Integrating pharmacology and in vivo cancer models in preclinical and clinical drug development

https://doi.org/10.1016/j.ejca.2004.01.003Get rights and content

Abstract

Historically, cancer drug development has been a roller coaster. Numerous agents have shown exciting activity in preclinical models and yet have had minimal activity clinically. These disappointments have led to reasonable scepticism about the true value of both syngeneic and xenograft rodent tumour models in accurately identifying agents that will have important clinical utility. Whereas the development of newer techniques, including transgenic mouse models of cancer, offers the potential to develop more predictive models, the role of such mice in cancer drug development is not yet validated. To advance in our understanding of predictive model systems it may be wise to analyse both the successes and the failures of conventional models in order to understand some of their limitations and perhaps to avoid making the same mistakes in the future. Here we review the value and limitations of xenograft models, and the role of integrating preclinical pharmacology in developing new treatments for solid tumours of childhood.

Introduction

The evaluation of antitumour agents in immune-deficient mice (athymic nude or severe combined immunodeficient (scid) mice) transplanted with human tumours is the major model system for drug development. In its most simple iteration, tumours are grown subcutaneously, and the model allows rapid and quantifiable assessment of antitumour activity relative to mouse toxicity. Logically, precedence should be given to those agents that show the greatest antitumour activity in the preclinical setting, assuming the preclinical data are predictive of drug activity in human studies. The challenge lies in being able to extrapolate these results to the clinic. Indeed, can this ever be done with any degree of confidence? The extensive screening for over 10 years by the National Cancer Institute (NCI) suggests only a moderate predictive value for their xenograft models, and even less concordance between in vitro testing data and clinical utility [1]. In this analysis, xenograft tumours derived from a particular cancer type did not predict for activity in the respective clinical disease; rather broad-spectrum activity in the preclinical models was associated with greater clinical activity. Interestingly, these results recapitulate those from syngeneic rodent tumour models used in the NCI screening programme before 1985, where clinical activity was associated with a high response rate in five of eight unrelated solid tumour models. The deficiency in all of these studies has been an inability or failure to relate tumour-response data to clinically achievable drug systemic exposures (i.e. studies on pharmacokinetics were not undertaken).

There are many reasons why preclinical results do not predict human efficacy. Here we will focus on differences in interspecies pharmacology. However, it is clear that the design of therapeutic clinical trials often fails to build upon the strong preclinical leads that may guide aspects of clinical trials' design, such as scheduling of drug administration. Conversely, criteria used to advance an agent in preclinical trials may not be as stringent as those used to evaluate response rates in the clinical setting. For example, 58% inhibition of tumour growth, a criterion used by NCI for assessing the activity of a drug against xenograft tumour models, represents progressive disease in a clinical trial. Our data suggest that if certain aspects of the study design are given careful consideration, it may be valid to predict clinical results derived from preclinical work with the xenograft model. These aspects include (a) the development of early-passage models of the appropriate human cancer, rather than the use of ‘ancient’ cell lines that have been in culture for decades; (b) the use of clinically relevant response criteria to evaluate a new entity; (c) the assessment of tumour responsiveness relative to drug systemic exposure; and (d) a rational consideration of the major/minor strengths and weaknesses of the model (i.e. all models have certain limitations). Here we review our experience using models of childhood cancers in drug development.

Section snippets

Retrospective studies: validation of tumour models

Xenograft tumour models, in which a human cancer is transplanted into immune-deficient mice, have been explored since the mid 1960s, but became more frequently used after the identification of the athymic nude mutant mouse which is deficient in T cells 2, 3. The more recent discovery of other immune-deficient mouse strains has further expanded the options for host transplantation. For example, the non-obese diabetic (NOD) scid mouse has proved useful for the propagation and testing of agents

Prospective use of xenograft models

The first prospective use of xenograft data followed perhaps our observation that melphalan had very significant activity against the ‘diagnosis’ panel of rhabdomyosarcomas [7]. In a phase II clinical trial in 13 patients who had failed standard chemotherapy, melphalan demonstrated marginal activity (one partial response). However, pharmacokinetic analysis showed that exposure to this agent in children was essentially similar to that in mice at doses inducing tumour regression. Consequently, we

Acknowledgments

We thank the numerous research technologists, postdoctoral fellows and colleagues at St. Jude who have contributed to work summarised in this article, which was conducted over the last 25 years. These studies have been supported through PHS award CA23099 from the National Cancer Institute, and by the American, Lebanese, Syrian, Associated Charities (ALSAC).

References (31)

  • P.J. Houghton et al.

    Drug sensitivity and resistance in the xenograft model

  • J.A. Houghton et al.

    Melphalana potential new agent in the treatment of childhood rhabdomyosarcoma

    Cancer Treat. Rep.

    (1985)
  • M.E. Horowitz et al.

    Phase II testing of melphalan in children with newly diagnosed rhabdomyosarcomaa model for anticancer drug development

    J. Clin. Oncol.

    (1988)
  • P.J. Houghton et al.

    Evaluation of a novel bis-naphthalimide anticancer agent, DMP 840, against human xenografts derived from adult, juvenile, and pediatric cancers

    Cancer Chemother. Pharmacol.

    (1994)
  • P.J. Houghton et al.

    Therapeutic efficacy of the cyclopropylpyrroloindole, carzelesin, against xenografts derived from adult and childhood solid tumors

    Cancer Chemother. Pharmacol.

    (1995)
  • Cited by (192)

    • Integrated pharmacokinetic/pharmacodynamic/efficacy analysis in oncology: importance of pharmacodynamic/efficacy relationships

      2023, Overcoming Obstacles in Drug Discovery and Development: Surmounting the Insurmountable-Case Studies for Critical Thinking
    • Translating complexity and heterogeneity of pancreatic tumor: 3D in vitro to in vivo models

      2021, Advanced Drug Delivery Reviews
      Citation Excerpt :

      This might be explained by their ability to adapt through constant passaging in in vitro culturing conditions, and select the most proliferative and resistant variants to survive. As a result, these cell lines have limited capacity to represent the real tumor of origin [39,236–238,241,242]. Additionally, significant variations in proteomics are observed between the cultured cell lines and the patients tumors [202].

    View all citing articles on Scopus
    View full text