Search

Purpose The diagnostic test for ALK rearrangement in non–small-cell lung cancer (NSCLC) for crizotinib treatment is currently done on tumor biopsies or fine-needle aspirations. We evaluated whether ALK rearrangement diagnosis could be performed by using circulating tumor cells (CTCs). Patients and Methods The presence of an ALK rearrangement was examined in CTCs of 18 ALK-positive and 14 ALK-negative patients by using a filtration enrichment technique and filter-adapted fluorescent in situ hybridization (FA-FISH), a FISH method optimized for filters. ALK-rearrangement patterns were determined in CTCs and compared with those present in tumor biopsies. ALK-rearranged CTCs and tumor specimens were characterized for epithelial (cytokeratins, E-cadherin) and mesenchymal (vimentin, N-cadherin) marker expression. ALK-rearranged CTCs were monitored in five patients treated with crizotinib. Results All ALK-positive patients had four or more ALK-rearranged CTCs per 1 mL of blood (median, nine CTCs per 1 mL; range, four to 34 CTCs per 1 mL). No or only one ALK-rearranged CTC (median, one per 1 mL; range, zero to one per 1 mL) was detected in ALK-negative patients. ALK-rearranged CTCs harbored a unique (3′5′) split pattern, and heterogeneous patterns (3′5′, only 3′) of splits were present in tumors. ALK-rearranged CTCs expressed a mesenchymal phenotype contrasting with heterogeneous epithelial and mesenchymal marker expressions in tumors. Variations in ALK-rearranged CTC levels were detected in patients being treated with crizotinib. Conclusion ALK rearrangement can be detected in CTCs of patients with ALK-positive NSCLC by using a filtration technique and FA-FISH, enabling both diagnostic testing and monitoring of crizotinib treatment. Our results suggest that CTCs harboring a unique ALK rearrangement and mesenchymal phenotype may arise from clonal selection of tumor cells that have acquired the potential to drive metastatic progression of ALK-positive NSCLC.

Purpose To evaluate the impact of a predefined gene expression–based classifier for clinical risk estimation and cytotoxic treatment decision making in neuroblastoma patients. Patients and Methods Gene expression profiles of 440 internationally collected neuroblastoma specimens were investigated by microarray analysis, 125 of which were examined prospectively. Patients were classified as either favorable or unfavorable by a 144-gene prediction analysis for microarrays (PAM) classifier established previously on a separate set of 77 patients. PAM classification results were compared with those of current prognostic markers and risk estimation strategies. Results The PAM classifier reliably distinguished patients with contrasting clinical courses (favorable [n = 249] and unfavorable [n = 191]; 5-year event free survival [EFS] 0.84 ± 0.03 v 0.38 ± 0.04; 5-year overall survival [OS] 0.98 ± 0.01 v 0.56 ± 0.05, respectively; both P < .001). Moreover, patients with divergent outcome were robustly discriminated in both German and international cohorts and in prospectively analyzed samples (P ≤ .001 for both EFS and OS for each). In subgroups with clinical low-, intermediate-, and high-risk of death from disease, the PAM predictor significantly separated patients with divergent outcome (low-risk 5-year OS: 1.0 v 0.75 ± 0.10, P < .001; intermediate-risk: 1.0 v 0.82 ± 0.08, P = .042; and high-risk: 0.81 ± 0.08 v 0.43 ± 0.05, P = .001). In multivariate Cox regression models based on both EFS and OS, PAM was a significant independent prognostic marker (EFS: hazard ratio [HR], 3.375; 95% CI, 2.075 to 5.492; P < .001; OS: HR, 11.119, 95% CI, 2.487 to 49.701; P < .001). The highest potential clinical impact of the classifier was observed in patients currently considered as non–high-risk (n = 289; 5-year EFS: 0.87 ± 0.02 v 0.44 ± 0.07; 5-year OS: 1.0 v 0.80 ± 0.06; both P < .001). Conclusion Gene expression–based classification using the 144-gene PAM predictor can contribute to improved treatment stratification of neuroblastoma patients.

