Françoise Renard

Clinical outcome of breast cancer patients with liver metastases alone in the anthracycline-taxane era: a retrospective analysis of two prospective, randomised metastatic breast cancer trials.

Liver metastases have long been known to indicate an unfavourable disease course in breast cancer (BC). However, a small subset of patients with liver metastases alone who were treated with pre-taxane chemotherapy regimens was reported to have longer survival compared with patients with liver and metastases at other sites. In the present study, we examined the clinical outcome of breast cancer patients with liver metastases alone in the context of two phase III European Organisation for Research and Treatment of Cancer (EORTC) trials which compared the efficacy of doxorubicin (A) versus paclitaxel (T) (trial 10923) and of AC (cyclophosphamide) versus AT (trial 10961), given as first-line chemotherapy in metastatic BC patients. The median follow-up for the patients with liver metastases was 90.5 months in trial 10923 and 56.6 months in trial 10961. Patients with liver metastases alone comprised 18% of all patients with liver metastases, in both the 10923 and 10961 trials. The median survival of patients with liver metastases alone and liver plus other sites of metastases were 22.7 and 14.2 months (log rank test, P=0.002) in trial 10923 and 27.1 and 16.8 months (log rank test, P=0.19) in trial 10961. The median TTP (time to progression) for patients with liver metastases alone was also longer compared with the liver plus other sites of metastases group in both trials: 10.2 versus 8.8 months (log rank test, P=0.02) in trial 10923 and 8.3 versus 6.7 months (log rank test, P=0.37) in trial 10961. Most patients with liver metastases alone have progression of their disease in their liver again (96 and 60% of patients in trials 10923 and 10961, respectively). Given the high prevalence of breast cancer, improved detection of liver metastases, encouraging survival achieved with currently available cytotoxic agents and the fact that a significant portion of patients with liver metastases alone have progression of their tumour in the liver again, a more aggressive multimodality treatment approach through prospective clinical trials seems worth exploring in this specific subset of women.

Endometrial disorders in 406 breast cancer patients on tamoxifen: the case for less intensive monitoring

OBJECTIVES To describe the endometrial appearance in postmenopausal breast cancer patients on tamoxifen and to assess a routine surveillance scheme for endometrial lesions. STUDY DESIGN Three hundred and seventeen postmenopausal breast cancer women already on tamoxifen at the start of the study (group I) and 89 breast cancer women assessed before any tamoxifen intake (group II) underwent an initial and then yearly scans with transvaginal ultrasonography, followed by an hysteroscopy and biopsy for women with an endometrium thickened above 8mm. Endometrial thickness was also measured in 823 women with no breast cancer nor tamoxifen intake (group III). RESULTS Initial mean endometrial thickness was 8.2mm in group I, 4.4mm in group II and 3.4mm in group III (P<0.001). Eighteen percent endometrial lesions were found in group I and 3.3% in group II. We observed a significant association between endometrial pathology and both cumulated dose and total duration. Polyps were the most frequent and first to appear pathology. Five cancers were detected in group I, and all of them had taken tamoxifen for more than 3 years. CONCLUSION Our surveillance scheme could be lightened; an acceptable screening scheme might include a baseline assessment before the start of tamoxifen and, if normal, yearly screening after 3 years of tamoxifen therapy, yearly surveillance for women with an abnormal baseline assessment and immediate investigation for symptomatic women.

The body site distribution of melanocytic naevi in 6–7 year old European children

The number and size of melanocytic naevi are the main predictors of cutaneous melanoma. Naevus development per unit of skin surface is greatest during childhood. We assessed the body distribution of naevi 2–4.9 mm and ⩾ 5 mm in 649 European children aged 6–7 years old from Brussels (Belgium), Bochum (Germany), Lyon (France) and Rome (Italy). The numbers of naevi 2–4.9 mm and naevi ⩾ 5 mm were strongly correlated, especially on the trunk. For naevi 2–4.9 mm, the highest relative densities were found on the face, back, shoulders and the external surface of the arms. The lowest relative densities were found on the hands, legs, feet and abdomen. The relative density of naevi ⩾ 5 mm was higher on the trunk than on any other body site. Similar body distributions were observed in both sexes and at each centre. The body site distribution of naevi 2–4.9 mm seemed to parallel the usual sun exposure patterns of young European children. It is suggested that the development of naevi ⩾ 5 mm might be a marker of the vulnerability of melanocytes to the harmful effects of solar radiation. Vulnerability would be maximal on the back, and would decrease from proximal to distal skin areas, with melanocytes of the hands and feet having the lowest vulnerability. The number of naevi acquired on a specific area of skin would result from the combined effects of local vulnerability to solar radiation and local sun exposure history. The origin of acquired body site differences in the susceptibility of melanocytes to ultraviolet radiation is unknown, although it seems to parallel the body site density of sensory innervation.

