Kathryn Harper

Early dissemination seeds metastasis in breast cancer

Accumulating data suggest that metastatic dissemination often occurs early during tumour formation, but the mechanisms of early metastatic spread have not yet been addressed. Here, by studying metastasis in a HER2-driven mouse breast cancer model, we show that progesterone-induced signalling triggers migration of cancer cells from early lesions shortly after HER2 activation, but promotes proliferation in advanced primary tumour cells. The switch from migration to proliferation was regulated by increased HER2 expression and tumour-cell density involving microRNA-mediated progesterone receptor downregulation, and was reversible. Cells from early, low-density lesions displayed more stemness features, migrated more and founded more metastases than cells from dense, advanced tumours. Notably, we found that at least 80% of metastases were derived from early disseminated cancer cells. Karyotypic and phenotypic analysis of human disseminated cancer cells and primary tumours corroborated the relevance of these findings for human metastatic dissemination.

Mechanism of early dissemination and metastasis in Her2+ mammary cancer

Metastasis is the leading cause of cancer-related deaths; metastatic lesions develop from disseminated cancer cells (DCCs) that can remain dormant. Metastasis-initiating cells are thought to originate from a subpopulation present in progressed, invasive tumours. However, DCCs detected in patients before the manifestation of breast-cancer metastasis contain fewer genetic abnormalities than primary tumours or than DCCs from patients with metastases. These findings, and those in pancreatic cancer and melanoma models, indicate that dissemination might occur during the early stages of tumour evolution. However, the mechanisms that might allow early disseminated cancer cells (eDCCs) to complete all steps of metastasis are unknown. Here we show that, in early lesions in mice and before any apparent primary tumour masses are detected, there is a sub-population of Her2+p-p38lop-Atf2loTwist1hiE-cadlo early cancer cells that is invasive and can spread to target organs. Intra-vital imaging and organoid studies of early lesions showed that Her2+ eDCC precursors invaded locally, intravasated and lodged in target organs. Her2+ eDCCs activated a Wnt-dependent epithelial–mesenchymal transition (EMT)-like dissemination program but without complete loss of the epithelial phenotype, which was reversed by Her2 or Wnt inhibition. Notably, although the majority of eDCCs were Twist1hiE-cadlo and dormant, they eventually initiated metastasis. Our work identifies a mechanism for early dissemination in which Her2 aberrantly activates a program similar to mammary ductal branching that generates eDCCs that are capable of forming metastasis after a dormancy phase.

Macrophages orchestrate breast cancer early dissemination and metastasis

Cancer cell dissemination during very early stages of breast cancer proceeds through poorly understood mechanisms. Here we show, in a mouse model of HER2+ breast cancer, that a previously described sub-population of early-evolved cancer cells requires macrophages for early dissemination. Depletion of macrophages specifically during pre-malignant stages reduces early dissemination and also results in reduced metastatic burden at end stages of cancer progression. Mechanistically, we show that, in pre-malignant lesions, CCL2 produced by cancer cells and myeloid cells attracts CD206+/Tie2+ macrophages and induces Wnt-1 upregulation that in turn downregulates E-cadherin junctions in the HER2+ early cancer cells. We also observe macrophage-containing tumor microenvironments of metastasis structures in the pre-malignant lesions that can operate as portals for intravasation. These data support a causal role for macrophages in early dissemination that affects long-term metastasis development much later in cancer progression. A pilot analysis on human specimens revealed intra-epithelial macrophages and loss of E-cadherin junctions in ductal carcinoma in situ, supporting a potential clinical relevance.Early dissemination of cancer cells has been reported to occur in certain breast cancer models. Here the authors show that intra-epithelial macrophages in the early pre-cancer lesions drive early cancer cell dissemination through Wnt-1 secretion and that such events impact the later development of metastasis.

