Dangers of using "optimal" cutpoints in the evaluation of prognostic factors

The value of metabolic parameters and textural analysis in predicting prognosis in locally advanced cervical cancer treated with chemoradiotherapy

Objective

The aim of the study was to assess the impact of clinical and metabolic parameters derived from ¹⁸F-FDG PET/CT (positron emission tomography–computed tomography) in patients with locally advanced cervical cancer (LACC) on prognosis.

Methods

Patients with LACC of stage IB2-IVA treated by primary radiochemotherapy followed by brachytherapy were enrolled in this retrospective study. Indexes derived from standardized uptake value (SUV), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and textural features of the primary tumor were measured for each patient. Overall survival (OS) and recurrence-free survival (RFS) rates were calculated according to Kaplan–Meier and survival curves were compared using the log-rank test. Uni- and multivariate analyses were performed using the Cox regression model.

Results

A total of 116 patients were included. Median follow-up was 58 months (range: 1–129). A total of 36 (31%) patients died. Five-year OS and RFS rates were 69 and 60%, respectively. Univariate analyses indicated that FIGO stage, the presence of hydronephrosis, high CYFRA 21.1 levels, and textural features had a significant impact on OS and RFS. MTV as well as SCC-Ag concentration were also significantly associated with OS. On multivariate analysis, the presence of hydronephrosis, CYFRA 21.1, and sphericity were independent prognostics factors for OS and RFS. Also, SCC-Ag level, MTV, and GLZLM (gray-level zone length matrix) ZLNU (zone length non-uniformity) were significantly associated with OS.

Conclusion

Classical prognostic factors and tumor heterogeneity on pretreatment PET/CT were significantly associated with prognosis in patients with LACC.

Performance Evaluation of Parametric and Nonparametric Methods When Assessing Effect Measure Modification

Effect measure modification is often evaluated using parametric models. These models, although efficient when correctly specified, make strong parametric assumptions. While nonparametric models avoid important functional form assumptions, they often require larger samples to achieve a given accuracy. We conducted a simulation study to evaluate performance tradeoffs between correctly specified parametric and nonparametric models to detect effect modification of a binary exposure by both binary and continuous modifiers. We evaluated generalized linear models and doubly robust (DR) estimators, with and without sample splitting. Continuous modifiers were modeled with cubic splines, fractional polynomials, and nonparametric DR-learner. For binary modifiers, generalized linear models showed the greatest power to detect effect modification, ranging from 0.42 to 1.00 in the worst and best scenario, respectively. Augmented inverse probability weighting had the lowest power, with an increase of 23% when using sample splitting. For continuous modifiers, the DR-learner was comparable to flexible parametric models in capturing quadratic and nonlinear monotonic functions. However, for nonlinear, nonmonotonic functions, the DR-learner had lower integrated bias than splines and fractional polynomials, with values of 141.3, 251.7, and 209.0, respectively. Our findings suggest comparable performance between nonparametric and correctly specified parametric models in evaluating effect modification.

Periodic Repolarization Dynamics Identifies ICD-responders in Non-ischemic Cardiomyopathy: A DANISH Substudy

Background: Identification of patients with non-ischemic cardiomyopathy who benefit from prophylactic implantation of a cardioverter-defibrillator (ICD) remains an unmet clinical need. We hypothesized that periodic repolarization dynamics (PRD), a marker of repolarization instability associated with sympathetic activity, could be used to identify patients that benefit from prophylactic ICD-implantation. Methods: Heart-failure (DANISH) study, in which patients with non-ischemic cardiomyopathy, left-ventricular ejection fraction (LVEF) ≤35% and elevated N-terminal pro-brain natriuretic peptides (NT-proBNP) were randomized to ICD-implantation or control group. Patients were included in the PRD-substudy if they had a 24-hour Holter monitor recording at baseline with technically acceptable ECG signals during the night hours (00:00-06.00 AM). PRD was assessed using wavelet analysis according to previously validated methods. Primary endpoint was all-cause mortality. Cox-regression models were adjusted for age, sex, NT-proBNP, estimated glomerular filtration rate, LVEF, atrial fibrillation, ventricular pacing, diabetes mellitus, cardiac resynchronization therapy and mean heart rate. We proposed PRD ≥10deg² as exploratory cut-off value for ICD-implantation. Results: Seven-hundred and forty-eight of the 1,116 DANISH patients qualified for the PRD-substudy. During a mean follow-up period of 5.1±2.0 years, 82 of 385 patients died in the ICD group and 85 of 363 patients died in the control group (p-value=0.40). In Cox-regression analysis, PRD was independently associated with mortality (HR 1.28 [1.09-1.50] per SD increase; p-value = 0.003). Moreover, PRD was significantly associated with mortality in the control group (HR 1.51 [1.25-1.81]; p<0.001) but not in the ICD-group 1.04 [0.83-1.54]; p-value=0.71). There was a significant interaction between PRD and the effect of ICD-implantation on mortality (p-value 0.008), with patients with higher PRD having the greater benefit in terms of mortality reduction. ICD-implantation was associated with an absolute mortality reduction of 17.5% in the 280 patients with PRD ≥10deg² (HR 0.54 [0.34-0.84]; p-value=0.006; number needed to treat 6), but not in the 468 patients with PRD<10deg² (HR 1.17 [0.77-1.78]; p-value=0.46; p-value for interaction 0.01). Conclusions: Increased PRD identified patients with non-ischemic cardiomyopathy, where prophylactic ICD-implantation led to significant mortality reduction.

Prognostic value of selected one-, two- and three-dimensional and Doppler echocardiographic methods to assess severity in dogs with myxomatous mitral valve disease

INTRODUCTION

Although challenging, predicting outcomes in dogs with myxomatous mitral valve disease (MMVD) is of importance to owners and veterinarians. Our aim was to compare the prognostic value of selected one-, two-, and three-dimensional and Doppler echocardiographic methods to predict cardiac-related and all-cause mortality in MMVD dogs. We hypothesize that three-dimensional methods would better predict prognosis than one- and two-dimensional methods.

ANIMALS

One-hundred thirty-eight privately-owned dogs with MMVD.

MATERIALS AND METHODS

Cox proportional Hazard analyses and Kaplan-Meier curves were used to investigate the predictive value of 14 variables; left atrial (LA) volume indexed to body weight (BW) measured by real-time three-dimensional echocardiography (RT3DE) and calculated using Simpson's modified method of discs (SMOD) and the area-length method; LA diameter in short-axis and long-axis to aortic in short-axis ratio, effective regurgitant orifice area indexed to body surface area measured in RT3DE en face view and calculated using four-chamber (4Ch) and two-chamber views alone or in combination; percentage increase in left ventricular end-diastolic and systolic internal diameters; fractional shortening; E wave velocity; regurgitant jet area/LA area.

RESULTS

All 14 variables were significantly predictive of cardiac-related mortality, and 11 were predictive of all-cause mortality. The prognosis was best predicted by LA volume/BW estimated by SMOD or RT3DE, consistently showing the highest predictive value in all analyses.

CONCLUSIONS

Left atrial volume calculated by SMOD showed a similar predictive value compared to RT3DE. Performing SMOD from apical 4Ch images should be considered an alternative to RT3DE for echocardiographic examinations where prognostication of disease outcome is sought.

Circulating Tumor Cells Enumeration from the Portal Vein for Risk Stratification in Early Pancreatic Cancer Patients

Simple Summary

Effective biomarkers are needed to enable personalized medicine for pancreatic cancer patients. This study analyzes the prognostic value, in early pancreatic cancer, of circulating tumor cells and clusters from the central venous catheter and portal blood. Circulating tumor cells were isolated using an immunomagnetic selection and were detected by microscopy using immunocytochemistry staining. In conclusion, the circulating tumor cell number in portal blood identifies a death risk in patients with early pancreatic cancer.

Abstract

Background. Effective biomarkers are needed to enable personalized medicine for pancreatic cancer patients. This study analyzes the prognostic value, in early pancreatic cancer, of single circulating tumor cell (CTC) and CTC clusters from the central venous catheter (CVC) and portal blood (PV). Methods. In total, 7 mL of PV and CVC blood from 35 patients with early pancreatic cancer were analyzed. CTC were isolated using a positive immunomagnetic selection. The detection and identification of CTC were performed by immunocytochemistry (ICC) and were analyzed by Epi-fluorescence and confocal microscopy. Results. CTC and the clusters were detected both in PV and CVC. In both samples, the CTC number per cluster was higher in patients with grade three or poorly differentiated tumors (G3) than in patients with well (G1) or moderately (G2) differentiated. Patients with fewer than 185 CTC in PV exhibited a longer OS than patients with more than 185 CTC (24.5 vs. 10.0 months; p = 0.018). Similarly, patients with fewer than 15 clusters in PV showed a longer OS than patients with more than 15 clusters (19 vs. 10 months; p = 0.004). These significant correlations were not observed in CVC analyses. Conclusions. CTC presence in PV could be an important prognostic factor to predict poor prognosis in early pancreatic cancer. In addition, the number of clustered-CTC correlate to a tumor negative differentiation degree and, therefore, could be used as a diagnostic biomarker for pancreatic cancer.

