Prognostic model incorporating immune checkpoint genes to predict the immunotherapy efficacy for lung adenocarcinoma: a cohort study integrating machine learning algorithms.

This study aimed to develop and validate a nomogram based on immune checkpoint genes (ICGs) for predicting prognosis and immune checkpoint blockade (ICB) efficacy in lung adenocarcinoma (LUAD) patients. A total of 385 LUAD patients from the TCGA database and 269 LUAD patients in the combined dataset (GSE41272 + GSE50081) were divided into training and validation cohorts, respectively. Three different machine learning algorithms including random forest (RF), least absolute shrinkage and selection operator (LASSO) logistic regression analysis, and support vector machine (SVM) were employed to select the predictive markers from 82 ICGs to construct the prognostic nomogram. The X-tile software was used to stratify patients into high- and low-risk subgroups based on the nomogram-derived risk scores. Differences in functional enrichment and immune infiltration between the two subgroups were assessed using gene set variation analysis (GSVA) and various algorithms. Additionally, three lung cancer cohorts receiving ICB therapy were utilized to evaluate the ability of the model to predict ICB efficacy in the real world. Five ICGs were identified as predictive markers across all three machine learning algorithms, leading to the construction of a nomogram with strong potential for prognosis prediction in both the training and validation cohorts (all AUC values close to 0.800). The patients were divided into high- (risk score ≥ 185.0) and low-risk subgroups (risk score < 185.0). Compared to the high-risk subgroup, the low-risk subgroup exhibited enrichment in immune activation pathways and increased infiltration of activated immune cells, such as CD8 + T cells and M1 macrophages (P < 0.05). Furthermore, the low-risk subgroup had a greater likelihood of benefiting from ICB therapy and longer progression-free survival (PFS) than did the high-risk subgroup (P < 0.05) in the two cohorts receiving ICB therapy. A nomogram based on ICGs was constructed and validated to aid in predicting prognosis and ICB treatment efficacy in LUAD patients.

Predictive Model to Identify the Long Time Survivor in Patients with Glioblastoma: A Cohort Study Integrating Machine Learning Algorithms.

We aimed to develop and validate a predictive model for identifying long-term survivors (LTS) among glioblastoma (GB) patients, defined as those with an overall survival (OS) of more than 3 years. A total of 293 GB patients from CGGA and 169 from TCGA database were assigned to training and validation cohort, respectively. The differences in expression of immune checkpoint genes (ICGs) and immune infiltration landscape were compared between LTS and short time survivor (STS) (OS<1.5 years). The differentially expressed genes (DEGs) and weighted gene co-expression network analysis (WGCNA) were used to identify the genes differentially expressed between LTS and STS. Three different machine learning algorithms were employed to select the predictive genes from the overlapping region of DEGs and WGCNA to construct the nomogram. The comparison between LTS and STS revealed that STS exhibited an immune-resistant status, with higher expression of ICGs (P<0.05) and greater infiltration of immune suppression cells compared to LTS (P<0.05). Four genes, namely, OSMR, FMOD, CXCL14, and TIMP1, were identified and incorporated into the nomogram, which possessed good potential in predicting LTS probability among GB patients both in the training (C-index, 0.791; 0.772-0.817) and validation cohort (C-index, 0.770; 0.751-0.806). STS was found to be more likely to exhibit an immune-cold phenotype. The identified predictive genes were used to construct the nomogram with potential to identify LTS among GB patients.

A Scoring System to Select the Candidates for Adjuvant Chemotherapy Alone in High-Risk Early-Stage Cervical Cancer Patients With Pelvic Lymph Node Metastases After Surgery.

