A deep learning-based radiomics approach to predict head and neck tumor regression for adaptive radiotherapy.

Early regression-the regression in tumor volume during the initial phase of radiotherapy (approximately 2 weeks after treatment initiation)-is a common occurrence during radiotherapy. This rapid radiation-induced tumor regression may alter target coordinates, necessitating adaptive radiotherapy (ART). We developed a deep learning-based radiomics (DLR) approach to predict early head and neck tumor regression and thereby facilitate ART. Primary gross tumor volume (GTVp) was monitored in 96 patients and nodal GTV (GTVn) in 79 patients during treatment. All patients underwent two computed tomography (CT) scans: one before the start of radiotherapy for initial planning and one during radiotherapy for boost planning. Patients were assigned to regression and nonregression groups according to their median tumor regression rate (ΔGTV/treatment day from initial to boost CT scan). We input a GTV image into the convolutional neural network model, which was pretrained using natural image datasets, via transfer learning. The deep features were extracted from the last fully connected layer. To clarify the prognostic power of the deep features, machine learning models were trained. The models then predicted the regression and nonregression of GTVp and GTVn and evaluated the predictive performance by 0.632 + bootstrap area under the curve (AUC). Predictive performance for GTVp regression was highest using the InceptionResNetv2 model (mean AUC = 0.75) and that for GTVn was highest using NASNetLarge (mean AUC = 0.73). Both models outperformed the handcrafted radiomics features (mean AUC = 0.63 for GTVp and 0.61 for GTVn) or clinical factors (0.64 and 0.67, respectively). DLR may facilitate ART for improved radiation side-effects and target coverage.

Assessment of a computed tomography-based radiomics approach for assessing lung function in lung cancer patients.

PURPOSE: We aimed to assess radiomics approaches for estimating three pulmonary function test (PFT) results (forced expiratory volume in one second [FEV1], forced vital capacity [FVC], and the ratio of FEV1 to FVC [FEV1/FVC]) using data extracted from chest computed tomography (CT) images. METHODS: This retrospective study included 85 lung cancer patients (mean age, 75 years ±8; 69 men) who underwent stereotactic body radiotherapy between 2012 and 2020. Their pretreatment chest breath-hold CT and PFT data before radiotherapy were obtained. A total of 107 radiomics features (Shape: 14, Intensity: 18, Texture: 75) were extracted using two methods: extraction of the lung tissue (<-250 HU) (APPROACH 1), and extraction of small blood vessels and lung tissue (APPROACH 2). The PFT results were estimated using the least absolute shrinkage and selection operator regression. Pearson's correlation coefficients (r) were determined for all PFT results, and the area under the curve (AUC) was calculated for FEV1/FVC (<70 %). Finally, we compared our approaches with the conventional formula (Conventional). RESULTS: For the estimated FEV1/FVC, the Pearson's r were 0.21 (P =.06), 0.69 (P <.01), and 0.73 (P <.01) for Conventional, APPROACH 1, and APPROACH 2, respectively; the AUCs for FEV1/FVC (<70 %) were 0.67 (95 % confidence interval [CI]: 0.55, 0.79), 0.82 (CI: 0.72, 0.91; P =.047) and 0.86 (CI: 0.78, 0.94; P =.01), respectively. CONCLUSIONS: The radiomics approach performed better than the conventional equation and may be useful for assessing lung function based on CT images.

Impact of feature selection methods and subgroup factors on prognostic analysis with CT-based radiomics in non-small cell lung cancer patients.

BACKGROUND: Radiomics is a new technology to noninvasively predict survival prognosis with quantitative features extracted from medical images. Most radiomics-based prognostic studies of non-small-cell lung cancer (NSCLC) patients have used mixed datasets of different subgroups. Therefore, we investigated the radiomics-based survival prediction of NSCLC patients by focusing on subgroups with identical characteristics. METHODS: A total of 304 NSCLC (Stages I-IV) patients treated with radiotherapy in our hospital were used. We extracted 107 radiomic features (i.e., 14 shape features, 18 first-order statistical features, and 75 texture features) from the gross tumor volume drawn on the free breathing planning computed tomography image. Three feature selection methods [i.e., test-retest and multiple segmentation (FS1), Pearson's correlation analysis (FS2), and a method that combined FS1 and FS2 (FS3)] were used to clarify how they affect survival prediction performance. Subgroup analysis for each histological subtype and each T stage applied the best selection method for the analysis of All data. We used a least absolute shrinkage and selection operator Cox regression model for all analyses and evaluated prognostic performance using the concordance-index (C-index) and the Kaplan-Meier method. For subgroup analysis, fivefold cross-validation was applied to ensure model reliability. RESULTS: In the analysis of All data, the C-index for the test dataset is 0.62 (FS1), 0.63 (FS2), and 0.62 (FS3). The subgroup analysis indicated that the prediction model based on specific histological subtypes and T stages had a higher C-index for the test dataset than that based on All data (All data, 0.64 vs. SCCall, 060; ADCall, 0.69; T1, 0.68; T2, 0.65; T3, 0.66; T4, 0.70). In addition, the prediction models unified for each T stage in histological subtype showed a different trend in the C-index for the test dataset between ADC-related and SCC-related models (ADCT1-ADCT4, 0.72-0.83; SCCT1-SCCT4, 0.58-0.71). CONCLUSIONS: Our results showed that feature selection methods moderately affected the survival prediction performance. In addition, prediction models based on specific subgroups may improve the prediction performance. These results may prove useful for determining the optimal radiomics-based predication model.