Cut points and contexts

In research, policy, and practice, continuous variables are often categorized. Statisticians have generally advised against categorization for many reasons, such as loss of information and precision as well as distortion of estimated statistics. Here, a different kind of problem with categorization is considered: the idea that, for a given continuous variable, there is a unique set of cut points that is the objectively correct or best categorization. It is shown that this is unlikely to be the case because categorized variables typically exist in webs of statistical relationships with other variables. The choice of cut points for a categorized variable can influence the values of many statistics relating that variable to others. This essay explores the substantive trade‐offs that can arise between different possible cut points to categorize a continuous variable, making it difficult to say that any particular categorization is objectively best. Limitations of different approaches to selecting cut points are discussed. Contextual trade‐offs may often be an argument against categorization. At the very least, such trade‐offs mean that research inferences, or decisions about policy or practice, that involve categorized variables should be framed and acted upon with flexibility and humility.

In practical settings, the choice of cut points for categorizing a continuous variable is likely to entail trade‐offs across multiple statistical relationships between the categorized variable and other variables. These trade‐offs mean that no single categorization is objectively best or correct.

Clinicopathological significance of HSP70 expression in gastric cancer: a systematic review and meta-analysis

Background

Heat shock protein 70 (HSP70) has been associated with the clinicopathological characteristics and prognosis of many cancers types, implying that it is a potential cancer biomarker. However, no consensus has been reached regarding its clinicopathological and prognostic significance in patients with gastric cancer. To address this gap, we performed a systematic review and meta-analysis.

Methods

We searched PubMed, Embase, and the Cochrane Library for full-text literature according to the eligibility criteria. We used the odds ratio and hazard ratio as the suitable parameters to evaluate the clinicopathological and prognostic significance of HSP70. The statistical analysis was performed using STATA 15.0.

Results

After inclusion and exclusion of studies based on the eligibility criteria, data of 1,307 patients with gastric cancer from 9 studies were finally included. The pooled outcomes implied that HSP70 expression was significantly correlated with higher differentiation degrees, intestinal gastric cancer, and lymphovascular invasion but not with age, gender, depth of invasion, Helicobacter pylori infection, lymph node invasion, TNM stages, and metastasis. The pooled HR showed no significant correlation between HSP70 expression and overall survival of gastric cancer patients.

Conclusions

Our meta-analysis showed that HSP70 plays a complicated role in the development of gastric cancer. It may be directly engaged in tumour differentiation and distant invasion but cannot be considered a biomarker for predicting the prognosis of gastric cancer.

Anxiety and Perceived Risk in Red Cross Volunteer Personnel Facing the Coronavirus Disease 2019 Pandemic

In the current situation of sanitary emergencies, humanitarian organizations and their volunteers are playing an important role in the coronavirus disease 2019 (COVID-19) pandemic. A study is proposed that includes a network of volunteers who perform humanitarian activities during the COVID-19 pandemic to assess anxiety, perceived risk, and response behaviors and to explore their relationship with sociodemographic variables. For data collection, an online questionnaire was developed through the Google Forms® platform, where the perceived risk, anxiety, and behavioral responses of the general population to the Influenza A (H1N1) pandemic were assessed. The survey presented is a modified version of that survey adapted for COVID-19. This adaptation was endorsed by an experts committee made up of the health chief of the Ecuadorian Red Cross, the focus point of operations from the International Federation of the Red Cross in Ecuador, and a member from the Health Unit of the Americas Regional Office of the International Federation of the Red Cross. A significant relationship has been shown between the job situation and perceived risk and anxiety, being the staff who worked full time away from home, which was exposed to greater risk and anxiety. Both perceived risk and perceived anxiety are very high (according to a 5-point Likert scale). Knowing these data from this first-line personnel will allow adopting measures that could be beneficial for stress management and, therefore, contribute to the well-being and support of these humanitarian and volunteer organizations in the worldwide response to COVID-1 9.

The impact of G protein-coupled oestrogen receptor 1 on male breast cancer: a retrospective analysis

Introduction

The G protein-coupled oestrogen receptor 1 (GPER-1) is a potential prognostic marker in breast cancer. However, its role in male breast cancer (MBC) is still unknown. This study evaluates the expression of GPER-1 in MBC samples and correlates these data with clinical and pathological parameters including patients' survival.

Material and methods

For this retrospective analysis of a prospectively maintained cohort of patients with MBC, we examined 161 specimens for GPER-1 expression using immunohistochemistry. An immunoreactive score (IRS) was calculated based on staining intensity and the percentage of positive tumour cells. Then, we correlated GPER-1 IRS with clinical and pathological parameters, and overall and relapse-free survival.

Results

About 40% of MBC samples were positive for GPER-1 expression (IRS ≥ 4). There was no significant correlation with clinicopathological parameters, such as hormone receptor status or grading. However, a statistical trend was observed for tumour size (≥ 2 cm, p = 0.093). Kaplan-Meier survival analysis revealed no significant correlation with relapse-free survival. However, there was a significant correlation with overall survival, but when we adjusted the log-rank p-value to compensate for the cut-off point optimization method, it rose above 0.1. Additionally, GPER-1-positive patients were older at diagnosis. When adjusted for age by multivariable Cox regression analysis, the significance of GPER-1 status for survival was further reduced.

Conclusions

We found no significant prognostic value of GPER-1 in this MBC cohort as anticipated from studies on female BC. Future studies with higher sample size are needed to further verify a potential sex-specific role of GPER-1.

Predicting personalized cumulative live birth following in vitro fertilization

OBJECTIVE

To develop in vitro fertilization (IVF) prediction models to estimate the individualized chance of cumulative live birth at two time points: pretreatment (i.e., before starting the first complete cycle of IVF) and posttreatment (i.e., before starting the second complete cycle of IVF in those couples whose first complete cycle was unsuccessful).

DESIGN

Population-based cohort study.

SETTING

National data from the Society for Assisted Reproductive Technology (SART) Clinic Outcome Reporting System.

PATIENT(S)

Based on 88,614 women who commenced IVF treatment using their own eggs and partner's sperm in SART member clinics.

INTERVENTION(S)

Not applicable.

MAIN OUTCOME MEASURE(S)

The pretreatment model estimated the cumulative chance of a live birth over a maximum of three complete cycles of IVF, whereas the posttreatment model did so over the second and third complete cycles. One complete cycle included all fresh and frozen embryo transfers resulting from one episode of ovarian stimulation. We considered the first live birth episode, including singletons and multiple births.

RESULT(S)

Pretreatment predictors included woman's age (35 years vs. 25 years, adjusted odds ratio 0.69, 95% confidence interval 0.66-0.73) and body mass index (35 kg/m² vs. 25 kg/m², adjusted odds ratio 0.75, 95% confidence interval 0.72-0.78). The posttreatment model additionally included the number of eggs from the first complete cycle (15 vs. 9 eggs, adjusted odds ratio 1.10, 95% confidence interval 1.03-1.18). According to the pretreatment model, a nulliparous woman aged 34 years with a body mass index of 23.3 kg/m², male partner infertility, and an antimüllerian hormone level of 3 ng/mL has a 61.7% chance of having a live birth over her first complete cycle of IVF (and a cumulative chance over three complete cycles of 88.8%). If a live birth is not achieved, according to the posttreatment model, her chance of having a live birth over the second complete cycle 1 year later (age 35 years, number of eggs 7) is 42.9%. The C-statistic for all models was between 0.71 and 0.73.

CONCLUSION(S)

The focus of previous IVF prediction models based on US data has been cumulative live birth excluding cycles involving frozen embryos. These novel prediction models provide clinically relevant estimates that could help clinicians and couples plan IVF treatment at different points in time.

Data augmentation based on waterfall plots to increase value of response data generated by small single arm Phase II trials

Waterfall plots have been increasingly used to visualize tumor response measurements in Phase II clinical trials. Despite the growing popularity of waterfall plots, quantitative summaries and distribution features of the data indicating antitumor activities are typically not reported. Statistical summaries from the raw and augmented data may provide valuable information for understanding such features. This issue has not been discussed adequately in the literature or fully recognized within the oncology community. In this article, we propose to augment the data using a statistical distribution system. Summary statistics of the data set corresponding to waterfall plot can be calculated using the original sample of the tumor changes or the augmentation sample, which may lead to additional insights into the treatment effect. We demonstrate the proposed method in numerical studies and in a Phase II clinical trial investigating the efficacy of a treatment for ovarian carcinoma. We recommend the proposed statistical analyses for making inferences in addition to the waterfall plot visualization.

Downregulation of MTSS1 in acute myeloid leukemia is associated with a poor prognosis, chemotherapy resistance, and disease aggressiveness

Despite recent approval of targeted drugs for acute myeloid leukemia (AML) therapy, chemotherapy with cytosine arabinoside and anthracyclines remains an important pillar of treatment. Both primary and secondary resistance are frequent and associated with poor survival, yet the underlying molecular mechanisms are incompletely understood. In previous work, we identified genes deregulated between diagnosis and relapse of AML, corresponding to therapy naïve and resistant states, respectively. Among them was MTSS1, whose downregulation is known to enhance aggressiveness of solid tumors. Here we show that low MTSS1 expression at diagnosis was associated with a poor prognosis in AML. MTSS1 expression was regulated by promoter methylation, and reduced by cytosine arabinoside and the anthracycline daunorubicin. Experimental downregulation of MTSS1 affected the expression of numerous genes. It induced the DNA damage response kinase WEE1, and rendered human AML cell lines more resistant to cytosine arabinoside, daunorubicin, and other anti-cancer drugs. Mtss1 knockdown in murine MLL-AF9-driven AML substantially decreased disease latency, and increased leukemic burden and ex vivo chemotherapy resistance. In summary, low MTSS1 expression represents a novel factor contributing to disease aggressiveness, therapy resistance, and poor outcome in AML.

