Direct modeling of the crude probability of cancer death and the number of life years lost due to cancer without the need of cause of death: a pseudo-observation approach in the relative survival setting.

In population-based cancer studies, net survival is a crucial measure for population comparison purposes. However, alternative measures, namely the crude probability of death (CPr) and the number of life years lost (LYL) due to death according to different causes, are useful as complementary measures for reflecting different dimensions in terms of prognosis, treatment choice, or development of a control strategy. When the cause of death (COD) information is available, both measures can be estimated in competing risks setting using either cause-specific or subdistribution hazard regression models or with the pseudo-observation approach through direct modeling. We extended the pseudo-observation approach in order to model the CPr and the LYL due to different causes when information on COD is unavailable or unreliable (i.e., in relative survival setting). In a simulation study, we assessed the performance of the proposed approach in estimating regression parameters and examined models with different link functions that can provide an easier interpretation of the parameters. We showed that the pseudo-observation approach performs well for both measures and we illustrated their use on cervical cancer data from the England population-based cancer registry. A tutorial showing how to implement the method in R software is also provided.

On comparison of net survival curves.

BACKGROUND: Relative survival analysis is a subfield of survival analysis where competing risks data are observed, but the causes of death are unknown. A first step in the analysis of such data is usually the estimation of a net survival curve, possibly followed by regression modelling. Recently, a log-rank type test for comparison of net survival curves has been introduced and the goal of this paper is to explore its properties and put this methodological advance into the context of the field. METHODS: We build on the association between the log-rank test and the univariate or stratified Cox model and show the analogy in the relative survival setting. We study the properties of the methods using both the theoretical arguments as well as simulations. We provide an R function to enable practical usage of the log-rank type test. RESULTS: Both the log-rank type test and its model alternatives perform satisfactory under the null, even if the correlation between their p-values is rather low, implying that both approaches cannot be used simultaneously. The stratified version has a higher power in case of non-homogeneous hazards, but also carries a different interpretation. CONCLUSIONS: The log-rank type test and its stratified version can be interpreted in the same way as the results of an analogous semi-parametric additive regression model despite the fact that no direct theoretical link can be established between the test statistics.

Selection of Non-small Cell Lung Cancer Patients for Intercalated Chemotherapy and Tyrosine Kinase Inhibitors.

BACKGROUND: When treating patients with advanced non-small cell lung cancer (NSCLC) with tyrosine kinase inhibitors and chemotherapy, intercalated schedule with time separation between the two classes of drugs should avoid their mutual antagonism. In a survey of published trials, we focus on relation between eligibility criteria and effectiveness of intercalated treatment. METHODS: Published documents were identified using major medical databases, conference proceedings and references of published trials. Median progression-free survival (PFS) was taken as the basic parameter of treatment efficacy. Correlation between characteristics of patients and median PFS was assessed through the Pearson's correlation coefficient and the coefficient of determination, separately for first-line and second-line setting. RESULTS: The series includes 11 single-arm trials and 18 randomized phase II or phase III trials with a total of 2903 patients. Treatment-naive patients or those in progression after first-line treatment were included in 16 and 13 trials, respectively. In 14 trials, only patients with non-squamous histology were eligible. Proportion of patients with non-squamous carcinoma (in first-line setting), proportion of never-smokers (both in first- and second-line setting) and proportion of epidermal growth factor receptor (EGFR) mutant patients (both in first- and second-line setting) showed a moderate or strong correlation with median PFS. In six trials of intercalated treatment applied to treatment-naive EGFR-mutant patients, objective response was confirmed in 83.1% of cases and median PFS was 18.6 months. CONCLUSIONS: Most suitable candidates for intercalated treatment are treatment-naive patients with EGFR-mutant tumors, as determined from biopsy or liquid biopsy. For these patients, experience with intercalated treatment is most promising and randomized trials with comparison to the best standard treatment are warranted.

On estimation in relative survival.

Estimation of relative survival has become the first and the most basic step when reporting cancer survival statistics. Standard estimators are in routine use by all cancer registries. However, it has been recently noted that these estimators do not provide information on cancer mortality that is independent of the national general population mortality. Thus they are not suitable for comparison between countries. Furthermore, the commonly used interpretation of the relative survival curve is vague and misleading. The present article attempts to remedy these basic problems. The population quantities of the traditional estimators are carefully described and their interpretation discussed. We then propose a new estimator of net survival probability that enables the desired comparability between countries. The new estimator requires no modeling and is accompanied with a straightforward variance estimate. The methods are described on real as well as simulated data.

