Missing data in breast cancer: Relationship with survival in national databases.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e19114-e19114
Author(s):  
Jennifer Kay Plichta ◽  
Christel N. Rushing ◽  
Holly C. Lewis ◽  
Dan G. Blazer ◽  
Terry Hyslop ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Missing Data ◽  
Proportional Hazards ◽  
Cancer Registries ◽  
Cox Proportional Hazards ◽  
Future Research ◽  
Risk Of Death ◽  
Cox Proportional Hazards Models ◽  
N Stage ◽  
National Databases

e19114 Background: National cancer registries are valuable tools used to analyze patterns of care and clinical oncology outcomes; yet, patients with missing data may impact the accuracy and generalizability of these data. We sought to evaluate the association between missing data and overall survival (OS). Methods: Using the NCDB and SEER, we compared data missingness among patients diagnosed with invasive breast cancer from 2010-2014. Key variables included: demographic variables (age, race, ethnicity, insurance, education, income), tumor variables (grade, ER, PR, HER2, TNM stage), and treatment variables (surgery in both databases; chemotherapy and radiation in NCDB). OS was compared between those with and without missing data via Cox proportional hazards models. Results: Overall, 775,996 patients in the NCDB and 263,016 in SEER were identified; missingness of at least 1 key variable was 29% and 13%, respectively. Of those, the majority were missing a tumor variable (NCDB 80%; SEER 88%), while demographic and treatment variables were missing less often. When compared to patients with complete data, missingness was associated with a greater risk of death; NCDB 17% vs. 14% (HR 1.23, 99% CI 1.21-1.25) and SEER 27% vs 14% (HR 2.11, 99% CI 2.05-2.18). Rate of death was similar whether the patient was missing 1 or ≥2 variables. When stratified by the type of missing variable, differences in OS between those with and without missing data in the NCDB were small. In SEER, reductions in OS were largest for those missing tumor variables (HR 2.26, 99% CI 2.19-2.33) or surgery data (HR 3.84, 99% CI 3.32-4.45). Among the tumor variables specifically, few clinically meaningful differences in OS were noted in the NCDB, while the most significant differences in SEER were noted in T and N stage (table). Conclusions: Missingness of select variables is associated with a worse OS and is not uncommon within large national cancer registries. Therefore, researchers must use caution when choosing inclusion/exclusion criteria for outcomes studies. Future research is needed to elucidate which patients are most often missing data and why OS differences are observed. [Table: see text]

scholarly journals Changes in Metabolic Syndrome Status and Breast Cancer Risk: A Nationwide Cohort Study

Cancers ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 1177
Author(s):  
In Young Choi ◽  
Sohyun Chun ◽  
Dong Wook Shin ◽  
Kyungdo Han ◽  
Keun Hye Jeon ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Metabolic Syndrome ◽  
Proportional Hazards ◽  
Screening Program ◽  
Cox Proportional Hazards ◽  
Cox Proportional Hazards Models ◽  
Hazard Ratios ◽  
Increased Risk ◽  
Health Screening Program ◽  
Design And Methods

Objective: To our knowledge, no studies have yet looked at how the risk of developing breast cancer (BC) varies with changes in metabolic syndrome (MetS) status. This study aimed to investigate the association between changes in MetS and subsequent BC occurrence. Research Design and Methods: We enrolled 930,055 postmenopausal women aged 40–74 years who participated in a biennial National Health Screening Program in 2009–2010 and 2011–2012. Participants were categorized into four groups according to change in MetS status during the two-year interval screening: sustained non-MetS, transition to MetS, transition to non-MetS, and sustained MetS. We calculated multivariable-adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for BC incidence using the Cox proportional hazards models. Results: At baseline, MetS was associated with a significantly increased risk of BC (aHR 1.11, 95% CI 1.06–1.17) and so were all of its components. The risk of BC increased as the number of the components increased (aHR 1.46, 95% CI 1.26–1.61 for women with all five components). Compared to the sustained non-MetS group, the aHR (95% CI) for BC was 1.11 (1.04–1.19) in the transition to MetS group, 1.05 (0.96–1.14) in the transition to non-MetS group, and 1.18 (1.12–1.25) in the sustained MetS group. Conclusions: Significantly increased BC risk was observed in the sustained MetS and transition to MetS groups. These findings are clinically meaningful in that efforts to recover from MetS may lead to reduced risk of BC.

