Who can benefit from postmastectomy radiotherapy among HR+/HER2- T1-2 N1M0 breast cancer patients? An explainable machine learning mortality prediction based approach.

Objective: The necessity of postmastectomy radiotherapy(PMRT) for patients with HR+/HER2 T1-2 N1M0 breast cancer remains controversial. We want to use explainable machine learning to learn the feature importance of the patients and identify the subgroup of the patients who may benefit from the PMRT. Additionally, develop tools to provide guidance to the doctors. Methods: In this study, we trained and validated 2 machine learning survival models: deep learning neural network and Cox proportional hazard model. The training dataset consisted of 35,347 patients with HR+/HER2- T1-2 N1M0 breast cancer who received mastectomies from the SEER database from 2013 to 2018. The performance of survival models were assessed using a concordance index (c-index).Then we did subgroup analysis to identify the subgroup who could benefit from PMRT. We also analyzed the global feature importance for the model and individual feature importance for individual survival prediction. Finally, we developed a Cloud-based recommendation system for PMRT to visualize the survival curve of each treatment plan and deployed it on the Internet. Results: A total of 35,347 patients were included in this study. We identified that radiotherapy improved the OS in patients with tumor size >14mm and age older than 54: 5-year OS rates of 91.9 versus 87.2% (radio vs. nonradio, P <0.001) and cohort with tumor size >14mm and grade worse than well-differentiated, 5-year OS rates of 90.8 versus 82.3% (radio vs. nonradio, P <0.001).The deep learning network performed more stably and accurately in predicting patients survival than the random survival forest and Cox proportional hazard model on the internal test dataset (C-index=0.776 vs 0.641) and in the external validation(C-index=0.769 vs 0.650).Besides, the deep learning model identified several key factors that significantly influence patient survival, including tumor size, examined regional nodes, age at 45-49 years old and positive regional nodes (PRN). Conclusion: Patients with tumor size >14mm and age older than 54 and cohort with tumor size >14mm and grade worse than well-differentiated could benefit from the PMRT. The deep learning network performed more stably and accurately in predicting patients survival than Cox proportional hazard model on the internal test. Besides, tumor size, examined regional nodes, age at 45-49 years old and PRN are the most significant factors to the overall survival (OS).

Development and validation of machine learning models to predict survival of patients with resected stage-III NSCLC.

Objective: To compare the performance of three machine learning algorithms with the tumor, node, and metastasis (TNM) staging system in survival prediction and validate the individual adjuvant treatment recommendations plan based on the optimal model. Methods: In this study, we trained three machine learning madel and validated 3 machine learning survival models-deep learning neural network, random forest and cox proportional hazard model- using the data of patients with stage-al3 NSCLC patients who received resection surgery from the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) database from 2012 to 2017,the performance of survival predication from all machine learning models were assessed using a concordance index (c-index) and the averaged c-index is utilized for cross-validation. The optimal model was externally validated in an independent cohort from Shaanxi Provincial People's Hospital. Then we compare the performance of the optimal model and TNM staging system. Finally, we developed a Cloud-based recommendation system for adjuvant therapy to visualize survival curve of each treatment plan and deployed on the internet. Results: A total of 4617 patients were included in this study. The deep learning network performed more stably and accurately in predicting stage-iii NSCLC resected patients survival than the random survival forest and Cox proportional hazard model on the internal test dataset (C-index=0.834 vs. 0.678 vs. 0.640) and better than TNM staging system (C-index=0.820 vs. 0.650) in the external validation. The individual patient who follow the reference from recommendation system had superior survival compared to those who did not. The predicted 5-year-survival curve for each adjuvant treatment plan could be accessed in the recommender system via the browser. Conclusion: Deep learning model has several advantages over linear model and random forest model in prognostic predication and treatment recommendations. This novel analytical approach may provide accurate predication on individual survival and treatment recommendations for resected Stage-iii NSCLC patients.

High mobility group box 1 promotes radioresistance in esophageal squamous cell carcinoma cell lines by modulating autophagy.

Resistance to radiotherapy results in relapse and treatment failure in locally advanced esophageal squamous cell carcinoma (ESCC). High mobility group box 1 (HMGB1) is reported to be associated with the radioresistance in bladder and breast cancer. However, the role of HMGB1 in the radiotherapy response in ESCC has not been fully elucidated. Here, we investigated the role of HMGB1 to radioresistance in ESCC clinical samples and cell lines. We found that HMGB1 expression was associated with tumor recurrence after postoperative radiotherapy in locally advanced ESCC patients. HMGB1 knockdown in ESCC cells resulted in increased radiosensitivity both in vitro and in vivo. Autophagy level was found depressed in HMGB1 inhibition cells and activation of autophagy brought back cell's radioresistance. Our results demonstrate that HMGB1 activate autophagy and consequently promote radioresistance. HMGB1 may be used as a predictor of poor response to radiotherapy in ESCC patients. Our finding also highlights the importance of the utility of HMGB1 in ESCC radiosensitization.

