Whither surgical quality assurance of breast cancer surgery (surgical margins and local recurrence) after paterson.

PURPOSE: The Kennedy report into the actions of the disgraced Breast Surgeon, Paterson focussed on issues of informed consent for mastectomy, management of surgical margins and raised concerns about local recurrence rates and the increasing emphasis on cosmesis after mastectomy for breast cancer. This article assesses whether Kennedy's recommendations apply to the UK as a whole and how to address these issues. New GMC advice on consent and newer nonevidenced innovations in immediate reconstruction have altered the level of informed consent required. Patients deserve a better understanding of the issues of oncological versus cosmetic outcomes on which to base their decisions. Involvement of the whole multidisciplinary team including Oncologists is necessary in surgical planning. Failure to obtain clear microscopic margins at mastectomy leads to an increased local recurrence, yet has received little attention in the UK. Whereas, other countries have used surgical quality assurance audits to reduce local recurrence; local recurrence rates are not available and the extent of variation across the UK in margin involvement after surgery, its management and relationship to local recurrence needs auditing prospectively to reduce unnecessary morbidity. To reassure public, patients and the NHS management, an accreditation system with more rigour than NHSBSP QA and peer review is now required. Resource and efforts to support its introduction will be necessary from the Royal College of Surgeons and the Association of Breast Surgeons. New innovations require careful evaluation before their backdoor introduction to the NHS. Private Hospitals need to have the same standards imposed.

Impact of Oncotype DX breast Recurrence Score testing on adjuvant chemotherapy use in early breast cancer: Real world experience in Greater Manchester, UK.

BACKGROUND: The National Institute for Health and Clinical Excellence (NICE) recommended the Oncotype DX® Breast Recurrence Score® (RS) assay as an option for informing adjuvant chemotherapy decisions in node-negative, oestrogen receptor (ER)+, human epidermal growth factor receptor 2 (HER2)-negative early breast cancer assessed to be at intermediate risk of recurrence based on clinicopathological factors. We evaluated the impact of RS testing on adjuvant chemotherapy decision-making in routine clinical practice in a UK Cancer Network. METHODS: RS testing was performed in 201 females with newly diagnosed, ER+, HER2-negative, invasive breast cancer who underwent breast surgery with curative intent, were calculated to have a >3% overall survival benefit at 10 years from adjuvant chemotherapy based on PREDICT, and were considered for adjuvant chemotherapy. The impact of RS testing on adjuvant treatment decisions/associated cost was assessed. RESULTS: In all patients, the multi-disciplinary team recommended chemotherapy but the RS result allowed 127/201 patients (63.2%) to avoid unnecessary adjuvant chemotherapy. Amongst ER+, HER2-negative, node-negative patients (eligible for Oncotype DX testing in UK guidelines), 60.3% were spared chemotherapy. In node-positive patients, the assay reduced the use of chemotherapy by 69.2%. The use of RS testing to guide treatment in these 201 patients was associated with significant cost saving (when considering the cost of RS testing for all patients plus chemotherapy and its associated cost for 74 patients). CONCLUSIONS: Incorporating RS testing into routine clinical practice for selected node-negative and node-positive breast cancer patients significantly reduces the use of chemotherapy (p < 0.001) with its associated morbidity and costs.

Prediction of margin involvement and local recurrence after skin-sparing and simple mastectomy.