scholarly journals Migration and work in postwar Australia: mortality profile comparisons between Australian and Italian workers exposed to blue asbestos at Wittenoom

2017 ◽  
Vol 75 (1) ◽  
pp. 29-36 ◽  
Author(s):  
Alison Reid ◽  
Enzo Merler ◽  
Susan Peters ◽  
Nimashi Jayasinghe ◽  
Vittoria Bressan ◽  
...  
Keyword(s):  
Proportional Hazards ◽  
Asbestos Exposure ◽  
Cancer Registries ◽  
Cox Proportional Hazards ◽  
Cumulative Exposure ◽  
Foreign Born ◽  
Male Population ◽  
Cox Proportional Hazards Models ◽  
National Mortality ◽  
Milling Operation

ObjectivesThree hundred and thirty thousand Italians arrived in Australia between 1945 and 1966, many on assisted passage schemes where the worker agreed to a 2-year unskilled employment contract. Italians were the largest of 52 migrant groups employed at the Wittenoom blue asbestos mining and milling operation. We compare mortality from asbestos-related diseases among Italian and Australian workers employed at Wittenoom.MethodsA cohort of 6500 male workers was established from employment records and followed up at state and national mortality and cancer registries. SMRs were calculated to compare mortality with the Western Australian male population. Time-varying Cox proportional hazards models compared the risk of mesothelioma between Australian and Italian workers.Results1031 Italians and 3465 Australians worked at Wittenoom between 1943 and 1966. Duration of employment was longer for the Italian workers, although the concentration of exposure was similar. The mesothelioma mortality rate per 100 000 was higher in Italians (184, 95% CI 148 to 229) than Australians (128, 95% CI 111 to 149). The risk of mesothelioma was greater than twofold (HR 2.27, 95% CI 1.43 to 3.60) in Italians at the lowest asbestos exposure category (<10 fibre years/per mL).ConclusionsA hierarchy in migration, isolation and a shortage of workers led to Italians at Wittenoom incurring higher cumulative exposure to blue asbestos and subsequently a greater rate of malignant mesothelioma than Australian workers.ImpactPoor working conditions and disparities between native and foreign-born workers has had a detrimental and differential impact on the long-term health of the workforce.

Specific Adverse Events Predict Survival Benefit in Patients Treated With Tamoxifen or Aromatase Inhibitors: An International Tamoxifen Exemestane Adjuvant Multinational Trial Analysis

2013 ◽  
Vol 31 (18) ◽  
pp. 2257-2264 ◽  
Author(s):  
Duveken B.Y. Fontein ◽  
Caroline Seynaeve ◽  
Peyman Hadji ◽  
Elysée T.M. Hille ◽  
Willemien van de Water ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Adverse Events ◽  
Endocrine Therapy ◽  
Proportional Hazards ◽  
Disease Free Survival ◽  
Distant Metastases ◽  
Cox Proportional Hazards ◽  
Endocrine Treatment ◽  
Survival Outcomes ◽  
Cox Proportional Hazards Models

