Causes of death in women with breast cancer: a risks and rates study on a population-based cohort.

Introduction: The increasing survival of patients with breast cancer has prompted the assessment of mortality due to all causes of death in these patients. We estimated the absolute risks of death from different causes, useful for health-care planning and clinical prediction, as well as cause-specific hazards, useful for hypothesis generation on etiology and risk factors. Materials and methods: Using data from population-based cancer registries we performed a retrospective study on a cohort of women diagnosed with primary breast cancer. We carried out a competing-cause analysis computing cumulative incidence functions (CIFs) and cause-specific hazards (CSHs) in the whole cohort, separately by age, stage and registry area. Results: The study cohort comprised 12,742 women followed up for six years. Breast cancer showed the highest CIF, 13.71%, and cardiovascular disease was the second leading cause of death with a CIF of 3.60%. The contribution of breast cancer deaths to the CIF for all causes varied widely by age class: 89.25% in women diagnosed at age <50 years, 72.94% in women diagnosed at age 50-69 and 48.25% in women diagnosed at age ≥70. Greater CIF variations were observed according to stage: the contribution of causes other than breast cancer to CIF for all causes was 73.4% in women with stage I disease, 42.9% in stage II-III and only 13.2% in stage IV. CSH computation revealed temporal variations: in women diagnosed at age ≥70 the CSH for breast cancer was equaled by that for cardiovascular disease and "other diseases" in the sixth year following diagnosis, and an early peak for breast cancer was identified in the first year following diagnosis. Among women aged 50-69 we identified an early peak for breast cancer followed by a further peak near the second year of follow-up. Comparison by geographic area highlighted conspicuous variations: the highest CIF for cardiovascular disease was more than 70% higher than the lowest, while for breast cancer the highest CIF doubled the lowest. Conclusion: The integrated interpretation of absolute risks and hazards suggests the need for multidisciplinary surveillance and prevention using community-based, holistic and well-coordinated survivorship care models.

Excess risk of subsequent malignant neoplasms in adolescent and young adult cancer survivors: Results from the first Italian population-based cohort.

BACKGROUND: Evidence about late effects in adolescent and young adult (AYA) cancer survivors is scarce. This study assessed the risk of subsequent malignant neoplasms (SMNs) to identify the most common SMNs to be considered in follow-up care. METHODS: Population-based cancer registries retrospectively identified first primary tumors (between 1976 and 2013) and SMNs in AYAs (15-39 years old at their cancer diagnosis). AYA cancer survivors were those alive at least 5 years after their first cancer diagnosis. The excess risk of SMNs was measured as standardized incidence ratios (SIRs) and absolute excess risk together with the cumulative incidence of SMNs. RESULTS: The cohort included 67,692 AYA cancer survivors. The excess risk of developing any SMN (SIR, 1.6; 95% confidence interval, 1.5-1.7) was 60%. The excess risk of SMNs was significantly high for survivors of lymphomas; cancers of the breast, thyroid, female genital tract, digestive organs, gonads, and urinary tract; and melanomas. The cumulative incidence of all SMNs in AYA cancer survivors within 25 years of their first cancer diagnosis was approximately 10%. Subsequent tumors contributing to approximately 60% of all SMNs were breast cancer, colorectal cancer, corpus uteri cancer, and ovarian cancer in females and colorectal cancer, bladder cancer, prostate cancer, lung cancer, and lymphomas in males. CONCLUSIONS: These results highlight the need to personalize follow-up strategies for AYA cancer survivors.

Adolescent and Young Adult Cancer Survivors: Design and Characteristics of the First Nationwide Population-Based Cohort in Italy.

