Diagnostic study on clinical feasibility of an AI-based diagnostic system as a second reader on mobile CT images: a preliminary result.

Background: Artificial intelligence (AI) has breathed new life into the lung nodules detection and diagnosis. However, whether the output information from AI will translate into benefits for clinical workflow or patient outcomes in a real-world setting remains unknown. This study was to demonstrate the feasibility of an AI-based diagnostic system deployed as a second reader in imaging interpretation for patients screened for pulmonary abnormalities in a clinical setting. Methods: The study included patients from a lung cancer screening program conducted in Sichuan Province, China using a mobile computed tomography (CT) scanner which traveled to medium-size cities between July 10th, 2020 and September 10th, 2020. Cases that were suspected to have malignant nodules by junior radiologists, senior radiologists or AI were labeled a high risk (HR) tag as HR-junior, HR-senior and HR-AI, respectively, and included into final analysis. The diagnosis efficacy of the AI was evaluated by calculating negative predictive value and positive predictive value when referring to the senior readers' final results as the gold standard. Besides, characteristics of the lesions were compared among cases with different HR labels. Results: In total, 251/3,872 patients (6.48%, male/female: 91/160, median age, 66 years) with HR lung nodules were included. The AI algorithm achieved a negative predictive value of 88.2% [95% confidence interval (CI): 62.2-98.0%] and a positive predictive value of 55.6% (95% CI: 49.0-62.0%). The diagnostic duration was significantly reduced when AI was used as a second reader (223±145.6 vs. 270±143.17 s, P<0.001). The information yielded by AI affected the radiologist's decision-making in 35/145 cases. Lesions of HR cases had a higher volume [309.9 (214.9-732.5) vs. 141.3 (79.3-380.8) mm3, P<0.001], lower average CT number [-511.0 (-576.5 to -100.5) vs. -191.5 (-487.3 to 22.5), P=0.010], and pure ground glass opacity rather than solid. Conclusions: The AI algorithm had high negative predictive value but low positive predictive value in diagnosing HR lung lesions in a clinical setting. Deploying AI as a second reader could help avoid missed diagnoses, reduce diagnostic duration, and strengthen diagnostic confidence for radiologists.

Increased microRNA-30a levels in bronchoalveolar lavage fluid as a diagnostic biomarker for lung cancer.

BACKGROUND: MicroRNA-30a (miRNA-30a) levels have been shown to increase in the plasma of lung cancer patients. Herein, we evaluated the miRNA-30a levels in the bronchoalveolar lavage fluid (BALF) of lung cancer patients as a potential biomarker for lung cancer diagnosis. METHODS: BALF miRNA-30a expression of 174 subjects was quantified using quantitative real-time reverse transcription-polymerase chain reaction and compared between lung cancer patients and control patients with benign lung diseases. Moreover, its diagnostic value was evaluated by performing receiver operating characteristic (ROC) curve analysis. RESULTS: The relative BALF miRNA-30a expression was significantly higher in the lung cancer patients than in the controls (0.74 ±  0.55 versus 0.07 ±  0.48, respectively, p < 0.001) as well as in lung cancer patients with stage I-IIA disease than in those with stage IIB-IV disease (0.98 ±  0.64 versus 0.66 ±  0.54, respectively, p < 0.05). Additionally, miRNA-30a distinguished benign lung diseases from lung cancers, with an area under the ROC curve (AUC) of 0.822. ROC analysis also revealed an AUC of 0.875 for the Youden index-based optimal cut-off points for stage I-IIA adenocarcinoma. Thus, increased miRNA-30a levels in BALF may be a useful biomarker for non-small-cell lung cancer diagnosis.

Noninvasive genotyping and monitoring of anaplastic lymphoma kinase (ALK) rearranged non-small cell lung cancer by capture-based next-generation sequencing.

