Drug Repurposing in cancer: the search for new treatments

“The average cost of developing a new drug and taking it to the market is estimated at $2.56 billion”

Many of us have had to look for transferable skills within our CVs to apply for new positions. Cancer researchers are now looking at established drugs, used outside of cancer, and investigating if they have ‘transferable skills’ in the field of oncology. The average cost of developing a new drug and taking it to the market is estimated at $2.56 billion with almost 88% of new drugs having their development abandoned (1). With these remarkable costs and the terrible success rates it is no surprise that scientists are looking to drugs we currently have, that we know are safe in humans, and investigating if they have further potential. Maybe the new breakthrough treatment for cancer will be a drug we already have in our local pharmacy. 

In 2011 Cancer researchers said, “a drug productivity crisis emerged in the pharmaceutical research and development area (R&D)” (2) with funding increasing to R&D without an increase in developed new drugs. A potential solution is to repurpose drugs we already have and that we know are safe in humans and use them to treat cancer. This is known as drug repurposing. The problem is finding which drugs could have potential to treat cancer. Researchers have come up with methods to ‘screen’ large numbers of drugs against cancer cells in the lab. Put simply, this involves growing cancer cells in the lab, and testing drugs ability to kill or prevent the growth of the cells. When cells are tested outside of their normal biological context, i.e., cancer cells grown in a lab rather than in a body, this is known as in vitro.  Drugs, which work in vitro, are then investigated for why they work and whether they might work in people. 

However, there are shortcomings to these methods. They can be quite laborious, and they are not ‘hypothesis driven’. This means that they test a large number of drugs sometimes without prior evidence or rationale that they might work. Additionally, in vitro testing of drugs does not take into account important clinical factors that affect whether a drug will work in the real world. For example, a lot of these drugs which are tested in vitro are metabolised in the body into non-active metabolites. This means that the drug may work to kill cancer cells in vitro but when given to a person their body metabolises that drug quickly and it never actually gets a chance to affect the tumour.

However, researchers have come up with ingenious methods to use ‘big data’ and computational models to predict which drugs might work.  ‘Big data’ is a slightly confusing term. It is used to refer to large data sets that are so complex that traditional data management tools cannot handle the data. The amazing thing about cancer research now is how much data is publicly available.

For example, The Cancer Genome Atlas project (4) collected over 11,000 tumours from 33 of the most prevalent cancer types and provides huge amounts of molecular data from those tumours. Anyone can access this data. When repurposing an already used drug there is not only data about the disease itself (say a specific cancer type) there is also data on the drug and the effect of the drug on humans, such as the chemical structure, drug targets, side effects and effects on gene expression. All this data is too much to ‘manually’ go through, certainly for multiple drugs at once. That is where computational models come in. 

Computer models can be created that can look at a particular type of cancer and can then look at the available drugs and hypothesise which might affect that disease. This can then be validated by researchers and greatly speed up the process compared to a trial and error approach. 

In one success story of this method, researchers used computational models to predict that a drug called decitabine would inhibit the growth of tumours that were dependent on a particular mutation in the KRAS gene known as K-RASG12V (4). KRAS is a protein involved in telling a cell to grow and divide. Some cancers have mutations in the gene that codes for KRAS, such as the K-RASG12V mutation. Out of the cancers that have this mutation another subset of those are reliant on this mutation for growth. This means that if you target that protein you can inhibit that tumours growth. Mottini and colleagues were able to use a validated gene ‘signature’ to predict which tumours would be dependent on this mutation to survive. They then used a computational model using available data on current drugs to predict which drugs would target this pathway and prevent the growth of these tumours. Decitabine, a drug already used to treat myelodysplastic syndromes and Acute Myeloid Leukaemia, came out of this model. They then validated that decitabine does work in these tumours and they could even predict how much it would work based on how ‘dependent’ the tumours were on the KRAS mutation. 

This is just one success story. For more, and for way more information than I went into, check out this review (5). The complexity of cancer and the drugs we use to treat it means that these ‘big data’ approaches are becoming more common to attempt to provide new treatment options. I am hugely hopeful for the future of oncology. With better technology, better ‘artificial intelligence’ and better understanding of the biology we should continue to see improved outcomes for patients.

References;

1. DiMasi JA, Grabowski HG, Hansen RW. Innovation in the pharmaceutical industry: new estimates of R&D costs. Journal of health economics. 2016 May 1;47:20-33. Accessed from: https://www.sciencedirect.com/science/article/pii/S0167629616000291?via%3Dihub

2. Pammolli, F., Magazzini, L. & Riccaboni, M. The productivity crisis in pharmaceutical R&D. Nat Rev Drug Discov 10, 428–438 (2011). https://doi.org/10.1038/nrd3405 

3. Cancer Genome Atlas Research Network, The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet. 2013 Oct;45(10):1113-20. doi: 10.1038/ng.2764. PMID: 24071849; PMCID: PMC3919969.

4. Mottini C, Tomihara H, Carrella D, Lamolinara A, Iezzi M, Huang JK, Amoreo CA, Buglioni S, Manni I, Robinson FS, Minelli R, Kang Y, Fleming JB, Kim MP, Bristow CA, Trisciuoglio D, Iuliano A, Del Bufalo D, Di Bernardo D, Melisi D, Draetta GF, Ciliberto G, Carugo A, Cardone L. Predictive Signatures Inform the Effective Repurposing of Decitabine to Treat KRAS-Dependent Pancreatic Ductal Adenocarcinoma. Cancer Res. 2019 Nov 1;79(21):5612-5625. doi: 10.1158/0008-5472.CAN-19-0187. Epub 2019 Sep 5. PMID: 31492820.

5. Mottini C, Napolitano F, Li Z, Gao X, Cardone L. Computer-aided drug repurposing for cancer therapy: Approaches and opportunities to challenge anticancer targets. Semin Cancer Biol. 2021 Jan;68:59-74. doi: 10.1016/j.semcancer.2019.09.023. Epub 2019 Sep 25. PMID: 31562957.

Author

  • Dr Craig Davison has a PhD in Medicine from Queens University Belfast (QUB). Craig has published research investigating novel treatment strategies targeting nucleotide metabolism. Craig is currently working as a postdoctoral research fellow and is passionate about science communication.