Purpose Specific adverse events (AEs) associated with endocrine therapy and related to depletion or blocking of circulating estrogens may be related to treatment efficacy. We investigated the relationship between survival outcomes and specific AEs including vasomotor symptoms (VMSs), musculoskeletal adverse events (MSAEs), and vulvovaginal symptoms (VVSs) in postmenopausal patients with breast cancer participating in the international Tamoxifen Exemestane Adjuvant Multinational (TEAM) trial. Patients and Methods Primary efficacy end points were disease-free survival (DFS), overall survival (OS), and distant metastases (DM). VMSs, MSAEs, and VVSs arising in the first year of endocrine treatment were considered. Patients who did not start or who discontinued their allocated therapy and/or had an event (recurrence/death) within 1 year after randomization were excluded. Landmark analyses and time-dependent multivariate Cox proportional hazards models assessed survival differences up to 5 years from the start of treatment. Results A total of 9,325 patients were included. Patients with specific AEs (v nonspecific or no AEs) had better DFS and OS (multivariate hazard ratio [HR] for DFS: VMSs, 0.731 [95% CI, 0.618 to 0.866]; MSAEs, 0.826 [95% CI, 0.694 to 0.982]; VVSs, 0.769 [95% CI, 0.585 to 1.01]; multivariate HR for OS: VMSs, 0.583 [95% CI, 0.424 to 0.803]; MSAEs, 0.811 [95% CI, 0.654 to 1.005]; VVSs, 0.570 [95% CI, 0.391 to 0.831]) and fewer DM (VMSs, 0.813 [95% CI, 0.664 to 0.996]; MSAEs, 0.749 [95% CI, 0.601 to 0.934]; VVSs, 0.687 [95% CI, 0.436 to 1.085]) than patients not reporting these symptoms. Increasing numbers of specific AEs were also associated with better survival outcomes. Outcomes were unrelated to treatment allocation. Conclusion Certain specific AEs are associated with superior survival outcomes and may therefore be useful in predicting treatment responses in patients with breast cancer treated with endocrine therapy.

Acute myeloid leukemia (AML) in older women after adjuvant breast cancer therapy

2006 ◽  
Vol 24 (18_suppl) ◽  
pp. 560-560 ◽  
Author(s):  
D. A. Patt ◽  
Z. Duan ◽  
G. Hortobagyi ◽  
S. H. Giordano
Keyword(s):  
Breast Cancer ◽  
Adjuvant Chemotherapy ◽  
Older Women ◽  
Proportional Hazards ◽  
Cox Proportional Hazards ◽  
Significant Independent Predictor ◽  
Kaplan Meier ◽  
Cox Proportional Hazards Models ◽  
Adjuvant Breast Cancer ◽  
Comorbidity Scores

560 Background: Adjuvant chemotherapy for breast cancer is associated with the development of secondary AML, but this risk in an older population has not been previously quantified. Methods: We queried data from the Surveillance, Epidemiology, and End Results-Medicare (SEER-Medicare) database for women who were diagnosed with nonmetastatic breast cancer from 1992–1999. We compared the risk of AML in patients with and without adjuvant chemotherapy (C), and by differing C regimens. The primary endpoint was a claim with an inpatient or outpatient diagnosis of AML (ICD-09 codes 205–208). Risk of AML was estimated using the method of Kaplan-Meier. Cox proportional hazards models were used to determine factors independently associated with AML. Results: 36,904 patients were included in this observational study, 4,572 who had received adjuvant C and 32,332 who had not. The median patient age was 75.3 (66.0–103.3). The median follow up was 63 months (13–132). Patients who received C were significantly younger, had more advanced stage disease, and had lower comorbidity scores (p<0.001). The unadjusted risk of developing AML at 10 years after any adjuvant C for breast cancer was 1.6% versus 1.1% for women who had not received C. The adjusted HR for AML with adjuvant C was 1.72 (1.16–2.54) compared to women who did not receive C. HR for radiation was 1.21 (0.86–1.70). HR was higher with increasing age but p>0.05. An analysis was performed among women who received C. When compared to other C regimens, anthracycline-based therapy (A) conveyed a significantly higher hazard for AML HR 2.17 (1.08–4.38), while patients who received A plus taxanes (T) did not have a significant increase in risk HR1.29 (0.44–3.82) nor did patients who received T with some other C HR 1.50 (0.34–6.67). Another significant independent predictor of AML included GCSF use HR 2.21 (1.14–4.25). In addition, increasing A dose was associated with higher risk of AML (p<0.05). Conclusions: There is a small but real increase in AML after adjuvant chemotherapy for breast cancer in older women. The risk appears to be highest from A-based regimens, most of which also contained cyclophosphamide, and may be dose-dependent. T do not appear to increase risk. The role of GCSF should be further explored. No significant financial relationships to disclose.