Noninvasive genotyping of driver genes and monitoring of tumor dynamics help make better personalized therapeutic decisions. However, neither PCR-based assays nor amplicon-based targeted sequencing can detect fusion genes like anaplastic lymphoma kinase (ALK) rearrangements in blood samples. To investigate the feasibility and performance of capture-based sequencing on ALK fusion detection, we developed a capture-based targeted sequencing panel to detect and quantify rearrangement events, along with other driver mutation variants in plasma. In this perspective study, we screened 364 patients with advanced non-small cell lung cancer (NSCLC) for ALK rearrangements, and collected blood samples from 24 of them with confirmed ALK rearrangements based on their tissue biopsies. ALK rearrangements were successfully detected in 19 of 24 patients at baseline with 79.2% (95% CI 57.9%, 92.9%) sensitivity and 100% (36/36) specificity. Among the 24 patients, we obtained longitudinal blood samples from 7 of them after either chemotherapy and/or Crizotinib treatment for disease monitoring. The by-sample detection rate of ALK rearrangements after treatment drops to 69.2% (9 of 13). In addition to detecting ALK rearrangements, we also detected 3 Crizotinib resistant mutations, ALK L1152R, ALK I1171T and ALK L1196M from patient P4. ctDNA concentration correlates with responses and disease progression, reflecting its ability as a biomarker. Our findings suggest capture-based sequencing can detect and quantify ALK rearrangements as well as other somatic mutations, including mutations mediated drug resistance, in plasma with high sensitivity, paving the way for its application in identifying driver fusion genes and monitoring tumor dynamics in the clinic.

Prognostic value of CD44 variant exon 6 expression in non-small cell lung cancer: a meta-analysis.

BACKGROUND: CD44v6 (CD44 variant exon 6) is the chief CD44 variant isoform regulating tumor invasion, progression, and metastasis. The prognostic value of CD44v6 expression in non small cell lung cancer (NSCLC) has been evaluated in many studies, but the results have remained controversial. Thus, we performed a meta- analysis of currently available studies to investigate the prognostic value of CD44v6 expression in NSCLC patients and the relationship between the expression of CD44v6 and clinicopathological features. MATERIALS AND METHODS: Two independent reviewers searched the relevant literature in Pubmed, Medline and Embase from 1946 to January 2014. Overall survival (OS) and various clinicopathological features were collected from included studies. This meta-analysis was accomplished using STATA 12.0 and Revman 5.2 software. Pooled hazard ratios (HRs) with 95% confidence intervals (95%CIs) were calculated to estimate the effects. RESULTS: A total of 921 NSCLC patients from ten studies met the inclusion criteria. The results showed that CD44v6 high expression was a prognostic factor for poor survival (HR=1.91, 95%CI=1.12-3.26, p<0.05). With respect to clinicopathological features, CD44v6 high expression was related to histopathologic type (squamous cell carcinoma versus adenocarcinoma: OR=2.72, 95%CI=1.38-5.38, p=0.004), and lymph node metastasis (OR=3.02, 95%CI=1.93-4.72, p<0.00001). CONCLUSIONS: Our results suggested CD44v6 high expression as a poor prognostic factor for NSCLC, and CD44v6 expression is associated with lymph node metastasis and histopathologic type. Therefore, CD44v6 expression can be used as a novel prognostic marker in NSCLC cases.

[Construction of sh-rpS6 lentivirus vectors and its effect on proliferation in lung adenocarcinoma A549 cell lines].

OBJECTIVE: To construct the sh-rpS6 lentivirus vector targeting ribosomal protein S6 (rpS6) and explore its effect on proliferation in lung adenocarcinoma A549 cell lines. METHODS: Sequences targeting the rpS6 gene were selected. The double strand shRNA oligo was ligated to pGCsil-GFP lentivirus vector and transformed into E. coli. The resulting recombinant vector was verified by sequencing. After transfection and lentivirus packing, the viral particles were collected and infected A549 cells. After selection of GFP positive cells by FACS, mRNA and protein expression levels of rpS6 were determined by real time PCR and Western blot. In the following experiment, the proliferation changes of A549 cell lines after the interference by sh-rpS6 was investigated by using CCK-8 kit. RESULTS: The sequencing result confirmed that pGCsil-sh-rpS6-GFP vector was successfully developed. Stably transfected A549 cell lines by sh-rpS6 were selected through FACS, with a selection ratio of 86.80%. The silencing effects of sh-rpS6 were determined by real time PCR and Western blot, suggesting that mRNA and protein expression of rpS6 in the targeted cells reduced by (79.72 +/- 6.83) % and (83.77 +/- 12.13) %, significantly lower than those of control groups. In vitro showed the cell proliferation with sh-rpS6 was significantly slower than that of controls (P < 0.05). CONCLUSION: The constructed sh-rpS6 lentivirus vector could inhibit the expression of rpS6 in A549 cell lines effectively and significantly slow the cell proliferation in vitro.

[Construction and expression of recombinant lentiviral vectors of AKT2,PDK1 and BAD].

