Breast cancer time to treatment in Martinique: predictive factors and effect on survival.

BACKGROUND: Martinique is the second French Region with the lowest physician-to-population ratio, which may affect waiting times for access to care. OBJECTIVES: To assess (i) factors influencing waiting times from diagnosis to cancer-related treatments in breast cancer women in Martinique, and (ii) the impact of waiting times on patients' survival. STUDY DESIGN: Retrospective observational study. METHODS: Data on women diagnosed with invasive breast cancer between 1st January 2013 and 31st December 2017 and initially treated by surgery were extracted from the Martinique population-based registry. A cox model was performed to find predictive factors for waiting times. A log-rank test was used to compare time-to-treatment between groups. RESULTS: In total, 713 patients were included (mean age: 58 ± 13). Median time from diagnosis to surgery was 40 [25-60] days. Age at diagnosis was found to predict variations in waiting times. Patients > 75 had longer waiting time to surgery than those < 40 or [40-50] (P = 0.016 and P < 0.001, respectively). Women with a time-to-treatment ≥ 4 months had a significant lower survival (P < 0.01). CONCLUSIONS: Specific interventions are needed to improve waiting time from diagnosis to initial treatment, as they are longer than recommended and affect survival time.

Epidemiology and survival of cervical cancer in the French West-Indies: data from the Martinique Cancer Registry (2002-2011).

BACKGROUND: The Caribbean ranks seventh among world regions most affected by cervical cancer. Social health inequalities, such as differences in access to screening services, engender disparities in incidence and mortality between low- and middle-income countries and industrialized countries. The French National Cancer Plan 2014-2019 focuses on reducing inequalities in cervical cancer. OBJECTIVE: The aim of this study was to describe the geographical distribution and overall survival of cervical cancer, based on data from a population-based cancer registry in Martinique (French West-Indies). METHODS: We included all cases of cervical cancer diagnosed between 2002 and 2011. The geographical distribution was described by zone of residence and by aggregated units for statistical information (IRIS). Based on the results of the model, standardized incidence rates (SIRs) were calculated using a Gamma Poisson model. Survival rates were calculated using the Kaplan-Meier method. Cox proportional hazards models were used to investigate the risk factors for cervical cancer mortality. RESULTS: A total of 1253 cases were analyzed (947 in situ tumors and 306 invasive cancers). 1230 cases with geolocalization were used to map the distribution of the incidence of in situ and invasive cervical cancers. Five IRIS were significantly over-incident. The 5-year overall survival rate was 55%, with a median survival of 6.5 years [95% CI: 4.9-10.1]. Multivariate analysis confirmed age at diagnosis (HR = 2.15 [1.50-3.09]; p < 0.0001), FIGO stage (HR = 3.53 [2.50-4.99]; p < 0.0001) and zone of residence (HR = 1.51 [1.06-2.13]; p = 0.02) as risk factors. CONCLUSIONS: Prognostic factors suggest that cervical cancer needs to be diagnosed at an early stage. Our results could allow cervical cancer screening programs to clearly identify geographical areas that would benefit from targeted interventions with a view to reducing incidence and mortality of cervical cancer in the Caribbean.