Plate Reader-Based Analytical Method for the Size Distribution of Submicron-Sized Protein Aggregates Using Three-Dimensional Homodyne Light Detection.

The assessment of aggregates is essential in biopharmaceutical development. Although submicron-sized aggregates are considered to have a potential immunogenicity risk, analytical techniques are limited. In this study, we present a new analytical technique using three-dimensional homodyne light detection (3D-HLD). In this system, submicron-sized particles are quantified by combining the reflected light detection of each particle by high-speed 3D scan and then enhancing the amplitude of the reflected light using HLD. The particle concentrations and size distributions of human tetanus immune globulin (TIG) aggregates generated by stirring were measured using 3D-HLD. Both concentrations and distributions were comparable to those obtained via resonant mass measurement (RMM), a technique commonly used for submicron-sized particle measurement. Aiming at feasibility assessment of 3D-HLD for the high-through-put formulation development, 30 formulations of TIG and rituximab under agitation stress were analyzed by 3D-HLD. The results showed that 3D-HLD can automatically and simultaneously assess the aggregate concentrations and size distributions of at least 90 samples. This study demonstrates that 3D-HLD can be used for submicron-sized aggregate analysis as an orthogonal method to RMM and also as a screening tool during formulation development.

Molecular and immunohistochemical studies in expression of voltage-dependent Ca2+ channels in dorsal root ganglia from streptozotocin-induced diabetic mice.

We have recently demonstrated that intrathecal injection of a selective P/Q-type blocker of the voltage-dependent Ca(2+) channels (VDCCs) significantly inhibited the mechanical hyperalgesia in streptozotocin (STZ)-induced diabetic mice, its antinociceptive effect being greater than in controls. In this study, to further clarify the underlying mechanism of the STZ-induced hyperalgesia, we investigated the expression level of the VDCC alpha1A and alpha1B subunits in the dorsal root ganglia (DRGs) and the dorsal spinal cord under this hyperalgesia. Real-time PCR analysis showed mRNA expression of alpha1A (P/Q-type), but not alpha1B (N-type), was significantly increased in the DRGs from the STZ-induced diabetic mice. On the other hand, gene expression of both alpha1 subunits was not changed in the dorsal part of the spinal cord. In diabetic DRG neurons, the number of large nerve cells was significantly reduced, whereas small neurons were significantly increased. Immunohistochemical study demonstrated the alpha1A-positive neurons, but not alpha1B-positive neurons, increased significantly greater in diabetic DRGs than in control in all cell size. These results indicate that an alteration in expression of P/Q-type VDCCs, especially in the small and medium-diameter primary afferent fibers, in pain pathways ascending input to the spinal cord may be involved in hypersensitivity in STZ-induced diabetes.

'Click chemistry' on polysaccharides: a convenient, general, and monitorable approach to develop (1-->3)-beta-D-glucans with various functional appendages.

(1-->3)-beta-D-Glucans having various functional appendages (lactoside, ferrocene, pyrene, and porphyrin) can be prepared in an convenient, quantitative, and regioselective manner through regioselective bromination-azidation of curdlan to afford 6-azido-6-deoxycurdlan followed by chemoselective [3+2]-cycloadditions with various functional modules bearing a terminal alkyne group. The ability to monitor reaction conversions is an additional advantage of this synthetic approach over the conventional direct modifications on polysaccharides; the reaction can be readily monitored based on the intensity of azido peaks in the in situ attenuated total reflection infrared spectra.

Photo-control of the polyplexes formation between DNA and photo-cation generatable water-soluble polymers.

