An efficient alternative to the stratified Cox model analysis

2012 ◽  
Vol 31 (17) ◽  
pp. 1849-1856 ◽  
Author(s):  
Devan V. Mehrotra ◽  
Shu-Chih Su ◽  
Xiaoming Li
Keyword(s):  
Cox Model ◽  
Model Analysis ◽  
Efficient Alternative ◽  
Stratified Cox Model

scholarly journals Anthracyclines in Early Breast Cancer: The ABC Trials—USOR 06-090, NSABP B-46-I/USOR 07132, and NSABP B-49 (NRG Oncology)

2017 ◽  
Vol 35 (23) ◽  
pp. 2647-2655 ◽  
Author(s):  
Joanne L. Blum ◽  
Patrick J. Flynn ◽  
Greg Yothers ◽  
Lina Asmar ◽  
Charles E. Geyer ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Early Breast Cancer ◽  
Cox Model ◽  
Growth Factor Receptor ◽  
Disease Free Survival ◽  
Invasive Disease ◽  
Efficacy Analysis ◽  
Oncology Research ◽  
Stratified Cox Model ◽  
To Receive

Purpose Docetaxel and cyclophosphamide (TC) was superior to doxorubicin and cyclophosphamide (AC) in a trial in early breast cancer. However, activity of TC relative to AC regimens with a taxane (TaxAC) is unknown. Methods In a series of three adjuvant trials, women were randomly assigned to TC for six cycles (TC6) or to a standard TaxAC regimen. US Oncology Research (USOR) 06-090 compared TC6 with docetaxel, doxorubicin, and cyclophosphamide (TAC6). National Surgical Adjuvant Breast and Bowel Project (NSABP) B-46-I/USOR 07132 compared TC6, TAC6, or TC6 plus bevacizumab. NSABP B-49 compared TC6 with several standard AC and taxane combination regimens. Before any analysis of individual trials, a joint efficacy analysis of TC versus the TaxAC regimens was planned, with invasive disease-free survival (IDFS) as the primary end point. Patients who received TC6 plus bevacizumab on NSABP B-46-I/USOR 07132 were not included. A hazard ratio (HR) from a stratified Cox model that exceeded 1.18 for TC6 versus TaxAC was predefined as inferiority for TC6. The prespecified interim monitoring plan was to report for futility if the HR was > 1.18 when 334 IDFS events were observed (50% of 668 events required for definitive analysis). Results A total of 2,125 patients were randomly assigned to receive TC6 regimens and 2,117 patients were randomly assigned to receive TaxAC regimens. The median follow-up time was 3.3 years. There were 334 IDFS events, and the HR for TC6 versus TaxAC was 1.202 (95% CI, 0.97 to 1.49), which triggered early reporting for futility. The 4-year IDFS was 88.2% for TC6 and was 90.7% for TaxAC ( P = .04). Tests for treatment interaction by protocol, hormone receptor status, and nodal status were negative. Conclusion The TaxAC regimens improved IDFS in patients with high-risk human epidermal growth factor receptor 2–negative breast cancer compared with the TC6 regimen.

scholarly journals Train Performance Analysis Using Heterogeneous Statistical Models

Atmosphere ◽  
2021 ◽  
Vol 12 (9) ◽  
pp. 1115
Author(s):  
Jianfeng Wang ◽  
Jun Yu
Keyword(s):  
Statistical Models ◽  
High Speed ◽  
Cox Model ◽  
Markov Chain Model ◽  
Chain Model ◽  
Time Varying ◽  
Weather Variables ◽  
Winter Climate ◽  
Stratified Cox Model ◽  
Snow Precipitation

This study investigated the effect of a harsh winter climate on the performance of high-speed passenger trains in northern Sweden. Novel approaches based on heterogeneous statistical models were introduced to analyse the train performance to take time-varying risks of train delays into consideration. Specifically, the stratified Cox model and heterogeneous Markov chain model were used to model primary delays and arrival delays, respectively. Our results showed that weather variables including temperature, humidity, snow depth, and ice/snow precipitation have a significant impact on train performance.

scholarly journals PIN67 Extended Cox Model Analysis With Non-Proportional Hazards Applied in Real-World Observational Data

2012 ◽  
Vol 15 (4) ◽  
pp. A249
Author(s):  
X. Ji ◽  
X. Gao ◽  
J.W. Baddley ◽  
R. Chambers ◽  
C.T. Solem ◽  
...  
Keyword(s):  
Observational Data ◽  
Real World ◽  
Proportional Hazards ◽  
Cox Model ◽  
Model Analysis ◽  
Extended Cox Model

Prospective pooled analysis of six phase III trials investigating duration of adjuvant (adjuv) oxaliplatin-based therapy (3 vs 6 months) for patients (pts) with stage III colon cancer (CC): The IDEA (International Duration Evaluation of Adjuvant chemotherapy) collaboration.