UNLABELLED: Skin-sparing mastectomy (SSM) facilitates immediate breast reconstruction. We investigated locoregional recurrence rates after SSM compared with simple mastectomy and the factors predicting oncological failure. METHODS: Patients with early breast cancer that underwent mastectomy between 2000 and 2005 at a single institution were studied to ascertain local and systemic recurrence rates between groups. Kaplan-Meier curves and log-rank test were used to evaluate disease-free survival. RESULTS: Patients (n = 577) underwent simple mastectomy (80%) or SSM (20%). Median follow up was 80 months. Patients undergoing SSM were of younger average age, less often had involved lymph nodes (22% vs 44%, p < 0.001), more often had DCIS present (79% vs 53%, p < 0.001) and involved margins (29% vs 15%, p = 0.001). Involved surgical margins were associated with large size (p = 0.001). The 8-year local recurrence (LR) rates were 7.9% for SSM and 5% for simple mastectomy respectively (p = 0.35). Predictors of locoregional recurrence were lymph node involvement (HR 8.0, for >4 nodes, p < 0.001) and involved surgical margins (HR 3.3, p = 0.002). In node negative patients, SSM was a predictor of locoregional recurrence (HR 4.8 [1.1, 19.9], p = 0.033). CONCLUSION(S): Delayed reconstruction is more appropriate for node positive early breast cancer after post-mastectomy radiotherapy. Re-excision of involved margins is essential to prevent local recurrence after mastectomy.

PO-21 - Stromal fibroblasts in preinvasive breast cancer (ductal carcinoma in situ, DCIS) demonstrate a cancer-like procoagulant phenotypic switch that may facilitate invasion.

INTRODUCTION: Ductal carcinoma in-situ (DCIS) is a preinvasive breast cancer where cancer cells remain confined within the ductal basement membrane. However, genotypic changes have been identified in stroma surrounding DCIS, outside the basement membrane. Stromal fibroblasts undergo phenotypic change in cancer to promote tumour angiogenesis, proliferation, immunosuppression and metastasis and in vivo can induce invasion of DCIS. Phenotypic changes in DCIS stromal fibroblasts may potentially act as a precursor for invasion. AIM: To determine if stromal fibroblasts in DCIS have procoagulant changes similar to those seen in cancer-associated fibroblasts in invasive breast cancer. MATERIALS AND METHODS: As part of the prospective cohort study CHAMPion (Cancer induced Hypercoagulabulity as a Marker of Prognosis), patients with DCIS (n=72) and invasive breast cancer (n=292) were recruited. Stromal fibroblasts in tumour and corresponding normal breast tissue (distant from the cancer) were quantified (percentage IHC stained) for tissue factor (TF), thrombin, PAR1 and PAR2. Fibroblasts were identified morphologically, at a minimum distance of 0.2mm from ductal tissue, to avoid myoepithelial scoring. Scoring was performed in duplicate by two independent pathologists. RESULTS: Fibroblast TF expression was present in normal breast tissue (mean 43% ([SD 27%]) but markedly increased in DCIS (mean 62% [SD 27%], p=0.002). Fibroblast TF expression was further increased in invasive breast cancer (mean 74% [SD 23%], normal vs invasion, p<0.001; DCIS vs invasion, p=0.03). Fibroblast thrombin and PAR2, but not PAR1, expression was increased in DCIS compared to normal (thrombin: 60% vs 42%, p<0.001; PAR2: 58% vs 41%, p=0.002), however no further significant increase was seen in invasive cancer (thrombin 63%, PAR2 61%). Fibroblast tissue factor correlated with fibroblast thrombin expression (p<0.001, r=0.4) and fibroblast PAR2 expression (p<0.001, r=0.5), with thrombin and PAR2 expression also correlating (p<0.001, r=0.4). CONCLUSIONS: Procoagulant phenotypic changes, in terms of increased TF, thrombin and PAR2 expression, occur in stromal fibroblasts at the preinvasive stage. It needs to be determined if this change is functional and therefore a potential therapeutic target for preventing transition to invasion.

PO-36 - Thrombin Inhibition Preoperatively (TIP) in early breast cancer, the first clinical trial of NOACs as an anti-cancer agent: trial methodology.