A New Prognostic Index PDPI for the Risk of Pneumonia Among Patients With Diabetes

Objective

Risk factors for the development of pneumonia among patients with diabetes mellitus are unclear. The aim of our study was to elucidate the potential risk factors and attempt to predict the probability of pneumonia based on the history of diabetes.

Methods

We performed a population-based, prospective multicenter cohort study of 1,043 adult patients with diabetes in China during 2017–2019. Demographic information, comorbidities, or laboratory examinations were collected.

Results

The study included 417 diabetic patients with pneumonia and 626 no-pneumonia-onset diabetic patients. The predictive risk factors were chosen on the basis of a multivariate logistic regression model to predict pneumonia among patients with diabetes including male sex [odds ratio (OR) = 1.72, 95% confidence interval (CI): 1.27–2.33, p < 0.001], age ≥ 75 years (OR = 2.31, 95% CI: 1.61–3.31, p < 0.001), body mass index < 25 (OR = 2.59, 95% CI: 1.92–3.50, p < 0.001), chronic obstructive pulmonary disease (OR = 6.58, 95% CI: 2.09–20.7, p = 0.001), hypertension (OR = 4.27, 95% CI: 3.12–5.85, p < 0.001), coronary heart disease (OR = 2.98, 95% CI: 1.61–5.52, p < 0.001), renal failure (OR = 1.82, 95% CI: 1.002–3.29, p = 0.049), cancer (OR = 3.57, 95% CI: 1.80–7.06, p < 0.001), use of insulin (OR = 2.28, 95% CI: 1.60–3.25, p < 0.001), and hemoglobin A1c ≥ 9% (OR = 2.70, 95% CI: 1.89–3.85, p < 0.001). A predictive nomogram was established. This model showed c-statistics of 0.811, and sensitivity and specificity were 0.717 and 0.780, respectively, under cut-off of 125 score.

Conclusion

We designed a clinically predictive tool for assessing the risk of pneumonia among adult patients with diabetes. This tool stratifies patients into relevant risk categories and may provide a basis for individually tailored intervention for the purpose of early prevention.

Clinicopathological and Prognostic Characteristics in Dedifferentiated/Poorly Differentiated Chordomas: A Pooled Analysis of Individual Patient Data From 58 Studies and Comparison With Conventional Chordomas

Background

Currently, the clinicopathological and prognostic characteristics of dedifferentiated chordoma (DC) and poorly differentiated chordoma (PDC) remain poorly understood. In this study, we sought to characterize clinicopathological parameters in a large PDC/DC cohort and determine their correlations with progression-free survival (PFS) and overall survival (OS) of patients. We also attempted to compare clinical features between PDC/DC and conventional chordoma (CC).

Methods

Literature searches (from inception to June 01, 2020) using Medline, Embase, Google Scholar and Wanfang databases were conducted to identify eligible studies according to predefined criteria. The local database at our center was also retrospectively reviewed to include CC patients for comparative analysis.

Results

Fifty-eight studies from the literature and 90 CC patients from our local institute were identified; in total, 54 PDC patients and 96 DC patients were analyzed. Overall, PDC or DC had distinct characteristics from CC, while PDC and DC shared similar clinical features. Adjuvant radiotherapy and chemotherapy were associated with both PFS and OS in PDC patients in the univariate and/or multivariate analyses. In the DC cohort, tumor resection type, adjuvant chemotherapy and tumor dedifferentiation components significantly affected PFS, whereas none of them were predictive of outcome in the multivariate analysis. By analyzing OS, we found that surgery, resection type and the time to dedifferentiation predicted the survival of DC patients; however, only surgery remained significant after adjusting for other covariables.

Conclusions

These data may offer useful information to better understand the clinical characteristics of PDC/DC and may be helpful in improving the outcome prediction of patients.

Phase I prognostic online (PIPO): A web tool to improve patient selection for oncology early phase clinical trials

PURPOSE

Patient selection in phase 1 clinical trials (Ph1t) continues to be a challenge. The aim of this study was to develop a user-friendly prognostic calculator for predicting overall survival (OS) outcomes in patients to be included in Ph1t with immune checkpoint inhibitors (ICIs) or targeted agents (TAs) based on clinical parameters assessed at baseline.

METHODS

Using a training cohort with consecutive patients from the VHIO phase 1 unit, we constructed a prognostic model to predict median OS (mOS) as a primary endpoint and 3-month (3m) OS rate as a secondary endpoint. The model was validated in an internal cohort after temporal data splitting and represented as a web application.

RESULTS

We recruited 799 patients (training and validation sets, 558 and 241, respectively). Median follow-up was 21.2 months (m), mOS was 10.2 m (95% CI, 9.3-12.7) for ICIs cohort and 7.7 m (95% CI, 6.6-8.6) for TAs cohort. In the multivariable analysis, six prognostic variables were independently associated with OS - ECOG, number of metastatic sites, presence of liver metastases, derived neutrophils/(leukocytes minus neutrophils) ratio [dNLR], albumin and lactate dehydrogenase (LDH) levels. The phase 1 prognostic online (PIPO) calculator showed adequate discrimination and calibration performance for OS, with C-statistics of 0.71 (95% CI 0.64-0.78) in the validation set. The overall accuracy of the model for 3m OS prediction was 87.2% (95% CI 85%-90%).

CONCLUSIONS

PIPO is a user-friendly objective and interactive tool to calculate specific survival probabilities for each patient before enrolment in a Ph1t. The tool is available at https://pipo.vhio.net/.

Expression of mammalian sterile 20-like kinase 1 and 2 and Yes-associated protein 1 proteins in triple-negative breast cancer and the clinicopathological significance

BACKGROUND AND AIM

Mammalian sterile 20-like kinase 1 and 2 (MST1/2) and Yes-associated protein 1 (YAP1) are the core molecules of the Hippo signaling pathway, which have been found to be unbalanced in the occurrence of tumors and promote the development of the lesions. The present study aimed to investigate the expression of MST1/2 and YAP1 proteins in triple-negative breast cancer (TNBC) and their clinicopathological significance.

METHODS

Immunohistochemistry was used to detect the expression level of protein in tissues. According to the percentage of positive cells and staining intensity, the expression intensity of MST1/2 and YAP1 proteins in the tissue samples was scored, and the correlation between MST1/2 and the clinicopathological features of TNBC were discussed.

RESULTS

The expression of MST1/2 and YAP1 was associated with histological grade, metastasis, lymph node metastasis stage, and tumor node metastasis stage. The overexpression of YAP1 predicted a poor prognosis in terms of overall survival and disease-free survival time. The MST1/2 expression was associated with improved overall survival and disease free survival of the patients.

CONCLUSION

MST1/2 and YAP1 may be used as prognostic indicators to evaluate the recurrence of TNBC and might become one of the new targets for breast cancer treatment.

The potential value of the neutrophil to lymphocyte ratio for early differential diagnosis and prognosis assessment in patients with aortic dissection

OBJECTIVES

The aim of the study was to assess the diagnostic performance and clinical utility of the neutrophil to lymphocyte ratio (NLR) in patients with suspected aortic dissection (AD) and investigate its role in predicting in-hospital mortality in AD.

METHODS

NLR values were calculated and compared in 467 consecutive patients with initially suspected AD. A receiver operating characteristic (ROC) curve analysis with the area under the curve (AUC) was used to evaluate the discriminative accuracy and predictive capability of the NLR for AD. Clinical utility was determined by decision curve analysis (DCA). The association between NLR and in-hospital mortality was investigated by logistic regression analyses in patients diagnosed with AD.

RESULTS

The NLR was significantly elevated in patients with AD, and the optimal cut-off point for the NLR to distinguish AD from other acute chest pain diseases was 5.67 [AUC (95% CI): 0.877 (0.844-0.905)]. We recommended an NLR of 2.43 as the appropriate cut-off point with 96.9% sensitivity and a negative likelihood ratio (LR) of 0.09 to satisfy clinical requirements for diagnosis. DCA showed that the use of NLR had a positive net benefit. The deceased patients with AD had a higher NLR than the discharged patients. Moreover, the NLR was an independent predictor of in-hospital mortality for AD [adjusted odds ratio (OR): 1.084 (1.029-1.142)], and patients with higher NLR values tended to have a higher risk of in-hospital mortality. The optimal cut-off point for the NLR to predict in-hospital mortality was 9.20 [AUC (95% CI): 0.695 (0.619-0.765)].

CONCLUSIONS

As an easily available and inexpensive parameter, the NLR could serve as a valuable clinical biomarker for early differential diagnosis and prognosis assessment of AD.

Appropriate analyses of bimodal substance use frequency outcomes: a mixture model approach

Background: In addiction research, outcome measures are often characterized by bimodal distributions. One mode can be for individuals with low substance use and the other mode for individuals with high substance use. Applying standard statistical procedures to bimodal data may result in invalid inference. Mixture models are appropriate for bimodal data because they assume that the sampled population is composed of several underlying subpopulations.Objectives: To introduce a novel mixture modeling approach to analyze bimodal substance use frequency data.Methods: We reviewed existing models used to analyze substance use frequency outcomes and developed multiple alternative variants of a finite mixture model. We applied all methods to data from a randomized controlled study in which 30-day alcohol abstinence was the primary outcome. Study data included 73 individuals (38 men and 35 women). Models were implemented in the software packages SAS, Stata, and Stan.Results: Shortcomings of existing approaches include: 1) inability to model outcomes with multiple modes, 2) invalid statistical inferences, including anti-conservative p-values, 3) sensitivity of results to the arbitrary choice to model days of substance use versus days of substance abstention, and 4) generation of predictions outside the range of common substance use frequency outcomes. Our mixture model variants avoided all of these shortcomings.Conclusions: Standard models of substance use frequency outcomes can be problematic, sometimes overstating treatment effectiveness. The mixture models developed improve the analysis of bimodal substance use frequency.