592 Background: About 50% of ER positive (ERpos) BC are resistant to hormone treatment. In absence of ligand, ERs are evenly distributed in nuclei in normal tissue. Upon ligand binding, ERs dimerizes and form a discrete focal subnuclear distribution pattern (FDP), which are associated with transcriptional activation of ER. This ER FDP is observed in BC. We hypothesized that in BC that the presence/absence of ER in the FDP could predict antiestrogen activity. This study describes an immunohistochemistry (IHC) method, by which a biomarker could be developed to investigate this hypothesis. Methods: 303 archived BC biopsies were obtained along with clinical and pathology data. Biopsies were analyzed for standard HES, ER, progesterone receptor (PR) and Ki67. PR positivity was determined with distinct antibodies to the A and B isoforms. The A-ER and D-ER nuclear patterns were analyzed at x1000 magnification. Results: Mean age; 58 (17 -89). Histology: ductal 85% lobular 13%, other 2%; 84% ERpos and 84% either PRApos or PRBpos, 7% ERpos and 7% PRApos or PRBpos only. All but 3 ERpos cases had received AntiEs; Adj. chemotherapy 51%, Stage: I 51%, II 43%, III 6 %. Grade: I 24%, II 51%, III 25%. Median follow up 31 months (mo). Local or distant progression (PD) was 19%. ER status was D-ER in 78% and A-ER in 22% of the biopsies and PD was associated with the D-ER pattern (p = 0.06), e.g. the hypothetically non-functional pattern D-ER was associated with a higher rate of PD (4 vs 35). With DFS defined as time to PD or death (5 year cut off), ERpos was better that ERneg (HR= 0.38, p = 0.016). For ERpos BC, A-ER was numerically better than D-ER (HR =0.3, p=0.24 two-sided). In univariate analyses, A-ER pattern was associated with higher grade (p=0.035), anisonucleosis (p=0.06), mitotic index (p=0.02), and Ki67 (=0.08). No association was found between ER pattern and age, stage or HER2 status. Conclusions: This study supports the hypothesis that AntiEs are mainly active in BC with A-ER pattern, which is targetable by AntiEs. A larger number of events are needed to reach significance in time-dependent analyses and confirmatory studies are warranted.

10556 Background: ALK high copy number (HCN) seems to be a frequent event, described in 13-17% of Non-small cell lung cancer (NSCLC). The goal of this study was to describe ALK genomic aberrations on NSCLC patients and cell lines, to explore the ALK HCN response to crizotinib through in vitro assays and to report three patients case. Methods: 191 Paraffin embedded specimens from advanced NSCLC patients and 27 NSCLC cancer cell lines were screened for ALK copy number by fluorescent in situ hybridization (FISH). Crizotinib sensitivity was evaluated in 9 cell lines through WST1 assays and clonogenic tests. Three patients exhibiting ALK HCN were assessed for response to crizotinib. Results: EML4-ALK translocation was present in 22 pts (11.5%). 21 pts (11%) exhibited over 6 copies of ALK. 6 (22%) cell lines displayed more than 5 copies of ALK, 19 (70%) presented a gain of 3 or 4 ALK copy number, only one cell line exhibited normal ALK copies and one harbored EML4-ALK translocation. FISH with CEP2 revealed a polysomy of chromosome 2 in cases with ALK HCN.Out of the 9 cell lines tested, 4 ALK HCN cell lines (H661, A427, BEN, H1299) exhibited increased sensitivity for crizotinib vs. 3 low ALK copy number (LCN) cell lines (H1975, H1651, H1650) with a low sensitivity. Median IC50 with crizotinib values was1750 nM [300-2800nM] in ALK HCN cell lines vs 4500 nM [800-8000nM] in ALK LCN cell lines, p=0.35. 3 patients with ALK HCN tumor received crizotinib ( in 4th , 5th and 6th -line therapy) for 2, 3 and 5 months with stable disease as best response and clinical benefit in 2 pts. Conclusions: ALK HCN may predict sensitivity to crizotinib. A clinical study is planned in ALK HCN pts.

10069 Background: Endometrial stromal tumours are rare uterine lesions comprising undifferentiated endometrial sarcoma (UES), endometrial stromal sarcoma low grade (ESS) and stromal nodule (SN). Co...