Search

BACKGROUND: Somatic and germline variants are not distinguishable by circulating tumor DNA (ctDNA) testing without analyzing non-tumor samples. Although confirmatory germline testing is clinically relevant, the criteria for selecting presumed germline variants have not been established in ctDNA testing. In the present study, we aimed to evaluate the prevalence of pathogenic germline variants in clinical ctDNA testing through their variant allele fractions (VAFs). METHODS: A total of consecutive 106 patients with advanced solid tumors who underwent ctDNA testing (Guardant360®) between January 2018 and March 2020 were eligible for this study. To verify the origin of pathogenic variants reported in ctDNA testing, germline sequencing was performed using peripheral blood DNA samples archived in the Clinical Bioresource Center in Kyoto University Hospital (Kyoto, Japan) under clinical research settings. RESULTS: Among 223 pathogenic variants reported in ctDNA testing, the median VAF was 0.9% (0.02-81.8%), and 88 variants with ≥ 1% VAFs were analyzed in germline sequencing. Among 25 variants with ≥ 30% VAFs, seven were found in peripheral blood DNA (BRCA2: n = 6, JAK2: n = 1). In contrast, among the 63 variants with VAFs ranging from 1 to < 30%, only one variant was found in peripheral blood DNA (TP53: n = 1). Eventually, this variant with 15.6% VAF was defined to be an acquired variant, because its allelic distribution did not completely link to those of neighboring germline polymorphisms. CONCLUSION: Our current study demonstrated that VAFs values are helpful for selecting presumed germline variants in clinical ctDNA testing.

Clinical cancer genome sequencing detects oncogenic variants that are potential targets for cancer treatment, but it also detects variants of unknown significance. These variants may interact with each other to influence tumor pathophysiology, however, such interactions have not been fully elucidated. Additionally, the effect of target therapy for those variants also unclarified. In this study, we investigated the biological functions of a HER2 mutation (G776S mutation) of unknown pathological significance, which was detected together with APC mutation by cancer genome sequencing of samples from a colorectal cancer (CRC) patient. Transfection of the HER2 G776S mutation alone slightly increased the kinase activity and phosphorylation of HER2 protein, but did not activate HER2 downstream signaling or alter the cell phenotype. On the other hand, the HER2 G776S mutation was shown to have strong oncogenic potential when loss of APC function was accompanied. We revealed that loss of APC function increased Wnt pathway activity but also increased RAS-GTP, which increased ERK phosphorylation triggered by HER2 G776S transfection. In addition, afatinib, a pan-HER tyrosine kinase inhibitor, suppressed tumor growth in xenografts derived from HER2 G776S-transfected CRC cells. These findings suggest that this HER2 mutation in CRC may be a potential therapeutic target.

Sudden deterioration of condition in patients with various diseases, such as cardiopulmonary arrest, may result in poor outcome even after resuscitation. Early detection of deterioration is important in medical and long-term care settings, regardless of the acute or chronic phase of disease. Early detection and appropriate interventions are essential before resuscitating measures are required. Among the vital signs that indicate the general condition of a patient, respiratory rate has a greater ability to predict serious events such as thromboembolism and sepsis than heart rate and blood pressure, even in early stages. Despite its importance, however, respiratory rate is frequently overlooked and not measured, making it a neglected vital sign. To facilitate the measurement of respiratory rate, a non-invasive method of detecting respiratory sounds was developed based on deep learning technology, using a built-in microphone in a smartphone. Smartphones attached to the bed headboards of 20 participants undergoing polysomnography (PSG) at Kyoto University Hospital recorded respiratory sounds. Sound data were synchronized with overnight respiratory information. After excluding periods of abnormal breathing on the PSG report, sound data were processed for each 1-minute period. Expiration sound was determined using the pressure flow sensor signal on PSG. Finally, a model to identify the expiration section from the sound information was created using a deep learning algorithm from the convolutional Long Short Term Memory network. The accuracy of the learning model in identifying the expiratory section was 0.791, indicating that respiratory rate can be determined using the microphone in a smartphone. By collecting data from more patients and improving the accuracy of this method, respiratory rates could be more easily monitored in all situations, both inside and outside the hospital.

