Long-term survival with non-proportional hazards: results from the Dutch Gastric Cancer Trial

Putter, Hein; Sasako, Mitsuru; Hartgrink, H.H.; Velde, Cornelis J.H. van de; Houwelingen, Hans C. van

doi:10.1002/sim.2143

Cited by 67 publications

(61 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clinical T-category, Gleason score, the use of hormonal treatment, and radiotherapy dose were modeled as categorical variables and age, pretreatment PSA, the proportion of positive diagnostic biopsy cores, and HP 10 as continuous variables. The assumption of proportional hazards was tested for each covariate first by univariate Cox analysis and examination of the Schoenfeld residuals and then using a bivariate Cox model, including both the covariate and its interaction with time (19,20).…”

Section: Sample Size and Data Analysismentioning

confidence: 99%

Tumor Hypoxia Predicts Biochemical Failure following Radiotherapy for Clinically Localized Prostate Cancer

Milosevic

Warde

Ménard

et al. 2012

Clinical Cancer Research

233

202

View full text Add to dashboard Cite

Purpose: Tumor hypoxia is an important determinant of outcome in many human malignancies and is associated with treatment resistance and metastases. The aim of this study was to determine the effect of hypoxia in patients with prostate cancer treated with radiotherapy.Experimental Design: Tumor hypoxia was measured in 247 patients with clinically localized prostate cancer before radiotherapy, with or without hormonal therapy. The median pO 2 was 6.8 mm Hg and the median hypoxic percentage less than 10 mm Hg (HP 10 ) was 63%. The median follow-up was 6.6 years.Results:The 5-year biochemical relapse-free rate (bRFR) was 78%. Prostrate-specific antigen and Gleason score were both associated with biochemical relapse and formed a baseline clinical model. The effect of hypoxia was found to vary with the duration of patient follow-up. HP 10 , when added to the clinical model, was an independent predictor of early bRFR (P ¼ 0.019). The relationship between hypoxia and early bRFR was more pronounced when the analysis was restricted to 142 patients with bulk tumor at the site of the oxygen measurements (P ¼ 0.004). Hypoxia was the only factor predictive of local recurrence in 70 patients who had biopsies conducted during follow-up (P ¼ 0.043), again with the effect being greatest early after completing treatment.Conclusions: This is the largest clinical study of prostate cancer hypoxia with direct measurement of tumor oxygen levels. It shows that hypoxia is associated with early biochemical relapse after radiotherapy and also with local recurrence in the prostate gland.

show abstract

Section: Sample Size and Data Analysismentioning

confidence: 99%

Tumor Hypoxia Predicts Biochemical Failure following Radiotherapy for Clinically Localized Prostate Cancer

Milosevic

Warde

Ménard

et al. 2012

Clinical Cancer Research

233

202

View full text Add to dashboard Cite

show abstract

“…These steps are important parts of the process of deriving a parsimonious model which also fits the data adequately. Reanalysis data from the Dutch Gastric Cancer Trial, Putter et al (2005) present another approach with much emphasis on estimating the baseline cumulative hazard function. However, they do not try to model the effect of continuous variables, e.g.…”

Section: Introductionmentioning

confidence: 99%

A New Proposal for Multivariable Modelling of Time‐Varying Effects in Survival Data Based on Fractional Polynomial Time‐Transformation

2007

View full text Add to dashboard Cite

The Cox proportional hazards model has become the standard for the analysis of survival time data in cancer and other chronic diseases. In most studies, proportional hazards (PH) are assumed for covariate effects. With long-term follow-up, the PH assumption may be violated, leading to poor model fit. To accommodate non-PH effects, we introduce a new procedure, MFPT, an extension of the multivariable fractional polynomial (MFP) approach, to do the following: (1) select influential variables; (2) determine a sensible dose-response function for continuous variables; (3) investigate time-varying effects; (4) model such time-varying effects on a continuous scale. Assuming PH initially, we start with a detailed model-building step, including a search for possible non-linear functions for continuous covariates. Sometimes a variable with a strong short-term effect may appear weak or non-influential if 'averaged' over time under the PH assumption. To protect against omitting such variables, we repeat the analysis over a restricted time-interval. Any additional prognostic variables identified by this second analysis are added to create our final time-fixed multivariable model. Using a forward-selection algorithm we search for possible improvements in fit by adding time-varying covariates. The first part to create a final time-fixed model does not require the use of MFP. A model may be given from 'outside' or a different strategy may be preferred for this part. This broadens the scope of the time-varying part. To motivate and illustrate the methodology, we create prognostic models from a large database of patients with primary breast cancer. Non-linear time-fixed effects are found for progesterone receptor status and number of positive lymph nodes. Highly statistically significant time-varying effects are present for progesterone receptor status and tumour size.

show abstract

“…For the Kaplan-Meier analysis in dependence of riluzole use, the Gehan-Breslow-Wilcoxon test was performed additionally, which emphasises the information at the beginning of the survival curves. A multivariate Cox regression model for non-proportional hazards was used to study the effect of riluzole on ALS survival [11]. Riluzole exposure was expressed as a categorical variable (riluzole use = 1, no riluzole = 0).…”

Section: Methodsmentioning

confidence: 99%

Epidemiology of Amyotrophic Lateral Sclerosis and Effect of Riluzole on Disease Course

et al. 2015

View full text Add to dashboard Cite

Objectives: To assess the epidemiology of ALS in Austria and to evaluate the long-term effect of riluzole treatment on survival. Methods: Hospital discharge and riluzole prescription databases were used to identify ALS cases from January 2008 to June 2012. Using the capture-recapture method we evaluated the incidence and prevalence of ALS and patients' survival in dependence of age, gender and riluzole treatment. Results: The corrected incidence and prevalence of ALS were 3.13/100,000 person-years (95% CI, 2.77 to 3.50) and 9.14/100,000 persons (95% CI, 8.53 to 9.79), respectively. Median survival from diagnosis was 676 days (95% CI, 591 to 761). A younger age at diagnosis was associated with a longer survival. Gender did not appear to affect survival time. Riluzole therapy was associated with a survival advantage only for the initial treatment period. The adjusted hazard ratio of mortality for using riluzole increased continually over time resulting in an apparent reversal of its beneficial effect after 6 months of therapy. Conclusions: We report incidence and prevalence estimates that are on the upper end of the wide range discussed in literature. Riluzole seems to exert a beneficial effect only in the first 6 months of therapy.

show abstract

Long-term survival with non-proportional hazards: results from the Dutch Gastric Cancer Trial

Cited by 67 publications

References 12 publications

Tumor Hypoxia Predicts Biochemical Failure following Radiotherapy for Clinically Localized Prostate Cancer

Tumor Hypoxia Predicts Biochemical Failure following Radiotherapy for Clinically Localized Prostate Cancer

A New Proposal for Multivariable Modelling of Time‐Varying Effects in Survival Data Based on Fractional Polynomial Time‐Transformation

Epidemiology of Amyotrophic Lateral Sclerosis and Effect of Riluzole on Disease Course

Contact Info

Product

Resources

About