Photo-cation generatable water-soluble polymers (Mw: approximately 1x10(5)) were prepared by radical copolymerization of N,N-dimethylacrylamide and vinyl monomer of triphenylmethane leucohydroxide (malachite green), which generate a cation upon ultraviolet light (UV) irradiation at wavelengths of between 290 and 410 nm. The malachite green contents of the copolymers were 3.6 (0.4 mol%), 7.9 (0.7 mol%), and 15.0 (2.7 mol%) per molecule. When an aqueous solution of the photo-cationized copolymers generated by UV irradiation was mixed with a Tris-HCl buffer (pH 7.4) of DNA (pGL3-control plasmid), dynamic light scattering (DLS) measurements showed the formation of polyplexes between the cationic copolymers and anionic DNA by non-specific electrostatic interaction, which was visualized with a confocal laser scanning microscopy (CLMS). Their mean cumulant diameter was about 150 nm with low polydispersity irrespective of the malachite green content in the copolymers. In the copolymer with the lowest malachite green content, almost all of the amount of the polyplexes was maintained by repeated UV irradiation, but the amount gradually decreased in the dark at 37 degrees C due to dissociation of the polyplexes, synchronized with the neutralization of the cation form of malachite green. The photo-cation generatable copolymers designed here can undergo photo-induced formation of the polyplexes with DNA and thermal polyplex dissociation, which may be used as a model for a novel photo-induced gene delivery system into cells.

Lactose-appended schizophyllan is a potential candidate as a hepatocyte-targeted antisense carrier.

A schizophyllan (beta-1,3-glucan) derivative carrying lactose-appendages prepared by reductive amination can form stable macromolecular complexes with polynucleotides, shows excellent affinity with a lactose-binding lectin, and effectively mediates gene transfection into hepatocytes.

Single-walled carbon nanotubes acquire a specific lectin-affinity through supramolecular wrapping with lactose-appended schizophyllan.

Single-walled carbon nanotubes can be entrapped within a helical superstructure composed of schizophyllan bearing lactoside-appendages to show an excellent water-solubility as well as a specific lectin-affinity.

Low Mw sulfated curdlan with improved water solubility forms macromolecular complexes with polycytidylic acid.

The water solubility of curdlan was enhanced by partial sulfation at O-6. Circular dichroism measurements suggest that the sulfated curdlan with the degree of substitution (DS) from 0 to 8.7 mol% forms macromolecular complexes with polycytidylic acid (poly(C)). Although the thermal stability of the complexes decreased with increase in DS, this could be overlapped by addition of NaCl in the concentration above that of serum. The results clearly indicate that the drawback arising from the electrostatic repulsion between the anionic charges can be partially compensated by the presence of salt. Furthermore, the polynucleotide chain complexed with the sulfated curdlan was protected from the enzymatic hydrolysis, corroborating the assumption that the sulfated curdlan has an ability to bind poly(C).

Polysaccharide-polynucleotide complexes (15): thermal stability of schizophyllan (SPG)/poly(C) triple strands is controllable by alpha-amino acid modification.

Schizophyllan (SPG), a beta-1,3-glucan polysaccharide which is known to form macromolecular complexes with certain polynucleotides, was modified by a reductive amination method with alpha-amino acids (Arg, Lys, and Ser). The thermal stability of the complexes as estimated by T(m) was enhanced in SPG-Arg and SPG-Lys conjugates which have pI values higher than the pH of the medium (8.0). The T(m) shift increased with the increase in the percentage of alpha-amino acid introduced and the highest T(m) values attained were 64 degrees C for SPG-Arg conjugate and 62 degrees C for SPG-Lys conjugate, which are higher by 13 and 11 degrees C, respectively, than those of the unmodified SPG+poly(C) complex. In the SPG-Ser conjugate with a pI lower than the medium pH (8.0), the T(m) values decreased with an increase in the percentage of Ser. Formation of the macromolecular complex was no longer detected above 13.2% Ser. The findings indicate that the T(m) values are easily controllable by the type and percentage of the introduced alpha-amino acids. We believe, therefore, that the present conjugates, consisting of naturally originated SPG and alpha-amino acids, provide an important lead for developing nontoxic artificial vectors and to control the affinity with polynucleotides in response to medium pH and temperature.

Schizophyllans carrying oligosaccharide appendages as potential candidates for cell-targeted antisense carrier.

Schizophyllans carrying beta-lactoside and alpha-mannoside appendages were prepared from native schizophyllan through NaIO4 oxidation followed by reductive amination using aminoethyl-beta-lactoside and alpha-mannoside, respectively. The resulting schizophyllans form stable macromolecular complexes with polynucleotides, such as poly(C) and poly(dA). Specific affinity between these macromolecular complexes and saccharide-binding proteins was demonstrated by surface plasmon resonance and agarose gel staining assays. beta-lactoside-appended schizophyllan enhanced an antisense activity in hepatocytes which express lactoside-binding proteins on their cell surfaces.