Reappraisal of the T Category for Solitary Intrahepatic Cholangiocarcinoma by Tumor Size in 611 Early-Stage (T1-2N0M0) Patients After Hepatectomy: a Surveillance, Epidemiology, and End Results (SEER) Analysis

BACKGROUND

The association between tumor size and survival in patients with intrahepatic cholangiocarcinoma (ICC) after hepatectomy is controversial, and the T category in the American Joint Committee on Cancer (AJCC) stage for ICC is a topic of debate.

METHODS

Data from 611 T1-2N0M0 ICC patients classified by the AJCC 8th edition who underwent hepatectomy were extracted from the Surveillance, Epidemiology, and End Results (SEER) database during 1988-2015. Cancer-specific survival was evaluated using Kaplan-Meier analysis. The optimal cutoff value of solitary tumor size was used an adjusted p value approach to discriminating patient survival.

RESULTS

In the AJCC 8th staging system, using a 5-cm cut-off value of tumor size for solitary ICC without vascular invasion (S/VI-) was not associated with survival in T1 category (p = 0.201), and multifocal ICC with vascular invasion had a worse survival than solitary ICC with vascular invasion (S/VI+) in T2 category (p = 0.014). Tumor size was a prognostic factor for both S/VI- and S/VI+, the optimal cutoff value of tumor size was obtained 8 cm for S/VI- and 3 cm for S/VI+. S/VI- ≤ 8 cm had a similar survival to S/VI+ ≤ 3 cm (p = 0.126), S/VI- > 8 cm had a similar survival to S/VI+ > 3 cm (p = 0.655), and multifocal ICC had a similar survival with S/VI- > 8 cm (p = 0.159) and S/VI+ > 3 cm (p = 0.196). When the cohort was divided into two groups-new T1 (S/VI- ≤ 8 cm and S/VI+ ≤ 3 cm) and new T2 (S/VI- > 8 cm, S/VI+ > 3 cm and multifocal ICC)-significant survival difference was observed (p < 0.0001).

CONCLUSIONS

The discriminatory power of the AJCC 8th edition for solitary ICC could be further enhanced by subdividing tumors according to size and vascular invasion (8 cm for S/VI- and 3 cm for S/VI+).

Pleuroparenchymal fibroelastosis in idiopathic pulmonary fibrosis: Survival analysis using visual and computer-based computed tomography assessment

BACKGROUND

Idiopathic pulmonary fibrosis (IPF) and pleuroparenchymal fibroelastosis (PPFE) are known to have poor outcomes but detailed examinations of prognostic significance of an association between these morphologic processes are lacking.

METHODS

Retrospective observational study of independent derivation and validation cohorts of IPF populations. Upper-lobe PPFE extent was scored visually (vPPFE) as categories of absent, moderate, marked. Computerised upper-zone PPFE extent (cPPFE) was examined continuously and using a threshold of 2·5% pleural surface area. vPPFE and cPPFE were evaluated against 1-year FVC decline (estimated using mixed-effects models) and mortality. Multivariable models were adjusted for age, gender, smoking history, antifibrotic treatment and diffusion capacity for carbon monoxide.

FINDINGS

PPFE prevalence was 49% (derivation cohort, n = 142) and 72% (validation cohort, n = 145). vPPFE marginally contributed 3-14% to variance in interstitial lung disease (ILD) severity across both cohorts.In multivariable models, marked vPPFE was independently associated with 1-year FVC decline (derivation: regression coefficient 18·3, 95 CI 8·47-28·2%; validation: 7·51, 1·85-13·2%) and mortality (derivation: hazard ratio [HR] 7·70, 95% CI 3·50-16·9; validation: HR 3·01, 1·33-6·81). Similarly, continuous and dichotomised cPPFE were associated with 1-year FVC decline and mortality (cPPFE ≥ 2·5% derivation: HR 5·26, 3·00-9·22; validation: HR 2·06, 1·28-3·31). Individuals with cPPFE ≥ 2·5% or marked vPPFE had the lowest median survival, the cPPFE threshold demonstrated greater discrimination of poor outcomes at two and three years than marked vPPFE.

INTERPRETATION

PPFE quantification supports distinction of IPF patients with a worse outcome independent of established ILD severity measures. This has the potential to improve prognostic management and elucidate separate pathways of disease progression.

FUNDING

This research was funded in whole or in part by the Wellcome Trust [209,553/Z/17/Z] and the NIHR UCLH Biomedical Research Centre, UK.

The evaluation of the 8th and 7th edition of the American joint committee on cancer tumor classification for distal cholangiocarcinoma: the proposal of a modified new tumor classification

BACKGROUND

The 7th and 8th editions of the American Joint Committee on Cancer (AJCC) tumor (T) classification of distal cholangiocarcinoma (DCC), which are based on either layer or depth, may not accurately stratify patient survival.

METHODS

A total of 121 patients who underwent resection for DCC between 2002 and 2016 were analyzed. The impact of the AJCC staging system on survival was examined and a new T classification was established based on independent prognostic factors.

RESULTS

Regarding overall survival, the optimal depth of invasion (DOI) cut-off value (8 mm) was the only independent prognostic factor. Regarding the relapse-free survival (RFS), a DOI >8 mm, portal vein (PV) invasion, and duodenal or pancreatic invasion were independent prognostic factors. A new T classification was developed as follows: T1, no invasion of adjacent organs; T2, invasion of the duodenum or pancreas; T3, invasion >8 mm into the bile duct wall; and T4, invasion of the PV or arteries. There were no significant differences in RFS according to the 8th edition of the AJCC. However, significant differences were observed in the RFS between T1 and T2 and between T2 and T3.

CONCLUSION

A new T classification based on the layer and depth may be more feasible.

Developing a Clinical Prediction Score: Comparing Prediction Accuracy of Integer Scores to Statistical Regression Models

Researchers often convert prediction tools built on statistical regression models into integer scores and risk classification systems in the name of simplicity. However, this workflow discards useful information and reduces prediction accuracy. We, therefore, investigated the impact on prediction accuracy when researchers simplify a regression model into an integer score using a simulation study and an example clinical data set. Simulated independent training and test sets (n = 1000) were randomly generated such that a logistic regression model would perform at a specified target area under the receiver operating characteristic curve (AUC) of 0.7, 0.8, or 0.9. After fitting a logistic regression with continuous covariates to each data set, continuous variables were dichotomized using data-dependent cut points. A logistic regression was refit, and the coefficients were scaled and rounded to create an integer score. A risk classification system was built by stratifying integer scores into low-, intermediate-, and high-risk tertiles. Discrimination and calibration were assessed by calculating the AUC and index of prediction accuracy (IPA) for each model. The optimism in performance between the training set and test set was calculated for both AUC and IPA. The logistic regression model using the continuous form of covariates outperformed all other models. In the simulation study, converting the logistic regression model to an integer score and subsequent risk classification system incurred an average decrease of 0.057-0.094 in AUC, and an absolute 6.2%-17.5% in IPA. The largest decrease in both AUC and IPA occurred in the dichotomization step. The dichotomization and risk stratification steps also increased the optimism of the resulting models, such that they appeared to be able to predict better than they actually would on new data. In the clinical data set, converting the logistic regression with continuous covariates to an integer score incurred a decrease in externally validated AUC of 0.06 and a decrease in externally validated IPA of 13%. Converting a regression model to an integer score decreases model performance considerably. Therefore, we recommend developing a regression model that incorporates all available information to make the most accurate predictions possible, and using the unaltered regression model when making predictions for individual patients. In all cases, researchers should be mindful that they correctly validate the specific model that is intended for clinical use.

The prognostic utility of pre-treatment neutrophil-to-lymphocyte-ratio (NLR) in colorectal cancer: A systematic review and meta-analysis

Background

Inflammation is a hallmark of cancer, and systemic markers of inflammation are increasingly recognised as negative prognostic factors for clinical outcome. Neutrophil‐to‐lymphocyte ratio (NLR) is readily available from routine blood testing of patients diagnosed with cancer.

Methods

Peer‐reviewed publications from PubMed/MEDLINE, Web of Science and EMBASE were identified according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) guidelines. Hazard ratios (HR) for overall survival (OS) and surrogate endpoints (SE; comprising disease‐, recurrence‐ and progression‐free survival) were pooled using a random effects model. Additional analysis was carried out to further investigate NLR as an independent prognostic factor and account for heterogeneity.

Results

Seventy‐one eligible papers comprising 32,788 patients were identified. High NLR was associated with poor clinical outcomes. Significant publication bias was observed, and larger studies also adjusted for more covariates. Correcting for publication bias in multivariate studies brought our best estimate for true effect size to HR = 1.57 (95% CI 1.39–1.78; p < 0.0001) for OS and to HR = 1.38 (95% CI 1.16–1.64; p = 0.0003) for SE.