A measure of explained variation for event history data.

There is no shortage of proposed measures of prognostic value of survival models in the statistical literature. They come under different names, including explained variation, correlation, explained randomness, and information gain, but their goal is common: to define something analogous to the coefficient of determination R(2) in linear regression. None however have been uniformly accepted, none have been extended to general event history data, including recurrent events, and many cannot incorporate time-varying effects or covariates. We present here a measure specifically tailored for use with general dynamic event history regression models. The measure is applicable and interpretable in discrete or continuous time; with tied data or otherwise; with time-varying, time-fixed, or dynamic covariates; with time-varying or time-constant effects; with single or multiple event times; with parametric or semiparametric models; and under general independent censoring/observation. For single-event survival data with neither censoring nor time dependency it reduces to the concordance index. We give expressions for its population value and the variance of the estimator and explore its use in simulations and applications. A web link to R software is provided.

Has equity in relative survival improved over time in Finland - a methodological exercise.

BACKGROUND: Population-based relative survival is widely used as a method of monitoring the success of cancer control. This success may not be relevant only for an entire country but also regional developments over time are of interest. It would not only be important that the relative survival improved but also that the differences between regions decreased over time. METHODS: In this paper the authors show how relative survival methods can be used to study such differences. In addition to standard methods, some more recently introduced approaches are used, most notably a method for checking the goodness of fit of the relative survival model. This gives confidence in the obtained results and provides additional insight when assumptions are not met. RESULTS: An analysis of cancers of the colon and ovary by cancer control region in Finland in 1953-2003 shows an overall improvement in relative survival, accompanied in colon cancer also by a decrease of differences in relative survival between the regions. Thus, the desired course was observed in colon cancer but not in cancer of the ovary. CONCLUSIONS: These results, applied to further sites, should lead to investigation of differences in cancer control policies between regions.

An approach to estimation in relative survival regression.

The goal of relative survival methodology is to compare the survival experience of a cohort with that of the background population. Most often an additive excess hazard model is employed, which assumes that each person's hazard is a sum of 2 components--the population hazard obtained from life tables and an excess hazard attributable to the specific condition. Usually covariate effects on the excess hazard are assumed to have a proportional hazards structure with parametrically modelled baseline. In this paper, we introduce a new fitting procedure using the expectation-maximization algorithm, treating the cause of death as missing data. The method requires no assumptions about the baseline excess hazard thus reducing the risk of bias through misspecification. It accommodates the possibility of knowledge of cause of death for some patients, and as a side effect, the method yields an estimate of the ratio between the excess and the population hazard for each subject. More importantly, it estimates the baseline excess hazard flexibly with no additional degrees of freedom spent. Finally, it is a generalization of the Cox model, meaning that all the wealth of options in existing software for the Cox model can be used in relative survival. The method is applied to a data set on survival after myocardial infarction, where it shows how a particular form of the hazard function could be missed using the existing methods.

Ljubljana nomograms for predicting the likelihood of non-sentinel lymph node metastases in breast cancer patients with a positive sentinel lymph node.

Several tools for predicting the likelihood of non-sentinel lymph node (non-SLN) involvement in SLN positive breast cancer patients have been created so far.The aim of our study was to create and validate different nomograms for predicting the likelihood of non-SLN involvement that would be applicable in different institutions and that would also include the results of the preoperative US examination of the axilla. From January 2000 to January 2009, 534 breast cancer patients underwent axillary lymph node dissection (ALND) due to metastatic SLN at our institution. Using logistic regression results three nomograms differing in the inclusion of the results of intraoperative examination of SLN were created. The nomograms were validated using bootstrap methods. In all three nomograms, US examination of the axilla was a powerful independent variable. Other variables included(different in different nomograms) were tumor size, lymphovascular invasion, metastasis size in SLN, number of negative and number of positive SLNs. Mean absolute error and mean area under the ROC curve equals to 0.016 and 0.77 for the first, 0.023 and 0.75 for the second and 0.014 and 0.79 for the third nomogram. Three nomograms for predicting the likelihood of non-SLN metastases including the results of the preoperative US examination of the axilla were created at our institution. They differ in the inclusion of the results of intraoperative examination of SLNs and are thus applicable in different institutions. The validation results seem promising and omission of completion ALND might be considered in patients with the probability of having non-SLN metastases of 10% or less.