INTRODUCTION: Breast cancer is associated with a 3-4 fold increased risk of VTE. These patients have a 4-fold lower survival than those remaining free of VTE, implying VTE is a surrogate marker for aggressive cancer. Tumour expression of thrombin pathway markers are increased in the oestrogen receptor negative (ER-), high Ki67, more aggressive breast cancer subtypes. In in vitro and in vivo studies, the thrombin pathway promotes cancer growth and metastases, highlighting the potential role of the thrombin pathway as a therapeutic target in cancer. AIM: To determine whether 14days of a preoperative oral Factor Xa inhibitor (Rivaroxaban) in oestrogen receptor negative early breast cancer patients results in inhibition of tumour proliferation as determined by a reduction in tumour Ki67 from baseline (pre-treatment) to 14days post treatment start (at time of surgical excision). RATIONALE: The TF-VIIa-FXa complex activates Protease Activated Receptor (PAR)2 to induce angiogensis, growth factors and tumour cell migration. Thrombin, in part via PAR1, induces angiogenesis, tumour cell proliferation as well as in vivo metastasis. In early breast cancer, TF and PAR2 expression is increased in the stroma, particularly in the more aggressive ER-, high Ki67 (proliferation) cancers. Rivaroxaban is an orally active direct Factor Xa inhibitor. Through inhibition of the TF-FVIIa-FXa complex, it can downregulate TF-FVIIa-FXa activation of PAR2, and inhibit conversion of prothrombin to thrombin. We hypothesise that 14days of Rivaroxaban will reduce breast cancer proliferation, as a surrogate marker for anti-cancer efficacy, in early breast cancer patients awaiting resectional surgery. RESULTS: Trial methodology: A multi-centre phase II preoperative 'Window-of Opportunity' randomised controlled trial of Rivaroxaban compared to no treatment in ER-, stage I-III early breast cancer patients. Patients will be randomised 1:1:1 (Rivaroxaban 20mg od: Rivaroxaban 10mg od: no treatment) and receive 14 (+/-3) days of treatment in the window between diagnosis and surgery. Randomisation will be blinded to pathologists, but not to patients or clinicians. Primary analysis will be based on the two Rivaroxaban arms being combined to form a Rivaroxaban: no treatment, 2:1 trial design, with change in Ki67 from baseline (pre) to post Rivaroxaban/no treatment (post) being the primary endpoint, and the no treatment arm acting as a reference group. Subgroup analysis of the Rivaroxaban arm (20mg od:10mg od) will allow assessment of dose-response. ACKNOWLEDGEMENT: Funder: National Institute for Health Research Eudract No: 2014-004909-33 REC Number: 15/NW/0406 UKCRN ID: 19731 Expected commencement: January 2016. For further information please contact Chief Investigator: cliona.kirwan@manchester.ac.uk.

A randomised pilot study comparing 13 G vacuum-assisted biopsy and conventional 14 G core needle biopsy of axillary lymph nodes in women with breast cancer.

AIM: To compare the acceptability, safety, and feasibility of vacuum-assisted biopsy (VAB) and core needle biopsy (CNB) of axillary lymph nodes in women with breast cancer. MATERIALS AND METHODS: This parallel, non-blinded, randomised study was approved by the National Research Ethics Service. Following written informed consent, women with abnormal appearing axillary lymph nodes and radiologically malignant breast masses were randomised 1:1 to lymph node sampling under local anaesthetic with either 14 G CNB or 13 G VAB in a single UK centre. Primary outcomes were study uptake rate and patient willingness to undergo a repeat procedure if necessary. Procedure duration, immediate and post-procedure pain scores, diagnostic yield, complications, and surgical histopathology were recorded. RESULTS: Ninety-five women were approached; 81 (85.3%) consented and were randomised. Forty underwent CNB; 40 underwent VAB; one was excluded. Median age was 57 years. The median procedure time was 2 minutes for both techniques. The median number of samples obtained was three for CNB and four for VAB. Median pain scores for the procedure and first 3 days were 1/10 and 1/10 for CNB and 1/10 and 2/10 for VAB (p=0.11 and 0.04). More women were prepared to undergo repeat CNB compared to VAB, but the difference was not significant (38/39 versus 33/39; p=0.11). Two patients developed a haematoma after VAB. One CNB and six VABs failed to yield adequate tissue (p=0.11), but the sensitivity was similar at 79% and 78%. CONCLUSION: Study uptake was high. Acceptability of the two procedures was similar, but VAB was associated with more post-procedure pain. The sensitivity appears to be similar.

