Breast cancer management and outcome according to surgeon's affiliation: a population-based comparison adjusted for patient's selection bias.

BACKGROUND: Studies have reported that breast cancer (BC) units could increase the quality of care but none has evaluated the efficacy of alternative options such as private BC networks, which is our study objective. PATIENTS AND METHODS: We included all 1404 BC patients operated in the public unit or the private network and recorded at the Geneva Cancer Registry between 2000 and 2005. We compared quality indicators of care between the public BC unit and the private BC network by logistic regression and evaluated the effect of surgeon's affiliation on BC-specific mortality by the Cox model adjusting for the propensity score. RESULTS: Both the groups had high care quality scores. For invasive cancer, histological assessment before surgery and axillary lymph node dissection when indicated were less frequent in the public sector (adjusted odds ratio (OR): 0.4, 95% confidence interval (CI) 0.3-0.7, and OR: 0.4, 95% CI 0.2-0.8, respectively), while radiation therapy after breast-conserving surgery was more frequent (OR: 2.5, 95% CI 1.4-4.8). Surgeon affiliation had no substantial effect on BC-specific mortality (adjusted hazard ratio (HR): 0.8, 95% CI 0.5-1.4). CONCLUSIONS: This study suggests that private BC networks could be an alternative to public BC units with both structures presenting high quality indicators of BC care and similar BC-specific mortality.

Impact of a positive family history on diagnosis, management, and survival of breast cancer: different effects across socio-economic groups.

BACKGROUND: This study aims to investigate whether increased awareness of breast cancer, due to a positive family history (FH), reduces diagnostic, therapeutic, and survival differences between women of low versus high socio-economic status (SES). METHODS: All breast cancer patients registered between 1990 and 2005 at the population-based Geneva Cancer Registry were included. With multivariate logistic and Cox regression analysis, we estimated the impact of SES and FH on method of detection, treatment, and mortality from breast cancer. RESULTS: SES discrepancies in method of detection and suboptimal treatment, as seen among women without a FH, disappeared in the presence of a positive FH. SES differences in stage and survival remained regardless of the presence of a positive FH. Overall, positive FH was associated with better survival. This effect was the strongest in women of high SES (age-adjusted Hazard Ratio [HR(ageadj)] 0.54 [0.3-1.0]) but less pronounced in women of middle (0.77 [0.6-1.0]), and absent in women of low SES (0.80 [0.5-1.2]). CONCLUSION: A positive FH of breast cancer may reduce SES differences in access to screening and optimal treatment. However, even with better access to early detection and optimal treatment, women of low SES have higher risks of death from their disease than those of high SES.

Characteristics and outcome of prostate cancer with PSA <4 ng/ml at diagnosis: a population-based study.

This population-based study aims to assess prognosis of prostate cancer diagnosed with prostate-specific antigen (PSA) levels <4 ng/ml in routine care. Materials and methods We compared prostate cancer patients with low PSA values (n=59) with other prostate cancer patients (n=1330) by logistic regression and the Cox model using data from the Geneva Cancer Registry. Results Patients with low PSA values more frequently had early-stage and well differentiated tumours. Nevertheless, 35% presented with aggressive tumour characteristics or metastases. After adjustment for other prognostic factors, prostate cancer-specific mortality was similar for both groups (hazard ratio: 1.1; 95%CI: 0.6-2.2). Conclusion We conclude that cancer with low PSA values at diagnosis is not indolent.

Characteristics and outcome of prostate cancer with PSA <4 ng/ml at diagnosis: a population-based study

This population-based study aims to assess prognosis of prostate cancer diagnosed with prostate-specific antigen (PSA) levels <4 ng/ml in routine care. Materials and methods We compared prostate cancer patients with low PSA values (n=59) with other prostate cancer patients (n=1330) by logistic regression and the Cox model using data from the Geneva Cancer Registry. Results Patients with low PSA values more frequently had early-stage and well differentiated tumours. Nevertheless, 35% presented with aggressive tumour characteristics or metastases. After adjustment for other prognostic factors, prostate cancer-specific mortality was similar for both groups (hazard ratio: 1.1; 95%CI: 0.6-2.2). Conclusion We conclude that cancer with low PSA values at diagnosis is not indolent (AU)

Effectiveness of organised versus opportunistic mammography screening.