Corrigendum: Mechanism of early dissemination and metastasis in Her2 + mammary cancer

This corrects the article DOI: 10.1038/nature20609

Abstract 3233: Macrophages orchestrate early dissemination of HER2+ cancer cells

Metastatic dissemination was proposed to occur only in invasive cancer. However, detection of cancer cells in the bone marrow of patients with pre-invasive breast ductal carcinoma in situ (DCIS) showed that dissemination could, through unknown mechanisms, take place earlier. Late evolved invasive breast cancer cells disseminated by recruiting circulating monocyte-derived tumor associated macrophages (TAMs). We therefore investigated whether macrophages might actively participate in early dissemination. We show that pre-malignant cancer cells in MMTV-HER2 mice produced CCL2 in an NFkB dependent manner and thereby recruited resident mammary tissue macrophages into the duct, months before monocyte-derived TAMs arrived. Intra-epithelial mammary tissue macrophages induced an epithelial-to-mesenchymal transition in early HER2+ cells in a Wnt-dependent manner and thereby drove early dissemination. Consequently, depletion of macrophages from pre-malignant MMTV-HER2 mice significantly reduced early dissemination of pre-malignant cancer cells as measured by reduction in levels of circulating cancer cells and disseminated cancer cells in lungs. Interestingly, macrophage depletion during pre-malignant stages alone was sufficient to reduce the onset of late metastasis in MMTV-HER2 mice, indicating that early disseminating cancer cells contributed to late metastasis formation. In humans DCIS samples we frequently found intra-epithelial macrophages inside of ducts and their presence correlated with reduced E-Cadherin levels. Finally, a first pilot study indicated that those patients whose DCIS lesions contained macrophage +/E-Cadherin low microenvironments frequently had DCCs in the bone marrow. Taken together, we reveal that resident mammary tissue macrophages can promote early dissemination, and unravel a mechanism how early cancer spread might proceed in breast cancer patients and in other cancers where metastasis are diagnosed and the primary tumor is never detected (cancer of unknown primary). We also propose that early DCCs play a long-term causal role in metastasis development when patients are diagnosed in late stages, implying that the heterogeneity of metastases is higher than we might anticipate. Citation Format: Nina Linde, Arthur Mortha, Nicole Saenger, Maria Soledad Sosa, Kathryn Harper, Ethan Tardio, Tanja Fehm, Thomas Karn, Miriam Merad, Julio A. Aguirre-Ghiso. Macrophages orchestrate early dissemination of HER2+ cancer cells. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3233.

Abstract A56: Identification of an ErbB2+ early disseminating cancer cell subpopulation with metastatic potential

Abstracts: AACR Special Conference on Tumor Metastasis; November 30-December 3, 2015; Austin, TX Recent clinical evidence suggests that cells are capable of disseminating from pre-invasive breast cancer lesions. It has been proposed that these early disseminating cancer cells (eDCCs) lodge into secondary sites, enter a state of dormancy, escape conventional therapy, and potentially evolve through different pathways from the primary tumor to form metastases. However, the mechanisms that endow these early cancer cells to disseminate and their behavior in target organs remain unknown. Here we show that ErbB2+ early cancer cells, while non-tumorigenic are highly efficient in disseminating. While a majority of these eDCCs remain dormant in target organs they are still endowed with a metastatic capacity. We identified an ErbB2HIGH/p38LOW/E-cadherinLOW/PRHI population of early cancer cells, which undergo a Wnt-dependent EMT and disseminate. Based on this signature we propose that these early cancer cells tap into motility programs similar to what is seen during branching morphogenesis (PRHI/WntHI/p38LOW) in order to disseminate. Using intra-vital imaging of transgenic ErbB2-T-CFP pre-malignant lesions we imaged this process at high resolution revealing that ErbB2HIGH/p38LOW/E-cadherinLOW cells display activate cancer cell motility, invadopodia formation and intravasation capacity. This correlated with CK8/18+/HER2+ circulating cancer cells as well as eDCCs in the lung and bone marrow of mice, a process that was enhanced following 2 weeks of systemic p38α/β inhibition. We conclude that during stages previously considered non-invasive, some HER2+ cancer cells can aberrantly activate an invasive morphogenetic program causing early dissemination that is normally restricted by active p38 signaling. These eDCCs carry latent metastatic initiating capacity and may contribute to relapse, altering our understanding of metastasis onset and how it might be targeted effectively. Citation Format: Kathryn Harper, Maria Soledad Sosa, Hedayatollah Hosseini, Alvaro Avivar Valderas, Chandandaneep Nagi, Roger Davis, Christoph Klein, David Entenberg, John Condeelis, Eduardo Farias, Julio Aguirre-Ghiso. Identification of an ErbB2+ early disseminating cancer cell subpopulation with metastatic potential. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Metastasis; 2015 Nov 30-Dec 3; Austin, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(7 Suppl):Abstract nr A56.

