 13.2 nm). The mixed micelles efficiently protected DSF from degradation in plasma, with 58% remaining after 48 h, while almost 90% of DSF was degraded after the same period for the DSF solution (DSF-sol), which was used as a control. The pharmacokinetics study showed that the maximum plasma concentration and bioavailability of DSF were improved by using the DSF-MMs (2 and 2.5 times that of the DSF-sol). The TIRs (tumor inhibition rates) of 5-FU, DSF-sol, and DSF-MMs were 63.46, 19.57, and 69.98%, respectively, implying that DSF-MMs slowed the growth of a H22 xenograft tumor model effectively.

A parenteral docetaxel-loaded lipid microsphere with decreased 7-epidocetaxel conversion in vitro and in vivo.

The purpose of the study was to develop a parenteral docetaxel lipid microsphere to inhibit its 7-epidocetaxel conversion in vitro and in vivo. 7-epidocetaxel conversion as the main indicator was investigated to optimize the formulation and process. 10% medium-chain triglyceride/long-chain triglyceride (3:1) as the oil phase, egg lecithin E80 as the emulsifier and 0.02% NaHSO3 as the acidity regulator were selected to prepare docetaxel lipid microsphere. This study found that pH and temperature were dominant factors on the epimerization of docetaxel in lipid microsphere, and that optimum conditions were a pH of 5.3 and thermal sterilization conditions of 121°Cautoclaving for 8min. According to the degradation kinetics, docetaxel lipid microsphere had a wider pH range where 7-epidocetaxel(%) stayed at low levels than Docetaxel for Injection, and might improve the docetaxel stability by loading drug in lecithin layer instead of altering the degradation mechanism. Docetaxel lipid microsphere decreased epimerization in plasma in vitro obviously. Pharmacokinetics of docetaxel and 7-epidocetaxel were investigated to quantify the 7-epidocetaxel conversion in vivo. The resulrs indicated that there was less conversion of docetaxel in lipid microspheres than in Docetaxel for Injection. The convert ratios were 0.61% and 3.04% respectively. In conclusion, lipid microsphere is a promising delivery system for intravenous administration of docetaxel with decreased 7-epidocetaxel conversion.

Improved tumor tissue penetration and tumor cell uptake achieved by delayed charge reversal nanoparticles.

The high affinity of positively charged nanoparticles to biological interfaces makes them easily taken up by tumor cells but limits their tumor permeation due to non-specific electrostatic interactions. In this study, polyion complex coated nanoparticles with different charge reversal profiles were developed to study the influence of charge reversal profile on tumor penetration. The system was constructed by polyion complex coating using micelles composed of poly (lysine)-b-polycaprolactone (PLys-b-PCL) as the cationic core and poly (glutamic acid)-g- methoxyl poly (ethylene glycol) (PGlu-g-mPEG) as the anionic coating material. Manipulation of charge reversal profile was achieved by controlling the polymer chain entanglement and electrostatic interaction in the polyion complex layer through glutaraldehyde-induced shell-crosslinking. The delayed charge reversal nanoparticles (CTCL30) could maintain negatively charged in pH 6.5 PBS for at least 2h and exhibit pH-responsive cytotoxicity and cellular uptake in an extended time scale. Compared with a faster charge reversal counterpart (CTCL70) with similar pharmacokinetic profile, CTCL30 showed deeper penetration, higher in vivo tumor cell uptake and stronger antitumor activity in vivo (tumor inhibition rate: 72.3% vs 60.2%, compared with CTCL70). These results indicate that the delayed charge reversal strategy could improve therapeutic effect via facilitating tumor penetration. STATEMENT OF SIGNIFICANCE: Here, the high tumor penetration capability of PEG-coated nanoparticles and the high cellular uptake of cationic nanoparticles were combined by a delayed charge reversal drug delivery system. This drug delivery system was composed of a drug-loading cationic inner core and a polyion complex coating. Manipulation of charge reversal profile was realized by varying the crosslinking degree of the shell of the cationic inner core, through which changed the strength of the polyion complex layer. Nanoparticles with delayed charge reversal profile exhibited improved tumor penetration, in vivo tumor cell uptake and in vivo tumor growth inhibition effect although they have similar pharmacokinetic and biodistribution behaviors with their instant charge reversal counterpart.