Brain metastases in breast cancer: Different survival by biological subtype and Ki67 expression

2009 ◽  
Vol 27 (15_suppl) ◽  
pp. 1069-1069
Author(s):  
D. Sartori ◽  
M. Bari ◽  
G. L. Pappagallo ◽  
F. Rosetti ◽  
S. Olsen ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Triple Negative ◽  
Proportional Hazards ◽  
Fold Increase ◽  
Cox Proportional Hazards ◽  
Her2 Overexpression ◽  
Cns Metastases ◽  
Risk Of Death ◽  
Pathologic Features ◽  
Increased Risk

1069 Background: Ten to 15% of patients (pts) with breast cancer will be diagnosed with central nervous system (CNS) metastases, and autopsy series suggest that up to 30% of pts have evidence of CNS disease at the time of death. The idenfication of factors that may predispose to CNS metastasis may help lead to earlier detection and possibly to improvement in disease management. Methods: Breast cancer pts with CNS metastases were identified within a database of 1300 breast cancer diganoses from 1995 to 2007 at the Department of Oncology, Azienda ULSS 13 VE. Pathologic features of tumor samples were examined using standard immunohistochemical assays. Results: Fifty-one pts with CNS metastases were identified. Median age at primary breast cancer diagnosis was 49 years (range, 28–78); median time to CNS metastases was 45 months (range, 3–244). HER2 overexpression was found in tumors from 25 pts (49.0%); 23 pts had tumors lacking overexpression of HER2, estrogen receptors (ER), and progesterone receptors (PgR) (ie, “triple negative” disease). Overexpression of p53 (at least 20% tumor cells positive), Ki67 (at least 20%), and BCL2 (at least 30%) were detected in tumors from 16 pts (31.4%), 32 pts (62.7%), and 14 pts (27.5%), respectively. Median survival from CNS involvement was 3.67 months (95% CI 2.05–5.28), with 24.4% and 15.3% of patients estimated to be alive at 12 and 24 months, respectively (Kaplan-Meier product limit method). A Cox proportional hazards analysis found that Ki67 overexpression was the only factor independently associated with a significantly increased risk of death (2.7-fold increase, p=0.028), while triple negative status was associated with a 1.8-fold increase in the risk of death (P=0.08) (Table). Conclusions: In our series of breast cancer pts with CNS metastases, nearly all had either HER2 overexpression or triple-negative disease. Pts whose tumors had higher proliferative indices, assessed by Ki67, had the poorest prognosis. [Table: see text] [Table: see text]

Antiandrogen withdrawal (AAWD) in patients (pts) with prostate cancer (PCa): Retrospective analysis of data from the South Australian Prostate Cancer Clinical Outcome Collaborative (SA-PCCOC).

2015 ◽  
Vol 33 (7_suppl) ◽  
pp. 240-240
Author(s):  
Sina Vatandoust ◽  
Ganessan Kichenadasse ◽  
Michael E O'Callaghan ◽  
Tina Kopsaftis ◽  
Scott Walsh ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Overall Survival ◽  
Proportional Hazards ◽  
South Australia ◽  
Progression Free Survival ◽  
Cox Proportional Hazards ◽  
Risk Of Death ◽  
Cox Proportional Hazards Models ◽  
Australian State ◽  
Chi Squared

