ABSTRACT
Background: Stereotactic body radiotherapy (SBRT) has emerged as an alternative approach for patients with hepatocellular carcinoma (HCC), and we aim to find potential prognostic biomarkers for HCC patients who received SBRT. Methods: In this study, we retrospectively analyzed HCC patients who underwent SBRT in our institution from January 2018 to December 2022. The inflammatory parameters, along with baseline patients' characteristics were collected to elucidate the potential relationship with survival benefits and liver toxicities. Results: Overall, 35 patients were enrolled in our study. For the efficacy population (25 patients who underwent SBRT for primary liver lesions), the objective response rate (ORR) and disease control rate (DCR) were 60% and 100%, respectively. The median progression-free survival (PFS) was 9.9 months [95% confidence interval (CI) 5.6-14.1 months], and the median overall survival (OS) was 18.5 months (95% CI 14.2-22.8 months). We further confirmed that higher baseline lymphocyte-C-reactive protein ratio (LCR) (≥2361.11) was positively related to both longer PFS (12.0 vs 4.3 months, P = 0.002) and OS (21.9 vs 11.4 months, P = 0.022). Moreover, patients with diabetes and higher alpha-fetoprotein (AFP) (≥400 ng/mL) were also found to be associated with worse OS. The most common hepatotoxicity was elevated gamma-glutamyl transferase (GGT) (84.0%). Conclusion: In conclusion, for patients with inoperable HCC, SBRT resulted in satisfactory local control, survival benefits, and acceptable liver toxicity. Pre-radiotherapy LCR might be an independent and readily available predictor for survival, which facilitates us to find the most appropriate treatment options.
ABSTRACT
Objectives: To assess the safety and efficacy of anlotinib (a multi-targeted tyrosine kinase inhibitor) combined with toripalimab (a PD-1 monoclonal antibody) in the treatment of unresectable biliary tract cancer (BTC). Methods: In this prospective, single-arm, single-centre exploratory clinical study, patients with locally progressed or metastatic BTC were included. Patients were treated with anlotinib (12 mg, PO, QD, for 2 weeks and then stopped for a week, 21 days for a cycle) and toripalimab (240 mg, IV, Q3W). The primary endpoint of the study was the objective response rate (ORR), as defined in RECIST version 1.1 criteria. Results: In this study, 15 BTC patients who met the criteria were enrolled. The ORR was 26.7%, the median progression-free survival (mPFS) was 8.6 months (95% CI: 2.1-15.2), the median overall survival (mOS) was 14.53 months (95% CI: 0.8-28.2) and the disease control rate (DCR) was 87.6%. A patient with hilar cholangiocarcinoma was successfully converted after three cycles of treatment and underwent surgical resection. Furthermore, patient gene sequencing revealed that STK11 was mutated more frequently in patients with poor outcomes. In addition, patients with a CD8/Foxp3 ratio > 3 had a longer survival than those with a CD8/Foxp3 ratio ≤ 3 (P = 0.0397). Conclusions: In patients with advanced BTC, the combination of anlotinib and toripalimab demonstrated remarkable anti-tumor potential, with increased objective response rates (ORR), longer overall survival (OS) and progression-free survival (PFS). Moreover, STK11 and CD8/Foxp3 may be as biomarkers that can predict the effectiveness of targeted therapy in combination with immunotherapy.
ABSTRACT
The application of machine learning in wavefront reconstruction has brought great benefits to real-time, non-invasive, deep tissue imaging in biomedical research. However, due to the diversity and heterogeneity of biological tissues, it is difficult to train the dataset with a unified model. In general, the utilization of some unified models will result in the specific sample falling outside the training set, leading to low accuracy of the machine learning model in some real applications. This paper proposes a sensorless wavefront reconstruction method based on transfer learning to overcome the domain shift introduced by the difference between the training set and the target test set. We build a weights-sharing two-stream convolutional neural network (CNN) framework for the prediction of Zernike coefficient, in which a large number of labeled randomly generated samples serve as the source-domain data and the unlabeled specific samples serve as the target-domain data at the same time. By training on massive labeled simulated data with domain adaptation to unlabeled target-domain data, the network shows better performance on the target tissue samples. Experimental results show that the accuracy of the proposed method is 18.5% higher than that of conventional CNN-based method and the peak intensities of the point spread function (PSF) are more than 20% higher with almost the same training time and processing time. The better compensation performance on target sample could have more advantages when handling complex aberrations, especially the aberrations caused by various histological characteristics, such as refractive index inhomogeneity and biological motion in biological tissues.
ABSTRACT
Non-invasive, real-time imaging and deep focus into tissue are in high demand in biomedical research. However, the aberration that is introduced by the refractive index inhomogeneity of biological tissue hinders the way forward. A rapid focusing with sensor-less aberration corrections, based on machine learning, is demonstrated in this paper. The proposed method applies the Convolutional Neural Network (CNN), which can rapidly calculate the low-order aberrations from the point spread function images with Zernike modes after training. The results show that approximately 90 percent correction accuracy can be achieved. The average mean square error of each Zernike coefficient in 200 repetitions is 0.06. Furthermore, the aberration induced by 1-mm-thick phantom samples and 300-µm-thick mouse brain slices can be efficiently compensated through loading a compensation phase on an adaptive element placed at the back-pupil plane. The phase reconstruction requires less than 0.2 s. Therefore, this method offers great potential for in vivo real-time imaging in biological science.