Lobaplatin promotes radiosensitivity, induces apoptosis, attenuates cancer stemness and inhibits proliferation through PI3K/AKT pathway in esophageal squamous cell carcinoma.

Radiotherapy is one of the common treatments for esophageal squamous cell carcinoma (ESCC). Yet, local recurrence led by radioresistance is still not solved. Lobaplatin (LBP) is known to have powerful clinical anti-tumor activities in various tumors, but its effect in radiotherapy is rarely studied. Here we report that LBP is a promising radiosensitizer for ESCC. We treated ESCC cells with LBP and radiation, both separately and in combination. Untreated cells were set as control groups. We found that LBP inhibited ESCC cell growth and enhanced their radiosensitivity. LBP also impeded the tumor growth in vivo. LBP combined with radiation significantly increased ESCC cell apoptosis. Meanwhile, LBP obviously decreased the expression of cancer stem cells biomarker CD271 both in vitro and in vivo. The molecular mechanism was related to the downregulation of Bcl-2/Bax ratio, PI3K and p-AKT (Ser473) expression. Taken together, our findings indicated that LBP could enhance the radiosensitivity of ESCC cells by increasing radiation-induced apoptosis, attenuating cancer stemness and inhibiting PI3K/AKT pathway. These results provide a foundation for the combined therapy of radiation and LBP for ESCC in clinical practice.

Effect of sinomenine hydrochloride on radiosensitivity of esophageal squamous cell carcinoma cells.

Radiation therapy is one of the most important treatments for unresectable and locally advanced esophageal squamous cell carcinoma (ESCC), however, the response to radiotherapy is sometimes limited by the development of radioresistance. Sinomenine hydrochloride (SH) has anticancer activity, but its effect on the radiosensitivity of ESCC is unclear. We determined the effect of SH on the radiosensitivity of ESCC cells and elucidated its potential radiosensitization mechanisms in vitro and in vivo. ESCC cells were subjected to SH and radiation, both separately and in combination. Untreated cells served as controls. The CCK­8 assay was used to evaluate cell proliferation, and the clonogenic assay to estimate radiosensitization. Flow cytometry was used to investigate cell cycle phases and cell apoptosis. Bcl­2, Bax, cyclin B1, CDK1, Ku86, Ku70, and Rad51 expression was evaluated using western blotting. In vivo, tumor xenografts were created using BALB/c nude mice. Tumor­growth inhibition was recorded, and Ki­67 and Bax expression in the tumor tissues was assessed using immunohistochemistry. SH inhibited ESCC cell growth and markedly increased their radiosensitivity by inducing G2/M phase arrest. SH combined with radiation therapy significantly increased ESCC cell apoptosis. The molecular mechanism by which SH enhanced radiosensitivity of ESCC cells was related to Bcl­2, cyclin B1, CDK1, Ku86, Ku70, and Rad51 downregulation and Bax protein expression upregulation. SH combined with radiation considerably delayed the growth of tumor xenografts in vivo. Immunohistochemical analysis showed that in the SH combined with radiation group, the expression of Bax was significantly higher while that of Ki­67 was lower than the expressions in the control groups. Taken together, our findings showed that SH could improve the sensitivity of radiation in ESCC cells by inducing G2/M phase arrest, promoting radiation­induced apoptosis and inhibiting DSB­repair pathways. SH appears to be a prospective radiosensitizer for improving the efficacy of radiotherapy for ESCC.

PTENP1 inhibits the growth of esophageal squamous cell carcinoma by regulating SOCS6 expression and correlates with disease prognosis.

PTEN pseudogene (PTENP1) has a tumor suppressive role in multiple cancers. However, its involvement in esophageal squamous cell carcinoma (ESCC) remains largely unknown. In this study, we set out to identify the role of PTENP1 in the development of ESCC. Gene Expression Omnibus database was employed to investigate the expression of PTENP1 in ESCC. sRNA target Database (StarBase v2.0) was used to query the downstream of PTENP1. Next, both in vitro and in vivo experiments were employed to explore the function. Cell proliferation was evaluated by CCK-8, soft agar, and colony formation assays. Expression of relative genes was assessed by quantitative real-time PCR (qRT-PCR) and Western blotting. 3'UTR luciferase assay was used to confirm the miRNA binding. The clinical significance of PTENP1 was further validated by immunohistochemistry (IHC) and correlation with clinicopathological indicators in additional samples (n = 93). We found expression of PTENP1 in ESCC was lower than that in the corresponding adjacent normal tissues (n = 17). Overexpression of PTENP1 in Eca109 and TE-1 cells resulted in inhibited proliferation and altered expression of SOCS6-p-STAT3-HIF-1α pathway both in vitro and in vivo. Subsequent IHC reported a similar trend in human ESCC samples. 3'UTR luciferase assay demonstrated that PTENP1 3'UTR decoyed miR-17-5p from binding to SOCS6. Moreover, PTENP1 expression was correlated with clinicopathological indicators to varying degrees, including histological grade, TNM stage, infiltration depth, lymph node metastasis, and overall survival. Taken together, these results suggested an anti-oncogenic role of PTENP1. Meanwhile, PTENP1 may also serve as a candidate of prognostic indicator for ESCC patients.

