A framework for the development of effective anti-metastatic agents

Abstract

Most cancer-related deaths are a result of metastasis, and thus the importance of this process as a target of therapy cannot be understated. By asking 'how can we effectively treat cancer?', we do not capture the complexity of a disease encompassing >200 different cancer types - many consisting of multiple subtypes - with considerable intratumoural heterogeneity, which can result in variable responses to a specific therapy. Moreover, we have much less information on the pathophysiological characteristics of metastases than is available for the primary tumour. Most disseminated tumour cells that arrive in distant tissues, surrounded by unfamiliar cells and a foreign microenvironment, are likely to die; however, those that survive can generate metastatic tumours with a markedly different biology from that of the primary tumour. To treat metastasis effectively, we must inhibit fundamental metastatic processes and develop specific preclinical and clinical strategies that do not rely on primary tumour responses. To address this crucial issue, Cancer Research UK and Cancer Therapeutics CRC Australia formed a Metastasis Working Group with representatives from not-for-profit, academic, government, industry and regulatory bodies in order to develop recommendations on how to tackle the challenges associated with treating (micro)metastatic disease. Herein, we describe the challenges identified as well as the proposed approaches for discovering and developing anticancer agents designed specifically to prevent or delay the metastatic outgrowth of cancer.

Fig. 1. Overview of metastasis.

Metastasis is a complex multistep process, and the very concept of designing a metastasis-specific therapeutic must consider which part of the process is best to target. Given that metastases are derived mainly from invasive tumours, therapeutic efforts have often targeted the intrinsic invasive propensity of tumour cells^,. Tumour cell production of angiogenic factors and TGFβ can activate endothelial cells and fibroblasts to remodel tissues and promote tumour cell invasion of stromal-modified spaces. Targeting stromal elements in cancers remains an active area of research^–. Intravasation of tumour cells is promoted by binding to macrophages that cause transient permeability in the vasculature; thus, targeting tumour-associated macrophages might reduce the number of circulating tumour cells (CTCs). Multiple factors intrinsic to tumour cells (including epithelial-to-mesenchymal transition, production of proteases and migratory capacity) improve intravasation, often via effects on cell types including fibroblasts, neutrophils and macrophages. Most tumour cells that enter the vasculature die as a result of hydrodynamic physical damage or leukocyte attack. However, platelets can bind to and protect CTCs and improve their ability to establish secondary sites. Platelet–CTC aggregates settled at distant sites can release cytokines that attract granulocytes; targeting platelets or granulocyte recruitment can prevent metastasis. Additionally, abrogation of platelet–CTC binding, leading to a reduction in the number of circulating and potentially metastatic cells, might explain the suppression of metastasis by aspirin in breast and prostate cancer models. Survival and proliferation of newly deposited cancer cells in a metastatic site are arguably the most important stages of the metastatic process. Cancers with a propensity to metastasize do not grow in all organs, indicating that a limited number of organs provide a suitable stromal environment for their colonization. Preferred colonization sites, termed pre-metastatic niches, can be prepared in advance of the arrival of disseminated tumour cells through the actions of myeloid-derived suppressor cells (MDSCs) and tumour cell-derived extracellular vesicles (EVs), such as exosomes^,. Whether this process can provide novel therapeutic targets to limit the arrest and survival of metastatic cells remains unclear, with development of EV-specific drugs, for example, creating a challenge. Evidence also supports roles for neutrophils^, and MDSCs in metastatic colonization. Evasion of the antitumour immune response is another critical factor in metastatic colonization. No single tumour type seems to exhibit all these mechanisms; therefore, targeting any one stage of the metastatic process requires a tumour-specific understanding of the mechanisms involved.

Fig. 2. Dormancy and the metastatic niche.