BACKGROUND: Detailed comparison of effectiveness between organised and opportunistic mammography screening operating in the same country has seldom been carried out. PATIENTS AND METHODS: Prognostic indicators, as defined in the European Guidelines, were used to evaluate screening effectiveness in Switzerland. Matching of screening programmes' records with population-based cancer registries enabled to compare indicators of effectiveness by screening and detection modality (organised versus opportunistic screening, unscreened, interval cancers). Comparisons of prognostic profile were also drawn with two Swiss regions uncovered by service screening of low and high prevalence of opportunistic screening, respectively. RESULTS: Opportunistic and organised screening yielded overall little difference in prognostic profile. Both screening types led to substantial stage shifting. Breast cancer prognostic indicators were systematically more favourable in Swiss regions covered by a programme. In regions without a screening programme, the higher the prevalence of opportunistic screening, the better was the prognostic profile. CONCLUSIONS: Organised screening appeared as effective as opportunistic screening. Mammography screening has strongly influenced the stage distribution of breast cancer in Switzerland, and a favourable impact on mortality is anticipated. Extension of organised mammography screening to the whole of Switzerland can be expected to further improve breast cancer prognosis in a cost-effective way.

Hormonal therapy for oestrogen receptor-negative breast cancer is associated with higher disease-specific mortality.

BACKGROUND: Tamoxifen has a remarkable impact on the outcome of oestrogen receptor (ER)-positive breast cancer. Without proven benefits, tamoxifen is occasionally prescribed for women with ER-negative disease. This population-based study aims to estimate the impact of tamoxifen on the outcome of ER-negative disease. METHODS: We identified all women (n = 528) diagnosed with ER-negative invasive breast cancer between 1995 and 2005. With Cox regression analysis, we calculated breast cancer mortality risks of patients treated with tamoxifen compared with those treated without tamoxifen. We adjusted these risks for the individual probabilities (propensity scores) of having received tamoxifen. RESULTS: Sixty-nine patients (13%) with ER-negative disease were treated with tamoxifen. Five-year disease-specific survival for women treated with versus without tamoxifen were 62% [95% confidence interval (CI) 48% to 76%] and 79% (95% CI 75% to 83%), respectively (P(Log-rank) < 0.001). For ER-negative patients, risk of death from breast cancer was significantly increased in those treated with tamoxifen compared with patients treated without tamoxifen (adjusted hazard ratio = 1.7, 95% CI 1.1-2.9, P = 0.031). CONCLUSION: Our results show that patients with ER-negative breast cancer treated with tamoxifen have an increased risk of death from their disease. Tamoxifen use should be avoided for these patients.

Hormone replacement therapy, mammography screening and changing age-specific incidence rates of breast cancer: an ecological study comparing two European populations.

BACKGROUND: In 2003, for the first time, US breast cancer incidence rates have fallen. Experts argue whether this is due to the reduced uptake of screening mammography or to lower use of Hormone Replacement Therapy (HRT). This study aims to disentangle the respective impact of screening and HRT on age-incidence rates and histology of breast cancer, by comparing two populations with comparably high levels of screening mammography, but with different prevalence of HRT. METHODS: We included all invasive breast cancers recorded at the Geneva cancer registry (n = 4,909) and the Netherlands Cancer Registry (n = 152,428) between 1989-2003. We compared age-specific incidence rates and trends in histological subtyping between the two populations. RESULTS: Between 1989-1991, incidence rates increased with age in both populations. In 2001-2003, women aged 60-64 years showed highest incidence rates in Geneva, while in the Netherlands incidence rates continued to increase with age. The annual increase in ductal cancer incidence was similar in the Netherlands (2.3%) and Geneva (2.5%), but the annual increase in lobular cancer was sharper in Geneva (10%) than in the Netherlands (5%). CONCLUSION: The sharp differences in age distribution and histological subtyping of breast cancer between two European populations are not attributable to screening, since both populations have a high uptake of mammography screening. Since the prevalence of HRT use is very high in Geneva and rather low in the Netherlands, HRT may explain these discrepancies. However, other etiological factors and differences in histological assessment may also have played a role.

Recent increase of breast cancer incidence among women under the age of forty.

Using data from the Geneva Cancer Registry, we found that in 2002-2004, breast cancer incidence in women aged 25-39 years increased by 46.7% per year (95% CI: 7.1-74.0, P=0.015), which surveillance or detection bias may not fully explain.

Increased risk of acute myeloid leukaemia after treatment for breast cancer.