BACKGROUND: The aim was to build a risk scoring system to guide the adjuvant treatment for early-stage cervical cancer patients with pelvic lymph node (LN) metastases after surgery. METHODS: A cohort of 1213 early-stage cervical cancer patients with pelvic LN metastases (T1-2aN1M0) were selected from the NCI SEER database, of which 1040 patients received adjuvant external beam radiotherapy concurrent with chemotherapy (EBRT+Chemo) and 173 patients received adjuvant chemotherapy alone. The Cox regression analysis was applied to identify the risk factors associated with worse survival. The exp (ß) of each independent risk factors from multivariate analysis was assigned to develop the risk scoring system. The total cohort was divided into different risk subgroups accordingly and the efficacy of different adjuvant modalities in each risk subgroups was compared. RESULTS: The patients were divided into 3 risk subgroups (Low-risk: total score <7.20, Middle-risk:7.20≤ total score≤ 8.40, High-risk: total score<8.40) based on the scoring system incorporating 5 independent risk factors. The survival analysis suggested that low-risk (hazard ratio [HR]=1.046, 95% CI: 0.586-1.867; P= 0.879) and middle-risk patients (HR=0.709, 95% CI: 0.459-1.096; P =0.122) could not benefit more from EBRT+Chemo than Chemo alone. However, EBRT+Chemo remained the superiority to Chemo alone in the high-risk subgroup (HR=0.482, 95% CI: 0.294-0.791; P =0.003). CONCLUSION: A risk scoring system has been built to direct the adjuvant treatment for early-stage cervical cancer patients with pelvic LN metastases after surgery, where Chemo alone was totally enough for low-risk and middle-risk patients stratified by the model while EBRT+Chemo was still recommended for patients in the high-risk subgroup.

A population-based risk scoring system to individualize adjuvant treatment for stage IIIC endometrial cancer patients after surgery.

BACKGROUND: To develop a risk scoring system to tailor the adjuvant treatment for stage IIIC EC patients after surgery. METHODS: Data source was from the Surveillance, Epidemiology, and End Results (SEER) registry, where 3251 post-operative stage IIIC EC patients with different adjuvant treatment were included. Cox regression analysis was used to identify risk factors. The exp (ß) of each independent risk factors generating from the cox analysis was used to construct the risk scoring system, which was further utilized to divide the patients into different risk subgroups and the efficacy of different adjuvant modalities in each risk subgroups would be compared accordingly. RESULTS: Six independent risk factors were identified to develop the scoring system, which further divided the patients into three risk subgroups based on the total risk score (Low-risk≤8.46, 8.47 ≤ Middle-risk≤9.94, High-risk≥9.95). This study revealed that CRT was not superior to RT alone (HR:1.208, 95%CI: 0.852-1.741; P = 0.289) or CT alone (HR:1.260, 95%CI: 0.750-2.116; P = 0.382) in Low-risk subgroup. We also observed that CRT had a survival advantage over other treatment modalities in the Middle-risk subgroup (All P < 0.001), but CRT and CT alone to be superimposable in the High-risk subgroup (HR: 1.395, 95%CI: 0.878-2.216; P = 0.159). CONCLUSION: A risk scoring system has been developed to tailor the adjuvant treatment for stage IIIC EC patients after surgery, where RT or CT alone could be a substitute for CRT in Low-risk patients and CT alone was a potential alternative for High-risk patients while CRT remained to be the optimal choice for the Middle-risk patients.

Prognostic model for the exemption of adjuvant chemotherapy in stage IIIC endometrial cancer patients.

Background: This study aimed to develop a nomogram to predict the survival for stage IIIC endometrial cancer (EC) patients with adjuvant radiotherapy (ART) alone and personalize recommendations for the following adjuvant chemotherapy (ACT). Methods: In total, 746 stage IIIC EC patients with ART alone were selected from the Surveillance, Epidemiology, and End Results (SEER) registry. Cox regression analysis was performed to identify independent risk factors. A nomogram was developed accordingly, and the area under the receiver operating characteristic curve (AUC) and C-index were implemented to assess the predictive power. The patients were divided into different risk strata based on the total points derived from the nomogram, and survival probability was compared between each risk stratus and another SEER-based cohort of stage IIIC EC patients receiving ART+ACT (cohort ART+ACT). Results: Five independent predictors were included in the model, which had favorable discriminative power both in the training (C-index: 0.732; 95% CI: 0.704-0.760) and validation cohorts (C-index: 0.731; 95% CI: 0.709-0.753). The patients were divided into three risk strata (low risk <135, 135 ≤ middle risk ≤205, and high risk >205), where low-risk patients had survival advantages over patients from cohort ART+ACT (HR: 0.45, 95% CI: 0.33-0.61, P < 0.001). However, the middle- and high-risk patients were inferior to patients from cohort ART+ACT in survival (P < 0.001). Conclusion: A nomogram was developed to exclusively predict the survival for stage IIIC EC patients with ART alone, based on which the low-risk patients might be perfect candidates to omit the following ACT. However, the middle- and high-risk patients would benefit from the following ACT.

