Using highly sensitive microarray-based procedures, we identified eight microRNAs (miRNAs) showing robust differential expression between 31 laser-capture-microdissected nasopharyngeal carcinomas (NPCs) and 10 normal healthy nasopharyngeal epithelial samples. In particular, miRNA mir-29c was expressed at one-fifth the levels in tumors as in normal epithelium. In NPC tumors, the lower mir-29c levels correlated with higher levels of multiple mRNAs whose 3 UTRs can bind mir-29c at target sequences conserved across many vertebrates. In cultured cells, introduction of mir-29c down-regulated these genes at the level of mRNA and inhibited expression of luciferase encoded by vectors having the 3 UTRs of these genes. Moreover, for each of several genes tested, mutating the mir-29c target sites in the 3 UTR abrogated mir-29c-induced inhibition of luciferase expression. Most of the mir-29c-targeted genes identified encode extracellular matrix proteins, including multiple collagens and laminin ␥1, that are associated with tumor cell invasiveness and metastatic potential, prominent characteristics of NPC. Thus, we identify eight miRNAs differentially expressed in NPC and demonstrate the involvement of one in regulating genes involved in metastasis.microarray ͉ collagen ͉ metastasis ͉ miRNA M icroRNAs (miRNAs) are short (Ϸ22 nucleotides) noncoding RNAs involved in posttranscriptional silencing of target genes. In animals, miRNAs control expression of target genes by inhibiting translation, by degrading target mRNAs, or both, through binding to their 3Ј UTRs with varying degrees of sequence complementarity (1). miRNAs have been found to regulate genes involved in diverse biological functions, including development, differentiation, proliferation, and stress response (2). Recently, a growing number of miRNAs have been implicated in cancers, including mir-15 and mir-16 in B cell chronic lymphocytic leukemias (3, 4); mir-143 and mir-145 in colorectal cancer (5); mir- 17-5p, mir-21, mir-125b, mir-145, and mir-155 in breast cancer (6, 7); mir-19, mir-146, mir-181b, mir-221, mir-222, and mir-346 in thyroid cancer (8-10); and mir-21 in glioblastoma (11). A significant number of miRNAs also have been mapped to cancer-associated genomic regions (12). Expression of miRNA let-7 has been correlated with prognosis in lung cancer (13) and found to regulate Ras in the same tumor (14). Very recently, mir-10b has been shown to contribute to metastasis in breast cancer (15). Although many miRNAs have been implicated in regulating cancers, very few of their target genes, and hence their downstream mode of action, have been identified.We developed a sensitive microarray-based assay to profile miRNA expression and used it to analyze human miRNAs in laser-microdissected tumor and normal cells from biopsies of a highly invasive cancer, nasopharyngeal carcinoma (NPC), and site-matched normal tissues. Eight miRNAs were differentially expressed. One of them, mir-29c, down-regulated in NPC, was shown to target multiple mRNAs encoding extracellular ma...

To identify the molecular mechanisms by which EBVassociated epithelial cancers are maintained, we measured the expression of essentially all human genes and all latent EBV genes in a collection of 31 laser-captured, microdissected nasopharyngeal carcinoma (NPC) tissue samples and 10 normal nasopharyngeal tissues. Global gene expression profiles clearly distinguished tumors from normal healthy epithelium. Expression levels of six viral genes (EBNA1, EBNA2, EBNA3A, EBNA3B, LMP1, and LMP2A) were correlated among themselves and strongly inversely correlated with the expression of a large subset of host genes. Among the human genes whose inhibition was most strongly correlated with increased EBV gene expression were multiple MHC class I HLA genes involved in regulating immune response via antigen presentation. The association between EBV gene expression and inhibition of MHC class I HLA expression implies that antigen display is either directly inhibited by EBV, facilitating immune evasion by tumor cells, and/or that tumor cells with inhibited presentation are selected for their ability to sustain higher levels of EBV to take maximum advantage of EBV oncogenemediated tumor-promoting actions. Our data clearly reflect such tumor promotion, showing that deregulation of key proteins involved in apoptosis (BCL2-related protein A1 and Fas apoptotic inhibitory molecule), cell cycle checkpoints (AKIP, SCYL1, and NIN), and metastasis (matrix metalloproteinase 1) is closely correlated with the levels of EBV gene expression in NPC. (Cancer Res 2006; 66(16): 7999-8006)

Taken together, our data suggest that pathological margins and pathological tumor depth are major independent prognosticators not only for local tumor control, but also for DSS and OS.

Extensive composite defects of the lower jaw are defined as those that involve skin, mandible, oral mucosa, and soft tissues. The enormous size and multilayered nature of these defects challenge most of the current reconstructive techniques. For reconstruction of extensive composite mandibular defects in 36 advanced oral cancer patients, two free flaps were used simultaneously in a complementary fashion. The aim was to provide bone reconstruction and adequate soft-tissue coverage in an optimal form. Primary reconstruction was carried out in 34 of 36 cases. The fibula osteoseptocutaneous-radial forearm fasciocutaneous flap combination was most commonly used (n = 20), followed by the fibula osteoseptocutaneous-rectus abdominis myocutaneous flap (n = 11). The other combinations included the fibula osteoseptocutaneous-tensor fasciae latae, the fibula osteoseptocutaneous-rectus femoris, the iliac crest-radial forearm, and the iliac crest-tensor fasciae latae flaps. In 11 cases, the second free flaps were attached to the distal runoff of the first free flaps because of unavailability of recipient vessels. The mean operation time was 12 hours 10 minutes. The complete flap survival rate was 93 percent (67 of 72 flaps) with 2.8 percent total (2 of 72) and 4.2 percent partial (3 of 72) flap failures. Median follow-up time was 14 months, and 44 percent of the patients were alive at the time of evaluation, surviving an average of 36 months. The average survival time for those who died was 11.1 months. The authors believe that in selected cases the double free-flap procedure for one-stage reconstruction of massive mandibular defects is justified because it is safe and effective and improves the quality of life and the number of days spent outside of the hospital for these patients.