Molecular phenotypes of DCIS predict overall and invasive recurrence.

BACKGROUND: Molecular phenotypes of invasive breast cancer predict early recurrence. Ductal carcinoma in situ (DCIS) exhibits similar phenotypes, but their frequency and significance remain unclear. To determine whether DCIS molecular phenotypes predict recurrence, 314 women (median age 57.7 years) with primary DCIS who were screened or entered DCIS trials in a specialist breast unit from 1990 to 2010 were studied. PATIENTS AND METHODS: Expression of Ki67, estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) within primary DCIS was established using immunohistochemistry (IHC). Patients were subdivided into molecular phenotypes using IHC surrogates [Luminal A (ER/PR+HER2-), Luminal B (ER/PR+/HER2+), HER2 type (ER and PR-/HER2+) or triple negative (ER/PR/HER2)] and recurrence rates compared. RESULTS: Overall, there were 57 (18.2%) recurrences, 35 (11.2%) DCIS and 22 (7%) invasive cancer. A low rate of recurrence at 5 years was seen in Luminal A DCIS (7.6%), compared with 15.8%-36.1% in other phenotypes. Independent predictors of overall recurrence on multivariate analysis were involved (<1 mm) surgical margins (HR 4.31, P < 0.001), high-grade lesions (HR 2.28, P < 0.024) and molecular phenotype (HR 5.14, P = 0.001 for Luminal B; HR 6.46, P < 0.001 for HER2 type and HR 3.27, P = 0.028 for triple-negative disease compared with Luminal A DCIS). Independent predictors for invasive recurrence were high Ki67 expression (HR 1.04, P = 0.021) and molecular phenotype (HR 13.4, P = 0.014 for Luminal B; HR 11.4, P = 0.027 for HER2 type and HR 10.3, P = 0.031 for triple negative compared with Luminal A DCIS). CONCLUSIONS: DCIS molecular phenotype predicts for both overall and invasive recurrence. HER2 testing of DCIS could help clinicians individualise the treatment of patients with DCIS.

The frequency of presentation and clinico-pathological characteristics of symptomatic versus screen detected ductal carcinoma in situ of the breast.

INTRODUCTION: DCIS accounts for 20% of screen-detected breast cancers, but also presents symptomatically. Historically, approximately 5% of DCIS was thought to be symptomatic, but accurate evaluation of the presentation of symptomatic DCIS is needed to determine its incidence and tumour biology. METHODS: Clinico-pathological details of a consecutive series of patients presenting to a single breast-unit, with a pre-operative diagnosis of DCIS, were selected. Data included age, mode of presentation, pre-operative clinical and radiographical findings. The final tumour histology, operation, size, grade, ER status (and HER2 expression in invasive cases) were recorded. RESULTS: 375 patients had a pre-operative histological diagnosis of DCIS. 308 (82%) screen-detected (median age 59), 67 (18%) presented via symptomatic clinics (median age 50). At final histology 286 (74%) were pure DCIS, and 67 (23%) had an invasive focus. 43% (29/67) of symptomatic cases had an invasive focus at final histology versus 19% (60/308) screen-detected (p ≤ 0.001). 31% (9/29) of symptomatic, versus 10% (6/60) of screen-detected cases with invasion were node positive (p = 0.05). 45% (28/62) intermediate/high-grade symptomatic cases had an invasive focus at final histology, compared to 19% (57/297) intermediate/high-grade screen-detected cases. 86% (212/248) screen-detected pure DCIS was ER positive compared to 68% (26/38) symptomatically presenting pure DCIS (p ≤ 0.001). Overall, 13% (38/248) pure DCIS presented symptomatically (p = 0.001). CONCLUSIONS: Overall, thirteen percent of pure DCIS present symptomatically. Nearly half of symptomatically presenting DCIS at core biopsy has an occult invasive focus and is more frequently ER negative. Symptomatic DCIS with an invasive focus is more likely to have lymph node involvement.