Construction and validation of a prognostic model for stage IIIC endometrial cancer patients after surgery.

BACKGROUND: To explore the most predictive lymph node (LN) scheme for stage IIIC endometrial cancer (EC) patients after hysterectomy and develop a scheme-based nomogram. METHODS: Data from 2626 stage IIIC EC patients, diagnosed between 2010 and 2014, were extracted from the Surveillance, Epidemiology, and End Results (SEER) registry. The predictive ability of four LN schemes was assessed using C-index and Akaike information criterion (AIC). A nomogram based on the most predictive LN scheme was constructed and validated. The comparison of the predictive ability between nomogram and FIGO stage was conducted using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). RESULTS: FIGO stage (stage IIIC1/stage IIIC2) was not an independent risk factor for OS in stage IIIC EC patients (P = 0.672) and log odds of positive lymph nodes (LODDS) had the best predictive ability (C-index: 0.742; AIC: 8228.95). A nomogram based on LODDS was constructed and validated, which had a decent C-index of 0.742 (0.723-0.762). The nomogram showed a better predictive ability than that of the FIGO staging system. CONCLUSION: FIGO IIIC1/FIGO IIIC2 could not differentiate the prognosis for stage IIIC EC patients. We developed and validated a nomogram based on LODDS to predict OS for post-operative patients with stage IIIC EC.

A real-world, population-based study of the trends for incidence and prognosis in high-grade neuroendocrine tumor of cervix.

To explore the incidence and prognosis trends for high-grade cervical neuroendocrine tumor (HGCNET) and construct a nomogram to predict prognosis for HGCNET. Annual age-adjusted incidence of HGCNET from 1975 to 2015 was retrieved from the Surveillance, Epidemiology, and End Results program, the linear regression, poisson regression and annual percentage changes were used to assess the incidence trend. Also, trends for relative survival (RS) and overall survival (OS) in HGCNET patients from 1975 to 2015 were evaluated. From 1988 to 1975, 514 HGCNET patients were selected and divided into two cohorts with a ratio of 7:3. Nomogram to predict OS for these patients was constructed and validated. The incidence trend for HGCNET was unchanged in the past four decades (P = 0.734), but the proportion of HGCNET in diagnosed cervical cancer slightly increased from 0.9% in 1975 to 1.9% in 2015 (P < 0.001). The 5-year RS and OS for HGCNET in the study periods decreased steadily (RS: P = 0.009; OS: P = 0.008). Nomogram incorporating age, T stage, lymph-node positive, distant metastasis and surgery was constructed. The C-index of the nomogram was 0.716 (0.680-0.752), which was higher than the FIGO staging system. The incidence of HGCNET remained unchanged in the past four decades but the proportion of HGCNET has slightly increased. Besides, a steadily decreasing survival for HGCNET was observed in the study periods. A nomogram was constructed to better predict prognosis for HGCNET.

Comparison of Different Lymph Node Staging Schemes for Predicting Survival Outcomes in Node-Positive Endometrioid Endometrial Cancer Patients.

Objective: To compare the prognostic predictive performance of six lymph node (LN) staging schemes: American Joint Committee on Cancer (AJCC) N stage, number of retrieved lymph nodes (NRLN), number of positive lymph nodes (NPLN), number of negative lymph nodes (NNLN), lymph node ratio (LNR), and log odds of positive lymph nodes (LODDS) among node-positive endometrioid endometrial cancer (EEC) patients. Methods: A total of 3,533 patients diagnosed with node-positive EEC between 2010 and 2016 from the Surveillance, Epidemiology, and End Results (SEER) database were retrospectively analyzed. We applied X-tile software to identify the optimal cutoff value for different staging schemes. Univariate and multivariate Cox regression models were used to assess the relationships between different LN schemes and survival outcomes [disease-specific survival (DSS) and overall survival (OS)]. Moreover, Akaike information criterion (AIC) and Harrell concordance index (C-index) were used to evaluate the predictive performance of each scheme in both continuous and categorical patterns. Results: N stage (N1/N2) was not an independent prognostic factor for node-positive EEC patients based on multivariate analysis (DSS: p = 0.235; OS: p = 0.145). Multivariate model incorporating LNR demonstrated the most superior goodness of fit regardless of continuous or categorical pattern. Regarding discrimination power of the models, LNR outperformed other models in categorical pattern (OS: C-index = 0.735; DSS: C-index = 0.737); however, LODDS obtained the highest C-index in continuous pattern (OS: 0.736; DSS: 0.739). Conclusions: N stage (N1/N2) was unable to differentiate the prognosis for node-positive EEC patients in our study. However, LNR and LODDS schemes seemed to have a better predictive performance for these patients than other number-based LN schemes whether in DSS or OS, which revealed that LNR and LODDS should be more helpful in prognosis assessment for node-positive EEC patients than AJCC N stage.