This study evaluates the risk of acute myeloid leukaemia (AML) in patients treated for breast cancer. We included all 6360 breast cancer patients that were recorded at the Geneva Cancer Registry between 1970 and 1999. Patients were followed for AML occurrence until December 2000. We calculated standardized incidence ratios of AML and identified factors modifying the risk of AML by multivariate Cox analysis. Twelve (0.2%) patients developed AML. In general, patients treated for breast cancer had a 3.5-fold (95% confidence interval (CI): 1.8-6.0) increased risk of developing AML compared with the general population. In particular, patients who were older than 70 years at breast cancer diagnosis and those treated with radiotherapy (with or without chemotherapy) had a significantly increased risk of developing AML. This population-based study confirms that radiotherapy increases the risk of AML. Due to the relatively low number of women treated with chemotherapy without radiotherapy and due to the infrequency of the disease, the question of whether chemotherapy alone increases this risk of AML cannot yet be answered.

Impact of familial risk factors on management and survival of early-onset breast cancer: a population-based study.

This population-based study evaluates the impact of a strong family history of breast cancer on management and survival of women with early-onset disease. We identified all breast cancer patients

Poorer survival of elderly patients with ovarian cancer: a population-based study.

OBJECTIVE: To compare ovarian cancer survival in elderly and young patients. MATERIAL AND METHODS: Using the Geneva Cancer Registry, we identify women diagnosed with primary ovarian cancer between 1980 and 1998. We compared tumors characteristics, treatment patterns of young patients (70 years) by logistic regression. To evaluate the effect of age on prognosis, we compared disease specific survival by Cox proportional hazard analysis, taking into account other prognostic factors. RESULTS: This study included 285 patient aged 70 years and 451

The impact of coding process on observed cancer mortality trends in Switzerland.

Official cancer mortality in Switzerland decreased by about 16% over the 9-year period 1990-1998 and this trend has often been used to suggest that secondary prevention by screening for breast cancer could be useless. However, the clear downshift observed between 1994 and 1995 for some cancers, such as female breast and prostate, and the simultaneous change in ICD classification used by the Federal Office for Statistics in 1995 (ICD-8 to ICD-10) could be related, suggesting an impact of coding process on the observed trend. For every death occurred between 1980 and 1999, the death certificates have been retrieved, the cause of death has been recoded and site-specific mortality rates have been calculated again for each year during this period. As suggested, the trend appears to be overestimated: in order to be comparable with current rates, the mortality observed before 1995 should be lowered by about 7% for men and 5% for women. The error may be partially due to attributing the cause of death to co-morbidity factors not normally (and nowadays) defined as the underlying cause. Logically, the impact of such a miscoding is more important among older people and for cancer sites with long survival. For instance, the correction should be around 15% for female breast, 12% for prostate and up to 40% for testicular cancer.

Set-up of a population-based familial breast cancer registry in Geneva, Switzerland: validation of first results.

BACKGROUND: This article evaluates the accuracy of family history of breast and ovarian cancer among first-degree relatives of breast cancer patients, retrospectively collected during the setting up of a population-based family breast cancer registry. PATIENTS AND METHODS: Family histories of cancer for all women with breast cancer recorded at the Geneva Cancer Registry from 1990 to 1999 were retrospectively extracted from medical files. The accuracy of these family histories was validated among Swiss women born in Geneva: all 119 with a family history of breast (n = 110) or ovarian (n = 9) cancer and a representative sample of 100 women with no family history of breast or ovarian cancer. We identified the first-degree relatives of these women with information from the Cantonal Population Office. All first-degree relatives, resident in Geneva from 1970 to 1999, were linked to the cancer registry database for breast and ovarian cancer occurrence. Sensitivity, specificity and level of overall agreement (kappa) were calculated. RESULTS: Among 310 first-degree relatives identified, 61 had breast cancer and six had ovarian cancer recorded at the Geneva Cancer Registry. The sensitivity, specificity and kappa of the reported family histories of breast cancer were 98%, 97% and 0.97, respectively. For ovarian cancer, the sensitivity, specificity and kappa were 67%, 99%, and 0.66, respectively. CONCLUSIONS: This study indicates that retrospectively obtained family histories are very accurate for breast cancer. For ovarian cancer, family histories are less precise and may need additional verification.

Management of women with ductal carcinoma in situ of the breast: a population-based study.

