to 88.2%±22.6% (p = <0.001). BMI, blood systolic/dyastolic pressure, total cholesterol and triglycerides levels remained unchanged. In Group II at 6 months, fasting glucose level increased from 92.5±18.4 to 97.0±20.7 (p = <0.001); no significant changes were seen in other parameters. Conclusion: Results from this study indicate that a low-cost coun- selling programme can impact on metabolic parameters in breast cancer patients. Disclosure of Interest: No significant relationships. P086 Vitamin D levels in Swiss breast cancer survivors M. Baumann 1 , S. Dani 2 *, D. Dietrich 3 , A. Hochstrasser 4 , M.T. Mark 5 , W. Riesen 1 , T. Ruhstaller 6 , A.J. Templeton 7 , B. Thürlimann 6 . 1 Institute for Clinical Chemistry and Haematology, St. Gallen Cantonal Hospital, St. Gallen, Switzerland, 2 Department of Medical Oncology and Haematology, St. Gallen Cantonal Hospital, St. Gallen, Switzerland, 3 Statistical Office, Swiss Group for Clinical Cancer Research - SAKK, Berne, Switzerland, 4 Oncology Centre, Männedorf Hospital, Zurich, Switzerland, 5 Department of Oncology and Haematology, Graubuenden Cantonal Hospital, Chur, Switzerland, 6 Breast Centre St. Gallen, Cantonal Hospital, St. Gallen, Switzerland, 7 Department of Oncology, St. Claraspital Basel and Faculty of Medicine, University of Basel, Basel, Switzerland We conducted a cross-sectional study in 332 women in Eastern Switzerland with early, i.e. non-metastatic breast cancer. Cholecal- ciferol intake and serum calcidiol [25-hydroxyvitamin D, 25(OH)D] and calcitriol [1,25-dihydroxyvitamin D, 1,25(OH)2D] levels were measured at the first visit (baseline) and during a follow-up visit in median 210 days after the first visit. Patients presenting calcidiol deficiency were advised to take cholecalciferol supplemen- tation. At baseline 60 (18%) patients had vitamin D deficiency (calcidiol ≤ 50 nmol/l) whereas 70 (21%) had insufficiency (50– 74 nmol/l). Out of 121 patients with ongoing cholecalciferol supplementation at baseline, calcidiol deficiency and insufficiency was observed in 9 (7%) and 16 (13%) patients, respectively, whereas out of 52 patients with no supplementation, 15 (29%) patients had deficiency and 19 (37%) patients had insufficiency. Only 85 (26%) patients had optimal calcidiol levels (75–100 nmol/l) at baseline. Seasonal variation was significant for calcidiol ( p = 0.042) and calcitriol (p = 0.001) levels. Living in a rural area was associated with a higher median calcidiol concentration as compared to living in an urban area (87 nmol/l versus 72 nmol/l, p = 0.001). Regular sport activity was positively associated with calcidiol (p = 0.045). Body mass index was inversely related to both calcidiol and calcitriol (rho = -0.24, p < 0.001; rho = -0.23, p < 0.001, respectively). The levels of calcidiol and calcitriol were correlated (rho = 0.21, p < 0.001). Age and bone mineral density had no significant correlation to the levels of calcidiol. Follow-up calcidiol was available for 230 patients, 44 (19%) of whom had calcidiol deficiency and 47 (21%) had insufficiency; 25 (41.6%) initially calcidiol-deficient patients attained sufficient calcidiol levels, whereas 33 (16.5%) patients with sufficient baseline calcidiol levels became deficient. Only 67 (30%) patients presented optimal calcidiol at the follow-up. In conclusion, a remarkable fraction of the patients had serum calcidiol below (40%) or above (30%) optimal levels and only around 30% of patients had optimal levels. Calcidiol and calcitriol levels increased on cholecal- ciferol supplementation, but the usual supplementation regimens were not adequate to bring calcidiol to the optimal range for a large proportion of patients. Adequate vitamin D supplementation must be tailored according to individual characteristics, needs and preferences. Disclosure of Interest: No significant relationships. P087 Advanced stage breast cancer is less often diagnosed in women who attend breast cancer screening L. De Munck 1 *, J. Fracheboud 2 , G. de Bock 3 , S. Siesling 1 , M. Broeders 4 . 1 Netherlands Comprehensive Cancer Organisation, Utrecht, Netherlands, 2 University Medical Center Rotterdam, Rotterdam, Netherlands, 3 University of Groningen, University Medical Center Groningen, Groningen, Netherlands, 4 Department for Health Evidence, Radboudumc and Dutch Reference Centre for Screening, Nijmegen, Netherlands Aims: The contribution of screening mammography to a reduced rate of advanced breast cancer is still questioned. The aim of this study is to assess age-specific incidence rates of advanced breast cancer in women who attended the screening programme compared to non- attenders. Methods: All women, aged 49 and older, diagnosed with breast cancer between 2006 and 2011 were selected from the Netherlands Cancer Registry. Data were linked to the Netherlands breast cancer screening programme, including data of screened women between 2004 and 2011 to covera period of at least 24 months before breast cancer diagnosis. Screen-detected cancers and interval cancers were defined as screen-related cancers diagnosed <24 months after a screening examination, all other cancers were defined as non- screen-related cancers. Two definitions of advanced breast cancer were used: (1) advanced stage defined as stages III and IV cancers (versus stages 0, I, II), and (2) advanced T-stage defined as invasive tumour sizes ≥15 mm (versus <15mm or DCIS), irrespective of lymph node stage. Incidence rates were age-adjusted using the European Standard Population 2013 as a reference. Multivariable logistic regression was used to estimate ratios for advanced stage between subgroups. Results: In total 72,612 breast cancers were included, of which 44,246 were screen-related cancers (61%) including 32,158 screen-detected cancers (73%). Incidence of advanced stage cancer was 38/100.000 in the screened population compared to 94/100.000 in the non- screened population (p < 0.001). Applying the second definition of advanced stage cancer resulted in much higher incidence rates for both populations. However, advanced T-stage was still less often diagnosed in the screened population (169/100.000 vs 194/100.000; p < 0.001). Non-screen related cancers had a three times higher risk to be advanced stage cancer compared to screen-related cancer (OR: 2.90, 95%CI: 2.78–3.02). Conclusions: Advanced breast cancer incidence rates are substan- tially lower in a screened population compared to a non-screened population, supporting the stage shift related to early detection of breast cancer. As the different definitions foradvanced breast cancer led to significant different estimates of the incidence of advanced cancers, we recommend that in the evaluation of a breast cancer screening programme the definition of stage is clearly stated. Disclosure of Interest: No significant relationships. P088 Cancer incidence and mortality trends in urban cities of India V. Deshmane*, S. Koyande, S. Jadhav. Indian Cancer Society, Mumbai, India Introduction: The Mumbai Cancer Registry (MCR), India’s oldest population based cancer registry (PBCR) covers 603 sq. km. of Greater Mumbai as well as three other urban cities viz. Pune, Aurangabad and Nagpur. Here we report the increasing incidence in breast cancer in all the four PBCRs. Breast cancer is now the commonest cancer in women in all four cities. Mortality trend for Mumbai is also reported. Methods: Breast cancer incidence was calculated for the year 2012 in Mumbai, Pune (2014), Nagpur (2013) and Aurangabad (2015). Trends were studied for the following periods: Mumbai 1964–2012, Pune 1972–2014, Aurangabad 1990–2015 and Nagpur 1980–2013. 15th St.GallenInternational Breast Cancer Conference / The Breast 32S1 (2017) S22–S77 S51