Conditional survival for high-risk early-stage cervical cancer patients with lymph node metastasis after hysterectomy.

BACKGROUND: To estimate conditional survival (CS) for high-risk early-stage cervical cancer patients with lymph node metastasis after hysterectomy. METHODS: 1964 T1-2N1M0 cervical cancer patients who underwent primary hysterectomy from 2004 to 2015 were extracted from the Surveillance, Epidemiology, and End Result (SEER) Program. Univariate and multivariate cox regression analysis were used to identify independent risk factors. 5-year conditional disease-specific survival (CDS5) and 5-year conditional relative survival (CRS5) were estimated. CDS5 and CRS5 stratified by risk factors were further calculated. RESULTS: CDS5 and CRS5 increased from 71.0% and 73.7% at 0-year to 89.2% and 91.7% at 5-year, respectively. Inversely, the actuarial disease-specific survival and RS dropped from 71.0% and 73.7% at 5-year to 63.3% and 67.6% at 10-year, respectively. Patients with unfavorable factors had a bigger gap between actuarial survival and CS. Both CDS5 and CRS5 curves across stratas of each prognostic factor had a tendency to level off with time elapsing. Notably, CRS5 couldn't exceed 95% even after 5-year follow-up except for patients with grade I disease (CRS5 at 5-year: 100%) or tumor size less than 2 cm (CRS5 at 5-year: 96%). CONCLUSION: CS increased over time while actuarial survival decreased as time passed. Patients with unfavorable factors had bigger improvement in CS than those with favorable factors. Excess mortality still existed in these patients after 5-year follow-up compared to the general population except for patients with grade I disease or tumor size <2 cm, who might gradually decrease follow-up times after 5-year.

Methodological challenges to control for immortal time bias in addressing drug effects in type 2 diabetes.

There are multiple biases in using observational studies to examine treatment effects such as those from prevalent drug users, immortal time and drug indications. We used renin angiotensin system (RAS) inhibitors and statins as reference drugs with proven efficacies in randomized clinical trials (RCTs) and examined their effectiveness in the prospective Hong Kong Diabetes Registry using adjustment methods proposed in the literature. Using time-dependent exposures to drug treatments yielded greatly inflated hazard ratios (HR) regarding the treatment effects of these drugs for cardiovascular disease (CVD) in type 2 diabetes. These errors were probably due to changing indications to use these drugs during follow up periods, especially at the time of drug commencement making time-dependent analysis extremely problematic. Using time-fixed analysis with exclusion of immortal time and adjustment for confounders at baseline and/or during follow-up periods, the HR of RAS inhibitors for CVD was comparable to that in RCT. The result supported the use of the Registry for performing pharmacoepidemiological analysis which revealed an attenuated low low-density lipoprotein cholesterol related cancer risk with RAS inhibitors. On the other hand, time-fixed analysis with including immortal time and adjustment for confounders at baseline and/or during follow-up periods, the HR of statins for CVD was similar to that in the RCT. Our results highlight the complexity and difficulty in removing these biases. We call for validations of the methods to cope with immortal time and drug use indications before applying them to particular research questions, so to avoid making erroneous conclusions.

[Complications of hyoid suspension with Repose system on obstructive sleep apnea hypopnea syndrome].