A signature motif in LIM proteins mediates binding to checkpoint proteins and increases tumour radiosensitivity.

Tumour radiotherapy resistance involves the cell cycle pathway. CDC25 phosphatases are key cell cycle regulators. However, how CDC25 activity is precisely controlled remains largely unknown. Here, we show that LIM domain-containing proteins, such as FHL1, increase inhibitory CDC25 phosphorylation by forming a complex with CHK2 and CDC25, and sequester CDC25 in the cytoplasm by forming another complex with 14-3-3 and CDC25, resulting in increased radioresistance in cancer cells. FHL1 expression, induced by ionizing irradiation in a SP1- and MLL1-dependent manner, positively correlates with radioresistance in cancer patients. We identify a cell-penetrating 11 amino-acid motif within LIM domains (eLIM) that is sufficient for binding CHK2 and CDC25, reducing the CHK2-CDC25 and CDC25-14-3-3 interaction and enhancing CDC25 activity and cancer radiosensitivity accompanied by mitotic catastrophe and apoptosis. Our results provide novel insight into molecular mechanisms underlying CDC25 activity regulation. LIM protein inhibition or use of eLIM may be new strategies for improving tumour radiosensitivity.

HPV16 E6-E7 induces cancer stem-like cells phenotypes in esophageal squamous cell carcinoma through the activation of PI3K/Akt signaling pathway in vitro and in vivo.

High-risk human papillomavirus (HPV), especially HPV16, correlates with cancerogenesis of human esophageal squamous cell carcinoma (ESCC) and we have reported that HPV16 related with a poor prognosis of ESCC patients in China. We aim to investigate the potential role and mechanism of HPV16 in ESCC development and progress. Our following researches demonstrated that ESCC cells which were stably transfected by HPV16 E6-E7 lentiviral vector showed a remarkable cancer stem-like cells (CSCs) phenotype, such as: migration, invasion, spherogenesis, high expression of CSCs marker in ESCC---p75NTR, chemoresistance, radioresistance, anti-apoptosis ability in vitro and cancerogenesis in vivo. HPV16 E6-E7 induced PI3K/Akt signaling pathway activation and this affect could be effectively inhibited by LY294002, a specific PI3K inhibitor. It was also indicated that the inhibition of PI3K/Akt signaling pathway by PI3K and Akt siRNA reverse the effect which induced by HPV16 E6-E7 in ESCC cells. Taken together, the present study demonstrates that HPV16 E6-E7 promotes CSCs phenotype in ESCC cells through the activation of PI3K/Akt signaling pathway. Targeting the PI3K/Akt signaling pathway in HPV16 positive tissues is an available therapeutic for ESCC patients.

Human papillomavirus 16 infection predicts poor outcome in patients with esophageal squamous cell carcinoma.

BACKGROUND: Previous studies indicate that human papillomavirus 16 (HPV16) infection plays a pivotal role in the etiology of esophageal squamous cell carcinoma (ESCC). We aim to detect the influence of HPV16 infection on ESCC patient prognosis. PATIENTS AND METHODS: Immunohistochemical staining for HPV16 E6 oncoprotein, the low-affinity p75 neurotrophin receptor (p75NTR), and phosphatidylinositol 3-kinase (PI3K) was performed on 103 archived surgical specimens from patients with ESCC and 54 control samples from patients with benign esophageal tumor or inflammatory lesions. All patients were from the Shaan Xi Province, People's Republic of China. RESULTS: HPV16 E6 expression was significantly higher in the ESCC group (P<0.05). HPV16 E6 expression was significantly higher in men than in women (P<0.05). p75NTR expression was higher in those aged >56 years (P<0.05). PI3K expression was higher in those with a more advanced histopathological grade (P<0.05). There was a positive correlation between HPV16 E6 and p75NTR expression (r=0.547, P<0.001) and between p75NTR and PI3K expression (r=0.364, P<0.001). In 100 evaluable patients, the 5-year overall survival (OS) rate was 11%. In patients with ESCC, HPV16 E6 and PI3K expression were negatively correlated with the 3-year OS (P<0.05), 5-year OS (P<0.05), and progression-free survival (P<0.05). CONCLUSION: HPV16 infection likely contributes to the etiology of ESCC patients in Shaan Xi, People's Republic of China. HPV16 infection status and PI3K expression levels could be useful for predicting prognosis in patients with ESCC.