Abstract PR14: ErbB2-induced epithelium to mesenchyme transition (EMT) is blocked by p38α/β signaling to restrict early dissemination in breast cancer

Our aim is to better understand the mechanisms dictating the early dissemination of cells from pre-malignant lesions, as well as those controlling cell fate decisions at secondary sites to develop more effective therapies for eradicating these populations. The majority of breast cancer patients die from metastatic disease, and not from their primary tumor. The current paradigm argues that dissemination of tumor cells can only occur from invasive lesions observed during late stages of the disease. However, clinical evidence suggests that “pre-malignant” mammary epithelial cells can disseminate and accumulate in target organs like the bone. This can occur from non-invasive lesions such as those found in Ductal Carcinoma in Situ (DCIS) patients. Thus, “early” disseminated tumor cells (DTCs) might be critical contributors to metastasis, but the mechanisms behind these events remain largely unknown. We had shown that p38-alpha/beta signaling maintains mammary gland ductal architecture by promoting anoikis and preventing disorganization of the luminal and basal epithelial compartments. We propose that, in addition to regulating mammary gland morphogenesis, p38-alpha/beta regulates a switch controlling early dissemination. In clinical samples we have found that ErbB2+ DCIS or invasive lesions show a down-regulation of the p38-signaling pathway, suggesting a role for the inhibition of p38-alpha/beta during cancer progression. In MMTV-Neu mammary glands ErbB2 is inversely correlated to membranous beta-catenin (inactive beta-catenin). Further, loss of P-ATF2 (p38 target) also coincides with reduced E-cadherin in this ErbB2+ epithelium. We also reveal that inhibition of p38-alpha/beta in MCF10A-ErbB2 or “pre-malignant” MMTV-ErbB2 acini results in increased invasion and disruption of normal acinar structure. In the MCF10A-ErbB2 model this correlated with E-cadherin down-regulation and induction of an 11-gene Wnt signaling EMT signature (e.g. up-regulation of Wnt5a/b, Wnt11, SNAIL, TWIST, N-cadherin, ZEB2, and SLUG). This EMT signature was also induced simply by p38-alpha/beta inhibition or ErbB2 overexpression in parental MCF10A cells. Further, treatment of MCF10A cells with Wnt3a inhibited p38-alpha and ATF2 phosphorylation. Inhibition of p38-alpha/beta signaling in MMTV-ErbB2 mice strongly increased the accumulation of pre-malignant MECs (detected by surface ErbB2) in the lungs and bone marrow. However, at least in lungs, >80% of these DTCs remained dormant as measure by P-Rb staining. We propose that in pre-malignant lesions p38-alpha/beta acts as an inhibitor to ErbB2-induced EMT, anoikis resistance and early dissemination. We further propose that Wnt signaling executes the EMT and also the silencing of p38-alpha/beta signaling to allow for early dissemination. This abstract is also presented as Poster B74. Citation Format: Kathryn Harper, Maria Soledad Sosa, Alvaro Avivar-Valderas, Huei-Chi Wen, Julio Aguirre-Ghiso. ErbB2-induced epithelium to mesenchyme transition (EMT) is blocked by p38α/β signaling to restrict early dissemination in breast cancer. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Invasion and Metastasis; Jan 20-23, 2013; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2013;73(3 Suppl):Abstract nr PR14.

