A canine model is described to study the tolerance of the pancreas to intra-operative radiotherapy (IORT). The canine pancreas is a horseshoe-shaped organ. To create a homogeneous delivery of IORT to the whole pancreas surgical manipulation is necessary which may induce pancreatitis. A resection of the left and right lobes of the pancreas facilitates the delivery of IORT, reduces the risk of pancreatitis and will demonstrate, eventually, minimal functional changes in the exocrine and endocrine pancreas at an earlier stage. Sixteen beagles were used. Investigations before and after the reduction procedure were intravenous glucose tolerance tests, serum insulin levels, faecal fat excretion, blood chemistry tests and body weight. Eight weeks after the pancreas reduction 15 dogs underwent an IORT procedure in which 25, 30 or 35 Gy IORT was delivered to the pancreatic remnant. We conclude that the pancreas reduction technique used to study the effects of IORT to the canine pancreas is feasible without mortality or morbidity. Endocrine and exocrine pancreatic function remained normal with a minimal follow-up of 3 months.

Background: Triple-negative cancer constitutes one of the most challenging groups of breast cancer given its aggressive clinical behaviour, poor outcome and lack of targeted therapy. Until now, profiling techniques have not been able to distinguish between patients with good and poor outcome. Recent studies suggested an important role for stroma in tumour growth and progression. In colorectal-, oesophageal- and breast cancer, the tumour-stroma ratio was found to be of prognostic value. Objective: to evaluate the prognostic value of the tumour-stroma ratio in triple-negative breast cancer. Methods: During the period January 2004–2008, 124 consecutive triple negative breast cancer patients treated in our hospital were retrospective evaluated. Routine Haematoxylin-Eosin (H&E) stained histological sections were evaluated by two investigators (kappa 0.735) for stroma percentage, growth pattern (pushing margin), necrosis and amount of lymphocytic infiltrate. Patients with less than 50% stroma were classified as stroma-low and patients with ≥ 50% stroma were classified as stroma-high. Results: of 124 triple-negative breast cancer patients, 50 (40%) had a stroma-high and 74 (60%) had a stroma-low tumour. Survival analysis revealed a 5 years relapse free period (RFP) of 85% in the stroma-low and 45% in the stroma-high group. Overall survival (OS) was 89% for stroma-low and 65% in the patients with a stroma-high tumour. Both RFP and OS were significantly worse in patients with stroma-high tumours compared to stroma-low. In a multivariate cox-regression analysis, tumour stroma remained an independent prognostic variable for RFP (HR 2.39; 95% CI 1.07−5.29; p=0.033) and OS (HR 3.00; 95% CI 1.08−8.32; 0.034) when corrected for other clinical-pathological variables. In addition, tumour-stroma proved to be a strong prognostic variable when compared to nodal status, tumour grade and tumour size, with respectively a HR of 2.39 (95% CI 1.37−6.26) versus HR 2.18 (95% CI 1.06−4.48), HR 2.10 (95% CI 0.89−4.91) and HR 0.53 (95% CI 0.70−3.91) for RFP with comparable numbers for OS. Conclusion: Tumour-stroma ratio is a strong independent prognostic variable in triple-negative breast cancer. It is easy to determine, reproducible (kappa 0.735) and can be easily incorporated into routine histological examination. This parameter optimizes risk stratification and could be target for future therapies. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-09-06.

Background: The sentinel lymph node biopsy (SLNB) procedure is the method of choice for the identification and monitoring of regional lymph node metastases in patients with breast cancer. In the case of a positive sentinel node additional lymph node dissection is still warranted for regional control, though 40-65% have no additional axillary disease. Recent studies showed that after breast-conserving surgery, SLNB and adjuvant systemic therapy there is no significant difference between recurrence-free interval and overall survival if there are 2 or less positive axillary nodes. Purpose: Preoperative identification of patients with limited axillary disease (≤ 2 macrometastases) using ultrasonography (US). Method: Between January 2007 and August 2011, data from 1103 consecutive primary breast cancer patients who underwent surgery in our hospital were collected into a single database. Patients were selected by clinical tumours smaller than 50mm (cT1-2), no palpable adenopathy (cN0) and a maximum of 2 SLNs containing macrometastases. The variable of interest was ultrasonography of the axilla. The population was divided into two groups: the group with 2 or fewer positive nodes and the group with more than 2 positive nodes in the axilla after ALND. Results: After selection, 1060 patients remained of which 102 (9.6%) had more than 2 positive axillary nodes on SLNB or ALND. Selected by unsuspected US, the chance of having more than 2 positive lymph nodes is substantially lower (4.2%). This is significant on univariate and multivariate analysis. The chance of more than 2 positive lymph nodes was 12.8 times as big in case of a positive axillary US. When we select this subgroup even further by excluding the patients with extra capsular extension (ECE) on SLN, the chance of more than 2 positive lymph nodes is only 2.6%. Subdivided by clinical tumour size the chances are 0.96% in case of cT1 tumour and 7.0% in case of a cT2 tumour. For pathological T-status the numbers are respectively 0.87% en 5.0%. For pT1-2 this is 2.2%. Conclusion: The risk of more than 2 positive axillary nodes is relatively small in patients with cT1-2 breast cancer. Ultrasonography of the axilla helps in further identifying patients with a minimal risk of additional axillary disease, putting ALND up for discussion. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-01-07.