Cancer treatment-induced bone loss in premenopausal women: a need for therapeutic intervention?

Current clinical treatment guidelines recommend cytotoxic chemotherapy, endocrine therapy, or both (with targeted therapy if indicated) for premenopausal women with early-stage breast cancer, depending on the biologic characteristics of the primary tumor. Some of these therapies can induce premature menopause or are specifically designed to suppress ovarian function and reduce circulating estrogen levels. In addition to bone loss associated with low estrogen levels, cytotoxic chemotherapy may have a direct negative effect on bone metabolism. As a result, cancer treatment-induced bone loss poses a significant threat to bone health in premenopausal women with breast cancer. Clinical trials of antiresorptive therapies, such as bisphosphonates, have demonstrated the ability to slow or prevent bone loss in this setting. Current fracture risk assessment tools are based on data from healthy postmenopausal women and do not adequately address the risks associated with breast cancer therapy, especially in younger premenopausal women. We therefore recommend that all premenopausal women with breast cancer be informed about the potential risk of bone loss prior to beginning anticancer therapy. Women who experience amenorrhea should have bone mineral density assessed by dual-energy X-ray absorptiometry and receive regular follow-up to monitor bone health. Regular exercise and daily calcium and vitamin D supplementation are recommended. Women with a Z-score <-2.0 or Z-score ≤-1.0 and/or a 5-10% annual decrease in bone mineral density should be considered for bisphosphonate therapy in addition to calcium and vitamin D supplements.

Management of aromatase inhibitor-associated bone loss in postmenopausal women with breast cancer: practical guidance for prevention and treatment.

BACKGROUND: Bone mineral density (BMD)-based guidelines for bone-directed therapy in women with early breast cancer (EBC) appear inadequate for averting fractures, particularly during aromatase inhibitor (AI) therapy. Therefore, an algorithm was developed to better assess risk and direct treatment (Hadji P, Body JJ, Aapro MS et al. Practical guidance for the management of aromatase inhibitor-associated bone loss. Ann Oncol 2008; 19: 1407-1416). Here, we provide updated guidance on pharmacologic interventions to prevent/treat aromatase inhibitor-associated bone loss (AIBL). DESIGN: Systematic literature review identified recent advances in preventing/treating AIBL. Individual agents were assessed based on trial size, design, follow-up, and safety. RESULTS: Fracture risk factors in patients with EBC remain unchanged (Hadji P, Body JJ, Aapro MS et al. Practical guidance for the management of aromatase inhibitor-associated bone loss. Ann Oncol 2008; 19: 1407-1416). The World Health Organization Fracture Risk Assessment Tool algorithm includes fracture risk factors plus BMD but does not adequately address AIBL effects. Several antiresorptives can prevent/treat AIBL. However, concerns regarding compliance and long-term efficacy/safety remain. Overall, evidence is strongest for twice-yearly zoledronic acid (ZOL), and recent advances support additional anticancer benefits from ZOL. CONCLUSIONS: All patients initiating AIs need advice regarding exercise, calcium/vitamin D supplements, baseline BMD monitoring (when available), and bone-directed therapy if T-score <-2.0 or at least two fracture risk factors were observed. Patients with T-score > -2.0 and no risk factors should be managed based on BMD loss during years 1-2. Unsatisfactory compliance/decreasing BMD after 12-24 months on oral bisphosphonates should trigger a switch to i.v. bisphosphonate.

Safety of tibolone in the treatment of vasomotor symptoms in breast cancer patients--design and baseline data 'LIBERATE' trial.