2017 ◽  
Vol 35 (18_suppl) ◽  
pp. LBA1-LBA1 ◽  
Author(s):  
Qian Shi ◽  
Alberto F. Sobrero ◽  
Anthony Frank Shields ◽  
Takayuki Yoshino ◽  
James Paul ◽  
...  
Keyword(s):  
Cox Model ◽  
Disease Free Survival ◽  
Standard Of Care ◽  
Pooled Analysis ◽  
Phase Iii ◽  
Stage Iii ◽  
Differential Outcomes ◽  
Phase Iii Trials ◽  
Stratified Cox Model

LBA1 Background: Since 2004, 6 months (m) of oxaliplatin (oxali)-based treatment has been the standard of care as adjuvant therapy for stage III CC. Since oxali is associated with cumulative neurotoxicity, shorter duration of adjuv therapy could spare pts toxicity and lead to substantial saving in health expenditures. Methods: A prospective, pre-planned pooled analysis of 6 concurrently conducted randomized phase III trials (SCOT, TOSCA, Alliance/SWOG 80702, IDEA France (GERCOR/PRODIGE), ACHIEVE, HORG) was performed to evaluate the non-inferiority (NI) of 3m compared with 6m (ref) of adjuv FOLFOX/XELOX. The primary endpoint was disease-free survival (DFS), defined as time from enrolment to relapse, 2nd CRC, and death (all causes). NI was to be declared if the 2-sided 95% confidence interval (CI) for DFS hazard ratio (HR 3m v 6m) estimated by a stratified Cox model was below 1.12. NI was examined within regimen and stage subgroups as pre-planned. Results: The analysis included 12,834 pts from 12 countries, accrued from 6/07 to 12/15. Stage distribution: 13% T1-2, 66% T3, 21% T4; 28% N2; 40% received XELOX. G3+ neurotoxicity was higher in the 6m v 3m arm (16 v 3% FOLFOX, 9 v 3% XELOX, p<0.0001). With a median follow-up of 39 mos, 3263 DFS events were observed. Overall, the 3 year DFS rate was 74.6% (3m) and 75.5% (6m), with estimated DFS HR of 1.07 (95%CI, 1.00-1.15). The 3m v 6m DFS HRs were 1.16 (95%CI, 1.06-1.26) and 0.95 (95% CI, 0.85-1.06) for FOLFOX and XELOX treated pts, respectively. The 3m v 6m DFS HRs were 1.01 (95%CI, 0.90-1.12) in T1-3 N1, and 1.12 (95%CI, 1.03-1.23) for T4 or N2 pts. Conclusions: While NI was not established for the overall cohort, NI of 3m v 6m oxali-based adjuv therapy was supported for XELOX. As each IDEA trial treated varying proportions of pts with XELOX (0 to 75%), the regimen interaction likely produced the differential outcomes observed between individual studies. Certain substages (T1-3 N1) also showed NI for 3m v 6m. These data provide a framework for discussions on risks and benefits of individualized adjuv therapy approaches. Support: U10CA180821, U10CA180882, U10CA180888, U10CA180835, INCA, PHRC2009, EME 09/800/34, HTA 14/140/84, CRUK C1348/A15960, JFMC. Clinical trial information: NCT01150045.

scholarly journals Stratified Cox Model with Interactions in Analysis of Recurrent Events

2018 ◽  
Vol 3 (335) ◽  
pp. 207-218 ◽  
Author(s):  
Beata Bieszk‑Stolorz
Keyword(s):  
Recurrent Events ◽  
Statistical Data ◽  
Cox Model ◽  
Large Fraction ◽  
Study Data ◽  
Men And Women ◽  
Good Policy ◽  
Identification Of Persons ◽  
Stratified Cox Model ◽  
The Unemployed

The purpose of this paper is the assessment of relative intensity of exit from registered unemployment by means of the analysis of recurrent survival episodes and the comparison of these results with the results obtained for an individual episode. The stratified Cox model with interactions was used. Statistical data collected by labour offices indicate that a large fraction of the unemployed persons is registered multiple times. However, many of them resign from the mediation of labour offices and are subsequently removed from the register. In the article, the intensities of de‑registration due to various causes for men and women were compared. The study data came from the database of personal details of people registered by the Poviat Labour Office in Szczecin in 2013. The observation covered the records of their registration until the end of 2014. Gender of the unemployed persons influenced the intensity of de‑registrations in the first episodes, partially in the second and third ones, due to various causes, such as finding a job or removal from the register, whereas it did not influence the intensity of de‑registrations in the fourth and subsequent episodes. As for the other causes in the subsequent episodes, the differences were also not statistically significant. The proposed analysis may be important for implementing a good policy in the labour market. The identification of persons that resign from the mediation of the labour office is as interesting as the identification of these persons who find a job.

scholarly journals Progression-Free Survival at 24 Months (PFS24) and Complete Response at 30 Months (CR30) from Autologous Stem Cell Transplantation (ASCT) Should be Used As Surrogates for OS in Follicular Lymphoma (FL) Patients