Purpose: Adolescent and young adult (AYA, 15-39 years) cancer survivors (alive at least 5 years after cancer diagnosis) are less studied than younger and older cancer survivors and research on their late effects is limited. To facilitate research on long-term outcomes of AYA cancer survivors, we established, in Italy, a population-based AYA cancer survivors' cohort. This article describes the study design and main characteristics of this cohort. Methods: The cohort derives from population-based cancer registries (CRs). Each CR identified AYA cancer patients retrospectively. Treatment for first primary cancer and all health events from diagnosis to death can be traced through linkage with available health databases, such as hospital discharge records (HDRs), mortality files, and outpatient and pharmaceutical databases. Results: Thirty-four CRs participated to the cohort which overall includes 93,291 AYAs with cancer and 67,692 cancer survivors. First primary cancer distribution in AYA cancer survivors differs by sex and age groups because of the different cancer types diagnosed in AYAs. Almost 78% of AYA cancer survivors have HDRs and 14.8% also pharmaceutical and outpatient databases. Conclusion: This cohort will be used to study, for the first time in Italy, the pattern and excess risk of late effects in AYA cancer survivors. HDRs, outpatient and pharmaceutical databases will be used to define primary treatment to assess its impact on AYA cancer survivors' late effects. This cohort exploiting data sources already available at CRs, minimize the data collection effort and it will contribute to assess the feasibility of using administrative database to study cancer survivors' late effects.

[Childhood cancer incidence in Apulia Region (Southern Italy), 2003-2008]. / Incidenza dei tumori maligni infantili in Puglia, 2003-2008.

OBJECTIVES: to investigate, for the first time, the incidence rates of malignant childhood cancers (children aged 0-14 years) in Apulia Region (Southern Italy) in the period 2003-2008. DESIGN: to compute incidence rates of childhood cancers from Apulia Region Cancer Registry database compared with the corresponding results published in 2012 by the Italian cancer registries network (AIRTUM),where data from the Apulia population were not included, because not available. SETTING AND PARTICIPANTS: we selected all incident cases of malignant tumours (behaviour: /3 of ICD-O-3 classification) in children aged 0-14 registered in the Apulia cancer registry. Local health unit (LHU) of Lecce (section of the Apulia cancer registry) collected data from 2003 to 2006; LHU of Taranto, BT, and Brindisi collected data from 2006 to 2008. MAIN OUTCOME MEASURES: we computed crude, age specific, and directly standardised rates (DSR), with 95% confidence intervals, of all malignant tumours, all categories and 5 subgroups of the ICCC-3 classification; standardised incidence ratios (SIR) for all childhood malignant tumours using the rates of the AIRTUM Pool 2003-2008 as reference . RESULTS: incident cases were 183. DSR (x106) of all childhood malignant tumours are: Apulia Region 169.7 (95%CI 145.9- 196.4); Brindisi 160.4 (95%CI 106.2-232.9); BT 177.7 (95%CI 122.7-248.7); Lecce 144.3 (95%CI 111.1-184.2); Taranto 216.2 (95%CI 163.0-281.4). SIR estimates are: Apulia Region 102.9 (95%CI 88.5-119.0); Brindisi 100.2 (95%CI 66.6-144.9); BT 105.4 (95%CI 73.0-147.2); Lecce 85.5 (95%CI 66.0-109.0); Taranto 134.6 (95%CI 101.7-174.8). Main incidence measures for all ICCC-3 categories and five subgroups of childhood cancers in Apulia are also reported. CONCLUSIONS: in Apulia Region, we estimated a DSR for all childhood malignant tumours very close to that of the AIRTUM Pool. DSRs for each ICCC-3 category look comparable with the data from the national survey too. When data of each LHU were analysed, the SIR estimate makes it evident an excess of all malignant childhood cancers in the LHU of Taranto. Other results of particular cancers and specific age groups also provide suggestions for further investigations.

BMI variation increases recurrence risk in women with early-stage breast cancer.

AIMS: The prognostic role of BMI variation during and/or after treatments for early-stage breast cancer is still unknown. PATIENTS & METHODS: The χ(2) test was conducted to explore the correlation between breast cancer recurrence and BMI changes in 520 early-stage breast cancer patients. Cox proportional hazard models were used to analyze the association of BMI changes, baseline BMI, known prognostic factors and recurrences. RESULTS: BMI gain was significant determinant of recurrences (p = 0.0008). In multivariate analyses, BMI variation more than 5.71% was associated with higher rates of recurrences, as well as age less than 55 years, stage disease and molecular subtype. CONCLUSION: Women who experience BMI gain after breast cancer may be at increased risk of poor outcomes.