Many patients with a history of breast cancer (BC) will suffer from vasomotor symptoms, which can be induced or exacerbated by treatment with tamoxifen or aromatase inhibitors. The LIBERATE trial was designed as a randomized, double-blind, multicenter trial to demonstrate that tibolone 2.5mg/day (Livial) is non-inferior to placebo regarding BC recurrence in women with vasomotor symptoms surgically treated for primary BC within the last 5 years. Secondary objectives are effects on vasomotor symptoms as well as overall survival, bone mineral density and health-related quality of life. Mean age at randomization was 52.6 years, and the mean time since surgery was 2.1 years. The mean daily number of hot flushes and sweating episodes was 7.3 and 6.1, respectively. For the primary tumor, Stage IIA or higher was reported for >70% of the patients. In subjects whose receptor status was known, 78.2% of the tumors were estrogen receptors positive. At randomization, tamoxifen was given to 66.2% of all patients and aromatase inhibitors to 7%. Chemotherapy was reported by 5% at randomization. The adjuvant tamoxifen use in LIBERATE allows a comparison with the Stockholm trial (showing no risk of BC recurrence associated with hormone therapy), which was stopped prematurely subsequent to HABITS. The LIBERATE trial is the largest, ongoing, well-controlled study for treatment of vasomotor symptoms in BC patients.

Serum vascular endothelial growth factor in breast cancer.

Many studies have reported elevated serum concentrations of vascular endothelial growth factor (VEGF) in patients with cancer and claimed that the measurement of circulating VEGF is a surrogate marker of angiogenesis and/or metastasis. To determine the value of VEGF measurement in the diagnosis and prognosis of breast cancer, we measured levels in women with and without breast cancer. Platelet-depleted plasma VEGF levels were measured in premenopausal women at four-day intervals across the menstrual cycle, postmenopausal women and postmenopausal women who had undergone hysterectomy. Platelet-depleted plasma VEGF was also measured in pre- and postmenopausal women with early breast cancer (EBC) and levels compared with intratumoral levels, clinicopathological prognostic parameters and microvessel density. Levels of VEGF were determined using ELISA and immuno-histochemistry. Microvessel density was determined by immunohistochemical CD34 staining. Plasma VEGF in premenopausal women remained stable across the menstrual cycle except for a peak between days 8 and 12. VEGF levels in postmenopausal women were higher than in premenopausal women unless postmenopausal women had undergone hysterectomy. Amongst premenopausal women, levels of VEGF were high in 22 EBC patients when compared to normal premenopausal controls. No correlation was found between plasma and intratumoral VEGF, clinicopathological prognostic parameters or tumour microvessel density. The origin of circulating VEGF differs between pre- and postmenopausal women. Its measurement is unlikely to provide clinically useful diagnostic and prognostic information in women with early and advanced breast cancer.

Angiogenic characteristics of circulating and tumoural thrombospondin-1 in breast cancer.

In cancer models, thrombospondin-1 (TSP-1) has been shown to inhibit angiogenesis or promote metastasis by increasing adhesion of malignant cells to endothelium. To determine the role of TSP-1 in breast cancer and breast cancer angiogenesis, we have measured TSP-1 in plasma and tumour cytosols and compared levels to established clinicopathological prognostic parameters and intratumoural microvessel density. TSP-1 was measured, by radioimmunoassay, in plasma (pTSP-1) and tumour cytosols (cTSP-1) of women with early breast cancer (EBC) (n=71). pTSP-1 in EBC was compared to pTSP-1 levels in women with advanced breast cancer (ABC) (n=66), normal controls (n=77) and was correlated with prognostic features and microvessel density (MVD) (measured by CD31 immunostaining). cTSP-1 levels were compared to prognostic features and microvessel density. pTSP-1 in women with EBC (median 484, IQR 344-877 ng/ml) and ABC (median 588, IQR 430-952 ng/ml) were elevated when compared to normal controls (median 21, IQR 175-247) (p<0.001). Women with lymph node metastases (n=35) had higher levels of TSP-1 (median 799 ng/ml, IQR 455-943) than women who were node negative (median 343 ng/ml, IQR 267-514) (n=36) (p<0.05). Levels of pTSP-1 in EBC correlated with MVD (R=0.39, p<0.05). Levels of TSP-1 in tumour cytosols of women with EBC (median 1714, IQR 893-5283 ng/ml) correlated with microvessel density (R=0.46, p<0.01). Circulating levels of TSP-1 appear to be a marker of breast cancer aggressiveness and in breast cancer may have a pro-angiogenic rather than anti-angiogenic role.