OBJECTIVE: To construct human protein kinase B (ATK2), phosphoinositide-dependent kinase 1 (PDK1) and bcl-2-associated death protein (BAD) lentiviral expression vector, and to determine their expressions in 293T cells. METHODS: Total RNA was extracted from lung cancer tissues. The full-length coding regions of human ATK2, BAD and PDK1 cDNA were amplified via RT-PCR using specific primers, subcloned into PGEM-Teasy and then sequenced for confirmation. The full-length coding sequence was cut out with a specific restriction enzyme digest and subclone into pCDF1-MCS2-EF1-copGFP. The plasmids were transfected into 293T cells using the calcium phosphate method. The over expression of AKT2, BAD and PDK1 were detected by Western blot. RESULTS: AKT2, PDK1 and BAD were subcloned into pCDF1-MCS2-EF1-copGFP, with an efficiency of transfection of 100%, 95%, and 90% respectively. The virus titers were 6.7 x 10(6) PFU/mL in the supernatant. After infection, the proteins of AKT2, PDK1 and BAD were detected by Western blot. CONCLUSION: The lentivial vector pCDF1-MCS2-EF1-copGFP containing AKT2, BAD and PDK1 were successfully constructed and expressed in 293T cells.

[Application value of low-dose computed tomography for the screening of lung cancer in high-risk group].

OBJECTIVE: To analyze the value of low-dose computed tomography (LDCT) in early screening of lung cancer in high-risk group. METHODS: LDCT was performed in 1551 adults over 40-years old (male 1054, female 497) for the screening of lung cancer in Examination Center of West China Hospital, Sichuan University, and the subjects were divided into high-risk group and non-high-risk group according to risk factors of lung caner. The detection rates of lung lesion and lung cancer in these two groups were carefully analyzed. RESULTS: In the people over 40-years old,the detection rate of pulmonary lesions by LDCT was 54.41% (844/1551), the detection rate of nodules was 29.4% (456/1551). Detection rate of lung cancer was 0.51% (8/1551), while it was 1.21% (7/577) in high-risk group, and 2.01% (6/298) in heavy smoker group. The difference of lung cancer detection rate was significant (P < 0.05). When stratified with the number of risk factors, the detection rate of lung cancer by LDCT was 0.10% (1/974) in the subjects without any high-risk factors, 0.98% (5/506) in the subjects with one high-risk factor and 2.86% (2/70) in the subjects with more high-risk factors, and the differences were stastatically significant (P < 0.05). CONCLUSION: LDCT can detect asymptomatic lung cancer sensitively with low radiation dose, which could be helpful for the screening of lung cancer.

Prognostic impact of Raf-1 and p-Raf-1 expressions for poor survival rate in non-small cell lung cancer.

Overexpression of Raf-1 has commonly been observed in solid tumors including non-small cell lung cancer (NSCLC). The objective of this study was to investigate whether overexpression of Raf-1, phosphorylated-Raf-1 (p-Raf-1) or both correlates with poor survival rate in NSCLC patients and to explore associations between expression of these proteins and NSCLC cell fate both in vitro and in vivo. Expression of Raf-1 and p-Raf-1 were detected by immunohistochemistry in tumor specimens from 152 NSCLC patients and associations between their expression and the clinicopathological characteristics were assessed. Five-year median survival rate of patients were analyzed by Kaplan-Meier method, log-rank test and Cox regression. Cell fate was compared between normal tumor cells and those with Raf-1 silencing, in both the adenocarcinoma cell line A549 and xenografted mice that were infected with the A549 cell line. The incidence of overexpression of both Raf-1 and p-Raf-1 in NSCLC was much higher than normal control (P < 0.05), and the survival rate of patients with positive expression of Raf-1, p-Raf-1 or both was found to be significantly lower than the negative group (P < 0.05). Both univariate and multivariate analyses showed Raf-1 (P = 0.000, P = 0.010), p-Raf-1 (P = 0.004, P = 0.046), or both (P = 0.001, P = 0.016) was good prognostic markers for poor survival rate in NSCLC patients. Suppression of Raf-1 inhibited tumorigenesis by inducing apoptosis both in vitro and in vivo. These findings demonstrate that overexpression of Raf-1, p-Raf-1 or both could be considered as a new independent prognostic biomarker for poor survival rates for NSCLC patients.

[An analysis of the clinicopathological features and misdiagnosis of 16 adults pulmonary sequestration].

