Per Broberg

Magnetic resonance imaging reveals altered distribution of hepatic fat in children with type 1 diabetes compared to controls

INTRODUCTION Children with type 1 diabetes have been identified as a risk group for non-alcoholic fatty liver disease (NAFLD). The aim was to compare total hepatic fat fraction and fat distribution across Couinaud segments in children with type 1 diabetes and controls and the relation of hepatic fat to plasma and anthropometric parameters. METHODS Hepatic fat fraction and fat distribution across Couinaud segments were measured with magnetic resonance imaging (MRI) in 22 children with type 1 diabetes and 32 controls. Blood tests and anthropometric data were collected. RESULTS No children had NAFLD. Children with type 1 diabetes had a slightly lower hepatic fat fraction (median 1.3%) than controls (median 1.8%), and their fat had a different segmental distribution. The fat fraction of segment V was the most representative of the liver as a whole. An incidental finding was that diabetes patients treated with multiple daily injections of insulin (MDI) had a fat distribution more similar to controls than patients with continuous subcutaneous insulin infusion (CSII). CONCLUSIONS In children with type 1 diabetes, NAFLD may be less common than recent studies have suggested. Children with type 1 diabetes may have a lower fat fraction and a different fat distribution in the liver than controls. Diabetes treatment with MDI or CSII may affect liver fat, but this needs to be confirmed in a larger sample of patients. The heterogeneity of hepatic fat infiltration may affect results when liver biopsy is used for diagnosing fatty liver.

Serological evaluation of possible exposure to Ljungan virus and related parechovirus in autoimmune (type 1) diabetes in children.

Exposure to Ljungan virus (LV) is implicated in the risk of autoimmune (type 1) diabetes but possible contribution by other parechoviruses is not ruled out. The aim was to compare children diagnosed with type 1 diabetes in 2005–2011 (n = 69) with healthy controls (n = 294), all from the Jämtland County in Sweden, using an exploratory suspension multiplex immunoassay for IgM and IgG against 26 peptides of LV, human parechoviruses (HPeV), Aichi virus and poliovirus in relation to a radiobinding assay (RBA) for antibodies against LV and InfluenzaA/H1N1pdm09. Islet autoantibodies and HLA‐DQ genotypes were also determined. 1) All five LV‐peptide antibodies correlated to each other (P < 0.001) in the suspension multiplex IgM‐ and IgG‐antibody assay; 2) The LV‐VP1_31–60‐IgG correlated with insulin autoantibodies alone (P = 0.007) and in combination with HLA‐DQ8 overall (P = 0.022) as well as with HLA‐DQ 8/8 and 8/X subjects (P = 0.013); 3) RBA detected LV antibodies correlated with young age at diagnosis (P < 0.001) and with insulin autoantibodies (P < 0.001) especially in young HLA‐DQ8 subjects (P = 0.004); 4) LV‐peptide‐VP1_31–60‐IgG correlated to RBA LV antibodies (P = 0.009); 5) HPeV3‐peptide‐IgM and ‐IgG showed inter‐peptide correlations (P < 0.001) but only HPeV3‐VP1_1–30‐IgG (P < 0.001) and VP1_95–124‐IgG (P = 0.009) were related to RBA LV antibodies without relation to insulin autoantibody positivity (P = 0.072 and P = 0.486, respectively). Both exploratory suspension multiplex IgG to LV‐peptide VP1_31–60 and RBA detected LV antibodies correlated with insulin autoantibodies and HLA‐DQ8 suggesting possible role in type 1 diabetes. It remains to be determined if cross‐reactivity or concomitant exposure to LV and HPeV3 contributes to the seroprevalence. J. Med. Virol. 87:1130–1140, 2015.

Occupational sedentariness and breast cancer risk

Abstract Background: Epidemiological studies have indicated that physical activity reduces the risk of developing breast cancer. More recently, sedentary behavior has been suggested as a risk factor independent of physical activity level. The purpose of the present study was to investigate occupational sedentariness and breast cancer risk in pre- and postmenopausal women. Materials and methods: In a population-based prospective cohort study (n = 29 524), working history was assessed by a questionnaire between 1990 and 1992. Participants were classified as having: (1) sedentary occupations only; (2) mixed occupations or (3) non-sedentary occupations only. The association between occupational sedentariness and breast cancer incidence was analyzed by Cox regression, adjusted for known risk factors and participation in competitive sports. Results: Women with a working history of occupational sedentariness had a significantly increased risk of breast cancer (adjusted HR 1.20; 95% CI 1.05, 1.37) compared with those with mixed or non-sedentary occupations. The association was stronger among women younger than 55 years (adjusted HR 1.54; 95% CI 1.20, 1.96), whereas no association was seen in women 55 years or older. Adjustment for participation in competitive sports did not change the association. Conclusions: We found that occupational sedentariness was associated with increased breast cancer risk, especially in women younger than 55 years. This may be a modifiable risk factor by planning breaks during the working day. Whether this reduces the risk of breast cancer needs to be further studied.