Prevention of venous thromboembolism in surgical patients with breast cancer.

BACKGROUND: No randomized trial has yet studied venous thromboembolism (VTE) prophylaxis in patients undergoing surgery for breast cancer. METHODS: Relevant articles were identified using Medline searches. Secondary articles were identified from the reference lists of key papers. RESULTS AND CONCLUSION: The absence of randomized trials comparing different methods of VTE prophylaxis with controls makes an evidence-based consensus among breast cancer surgeons difficult. Intermittent pneumatic compression (IPC) and graduated compression (GC) are effective in reducing VTE without the haemorrhagic complications associated with heparin; their effects are additive. The authors suggest the following strategy. All patients undergoing surgery for breast cancer should receive both IPC and GC, with heparin reserved for those at very high risk. A controlled trial should randomize women to receive heparin or not, and all women should have both IPC and GC. The primary endpoints should be the development of VTE and/or haemorrhagic complications.

Cyclooxygenase-2 inhibition: effects on tumour growth, cell cycling and lymphangiogenesis in a xenograft model of breast cancer.

Cyclooxygenase-2 (COX-2) is associated with poor-prognosis breast cancer. We used a nude mouse xenograft model to determine the effects of COX-2 inhibition in breast cancer. Oestrogen receptor (ER)-positive MCF7/HER2-18 and ER-negative MDAMB231 breast cancer cell lines were injected into nude mice and allowed to form tumours. Mice then received either chow containing Celecoxib (a COX-2 inhibitor) or control and tumour growth measured. Tumour proliferation, apoptosis, COX-2, lymphangiogenesis and angiogenesis were assessed by immunohistochemistry (IHC), Western blotting or Q-PCR. Celecoxib inhibited median tumour growth in MCF7/HER2-18 (58.7%, P=0.029) and MDAMB231 (46.3%, P=0.0002) cell lines compared to control. Cyclooxygenase-2 expression decreased following Celecoxib treatment (MCF7/HER2-18 median control 65.3% vs treated 22.5%, P=0.0001). Celecoxib increased apoptosis in MCF7/HER2-18 tumours (TUNEL 0.52% control vs 0.73% treated, P=0.0004) via inactivation of AKT (median pAKT(ser473) 57.3% control vs 35.5% treated, P=0.0001--confirmed at Western blotting). Q-PCR demonstrated decreased podoplanin RNA (lymphangiogenesis marker) in the MCF7/HER2-18 - median 2.9 copies treated vs 66.6 control (P=0.05) and MDAMB231-treated groups--median 160.7 copies vs 0.05 control copies (P=0.015), confirmed at IHC. Cyclooxygenase-2 is associated with high levels of activated AKT(ser473) and lymphangiogenesis in breast cancer. Cyclooxygenase-2 inhibition decreases tumour growth, and may potentially decrease recurrence, by inactivating AKT and decreasing lymphangiogenesis.

Activated c-SRC in ductal carcinoma in situ correlates with high tumour grade, high proliferation and HER2 positivity.

Overexpression and/or activity of c-Src non-receptor tyrosine kinase is associated with progression of several human epithelial cancers including breast cancer. c-Src activity in 'pure' ductal carcinoma in situ (DCIS) was measured to assess whether this predicts recurrence and/or correlates with HER2 expression and other clinical parameters. Activated c-Src levels were evaluated in DCIS biopsies from 129 women, with median follow-up at 60 months. High levels of activated c-Src correlated with HER2 positivity, high tumour grade, comedo necrosis and elevated epithelial proliferation. In univariate analysis, high activated c-Src level associated with lower recurrence-free survival at 5 years (P=0.011). Thus, high c-Src activity may identify a subset of DCIS with high risk of recurrence or progression to invasive cancer where therapeutics targeting c-Src may benefit this patient subset.