High burden of breast cancer in Belgium: recent trends in incidence (1999-2006) and historical trends in mortality (1954-2006)

IntroductionIn Belgium, breast cancer mortality has been monitored since 1954, whereas cancer incidence data have only been made available for a few years. In this article we update historical trends of breast cancer mortality and describe the recent breast cancer incidence.MethodsIncidence data were extracted from the Belgium Cancer Registry from 2004 to 2006 for the Walloon and Brussels Regions and Belgium, and from 1999 to 2006 for the Flemish Region. The Directorate-general Statistics and Economic information provided the mortality data for the years 1954-1999 and 2004. The regional authorities of the Flemish and Brussels Regions provided the mortality data for the years 2000-2003 and 2005-2006.ResultsIn 2004, the World age-standardised breast cancer incidence for the whole of Belgium was 110 per 100, 000 person-years for all ages; and 172, 390 and 345 per 100, 000 person-years for the 35-49, 50-69, and 70+ age groups, respectively. The incidence rate was slightly higher in each age group in the Brussels Region. In Flanders, where the incidence could be observed during a longer period, an increase was observed until 2003 in the 50-69 age group, followed by a decrease. To the contrary, in the oldest age group, incidence continued to rise over the whole period, whereas no change in incidence was observed between 1999 and 2006 in the 35-49 age group.Mortality increased until the late 1980s and afterwards decreased in all regions and in age groups younger than 70. In women of 70 years and older, the decline began later.ConclusionsThe burden of breast cancer in Belgium is very high. In 2004, Belgium ranked first for the age-standardised incidence rate in Europe for all ages combined and in the 35-49 and 50-69 age groups. The impact of the known risk factors and of mammographic screening should be further studied. The mortality rate in Belgium ranked lower than incidence, suggesting favourable survival. Plausible explanations for the discrepancy between incidence and mortality are discussed.

Decline in breast cancer incidence in the Flemish region of Belgium after a decline in hormonal replacement therapy

BACKGROUND Breast cancer incidence rate in Belgian women was as high as 152.7 for 100 000 in 2003 (adjusted on European population). We made an estimation of the contribution of hormone replacement therapy (HRT) on breast cancer incidence from 1999 to 2005 in women aged 50-69 years in Flanders. METHODS Breast cancer data were extracted from the Belgium Cancer Registry. Drug consumption was computed from drug sales data. The fraction of breast cancers attributable to HRT was calculated by year, using the relative risks of the Million Women Study in the UK. RESULTS The proportion of women aged 50-69 years using HRT in Flanders increased since 1992, peaked at 20% in 2001, then decreased to 8% in 2008. The incidence of breast cancer in 100 000 women aged 50-69 years in Flanders increased from 332.8 in 1999 to 407.9 in 2003, then decreased to 366.1 in 2005; the variations were mostly noticeable for tumors <20 mm in size. The fraction of breast cancers attributed to HRT peaked at 11% in 2001 and decreased afterward. CONCLUSION The high level of breast cancer observed in the years 2001-2003 in Flanders can be partly attributed to the use of HRT. Since participation to mammography screening of Flemish women aged 50-69 years was still on the rise in 2003 and never exceeds 62%, the decrease in breast cancer incidence was likely to be due to the decrease in HRT use and not to screening saturation.

Evaluation of a low-invasive strategy for prostate cancer screening with prostate-specific antigen.

OBJECTIVES To evaluate a two-step strategy for the detection of prostate cancer within the context of serial screening and compare this strategy with other screening strategies. The optimal combination of tests proposed for prostate cancer screening remains undetermined, particularly when screening is repeated over time. METHODS A prospective serial prostate cancer screening study with follow-up to 55 months was performed in a general community screening clinic. One thousand seven hundred seven self-referred men, 50 to 75 years old, without a history of prostate cancer agreed to undergo screening for prostate cancer on an annual basis. Serum prostate-specific antigen (PSA) measurement was the first-step screening test. If the serum PSA test was positive, a standard urologic evaluation was performed. Biopsy was recommended only if a test other than serum PSA was suspicious for cancer. The outcome measures were the biopsy rate and prostate cancer detection rate. The comparisons with other studies were age-standardized to correct for differences in age distribution. RESULTS The biopsy and cancer detection rates after the first test were 7.0% and 2.0%, respectively. After 4 years of the study, the cumulative biopsy rate and cumulative cancer detection rate per enrolled man was 12% and 4.1%, respectively. The comparisons between studies revealed that screening strategies using serum PSA as a first-line test had similar detection rates but lower biopsy rates than strategies performing biopsy when one of several screening tests was positive. CONCLUSIONS A two-step screening strategy using serum PSA alone as the initial test seemed able to detect as many cancers as when all screening tests were used at the same time but reduced the number of unnecessary biopsies.

A design to improve the comparability of area maps: the example of the premature mortality in Belgium

Methods A diverging green-to-red colour scale is used, where the cut-off colour (yellow) represents a reference value with the same meaning across all maps: the average sex-specific mortality rate for the cause of death studied. To represent the relative disparity, rates are classified according to a geometric progression with a 1.1 step, meaning that each colour class has a mortality rate 1.1 times higher than the preceding one. The number of classes is determined by the disparity between the extreme districts: the larger the disparity, the more colour classes present and the sharper the colour contrast. The relative distance between the highest and the lowest classes is 1.1 The midpoint of each class is calculated as the average rate * (1.1) rank of the class, starting from the average. The legend of the maps displays the boundaries of each class.