Tumors can become lethal when they progress from preinvasive lesions to invasive carcinomas. Here, we identify candidate tumor progression genes using gene array analysis of preinvasive and invasive tumors from mice, which were then evaluated in human cancers. Immediate early response protein IEX-1, small stress protein 1 (HSPB8), and tumor necrosis factorassociated factor-interacting protein mRNAs displayed higher expression levels in invasive lesions than in preinvasive lesions using samples obtained by laser capture microdissection (LCM) from transgenic erbB2, ras, and cyclin D1 mice. LCM-isolated tissues from patient-matched normal, ductal carcinoma in situ, and invasive ductal carcinoma revealed similar increased expression in invasive human cancers compared with preinvasive and normal samples. These genes induced anchorage independence, increased cell proliferation, and protected against apoptosis, singly or in collaboration with erbB2. Surprisingly, they were all up-regulated by 17B-estradiol and cyclin D1, and cyclin D1 overexpression increased p300/CBP binding to their promoters, supporting the model that cyclin D1-estrogen receptor (ER) coactivator interactions may be important to its role in ER-positive breast cancer. Additionally, an irreversible dual kinase inhibitor of ErbB signaling inhibited expression of the same genes. The up-regulation of genes contributing to increased invasiveness of ER-positive cancers offers a novel explanation for the contribution of cyclin D1 to a worse prognosis in ER-positive cancers. As targets of estrogen, cyclin D1, and erbB2 signaling, these candidates offer insights into the nature of the second events involved in breast cancer progression, regulatory events contributing to invasion, and potential targets of combined inhibition of hormone and growth factor signaling pathways. (Cancer Res 2006; 66(24): 11649-58)

BackgroundThe use of 18F-FDG PET–CT (PET–CT) is widespread in many cancer types compared to sarcoma. We report a large retrospective audit of PET–CT in bone and soft tissue sarcoma with varied grade in a single multi-disciplinary centre. We also sought to answer three questions. Firstly, the correlation between sarcoma sub-type and grade with 18FDG SUVmax, secondly, the practical uses of PET–CT in the clinical setting of staging (during initial diagnosis), restaging (new baseline prior to definitive intervention) and treatment response. Finally, we also attempted to evaluate the potential additional benefit of PET–CT over concurrent conventional CT and MRI.MethodsA total of 957 consecutive PET–CT scans were performed in a single supra-regional centre in 493 sarcoma patients (excluding GIST) between 2007 and 2014. We compared, PET–CT SUVmax values in relation to histology and FNCCC grading. We compared PET–CT findings relative to concurrent conventional imaging (MRI and CT) in staging, restaging and treatment responses.ResultsHigh-grade (II/III) bone and soft tissue sarcoma correlated with high SUVmax, especially undifferentiated pleomorphic sarcoma, leiomyosarcoma, translocation induced sarcomas (Ewing, synovial, alveolar rhabdomyosarcoma), de-differentiated liposarcoma and osteosarcoma. Lower SUVmax values were observed in sarcomas of low histological grade (grade I), and in rare subtypes of intermediate grade soft tissue sarcoma (e.g. alveolar soft part sarcoma and solitary fibrous tumour). SUVmax variation was noted in malignant peripheral nerve sheath tumours, compared to the histologically benign plexiform neurofibroma, whereas PET–CT could clearly differentiate low from high-grade chondrosarcoma. We identified added utility of PET–CT in addition to MRI and CT in high-grade sarcoma of bone and soft tissues. An estimated 21% overall potential benefit was observed for PET–CT over CT/MRI, and in particular, in ‘upstaging’ of high-grade disease (from M0 to M1) where an additional 12% of cases were deemed M1 following PET–CT.ConclusionsPET–CT in high-grade bone and soft tissue sarcoma can add significant benefit to routine CT/MRI staging. Further prospective and multi-centre evaluation of PET–CT is warranted to determine the actual predictive value and cost-effectiveness of PET–CT in directing clinical management of clinically complex and heterogeneous high-grade sarcomas.

Overexpression of cyclin D1 is associated with many cancers, and its overexpression is especially associated with a poor prognosis in breast cancer. Paradoxically, cyclin D1 is known to enhance radiation sensitivity, a finding that has not yet been therapeutically exploited. Proposed cyclin D1 functions that could be involved in this effect include cyclin-dependent kinase (CDK)-dependent phosphorylation of retinoblastoma gene product (pRb), titration of p21/p27 complexes, and less well-characterized effects on gene expression. In this report, we sought to clarify the functions of cyclin D1 that might contribute to enhanced radiation sensitivity. Breast cancer cells stably overexpressing a cyclin D1 mutant (KE) that cannot interact with its CDK partners to phosphorylate pRb were as radiation sensitive as those expressing wild-type D1. Although cyclin D1 has been proposed to affect radiation sensitivity through interactions with p21, a cyclin D1 mutant defective for p21 interactions also increased radiation sensitivity. Cyclin D1 overexpression is generally confined to hormone receptor-positive breast cancers, wherein standard therapies include both radiation and hormonal therapies. Among several proposed CDK-independent cyclin D1 targets, we have identified heat shock protein B8 (HSPB8) as a target particularly associated with cyclin D1 and ER-positive tumors. We therefore evaluated its potential contribution to radiation sensitivity. Overexpression of HSPB8 markedly increased radiation sensitivity, and HSPB8 small interfering RNA blocked cyclin D1's enhancement of radiation sensitivity. Taken together, our results show that some of cyclin D1's effects on radiation sensitivity are CDK and p21 independent and identify HSPB8 as a candidate CDK-independent cyclin D1 target that can mediate its effects. [Cancer Res 2007;67(22): 10774-81]