Alternative lengthening of telomeres (ALT) is likely to be an important target for anticancer treatment as approximately 10% of cancers depend on this telomere maintenance mechanism for continued growth, and inhibition of ALT can cause cellular senescence. However, no ALT inhibitors have been developed for therapeutic use because of the lack of a suitable ALT activity assay and of known ALT-specific target molecules. Here we show that partially single-stranded telomeric (CCCTAA)(n) DNA circles (C-circles) are ALT specific. We provide an assay that is rapidly and linearly responsive to ALT activity and that is suitable for screening for ALT inhibitors. We detect C-circles in blood from ALT(+) osteosarcoma patients, suggesting that the C-circle assay (CC assay) may have clinical utility for diagnosis and management of ALT(+) tumors.

Some immortalized mammalian cell lines and tumors maintain or increase the overall length of their telomeres in the absence of telomerase activity by one or more mechanisms referred to as alternative lengthening of telomeres (ALT). Characteristics of human ALT cells include great heterogeneity of telomere size (ranging from undetectable to abnormally long) within individual cells, and ALT-associated PML bodies (APBs) that contain extrachromosomal telomeric DNA, telomere-speci®c binding proteins, and proteins involved in DNA recombination and replication. Activation of ALT during immortalization involves recessive mutations in genes that are as yet unidenti®ed. Repressors of ALT activity are present in normal cells and some telomerase-positive cells. Telomere length dynamics in ALT cells suggest a recombinational mechanism. Inter-telomeric copying occurs, consistent with a mechanism in which singlestranded DNA at one telomere terminus invades another telomere and uses it as a copy template resulting in net increase in telomeric sequence. It is possible that t-loops, linear and/or circular extrachromosomal telomeric DNA, and the proteins found in APBs, may be involved in the mechanism. ALT and telomerase activity can co-exist within cultured cells, and within tumors. The existence of ALT adds some complexity to proposed uses of telomererelated parameters in cancer diagnosis and prognosis, and poses challenges for the design of anticancer therapeutics designed to inhibit telomere maintenance.

a b s t r a c tAlternative Lengthening of Telomeres (ALT) activity can be deduced from the presence of telomere length maintenance in the absence of telomerase activity. More convenient assays for ALT utilize phenotypic markers of ALT activity, but only a few of these assays are potentially definitive. Here we assess each of the current ALT assays and their implications for understanding the ALT mechanism. We also review the clinical situations where availability of an ALT activity assay would be advantageous. The prevalence of ALT ranges from 25% to 60% in sarcomas and 5% to 15% in carcinomas. Patients with many of these types of ALT[+] tumors have a poor prognosis.

Human cancer cells maintain telomeres by telomerase activity (TA) or by alternative lengthening of telomeres (ALT). We proposed to define the prevalence of the two telomere maintenance mechanisms (TMM), to assess their association with histology, and to compare their prognostic relevance in a series of 93 patients with liposarcoma. ALT was detected by assaying ALT-associated promyelocytic leukemia nuclear bodies and TA was assayed using the telomeric repeat amplification protocol. ALT or TA was found in 25.9% or 26.6% of 139 tested liposarcoma lesions, respectively. Three lesions were ALT+/TA+ whereas f50% of lesions did not show any known TMM. TMM phenotype was consistent during disease progression. ALT was prevalent in dedifferentiated and in grade 3 liposarcomas whereas TA prevailed in most round-cell myxoid and in grade 2 liposarcomas. ALT and TA incidence was similar in primary and recurrent lesions whereas metastases were more frequently TA+ than ALT+ (59% versus 18%; P = 0.04). TMM presence negatively affected patient prognosis (P = 0.001): increased mortality was associated with positivity for TA (P = 0.038) or ALT (P < 0.0001) compared with TMM absence. ALT proved to be a stronger prognostic discriminant of increased mortality than TA even when adjusted for tumor location, grade, and histology (hazard ratio for cause-specific death, 3.58 versus 1.15). Our results indicate that ALT can support fully malignant liposarcomas and is associated with unfavorable disease outcome. (Cancer Res 2006; 66(17): 8918-24)