BACKGROUND: Increasing incidence of ductal carcinoma in situ (DCIS) confronts patients and clinicians with optimal treatment decisions. This multidisciplinary study investigates therapeutic modalities of DCIS in daily practice and provides recommendations on how to increase quality of care. PATIENTS AND METHODS: All women (n = 116) with unilateral DCIS recorded in the Geneva Cancer Registry from 1995 to 1999 were considered. Information concerned patient and tumor characteristics, treatment and outcome. Factors linked to therapy were determined using a case-control approach. Cases were women with treatment of interest and controls other women on the study. RESULTS: Most DCIS cases (62%) were discovered by mammography screening. Ninety (78%) women had breast-conserving surgery (BCS), 18 (16%) mastectomy and seven (6%) bilateral mastectomy. Eight (7%) patients had tumor-positive margins, 18 (16%) lymph node dissection and two (1.7%) chemotherapy. Twenty-five per cent of women with BCS had no radiotherapy, three had radiotherapy after mastectomy. Less than 50% underwent breast reconstruction after mastectomy. Method of discovery, multifocality, tumor localization, size and differentiation were linked to the use of BCS or lymph node dissection. CONCLUSIONS: Because of important disparities in DCIS management, recommendations are made to increase quality of care, in particular to prevent axillary dissection or bilateral mastectomy and to increase the use of radiotherapy after BCS.

Adjuvant chemotherapy for colon carcinoma with positive lymph nodes: use and benefit in routine health care practice.

In 1990, an international consensus was reached on the efficacy of adjuvant chemotherapy for lymph node positive (stage III) colon carcinoma (CC). This study evaluates the use and benefit of such therapy in routine health care practice. The study includes all patients with stage III CC treated by putative curative surgery (n = 182) recorded at the Geneva cancer registry between 1990 and 1996. Factors modifying chemotherapy use were determined by logistic regression, considering patients with chemotherapy as cases (n = 55) and others as controls (n = 127). The effect of chemotherapy on the 5-year survival was evaluated by the Cox model. Analyses were adjusted for possible confounders. The use of chemotherapy increased over the period (P(trend) < 0.001). Age strongly modulated chemotherapy use. In 1996, 54% of eligible patients received chemotherapy, this proportion fell to 13% after age 70. Decisions to use chemotherapy significantly depended on stage, grade and cancer site. The chance to be treated was non-significantly lower among individuals of low social class, widowed and foreigners. Chemotherapy significantly decreased mortality rates (Hazard ratio: 0.35, 95%CI: 0.18-0.68), independently of the prognostic factors and with similar benefit regardless of stage and age group. Strong beneficial effect of adjuvant chemotherapy on stage III CC can be achieved in routine practice. However, this study shows that it is probably not optimally utilised in Switzerland, particularly among the elderly.

Surveys on mammography frequency in Geneva.

Two surveys were conducted in Geneva, in 1991 and 1995, to assess the coverage of mammography before the introduction of a breast cancer screening programme. Women who attended for mammographies did so at their own request, or were referred by doctors (more by gynaecologists than general practitioners). In 1995, the total female population was around 200,000, of which the target population for screening (age group 50 to 69) was 46,000 persons. The total number of mammographies observed increased by 23%, with a higher increase for "routine" mammographies (+35%). However, these figures hide certain trends, with a high proportion of educated women performing regular breast investigations. The two year coverage rate for mammography among the 50 to 69 age group is estimated at 38%, which means that about 20,000 women within this target population in Geneva still escape screening.

Determinants of long term survival after surgery for cancer of the lung: A population-based study.

BACKGROUND: Even if some determinants of lung cancer (LC) prognosis have been established, their independent effect on long term survival remains to be seen. The objective of the current study was to identify the prognostic indicators of long term survival among LC patients treated by surgery. METHODS: All patients with LC recorded at the Geneva Cancer Registry between 1977 and 1987 were analyzed by logistic regression, considering as cases (n = 98) those patients alive 10 years after their initial diagnosis and as controls (n = 330) all other patients, excluding those who did not undergo putative curative surgery. The effect of each prognostic factor was evaluated after accounting for age and gender ("crude" effect) and also for other a priori confounding factors (adjusted effect). Additional models considered two staging variables simultaneously to identify the strongest staging determinant. Results were presented as relative risk estimates of long term (>/=10 years) survival. RESULTS: Age, histology, and stage of disease significantly influenced prognosis regardless of the confounding factors considered. Gender also emerged as a discriminated factor in LC outcome, with a 2.1-fold increased chance (95% confidence interval, 1.6-3.5) of long term survival for women compared with men. Method of discovery, presence of symptoms, period of diagnosis, socioeconomic status, and tumor differentiation did not appear to be associated with long term survival. Extent and size of the tumor were found to be the most reliable prognostic staging factors, whereas adenopathy had no effect after accounting for extension. CONCLUSIONS: The current population-based study quantifies the independent effect of the factors modifying the chances of curability in patients with LC. In particular, it provides additional evidence that gender strongly influences long term survival.