Conclusions

NLR is confirmed as an easily available prognostic biomarker in colorectal cancer, despite the limitations of some studies previously reporting this finding. As such, it should be routinely collected in prospective clinical trials. While more standardised and rigorous large‐scale studies are needed before high NLR can be fully assessed as an independent predictor of CRC progression and outcome, the data suggest that it may be used to highlight individuals with tumour‐promoting inflammatory context.

The prognostic significance of serum aspartate transaminase and gamma-glutamyl transferase in liver deceased donors

The impact of aspartate transaminases (AST) and gamma-glutamyl transferase (GGT) in serum of deceased donors on outcomes after liver transplantation (LT) is unclear. This study aimed to explore the relationship between donor highest AST value or first donor GGT value and graft survival. All consecutive patients who underwent a primary LT in a single center with available donor AST (N=1253) and GGT value (N=1152) were included. There was no significant association between donor AST and 90-day graft survival. We found a moderate association between GGT and 90-day graft survival. We found a significant interaction with a donor history of alcohol abuse (HAA). The risk of graft loss was associated with AST and GGT in donors with an HAA but remains unchanged in donors without HAA. There was no difference in graft survival according to donor AST or GGT with a cutoff ≥ 95^th percentile (475 UI/L for AST and 170 UI/L for GGT). However, graft survival was significantly decreased when donors combined GGT ≥ 170 UI/L and HAA (61% at one year). Hepatic grafts from donors with high AST or high GGT but without alcohol history and no additional risk factors can be transplanted in low-risk recipient.

A Longitudinal Model-Based Biomarker Analysis of Exposure Response in Adults with Pulmonary Tuberculosis

The identification of sensitive, specific, and reliable biomarkers that can be quantified in the early phases of tuberculosis treatment and predictive of long-term outcome is key for the development of an effective short-course treatment regimen. Time to positivity (TTP), a biomarker of treatment outcome against Mycobacterium tuberculosis, measures longitudinal bacterial growth in mycobacterial growth indicator tube broth culture and may be predictive of standard time to stable culture conversion (TSCC). In two randomized phase 2b trials investigating dose-ranging rifapentine (Studies 29 and 29X), 662 participants had sputum collected over 6 months where TTP, TSCC, and time to culture conversion were quantified. The goals of this post hoc study were to characterize longitudinal TTP profiles and to identify individual patient characteristics associated with delayed time to culture conversion. In order to do so, a nonlinear mixed-effects model describing longitudinal TTP was built. Independent variables associated with increased bacterial clearance (increased TTP), assessed by subject-specific and population-level trajectories, were higher rifapentine exposure, lower baseline grade of sputum acid-fast bacillus smear, absence of productive cough, and lower extent of lung infiltrates on radiographs. Importantly, sensitivity analysis revealed that major learning milestones in phase 2b trials, such as significant exposure-response and covariate relationships, could be detected using truncated TTP data as early as 6 weeks from start of treatment, suggesting alternative phase 2b study designs. The TTP model built depicts a novel phase 2b surrogate endpoint that can inform early assessment of experimental treatment efficacy and treatment failure or relapse in patients treated with shorter and novel TB treatment regimens, improving efficiency of phase 2 clinical trials. (The studies discussed in this paper have been registered at ClinicalTrials.gov under identifiers NCT00694629 and NCT01043575.)

Challenges and Opportunities in the Statistical Analysis of Multiplex Immunofluorescence Data

Simple Summary

Immune modulation is considered a hallmark of cancer initiation and progression, and has offered promising opportunities for therapeutic manipulation. Multiplex immunofluorescence (mIF) technology has enabled the tumor immune microenvironment (TIME) to be studied at an increased scale, in terms of both the number of markers and the number of samples. Another benefit of mIF technology is the ability to measure not only the abundance but also the spatial location of multiple cells types within a tissue sample simultaneously, allowing for assessment of the co-localization of different types of immune markers. Thus, the use of mIF technologies have enable researchers to characterize patient, clinical, and tumor characteristics in the hope of identifying patients whom might benefit from immunotherapy treatments. In this review we outline some of the challenges and opportunities in the statistical analyses of mIF data to study the TIME.

Abstract

Immune modulation is considered a hallmark of cancer initiation and progression. The recent development of immunotherapies has ushered in a new era of cancer treatment. These therapeutics have led to revolutionary breakthroughs; however, the efficacy of immunotherapy has been modest and is often restricted to a subset of patients. Hence, identification of which cancer patients will benefit from immunotherapy is essential. Multiplex immunofluorescence (mIF) microscopy allows for the assessment and visualization of the tumor immune microenvironment (TIME). The data output following image and machine learning analyses for cell segmenting and phenotyping consists of the following information for each tumor sample: the number of positive cells for each marker and phenotype(s) of interest, number of total cells, percent of positive cells for each marker, and spatial locations for all measured cells. There are many challenges in the analysis of mIF data, including many tissue samples with zero positive cells or “zero-inflated” data, repeated measurements from multiple TMA cores or tissue slides per subject, and spatial analyses to determine the level of clustering and co-localization between the cell types in the TIME. In this review paper, we will discuss the challenges in the statistical analysis of mIF data and opportunities for further research.

A Model-free Approach for Testing Association

The question of association between outcome and feature is generally framed in the context of a model based on functional and distributional forms. Our motivating application is that of identifying serum biomarkers of angiogenesis, energy metabolism, apoptosis, and inflammation, predictive of recurrence after lung resection in node-negative non-small cell lung cancer patients with tumor stage T2a or less. We propose an omnibus approach for testing association that is free of assumptions on functional forms and distributions and can be used as a general method. This proposed maximal permutation test is based on the idea of thresholding, is readily implementable and is computationally efficient. We demonstrate that the proposed omnibus tests maintain their levels and have strong power for detecting linear, nonlinear and quantile-based associations, even with outlier-prone and heavy-tailed error distributions and under nonparametric setting. We additionally illustrate the use of this approach in model-free feature screening and further examine the level and power of these tests for binary outcome. We compare the performance of the proposed omnibus tests with comparator methods in our motivating application to identify preoperative serum biomarkers associated with non-small cell lung cancer recurrence in early stage patients.

Should Therapeutic Drug Monitoring Based on the Vancomycin Area Under the Concentration-Time Curve Be Standard for Serious Methicillin-Resistant Staphylococcus aureus Infections?-No

In this counterpoint we critically appraise the evidence supporting therapeutic drug monitoring based on the vancomycin 24-hour area under the concentration-time curve (AUC24) for serious methicillin-resistant Staphylococcus aureus infections. We reveal methodologically weaknesses and inconsistencies in the data and suggest that, in the absence of clear and convincing evidence of benefit compared with modestly reducing trough targets, alternative strategies are more likely to result in superior safety and efficacy. These include focusing on fundamental antibiotic stewardship to limit vancomycin exposure overall, achieving earlier and more complete source control, and establishing alternative therapeutic options to vancomycin. Implementation of AUC24-based therapeutic drug monitoring will take resources away from these more promising, alternative solutions.

Incidences and Determinants of Functional Cure during Entecavir or Tenofovir Disoproxil Fumarate for Chronic Hepatitis B

BACKGROUND

Long-term incidences and baseline determinants of functional cure (HBsAg seroclearance) during entecavir (ETV) or tenofovir disoproxil fumarate (TDF) treatment are incompletely understood.

METHODS

This is an international multicenter cohort study of treatment-naïve chronic hepatitis B (CHB) patients who initiated on ETV or TDF without baseline malignancy. Patients were observed for HBsAg seroclearance until death or loss to follow-up. We calculated the incidences and explored the baseline determinants of HBsAg seroclearance using competing risk regression.

RESULTS

The analysis included 4,769 patients (median age, 50 years; 69.05% male), with a median follow-up of 5.16 years (26,614.47 person-years). HBsAg clearance occurred in 58 patients, yielding a 10-year cumulative incidence of 2.11% (95% CI, 1.54 -- 2.88%) and an annual rate of 0.22% (95% CI, 0.17--0.28%). Baseline predictors included low-level viremia with HBV DNA <2,000 IU/mL (adjusted sub-distribution HR [aSHR], 3.14; 95% CI, 1.80--5.49), elevated serum alanine aminotransferase (ALT) >200 U/L (aSHR, 3.68; 95% CI, 2.07--6.53), serum bilirubin (aSHR, 1.11 per mg/dL; 95% CI, 1.06--1.17), and fatty liver (aSHR, 1.84; 95% CI, 1.03--3.29).

CONCLUSION

HBsAg seroclearance rarely occurs in CHB patients treated with ETV or TDF and is associated with low-level viremia, ALT flare, bilirubin level, and fatty liver.

Prognostic Significance of Tumor-Associated Macrophages in Chondroblastoma and Their Association with Response to Adjuvant Radiotherapy

Objective

Chondroblastoma (CB) is a rare and locally growing cartilage-derived tumor. Currently, clinical implications of tumor-associated macrophages (TAMs) in CB remain unclear. In this study, we sought to analyze the relationship between TAM parameters (including densities of CD68+ and CD163+ cells as well as the CD163+/CD68+ ratio) and clinicopathological characteristics and survival of patients.

Methods

Immunohistochemistry was used to assess TAM subtypes for CD68 and CD163, as well as the expression levels of p53, CD34, and Ki-67 on tumor cells in 132 tissue specimens retrieved between July 2002 and April 2020. Then, TAM parameters were retrospectively analyzed for their associations with patient outcomes (local recurrence-free survival [LRFS] and overall survival [OS]) and clinicopathological features.