Trends in survival after childhood cancer in Slovenia between 1957 and 2007.

OBJECTIVES: To evaluate childhood cancer survival in Slovenia, to provide results comparable to ACCIS studies, and to study the effect of recorded variables on 5-year survival. METHODS: The data are registry-based and present a unique collection in terms of control and homogeneity. Survival was explored using Kaplan-Meier estimates and the Cox model. Restricted cubic splines were used to illustrate the nonlinearity of the age and year of diagnosis effect for the four chosen diagnoses. RESULTS: The data set includes 1827 children examined from 1957 to 2002 with the follow-up ending 2007. The overall 5-year survival increased from 0.26 (95%CI [0.21, 0.33]) before 1973 to 0.8 (95% CI [0.74, 0.85]) for patients diagnosed in the period 1998-2002. It is best for Hodgkin disease and leukemia; for non-Hodgkin lymphoma (NHL) and intracranial and intraspinal neoplasms (CNS) the rate of improvement has been slowing down since 1990. Survival is significantly associated with age at diagnosis for patients with leukemia, CNS, NHL, and neuroblastoma (p < .001), the association varies between diseases. Hazard decreases with age for children with CNS and NHL, increases for children with neuroblastoma, and is quadratic with its lowest point at the age of about 5 years for children with leukemia. CONCLUSIONS: The survival experience in Slovenia compares well with those of large samples in the United States SEER program 1975-1995 and the data collected by ACCIS from 62 population-based cancer registries in Europe. The hazard of dying has been decreasing constantly, mainly due to improvements in leukemia treatment.

Confidence intervals for the Mann-Whitney test.

The Mann-Whitney test is a commonly used non-parametric alternative of the two-sample t-test. Despite its frequent use, it is only rarely accompanied with confidence intervals of an effect size. If reported, the effect size is usually measured with the difference of medians or the shift of the two distribution locations. Neither of these two measures directly coincides with the test statistic of the Mann-Whitney test, so the interpretation of the test results and the confidence intervals may be importantly different. In this paper, we focus on the probability that random variable X is lower than random variable Y. This measure is often referred to as the degree of overlap or the probabilistic index; it is in one-to-one relationship with the Mann-Whitney test statistic. The measure equals the area under the ROC curve. Several methods have been proposed for the construction of the confidence interval for this measure, and we review the most promising ones and explain their ideas. We study the properties of different variance estimators and small sample problems of confidence intervals construction. We identify scenarios in which the existing approaches yield inadequate coverage probabilities. We conclude that the DeLong variance estimator is a reliable option regardless of the scenario, but confidence intervals should be constructed using the logit scale to avoid values above 1 or below 0 and the poor coverage probability that follows. A correction is needed for the case when all values from one sample are smaller than the values of the other. We propose a method that improves the coverage probability also in these cases.

Initial healing rates as predictive factors of venous ulcer healing: the use of a laser-based three-dimensional ulcer measurement.

There is a need for practical methods to predict the healing time of venous leg ulcers. In a prospective cohort study of 81 patients with venous leg ulcers, we used a recently described laser-based three-dimensional measurement of the ulcers at days 0 and 28 to estimate the predictive power of horizontal (HIHR) and vertical initial healing rates (VIHR) for wound healing by week 24. The rates were calculated by Gilman's equation [(A(1)-A(2))/((p(1)+p(2))/2)((0-4))] and by its modification [(V(1)-V(2))/((A(1)+A(2))/2)((0-4))], respectively. The influence of risk factors on both the initial healing rates was also studied. The HIHR and VIHR are important predictors of healing at 24 weeks. They are not influenced by age, ulcer duration, initial ulcer area, and insufficient sapheno-femoral junction, and/or calf perforating veins. Together with ulcer duration, they are independent predictors of the 24-week healing (the area under ROC curve equals to 0.9). VIHR gives us additional information and significantly improves the prediction of 24-week healing.

SAS and R functions to compute pseudo-values for censored data regression.

Recently, in a series of papers, a method based on pseudo-values has been proposed for direct regression modeling of the survival function, the restricted mean and cumulative incidence function with right censored data. The models, once the pseudo-values have been computed, can be fit using standard generalized estimating equation software. Here we present SAS macros and R functions to compute these pseudo-values. We illustrate the use of these routines and show how to obtain regression estimates for a study of bone marrow transplant patients.

Comparison of the algorithms classifying the ABC and GCB subtypes in diffuse large B-cell lymphoma.