Metastatic latency is more pronounced for certain types of cancer, notably breast and prostate cancer and melanoma. The detection of disseminated tumour cells (DTCs) in bone marrow aspirates, obtained long after eradication of the primary tumour, verifies the presence of dormant tumour cells and is predictive of disease recurrence^–. Dormant tumour cells are proposed to exist either as single cells in a state of cell cycle arrest or as small masses of cells that fail to expand into clinically detectable lesions, possibly owing to failure of angiogenesis, balanced rates of proliferation and apoptosis or effective immunosurveillance. Cells derived from these clusters that are too small to be detected by normal clinical imaging are presumably the source of circulating tumour cells detected in some patients after successful treatment of the primary tumour. The metastatic niche is likely to vary between different organs, reflecting the tissue-specific nature of the microenvironment in which the DTC is located; extensive crosstalk occurs between the tumour cells, stromal cells and extracellular matrix components of the niche. Various niches have been proposed, including the perivascular niche associated with the vasculature, the haematopoietic stem cell niche of the bone marrow and the osteoblastic niche in bone. The factors that maintain tumour cell dormancy in these niches are starting to be unravelled, with different extracellular matrix components, cytokines and other proteins being implicated for different cancer types and niches. Far less is understood about how dormancy is broken, which could occur following failure of immunosurveillance, in response to inflammation triggered by trauma or an infection or perhaps as a result of ageing-related deficiencies in tissue homeostasis. With regard to therapy, the challenge is to decide whether the aim should be to retain the DTCs in a dormant state or, instead, to disrupt their niche and dormancy, thereby rendering them susceptible to apoptotic or anoikic death and/or to chemotherapy.

Fig. 3. Development pathway for anti-metastatic agents.

The general process for development of anti-metastatic agents has the same fundamental basis as that used in the development of drugs with a direct antitumour mechanism of action, with some special considerations as highlighted in the figure and described as follows. In target identification and preclinical development, special consideration must be given to the functional relevance of the models being used, which should reflect human metastatic disease as much as possible; the role of the immune system in metastasis is a critical factor. The experimental conditions should also mimic those of the clinical setting. Drug discovery and subsequent preclinical testing strategies need to be designed to account for the fact that, in most cases, the anti-metastatic therapy under development will be given chronically in a healthier population of patients, such as those that have been cured of their primary disease but are at high risk of developing secondary tumours, necessitating oral administration and a risk–benefit profile lacking key toxicity liabilities. Other considerations, such as activity in several different preclinical models, an optimized pharmacokinetic (PK) profile and development of pharmacodynamic (PD) markers suitable for use in the clinic, are common to all cancer drug discovery and development programmes. Given the favourable risk–benefit profile necessary for anti-metastatic agents, an accelerated development approach can be taken by conducting initial phase I studies in healthy volunteers rather than the classical populations with advanced-stage cancer. The key aims of these studies are to determine the safety, PK profile and PD characteristics (ensuring the putative biomarkers developed can be measured in non-malignant tissues) in order to provide an early go or no-go decision point and ensure that the drug has the intended biological effects. To gain rapid biological proof of concept in patients with cancer, window-of-opportunity studies, in which a dose of the anti-metastatic agent is given before surgery to examine PD effects, can be considered. If validated surrogate end points of clinical efficacy are available, these can be used to substantially reduce development timelines and, provided agreement has been reached with appropriate regulatory bodies, support provisional approval. If successfully executed, this regulatory strategy will avoid the protracted clinical development timelines that are one of the greatest barriers to the development of anti-metastatic drugs. Provided that provisional approval is given, regulatory bodies will require further in-use continuous assessment, typically in confirmatory phase IV studies that can be funded using ongoing sales revenue. The aim of these larger-cohort and much longer duration clinical trials is to confirm that a pre-defined level of clinical benefit is achieved according to more traditional outcomes, such as overall survival. If provisional approval has not been given by regulators, then costly (in terms of both finance and time) randomized controlled phase III studies in large cohorts will be necessary to gain approval on the basis of standard clinical outcome measures. CE, Conformité Européene.