Results

TAM densities were significantly higher in axial chondroblastoma tissue than in extra-axial chondroblastoma tissue. Moreover, the number of CD163+ TAMs was positively correlated with tumor invasion of surrounding tissues and high expression of CD34 and Ki-67 on tumor cells, whereas CD163+ cell density and the CD163/CD68 ratio were negatively associated with patient response to adjuvant radiotherapy. Univariate Kaplan-Meier analysis revealed that the number of CD68+ and CD163+ lymphocytes was significantly associated with both LRFS and OS. Multivariate Cox regression analysis showed that CD163+ and CD68+ cell levels were independent prognostic factors of LRFS, while TAM data independently predicted OS. More importantly, in subgroup analysis based on three significant factors in univariate survival analysis (including tumor location, adjuvant radiotherapy, and surrounding tissue invasion by tumors), the TAM parameters still displayed good prognostic performance.

Conclusion

These data suggest that TAM may significantly affect the biological behavior of CB. We hypothesize that modulating the TAM level or polarization status in the microenvironment may be an effective approach for CB treatment.

Kaplan-Meier Curves, Cox Model, and P-Values Are Not Enough for the Prognostic Evaluation of Tumor Markers: Statistical Suggestions for a More Comprehensive Approach

The assessment of prognostic markers is key to the improvement of therapeutic strategies for cancer patients. Some promising markers may fail to be applied in clinical practice, or some useless markers may be applied, because of misleading results ensuing from inadequate planning of the study and/or from an oversimplified statistical analysis. This commentary illustrates and discusses the main issues involved in planning an effective clinical study and the subsequent statistical analysis for the prognostic evaluation of a cancer marker. Another aim is to extend the most applied statistical models (ie, those using Kaplan-Meier and Cox) to enable the choice of the best-suited methods for study endpoints. Specifically, for tumor-centered endpoints like tumor recurrence, the issue of competing risks is highlighted. For markers measured on a continuous numerical scale, a loss of relevant prognostic information may occur by setting arbitrary cutoffs; thus, the methods to analyze the original scale are explained. Furthermore, because the P-value is not a sufficient criterion to assess the usefulness of a marker in clinical practice, measures for evaluating the ability of the marker to discriminate between "good" and "bad" prognoses are illustrated. Several tumor markers are considered both in human and veterinary medicine. Given the similarity between markers for human breast cancer and canine mammary cancer, an application of the statistical methods discussed within the article to a public dataset from human breast cancer patients is shown.

Statistical Considerations in the Evaluation of Continuous Biomarkers

Discovery of biomarkers has been steadily increasing over the past decade. Although a plethora of biomarkers has been reported in the biomedical literature, few have been sufficiently validated for broader clinical applications. One particular challenge that may have hindered the adoption of biomarkers into practice is the lack of reproducible biomarker cut points. In this article, we attempt to identify some common statistical issues related to biomarker cut point identification and provide guidance on proper evaluation, interpretation, and validation of such cut points. First, we illustrate how discretization of a continuous biomarker using sample percentiles results in significant information loss and should be avoided. Second, we review the popular "minimal-P-value" approach for cut point identification and show that this method results in highly unstable P values and unduly increases the chance of significant findings when the biomarker is not associated with outcome. Third, we critically review a common analysis strategy by which the selected biomarker cut point is used to categorize patients into different risk categories and then the difference in survival curves among these risk groups in the same dataset is claimed as the evidence supporting the biomarker's prognostic strength. We show that this method yields an exaggerated P value and overestimates the prognostic impact of the biomarker. We illustrate that the degree of the optimistic bias increases with the number of variables being considered in a risk model. Finally, we discuss methods to appropriately ascertain the additional prognostic contribution of the new biomarker in disease settings where standard prognostic factors already exist. Throughout the article, we use real examples in oncology to highlight relevant methodologic issues, and when appropriate, we use simulations to illustrate more abstract statistical concepts.

Incidence and Risk Factors of Pannus after Mechanical Aortic Valve Replacement

BACKGROUND

 This study was conducted to evaluate the occurrence rate and risk factors of subaortic pannus (SAP) after bileaflet mechanical aortic valve (AV) replacement.

METHODS

 Between 1990 and 2014, 862 patients underwent primary AV replacement with bileaflet mechanical prosthesis. SAP was defined as (1) gradual increase in mean pressure gradient through mechanical AV without any evidence of motion limitation of the leaflets on echocardiography and (2) AV mean pressure gradient >40 mm Hg or AV peak velocity >4 m/s on echocardiography, and (3) any visible subaortic tissue ingrowth beneath the mechanical AV on echocardiography or computed tomography. Clinical and echocardiographic follow-up durations were 13.8 ± 8.0 and 10.7 ± 7.9 years, respectively.

RESULTS

 Mean age was 51.1 ± 12.1 years and concomitant surgeries were performed in 503 patients (58.4%). Overall survival at 10 and 20 years was 84.2 and 67.1%, respectively. SAP occurred in 33 patients, and in only 2 patients during the first 10 years after surgery. The cumulative incidence of SAP formation at 10, 20, and 25 years were 0.3, 5.0, and 9.9%, respectively. The Fine and Gray model demonstrated that small prosthetic valve size (hazard ratio [HR] [95% confidence interval, CI] = 0.738 [0.575-0.946]), young age (HR [95% CI] = 0.944 [0.909-0.981]), and concomitant mitral valve replacement (MVR) (HR [95% CI] = 3.863 [1.358-10.988]) were significant risk factors for the SAP formation.

CONCLUSIONS

 SAP occurred gradually over time with 10- and 20-year cumulative incidence of 0.3 and 5.0%, respectively. Young age, small prosthetic valve size, and concomitant MVR were risk factors for SAP formation. Therefore, we recommend efforts to select large prostheses for young patients requiring concomitant MVR.

Endometrial thickness in the prediction of neonatal adverse outcomes in frozen cycles for singleton pregnancies

RESEARCH QUESTION

Does endometrial thickness (EMT) predict adverse neonatal outcomes in singleton pregnancies after in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) frozen embryo transfer (FET)?

DESIGN

This retrospective study involved 13,383 women undergoing IVF/ICSI FET cycles between January 2010 and December 2018 in Women's Hospital of Zhejiang University. The primary outcome was preterm delivery (PTD). The secondary outcomes were small for gestational age (SGA), large for gestational age (LGA) and low birthweight (LBW).

RESULTS

A total of 13,383 FET cycles resulting in 5220 singleton live births and 8163 failed cycles were included. Multiple spline regression visualization showed an increasing risk of PTD and LBW for a thin EMT. By comparing multiple cut-off points using area under the curve, a cut-off point of 8 mm was identified, which was used to categorize EMT. A reference point of EMT greater than 8 mm was used; after adjusting for covariates, individuals with EMT less than 8 mm had an adjusted odds ratio of 1.75 (95% CI 1.30 to 2.34) for PTD, 1.57 (95% CI 1.09 to 2.26) for LBW, 0.97 (95% CI 0.63 to 1.50) for SGA and 1.04 (95% CI 0.79 to 1.37) for LGA. Additional analyses showed similar increasing risk with a thin endometrium for both PTD with and without caesarean section, and PTD with low and normal birthweight percentiles.

CONCLUSION

A clinical cut-off point of 8 mm has been identified, below which risk of PTD and LBW increases in women undergoing IVF/ICSI.

Association of Bone Metastatic Burden With Survival Benefit From Prostate Radiotherapy in Patients With Newly Diagnosed Metastatic Prostate Cancer: A Secondary Analysis of a Randomized Clinical Trial

IMPORTANCE

Prostate radiotherapy (RT) improves survival in men with low-burden metastatic prostate cancer. However, owing to the dichotomized nature of metastatic burden criteria, it is not clear how this benefit varies with bone metastasis counts and metastatic site.

OBJECTIVE

To evaluate the association of bone metastasis count and location with survival benefit from prostate RT.

DESIGN, SETTING, AND PARTICIPANTS

This exploratory analysis of treatment outcomes based on metastatic site and extent as determined by conventional imaging (computed tomography/magnetic resonance imaging and bone scan) evaluated patients with newly diagnosed metastatic prostate cancer randomized within the STAMPEDE trial's metastasis M1 RT comparison. The association of baseline bone metastasis counts with overall survival (OS) and failure-free survival (FFS) was assessed using a multivariable fractional polynomial interaction procedure. Further analysis was conducted in subgroups.

INTERVENTIONS

Patients were randomized to receive either standard of care (androgen deprivation therapy with or without docetaxel) or standard of care and prostate RT.

MAIN OUTCOMES AND MEASURES

The primary outcomes were OS and FFS.

RESULTS

A total of 1939 of 2061 men were included (median [interquartile range] age, 68 [63-73] years); 1732 (89%) had bone metastases. Bone metastasis counts were associated with OS and FFS benefit from prostate RT. Survival benefit decreased continuously as the number of bone metastases increased, with benefit most pronounced up to 3 bone metastases. A plot of estimated treatment effect indicated that the upper 95% CI crossed the line of equivalence (hazard ratio [HR], 1) above 3 bone metastases without a detectable change point. Further analysis based on subgroups showed that the magnitude of benefit from the addition of prostate RT was greater in patients with low metastatic burden with only nonregional lymph nodes (M1a) or 3 or fewer bone metastases without visceral metastasis (HR for OS, 0.62; 95% CI, 0.46-0.83; HR for FFS, 0.57; 95% CI, 0.47-0.70) than among patients with 4 or more bone metastases or any visceral/other metastasis (HR for OS, 1.08; 95% CI, 0.91-1.28; interaction P = .003; HR for FFS, 0.87; 95% CI, 0.76-0.99; interaction P = .002).