Recent trends in incidence of cervical cancer in several regions of south-western Europe.

In recent decades, most Western countries have experienced a decline in the incidence of invasive cervical cancer. More recently, a reversal of this trend has been noted in young women, especially in anglo-saxon countries. These trends have been attributed to the beneficial results of cervical cancer screening on the one hand, to the widespread increase of high risk sexual behaviors on the other. Recent trends in Latin European countries have not previously been studied. Time trends of incidence data from 10 regions of South-Western Europe (Geneva, Vaud, Calvados, Doubs, Bas-Rhin, Isère, Ragusa, Varese, Zaragoza and Navarra) were analysed by means of log-linear models. The data were provided for various periods of time between 1970 and 1990. The mean incidence rate varies threefold between Navarra where rates are the lowest to Calvados where they are the highest. The overall decrease rate is of the order of -3% per year but rate changes differ by age group. The reduction seems to concern mostly middle aged women. A definite trend could not be determined among younger women due to low case numbers. Thus, the hypothesis of a rising incidence in young women cannot be confirmed at this time.

Mortality and incidence of cancer among a cohort of self employed butchers from Geneva and their wives.

To investigate whether specific cancers are associated with the occupation of butcher, as has been reported from other countries, a historical prospective cohort study was undertaken. The cohort consisted of all self employed butchers (n = 552) and pork butchers (n = 310) born since 1880 who set up a shop in the canton of Geneva from 1901 to 1969, and of their wives (n = 887). The study group was followed up from 1901 to 1990 for general mortality, from 1942 to 1990 for cause specific mortality, and from 1970 to 1989 for incidence of cancer. There was no trace of 45 men (5%) and 52 women (6%). Compared with the general population of the canton of Geneva, butchers and pork butchers experienced a significant increase, taking into account 15 years of latency, in mortality from all causes (observed deaths (Obs) 540, expected deaths (Exp) 445.5, standardised mortality ratio (SMR) 121, 90% confidence interval (90% Cl) 113-130). There were significant excesses in incidence and mortality from colorectal cancer, cancer of the prostate, and all malignant neoplasms, and in incidence of cancer of the liver. The risk of lung cancer was significantly increased among pork butchers (SMR 176, 90% Cl 114-262; standardised incidence ratio (SIR) 231, 90% Cl 137-368) but not among butchers (SMR 92, 90% Cl 59-138; SIR 113, 90% Cl 67-179). There was also a significant increase in mortality from cancer of the larynx among butchers. For non-malignant causes of death significant excesses were found among all men for ischaemic heart disease, motor vehicle accidents, and cirrhosis of the liver. Analysis of subgroups showed a cluster of deaths from leukaemia among older butchers born between 1880 and 1899 (Obs 5, Exp 0.6, p < 0.0001). Exposure of pork butchers to polycyclic aromatic hydrocarbons during meat smoking, which was assessed in a contemporary study, might have contributed to their increased risk of lung cancer. The possible role of other factors, especially cigarette smoking, nitrosamines, and oncogenic viruses was discussed. Moreover, there was evidence from another contemporary study that butchers and pork butchers ate more animal fat, and probably more animal protein, than the average male population of Geneva. These results suggest that dietary factors could be implicated in the excesses of colorectal cancer, cancer of the prostate, and ischaemic heart disease. An increased risk for alcohol abuse might explain the excesses of liver cirrhosis, cancer of the liver, cancer of the larynx and motor vehicle accidents. Among all wives overall mortality was similar to that expected (SMR 100, 90% Cl 93-108) and there was no significant excess risk for any specific cancer nor for any non-malignant cause of death. Results for cancer of the cervix uteri, especially among subgroups, suggest an increased risk consistent with previous findings from other countries.