Blood ◽  
2015 ◽  
Vol 126 (23) ◽  
pp. 521-521
Author(s):  
Ana Jiménez Ubieto ◽  
Carlos Grande García ◽  
Lucrecia Yañez ◽  
Dolores Caballero ◽  
Silvana Novelli ◽  
...  
Keyword(s):  
High Risk ◽  
Salvage Therapy ◽  
Cox Model ◽  
Progression Free Survival ◽  
Model Analysis ◽  
First Line ◽  
End Points ◽  
Free Survival ◽  
The Status

Abstract Introduction: At present, determining OS (Overall Survival) remains the gold standard in clinical trial endpoints which evaluate the role of ASCT in FL. However, in the context of a disease characterized by very long median survivals with a continuous pattern of relapse, and still more with the advent of novel treatments, OS assessments seem elusive. In fact, PFS (Progression Free Survival) is the standard endpoint for new drug approvals in first-line FL, and some other earlier endpoints such 2-year PFS (Casulo et al , JCO 2015) or CR30 (Sargent, 2015) have been recently proposed as potential surrogates and as alternatives to PFS or OS as primary end-points in FL patients treated with 1st line chemoinmunoterapy. Objective: To assess if 2-year PFS and CR30 are feasible surrogates of OS in the setting of a very long follow-up series of FL patients treated with ASCT. Material and Methods: A total of 626 chemosensitive FL patients (mean age 47 years, male 49%) reported to the Spanish GELTAMO registry and intensified with ASCT between 1989 and 2007 were analyzed. The status of the disease at the moment of ASCT was either in 1st response [203 in 1st CR, 43% of them needing more than one therapy line to reach the CR, and 140 in 1st Partial Response (PR)] or in response after salvage therapy (174 in 2nd CR, 28 in 3rd CR and 81 in 2nd or 3rd PR). In 615 patients the status of the disease was evaluable after ASCT: 569 cases (92%) in CR, 27 cases (4%) in PR and 19 cases (3%) progressed or died. To assess 2-year PFS, two groups were defined: patients with progression of disease (POD) within 2 years from ASCT (early POD) and patients without progression within 2 years from ASCT. Cox model analysis was used to evaluate the association between early POD and OS from a risk - defining event, which is survival from time of POD for early progressors or from 2 years after ASCT for the reference group (Casulo et al, JCO 2015.Appendix). To asses CR30, two groups were defined: patients with and patients without CR at 30 months from ASCT. Cox model analysis was used to evaluate the association between being or not in CR at 30 months from ASCT. Results: Median follow-up is 12.2 years from ASCT and 14.2 years from diagnosis. Of the assessable patients, 31% were in the high-risk FLIPI group and 40% in the high-risk FLIPI 2 group. 30% of patients received rituximab prior to ASCT. Globally median PFS and median OS are 11 and 21 years, respectively. Patients transplanted in PR (n=221; 35%) had a worse OS than those transplanted in CR (n=405, 65%): HR 2.45 (95% CI, 2.2-2.7; P<10-5) (fig. 1). Similar findings were found in the subgroup of patients transplanted as first line therapy HR 2.59 (95% CI, 2.3-2.8; P<10-5,), or as salvage therapy HR 2.69 (95% CI, 2.2-2.7; P<10-5). Of the global series, 25% (n=154) had an early POD, 71% (n=447) didn't progress or die within 2 years from ASCT and 4% (n=25) died without POD less than 2 years after ASCT, and were excluded from the analysis. Of the 405 patients transplanted in CR, 16% (n=63) had an early POD, 80% (n=325) didn't progress or die within 2 years from ASCT and 4% (n=17) died without POD less than 2 years after ASCT, and were excluded from the analysis. Early POD is associated with reduced OS in all context [the global series: HR 6.8 (95% CI, 6.5-7.1; P=10-5) (fig. 2); the patients transplanted in CR: HR 8.9 (95% CI, 8.4-9.3; P<10-5 (fig. 2); where a "plateau" is found, and the patients transplanted in only PR: HR 3.9(95% CI, 3.4-4.3; P<10-5). In the subgroup of patients treated with rituximab before ASCT (n=179) both, the fact of being transplanted in CR and the absence of an early POD, remain associated with an improve OS: HR 1.82 (CI 95%, 1.5-2-1; P=0.05, and HR 7.5 (CI 95%, 7.2-7.8; P=10-5), respectively. In the same way; for patients responding to ASCT (n=596; 97%), the absence of a CR at 30 months from ASCT (n=161), is associated with a reduced survival: HR 9.8 (95% CI 9.5-10.3; P>10-5), fig. 3. Conclusion: 2-year PFS and CR30 could be used as subrogates for OS and as primary end points, not only in FL patients treated with 1st line chemoinmunoterapy, but also in FL intensified with an ASCT. To our knowledge this is the first study to establish that early relapse after ASCT is predictive of poor survival in FL patients; in both, patients treated or not with rituximab previously to the ASCT. This finding is even more evident in patients transplanted in CR. Likewise, best response achieved before ASCT is a robust prognostic factor for OS in FL patients. Figure 1. Figure 1. Figure 2. Figure 2. Figure 3. Figure 3. Disclosures No relevant conflicts of interest to declare.

Sign in / Sign up

Export Citation Format

Share Document