BACKGROUND: Tumor heterogeneity has been known to cause inter-assay discordance among next-generation sequencing (NGS) results. However, whether preclinical factors such as sample type, sample quality and analytical features of gene panel can affect the concordance between two different assays remains largely unexplored. METHODS: Replicate sets of DNA samples extracted from formalin-fixed paraffin-embedded tissues (FFPE) (n = 20) and fresh frozen (FF) tissues (n = 10) were herein analyzed using a tumor-only (TO) and paired tumor-normal (TN) gene panel in laboratories certified by the Clinical Laboratory Improvement Amendment. Reported variants from the TO and TN panels were then compared. Furthermore, additional FFPE samples were sequentially sliced from the same FFPE block and submitted to another TN panel assay. RESULTS: Substantial discordance (71.8%) was observed between the results of the two panels despite using identical DNA samples, with the discordance rate being significantly higher for FFPE samples (p < 0.05). Among the 99 variants reported only in the TO panel, 32.3% were consistent with germline variants, which were excluded in the TN panel, while 30.3% had an allele frequency of less than 5%, some of which were highly likely to be artificial calls. The comparison of two independent TN panel assay results from the same FFPE block also showed substantial discordance rate (55.3%). CONCLUSIONS: In the context of clinical settings, our comparative analysis revealed that inter-NGS assay discordance commonly occurred due to sample types and the different analytical features of each panel.

Designing a deep neural network model that integrates clinical images with other electronic medical records entails various preprocessing operations. Preprocessing of clinical images often requires trimming of parts of the lesions shown in the images, whereas preprocessing of other electronic medical records requires vectorization of these records; for example, patient age is often converted into a categorical vector of 10-year intervals. Although these preprocessing operations are critical to the performance of the classification model, there is no guarantee that the preprocessing step chosen is appropriate for model training. The ability to integrate these preprocessing operations into a deep neural network model and to train the model, including the preprocessing operations, can help design a multi-modal medical classification model. This study proposes integration layers of preprocessing, both for clinical images and electronic medical records, in deep neural network models. Preprocessing of clinical images is realized by a vision transformer layer that selectively adopts the parts of the images requiring attention. The preprocessing of other medical electrical records is performed by adopting full-connection layers and normalizing these layers. These proposed preprocessing-integrated layers were verified using a posttreatment visual acuity prediction task in ophthalmology as a case study. This prediction task requires clinical images as well as patient profile data corresponding to each patient's posttreatment logMAR visual acuity. The performance of a heuristically designed prediction model was compared with the performance of the prediction model that includes the proposed preprocessing integration layers. The mean square errors between predicted and correct results were 0.051 for the heuristic model and 0.054 for the proposed model. Experimental results showed that the proposed model utilizing preprocessing integration layers achieved nearly the same perform

BACKGROUND: Borderline resectable pancreatic cancer (BRPC) is a category of pancreatic cancer that is anatomically widely spread, and curative resection is uncommon with upfront surgery. Intensity-modulated radiation therapy (IMRT) is a form of radiation therapy that delivers precise radiation to a tumor while minimizing the dose to surrounding normal tissues. Here, we conducted a phase 2 study to estimate the curability and efficacy of neoadjuvant chemoradiotherapy using IMRT (NACIMRT) for patients with BRPC with arterial abutment (BRPC-A). METHODS: A total of 49 BRPC-A patients were enrolled in this study and were treated at our hospital according to the study protocol between June 2013 and March 2021. The primary endpoint was microscopically margin-negative resection (R0) rates and we subsequently analyzed safety, histological effect of the treatment as well as survivals among patients with NACIMRT. RESULTS: Twenty-nine patients (59.2%) received pancreatectomy after NACIMRT. The R0 rate in resection patients was 93.1% and that in the whole cohort was 55.1%. No mortality was encountered. Local therapeutic effects as assessed by Evans classification showed good therapeutic effect (Grade 1, 3.4%; Grade 2a, 31.0%; Grade 2b, 48.3%; Grade 3, 3.4%; Grade 4, 3.4%). Median disease-free survival was 15.5 months. Median overall survival in the whole cohort was 35.1 months. The only independent prognostic pre-NACIMRT factor identified was serum carbohydrate antigen 19-9 (CA19-9) > 400 U/ml before NACIMRT. CONCLUSIONS: NACIMRT showed preferable outcome without significant operative morbidity for BRPC-A patients. NACIMRT contributes to good local tumor control, but a high initial serum CA19-9 implies poor prognosis even after neoadjuvant treatment. TRIAL REGISTRATION: UMIN-CTR Clinical Trial: https://upload.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000011776 Registration number: UMIN000010113. Date of first registration: 01/03/2013.