Abstract SY17-02: Mechanisms driving dormancy of disseminated tumor cells

The vast majority of cancer related deaths are due to metastasis. However, our understanding of this process is limited. This is clearer when we consider that these lesions can develop years or decades after successful primary tumor treatment. These long periods of clinical remission can be explained by minimal residual disease (i.e. Disseminated Tumor Cells, DTCs) entering a non-productive or dormant state. However, our limited knowledge of these events has curtailed the development of strategies to prevent metastasis. Key questions in this field are: How does early dissemination start and what are the mechanisms? How does the tumor microenvironment aid this process? Are primary tumor (PT) niches responsible for DTCs growth or quiescence? What role does the microenvironment in the metastatic niche play in determining the timing of DTC dormancy? In patients, DTCs that are non-proliferative can be found in sites where they usually form secondary lesions or in sites where they never do. Thus despite being able to disseminate these DTCs are “growth-suppressed” by certain target organ microenvironments. Several mechanisms are proposed to explain clinical dormancy. The absence of proliferation markers in DTCs in patients and experimental models suggest that solitary DTCs are quiescent. However, angiogenic dormancy or immune system-mediated tumor mass dormancy might also be responsible for maintaining residual disease dormant. At least three potential scenarios might explain DTC dormancy. 1) DTCs from invasive cancers activate stress signals in response to the dissemination process and/or a growth suppressive target organ microenvironment, inducing dormancy. 2) Therapy and/or microenvironmental stress conditions acting on PT cells carrying specific gene signatures primes newly DTCs to enter dormancy. Thus, specific PT “stress microenvironments” might trigger long-term DTC dormancy. 3) Lesions pathologically defined as non-invasive carry cells able to undergo micro-invasion and disseminate. Here although these DTCs were able to disseminate and survive they are unfit for expansion in secondary sites. Perhaps occasional cell divisions allow them to progress via epigenetic and genetic pathways to a fully metastatic cell able to grow in secondary sites. We propose that DTC dormancy is ultimately a survival strategy that when blocked will prevent metastasis. Below we expand on the first two scenarios. For additional details on the third scenario please see. DTC dormancy and the target organ microenvironment. Solitary DTCs in target organs establish interactions with the ECM, immune cells and the vascular system. This and the pattern of metastasis proposed by the seed and soil theory suggests that microenvironments in the target organ can determine metastatic growth vs. dormancy. Our work shows that in squamous carcinoma cells (HEp3) reduced urokinase (uPA) receptor (uPAR) expression deactivated α5β1-integrins and this made these cells incapable of binding efficiently to fibronectin. This resulted in reduced Src-FAK and EGFR signaling but also in p38α/β activation. This caused these tumor cells to enter a state of dormancy characterized by a prolonged G0-G1 arrest. Others have reproduced these findings showing that loss of β1-integrin or FAK signaling in breast cancer models can also induce dormancy and that Src-MLKC signaling can prevent dormancy. In addition, an enriched collagen-I lung microenvironment can trigger intravenously delivered tumor cells to exit from dormancy as solitary cells. On the other hand environments rich in fibrillar collagen-I can induce quiescence of melanoma cells via activation of the discoidin domain receptor 2 and p15INK4b induction. These results imply that stress signaling induced either by therapies or by a restrictive (i.e. fibrotic or non-fibrotic target tissues depending on the tumor type) tissue microenvironments could activate dormancy (or its interruption) in DTCs. Our work in the HEp3 model also revealed that activation of p38α/β, while inhibiting ERK1/2 signaling activates an unfolded protein response (UPR). These signals induce survival and quiescence of dormant HEp3 (D-HEp3) cells. D-HEp3 cells in vivo enter a deep G 0 -G 1 arrest due to p21 and p27 upregulation. At least 3 