Purpose To determine the expressions of SARS-CoV-2 receptor angiotensin-converting enzyme 2 (ACE2) and type II transmembrane serine protease (TMPRSS2) genes in human and mouse ocular cells and comparison to other tissue cells. Methods Human conjunctiva and primary pterygium tissues were collected from pterygium patients who underwent surgery. The expression of ACE2 and TMPRSS2 genes was determined in human primary conjunctival and pterygium cells, human ocular and other tissue cell lines, mesenchymal stem cells as well as mouse ocular and other tissues by reverse transcription-polymerase chain reaction (RT-PCR) and SYBR green PCR.Results RT-PCR analysis showed consistent expression by 2 ACE2 gene primers in 2 out of 3 human conjunctival cells and pterygium cell lines. Expression by 2 TMPRSS2 gene primers could only be found in 1 out of 3 pterygium cell lines, but not in any conjunctival cells. Compared with the lung A549 cells, similar expression was noted in conjunctival and pterygium cells. In addition, mouse cornea had comparable expression of Tmprss2 gene and lower but prominent Ace2 gene expression compared with the lung tissue. Conclusion Considering the necessity of both ACE2 and TMPRSS2 for SARS-CoV-2 infection, our results suggest that conjunctiva would be less likely to be infected by SARS-CoV-2, whereas pterygium possesses some possibility of SARS-CoV-2 infection. With high and consistent expression of Ace2 and Tmprss2 in cornea, cornea rather than conjunctiva has higher potential to be infected by SARS-CoV-2. Precaution is necessary to prevent possible SARS-CoV-2 infection through ocular surface in clinical practice.

Background Epidermal growth factor receptor exon 20 insertion (EGFRex20ins) mutations represent approximately 4–12% of EGFR mutations and are generally refractory to the 1st and 2nd generation EGFR tyrosine kinase inhibitors (TKIs). Development of effective therapies for patients with EGFRex20ins mutant non-small-cell lung carcinoma (NSCLC) represents a great unmet need. Preclinical models have shown that osimertinib is active in NSCLC harboring EGFRex20ins, while the antitumor activity of osimertinib remains to be evaluated in patients with EGFRex20ins mutations. Methods Tumor genotyping was performed in 2316 Chinese NSCLC cases with targeted next generation sequencing (NGS) covering the whole exons of EGFR gene. The frequency and genetic characteristics of EGFRexon20ins mutations were analyzed. Furthermore, six patients with specific EGFRexon20ins mutations and receiving osimertinib 80 mg once daily were retrospectively included to assess the antitumor activity and safety of monotherapy osimertinib. Results EGFRex20ins mutations were identified in 4.8% (53/1095) of EGFR mutant NSCLC and 2.3% (53/2316) of all NSCLC cases. The most frequently identified EGFRexon20ins is A767_V769dup (17/53,32.1%). We found that the genetic characteristics of EGFRex20ins mutations in Chinese patients with NSCLC were comparable to those reported in Caucasian patients. Four patients with osimertinib therapy achieved partial response and the rest stable disease. Median progression free survival (PFS) was 6.2 months (95% confidence interval 5.0–12.9 months; range 4.9–14.6 months). The most common adverse events (AEs) were diarrhea (2/6), pruritis (2/6), stomatitis (1/6) and nausea (1/6). No grade 3 or more AEs were documented. Conclusions This study revealed that the genetic characteristics of EGFRex20ins mutations in Chinese patients with NSCLC were comparable to those reported in Caucasian patients. Furthermore, our study firstly demonstrated promising antitumor activity of osimertinib in certain EGFRex20ins mutant advanced NSCLC patients, indicating that osimertinib treatment for EGFRex20ins positive patients deserves further study. Electronic supplementary material The online version of this article (10.1186/s12885-019-5820-0) contains supplementary material, which is available to authorized users.

Retinal fundus diseases can lead to irreversible visual impairment without timely diagnoses and appropriate treatments. Single disease-based deep learning algorithms had been developed for the detection of diabetic retinopathy, age-related macular degeneration, and glaucoma. Here, we developed a deep learning platform (DLP) capable of detecting multiple common referable fundus diseases and conditions (39 classes) by using 249,620 fundus images marked with 275,543 labels from heterogenous sources. Our DLP achieved a frequency-weighted average F1 score of 0.923, sensitivity of 0.978, specificity of 0.996 and area under the receiver operating characteristic curve (AUC) of 0.9984 for multi-label classification in the primary test dataset and reached the average level of retina specialists. External multihospital test, public data test and tele-reading application also showed high efficiency for multiple retinal diseases and conditions detection. These results indicate that our DLP can be applied for retinal fundus disease triage, especially in remote areas around the world.