Repositioning approved drugs for the treatment of problematic cancers using a screening approach

Abstract

Advances in treatment strategies together with an earlier diagnosis have considerably increased the average survival of cancer patients over the last four decades. Nevertheless, despite the growing number of new antineoplastic agents introduced each year, there is still no adequate therapy for problematic malignancies such as pancreatic, lung and stomach cancers. Consequently, it is important to ensure that existing drugs used to treat other types of cancers, and potentially other diseases, are not overlooked when searching for new chemotherapy regimens for these problematic cancer types. We describe a screening approach that identifies chemotherapeutics for the treatment of lung and pancreatic cancers, based on drugs already approved for other applications. Initially, the 1280 chemically and pharmacologically diverse compounds from the Prestwick Chemical Library® (PCL) were screened against A549 (lung cancer) and PANC-1 (pancreatic carcinoma) cells using the PrestoBlue fluorescent-based cell viability assay. More than 100 compounds from the PCL were identified as hits in one or both cell lines (80 of them, being drugs used to treat diseases other than cancer). Selected PCL hits were further evaluated in a dose-response manner. Promising candidates for repositioning emanating from this study include antiparasitics, cardiac glycosides, as well as the anticancer drugs vorinostat and topotecan.

Fig 1. Schematic of the HTS assay representing a 384-well screening plate (left) and the respective data evaluation and visualization after normalization (right).

First two columns (pale blue): negative control (0.1% DMSO; no drug added); Last two columns (green): positive control (DOX, 10 μM, final DMSO concentration 0.1%); Middle columns (dark blue) test compounds (PCL drugs, 10 μM, final DMSO concentration 0.1%, 1 well/compound, 320 compounds/plate). The full PCL can be processed on four plates. Cell suspension (A549 cells shown here) is added to every well. The Z’-factor for plate validation is calculated using the fluorescence signals of the control wells (the green and blue shaded areas on the right side denote the average ± 3 x SD of the negative and the positive controls, respectively). In this example a Z’-factor of 0.747 is suitable to validate the plate.

Fig 2. PCL compounds identified as hits in A549 and/or PANC-1 cells, plot based on their mean scores from the primary screens.

A) all PCL hits; B) only hits which are used as antineoplastic drugs. Hits identified in only one of the cell lines are plotted and their score on the other is taken for zero. For all the names of the hit compounds, score values and pharmacological classification see S1 and S2 Tables.

Fig 3

Concentration-effect curves (means ± SD) for auranofin (A), daunorubicin and cytarabine (B) in A549 cells (PrestoBlue fluorescence assay, 72 h exposure). Curves fitting and graphs were prepared using GraphPad Prism 6.