To validate the applicability of a residual CHO host cell protein (HCP) test method for the in-process candidate therapeutic monoclonal antibody and its release control.

The coverage ratio of goat anti-CHO HCP polyclonal antibody to the HCPs from supernatant of fermented mock CHO cells was determined by fluorescent-labeled two-dimensional differential gel electrophoresis (2D-DIGE). The limit of detection (LOD) and limit of quantitation (LOQ) were determined statistically from results of 40 tests. The specificity, precision/intermediate precision, accuracy, linearity and robustness were validated according to ICH Q2. To verify the suitability of the method for in-process quality control, key HCP removal steps, and sample dilution ranges for each step were determined according to USP 〈1132〉.

The coverage ratio of goat anti-CHO HCP polyclonal antibody was 56%. The LOD and LOQ was 2.47 and 7.41 ng·mL^-1, respectively. The validation results showed that the specificity, repeatability/intermediate precision, accuracy, linearity and robustness of the selected CHO HCP test method all met the application requirements. With the detection results of serially diluted purification intermediates, the dilution factor range of intermediates and HCP clearance ratio of each purification step was confirmed, and protein A affinity chromatography was identified as the main clearance step for HCP. Deduced with the above results, the selected CHO HCP test method can be used as the in-process control method for protein A chromatography and release test for drug substance.

The selected CHO HCP test method can be used in the process control and drug substance release of candidate therapeutic mAbs.

To construct bispecific fusion protein from Glypican-3 and NKG2D, and to primarily evaluate the cytolytic properties against human hepatocellular carcinoma cell line HepG2 in vitro.

Using genetic engineering technology, bispecific fusion protein FabGC33-MICA (human MHC class Ⅰ chain-related A, FabGM) expression vector was obtained by fusion the extracellular domain of MICA to heavy chain's C-terminal of the Fab form of GC33, and was used to transfect FreeStyle^TM 293-F cells. FabGM fusion protein was purified from the cell supernatant by affinity chromatography. Octet QKe system (ForteBio) was applied to detect FabGM's affinity for its targets. Flow cytometry analysis was used to assess the avidity of FabGM fusion protein in HepG2 cells. The cytotoxicity of FabGM against HepG2 was monitored by IncuCyte Live Cell Imaging System. The secretion of interferon-γ (IFN-γ) by PBMCs was detected by ELISA assay.

FabGM was successfully purified and confirmed selectively combined with GPC3 and NKG2D. The SDS-PAGE assay confirmed the predicted molecular. The K_d values of FabGM to target molecular GPC3 and NKG2D was 1.14 and 561 nmol·L^-1 respectively. FabGM effectively mediated the targeted killing effect of PBMCs against HepG2. ELISA assay showed that IFN-γ level was significantly increased during the killing process compared with negative control.

Bispecific fusion protein FabGM effectively mediated PBMCs to kill HepG2 cells in vitro, providing experimental evidence for FabGM to become a candidate drug for hepatocellular carcinoma treatment.

To review the progress on the study of drug accelerated stability models and search for a more efficient, lower-cost and more reliable model.

The study of drug stability is an indispensable step in the process of drug development. The advantages and disadvantages of the currently used accelerated stability models (ASMs), including classical Arrhenius' model, Waterman's model and Clancy's models, are compared from the aspects of factors for consideration, extensiveness of suitable application for drug substances, length of experimental period and accuracy of prediction.

There is only the effect of temperature which is considered in the classical Arrhenius' model, the experimental period is 3~6 months and even more and the range of application is narrow in this model that only less than 50% of drug substances could be used to predict the stability correctly. The effects of temperature and humidity on stability are taken into account in the Waterman's model and Clancy's models. The experimental period needs only 2~3 weeks, in which nearly 71% of the stability of the temperature and humidity influencing factors could be interpreted by Waterman's model, while almost all of them could be interpreted by Clancy's model, and the accuracy of prediction is higher than that of Waterman's model.

Clancy's model is more efficient, lower-cost and more reliable model than the current model; but the Waterman model is more widely used than the Clancy model.

To identify the potential risk of N-nitrosamines in azithromycin and to establish corresponding method for confirmatory testing, providing reference for risk assessment by manufacturers.

In conformity with the relevant industry guidance of FDA and EMA, risk identification and control limits for potential N-nitrosamines in azithromycin was conducted based on the structure and synthetic route, as well as degradation test. A liquid chromatography-mass spectrometry method was subsequently established and applied to the confirmatory testing of N-nitrosodimethylamine in azithromycin injection of 61 batches from 15 different manufacturers.

N-nitrosodimethylamine was probably introduced from starting material or generated as the degradation product of azithromycin. In all cases, N-nitrosodimethylamine were found below the limit of quantitation.

The risk of potential N-nitrosamines in azithromycin injection was acceptable. This method proved sensitive, selective, accurate and suitable to the confirmatory testing and quality control of N-nitrosodimethylamine in azithromycin injection. The article also provided an example to the risk assessment and confirmatory testing of potential N-nitrosamines in chemically synthesized APIs and drug products.

To analyze the occurrence and clinical characteristics of bleeding induced by rivaroxaban in order to provide reference for clinical medication.

Bleeding adverse reaction literature from CNKI, VIP, Chinese Hospital Data Library, Wanfang, PubMed and EMBase since the listing of rivaroxaban were collected and analyzed.

A total of 51 effective literature of report cases regarding bleeding ADR induced by rivaroxaban, involving 59 patients, were identified and included in the analysis. The majority of patients were over 60 years old (51 cases, 86.44%). Most of bleeding occurred within 30 days after medication (27 cases, 45.76%). Cerebral hemorrhage was commonly seen (15 cases, 25.42%). Conventional treatment is mostly applied (26 cases, 44.07%). In terms of treatment outcome, most bleeding stopped within 1 to 7 days (29 cases, 49.15%). There were 6 deaths, the majority of which occurred within 14 days of admission (5 cases, 83.33%).

The bleeding ADR induced by rivaroxaban need to be highly valued in clinical, and drug monitoring need to be strengthened to reduce the occurrence of fatal bleeding ADR.

To provide a basis for drug selection, as well as safe and rational use of drugs, Mini health technology assessment was used for 4 antiangiogenic agents for the treatment of advanced NSCLC (bevacizumab, ramucirumab, anlotinib, recombinant human endostatin).

Mini health technology assessment was conducted for the 4 mentioned antiangiogenic agents based on ten aspects, such as pharmaceutical characteristics, effectiveness, safety, economy, medical insurance, national essential drug, storage conditions, expiration date, global use and corporate credibility.

Based on the above evaluation criteria, the scores of the four antiangiogenic agents ranged 61~72. The difference in scores was mainly in terms of effectiveness, safety, economy, and storage conditions. Arotinib scored high in the treatment of small cell lung cancer, and was recommended in the drug selection list of medical institutions.

Mini health technology assessment is a multi-dimensional evaluation of drugs, providing basis for hospital drug selection, safe and rational use. It is of relatively high promotion value.