RESEARCH NEWS & VIEWS cancer — the diameter of the tumour and the include changes that reduce the cell’s drive to cells dormant for a longer, but still temporary, number of lymph nodes containing cancer- divide, similar to those that limit proliferation period of time. We believe that a combined ous cells, which indicates whether the primary in cancer stem cells (for example, see ref. 6). strategy that concomitantly targets both the disease had spread at the start of treatment. These changes could be epigenetic — modu- tumour cells and the surrounding microenvi- However, even among women with small, lating gene expression without affecting the ronment has a higher probability of destroy- node-negative tumours, the risk of metastasis underlying DNA sequence — or genetic5. ing the dormant cells or of inducing a lifelong was about 10% over the 15-year period. Extrinsic factors include crosstalk between state of dormancy, and hence provides a higher Pan et al. acknowledge that there are different cell types in the surrounding micro- probability of a cure. A better understanding of several caveats to their analysis, including a environment, such as endothelial cells that line the crosstalk between the dormant cells, their lack of available data on how many women vessels, immune cells and fibroblasts, which surrounding cell types and the immune sys- completed their adjuvant treatment, and make up the structural framework of tissues. tem is crucial for developing effective micro- suboptimal treatment of people who had one A second dimension is vasculature-related environment-directed therapies. In addition, particular type of ER-positive breast cancer, dormancy, in which the tumour-cell popula- attention should be paid to the possibility dubbed HER2-positive. Nonetheless, it is clear tion is kept small because poor vascularization that dormant breast tumour cells undergo that, even after adjuvant endocrine therapy, in the region leads to a lack of nutrients and evolutionarily conserved programs that lead women with ER-positive, early-stage breast oxygen. A third dimension is immune-medi- to a stem-cell-like, prolonged resting state. cancer remain at persistent risk of recurrence ated dormancy, in which the immune system Maintaining this stem-cell status permanently for at least 20 years after the original diagno- controls an expanding tumour-cell population might be another way to prevent dormant cells sis. These findings, together with data from by continuously searching for and eliminating from reawakening. ■ another analysis of breast-cancer relapse2, cancerous cells. might have implications for long-term follow- If this dormancy status becomes unbalanced Giuseppe Curigliano is in the Department up strategies, and possibly for treatments. owing to changes in any of these factors, dor- of Hematology and Oncology, European One way to decrease the risk of relapse mant cells awaken and metastatic disease Institute of Oncology, University of Milan, might be to extend the duration of adjuvant develops. Thus, strategies to prevent relapse 20141 Milan, Italy. Fatima Cardoso is at the endocrine therapy — a strategy that is already should aim either to stop dormant cells from Breast Unit, Champalimaud Clinical Center, being tested. We wonder, however, whether awakening or to destroy them when dormant. Champalimaud Foundation, 1400–038 this approach will be sufficient to reduce or To achieve this goal, it would be beneficial to Lisbon, Portugal. avoid the risk of late metastasis. One trial3 identify those patients who will have a late e-mails: [email protected]; published in 2016 indicates that increasing relapse and to tailor a therapeutic strategy to [email protected] the duration of adjuvant therapy from five them. ER-positive breast cancers have a low 7 1. Pan, H. et al. N. Engl. J. Med. 377, 1836–1846 (2017). to ten years significantly improves rates of level of intratumoral cell diversity , such as 2. Colleoni, M. et al. J. Clin. Oncol. 34, 927–935 (2016). disease-free survival five years after the treat- distinct forms of genetic, epigenetic and func- 3. Goss, P. E. et al. N. Engl. J. Med. 375, 209–219 ment ceases, and lowers the incidence of can- tional diversity. A better understanding of (2016). 4. Colleoni, M. et al. Lancet Oncol. 19, 127–138 (2017). cer arising in the previously unaffected breast this diversity might lead to the identification 5. Goss, P. E. & Chambers, A. F. Nature Rev. Cancer 10, — but does not increase overall survival rates. of factors that enable certain cells to become 871–877 (2010). A second4 trial compared continuous adju- resistant to endocrine therapy and survive in 6. Tosoni, D. et al. EMBO Mol. Med. 9, 655–671 (2017). 7. Ellis, M. J. & Perou, C. M. Cancer Discov. 