Different immunohistochemical algorithms for the classification of the activated B-cell (ABC) and germinal center B-cell (GCB) subtypes of diffuse large B-cell lymphoma (DLBCL) are applied in different laboratories. In the present study, 127 patients with DLCBL were investigated, all treated with rituximab and cyclophosphamide, hydroxydaunorubicin, oncovin and prednisone (CHOP) or CHOP-like regimens between April 2004 and December 2010. Multi-tumor tissue microarrays were prepared and were tested according to 4 algorithms: Hans; modified Hans; Choi; and modified Choi. For 39 patients, the flow cytometric quantification of CD19 and CD20 antigen expression was performed and the level of expression presented as molecules of equivalent soluble fluorochrome units. The Choi algorithm was demonstrated to be prognostic for OS and classified patients into the GCB subgroup with an HR of 0.91. No difference in the expression of the CD19 antigen between the ABC and GCB groups was observed, but the ABC subtype exhibited a decreased expression of the CD20 antigen compared with the GCB subtype.

The effect of CD34+ cell telomere length and hTERT expression on the outcome of autologous CD34+ cell transplantation in patients with chronic heart failure.

Age-related telomere attrition in stem/progenitor cells may diminish their functional capacity and thereby impair the outcome of cell-based therapies. The aim of the present study was to investigate the effect of CD34+ cell telomere length and hTERT expression on the clinical outcome of autologous CD34+ cell transplantation. We studied 43 patients with cardiomyopathy. Their peripheral blood CD34+ cells were mobilized with granulocyte colony-stimulating factor, enriched by immunoselection and delivered transendocardially. Relative telomere length and expression levels of hTERT were measured using a real-time PCR assay. Immunoselected CD34+ cells had longer telomere length compared to leukocytes in leukapheresis products (p=0.001). In multivariate analysis, CD34+ cell telomere length was not associated with the clinical outcome (b=3.306, p=0.540). While hTERT expression was undetectable in all leukapheresis products, 94.4% of the CD34+ enriched cell products expressed hTERT. Higher CD34+hTERT expression was associated with a better clinical outcome on univariate analysis (b=87.911, p=0.047). Our findings demonstrate that CD34+ cell telomere length may not influence the clinical outcome in cardiomyopathy patients treated with autologous CD34+ cell transplantation. Larger studies are needed to validate the impact of the CD34+hTERT expression on the clinical outcome of autologous CD34+ cell transplantation.

Effect of repeated passive anterior loading on human knee anterior laxity.

BACKGROUND: Increased knee anterior laxity results when the anterior cruciate ligament is injured. This increased laxity can cause knee dysfunction. Until recently this laxity was believed to be only diminished through surgery. But recent findings indicate that knee anterior laxity may be decreased with repeated loading of the knee. OBJECTIVE: The purpose of this study was to test the hypothesis that regular passive anterior loading of the uninjured human knee would enhance its stiffness. STUDY DESIGN: Randomized controlled trial. METHODS: Knee anterior laxity was tested using an arthrometer in 22 young, uninjured females before, during and after a 3 month period during which passive anterior loading was applied by a trained physiotherapist over 5 sessions per week to a randomly assigned knee. RESULTS: Knee anterior laxity was not affected by the passive anterior loading of the knee. CONCLUSIONS: Given that in this study repeated passive loading of the knee did not change knee anterior laxity, it would be easy to conclude that this training is ineffective and no further research is required. We caution against this given the relatively short duration and possibly insufficient intensity of the training and the population studied; individuals with normal joint laxity. We recommend that future research be performed that consists of individuals with lax joints who receive training for prolonged periods.

Inter-rater reliability of the GNRB® knee arthrometer.

BACKGROUND: In clinical settings, where arthrometers are mainly used by different clinicians, knowing the inter-rater reliability of the instrument is crucial in order for the results from different examiners to be accurately interpreted and limitations fully understood. The aim of this study was to evaluate the inter-rater reliability of the GNRB® knee arthrometer. METHODS: Knee anterior laxity in both knees was tested in a group of young, uninjured subjects (N=27, 13 females) by two examiners. Knee anterior laxity was calculated at test forces of 134 N and 250 N with values presented for the unstandardised and standardised conditions (relative to patellar stabilisation force). RESULTS: The ICCs ranged from 0.220 to 0.424. CONCLUSIONS: The inter-rater reliability of the GNRB® knee arthrometer is low.