Pancreas volume and fat fraction in children with Type 1 diabetes

People with Type 1 diabetes have smaller pancreases than healthy individuals. Several diseases causing pancreatic atrophy are associated with pancreatic steatosis, but pancreatic fat in Type 1 diabetes has not been measured. This cross‐sectional study aimed to compare pancreas size and fat fraction in children with Type 1 diabetes and controls.

Sample size re-assessment leading to a raised sample size does not inflate type I error rate under mild conditions

BackgroundOne major concern with adaptive designs, such as the sample size adjustable designs, has been the fear of inflating the type I error rate. In (Stat Med 23:1023-1038, 2004) it is however proven that when observations follow a normal distribution and the interim result show promise, meaning that the conditional power exceeds 50%, type I error rate is protected. This bound and the distributional assumptions may seem to impose undesirable restrictions on the use of these designs. In (Stat Med 30:3267-3284, 2011) the possibility of going below 50% is explored and a region that permits an increased sample size without inflation is defined in terms of the conditional power at the interim.MethodsA criterion which is implicit in (Stat Med 30:3267-3284, 2011) is derived by elementary methods and expressed in terms of the test statistic at the interim to simplify practical use. Mathematical and computational details concerning this criterion are exhibited.ResultsUnder very general conditions the type I error rate is preserved under sample size adjustable schemes that permit a raise. The main result states that for normally distributed observations raising the sample size when the result looks promising, where the definition of promising depends on the amount of knowledge gathered so far, guarantees the protection of the type I error rate. Also, in the many situations where the test statistic approximately follows a normal law, the deviation from the main result remains negligible. This article provides details regarding the Weibull and binomial distributions and indicates how one may approach these distributions within the current setting.ConclusionsThere is thus reason to consider such designs more often, since they offer a means of adjusting an important design feature at little or no cost in terms of error rate.

Impact of a paternal origin of germline BRCA1/2 mutations on the age at breast and ovarian cancer diagnosis in a Southern Swedish cohort.

Three studies have reported that BRCA1/2 mutations of paternal origin confer an earlier age at breast cancer diagnosis compared with maternal origin. The primary aim of this study was to investigate the impact of parental origin of BRCA1/2 mutations on age at breast and ovarian cancer diagnosis. This study included 577 female BRCA1/2 mutation carriers. All BRCA1/2 mutation carriers belonged to families registered between 1993 and 2011 at the Oncogenetic Clinic at Skånes University Hospital, Lund, Sweden. Cox proportional hazard ratios were used to analyze time to breast or ovarian cancer diagnosis. A novel finding was that carriers of BRCA1 mutations of paternal origin were 4 years older at age of ovarian cancer (P = 0.009) compared with those carrying a BRCA1 mutation of maternal origin. BRCA1 carriers with mutations of paternal origin were 4 years younger at breast cancer diagnosis (P = 0.017) compared with those carrying a BRCA1 mutation of maternal origin, which is in agreement with three previous studies. Both findings were adjusted for of year of inclusion, birth date, and oral contraceptive pill use. No associations between parental origin of BRCA2 mutations and time to breast or ovarian cancer diagnosis were found. An attempt to handle a potential selection bias regarding use of oral contraceptives was made using multiple imputations by chained equations. The observed age difference may allow a greater understanding of mechanisms associated with the differences in cancer penetrance in BRCA1/2 mutation carriers, some of which may depend on paternal origin.

Conditional estimation in two-stage adaptive designs

We consider conditional estimation in two-stage sample size adjustable designs and the consequent bias. More specifically, we consider a design which permits raising the sample size when interim results look rather promising, and which retains the originally planned sample size when results look very promising. The estimation procedures reported comprise the unconditional maximum likelihood, the conditionally unbiased Rao-Blackwell estimator, the conditional median unbiased estimator, and the conditional maximum likelihood with and without bias correction. We compare these estimators based on analytical results and a simulation study. We show how they can be applied in a real clinical trial setting.