OBJECTIVE: To explore the clinicopathological features of adult pulmonary sequestration and summarize the misdiagnosis experiences. METHODS: Data of 16 cases of adult pulmonary sequestration (18 years), who were confirmed by surgery and biopsy in our hospital were collected and reviewed. RESULTS: The median age of all the patients was 38.5 years. The female seemed to be more likely to suffer from adult pulmonary sequestration (n = 12) with cough to be the most frequent symptom (n = 9). CT scans revealed most of the lesions were located in the left lower lobes of the lungs (n = 9). Half of the lesions were characterized by pulmonary cyst-like changes and/or multiple cystic bronchiectasis (n = 8), followed by soft tissue mass in or out of the lung fields (n = 7). Enhanced CT scans showed abnormal arteries from the systemic circulation. Only two cases were diagnosed as pulmonary sequestration correctly in the primary diagnosis. The remaining were mostly misdiagnosed as pulmonary cyst-like changes with bronchiectasis (n = 6) or tumors (n = 6). According to the findings during surgery, 13 cases were intralobar pulmonary sequestrations; 3 cases were extralobars, whose tissues were all detected dysplasia and chronic inflammatory by histopathological examinations. CONCLUSIONS: The misdiagnosis rate of pulmonary sequestration is high because of its non-specific clinical symptoms. Since it is characterized by abnormal arteries and pulmonary dysplasia, enhanced CT scans should be used as a preferred screening method for suspected cases, especially for those middleaged patients with cystic or mass-like lesions in the left lower lobes of the lungs.

Overexpression of Bcl-2-associated death inhibits A549 cell growth in vitro and in vivo.

The importance of apoptosis during the process of inhibiting tumorigenesis has been recognized. The role of BH3-only proapoptotic protein Bcl-2-associated death (BAD) in tumor growth remains controversial. The aim of this study was to explore the role of BAD in lung cancer cells. Our study showed that expression of BAD was upregulated in A549 cells by a recombinant lentivirus overexpressing BAD. In vitro, BAD overexpression significantly inhibited A549 cell proliferation and induced apoptosis in cell proliferation and apoptosis assays, respectively. The effect of BAD on A549 cells was studied in tumor xenograft of nude mice and the results showed that the tumor volume in the experimental group was smaller than the control groups. Further, immunohistochemical technique was used to determine the cell proliferation and apoptosis status of the lung tumor xenograft cells. This demonstrated that the in vivo and in vitro results were consistent. Taken together, our results indicate that overexpression of BAD inhibits the growth of A549 cells in vitro and in vivo, through inhibiting cell proliferation and inducing apoptosis. Thus, BAD could be a potential therapeutic target.

[The expression and clinical significance of AKT2, phosphorylated AKT2 in non-small cell lung cancer].

OBJECTIVE: To study the expression and clinical significance of AKT2, phosphorylated AKT2 (p-AKT2) in non-small cell lung cancer (NSCLC). METHODS: The tumor tissues were obtained from 137 cases of NSCLC, the expressions of AKT2 and p-AKT2 in the tissues were measured by immunohistochemistry. The statistic analysis was carried on to study the correlation of AKT2, p-AKT2 expression to the type of lung cancer, TNM stage, pathological grading. Survival analysis was also studied. RESULTS: The positive rates of AKT2 in lung adenocarcinoma and squamous carcinoma were 60.5% and 54.1% respectively (P > 0.05). While the positive rates of p-AKT2 in lung adenocarcinoma and squamous carcinoma were 68.4% and 47.5% (P < 0.05). The expressions of AKT2 and p-AKT2 were not correlated with age, gender, TNM stage and cell differentiation degree. Further more, survival analysis revealed that 5-year survival rate and median survival time for the patients with positive expression of p-AKT2 were significantly poorer than those with negative expression (20% vs 56%, (28.464 +/- 2.235) months vs (39.214 +/- 3.075) months, P < 0.053, while there were no significant differences with regard to AKT2 expression. CONCLUSION: The positive expression of p-AKT2 in lung adenocarcinoma was higher than that in lung squamous carcinomas. p-AKT2 may be a prognostic factor for non-small cell lung cancer.

[Co-expression and prognostic significance of hnRNP B1 and CD44 in non-small cell lung cancers].

OBJECTIVE: To study the co-expression and prognostic significance of hnRNP B1 and CD44 in non-small cell lung cancers (NSCLC). METHODS: The co-expressions of CD44 and hnRNP B1 in the tissues from 88 cases of NSCLC were measured by immunohistochemistry. The relationship between the expressions and the prognosis of NSCLC was analysed. RESULTS: The NSCLC had a high expression of hnRNP B1 (68.18%) and CD44 (52.27%). The cells had the longest average life [(59.607 +/- 4.092) months] in the conditions of high expressed hnRNP B1 and low expressed CD44, and shortest average life [(21.357 +/- 3.545) months] in the conditions of low expressed hnRNP B1 and high expressed CD44. CONCLUSION: Combined detection of hnRNP B1 and CD44 can help forecast the prognosis of NSCLC.