3, 27–34 vant therapy between years five and ten with metastatic niches. (2013). an ‘on–off’ therapy that aimed to resensitize Perhaps the major effect of extended adju- cancer cells that might have become resistant vant endocrine therapy is to keep tumour This article was published online on 29 January 2018. to the therapy. It found no difference in rates of metastasis-free survival on completion of either treatment. ORGANIC CHEMISTRY Longer follow-ups are needed to better understand the effects of extended adjuvant endocrine therapy, because of the slowly pro- gressive nature of the disease. Nonetheless, it Reactive carbon species is clear that, although extending the duration of therapy can play a part in preventing late relapses, it might well need to be given for the tamed for synthesis rest of a woman’s life to be effective. This would raise issues of toxicity, compliance and cost. A highly reactive form of carbon, known as a carbyne, holds great promise for To best determine other possible ways to organic synthesis, but has been difficult to prepare. Reactions that produce reduce the risk of relapse, we must consider carbyne equivalents now unleash this synthetic potential. See Letter p.86 what might cause the dormant tumour cells from which metastasis arises to reawaken after many years. Dormant cells have escaped ROHAN E. J. BECKWITH to new types of C–C bond-formation reaction. destruction by the body’s immune system Conventional approaches to C–C bond and entered a microenvironment that sup- he basis of organic chemistry is the study formation typically involve well-studied, ports their survival5. Once in this niche, of carbon-containing compounds with reactive carbon species. Carbynes, however, many mechanisms might underlie dormancy the aim of manipulating carbon atoms have been largely unexplored for C–C bond — indeed, dormancy can be thought of as a Tto generate new molecules through the forma- formation because their high reactivity makes multidimensional status, which involves mul- tion of carbon–carbon (C–C) bonds. On page them challenging to prepare and handle. Once tiple factors (Fig. 1). 86, Wang et al.1 report a method for harnessing formed, carbynes react with each other, with One dimension is cellular dormancy, in a reactive form of carbon known as a carbyne solvent molecules and with other substrates which intrinsic or extrinsic factors drive cells (Fig. 1a), which has been underused in syn- in an uncontrolled manner, producing myriad into a resting state. Intrinsic factors might thetic chemistry. The findings open the door products. This has limited their applications, 36 | NATURE | VOL 554 | 1 FEBRUARY©201 82018Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved. ©2018 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved. NEWS & VIEWS RESEARCH and even efforts to study these species. a Carbynes have previously been formed as C complexes with metals, which can then be Carbyne decomposed to release the free carbyne (see b refs 2 and 3, for example). By conducting such Mask decompositions in water at room temperature, researchers have prepared simple compounds Light EtO C Mask 2 Mask that contain C–C triple bonds from the reac- EtO C 4 EtO C 2 tions of free carbynes with each other . The 2 Precursor formation of undesired side products in these Mask Carbyne equivalent Mask reactions was minimized because the rate of reaction of the free carbynes with water c N OTf was several times slower than the rate of the 2 H 5 I O carbyne–carbyne reaction . However, product Me EtO C Me yields were low, and the method has limited 2 O CO Et Me 2 Me CO Et applications for synthesis. Catalyst, base, blue light, H 2 Wang et al. have devised a clever means of Precursor room temperature 32% yield accessing equivalents of carbynes, an approach that has broad synthetic utility. The authors Me prepared stable precursor compounds that d Me OTf contain two ‘masking’ groups (Fig. 1b). These N2 N2 precursors can be activated by a catalyst in I O Me a light-mediated process so that one of the EtO2C O CO Et EtO2C masks is released, generating a carbyne equiv- 2 Catalyst, base, white light, Me room temperature Diazoacetate alent. Further activation allows the carbyne 73% yield equivalent to react with substrate molecules, losing the second mask and forming three new Figure 1 | Carbyne equivalents for synthetic chemistry. a, Carbynes are reactive carbon-containing bonds in a single synthetic step. species with potential uses in organic synthetic chemistry. Dots indicate electrons; black sphere represents The authors demonstrated the power of this any organic chemical group. b, Wang et al.1 report a strategy that allows equivalents of carbynes to be single-step approach by transforming isobutyl- prepared from precursor compounds that contain two ‘masking’ groups. One mask is removed in a light- benzene (a simple hydrocarbon that contains mediated process, generating the carbyne equivalent; square brackets indicate that the carbyne equivalent a benzene ring) into a more-complex system, is formed transiently.