transcription factors (TFs), p53, BHLHB3/41/Sharp1 and NR2F1 were induced by p38α/β and required for dormancy in vivo. Bone marrow derived dormant HEp3 cells displayed a low ERK/p38 signaling ratio and induction of these TFs. Interestingly, metastasis suppressor genes (MSGs) like MKK4 and MKK6 activate p38, BHLHB3 is a target of p38 and Nm23-H1 (another MSG) appears to function via the downregulation of EDG2 LPA receptor signaling through ERK1/2. Thus, different mechanisms converge on the regulation of the ERK/p38 signaling ratio to dictate DTC fate. Overall these studies provide clues as to potential mechanisms to induce or maintain dormancy or eradicate DTCs by targeting their survival signals. Primary tumor microenvironments as determinants of DTC fate. Gene signatures identified in PTs predict long-term metastatic relapse more than a decade later and in the absence of the PT carrying the signature. Gene profiles from the tumor stroma also predict patient outcome. This suggests that a crosstalk between PT cells and their microenvironment in primary sites can dictate distant disease progression. One interpretation is that genes signatures in the PT and the microenvironment determine DTC fate. Since symptomatic metastasis show homogeneous progression, the gene signatures in the PT might provide information on how those individual or groups of genes influence dormancy of DTCs. Modeling how the genes influence DTC survival and quiescence or subsequent angiogenesis or interaction with the immune system might reveal how they regulate minimal residual disease. Importantly, determining whether signatures derived from circulating tumor cells (CTCs) (i.e. recently intravasated tumor cells) are more or equally informative than the PT signatures might justify characterizing CTCs vs. DTCs (i.e. CTCs that already lodged and reside in target organs). Another potential example of the primary tumor influencing DTC behavior proposed that CTCs might return to the PT in a self-seeding process and this helps “breed” more aggressive variants that colonize target organs. These studies showed that aggressive variants of MDA-MB-231 breast cancer cells were highly efficient in disseminating and cross-seeding contra lateral tumors. The less aggressive variants of different cancer cell lines were less efficient in the seeding self/cross-seeding process. These data suggest that development of a more aggressive metastatic progeny requires the ability of PTs to attract their own CTCs back and that these tumor cells can efficiently re-colonize the PT. It is unclear how this happens after PT surgery, but it is possible that these events take place when multiple metastases co-exist. It will be interesting to determine whether CTCs from patients with early and advanced lesions carry already these “self-seeding” signatures. An additional scenario of a PT microenvironment influencing DTC behavior is that of systemic tumor instigation. This model proposes that PTs can influence the growth of otherwise-indolent DTCs or micrometastases. This is though to occur by mobilizing bone-marrow cells into the stroma of the indolent lesions. These results suggest that growth and proliferation of poorly aggressive tumors (dormant DTCs and/or micrometastasisβ) can be regulated on a systemic level by endocrine factors released by certain instigating tumors. However, it remains unclear how metastasis are instigated years or decades after patients underwent PT surgery. Regardless of their origin we propose that characterization of DTCs would be the most relevant because they carry the aggregate information of their origin, how treatment influenced their adaptation and/or selection and ultimately how the target organ dictated their adaptation and/or selection. The challenges presented by the problem of cancer dormancy are significant and studying DTCs and dormant disease is difficult. But the benefits of unraveling the inherent complexity of this step of metastasis biology should be of great impact for cancer patients. Grant Support: Samuel Waxman Cancer Research Foundation Tumor Dormancy Program, NIH/National Cancer Institute (CA109182, CA163131), NIEHS (ES017146) and NYSTEM grant to J.A.A-G. M.S.S. is supported by a DoD-BCRP Grant-10904826. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr SY17-02. doi:1538-7445.AM2012-SY17-02