Statin use and breast cancer survival: A Swedish nationwide study

Background A sizeable body of evidence suggests that statins can cease breast cancer progression and prevent breast cancer recurrence. The latest studies have, however, not been supportive of such clinically beneficial effects. These discrepancies may be explained by insufficient power. This considerably sized study investigates the association between both pre- and post-diagnostic statin use and breast cancer outcome. Methods A Swedish nation-wide retrospective cohort study of 20,559 Swedish women diagnosed with breast cancer (July 1st, 2005 through 2008). Dispensed statin medication was identified through the Swedish Prescription Registry. Breast cancer related death information was obtained from the national cause-of-death registry until December 31st, 2012. Cox regression models yielded hazard ratios (HR) and 95% confidence intervals (CI) regarding associations between statin use and breast cancer-specific and overall mortality. Results During follow-up, a total of 4,678 patients died, of which 2,669 were considered breast cancer related deaths. Compared to non- or irregular use, regular pre-diagnostic statin use was associated with lower risk of breast cancer related deaths (HR=0.77; 95% CI 0.63–0.95, P=0.014). Similarly, post-diagnostic statin use compared to non-use was associated with lower risk of breast cancer related deaths (HR=0.83; 95% CI 0.75–0.93, P=0.001). Conclusion This study evidently supports the notion that statin use is protective regarding breast cancer related mortality in agreement with previous Scandinavian studies, although less so with studies in other populations. These disparities should be further investigated to pave the way for future clinical trials investigating the role of statins in breast cancer.

Abstract P1-12-10: Tamoxifen treated patients have a better survival than patients treated with aromatase inhibitors - A population based registry study in Sweden

Background. Randomised trials suggest that therapy with aromatase inhibitors improves survival in breast cancer compared with tamoxifen therapy in postmenopausal cases with hormone receptor positive breast cancer. Whether these results from randomized studies transform into the general population is unknown. We have therefore compared survival for all breast cancer cases in Sweden diagnosed 2000-2008 (n=54406) who received adjuvant antihormonal therapy. Material and methods. The study includes all women with BC diagnosed in Sweden between 2000 through 2008 (n=54406). The women had no previous cancer diagnosis during the period of 1958-1999. Dates of birth, BC diagnosis and TNM-stage where directly extracted from the cancer registry. The womens antihormonal therapy was gathered from the Swedish Prescription Registry (22213 women were on antihormonal therapy). Information regarding the cause of death and date of death was obtained from the Cause of Death Registry and tbe Swedish Population Register up until the 31st of December 2012 and 31st of December 2013 respectively. The breast cancer death and overall death have been calculated and the survival was compared between tamoxifen and aromatase inhibitor treated breast cancer patients. Analyses were adjusted for TNM-stage and age at diagnosis and restricted to women aged 50 and above. Results. Patients being treated with tamoxifen had a better breast cancer prognosis compared with aromatase inhibitor treated patients (HR 0.54, 95%CI 0.48-0.61). Restricting the analysis to stage 1 disease confirmed a better prognosis for tamoxifen treated women (HR 0.48, 95%CI 0.34-0.66). A better prognosis could be seen in all age strata studies, 50-60.61-70.71-90. The findings for overall survival gave similar results. Conclusion .This population based observational study show that women treated with aromatase inhibitors have a worse overall and breast cancer specific survival compared with tamoxifen treated women regardless of age and tumor stage.

Abstract 865: Physical inactivity increases the risk of endometrial cancer and premenopausal breast cancer

Background. Epidemiological studies indicate that physical activity reduces the risk of cancer. Physical inactivity or sedentary behavior, has recently been suggested as a risk factor independent of physical activity level. Breast and endometrial cancer incidence have both been associated with physical activity, and endometrial cancer incidence with physical inactivity/sedentary behavior. The purpose of the present study was to investigate physical inactivity as a risk factor for breast cancer, divided into pre- and postmenopausal subtypes, and for endometrial cancer, and to explore possible dose-response relations regarding the level of physical inactivity. Methods. In a population-based prospective cohort study, 29 520 women in the southern part of Sweden, in ages between 25 and 64 years, participated. A questionnaire-based survey was performed 1990-92. Their reported professions were classified as sedentary or not which together with reported participation in competitive sports constituted the basis for classification into three activity-levels; (1) physical inactivity, defined as having sedentary occupation and no participation in competitive sports, (2) partly inactive, defined as either having a sedentary occupation or non- participation in competitive sports, (3) Physical active, defined as not having a sedentary occupation and participation in competitive sports. The association between physical inactivity and pre and postmenopausal breast and endometrial cancer incidence were analyzed by Cox regression, adjusted for hormonal factors, family history of cancer, body mass index (BMI) and age. Results. Physical inactive women had a significantly increased risk of endometrial cancer (HR = 2.41, 95% CI 1.14-5.11) and premenopausal breast cancer (HR = 2.40, 95% CI 1.14-5.02), compared with women who were less physical inactive. No such association was found for postmenopausal breast cancer. Analysis of linear trend showed a significant dose-response relationship with increased risk of both premenopausal breast cancer (ptrend 0.02) and endometrial cancer (ptrend