Use of the hollow fiber assay for the evaluation of DNA damaging agents

• Published in: Journal of pharmacological and toxicological methods 2011-11, Vol.64 (3), p.226-232
• Main Authors: Veiga, J. Pedro, Cooper, Patricia A., Pors, Klaus, Patterson, Laurence H., Bibby, Michael C., Shnyder, Steven D.
• Format: Article
• Language: English
• Subjects: Animals; Antineoplastic Agents - chemistry; Antineoplastic Agents - pharmacology; Antitumour drug screening; Cell Line, Tumor; DNA - drug effects; DNA break; DNA Breaks, Double-Stranded; Double-stranded; Doxorubicin; Doxorubicin - pharmacology; Drug Discovery - instrumentation; Drug Discovery - methods; Drug Evaluation, Preclinical - instrumentation; Drug Evaluation, Preclinical - methods; Drug Screening Assays, Antitumor - instrumentation; Drug Screening Assays, Antitumor - methods; Duocarmycin; Female; Hollow fiber assay; Humans; Mice; Mice, Inbred BALB C
• ISSN: 1056-8719; 1873-488X
• DOI: 10.1016/j.vascn.2011.04.006

The preclinical development and clinical progression of potential anticancer agents are highly time and resource-intensive. Traditionally, promising compounds in vitro undergo further screening in xenograft models, a long process that uses large numbers of animals. In order to hasten compound progression, the hollow fiber assay (HFA) was developed by the US National Cancer Institute as an additional filtering step in drug development, bridging the gap between in vitro and xenograft compound screening. The HFA demonstrates a good correlation in terms of clinical predictivity, and has significant reduction and refinement benefits for animal usage. In addition, the assay enables the study of basic pharmacological properties of compounds under investigation. The HFA has been mainly used as a rapid in vivo cytotoxicity screen, but has also been shown to be amenable to study drug/target interactions in vivo. One of the challenges of the HFA is the small sample sizes obtained, which can limit sensitivity. Here we specifically focus on the detection of DNA double-strand breaks, monitoring the effects of standard and novel anti-cancer agents on human lung, colon and breast cancer cell lines using immunoblotting and flow cytometry techniques for γ-H2A.X. This presented a further challenge due to the low abundance of the target event. We found a good correlation between techniques in terms of rate of detection and sensitivity confirming the ability to use the HFA for detection of these specific drug–target interactions. The results demonstrate good sensitivity and predictability for drug behavior in an assay where cell number is limited. In contrast to conventional xenograft studies, this short-term assay also enables analysis of pharmacodynamic endpoints in tumor cells in vivo. Importantly, there is a significant impact on reduction and refinement of the use of animals in incorporating this assay into the drug development process.