Cox Regression Model for Interval-Censored Data in Breast Cancer Follow-up Studies

2003 ◽  
Author(s):  
George Y. Wong
Keyword(s):  
Breast Cancer ◽  
Regression Model ◽  
Censored Data ◽  
Cox Regression ◽  
Interval Censored Data ◽  
Cox Regression Model ◽  
Follow Up Studies ◽  
Interval Censored

Evaluation of microRNA-10b expression as a novel predictive marker of metastases development and patients’ survival in breast cancer.

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. 576-576
Author(s):  
Paola Parrella ◽  
Raffaela Barbano ◽  
Barbara Pasculli ◽  
Andrea Fontana ◽  
Massimiliano Copetti ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Regression Model ◽  
Cox Regression ◽  
Rna Extraction ◽  
Distant Metastases ◽  
Pathological Examination ◽  
Breast Cancer Cell Lines ◽  
Cox Regression Model ◽  
Increased Risk

576 Background: MicroRNA-10b was found highly expressed in metastatic breast cancer cell lines and able to generate metastases in mice models. The aim of this study is to evaluate the putative association between miR10b expression and disease progression. Methods: We selectedfrom our tumor bank 150 consecutive breast cancers with at least three years follow up. For each case frozen paired tumor and normal tissue and complete clinical data were available. Pathological examination was performed to ensure that each tumour sample contained more than 70% of cancer cells resulting in 114 samples suitable for RNA extraction. RNA quality was measured and only samples with RIN≥7.0 were analyzed (n=101) by a relative quantification method. Results: miR10b relative expression in tumor to normal samples (RERs) was significantly higher in the subgroup of patients with metastases (median 0.25 IQR 0.11-1.02) as compared with patients without metastases (median 0.09 IQR 0.04-0.29) (P=0.023 Mann Whitney Test). The association between miR-10b RERs and survival was evaluated in the group of patients without metastases at diagnosis (n=90). In univariate Cox regression model, patients with high miR-10b RERs had a higher risk of distant metastases development (HR 4.91, P=0.02) and disease related death (HR 6.02; P=0.01). In a multivariate Cox regression model adjusted for tumor size, lymph node metastases, grade, ER, PgR status, and Ki67 labeling index (n=79), higher miR-10b RERs were still associated with increased risk of distant metastases development (HR18.84; P<0.001) and disease related death (HR 13.39; P=0.003) (Table). Conclusions: We show that in breast cancer patients miR-10b expression is associated with worse prognosis on a short term follow up. These results suggest that miR-10b expression could be used for individual patient’s risk assessment and perhaps as potential therapeutical target. [Table: see text]

scholarly journals FRI0181 ORGAN DAMAGE FREE SURVIVAL IN SOUTHERN CHINESE PATIENTS WITH ACTIVE LUPUS NEPHRITIS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 674.1-674
Author(s):  
C. C. Mok ◽  
C. S. Sin ◽  
K. C. Hau ◽  
T. H. Kwan
Keyword(s):  
Lupus Nephritis ◽  
Regression Model ◽  
Cox Regression ◽  
Organ Damage ◽  
Renal Survival ◽  
Cox Regression Model ◽  
Free Survival ◽  
Renal Response

Background:The goals of treatment of lupus nephritis (LN) are to induce remission, retard the progression of chronic kidney disease, prevent organ complications and ultimately reduce mortality. Previous cohort studies of LN have mainly focused on the risk of mortality and development of end stage renal failure (ESRF) (renal survival). The cumulative frequency of LN patients who survive without organ damage, which correlates better with the balance between treatment efficacy and toxicity, as well as quality of life, has not been well studied.Objectives:To study the organ damage free survival and its predictive factors in patients with active LN.Methods:Consecutive patients who fulfilled ≥4 ACR/SLICC criteria for SLE and with biopsy proven active LN between 2003 and 2018 were retrospectivey analyzed. Those with organ damage before LN onset were excluded. Data on renal parameters and treatment regimens were collected. Complete renal response (CR) was defined as normalization of serum creatinine (SCr), urine P/Cr (uPCR) <0.5 and inactive urinary sediments. Partial renal response (PR) was defined as ≥50% reduction in uPCR and <25% increase in SCr. Organ damage of SLE was assessed by the ACR/SLICC damage index (SDI). The cumulative risk of having any organ damage or mortality since LN was studied by Kaplan-Meier’s analysis. Factors associated with a poor outcome were studied by a forward stepwise Cox regression model, with entry of covariates with p<0.05 and removal with p>0.10.Results:273 LN patients were identified but 64 were excluded (organ damage before LN onset). 211 LN patients were studied (92% women; age at SLE 30.4±13.5 years; SLE duration at LN 1.9±3.1years). 47 (22%) patients had nephrotic syndrome and 60 (29%) were hypertensive. Histological LN classes was: III/IV±V (75.1%), I/II (7.8%) and pure V (17.1%) (histologic activity and chronicity score 7.0±4.2 and 1.8±1.5, respectively). Induction regimens were: prednisolone (33.1±17.5mg/day) in combination with intravenous cyclophosphamide (CYC) (21.4%; 1.0±0.2g per pulse), oral CYC (8.6%; 96.4±37.8mg/day), azathioprine (AZA) (14.3%; 78.6±25.2mg/day), mycophenolate mofetil (MMF) (22.8%; 1.9±0.43g/day) and tacrolimus (TAC) (17.1%; 4.3±1.1mg/day). After a follow-up of 8.6±5.4 years, 94(45%) patient developed organ damage (SDI≥1) and 21(10%) patients died. The commonest organ damage was renal (36.3%) and musculoskeletal (17.9%), and the causes of death were: infection (38.1%), malignancy (19.0%), cardiovascular events (9.5%) and ESRF complications (9.5%). At last visit, 114 (55%) patients survived without any organ damage. The cumulative organ damage free survival at 5, 10 and 15 years after renal biopsy was 73.5%, 59.6% and 48.3%, respectively. The 5, 10 and 15-year renal survival rate were 95.2%, 92.0% and 84.1% respectively. In a Cox regression model, nephritic relapse (HR 3.72[1.78-7.77]), proteinuric relapse (HR 2.30[1.07-4.95]) and older age (HR 1.89[1.05-3.37]) were associated with either organ damage or mortality, whereas CR (HR 0.25[0.12-0.50]) at month 12 were associated with organ damage free survival. Baseline SCr, uPCR and histological LN classes were not significantly associated with a poor outcome. Among patients with class III/IV LN, the long-term organ damage free survival were not significantly different in users of MMF (reference) from CYC (IV/oral) (HR 1.45[0.76- 2.75]) or TAC (HR 1.03[0.26-1.62]) as induction therapy.Conclusion:Organ damage free survival is achieved in 55% of patients with active LN upon 9 years of follow-up. CYC/MMF/TAC based induction regimens did not differ for the long-term outcome of LN. Targeting complete renal response and preventing renal relapses remain important goals of LN treatment.Acknowledgments:NILDisclosure of Interests:None declared

Sign in / Sign up

Export Citation Format

Share Document