CONCLUSIONS AND RELEVANCE

In this exploratory analysis of a randomized clinical trial, bone metastasis count and metastasis location based on conventional imaging were associated with OS and FFS benefit from prostate RT in M1 disease.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT00268476; ISRCTN.com Identifier: ISRCTN78818544.

Analysis of the relationship between surgeon procedure volume and complications after total knee arthroplasty using a propensity-matched cohort study

Objectives

This study aimed to identify a threshold in annual surgeon volume associated with increased risk of revision (for any cause) and deep infection requiring surgery following primary elective total knee arthroplasty (TKA).

Design

A propensity score matched cohort study.

Setting

Ontario, Canada.

Participants

169 713 persons who received a primary TKA between 2002 and 2016, with 3-year postoperative follow-up.

Main outcome measures

Revision arthroplasty (for any cause), and the occurrence of deep surgical infection requiring surgery.

Results

Based on restricted cubic spline analysis, the threshold for increased probability of revision and deep infection requiring surgery was <70 cases/year. After matching of 51 658 TKA recipients from surgeons performing <70 cases/year to TKA recipients from surgeons with greater than 70 cases/year, patients in the former group had a higher rate of revision (for any cause, 2.23% (95% Confidence Interval (CI) 1.39 to 3.07) vs 1.70% (95% CI 0.85 to 2.55); Hazard Ratio (HR) 1.33, 95% CI 1.21 to 1.47, p<0.0001) and deep infection requiring surgery (1.29% (95% CI 0.44 to 2.14) vs 1.09% (95% CI 0.24 to 1.94); HR 1.33, 95% CI 1.17 to 1.51, p<0.0001).

Conclusions

For primary TKA recipients, cases performed by surgeons who had performed fewer than 70 TKAs in the year prior to the index TKA were at 31% increased relative risk of revision (for any cause), and 18% increased relative risk for deep surgical infection requiring surgery, at 3-year follow-up.

GATA3 as an Adjunct Prognostic Factor in Breast Cancer Patients with Less Aggressive Disease: A Study with a Review of the Literature

BACKGROUND

GATA binding protein 3 (GATA3) expression is positively correlated with estrogen receptor (ER) expression, but its prognostic value as an independent factor remains unclear. Thus, we undertook the current study to evaluate the expression of GATA3 and its prognostic value in a large series of breast carcinomas (BCs) with long-term follow-up.

METHODS

A total of 702 consecutive primary invasive BCs resected between 1989 and 1993 in our institution were arranged in tissue microarrays, immunostained for ER, progesterone receptor (PR), ki-67, HER2, p53, and GATA3, and scored. Clinico-pathological data were retrospectively collected.

RESULTS

GATA3 was evaluable in 608 (87%) of the 702 cases; it was positive in 413 (68%) cases and negative in 195 (32%) cases. GATA3 positivity was significantly associated with lower grade (p < 0.0001), size (p = 0.0463), stage (p = 0.0049), ER+ (p < 0.0001), PR+ (p < 0.0001), HER2- (p = 0.0175), and p53 wild-type pattern (p < 0.0001). The median follow-up was 183 months, GATA3 positivity was associated with better overall survival (HR 0.70, p = 0.001), and its prognostic value was retained in a multivariate analysis. The association with better overall survival was stronger in patients with grade 1-2, pT1-2, pN0, stage I-II, ER+, PR+, ki-67 < 20%, HER2-, a wild-type p53 immunohistochemical pattern, and in luminal B BC.

CONCLUSIONS

Our findings indicate that GATA3 is a positive prognostic marker in BC patients, especially in patients with biologically less aggressive BC. Incorporating GATA3 immunohistochemistry into routine practice could help further stratify BC patients for their risk.

CXCL13 and CXCL9 CSF Levels in Central Nervous System Lymphoma-Diagnostic, Therapeutic, and Prognostic Relevance

Background: Diagnostic delay and neurologic deterioration are still a problem for the treatment of rapidly progressing CNS lymphoma (CNSL); there is an unmet need for a diagnostic test with a high diagnostic yield and limited risk, minimizing the time to the initiation of effective treatment.

Methods: In this prospective monocentric study, we analyzed the utility of CXCL13 and CXCL9 as diagnostic, therapeutic and prognostic biomarkers for CNSL. Cerebrospinal fluid (CSF) from 155 consecutive patients admitted with brain lesions of various origins was collected. Levels of CXCL13 and CXCL9 were analyzed by ELISA. Additionally, CSF was analyzed during CNSL disease course (relapse, remission, progress) in 17 patients.

Results: CXCL13 and CXCL9 CSF levels were significantly increased in patients with CNSL compared to control patients with lesions of other origin. Using logistic regression and a minimal-p-value approach, a cut-off value of 80 pg/ml for CXCL13 shows high sensitivity (90.7%) and specificity (90.1%) for the diagnosis of active CNSL. CXCL9 at a cut-off value of 84 pg/ml is less sensitive (61.5%) and specific (87.1%). Both cytokines correlate with the clinical course and response to therapy.

Conclusions: Our results confirm the excellent diagnostic potential of CXCL13 and introduce CXCL9 as a novel albeit less powerful marker for PCNSL.

Significance of the Overexpression of Substance P and Its Receptor NK-1R in Head and Neck Carcinogenesis: A Systematic Review and Meta-Analysis

The objective of our study has been, through a systematic review and meta-analysis, to increase the scientific evidence on the implications of SP and its receptor NK-1R in head and neck carcinogenesis. We searched studies published before May-2020 without date and publication language restrictions (PubMed, Embase, Web of Science, Scopus). We evaluated the quality of the studies included (QUIPS tool). We performed heterogeneity, sensitivity, small-study effects, and subgroup analyses. A total 16 studies and 1308 cases met inclusion criteria. Qualitative evaluation demonstrated that not all studies were performed with the same scientific rigor, finding the greatest risk of bias in the study confounding and prognostic factors measurement domains. Quantitative evaluation showed a greater SP/NK-1R overexpression in malignant head and neck lesions compared to benign lesions (p = 0.02), and that expression was observed in malignant salivary gland pathology. Likewise, we found a higher overexpression of NK-1R compared to SP (p = 0.02). In conclusion, the results of this systematic review and meta-analysis show evidence that the upregulation of SP and NK-1R are oncogenic events involved in head and neck carcinogenesis, probably acting in the early stages of malignization. In addition, there is evidence of a greater relevance of the upregulation of the NK-1R receptor compared to SP, which highlights the interest in deepening the development of targeted therapies on the receptor. Future studies assessing the relationships between SP/NK-1R among subjects with head and neck tumors could consider the recommendations given in this systematic review and meta-analysis to improve and standardize future research.

Tertiary Lymphoid Structure-B Cells Narrow Regulatory T Cells Impact in Lung Cancer Patients

The presence of tertiary lymphoid structures (TLS) in the tumor microenvironment is associated with better clinical outcome in many cancers. In non-small cell lung cancer (NSCLC), we have previously showed that a high density of B cells within TLS (TLS-B cells) is positively correlated with tumor antigen-specific antibody responses and increased intratumor CD4⁺ T cell clonality. Here, we investigated the relationship between the presence of TLS-B cells and CD4⁺ T cell profile in NSCLC patients. The expression of immune-related genes and proteins on B cells and CD4⁺ T cells was analyzed according to their relationship to TLS-B density in a prospective cohort of 56 NSCLC patients. We observed that tumor-infiltrating T cells showed marked differences according to TLS-B cell presence, with higher percentages of naïve, central-memory, and activated CD4⁺ T cells and lower percentages of both immune checkpoint (ICP)-expressing CD4⁺ T cells and regulatory T cells (Tregs) in the TLS-B^high tumors. A retrospective study of 538 untreated NSCLC patients showed that high TLS-B cell density was even able to counterbalance the deleterious impact of high Treg density on patient survival, and that TLS-B^high Treg^low patients had the best clinical outcomes. Overall, the correlation between the density of TLS-B^high tumors with early differentiated, activated and non-regulatory CD4⁺ T cell cells suggest that B cells may play a central role in determining protective T cell responses in NSCLC patients.

Re-Defining the Population-Specific Cut-Off Mark for Vitamin A Deficiency in Pre-School Children of Malawi

Retinol Binding Protein (RBP) is responsible for the transport of serum retinol (SR) to target tissue in the body. Since RBP is relatively easy and cheap to measure, it is widely used in national Micronutrient Surveys (MNS) as a proxy for SR to determine vitamin A status. By regressing RBP concentration against SR concentration measured in a subset of the survey population, one can define a population-specific threshold concentration of RBP that indicates vitamin A deficiency (VAD). However, the relationship between RBP and SR concentrations is affected by various factors including inflammation. This study, therefore, aimed to re-define the population-specific cut-off for VAD by examining the influence of inflammation on RBP and SR, among pre-school children (PSC) from the 2015-16 Malawi MNS. The initial association between RBP and SR concentrations was poor, and this remained the case despite applying various methods to correct for inflammation. The World Health Organization (WHO) recommends the threshold of 0.7 µmol/L to define VAD for SR concentrations. Applying this threshold to the RBP concentrations gave a VAD prevalence of 24%, which reduced to 10% after inflammation adjustments following methods developed by the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA). Further research is required to identify why SR and RBP were poorly associated in this population. Future MNS will need to account for the effect of inflammation on RBP to measure the prevalence of VAD in Malawi.