Outcomes of peritoneal dialysis patients and switching to hemodialysis: a competing risks analysis.

BACKGROUND: We performed a review of a large incident peritoneal dialysis cohort to establish the impact of current practice and that of switching to hemodialysis. METHODS: Patients starting peritoneal dialysis between 2004 and 2010 were included and clinical data at start of dialysis recorded. Competing risk analysis and Cox proportional hazards model with time-varying covariate (technique failure) were used. RESULTS: Of 286 patients (median age 57 years) followed for a median of 24.2 months, 76 were transplanted and 102 died. Outcome probabilities at 3 and 5 years respectively were 0.69 and 0.53 for patient survival (or transplantation) and 0.33 and 0.42 for technique failure. Peritonitis caused technique failure in 42%, but ultrafiltration failure accounted only for 6.3%. Davies comorbidity grade, creatinine and obesity (but not residual renal function or age) predicted technique failure. Due to peritonitis deaths, technique failure was an independent predictor of death hazard. When successful switch to hemodialysis (surviving more than 60 days after technique failure) and its timing were analyzed, no adverse impact on survival in adjusted analysis was found. However, hemodialysis via central venous line was associated with an elevated death hazard as compared to staying on peritoneal dialysis, or hemodialysis through a fistula (adjusted analysis hazard ratio 1.97 (1.02 - 3.80)). CONCLUSIONS: Once the patients survive the first 60 days after technique failure, the switch to hemodialysis does not adversely affect patient outcomes. The nature of vascular access has a significant impact on outcome after peritoneal dialysis failure.

Direct microscopic monitoring of initial and dynamic clot lysis using plasmin or rt-PA in an in vitro flow system.

INTRODUCTION: Plasmin is a direct-acting thrombolytic agent with a favorable safety profile upon intra-arterial delivery in pre-clinical and phase I studies. However, the thrombolytic efficacy of plasmin, relative to that of rt-PA, remains to be established. We have compared the dynamics of clot lysis with plasmin or rt-PA in an in vitro perfusion system, in which thrombolytic agent is administered locally, allowed to induce lysis for short intervals, then washed with plasma in a re-circulation circuit. MATERIALS AND METHODS: Whole blood human clots were prepared in observation chambers, exposed to plasmin or rt-PA at equimolar concentrations (1.2/1.0, 1.8/1.5 and 2.4/2.0 mg/ml) for measured intervals of time, followed by perfusion with human plasma. Clot size was monitored by digital analysis of sequential photographs obtained through an optical microscope. RESULTS: Plasma perfusion after incubation with thrombolytic agent rapidly removed superficial clot fragments. This initial decrease in clot size was greater with plasmin than with rt-PA when tested at the highest concentrations of agent (0.63 ± 0.11 vs. 0.30 ± 0.11, p=0.001 for clots with non-cross-linked fibrin and 0.53 ± 0.15 vs. 0.14 ± 0.15, p=0.02, for clots with cross-linked-fibrin). Subsequent clot lysis during plasma flow was greater after prior incubation with rt-PA. Longer incubation times of plasmin resulted in larger portions of the clot being washed free. Repeated plasmin incubations and plasma perfusions of a clot successfully induced stepwise reductions in clot size. CONCLUSIONS: Initial clot lysis is greater with direct exposure using plasmin than rt-PA. During washout and circulation with plasma, rt-PA induced continued clot lysis, while plasmin lysis was curtailed, presumably because of plasmin inhibition.

Robotic knee laxity testing: reliability and normative data.

PURPOSE: To evaluate the reliability of the GeNouRoB knee arthrometer and present normative values of knee anterior laxity using this device on young females. METHODS: Anterior laxity in both knees was tested in two groups of young, uninjured females using the hamstrings electromyography biofeedback feature of the device. There were 13 participants in the group tested for reliability and 23 for the normative study. Laxity (mm of movement of the proximal tibia in the anterior direction relative to the femur) was calculated at test forces of 134 N and 250 N with values presented for the unstandardised and standardised (relative to stabilisation force) conditions. RESULTS: The relative reliability (95% limits of agreement) of the device for laxity at a test force of 134 N was 2 to 3mm. Left knee anterior laxity was almost 1mm greater than the right. CONCLUSIONS: The relative reliability of the GeNouRoB arthrometer is comparable to the KT device. In agreement with previous work on the nonrobotic KT arthrometer, the knee anterior laxity values found with the GeNouRoB are greater in the left as compared to the right knee.