OBJECTIVE: To investigate the complications of hyoid suspension with Repose system on obstructive sleep apnea hypopnea syndrome (OSAHS) and to find out the effective prevention strategies. METHODS: Forty four patients diagnosed by polysomnography as OSAHS were received hyoid suspension with Repose system from June 2005 to July 2009. The intraoperative and postoperative complications were analyzed retrospectively. The patients who reported abnormal swallowing were evaluated with water drinking test and video fluorography swallow study to assess biomechanical changes in swallowing. RESULTS: Incidence rate of perioperative titanium nial amotio was 15.9% (7/44). It was avoided by implanting titanium nial again. No perioperative complications occurred such as injury of superior laryngeal nerve, blood vessel and thyrohyoid membrane, fracture of hyoid bone, suture break. All patients developed dysfunctions of pronunciation and swallowing, edema of mouth floor, dysfunctions of movement of tongue that could relieved gradually after three days on most patients. No postoperative complications occurred such as titanium nial amotio, fat liquoring, edema of mouth floor, hematoma, infection, foreign body reaction, injury of root apex of anterior tooth, dysfunctions of movement and sensation of tongue, suture break and death. All patients had over 2 year postoperative follow up. Incidence rate of dysfunctions of pronunciation and swallowing was 4.5% (2/44) and 15.9% (7/44) respectively. Main situation for dysfunctions of pronunciation was speaking ambiguity. Major manifestations of swallowing abnormalities were occasional aspiration, food going down the wrong tube, food becoming stuck in the throat, deglutition with bowing head. Seven patients who had abnormal swallowing possessed normal water drinking test and occurred asynersis of hyoid movement and laryngeal elevation, but aspiration were not observed. Three patients presented obvious stagnation in epiglottic vallecula and sinus piriformis. CONCLUSIONS: Most complications of hyoid suspension with Repose system may be avoided or recovered on short term. Postoperative dysfunctions of pronunciation and swallowing may exist for a long time.

Diabetes, insulin and cancer risk.

There is a consensus that both type 1 and type 2 diabetes are associated with a spectrum of cancers but the underlying mechanisms are largely unknown. On the other hand, there are ongoing debates about the risk association of insulin use with cancer. We have briefly reviewed recent related research on exploration of risk factors for cancer and pharmacoepidemiological investigations into drug use in diabetes on the risk of cancer, as well as the current understanding of metabolic pathways implicated in intermediary metabolism and cellular growth. Based on the novel findings from the Hong Kong Diabetes Registry and consistent experimental evidence, we argue that use of insulin to control hyperglycemia is unlikely to contribute to increased cancer risk and that dysregulations in the AMP-activated protein kinase pathway due to reduced insulin action and insulin resistance, the insulin-like growth factor-1 (IGF-1)-cholesterol synthesis pathway and renin-angiotensin system, presumably due to reduced insulin secretion and hyperglycemia, may play causal roles in the increased risk of cancer in diabetes. Further exploration into the possible causal relationships between abnormalities of these pathways and the risk of cancer in diabetes is warranted.

[Complications of tongue base reduction with radiofrequency tissue ablation on obstructive sleep apnea hypopnea syndrome].

OBJECTIVE: To investigate the complications of tongue base reduction with radiofrequency tissue ablation on patients with obstructive sleep apnea hypopnea syndrome (OSAHS) and find out the effective prevention strategies. METHODS: One hundred and ninety three OSAHS patients diagnosed by polysomnography were received tongue base reduction with radiofrequency tissue ablation between March 2008 and December 2009. The intraoperative and postoperative complications including bleeding, hematoma of tongue base, abscess of tongue base, altered taste, tongue numbness, deviation of tongue extension movement, dysfunctions of pronunciation and swallowing as well as the managements were analyzed retrospectively. RESULTS: No perioperative complications occurred. There were 186 cases with postoperative pain (96.4%), 155 cases with submandibular edema (80.3%). Nocturnal sudden cardiac death was encountered in 1 case and secondary bleeding in 1 case. There was no ulceration of tongue base mucose, hematoma or abscess of tongue base, altered taste, tongue numbness, tongue deviations, speech, swallowing and taste disorder after operation. The scale of postoperative pain claimed by patients was ranged between mild to moderate. Diclofenac suppository had analgesic effect for these patients. The quantity of bleeding in patient with secondary hemorrhage was so little that after proper treatment the bleeding was stopped and never happened again. Patient with nocturnal sudden cardiac death occurred at thirty-seven hours after operation, because of swelling and pain of tongue base aggravated sleep apnea and night hypoxemia inducing fatal arrhythmia. CONCLUSIONS: Postoperative pain and submandibular edema were 2 most common postoperative complications which can be easily controlled by antibiotics, Glucocorticoids and Diclofenac suppository. For those severe OSAHS patients accompanied by cardiopulmonary diseases, the tongue base reduction with radiofrequency tissue ablation can induce nocturnal sudden cardiac death. It is important to pay more attention on arrhythmias at night in severe OSAHS patients.