240 Background: In 15-30% of pts with metastatic PCa who progress on Maximal Androgen Blockade (MAB), withdrawal of the antiandrogen agent (AAWD) and continuing the LHRH agonist alone, leads to PSA decreases of ≥50% and prolonged progression free survival. Here we describe patient and disease characteristics, treatment history and outcomes of pts who have been managed with AAWD. Methods: Data were obtained from SA-PCCOC (a longitudinal, observational registry of biopsy-proven PCa cases, throughout the Australian state of South Australia since 1998). Proportions were compared using a Chi squared test. A multivariable model used competing risks (Fine and Gray) and Cox proportional Hazards models to assess overall survival and Prostate cancer specific mortality (PCSM). Survival was calculated from the date of rising PSA for patients on LHRH and AA. Results: 140 pts were found to have MAB. Of these, 31(22.1%) had AAWD. In the AAWD group, median age was 81y (51-95). Age at diagnosis, Gleason score at biopsy and diagnostic PSA were not significantly different amongst the two groups. Treatment PSA was significantly lower in the AAWD group (20.55 (range 0.6-9,995) vs 50.50 (range 0.95-4378) p= 0.02). There was a significant association of AAWD with PCSM (sHR 0.35, 95% CI 0.16-0.76; p = 0.008). Also significant in the model was prior time on hormones (sHR [per month increase] 0.96 95% CI 0.95-0.98, p<0.001). There was also a significant association of AAWD with overall survival (HR 0.22, 95% CI 0.10-0.46; p <0.001). Again, prior time on hormones was also significant (HR [per month increase] 0.96 95% CI 0.95-0.98, p<0.001). Multivariate analysis was performed on data from 80 pts (60 pts omitted due to missing data). Conclusions: Pts in whom AAWD was used were older and had lower treatment PSA. In this small cohort, AAWD was associated with both reduced PCSM and overall risk of death. The time spent on MAB also appeared to be significant. This retrospective observational study may be subject to confounding, however the observation warrants further investigation in larger cohorts and in a prospective setting.

The role of ALBI and IGF-CTP score in refining prognostication of HCC.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e16659-e16659
Author(s):  
Sunyoung S. Lee ◽  
Yehia I. Mohamed ◽  
Aliya Qayyum ◽  
Manal Hassan ◽  
Lianchun Xiao ◽  
...  
Keyword(s):  
Overall Survival ◽  
Proportional Hazards ◽  
Cox Proportional Hazards ◽  
Survival Prediction ◽  
Score System ◽  
Risk Of Death ◽  
U Statistics ◽  
Cox Proportional Hazards Models ◽  
Ctp Score

e16659 Background: Child-Turcotte-Pugh (CTP) score is widely used in the assessment of prognosis of HCC and CTP-A is the standard criterion for active therapy and clinical trials entry. Recently, ALBI and insulin-like growth factor-1 (IGF)-CTP scores have been reported to improve survival prediction over CTP score. However, comparative studies to compare both scores and to integrate IGF into Albi score are lacking. Methods: After institutional board approval, data and samples were prospectively collected. 299 HCC patients who had data to generate both IGF-CPG and Albi index were used. The ALBI index, and IGF score were calculated, Cox proportional hazards models were fitted to evaluation the association between overall survival (OS) and CTP, IGF-CTP, Albi and IGF, albumin, bilirubin. Harrell’s Concordance index (C-index) was calculated to evaluate the ability of the three score system to predict overall survival. And the U-statistics was used to compare the performance of prediction of OS between the score system. Results: OS association with CTP, IGF-CTP and Albi was performed (Table). IGF-CTP B was associated with a higher risk of death than A (HR = 1.6087, 95% CI: 1.2039, 2.1497, p = 0.0013), ALBI grade 2 was also associated with a higher risk of death than 1 (HR = 2.2817, 95% CI: 1.7255, 3.0172, p < 0.0001). IGF-1(analyzed as categorical variable) was independently associated with OS after adjusting for the effects of ALBI grade. Which showed IGF-1 ≤26 was significantly associated with poor OS, P = 0.001. Conclusions: Although ALBI grade and IGF-CTP score in this analysis had similar prognostic values in most cases, their benefits might be heterogenous in some specific conditions. We looked into corporation of IGF-1 into ALBI grade, IGF score with cutoff ≤26 which clearly refined OS prediction and better OS stratification of ALBI-grade.