Mediterranean diet assessment challenges: Validation of the Croatian Version of the 14-item Mediterranean Diet Serving Score (MDSS) Questionnaire

Mediterranean diet (MD) is among the most commonly investigated diets and recognized as one of the healthiest dietary patterns. Due to its complexity, geographical and cultural variations, it also represents a challenge for quantification. The aim of this cross-sectional study was to assess reliability and validity of the Croatian version of the 14-item Mediterranean Diet Serving Score (MDSS), using the Mediterranean Diet Adherence Screener (MEDAS) as the referent test. We included the exploratory sample of 360 medical students, and a confirmatory sample of 299 health studies students from the University of Split, Croatia. Test-retest reliability and validity of the MDSS were tested using intra-class correlation coefficients (ICC), while Cohen's kappa statistic was used to test correct classification of subjects into MD adherent/non-adherent category. A very good reliability was shown for the overall MDSS score (ICC = 0.881 [95% CI 0.843-0.909]), and a moderate reliability for the binary adherence (κ = 0.584). Concurrent validity of the MDSS was also better when expressed as a total score (ICC = 0.544 [0.439-0.629]) as opposed to the adherence (κ = 0.223), with similar result in the confirmatory sample (ICC = 0.510 [0.384-0.610]; κ = 0.216). Disappointingly, only 13.6% of medical students were adherent to the MD according to MDSS, and 19.7% according to the MEDAS questionnaire. Nevertheless, MDSS score was positively correlated with age (ρ = 0.179: P = 0.003), self-assessed health perception (ρ = 0.123; P = 0.047), and mental well-being (ρ = 0.139: P = 0.022). MDSS questionnaire is a short, reliable and reasonably valid instrument, and thus useful for assessing the MD adherence, with better results when used as a numeric score, even in the population with low MD adherence.

The ratio of gastric tube length to thorax length: a vital factor affecting leak after esophageal cervical anastomosis

Background

Esophagogastric anastomotic leak (AL) is a severe complication following esophageal resection. This study aims to explore preliminarily whether the ratio of the gastric conduit length to the thorax length can be regarded as a potential prognostic variable for AL, and if so, a cut-off value can be found to divide the patients into distinct risk groups.

Methods

We retrospectively reviewed the clinical data of 273 patients who underwent esophagectomy. The gastric conduit length, the thorax length, and other covariates were collected. Logistic regression was first conducted to probe the rationality of the ratio as a risk indicator of AL. Then the dichotomizing analysis was applied to find the optimal cut-off value.

Results

The incidence of AL was 12.5% (34/273). The coefficient of the ratio in the logistic regression equation was –7.901 with P<0.001, which indicated that the larger the ratio, the smaller the risk of AL. Further smoothed scatter plots revealed that a potential step function of the ratio of AL incidence exists, of which the steep part ranges from 1.74 to 1.90. Results of the accurate cut-off value search through a minimum P value approach give the optimal dichotomization point of 1.79.

Conclusions

The ratio of the gastric conduit length to the thorax length can reflect the tension in the anastomosis. The research proposes that surgeons can control the length of the gastric conduit during reconstruction to reduce the tension in the anastomosis and thus lead to a decrease in the incidence of AL.

AUCs and 123s: a critical appraisal of vancomycin therapeutic drug monitoring in paediatrics

The revised vancomycin guidelines recommend implementing AUC24-based therapeutic drug monitoring (TDM) using Bayesian methods in both adults and paediatrics. The motivation for this change was accumulating evidence showing aggressive dosing to achieve high troughs, as recommended in the first guidelines for adults and extrapolated to paediatrics, is associated with increased nephrotoxicity without improving clinical outcomes. AUC24-based TDM requires substantial resources that may need to be diverted from other valuable interventions. It can therefore be justified only after certain assumptions are shown to be true: (i) there is a clear relationship between vancomycin efficacy and/or toxicity and the proposed therapeutic range; and (ii) maintaining exposure within the target range with AUC24-based TDM improves clinical outcomes and/or decreases toxicity. In this review, we critically appraise the scientific basis for these assumptions. We find studies evaluating the relationship between vancomycin AUC24/MIC and efficacy in adults and children do not offer strong support for the recommended lower limit of the proposed therapeutic range (i.e. AUC24/MIC ≥400). Nephrotoxicity in children increases in a stepwise manner along the vancomycin exposure continuum but it is unclear if one parameter (AUC24 versus trough) is a superior predictor. Overall, evidence in children suggests good-to-excellent correlation between AUC24 and trough. Most importantly, there is no convincing evidence that the method of vancomycin TDM has a causal role in improving efficacy or reducing toxicity. These findings question the need to transition to resource-intensive AUC24-based TDM over retaining trough-based TDM with lower targets to minimize nephrotoxicity in paediatrics.

Prognostic value of preoperative lymphocyte-to-monocyte ratio in oral cancer patients and establishment of a dynamic nomogram

OBJECTIVE

To assess the association of preoperative lymphocyte-to-monocyte ratio (LMR) and overall survival (OS) in patients with oral cancer and develop a dynamic nomogram for individualized survival prediction.

METHOD

The prognostic value of LMR was evaluated in a large-scale cohort with 651 postoperative patients with oral cancer between January 2010 and December 2017. Propensity score-matched (PSM) analysis and inverse probability of treatment weighting (IPTW) analysis were performed to further verify the prognostic value of LMR. A dynamic nomogram was then developed based on the LMR and clinicopathological features, and its predictive performance and clinical utility were evaluated.

RESULTS

A high LMR was significantly associated with better OS of patients with oral cancer (HR = 0.65; 95% CI = 0.44-0.98). The similar association was also observed in the PSM and IPTW analyses. Moreover, compared with TNM staging system, the dynamic nomogram based on the LMR exhibited more excellent predictive performance (0.72 versus 0.64, p < .001), with calibration curves (1,000 bootstrap resamples) suggesting good match between the actual and predicted probabilities. Decision curve analyses (DCAs) showed a more significant positive net benefit in the practical ranges of threshold probabilities using the dynamic nomogram.

CONCLUSION

Preoperative LMR may serve as an easily accessible and non-invasive prognostic biomarker for predicting the prognosis of patients with oral cancer. A dynamic nomogram based on the LMR may show more convenience in survival prediction for patients with oral cancer. Further future studies are warranted to confirm our findings.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

Background: Although many works have supported the utility of PET radiomics, several authors have raised concerns over the robustness and replicability of the results. This study aimed to perform a systematic review on the topic of PET radiomics and the used methodologies. Methods: PubMed was searched up to 15 October 2020. Original research articles based on human data specifying at least one tumor type and PET image were included, excluding those that apply only first-order statistics and those including fewer than 20 patients. Each publication, cancer type, objective and several methodological parameters (number of patients and features, validation approach, among other things) were extracted. Results: A total of 290 studies were included. Lung (28%) and head and neck (24%) were the most studied cancers. The most common objective was prognosis/treatment response (46%), followed by diagnosis/staging (21%), tumor characterization (18%) and technical evaluations (15%). The average number of patients included was 114 (median = 71; range 20–1419), and the average number of high-order features calculated per study was 31 (median = 26, range 1–286). Conclusions: PET radiomics is a promising field, but the number of patients in most publications is insufficient, and very few papers perform in-depth validations. The role of standardization initiatives will be crucial in the upcoming years.

Serum levels of soluble programmed death-ligand 1 (sPD-L1): A possible biomarker in predicting post-treatment outcomes in patients with early hepatocellular carcinoma

BACKGROUND

There have been great advances in hepatocellular carcinoma management over the last years. However, there are still no prognostic biomarkers that can identify patients who will benefit the most from curative treatments. We aimed to investigate whether sPD-L1 levels measured before curative treatment is a prognostic biomarker of survival in patients with HCC.

METHODS

HCC patients from a prospectively collected database were selected and soluble programmed death-ligand1(sPD-L1) levels were determined. The association of sPD-L1 levels and overall survival (OS) and disease-free survival (DFS) was assessed.

RESULTS

One hundred twenty-one patients with HCC were included. The best cut-off value of sPD-L1 for both DFS and OS was 96 pg/mL. Patients with a high sPD-L1 value (>96 pg/mL) had a shorter disease free survival and OS (hazard ratio 5.42, 95% confidence interval 2.28-12.91, p < 0.001, and hazard ratio 9.67, 95% confidence interval 4.33-21.59, p < 0.001). High sPD-L1 levels were associated with mortality independently from other known survival predictors. We found a positive correlation between sPD-L1 and PD-L1 expression in cancer cells (p = 0.01). In 16 out of 38 patients, sPD-L1 levels decreased from baseline value on week 6 after treatment and in 22 out of 38 patients, sPD-L1 levels increased from the baseline value. However, fluctuations of sPD-L1 in time had no influence on survival (p = 0.148).

CONCLUSION

We conclude that a high sPD-L1 level is a biomarkerfor a poor outcome in HCC. The predictive value of sPD-L1 levels for a successful anti-PD1/PD-L1 therapy should be investigated in the future.