Biopsies of metastatic tissue are increasingly being performed. Bone is the most frequent site of metastasis in breast cancer patients, but bone remains technically challenging to biopsy. Difficulties with both tissue acquisition and techniques for analysis of hormone receptor status are well described. Bone biopsies can be carried out by either by standard posterior iliac crest bone marrow trephine/aspiration or CT-guided biopsy of a radiologically evident bone metastasis. The differential yield of these techniques is unknown. Results from three prospective studies of similar methodology were pooled. Patients underwent both an outpatient posterior iliac crest bone marrow trephine/aspiration and a CT-guided biopsy of a radiologically evident bone metastasis. Samples were assessed for the presence of malignant cells and where possible also for estrogen (ER) and progesterone receptor (PgR) expression. 40 patients were enrolled. Bone marrow aspiration/trephine biopsy was completed in 39/40 (97.5%) and CT-guided biopsy was completed in 34/40 (85%) of patients. Sufficient tumor cells for hormone receptor analysis were available in 19/39 (48.8%) and 16/34 (47%) of and bone marrow aspiration/trephine and CT-guided biopsies, respectively. Significant discordance in ER and PgR between the primary and the bone metastasis was also seen. Nine patients had tissue available from both bone marrow and CT-guided bone biopsies. ER and PgR concordance between these sites was 100 and 78%, respectively. Performing studies on human bone metastases is technically challenging, with relatively low yields regardless of technique. Given resource issues and similar success rates when comparing both techniques, bone marrow examination may be utilized first and if inadequate tissue is obtained, CT-guided biopsies can then be used.

The onset of osseous metastases during the course of colorectal cancer is not common. Although rare, they usually appear in the axial skeleton. In our report, we refer to the case of a 48-year-old patient who presented with colon cancer and eventually developed a solitary bone metastasis in the upper end of left tibia. At the time of diagnosis and staging investigations, the patient had only a primary disease.

Summary Two patients with dermatomyositis presented with violaceous lichenoid eruptions on the light exposed areas. Histology and direct immunofluorescence showed the typical features of lichen planus. The rash in dermatomyositis usually consists of a purplish‐red heliotrope erythema on the face, erythematous changes on the hands and backs of the fingers with diffuse redness and shining of the nail folds (Rowell, 1979). A pathognomonic sign in dermatomyositis is the Gottron papule, a violaceous flat‐topped lesion over the dorsal interphalangeal joints and occasionally a diffuse lichenoid eruption occurs which may be mistaken for lichen planus (Braverman, 1983). However, there have been no previous reports of the histology and immunofluorescence findings in these patients. Two cases are reported showing clinically a violaceous lichenoid eruption on the dorsal aspect of both hands with the histology and direct immunofluorescence typical of lichen planus. The association of lichen planus and dermatomyositis is discussed.

e14154 Background: The timing of chemotherapy (adjuvant vs neoadjuvant) for patients undergoing potentially curative hepatic resection of metastasis from colorectal cancer origin is controversial. Methods: We performed a retrospective review of all patients at TOHCC who underwent successful surgical resection of hepatic metastasis from CRC between 2002-2009. Patients receiving intra-operative radiofrequency ablation (RFA) as part of their management were included. Factors associated with overall survival were evaluated. Results: 168 patients had median age 62 years (31-84), 57/43% male/female. Median pre-op CEA was 5.5 ng/ml (0.3 - 529.2). Primary tumor location included 40% left, 35% rectum, 20% right, 5% transverse. 61.5% patients were node positive. Following hepatectomy 10.2% had positive resection margins. Intra-operative RFA was used in 26 (15.4%) patients. Chemotherapy was administered in a neo-adjuvant (47.6%), adjuvant (58.4%) or “perioperative” (both neoadjuvant and adjuvant; 36.1% ) setting. Use of intraoperative RFA vs. no RFA was not associated with overall survival (HR: 0.94 [95%CI: 0.49-1.80],p=0.85). In both univariate and multivariate analysis, the only adverse factor found associated with survival was R1 status (HR 2.38 [95% CI 1.49-5.0], p=0.0198). Other baseline variables (age, gender, node status, histologic grade, primary tumour site, timing of metastatic diagnosis [metachronous vs. synchronous]) did not reach statistical significance. Conclusions: While statistical significance was not achieved, ability to demonstrate benefit of chemotherapy may have been limited by sample size. Hazard ratios suggest trend to greatest benefit with use of post-operative adjuvant chemotherapy. 3y and 5y survival data is encouraging. Use of RFA where required as an adjunct to hepatic resection appeared as effective as resection of all disease in this series. Recurrence and morbidity data will be presented. [Table: see text]