scholarly journals Characteristics and outcomes of an international cohort of 400,000 hospitalised patients with Covid-19

2021 ◽  
Author(s):  
◽  
Christiana Kartsonaki
Keyword(s):  
Clinical Outcomes ◽  
Proportional Hazards ◽  
Invasive Mechanical Ventilation ◽  
Case Fatality ◽  
Cox Proportional Hazards ◽  
Risk Of Death ◽  
Cox Proportional Hazards Models ◽  
Hospitalised Patients ◽  
Male Sex ◽  
Viable Model

Background: Policymakers need robust data to respond to the COVID-19 pandemic. We describe demographic features, treatments and clinical outcomes in the International Severe Acute Respiratory and emerging Infection Consortium (ISARIC) COVID-19 cohort, the world's largest international, standardised cohort of hospitalised patients. Methods: The dataset analysed includes COVID-19 patients hospitalised between January 2020 and May 2021. We investigated how symptoms on admission, comorbidities, risk factors, and treatments varied by age, sex, and other characteristics. We used Cox proportional hazards models to investigate associations between demographics, symptoms, comorbidities, and other factors with risk of death, admission to intensive care unit (ICU), and invasive mechanical ventilation (IMV). Findings: 439,922 patients with laboratory-confirmed (91.7%) or clinically-diagnosed (8.3%) SARS-CoV-2 infection from 49 countries were enrolled. Age (adjusted hazard ratio [HR] per 10 years 1.49 [95% CI 1.49-1.50]) and male sex (1.26 [1.24-1.28]) were associated with a higher risk of death. Rates of admission to ICU and use of IMV increased with age up to age 60, then dropped. Symptoms, comorbidities, and treatments varied by age and had varied associations with clinical outcomes. Tuberculosis was associated with an 86% higher risk of death, and HIV with an 87% higher risk of death. Case fatality ratio varied by country partly due to differences in the clinical characteristics of recruited patients. Interpretation: The size of our international database and the standardized data collection method makes this study a reliable and comprehensive international description of COVID-19 clinical features. This is a viable model to be applied to future epidemics.

Early hemoglobin decline and survival prognosis in epoetin alfa studies

2009 ◽  
Vol 27 (15_suppl) ◽  
pp. 9625-9625
Author(s):  
J. A. Berlin ◽  
P. J. Bowers ◽  
S. Rao ◽  
S. Sun ◽  
K. Liu ◽  
...  
Keyword(s):  
Proportional Hazards ◽  
Cox Proportional Hazards ◽  
Time Dependent ◽  
Epoetin Alfa ◽  
Survival Prognosis ◽  
Proportional Hazards Models ◽  
Risk Of Death ◽  
Cox Proportional Hazards Models ◽  
Level Data ◽  
Increased Risk

9625 Background: When cancer patients (pts) with chemotherapy-induced anemia (CIA) respond to erythropoietic-stimulating agents (ESA), hemoglobin (Hb) typically increases within 4–8 weeks. This exploratory analysis examined whether mortality differs depending on Hb response after 4 or 8 weeks of epoetin alfa (EPO) treatment or depending on transfusion. Methods: Pt-level data were analyzed from 31 randomized studies (7,215 pts) of epoetin alfa vs non-EPO (15 studies) or placebo (16 studies) in pts with CIA. A landmark analysis was used; Hb change was set at a specific time (4 and 8 weeks) and subsequent survival was examined separately for EPO and placebo. Pts were categorized as “Hb increased” (>0.5 g/dL), “Hb decreased” (>0.5 g/dL), or “Hb stable” (within ±0.5 g/dL) compared to baseline. Hb stable was compared to other Hb change categories with Cox proportional hazards models, stratified by study and adjusted for potential confounders. Results: The hazard ratio (HR) for Hb decreased versus Hb stable at 4 weeks was 1.44 for EPO (95% CI: 1.04, 1.99), indicating worse survival for pts with a decline in Hb. This association was weaker for placebo (HR: 1.12; 95% CI: 0.74, 1.67). Increased risk with declining Hb in EPO-treated pts was most pronounced in studies that maintained Hb ≥12 g/dL or treated pts for >12–16 weeks (1,876 pts). Patterns were similar using the 8-week landmark. In both EPO-treated and placebo pts, transfusion increased the rate of on-study death ∼3.5 fold (treating transfusion as a time-dependent variable). Conclusions: These exploratory findings suggest that both decreased Hb after 4 or 8 weeks of EPO treatment and transfusion are associated with increased risk of death. In spite of adjustment for other prognostic factors, it is likely that this association reflects poorer underlying prognosis of pts whose Hb fails to respond. ESAs should be discontinued in the absence of a Hb response. [Table: see text]

Differences between taxanes in time on therapy and associated discontinuation events in metastatic breast cancer: Results from a U.S. claims analysis.

2013 ◽  
Vol 31 (26_suppl) ◽  
pp. 158-158 ◽  
Author(s):  
Rex W. Force ◽  
Brooke A. Pugmire ◽  
Gary Binder ◽  
Cindy Duval ◽  
Deya Corzo ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Metastatic Breast Cancer ◽  
Median Time ◽  
Proportional Hazards ◽  
Metastatic Breast ◽  
Cox Proportional Hazards ◽  
Claims Analysis ◽  
Insurance Plan ◽  
Cox Proportional Hazards Models ◽  
Clinical Benefits

158 Background: Taxanes in metastatic breast cancer (mBC) have demonstrated clinical benefits but have different toxicity profiles. This study evaluated differences in time to discontinuation (D/C) between taxanes overall and by specific events associated with D/C (i.e., death, adverse events (AE), subsequent therapy). Methods: Women receiving >1 dose of either paclitaxel (P), docetaxel (D), or nab-paclitaxel (nab-P) between 2006-2009 were identified by ICD-9 codes indicative of mBC from a commercial payer claims database. Logistic regression and Cox-proportional hazards models were used to compare rates of D/C and time to D/C, respectively. Control variables were age, use of prior or concurrent chemo, and comorbidity score. Drug D/C was defined as no taxane administration for 35 days. AE within 35-days after D/C were considered taxane-associated. Death or disenrollment from the insurance plan within 90 days and initiation of subsequent therapy within 60 days after D/C were considered temporally associated with D/C. Results: 4,503 pts with mBC were included (2,599 received D; 1,643 received P, and 261 received nab-P). The most common reason for D/C in all pts was AEs (37.8%), initiation of subsequent therapy (34.4%), and death or disenrollment (8.4%). A higher proportion of pts on D (29.4%) and P (17.5%) had D/C due to neutropenia vs nab-P (6.9%, p<0.001 for both comparisons). A higher proportion of pts on nab-P (12.3%) continued on therapy until death/disenrollment vs D (8.9%, p 0.046) or P (7.1%, p 0.023). Similarly, a higher proportion of pts on nab-P (51.2%) initiated subsequent therapy at D/C of taxane vs D (31%) or P (37.3%; p<0.01 for both comparisons). Median time to D/C was 85 days for each of D and P, and 127 days for nab-P (p<0.05). Median time to D/C associated with AEs, neutropenia, death/disenrollment or subsequent therapy were significantly longer for nab-P compared with D or P (p<0.05). Conclusions: Women treated for mBC stayed on nab-P therapy 50% longer than with other taxanes. This time on therapy advantage also indicated that pts on nab-P were less likely to D/C due to AEs, more likely to stay on therapy, and more likely to be candidates for subsequent therapies.

Sign in / Sign up

Export Citation Format

Share Document